From owner-freebsd-stable Sun Jul 27 06:39:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id GAA05304 for stable-outgoing; Sun, 27 Jul 1997 06:39:55 -0700 (PDT) Received: from sev.mtelecom.ru (gw2.mtelecom.ru [195.90.159.22]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA05299 for ; Sun, 27 Jul 1997 06:39:51 -0700 (PDT) Received: (from uucp@localhost) by sev.mtelecom.ru (8.8.4/8.8.4) with UUCP id RAA10590 for freebsd-stable@FreeBSD.ORG; Sun, 27 Jul 1997 17:42:07 +0400 (MSD) Received: (from seva@localhost) by anka.da.mtelecom.ru (8.8.3/8.8.3) id RAA27488; Sun, 27 Jul 1997 17:25:22 +0400 (MSD) Date: Sun, 27 Jul 1997 17:25:22 +0400 (MSD) From: Seva Semenov Message-Id: <199707271325.RAA27488@anka.da.mtelecom.ru> To: freebsd-stable@FreeBSD.ORG Subject: Re: mount_msdos -- is it safe yet? References: <1431.869617688@time.cdrom.com> X-Newsreader: TIN [UNIX 1.3 unoff BETA release 961203] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk In article <1431.869617688@time.cdrom.com> you wrote: >> How about LFN and VFAT32 support? (I don't have the time to do it >> myself) H>They're waiting for someone who actually has the time. :) H> Jordan Yes.. btw.. is it safe to mount_nfs in 2.2.2-RELEASE? I was waiting for some1 who report it, but it seems like this only my problem. Coz I've getted and tested 4 computers - 486dx100/8M/Cnet en2000 - p120/16M/3Com595 - notebook p100/16Mb/Compex PE200 - .... forget... something with 32Mb and discovered something ugly. step 1. get 2.2.1 or 2.2.2-RELEASE from ftp.freebsd.org. step 2. install it step 3. mount_nfs some big directory from another server step 4 in ttyv0 bash$ du /nfs/ at the same time in ttyv1 remove any big direcory by _mc_ (no matter vertion 3.1 or 4.0) but not just "rm -r /nfs/big.dir" in 5-10 seconds i always getting the same result: system rebooting. No1 notify me that mount_nfs in RELEASES for tests only.... From owner-freebsd-stable Sun Jul 27 09:08:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA10839 for stable-outgoing; Sun, 27 Jul 1997 09:08:37 -0700 (PDT) Received: from mh1.cts.com (root@mh1.cts.com [205.163.24.66]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA10833 for ; Sun, 27 Jul 1997 09:08:34 -0700 (PDT) Received: from io.cts.com (io.cts.com [198.68.174.34]) by mh1.cts.com (8.8.5/8.8.5) with ESMTP id JAA10109 for ; Sun, 27 Jul 1997 09:08:33 -0700 (PDT) Received: (from mdavis@localhost) by io.cts.com (8.8.5/8.8.5) id JAA01783 for freebsd-stable@freebsd.org; Sun, 27 Jul 1997 09:08:29 -0700 (PDT) From: Morgan Davis Message-Id: <199707271608.JAA01783@io.cts.com> Subject: Support for ATAPI IDE-CDROMs To: freebsd-stable@freebsd.org Date: Sun, 27 Jul 1997 09:08:29 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk OK, what's the trick for getting an ATAPI IDE CD-ROM drive to mount? I've looked through the Handbook, FAQ, etc. Here's what I have so far: 1. wdc0, wdc1, ATAPI, wcd0 and wcd1 in kernel. 2. sh MAKEDEV wcd0 wcd1 When doing a mount_cd9660 /dev/wcd0c /cdrom I get "Device not configured". Physically, the IDE HDD's are on the Primary IDE Channel. The CD-ROM is set to be the master device on the Secondary IDE Channel. From DOS, the all the drives can be seen, including the CD-ROM drive using standard ATAPI drivers. So I know that the problem is related to software/configuration settings. Any hints? --Morgan From owner-freebsd-stable Sun Jul 27 12:34:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA18254 for stable-outgoing; Sun, 27 Jul 1997 12:34:10 -0700 (PDT) Received: from shell.uniserve.com (tom@shell.uniserve.com [204.244.210.252]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA18249 for ; Sun, 27 Jul 1997 12:34:08 -0700 (PDT) Received: from localhost (tom@localhost) by shell.uniserve.com (8.8.5/8.8.5) with SMTP id MAA07831; Sun, 27 Jul 1997 12:32:15 -0700 (PDT) X-Authentication-Warning: shell.uniserve.com: tom owned process doing -bs Date: Sun, 27 Jul 1997 12:32:15 -0700 (PDT) From: Tom To: Morgan Davis cc: freebsd-stable@FreeBSD.ORG Subject: Re: Support for ATAPI IDE-CDROMs In-Reply-To: <199707271608.JAA01783@io.cts.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, 27 Jul 1997, Morgan Davis wrote: .. > Any hints? See LINT or GENERIC for additional ATAPI kernel options you are missing. > --Morgan > Tom From owner-freebsd-stable Sun Jul 27 14:52:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA23468 for stable-outgoing; Sun, 27 Jul 1997 14:52:55 -0700 (PDT) Received: from mh1.cts.com (root@mh1.cts.com [205.163.24.66]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA23463 for ; Sun, 27 Jul 1997 14:52:53 -0700 (PDT) Received: from io.cts.com (io.cts.com [198.68.174.34]) by mh1.cts.com (8.8.5/8.8.5) with ESMTP id OAA04913; Sun, 27 Jul 1997 14:52:51 -0700 (PDT) Received: (from mdavis@localhost) by io.cts.com (8.8.5/8.8.5) id OAA00938; Sun, 27 Jul 1997 14:52:49 -0700 (PDT) From: Morgan Davis Message-Id: <199707272152.OAA00938@io.cts.com> Subject: Re: Support for ATAPI IDE-CDROMs In-Reply-To: <9707271756.aa09217@salmon.maths.tcd.ie> from David Malone at "Jul 27, 97 05:56:18 pm" To: dwmalone@maths.tcd.ie (David Malone) Date: Sun, 27 Jul 1997 14:52:49 -0700 (PDT) Cc: freebsd-stable@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk David Malone writes: > > OK, what's the trick for getting an ATAPI IDE CD-ROM drive to mount? > > Do any messages about wcd0 or wcd1 show up when you're booting? No, none. Here's the output: FreeBSD 2.2.2-RELEASE #0: Sun Jul 27 08:27:14 PDT 1997 mdavis@io.cts.com:/u/src/sys/compile/IO CPU: i486DX (486-class CPU) real memory = 41943040 (40960K bytes) avail memory = 39563264 (38636K bytes) Probing for devices on PCI bus 0: chip0 rev 2 on pci0:5 Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 at 0x300-0x31f irq 9 on isa ed0: address 00:c0:0c:20:3f:c4, type NE2000 (16 bit) sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A sio2 at 0x3e8-0x3ef irq 5 on isa sio2: type 16550A lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 765 fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 406MB (832608 sectors), 826 cyls, 16 heads, 63 S/T, 512 B/S wdc0: unit 1 (wd1): wd1: 406MB (832608 sectors), 826 cyls, 16 heads, 63 S/T, 512 B/S wdc1 at 0x170-0x177 irq 15 on isa npx0 on motherboard npx0: INT 16 interface > If not then the kernel hasn't even detected them. This will atleast > tell you if you should be looking at hardware/kernel or higher level > stuff. Here's the relevant items from my kernel config file: controller wdc0 at isa? port "IO_WD1" bio irq 14 vector wdintr disk wd0 at wdc0 drive 0 disk wd1 at wdc0 drive 1 controller wdc1 at isa? port "IO_WD2" bio irq 15 vector wdintr #disk wd2 at wdc1 drive 0 #disk wd3 at wdc1 drive 1 options ATAPI #Enable ATAPI support for IDE bus device wcd0 #IDE CD-ROM Hmmm. Now what? Thanks for your assistance. --Morgan From owner-freebsd-stable Sun Jul 27 14:53:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA23536 for stable-outgoing; Sun, 27 Jul 1997 14:53:51 -0700 (PDT) Received: from eve.umiacs.umd.edu (eve.umiacs.umd.edu [128.8.120.38]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA23527; Sun, 27 Jul 1997 14:53:47 -0700 (PDT) Received: by eve.umiacs.umd.edu (8.8.5/UMIACS-0.9/04-05-88) id RAA21969; Sun, 27 Jul 1997 17:53:45 -0400 (EDT) Date: Sun, 27 Jul 1997 17:53:45 -0400 (EDT) Message-Id: <199707272153.RAA21969@eve.umiacs.umd.edu> From: "David A. Bader" To: freebsd-stable@freebsd.org Subject: Linux ELF binaries under FreeBSD 2.2-stable Reply-to: dbader@umiacs.umd.edu Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I'm really impressed -- I finally ran several Linux ELF binaries under FreeBSD 2.2-stable, and they ran perfectly. Does anyone know what the "correct" method for running the Linux compatibility modules? Currently, I have the "COMPAT_LINUX" option compiled into my kernel, which was adequate, but I noticed that the rc.conf option for Linux (which I haven't enabled) loads the linux_mod.o LKM. Which is the better method? Thanks, david David A. Bader, Ph.D. Office: 301-405-6755 Institute for Advanced Computer Studies FAX: 301-314-9658 A.V. Williams Building Internet: dbader@umiacs.umd.edu University of Maryland WWW: http://www.umiacs.umd.edu/~dbader College Park, MD 20742 From owner-freebsd-stable Sun Jul 27 15:24:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA24748 for stable-outgoing; Sun, 27 Jul 1997 15:24:02 -0700 (PDT) Received: from as5200-port-254.no.neosoft.com (as5200-port-254.no.neosoft.com [206.27.167.254]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA24714 for ; Sun, 27 Jul 1997 15:23:47 -0700 (PDT) Received: (from conrads@localhost) by as5200-port-254.no.neosoft.com (8.8.6/8.8.6) id RAA13229; Sun, 27 Jul 1997 17:18:09 -0500 (CDT) Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199707272152.OAA00938@io.cts.com> Date: Sun, 27 Jul 1997 17:18:06 -0500 (CDT) Organization: NeoSoft, Inc. From: Conrad Sabatier To: Morgan Davis Subject: Re: Support for ATAPI IDE-CDROMs Cc: freebsd-stable@FreeBSD.ORG Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On 27-Jul-97 Morgan Davis wrote: > >Here's the relevant items from my kernel config file: > >controller wdc0 at isa? port "IO_WD1" bio irq 14 vector wdintr >disk wd0 at wdc0 drive 0 >disk wd1 at wdc0 drive 1 > >controller wdc1 at isa? port "IO_WD2" bio irq 15 vector wdintr >#disk wd2 at wdc1 drive 0 >#disk wd3 at wdc1 drive 1 > >options ATAPI #Enable ATAPI support for IDE bus >device wcd0 #IDE CD-ROM > >Hmmm. Now what? > >Thanks for your assistance. You said your CD is master on the second controller? Try uncommenting the wd2 line. I would recommend using "options ATAPI_STATIC" as well. -- Conrad Sabatier http://www.neosoft.com/~conrads/ From owner-freebsd-stable Sun Jul 27 17:19:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA29590 for stable-outgoing; Sun, 27 Jul 1997 17:19:02 -0700 (PDT) Received: from mh1.cts.com (root@mh1.cts.com [205.163.24.66]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA29583 for ; Sun, 27 Jul 1997 17:18:58 -0700 (PDT) Received: from io.cts.com (io.cts.com [198.68.174.34]) by mh1.cts.com (8.8.5/8.8.5) with ESMTP id RAA17431 for ; Sun, 27 Jul 1997 17:18:57 -0700 (PDT) Received: (from mdavis@localhost) by io.cts.com (8.8.5/8.8.5) id RAA00184; Sun, 27 Jul 1997 17:18:56 -0700 (PDT) From: Morgan Davis Message-Id: <199707280018.RAA00184@io.cts.com> Subject: Re: Support for ATAPI IDE-CDROMs (SOLUTION) In-Reply-To: <199707271608.JAA01783@io.cts.com> from Morgan Davis at "Jul 27, 97 09:08:29 am" To: mdavis@io.cts.com (Morgan Davis) Date: Sun, 27 Jul 1997 17:18:56 -0700 (PDT) Cc: freebsd-stable@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The solution is to enable ATAPI_STATIC in your kernel config. Otherwise, the probing at boot time will not see that you've got an IDE CD-ROM drive, which means LKM support after booting will be unsuccessful. Thanks for all who helped. --Morgan Morgan Davis writes: > OK, what's the trick for getting an ATAPI IDE CD-ROM drive to mount? > I've looked through the Handbook, FAQ, etc. Here's what I have so > far: > > 1. wdc0, wdc1, ATAPI, wcd0 and wcd1 in kernel. > 2. sh MAKEDEV wcd0 wcd1 > > When doing a > > mount_cd9660 /dev/wcd0c /cdrom > > I get "Device not configured". > > Physically, the IDE HDD's are on the Primary IDE Channel. The CD-ROM > is set to be the master device on the Secondary IDE Channel. From > DOS, the all the drives can be seen, including the CD-ROM drive using > standard ATAPI drivers. So I know that the problem is related to > software/configuration settings. > > Any hints? > > --Morgan > From owner-freebsd-stable Sun Jul 27 18:27:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA01869 for stable-outgoing; Sun, 27 Jul 1997 18:27:50 -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 SAA01859 for ; Sun, 27 Jul 1997 18:27:05 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id KAA02571; Mon, 28 Jul 1997 10:56:51 +0930 (CST) From: Michael Smith Message-Id: <199707280126.KAA02571@genesis.atrad.adelaide.edu.au> Subject: Re: Linux ELF binaries under FreeBSD 2.2-stable In-Reply-To: <199707272153.RAA21969@eve.umiacs.umd.edu> from "David A. Bader" at "Jul 27, 97 05:53:45 pm" To: dbader@umiacs.umd.edu Date: Mon, 28 Jul 1997 10:56:51 +0930 (CST) Cc: freebsd-stable@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-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk David A. Bader stands accused of saying: > > Does anyone know what the "correct" method for running the Linux > compatibility modules? Currently, I have the "COMPAT_LINUX" option > compiled into my kernel, which was adequate, but I noticed that the > rc.conf option for Linux (which I haven't enabled) loads the > linux_mod.o LKM. Which is the better method? The LKM is "better" in that you don't have to rebuild your kernel to make it work. > David A. Bader, Ph.D. Office: 301-405-6755 -- ]] 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-freebsd-stable Sun Jul 27 23:52:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA13730 for stable-outgoing; Sun, 27 Jul 1997 23:52:52 -0700 (PDT) Received: from btp1da.phy.uni-bayreuth.de (btp1da.phy.uni-bayreuth.de [132.180.20.32]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA13716 for ; Sun, 27 Jul 1997 23:52:45 -0700 (PDT) Received: (from werner@localhost) by btp1da.phy.uni-bayreuth.de (8.8.6/8.7.3) id IAA16125; Mon, 28 Jul 1997 08:52:32 +0200 (MEST) Message-ID: <19970728085232.27774@btp1da.phy.uni-bayreuth.de> Date: Mon, 28 Jul 1997 08:52:32 +0200 From: Werner Griessl To: Babumpabajard Cc: stable@freebsd.org Subject: Re: Upgrading to stable References: <3.0.32.19970726012212.00968e60@mail.itw.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <3.0.32.19970726012212.00968e60@mail.itw.com>; from Babumpabajard on Sat, Jul 26, 1997 at 01:22:13AM -0400 Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, Jul 26, 1997 at 01:22:13AM -0400, Babumpabajard wrote: > I recently CVsup'ed the sources for stable, and successfully did a make > world. I then used the tutorial on upgrading from sources to re-do my /dev > dirctory, and did all the other things it notes, and restarted. When it > restarted it came up with the inablity to find /dev/wd0s1f device. > > I did not do a sh MAKEDEV wd0s1f. > > This error is becuase I did not do a sh MAKEDEV wd0s1f, right? > > Now, if I start in single user mode, I can read /dev, and there is no > wd0s1f, but I can't create it becuase it's mounting /dev as a read-only > file system. Am I correct in assuming the read-only file system error is > becuase I didn't completely re-make my devices? > > Can I turn off readonly mode? Do a "mount -u /" in single user mode to make it read/write. Werner > > How can I fix this? Can I? > > Thanks for all the help I've already received, and thanks for reading this :> > > Michael From owner-freebsd-stable Mon Jul 28 00:07:49 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA14518 for stable-outgoing; Mon, 28 Jul 1997 00:07:49 -0700 (PDT) Received: from btp1da.phy.uni-bayreuth.de (btp1da.phy.uni-bayreuth.de [132.180.20.32]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA14510 for ; Mon, 28 Jul 1997 00:07:44 -0700 (PDT) Received: (from werner@localhost) by btp1da.phy.uni-bayreuth.de (8.8.6/8.7.3) id JAA16348; Mon, 28 Jul 1997 09:06:48 +0200 (MEST) Message-ID: <19970728090647.30861@btp1da.phy.uni-bayreuth.de> Date: Mon, 28 Jul 1997 09:06:47 +0200 From: Werner Griessl To: Seva Semenov Cc: stable@freebsd.org Subject: Re: mount_msdos -- is it safe yet? References: <1431.869617688@time.cdrom.com> <199707271325.RAA27488@anka.da.mtelecom.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <199707271325.RAA27488@anka.da.mtelecom.ru>; from Seva Semenov on Sun, Jul 27, 1997 at 05:25:22PM +0400 Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sun, Jul 27, 1997 at 05:25:22PM +0400, Seva Semenov wrote: > In article <1431.869617688@time.cdrom.com> you wrote: > >> How about LFN and VFAT32 support? (I don't have the time to do it > >> myself) > H>They're waiting for someone who actually has the time. :) > H> Jordan > Yes.. > btw.. is it safe to mount_nfs in 2.2.2-RELEASE? > I was waiting for some1 who report it, but it seems > like this only my problem. > Coz I've getted and tested 4 computers > - 486dx100/8M/Cnet en2000 > - p120/16M/3Com595 > - notebook p100/16Mb/Compex PE200 > - .... forget... something with 32Mb > and discovered something ugly. > > step 1. > get 2.2.1 or 2.2.2-RELEASE from ftp.freebsd.org. > step 2. > install it > step 3. > mount_nfs some big directory from another server > step 4 > in ttyv0 > bash$ du /nfs/ > at the same time in ttyv1 remove any big direcory by _mc_ > (no matter vertion 3.1 or 4.0) but not just "rm -r /nfs/big.dir" > > in 5-10 seconds i always getting the same result: system rebooting. > > No1 notify me that mount_nfs in RELEASES for tests only.... > > Use option "nfsv2" to make it work, unfortunately nfsv3 is the default. Werner From owner-freebsd-stable Mon Jul 28 22:44:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id WAA11911 for stable-outgoing; Mon, 28 Jul 1997 22:44:12 -0700 (PDT) Received: from set.sands.com (dal99.metronet.com [192.245.137.99]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA11855 for ; Mon, 28 Jul 1997 22:43:58 -0700 (PDT) Received: from set.sands.com (tsprad@localhost [127.0.0.1]) by set.sands.com (8.8.5/8.8.5) with ESMTP id UAA02308 for ; Mon, 28 Jul 1997 20:58:37 -0500 (CDT) Message-Id: <199707290158.UAA02308@set.sands.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: stable@freebsd.org Subject: Re: mount_msdos -- is it safe yet? In-reply-to: Your message of "Mon, 28 Jul 1997 09:06:47 +0200." <19970728090647.30861@btp1da.phy.uni-bayreuth.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 28 Jul 1997 20:58:36 -0500 From: Ted Spradley Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk werner@btp1da.phy.uni-bayreuth.de said: > On Sun, Jul 27, 1997 at 05:25:22PM +0400, Seva Semenov wrote: [...] > > Yes.. > > btw.. is it safe to mount_nfs in 2.2.2-RELEASE? > > I was waiting for some1 who report it, but it seems > > like this only my problem. [...] > > Use option "nfsv2" to make it work, unfortunately nfsv3 is the > default. > Werner > Could this be the cause of my difficulty with amd in 2.2.2-RELEASE? With 2.2.1 and previous versions the example /etc/amd.map worked fine with AIX 3.2.5 and DEC OSF/1 servers, with 2.2.2 I get 'Permission denied' messages. I can mount the same filesystems from the command line with no special arguments. Does anyone know the magic to get amd to use these options? I never have figured out how to get amd to use the -P option to mount a Linux filesystem, so I have to put it in fstab and keep it mounted all the time. From owner-freebsd-stable Tue Jul 29 09:40:23 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA27713 for stable-outgoing; Tue, 29 Jul 1997 09:40:23 -0700 (PDT) Received: from mailgate.greenhills.co.uk (mailgate.greenhills.co.uk [195.11.194.180]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id JAA27698 for ; Tue, 29 Jul 1997 09:40:07 -0700 (PDT) Received: (qmail 2463 invoked by uid 982); 29 Jul 1997 16:37:49 -0000 Message-ID: <19970729173749.04184@webcrawler.com> Date: Tue, 29 Jul 1997 17:37:49 +0100 From: Martijn Koster To: www@FreeBSD.ORG, freebsd-stable@FreeBSD.ORG Cc: bp@webcrawler.com Subject: Handbook/sup-file bug, or "Staying unstable with FreeBSD" Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.76 Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk This must have been noticed before. http://www.freebsd.org/handbook/handbook.html says: This handbook covers ... Release 2.2.2 http://www.freebsd.org/handbook/handbook237.html#488 says: Use the cvsup program with . ftp://ftp.freebsd.org/pub/FreeBSD/FreeBSD-current/src/share/examples/cvsup/stable-supfile has: # The following line is for 2.1-stable. If you want 2.2-stable, change # "RELENG_2_1_0" to "RELENG_2_2". *default release=cvs tag=RELENG_2_1_0 This isn't very helpful -- I didn't spot that, and have obviously been downgrading my machine, no wonder everything and its brother breaks on make world. I suggest fixing the supfile (who cares about 2_1_0?), or at the very least mentioning in handbook237.html#488 that you need to make this change. Regards, -- Martijn Koster, m.koster@pobox.com From owner-freebsd-stable Thu Jul 31 04:34:57 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id EAA00691 for stable-outgoing; Thu, 31 Jul 1997 04:34:57 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA00630; Thu, 31 Jul 1997 04:34:42 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.6/8.6.9) id EAA16307; Thu, 31 Jul 1997 04:34:38 -0700 (PDT) Date: Thu, 31 Jul 1997 04:34:38 -0700 (PDT) Message-Id: <199707311134.EAA16307@silvia.HIP.Berkeley.EDU> To: committers@freebsd.org CC: current@freebsd.org, stable@freebsd.org Reply-to: committers@freebsd.org Subject: A whole new world From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Here is a patch for 2.2-stable. As I said elsewhere, I don't have a 3.0-machine anymore, so that has to wait until thud comes back online. I believe -current people will have a reasonably good success by checking out a 2.2 /usr/src/Makefile, applying the patch, and just using that Makefile (IIRC, the previous patch made -current and -stable src/Makefile's look almost the same) in conjunction with the following patches to src/share/mk: %%%%%%%%%%%%%%%%%%% Index: share/mk/bsd.lib.mk =================================================================== RCS file: /usr/cvs/src/share/mk/bsd.lib.mk,v retrieving revision 1.59 diff -u -r1.59 bsd.lib.mk --- bsd.lib.mk 1997/06/21 15:40:32 1.59 +++ bsd.lib.mk 1997/07/01 21:15:03 @@ -186,10 +186,12 @@ > $$TMP; \ mv $$TMP ${DEPENDFILE} .endif +.if !defined(NOEXTRADEPEND) _EXTRADEPEND:: echo lib${LIB}.so.${SHLIB_MAJOR}.${SHLIB_MINOR}: \ `${LDDESTDIRENV} ${CC} -shared -Wl,-f ${LDDESTDIR} ${LDADD}` \ >> ${DEPENDFILE} +.endif .if !target(install) .if !target(beforeinstall) Index: share/mk/bsd.prog.mk =================================================================== RCS file: /usr/cvs/src/share/mk/bsd.prog.mk,v retrieving revision 1.52 diff -u -r1.52 bsd.prog.mk --- bsd.prog.mk 1997/06/28 08:14:10 1.52 +++ bsd.prog.mk 1997/07/01 21:15:03 @@ -68,7 +68,7 @@ .endif .endif -.if defined(PROG) +.if defined(PROG) && !defined(NOEXTRADEPEND) _EXTRADEPEND: echo ${PROG}: `${CC} -Wl,-f ${CFLAGS} ${LDFLAGS} ${LDDESTDIR} \ ${LDADD:S/^/-Wl,/}` >> ${DEPENDFILE} %%%%%%%%%%%%%%%%%%% This isn't really necessary unless you are building -current on a very old system. If your source tree is -stable, you don't need the above patch (it won't apply anyway ;). Now, here is a patch to Makefile. %%%%%%%%%%%%%%%%%%% Index: Makefile =================================================================== RCS file: /usr/cvs/src/Makefile,v retrieving revision 1.109.2.8 diff -u -r1.109.2.8 Makefile --- Makefile 1997/07/31 08:32:42 1.109.2.8 +++ Makefile 1997/07/31 08:41:42 @@ -31,6 +31,9 @@ # /usr/share/mk. These include: # obj depend all install clean cleandepend cleanobj +.if (!make(world)) && (!make(buildworld)) && (!make(installworld)) +.MAKEFLAGS:= ${.MAKEFLAGS} -m ${.CURDIR}/share/mk +.endif # Put initial settings here. SUBDIR= @@ -113,8 +116,8 @@ .endif .endif -SUP?= sup -SUPFLAGS?= -v +SUP?= sup +SUPFLAGS?= -v # # While building tools for bootstrapping, we dont need to waste time on @@ -135,98 +138,151 @@ @echo "make world started on `LC_TIME=C date`" @echo "--------------------------------------------------------------" .if target(pre-world) + @echo @echo "--------------------------------------------------------------" @echo " Making 'pre-world' target" @echo "--------------------------------------------------------------" cd ${.CURDIR} && ${MAKE} pre-world - @echo .endif - @echo "--------------------------------------------------------------" - @echo " Making hierarchy" - @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} hierarchy + cd ${.CURDIR} && ${MAKE} buildworld + cd ${.CURDIR} && ${MAKE} installworld +.if target(post-world) @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding /usr/share/mk" + @echo " Making 'post-world' target" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} mk + cd ${.CURDIR} && ${MAKE} post-world +.endif @echo @echo "--------------------------------------------------------------" - @echo " Cleaning up the source tree" + @echo "make world completed on `LC_TIME=C date`" @echo "--------------------------------------------------------------" -.if defined(NOCLEAN) - @echo "Not cleaning anything! I sure hope you know what you are doing!" + +.if defined(MAKEOBJDIRPREFIX) +WORLDTMP= ${MAKEOBJDIRPREFIX}${.CURDIR}/tmp .else - cd ${.CURDIR} && ${MAKE} ${CLEANDIR} +WORLDTMP= /usr/obj${.CURDIR}/tmp .endif +STRICTTMPPATH= ${WORLDTMP}/sbin:${WORLDTMP}/usr/sbin:${WORLDTMP}/bin:${WORLDTMP}/usr/bin +TMPPATH= ${STRICTTMPPATH}:${PATH} + +# XXX COMPILER_PATH is needed for finding cc1, ld and as +# XXX GCC_EXEC_PREFIX is for *crt.o. It is probably unnecssary now +# tbat LIBRARY_PATH is set. We still can't use -nostdlib, since gcc +# wouldn't link *crt.o or libgcc if it were used. +# XXX LD_LIBRARY_PATH is for ld.so. It is also used by ld, although we don't +# want that - all compile-time library paths should be resolved by gcc. +# It fails for set[ug]id executables (are any used?). +COMPILER_ENV= BISON_SIMPLE=${WORLDTMP}/usr/share/misc/bison.simple \ + COMPILER_PATH=${WORLDTMP}/usr/libexec:${WORLDTMP}/usr/bin \ + GCC_EXEC_PREFIX=${WORLDTMP}/usr/lib/ \ + LD_LIBRARY_PATH=${WORLDTMP}${SHLIBDIR} \ + LIBRARY_PATH=${WORLDTMP}${SHLIBDIR}:${WORLDTMP}/usr/lib + +BMAKEENV= PATH=${TMPPATH} ${COMPILER_ENV} NOEXTRADEPEND=t +XMAKEENV= PATH=${STRICTTMPPATH} ${COMPILER_ENV} \ + CC='cc -nostdinc' # XXX -nostdlib + +# used to compile and install 'make' in temporary build tree +IBMAKE= ${BMAKEENV} ${MAKE} DESTDIR=${WORLDTMP} +# bootstrap make +BMAKE= ${BMAKEENV} ${WORLDTMP}/usr/bin/${MAKE} DESTDIR=${WORLDTMP} +# cross make used for compilation +XMAKE= ${XMAKEENV} ${WORLDTMP}/usr/bin/${MAKE} DESTDIR=${WORLDTMP} +# cross make used for final installation +IXMAKE= ${XMAKEENV} ${WORLDTMP}/usr/bin/${MAKE} + +# +# buildworld +# +# Attempt to rebuild the entire system, with reasonable chance of +# success, regardless of how old your existing system is. +# +buildworld: +.if !defined(NOCLEAN) @echo -.if !defined(NOOBJDIR) @echo "--------------------------------------------------------------" - @echo " Rebuilding the obj tree" + @echo " Cleaning up the temporary build tree" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} obj - @echo + mkdir -p ${WORLDTMP} + chflags -R noschg ${WORLDTMP}/ + rm -rf ${WORLDTMP} .endif + @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding bootstrap tools" + @echo " Making make" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} bootstrap + mkdir -p ${WORLDTMP}/usr/bin + cd ${.CURDIR}/usr.bin/make && \ + ${IBMAKE} -I${.CURDIR}/share/mk ${OBJDIR} clean cleandepend depend && \ + ${IBMAKE} -I${.CURDIR}/share/mk ${MK_FLAGS} all install clean cleandepend @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding tools necessary to build the include files" + @echo " Making hierarchy" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} include-tools + cd ${.CURDIR} && ${BMAKE} hierarchy +.if !defined(NOCLEAN) @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding /usr/include" + @echo " Cleaning up the obj tree" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} includes + cd ${.CURDIR} && ${BMAKE} ${CLEANDIR} +.endif +.if !defined(NOOBJDIR) @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding tools needed to build the libraries" + @echo " Rebuilding the obj tree" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} lib-tools + cd ${.CURDIR} && ${BMAKE} obj +.endif @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding /usr/lib" + @echo " Rebuilding bootstrap tools" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} libraries + cd ${.CURDIR} && ${BMAKE} bootstrap @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding symorder, groff and zic(8)" + @echo " Rebuilding tools necessary to build the include files" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} build-tools + cd ${.CURDIR} && ${BMAKE} include-tools @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding dependencies" + @echo " Rebuilding /usr/include" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} depend + cd ${.CURDIR} && ${BMAKE} includes @echo @echo "--------------------------------------------------------------" - @echo " Building everything.." + @echo " Rebuilding tools needed to build the libraries" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} all + cd ${.CURDIR} && ${BMAKE} lib-tools @echo @echo "--------------------------------------------------------------" - @echo " Installing everything.." + @echo " Rebuilding /usr/lib" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} install + cd ${.CURDIR} && ${BMAKE} libraries @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding man page indexes" + @echo " Rebuilding all other tools needed to build the world" @echo "--------------------------------------------------------------" - cd ${.CURDIR}/share/man && ${MAKE} makedb -.if target(post-world) + cd ${.CURDIR} && ${BMAKE} build-tools @echo @echo "--------------------------------------------------------------" - @echo " Making 'post-world' target" + @echo " Rebuilding dependencies" @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} post-world -.endif + cd ${.CURDIR} && ${XMAKE} depend @echo @echo "--------------------------------------------------------------" - @echo "make world completed on `LC_TIME=C date`" + @echo " Building everything.." @echo "--------------------------------------------------------------" + cd ${.CURDIR} && ${XMAKE} all + +# +# installworld +# +# Installs everything compiled by a 'buildworld'. +# +installworld: + cd ${.CURDIR} && ${IXMAKE} reinstall # # reinstall @@ -242,11 +298,6 @@ cd ${.CURDIR} && ${MAKE} hierarchy @echo @echo "--------------------------------------------------------------" - @echo " Rebuilding /usr/share/mk" - @echo "--------------------------------------------------------------" - cd ${.CURDIR} && ${MAKE} mk - @echo - @echo "--------------------------------------------------------------" @echo " Installing everything.." @echo "--------------------------------------------------------------" cd ${.CURDIR} && ${MAKE} install @@ -256,7 +307,6 @@ @echo "--------------------------------------------------------------" cd ${.CURDIR}/share/man && ${MAKE} makedb - # # update # @@ -283,7 +333,6 @@ cd ${.CURDIR} && cvs -q update -P -d .endif - # # most # @@ -351,17 +400,17 @@ cd ${.CURDIR}/etc && ${MAKE} distrib-dirs # -# mk - update the /usr/share/mk makefiles. -# -mk: - cd ${.CURDIR}/share/mk && ${MAKE} install - -# # bootstrap - [re]build tools needed to run the actual build, this includes # tools needed by 'make depend', as some tools are needed to generate source # for the dependency information to be gathered from. # bootstrap: +.if defined(DESTDIR) + rm -f ${DESTDIR}/usr/src/sys + ln -s ${.CURDIR}/sys ${DESTDIR}/usr/src + cd ${.CURDIR}/include && find -dx . | cpio -dump ${DESTDIR}/usr/include + cd ${.CURDIR}/include && make symlinks +.endif cd ${.CURDIR}/usr.bin/make && ${MAKE} depend && \ ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} cd ${.CURDIR}/usr.bin/xinstall && ${MAKE} depend && \ @@ -373,11 +422,13 @@ # include-tools - generally the same as 'bootstrap', except that it's for # things that are specifically needed to generate include files. # -# XXX should be merged with bootstrap, it's not worth keeeping them seperate +# XXX should be merged with bootstrap, it's not worth keeeping them separate. +# Well, maybe it is now. We force 'cleandepend' here to avoid dependencies +# on cleaned away headers in ${WORLDTMP}. # include-tools: - cd ${.CURDIR}/usr.bin/rpcgen && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} depend all install ${CLEANDIR} ${OBJDIR} + cd ${.CURDIR}/usr.bin/rpcgen && ${MAKE} cleandepend depend && \ + ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} # # includes - possibly generate and install the include files. @@ -388,8 +439,7 @@ mtree -deU -f ${.CURDIR}/etc/mtree/BSD.include.dist \ -p ${DESTDIR}/usr/include .endif - cd ${.CURDIR}/include/rpcsvc && ${MAKE} all - cd ${.CURDIR}/include && ${MAKE} install + cd ${.CURDIR}/include && ${MAKE} all install cd ${.CURDIR}/gnu/include && ${MAKE} install cd ${.CURDIR}/gnu/lib/libreadline && ${MAKE} beforeinstall cd ${.CURDIR}/gnu/lib/libregex && ${MAKE} beforeinstall @@ -422,36 +472,31 @@ cd ${.CURDIR}/lib/libss && ${MAKE} beforeinstall cd ${.CURDIR}/lib/libscsi && ${MAKE} beforeinstall cd ${.CURDIR}/lib/libutil && ${MAKE} beforeinstall + cd ${.CURDIR}/lib/libz && ${MAKE} beforeinstall # # lib-tools - build tools to compile and install the libraries. # +# XXX gperf is required for cc +# XXX a new ld and tsort is required for cc lib-tools: - cd ${.CURDIR}/usr.bin/tsort && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/gnu/usr.bin/ld && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/gnu/usr.bin/as && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/usr.bin/ar && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/usr.bin/ranlib && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/usr.bin/nm && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/usr.bin/lex/lib && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/usr.bin/compile_et && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} && \ - rm -f /usr/sbin/compile_et - cd ${.CURDIR}/usr.bin/mk_cmds && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/gnu/usr.bin/bison && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/gnu/usr.bin/gperf && ${MAKE} depend && \ - ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} - cd ${.CURDIR}/gnu/usr.bin/cc && ${MAKE} depend && \ +.for d in \ + gnu/usr.bin/gperf \ + gnu/usr.bin/ld \ + usr.bin/tsort \ + gnu/usr.bin/as \ + gnu/usr.bin/bison \ + gnu/usr.bin/cc \ + usr.bin/ar \ + usr.bin/compile_et \ + usr.bin/lex/lib \ + usr.bin/mk_cmds \ + usr.bin/nm \ + usr.bin/ranlib \ + usr.bin/uudecode + cd ${.CURDIR}/$d && ${MAKE} depend && \ ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} +.endfor # # libraries - build and install the libraries @@ -501,15 +546,70 @@ # # build-tools - build and install any other tools needed to complete the # compile and install. +# ifdef stale +# bc and cpp are required to build groff. Otherwise, the order here is +# mostly historical, i.e., bogus. +# chmod is used to build gcc's tmpmultilib[2] at obscure times. +# endif stale +# XXX uname is a bug - the target should not depend on the host. # build-tools: .for d in \ + bin/cat \ + bin/chmod \ + bin/cp \ + bin/date \ + bin/dd \ + bin/echo \ + bin/expr \ + bin/hostname \ + bin/ln \ + bin/ls \ + bin/mkdir \ + bin/mv \ + bin/rm \ + bin/sh \ + bin/test \ + gnu/usr.bin/awk \ + gnu/usr.bin/bc \ + gnu/usr.bin/grep \ + gnu/usr.bin/groff \ + gnu/usr.bin/gzip \ + gnu/usr.bin/man/makewhatis \ + gnu/usr.bin/sort \ + gnu/usr.bin/texinfo \ share/info \ - gnu/usr.bin/texinfo \ + usr.bin/basename \ + usr.bin/cap_mkdb \ + usr.bin/chflags \ + usr.bin/cmp \ + usr.bin/col \ + usr.bin/cpp \ + usr.bin/expand \ + usr.bin/file2c \ + usr.bin/find \ + usr.bin/gencat \ + usr.bin/lorder \ + usr.bin/m4 \ + usr.bin/mkdep \ + usr.bin/paste \ + usr.bin/sed \ + usr.bin/size \ + usr.bin/soelim \ + usr.bin/strip \ usr.bin/symorder \ - usr.sbin/zic \ - gnu/usr.bin/awk \ - gnu/usr.bin/groff + usr.bin/touch \ + usr.bin/tr \ + usr.bin/true \ + usr.bin/uname \ + usr.bin/uuencode \ + usr.bin/vgrind \ + usr.bin/vi \ + usr.bin/wc \ + usr.bin/yacc \ + usr.sbin/chown \ + usr.sbin/mtree \ + usr.sbin/zic cd ${.CURDIR}/$d && ${MAKE} depend && \ ${MAKE} ${MK_FLAGS} all install ${CLEANDIR} ${OBJDIR} .endfor %%%%%%%%%%%%%%%%%%% Basically what it does is to split "world" into two parts, "buildworld" and "installworld", which can be run together or separately (or even on different machines). The "buildworld" target does not install anything outside /usr/obj; you still need to be root to run it (the default install rules want to set the ownership, etc.), but it's now possible to upgrade a bunch of -stable machines by running "make buildworld" on a 2.1.5 NFS server and then running "make installworld" on each of the clients, while not compromising the stability of the server at all. (Trust me, I've done exactly that last week. :) One thing this patch does not address (i.e., I tried, but gave up) is that your source tree still has to be called "/usr/src" -- there are far too many brokenness in the tree for me to fix by myself (and those can be hunted down/fixed independently of this project anyway). If that can be fixed, the above NFS server in the example above can even build binaries for -current and -stable machines simultaneously. If someone can take a stab at this, it is great. I would like people, especially those with very old systems, to try to upgrade your machine by using this Makefile. Everything (except for upgrading /etc) should be done with a single "make world", followed by a kernel rebuild and reboot; if it doesn't, then there is something wrong, and I want to know about it. Thanks! Satoshi From owner-freebsd-stable Thu Jul 31 12:42:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA28540 for stable-outgoing; Thu, 31 Jul 1997 12:42:02 -0700 (PDT) Received: from dragon.cmnsens.zoom.com (dragon.cmnsens.zoom.com [207.33.155.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA28534 for ; Thu, 31 Jul 1997 12:41:59 -0700 (PDT) Received: from cmnsens (cmnsens.cmnsens.zoom.com [207.33.155.2]) by dragon.cmnsens.zoom.com (8.8.6/8.8.6) with SMTP id MAA00453 for ; Thu, 31 Jul 1997 12:41:57 -0700 (PDT) Message-Id: <199707311941.MAA00453@dragon.cmnsens.zoom.com> From: "Mike Burgett" To: "freebsd-stable@freebsd.org" Date: Thu, 31 Jul 97 12:41:56 -0700 Reply-To: "Mike Burgett" Priority: Normal X-Mailer: PMMail 1.92 For OS/2 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Subject: -stable and /usr/sbin/ncrcontrol Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've been on a schedule of CVSup'ing 2.2-stable daily, and doing a 'make world install' every week or so. So far so good. This morning, I did my weekly build, and built a new kernel, and rebooted, so everything should be in sync. However, when I tried running '/usr/sbin/ncrcontrol -i', I get 'incompatible with kernel. Rebuild!' Is there something I'm missing here, or should this work? I've not tried using it before, so I have no idea if/when it ever worked here. (And yes, my system does include an NCR controller, an ASUS SC200, which has sd0, sd1, cd0, and st0 hung off of it...) Thanks, Mike From owner-freebsd-stable Thu Jul 31 13:22:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA01015 for stable-outgoing; Thu, 31 Jul 1997 13:22:16 -0700 (PDT) Received: from Octopussy.MI.Uni-Koeln.DE (Octopussy.MI.Uni-Koeln.DE [134.95.166.20]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id NAA00835; Thu, 31 Jul 1997 13:19:56 -0700 (PDT) Received: from x14.mi.uni-koeln.de (annexr2-45.slip.Uni-Koeln.DE) by Octopussy.MI.Uni-Koeln.DE with SMTP id AA12530 (5.67b/IDA-1.5); Thu, 31 Jul 1997 22:19:53 +0200 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.6/8.6.9) id WAA02227; Thu, 31 Jul 1997 22:18:30 +0200 (CEST) X-Face: " Date: Thu, 31 Jul 1997 22:18:30 +0200 From: Stefan Esser To: Mike Burgett Cc: "freebsd-stable@freebsd.org" , Stefan Esser Subject: Re: -stable and /usr/sbin/ncrcontrol References: <199707311941.MAA00453@dragon.cmnsens.zoom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.74 In-Reply-To: <199707311941.MAA00453@dragon.cmnsens.zoom.com>; from Mike Burgett on Thu, Jul 31, 1997 at 12:41:56PM -0700 Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Jul 31, Mike Burgett wrote: > I've been on a schedule of CVSup'ing 2.2-stable daily, and doing a 'make world > install' every week or so. So far so good. This morning, I did my weekly > build, and built a new kernel, and rebooted, so everything should be in sync. > > However, when I tried running '/usr/sbin/ncrcontrol -i', I get 'incompatible > with kernel. Rebuild!' The problem is that "ncrcontrol" accesses some data structure in the kernel, and whenever a kernel is built with a different definition of that structure, ncrcontrol has to be rebuilt, too. > Is there something I'm missing here, or should this work? I've not tried > using it before, so I have no idea if/when it ever worked here. Could you send me your kernel config file ? Kernel config options may effect the size of the above mentioned data structure, and the compilation of ncrcontrol can't guess the build directory of your kernel, and thus doesn't know which options you use in your kernel ... > (And yes, my system does include an NCR controller, an ASUS SC200, which has > sd0, sd1, cd0, and st0 hung off of it...) I had no doubt you really had a NCR SCSI controller ... :) Regards, STefan From owner-freebsd-stable Thu Jul 31 23:01:47 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA01387 for stable-outgoing; Thu, 31 Jul 1997 23:01:47 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA01339; Thu, 31 Jul 1997 23:01:30 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.6/8.6.9) id XAA18505; Thu, 31 Jul 1997 23:01:28 -0700 (PDT) Date: Thu, 31 Jul 1997 23:01:28 -0700 (PDT) Message-Id: <199708010601.XAA18505@silvia.HIP.Berkeley.EDU> To: current@freebsd.org, stable@freebsd.org CC: ports@freebsd.org Reply-to: ports@freebsd.org Subject: ports-current/packages-current discontinued From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk -----BEGIN PGP SIGNED MESSAGE----- Hi all, As I have mentioned in several places, I am now going to officially discontinue support of FreeBSD-current by the ports tree. The ports tree will now support FreeBSD-stable (the "2.2 branch"). As you know, the ports tree has always supported -current. Why change it now? Note that the "ports support -current" used to mean "ports support the next release" or "ports support the tree that is to become the next release, so we will have packages all built and ready when the release is made". It was true three years ago, when there was only one development tree (-current) that would become the next release. However, since then the tree was branched, the ports team has always been sandwiched between the need to support both -current and -stable. When the tree was initially branched, the promise was that the 2.1-stable branch was going to be short-lived and will only have one release, so we will just ship ports/packages from the previous CDROM. They will be only a few months old, and we'll move onto 2.2. Unfortunately, that didn't happen, the 2.1-stable tree lingered for almost two years until 2.2 was finally released (or 15 months since the 2.1 release), without getting any new support for ports. And we heard no end of it from the users, who upgraded from 2.1 to 2.1.5 and then 2.1.7 and still found the old moldy ports/packages in there. Or those who cvsup'd the ports tree and found that many things don't build on their system. Learning from that experience, we have tried to keep the ports tree compile both on 3.0-current and 2.2-stable when those two were branched. Many man-hours have gone into the tree for this effort, and we have succeeded to get it to work for more than half a year, despite some extensive changes in 3.0-current (like Garrett's network header updates). I have also provided a package containing the minimal set of utilities (the "2.2-stable upgrade kit") for 2.2.1R and 2.2.2R users so they can use the latest ports without even tracking 2.2-stable. Another thing I have done was to merge necessary changes from 3.0-current to 2.2-stable (e.g., install-info) to make the two systems reasonably close to each other. However, the recent rash of commits to the 3.0 branch without regards of consistency and compatibility with 2.2-stable made it necessary to make a decision. Also, it is often the case that -current doesn't even compile, which made it difficult for me to maintain a reasonable build machine. As we will have (at least) one more release from the 2.2-stable branch, it doesn't really make sense for us to follow the circus going on in 3.0-current now. We will make sure everything compiles and works in 2.2-stable, so when the next release comes out, we'll have a functional set of packgaes. (Also, people following -current are generally more capable of building stuff themselves. :) Note that the above doesn't mean we are not going to discontinue "#if __FreeBSD_version > ..." style of patches -- compatible porting in that manner has always been encouraged, and will continue to be so. (That will make it easier when we move over to that branch.) It just means that when it becomes impossible to support both branches without severe loads to our already scarce manpower and/or disgusting hacks, we will choose to support -stable. As many you have undoubtedly noticed, I am in the process of building packages for 2.2-stable and copying them to ftp.freebsd.org's new packages-stable directory. I will also move the ports tree on the ftp site from under "FreeBSD-current" to somewhere else to avoid confusion, and remove the "ports-current" symbolic link. The packages-current directory will eventually be deleted too (unless some kind soul steps forward to build them for me). This is truly a sad day. The irony of this all is that it really didn't have to happen, and it would have been possible for us to support both 3.0-current and 2.2-stable if more consideration has been made before some commits into -current. For instance, there is no reason why tcl-8.0beta2 had to go into the tree now, whether the tree is called -current or not. It is dubbed BETA of all things on the master site, and we haven't even had it in our ports area to test it. As some people have noted, there is no matching tk version in our ports tree either. The latest incident just reinforces my belief that it was a mistake in the first place to include something like tcl in the base distribution (note that it was also done without consensus back then), and one we may regret for a long time to come. Yours truly, Satoshi and the awesome Ports Team -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBM+F7mINA0SoeCNiJAQFxCgP/ZeTbsaIXqCgPOvBImJa3FOQdUjouo6IM jmbsoYqxgJc6bYZOPDxvyRWK2ne845FjHywSyssCNfQU82qgg2Vdot0JwXoKTGm/ 5x5ZG6nHCgRjXVEsEC6SqI3Aw9iwA0vgD+dRB5LLVhCMs/BbA5ejkSyxgq+RITKG LzQ6koCTHRg= =PHTA -----END PGP SIGNATURE----- From owner-freebsd-stable Fri Aug 1 02:27:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA11290 for stable-outgoing; Fri, 1 Aug 1997 02:27:50 -0700 (PDT) Received: from trifork.gu.net (trifork.gu.net [194.93.190.194]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA11285 for ; Fri, 1 Aug 1997 02:27:41 -0700 (PDT) Received: from localhost (localhost.gu.kiev.ua [127.0.0.1]) by trifork.gu.net (8.8.6/8.8.6) with SMTP id UAA03010 for ; Fri, 1 Aug 1997 20:23:38 +0300 (EEST) Date: Fri, 1 Aug 1997 20:23:38 +0300 (EEST) From: Andrew Stesin Reply-To: stesin@gu.net To: stable@freebsd.org Subject: *curses programs broken in 2.2-970713-RELENG ?! Message-ID: X-NCC-RegID: ua.gu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, yesterday I upgraded a machine at work to 2.2-970713-RELENG, and what the hell, all *curses programs stopped working! Like lynx, ncftp2, etc. (Pine 3.96 works, but it doesn't use curses AFAIK). Sypthoms: it seems that even `Enter' key doesn't work, I can't even enter ncftp2 command! (Pressing `Enter' causes repeating of the previous command or input hangs...) Vertical spacing/highlighting doesn't work either, all I get is a teletype-style mixture on the screen without line breaks. /bin/sh in `set -o emacs' mode works well at the same time. My surprize is unmeasurable. Any ideas or suggestions? Best regards, Andrew Stesin nic-hdl: ST73-RIPE From owner-freebsd-stable Fri Aug 1 02:29:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA11369 for stable-outgoing; Fri, 1 Aug 1997 02:29:43 -0700 (PDT) Received: from trifork.gu.net (trifork.gu.net [194.93.190.194]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA11359 for ; Fri, 1 Aug 1997 02:29:35 -0700 (PDT) Received: from localhost (localhost.gu.kiev.ua [127.0.0.1]) by trifork.gu.net (8.8.6/8.8.6) with SMTP id UAA03031 for ; Fri, 1 Aug 1997 20:25:32 +0300 (EEST) Date: Fri, 1 Aug 1997 20:25:32 +0300 (EEST) From: Andrew Stesin Reply-To: stesin@gu.net To: stable@freebsd.org Subject: Re: *curses programs broken in 2.2-970713-RELENG ?! In-Reply-To: Message-ID: X-NCC-RegID: ua.gu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Just a minor addition: ncftp2 in `line mode' works Ok! It understands all input, including arrow keys and Enter. I'll try to grab a more recent SNAP now and throw it in... On Fri, 1 Aug 1997, Andrew Stesin wrote: > Sypthoms: it seems that even `Enter' key doesn't work, > I can't even enter ncftp2 command! (Pressing `Enter' causes > repeating of the previous command or input hangs...) Vertical > spacing/highlighting doesn't work either, all I get is a teletype-style > mixture on the screen without line breaks. /bin/sh in `set -o emacs' > mode works well at the same time. > Best regards, Andrew Stesin nic-hdl: ST73-RIPE From owner-freebsd-stable Fri Aug 1 10:35:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA01376 for stable-outgoing; Fri, 1 Aug 1997 10:35:42 -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 KAA01336; Fri, 1 Aug 1997 10:35:18 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id SAA19408; Fri, 1 Aug 1997 18:30:21 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.6/8.8.6) id SAA07707; Fri, 1 Aug 1997 18:24:49 +0200 (CEST) Message-ID: <19970801182448.26268@klemm.gtn.com> Date: Fri, 1 Aug 1997 18:24:48 +0200 From: Andreas Klemm To: ports@freebsd.org Cc: current@freebsd.org, stable@freebsd.org Subject: Re: ports-current/packages-current discontinued References: <199708010601.XAA18505@silvia.HIP.Berkeley.EDU> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: Mutt 0.79 In-Reply-To: <199708010601.XAA18505@silvia.HIP.Berkeley.EDU>; from Satoshi Asami on Thu, Jul 31, 1997 at 11:01:28PM -0700 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, Jul 31, 1997 at 11:01:28PM -0700, Satoshi Asami wrote: > We will make sure everything compiles and > works in 2.2-stable, so when the next release comes out, we'll have a > functional set of packgaes. (Also, people following -current are > generally more capable of building stuff themselves. :) But unable to build really working 2.2 packages, because they only have -current :-/ > This is truly a sad day. The irony of this all is that it really > didn't have to happen, and it would have been possible for us to > support both 3.0-current and 2.2-stable if more consideration has been > made before some commits into -current. Well, wouldnīt it make more sense to discuss, to back out the changes that introduces these problems ? Very very sad is, that in my eyes tcl 8.x brings us really _nothing_ from the Operating systems view, but finally it breaks the whole concept, that -current ports can be used in 99% of cases in 2.2 and -current. We want to improve the Operating System -> thatīs FreeBSD, not compete with Linux to pickup the last bits and bytes, that break something. Why was TCL not brought into the ports collection ??? We already have several TCL versions there for backward compatibility ... why not put the newest stuff into there with respect to the problems. that now arise ? Generally we had the policy, to use the ports collection for new stuff, that belongs to FreeBSD. I think itīs time to re-think, what really belongs into the operating system and what not ... Woulnīt it be better to come to a decision, that such things like perl and tcl, _if_ they are needed in the base system, should be of the same version in -current and -stable ?! > For instance, there is no reason why tcl-8.0beta2 had to go > into the tree now, whether the tree is called -current or not. I fully agree, if I read now the big _disadvantages_ > It is dubbed BETA of all things on the master site, and we haven't > even had it in our ports area to test it. Then Iīd vote "back it out!" ;-) > As some people have noted, there is no matching tk version in our > ports tree either. The latest incident just reinforces my belief that > it was a mistake in the first place to include something like tcl in > the base distribution (note that it was also done without consensus > back then), and one we may regret for a long time to come. I think itīs time to "purify" our sources. I think itīs a bad idea, to have only a ports collection based on -STABLE. Ports should at least run with -STABLE and -CURRENT. I only run -current, create ports myself and Iīm unable to build up a -STABLE system to test my ports for 2.2-STABLE systems. And I donīt want to go back to -STABLE, because I have a SMP system and like the bleeding edge more ... Iīd like to say "peope, letīs try to find a way out of this mess" .... Andreas /// -- 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-freebsd-stable Fri Aug 1 11:35:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA04732 for stable-outgoing; Fri, 1 Aug 1997 11:35:34 -0700 (PDT) Received: from mailhost.anasazi.com (mailhost.anasazi.com [138.113.128.2]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id LAA04679; Fri, 1 Aug 1997 11:34:56 -0700 (PDT) Received: from chad.anasazi.com by mailhost.anasazi.com (5.65/3.7b) id AA26033; Fri, 1 Aug 97 11:34:06 -0700 Received: by chad.anasazi.com (5.65/3.7) id AA16893; Fri, 1 Aug 97 11:34:03 -0700 From: chad@anasazi.com (Chad R. Larson) Message-Id: <9708011834.AA16893@chad.anasazi.com> Subject: Re: ports-current/packages-current discontinued To: andreas@klemm.gtn.com (Andreas Klemm) Date: Fri, 1 Aug 1997 11:34:02 -0700 (MST) Cc: ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG In-Reply-To: <19970801182448.26268@klemm.gtn.com> from "Andreas Klemm" at Aug 1, 97 06:24:48 pm Reply-To: chad@anasazi.com X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > On Thu, Jul 31, 1997 at 11:01:28PM -0700, Satoshi Asami wrote: > > We will make sure everything compiles and > > works in 2.2-stable, so when the next release comes out, we'll have a > > functional set of packgaes. (Also, people following -current are > > generally more capable of building stuff themselves. :) > > > > [reference to TCL/TK deleted] > > It is dubbed BETA of all things on the master site, and we haven't > > even had it in our ports area to test it. > > Then I=B4d vote "back it out!" ;-) I'd second that. I'd like to see the "Release" of FreeBSD to contain the core system. User contributed stuff should all be integrated through the ports process. -crl -- Chad R. Larson (CRL22) Brother, can you paradigm? 602-870-3330 chad@anasazi.com chad@anasaz.UUCP chad@dcfinc.com Anasazi, Inc. - 7500 North Dreamy Draw Drive, Suite 120, Phoenix, Az 85020 From owner-freebsd-stable Sat Aug 2 02:28:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA13827 for stable-outgoing; Sat, 2 Aug 1997 02:28:09 -0700 (PDT) Received: from dfw-ix15.ix.netcom.com (dfw-ix15.ix.netcom.com [206.214.98.15]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA13814; Sat, 2 Aug 1997 02:28:04 -0700 (PDT) Received: (from smap@localhost) by dfw-ix15.ix.netcom.com (8.8.4/8.8.4) id EAA10704; Sat, 2 Aug 1997 04:27:32 -0500 (CDT) Received: from sjx-ca30-16.ix.netcom.com(204.31.235.176) by dfw-ix15.ix.netcom.com via smap (V1.3) id sma010692; Sat Aug 2 04:27:28 1997 Received: (from asami@localhost) by blimp.mimi.com (8.8.6/8.6.9) id CAA11125; Sat, 2 Aug 1997 02:27:25 -0700 (PDT) Date: Sat, 2 Aug 1997 02:27:25 -0700 (PDT) Message-Id: <199708020927.CAA11125@blimp.mimi.com> To: andreas@klemm.gtn.com CC: ports@freebsd.org, current@freebsd.org, stable@freebsd.org In-reply-to: <19970801182448.26268@klemm.gtn.com> (message from Andreas Klemm on Fri, 1 Aug 1997 18:24:48 +0200) Subject: Re: ports-current/packages-current discontinued From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * But unable to build really working 2.2 packages, because they * only have -current :-/ Well, I'll be doing thot from now on. Watch packages-stable on your nearest mirror site -- other than building packages for new/upgraded ports, I've been recompiling the entire tree every month or so, and that will still continue after switching to -stable. * Well, wouldn't it make more sense to discuss, to back out the * changes that introduces these problems ? You are absolutely right. * Why was TCL not brought into the ports collection ??? * We already have several TCL versions there for backward * compatibility ... why not put the newest stuff into there * with respect to the problems. that now arise ? Beats me. ;) * Wouln't it be better to come to a decision, that such * things like perl and tcl, _if_ they are needed in the * base system, should be of the same version in -current * and -stable ?! That's a big "if" there. I think the latest incident are making many people wonder "is it really worth having tcl in the base tree?". The problem with tcl is that it changes too much and often new versions are not backwards compatible with the old ones. It's not much of a problem if it isn't such an essential part of the FreeBSD environment -- but it is, and there are so many ports that depend on a particular version of tcl, it is simply too much of a pain to have "the" version in the base system yank the ports around every time it is upgraded. (The last non-continuation occurred between tcl73/tk36 and tcl74/tk40 -- but at that time, tcl was not part of the base tree so we could deal with it entirely in the ports side.) There are only two ways out of this, as far as I can tell; remove tcl from the base system (pst has done most of the work for this), or completely ignore the one in the base system and always use tcl from the ports collection (I'm not sure how hard this is -- we may need an enhanced version of LIB_DEPENDS or something). * Ports should at least run with -STABLE and -CURRENT. I only run Absolutely. I (and many others) have spend much time and effort on this; I would like to continue doing so. Satoshi P.S. By the way, every single one of the packages (that worked for -current) still compiled cleanly after I switched to -stable; the only ones that broke since my last -current build (about two weeks ago) was tk41 (broken by half-baked upgrade/switch of dependency, since fixed) and tclX75 (broken by tk41, haven't figured this one out yet). From owner-freebsd-stable Sat Aug 2 03:13:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA16003 for stable-outgoing; Sat, 2 Aug 1997 03:13:37 -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 DAA15998; Sat, 2 Aug 1997 03:13:30 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id TAA09852; Sat, 2 Aug 1997 19:43:14 +0930 (CST) From: Michael Smith Message-Id: <199708021013.TAA09852@genesis.atrad.adelaide.edu.au> Subject: Re: ports-current/packages-current discontinued In-Reply-To: <199708020927.CAA11125@blimp.mimi.com> from Satoshi Asami at "Aug 2, 97 02:27:25 am" To: asami@cs.berkeley.edu (Satoshi Asami) Date: Sat, 2 Aug 1997 19:43:14 +0930 (CST) Cc: andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@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-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I've stayed out of this thread long enough. 8) Satoshi has, IMHO, done the right thing, but for all the wrong reasons. Ports should have less (or no) "official" support on -current. If you are running -current, you should be able to take care of yourself. Naturally, merging fixes to help them run on -current is desirable, as it will make the cutover at the next major upgrade easier. > * Well, wouldn't it make more sense to discuss, to back out the > * changes that introduces these problems ? > > You are absolutely right. We've done this. Countless times. Nobody appears to have actually done very much about the results of these discussions. It is, however, apparent that moving in any _other_ direction that that eventually agreed on will only make things worse. > * Wouln't it be better to come to a decision, that such > * things like perl and tcl, _if_ they are needed in the > * base system, should be of the same version in -current > * and -stable ?! This indicates that you haven't even watched, let alone studied, the issues that have been raised in these discussions before. > There are only two ways out of this, as far as I can tell; remove tcl > from the base system (pst has done most of the work for this), or > completely ignore the one in the base system and always use tcl from > the ports collection (I'm not sure how hard this is -- we may need an > enhanced version of LIB_DEPENDS or something). The correct answer to this is, of course, that any port that requires a specific Tcl version or range of versions should require one of those versions out of the ports collection. A port failing to operate regardless of the Tcl version in the base distribution is _fundamentally_broken_, and should be fixed. -- ]] 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-freebsd-stable Sat Aug 2 04:14:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id EAA19703 for stable-outgoing; Sat, 2 Aug 1997 04:14:39 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA19665; Sat, 2 Aug 1997 04:14:24 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id MAA07164; Sat, 2 Aug 1997 12:30:14 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.6/8.8.6) id MAA02251; Sat, 2 Aug 1997 12:23:32 +0200 (CEST) Message-ID: <19970802122332.51323@klemm.gtn.com> Date: Sat, 2 Aug 1997 12:23:32 +0200 From: Andreas Klemm To: Satoshi Asami Cc: ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued References: <19970801182448.26268@klemm.gtn.com> <199708020927.CAA11125@blimp.mimi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: Mutt 0.79 In-Reply-To: <199708020927.CAA11125@blimp.mimi.com>; from Satoshi Asami on Sat, Aug 02, 1997 at 02:27:25AM -0700 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, Aug 02, 1997 at 02:27:25AM -0700, Satoshi Asami wrote: > * But unable to build really working 2.2 packages, because they > * only have -current :-/ > > Well, I'll be doing thot from now on. Watch packages-stable on your > nearest mirror site -- other than building packages for new/upgraded > ports, I've been recompiling the entire tree every month or so, and > that will still continue after switching to -stable. Hmmm, words like "if it ainīt broken, donīt fix it". The whole thing intorduces new words an phrases as "ports-stable" removes well known things like ports-current. Wouldnīt it be simpler to keep things as they are and to consolidate on a basis, that makes it again possible for you so as if nothing bad has happened ? ;-) > * Well, wouldn't it make more sense to discuss, to back out the > * changes that introduces these problems ? > > You are absolutely right. Shouldnīt we then wait for an agreement without changing things or do you want to push the discussion forward by introducing ports-stable ? > The problem with tcl is that it changes too much and often new > versions are not backwards compatible with the old ones. It's not > much of a problem if it isn't such an essential part of the FreeBSD > environment -- but it is, and there are so many ports that depend on a > particular version of tcl, it is simply too much of a pain to have > "the" version in the base system yank the ports around every time it > is upgraded. (The last non-continuation occurred between tcl73/tk36 > and tcl74/tk40 -- but at that time, tcl was not part of the base > tree so we could deal with it entirely in the ports side.) I think FreeBSD doesnīt have to use both scripting languages. Perl is _the_ straight forward enhancement of sh & Co., so that it should be ok to write utilities based on this. Desktop Managament systems or administration tool could be as well packaged as port with a proper dependency to the correct TCL/TK version. During system installation you could ask if the user wants to install the foo/bar administration tool package and FreeBSD would be purified from the bloat of huge "work in progress" packages ... > There are only two ways out of this, as far as I can tell; remove tcl > from the base system (pst has done most of the work for this), or > completely ignore the one in the base system and always use tcl from > the ports collection (I'm not sure how hard this is -- we may need an > enhanced version of LIB_DEPENDS or something). See above, Iīd vote now to remove TCP completely after seeing that results. > Absolutely. I (and many others) have spend much time and effort on > this; I would like to continue doing so. Thanks for your and all the others most valueable work. My big wish would be, not to debate about this too much, we all know our wonderful working ports mechanism, so then please use it when it is recommended. And I think it would be here. Would you all agree, to make ports of FreeBSD system utilities that rely on TCL ? Is it possible to remove TCL from the base system and keep the ports mechism as it is without introducing new declarations ? -- 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-freebsd-stable Sat Aug 2 06:28:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id GAA23466 for stable-outgoing; Sat, 2 Aug 1997 06:28:43 -0700 (PDT) Received: from 315-dialup-8.global2000.net (315-dialup-8.global2000.net [208.133.142.18]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA23460; Sat, 2 Aug 1997 06:28:38 -0700 (PDT) Message-Id: <199708021328.GAA23460@hub.freebsd.org> X-Mailer: XFMail 1.1 [p0] on FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <6046.869107933@time.cdrom.com> Date: Sat, 26 Jul 1997 23:47:15 -0400 (EDT) Organization: Griff Enterprises From: "Eric A. Griff" To: "Jordan K. Hubbard" Subject: Re: mw fails even more... Cc: freebsd-questions@freebsd.org, stable@freebsd.org, chokepnt@prima.ruhr.de, root@counterintelligence.cdrom.com, dg@root.com, Gary Kline Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On 17-Jul-97 Jordan K. Hubbard spewed out Re: mw fails even more...>> Now I'm e ven more at a loss re my upgrade to a K5 or K6 >> chip. Perhaps someone on the Core team--like you, David-- >> can give everybody advice on which CPU's do work flawlessly >> with BSD and which have known or suspected woes. > >Buy Intel and you'll have fewer worries. That's kinda the bottom >line. Not to knock Cyrix or AMD or any of Intel's competitors, but >it's just a simply fact of life that the various companies who orbit >the PC market are going to test first and most stringently with Intel >parts. > > Jordan Just my $.04 worth (inflation included) Agreed. I haven't had a make world failure here since switching from FreeBSD 3.0 back to 2.2. I have cvsuped maybe 20 times since the beginning of ap ril and done a world with every one of them, without fail. The CPU is a Intel P60, with 8M of ram, 24M swap and 900M disk space total. Of course it isn't state of the art, but it works, and seems to be 100% Intel Compatible :) ---------------------------------------------- Eric A. Griff RD#1 Box 372 Oneida, NY 13421 Phone: (315)495-2385 USA From owner-freebsd-stable Sat Aug 2 09:25:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA29897 for stable-outgoing; Sat, 2 Aug 1997 09:25:27 -0700 (PDT) Received: from ady.warp.starnets.ro (ady.warp.starnets.ro [193.226.124.33]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA29887; Sat, 2 Aug 1997 09:25:03 -0700 (PDT) Received: from localhost (ady@localhost) by ady.warp.starnets.ro (8.8.5/8.8.5) with SMTP id TAA05353; Sat, 2 Aug 1997 19:22:26 +0300 (EEST) Date: Sat, 2 Aug 1997 19:22:26 +0300 (EEST) From: Penisoara Adrian To: Michael Smith cc: Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <199708021013.TAA09852@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, On Sat, 2 Aug 1997, Michael Smith wrote: > Ports should have less (or no) "official" support on -current. If you > are running -current, you should be able to take care of yourself. Well, what about folks running 3.0-current for SMP reasons and that have not so much ideea about UN*X/FreeBSD hacking ? By chance it happens that myself too I am in the same position... > Naturally, merging fixes to help them run on -current is desirable, as > it will make the cutover at the next major upgrade easier. Not only that, the ports should be buildable on every version of FreeBSD, if possible; it comes to my mind something like GNU's autoconfig scripts... I'd really like to see *one* single ports tree, buildable on every FreeBSD version; this might be hard, I know, but I think it well worths working on it. Ady (@warp.starnets.ro) From owner-freebsd-stable Sat Aug 2 10:21:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA02212 for stable-outgoing; Sat, 2 Aug 1997 10:21:54 -0700 (PDT) Received: from labs.usn.blaze.net.au (root@labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA02207; Sat, 2 Aug 1997 10:21:45 -0700 (PDT) Received: from labs.usn.blaze.net.au (davidn@local [127.0.0.1]) by labs.usn.blaze.net.au (8.8.6/8.8.5) with ESMTP id DAA00921; Sun, 3 Aug 1997 03:20:13 +1000 (EST) Message-Id: <199708021720.DAA00921@labs.usn.blaze.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Michael Smith cc: asami@cs.berkeley.edu (Satoshi Asami), andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 19:43:14 +0930." <199708021013.TAA09852@genesis.atrad.adelaide.edu.au> 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: Sun, 03 Aug 1997 03:20:12 +1000 From: David Nugent Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > There are only two ways out of this, as far as I can tell; remove tcl > > from the base system (pst has done most of the work for this), or > > completely ignore the one in the base system and always use tcl from > > the ports collection (I'm not sure how hard this is -- we may need an > > enhanced version of LIB_DEPENDS or something). > > The correct answer to this is, of course, that any port that requires > a specific Tcl version or range of versions should require one of > those versions out of the ports collection. A port failing to operate > regardless of the Tcl version in the base distribution is > _fundamentally_broken_, and should be fixed. This is a gross over-simplification. The ports collection cannot ignore the rest of the system on which it is installed. That's the whole point of this discussion. Well, perhaps it isn't that much of an over-simplification, if you consider that all of these problems are fixed if you remove tcl from the base dist, in -current and in 2.2-stable as well (yes, this would be a lot of work, and yes, I'll volunteer to help). Just out of curiosity, why exactly was tcl8.0 *beta* added to the base distribution? I saw a message from Jordan which said "this will make my life easier" giving rather obscure reasons, but I can't express the amazement I experienced when I first saw the commit. This upgrade has caused me no end of problems, since it broke many tcl and tk apps I run on several systems. Yes, sure, this is -current, and we have to live with these inconveniences, but I'd be far more comfortable living with this particular inconvenience if the end result was the removal of tcl from the base distribution altogether, and moving it to ports. This is not an anti-tcl stance, but the opposite; I use tcl/tk widely myself, but I'd far prefer to track versions through the ports system than have to rely on the base distribution staying up-to-date, purely because many apps require different versions. The same with perl. The base distribution is simply not the place for these things; surely that much is obvious by now? Upgrading the base system's version to an untested beta with some very significant internal changes, even in -current, is the most stupid move I have seen in the FreeBSD project since my involvement. I can count the number of such incidents I've witnessed in the last couple of years on one hand, so it's not like the project is infested with stupidity. It was very ill-considered, and Satoshi's position here is critical. That he apparently got no say in this is incredible, to say the least. It seems obvious to an outsider that there are some very fundamental communication problems within the core team. Seeing this fixed is even more critical than where tcl or perl happen to reside. Regards, David -- David Nugent - Unique Computing Pty Ltd - Melbourne, Australia davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-freebsd-stable Sat Aug 2 10:54:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA03671 for stable-outgoing; Sat, 2 Aug 1997 10:54:29 -0700 (PDT) Received: from earth.mat.net (root@earth.mat.net [206.246.122.2]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id KAA03663; Sat, 2 Aug 1997 10:54:25 -0700 (PDT) Received: from Journey2.mat.net (journey2.mat.net [206.246.122.116]) by earth.mat.net (8.6.12/8.6.12) with SMTP id NAA17995; Sat, 2 Aug 1997 13:54:23 -0400 Date: Sat, 2 Aug 1997 13:54:35 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@Journey2.mat.net To: current@FreeBSD.ORG cc: ports@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <199708021720.DAA00921@labs.usn.blaze.net.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, 3 Aug 1997, David Nugent wrote: > Upgrading the base system's version to an untested beta with some very > significant internal changes, even in -current, is the most stupid move > I have seen in the FreeBSD project since my involvement. I can count the > number of such incidents I've witnessed in the last couple of years on one > hand, so it's not like the project is infested with stupidity. It was > very ill-considered, and Satoshi's position here is critical. That he > apparently got no say in this is incredible, to say the least. It seems > obvious to an outsider that there are some very fundamental communication > problems within the core team. Seeing this fixed is even more critical > than where tcl or perl happen to reside. [I trimmed out all the single names, you guys are all subscribed] Another thing to consider is the basic cleanliness of FreeBSD to tcl applications that don't arrive via the ports mechanism. Our present environment is such that even a well written tcl app hasn't the least chance of configuring itself correctly. In our respect for tcl, we've constructed a tcl-hostile system. ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 213 Lakeside Drive Apt T-1 | Greenbelt, MD 20770 | I run Journey2 and picnic, both FreeBSD (301) 220-2114 | version 3.0 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-stable Sat Aug 2 11:08:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA04334 for stable-outgoing; Sat, 2 Aug 1997 11:08:16 -0700 (PDT) Received: from critter.dk.tfs.com (critter.phk.freebsd.dk [195.8.133.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA04311; Sat, 2 Aug 1997 11:08:11 -0700 (PDT) Received: from critter.dk.tfs.com (localhost [127.0.0.1]) by critter.dk.tfs.com (8.8.6/8.8.5) with ESMTP id UAA01775; Sat, 2 Aug 1997 20:05:18 +0200 (CEST) To: David Nugent cc: Michael Smith , asami@cs.berkeley.edu (Satoshi Asami), andreas@klemm.gtn.com, ports@freebsd.org, current@freebsd.org, stable@freebsd.org From: Poul-Henning Kamp Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sun, 03 Aug 1997 03:20:12 +1000." <199708021720.DAA00921@labs.usn.blaze.net.au> Date: Sat, 02 Aug 1997 20:05:18 +0200 Message-ID: <1773.870545118@critter.dk.tfs.com> Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <199708021720.DAA00921@labs.usn.blaze.net.au>, David Nugent writes: >Just out of curiosity, why exactly was tcl8.0 *beta* added to the base >distribution? Because -current is "not yet release code" and tcl8.0b2 is "almost but not quite release code". That's why. It sounds to me like a lot of people should seriously reconsider if they ought to run -current :-( -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. Power and ignorance is a disgusting cocktail. From owner-freebsd-stable Sat Aug 2 11:34:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA05392 for stable-outgoing; Sat, 2 Aug 1997 11:34:37 -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 LAA05385; Sat, 2 Aug 1997 11:34:32 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost.cdrom.com [127.0.0.1]) by time.cdrom.com (8.8.6/8.6.9) with ESMTP id LAA15390; Sat, 2 Aug 1997 11:32:38 -0700 (PDT) To: David Nugent cc: Michael Smith , asami@cs.berkeley.edu (Satoshi Asami), andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sun, 03 Aug 1997 03:20:12 +1000." <199708021720.DAA00921@labs.usn.blaze.net.au> Date: Sat, 02 Aug 1997 11:32:38 -0700 Message-ID: <15386.870546758@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I have seen in the FreeBSD project since my involvement. I can count the > number of such incidents I've witnessed in the last couple of years on one > hand, so it's not like the project is infested with stupidity. It was > very ill-considered, and Satoshi's position here is critical. That he > apparently got no say in this is incredible, to say the least. It seems > obvious to an outsider that there are some very fundamental communication > problems within the core team. Seeing this fixed is even more critical > than where tcl or perl happen to reside. What most people here don't understand is that there was very fierce debate about this behind the scenes for some time before it was done, and what started as a TCL debate blew up into a whole "should FreeBSD have everything unbundled, from the compilers to perl, or should it bundle all the high level tools so that other system tools can be written which depend on them?" sort of fracas. One man's useful tool is another man's wasteful, unnecessary bloat, it seems, and rather than see an anti-bloatist campaign which would have led to the removal of tcl, perl, xntpd, tn3270 and a host of other utilities which are currently not deemed "essential" by the anti-bloatists, I think it was sort of deemed the lesser of two evils to just let the bloatists win the TCL debate. At least that's how I see it from my perspective - frankly, after the debate in question was over (which, again, most people here were spared), I decided I didn't even want to think about the issue for awhile and that's why I've been silent in the face of Satoshi's impassioned pleas - I don't want to go back to the bargaining table and have to decide which utilities will get the axe. Once you start with TCL, it will *not* stop there - I can only assure you of that. Perl will follow immediately behind, as will much other stuff (yes perl fans, there are many out there who consider your favorite utility language an evil, bloated monster which should not be bundled with FreeBSD at all). What we have now is a rough state of equilibrium between the two sides (who are fundamentally at odds as to what constitutes a reasonable bundling policy) and while TCL may be causing some grief, I think the bloatists are content with that state of affairs and nothing else is on the bundling horizon that I can see. Nuke TCL and you will swing the balance in the other direction, with a lot more than just TCL biting the dust as a result. Maybe that's not such a bad thing, but just so you understand how much of a "linchpin" issue this one is. Jordan From owner-freebsd-stable Sat Aug 2 11:39:17 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA05722 for stable-outgoing; Sat, 2 Aug 1997 11:39:17 -0700 (PDT) Received: from mail.scsn.net (scsn.net [206.25.246.12]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA05688; Sat, 2 Aug 1997 11:39:09 -0700 (PDT) Received: from rhiannon.scsn.net ([208.133.153.171]) by mail.scsn.net (Post.Office MTA v3.1 release PO203a ID# 0-32322U5000L100S10000) with ESMTP id AAA126; Sat, 2 Aug 1997 14:30:12 -0400 Received: (from root@localhost) by rhiannon.scsn.net (8.8.6/8.8.5) id OAA01081; Sat, 2 Aug 1997 14:38:46 -0400 (EDT) Message-ID: <19970802143845.26672@scsn.net> Date: Sat, 2 Aug 1997 14:38:45 -0400 From: "Donald J. Maddox" To: Poul-Henning Kamp Cc: David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued Reply-To: dmaddox@scsn.net References: <199708021720.DAA00921@labs.usn.blaze.net.au> <1773.870545118@critter.dk.tfs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <1773.870545118@critter.dk.tfs.com>; from Poul-Henning Kamp on Sat, Aug 02, 1997 at 08:05:18PM +0200 Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, Aug 02, 1997 at 08:05:18PM +0200, Poul-Henning Kamp wrote: > In message <199708021720.DAA00921@labs.usn.blaze.net.au>, David Nugent writes: > > >Just out of curiosity, why exactly was tcl8.0 *beta* added to the base > >distribution? > > Because -current is "not yet release code" and tcl8.0b2 is "almost but > not quite release code". > > That's why. > > It sounds to me like a lot of people should seriously reconsider if > they ought to run -current :-( > > -- > Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. > http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. > whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. > Power and ignorance is a disgusting cocktail. This seems to me to be a very silly attitude. Today's -current is tomorrow's release... If this foolishness of continuing to maintain tcl in the base tree is not stopped _now_, we will have to live with it for the next N releases. _Please_ end it *now*. From owner-freebsd-stable Sat Aug 2 12:25:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA07827 for stable-outgoing; Sat, 2 Aug 1997 12:25:54 -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 MAA07811; Sat, 2 Aug 1997 12:25:42 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost.cdrom.com [127.0.0.1]) by time.cdrom.com (8.8.6/8.6.9) with ESMTP id MAA15696; Sat, 2 Aug 1997 12:23:22 -0700 (PDT) To: dmaddox@scsn.net cc: David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 15:11:35 EDT." <19970802151135.60481@scsn.net> Date: Sat, 02 Aug 1997 12:23:21 -0700 Message-ID: <15692.870549801@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Most of this argument is just silly. Even the most vehement anti- > bloatists don't consider perl to be "an evil, bloated monster"; they just > consider _it's inclusion in the base distribution_ to be AEBM. While you I don't see the difference from the POV of this discussion so this paragraph of yours doesn't really parse for me. > In any case, I see none of this bloatist v. antibloatist propaganda > as cogent here. Tcl should not be part of the base system because it It's imminently cogent - this is NOT just a technical issue, it's an emotional one, and if you think that all software decisions are made on purely technical merits then I have a certain tower in Paris which I could make you a _great_ deal on. ;-) Jordan From owner-freebsd-stable Sat Aug 2 12:35:56 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA08362 for stable-outgoing; Sat, 2 Aug 1997 12:35:56 -0700 (PDT) Received: from mail.scsn.net (scsn.net [206.25.246.12]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA08326; Sat, 2 Aug 1997 12:35:46 -0700 (PDT) Received: from rhiannon.scsn.net ([208.133.153.148]) by mail.scsn.net (Post.Office MTA v3.1 release PO203a ID# 0-32322U5000L100S10000) with ESMTP id AAA190; Sat, 2 Aug 1997 15:25:42 -0400 Received: (from root@localhost) by rhiannon.scsn.net (8.8.6/8.8.5) id PAA01228; Sat, 2 Aug 1997 15:32:16 -0400 (EDT) Message-ID: <19970802153150.47714@scsn.net> Date: Sat, 2 Aug 1997 15:31:50 -0400 From: "Donald J. Maddox" To: "Jordan K. Hubbard" Cc: dmaddox@scsn.net, David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued Reply-To: dmaddox@scsn.net References: <19970802151135.60481@scsn.net> <15692.870549801@time.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79 In-Reply-To: <15692.870549801@time.cdrom.com>; from Jordan K. Hubbard on Sat, Aug 02, 1997 at 12:23:21PM -0700 Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, Aug 02, 1997 at 12:23:21PM -0700, Jordan K. Hubbard wrote: > > Most of this argument is just silly. Even the most vehement anti- > > bloatists don't consider perl to be "an evil, bloated monster"; they just > > consider _it's inclusion in the base distribution_ to be AEBM. While you > > I don't see the difference from the POV of this discussion so this > paragraph of yours doesn't really parse for me. Huh? > > > In any case, I see none of this bloatist v. antibloatist propaganda > > as cogent here. Tcl should not be part of the base system because it > > It's imminently cogent - this is NOT just a technical issue, it's > an emotional one, and if you think that all software decisions are > made on purely technical merits then I have a certain tower in Paris > which I could make you a _great_ deal on. ;-) No, in light of this discussion, it's clear that they are not. I humbly submit that they _should be_. Maybe you gentlemen of core should take a step back from the situation, take a few deep breaths, and reconsider this issue. From owner-freebsd-stable Sat Aug 2 12:36:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA08542 for stable-outgoing; Sat, 2 Aug 1997 12:36:26 -0700 (PDT) Received: from genghis.eng.demon.net (genghis.eng.demon.net [193.195.45.10]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA08458; Sat, 2 Aug 1997 12:36:09 -0700 (PDT) Received: from genghis.eng.demon.net [193.195.45.10] by genghis.eng.demon.net with esmtp (Exim 1.62 #1) id 0wujxY-00006R-00; Sat, 2 Aug 1997 20:35:28 +0100 To: Poul-Henning Kamp cc: ports@freebsd.org, current@freebsd.org, stable@freebsd.org Subject: Re: ports-current/packages-current discontinued Organization: Demon Internet Ltd. Reply-To: ade@demon.net In-reply-to: Your message of "Sat, 02 Aug 1997 21:15:33 +0200." <1987.870549333@critter.dk.tfs.com> Date: Sat, 02 Aug 1997 20:35:28 +0100 From: Ade Lovett Message-Id: Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Poul-Henning Kamp writes: > >In message , Ade Lovett writes: >>This seems to be something of an oversimplification. Whilst there >>are undoubtedly people who are much more likely to be better off >>running release (or perhaps -stable) code, there are quite a number >>of people who need features (SMP, for example) that are only present >>in -current. > >You know, I actualle don't see that as an excuse for running -current, >but if you insist, at least don't try to use that kind of argument >for trying to turn -current into -stable, OK ? I'm suggesting no such thing. What I am suggesting is that the 'base' FreeBSD system is now too big, and too complicated, for its own good. Perl and TCL shouldn't really be in the base system at all. Especially perl, given that it's version 4. Should I wish to install perl5, either direct from CPAN, or through ports, there's even no easy way to ensure that all traces of perl4 have been removed from my machine, which may (or may not) cause subtle problems in the future. Indeed, perl and tcl bring up an interesting discrepancy. If -current is containing newer versions of tcl than in -release or -stable, why isn't the same being done for perl? After all, now that perl 5 has been around for quite a while, and if perl is to remain part of the base system, shouldn't it be being incorporated into -current, with perl 4 being dropped? Having duplicate code, in the base system, and in ports, is just causing anguish all around. That much should be obvious :( Either perl belongs in the base system, or it doesn't. If it does, then the ports version should be killed off, if it doesn't, it needs to be removed from /usr/src. Similarly for tcl. Similarly for a whole host of other things. You (plural, not aimed at anyone in particular) can't have it both ways -- it just causes far too much grief in the long run. >>The problems come about when the base operating system contains >>components that really shouldn't be there at all. TCL, Perl certainly >>fall into this category, and there's probably quite a bit else which >>would be better off in either ports, or an 'additions' package. > >I consider "options" equally bad in protocols and operating systems. We already have "options". At install time I can choose to add a bunch of things to my machine should my heart so desire. games, proflibs, catpages, manpages etc.. Splitting up "bin" into two or more components, just in the same way that selecting "src" will offer sub-choices, depending on whether the binaries are absolutely definitely critical for system to run, or whether they are useful 'core' bolt-ons, doesn't seem to be too much of a change in direction. -aDe -- Ade Lovett, Demon Internet Ltd. From owner-freebsd-stable Sat Aug 2 12:43:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA08965 for stable-outgoing; Sat, 2 Aug 1997 12:43:26 -0700 (PDT) Received: from critter.dk.tfs.com (critter.phk.freebsd.dk [195.8.133.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA08955; Sat, 2 Aug 1997 12:43:17 -0700 (PDT) Received: from critter.dk.tfs.com (localhost [127.0.0.1]) by critter.dk.tfs.com (8.8.6/8.8.5) with ESMTP id VAA01989; Sat, 2 Aug 1997 21:15:33 +0200 (CEST) To: ade@demon.net cc: Poul-Henning Kamp , ports@freebsd.org, current@freebsd.org, stable@freebsd.org From: Poul-Henning Kamp Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 20:03:07 BST." Date: Sat, 02 Aug 1997 21:15:33 +0200 Message-ID: <1987.870549333@critter.dk.tfs.com> Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message , Ade Lovett writes: >>It sounds to me like a lot of people should seriously reconsider if >>they ought to run -current :-( > > >This seems to be something of an oversimplification. Whilst there >are undoubtedly people who are much more likely to be better off >running release (or perhaps -stable) code, there are quite a number >of people who need features (SMP, for example) that are only present >in -current. You know, I actualle don't see that as an excuse for running -current, but if you insist, at least don't try to use that kind of argument for trying to turn -current into -stable, OK ? >The problems come about when the base operating system contains >components that really shouldn't be there at all. TCL, Perl certainly >fall into this category, and there's probably quite a bit else which >would be better off in either ports, or an 'additions' package. I consider "options" equally bad in protocols and operating systems. -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. Power and ignorance is a disgusting cocktail. From owner-freebsd-stable Sat Aug 2 12:55:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA10049 for stable-outgoing; Sat, 2 Aug 1997 12:55:43 -0700 (PDT) Received: from genghis.eng.demon.net (genghis.eng.demon.net [193.195.45.10]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA09994; Sat, 2 Aug 1997 12:55:17 -0700 (PDT) Received: from genghis.eng.demon.net [193.195.45.10] by genghis.eng.demon.net with esmtp (Exim 1.62 #1) id 0wukGW-00007A-00; Sat, 2 Aug 1997 20:55:04 +0100 To: "Jordan K. Hubbard" cc: ports@freebsd.org, current@freebsd.org, stable@freebsd.org Subject: Re: ports-current/packages-current discontinued Organization: Demon Internet Ltd. Reply-To: ade@demon.net In-reply-to: Your message of "Sat, 02 Aug 1997 11:32:38 PDT." <15386.870546758@time.cdrom.com> Date: Sat, 02 Aug 1997 20:55:03 +0100 From: Ade Lovett Message-Id: Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk "Jordan K. Hubbard" writes: > >Nuke TCL and you will swing the balance in the other direction, with a >lot more than just TCL biting the dust as a result. Maybe that's not >such a bad thing, but just so you understand how much of a "linchpin" >issue this one is. Surely this would depend on how far things "bite the dust". Personally, I don't think it would be unreasonable to have a situation where the base system is made up of two distinct, but heavily inter-related parts, namely that chunk of the current core system that is definitely needed to run any kind of system at all, and other parts which aren't necessarily required, but which are stamped with a kind of "seal of approval" for use with FreeBSD. We then have a situation where both sides of the "bloat" are happy, those that want a minimalistic approach can install the required components and leave out the approved integrated packages, others that prefer a full-featured system, without having to compile/install anything else, can install both. In terms of what FreeBSD (the entity) is, however, both the required and extra package components should be considered as equally important. Possibly, a somewhat simplistic attitude, but certainly something to consider, no? -aDe -- Ade Lovett, Demon Internet Ltd. From owner-freebsd-stable Sat Aug 2 13:51:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA13430 for stable-outgoing; Sat, 2 Aug 1997 13:51:31 -0700 (PDT) Received: from critter.dk.tfs.com (critter.phk.freebsd.dk [195.8.133.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA13421; Sat, 2 Aug 1997 13:51:25 -0700 (PDT) Received: from critter.dk.tfs.com (localhost [127.0.0.1]) by critter.dk.tfs.com (8.8.6/8.8.5) with ESMTP id WAA02299; Sat, 2 Aug 1997 22:49:03 +0200 (CEST) To: ade@demon.net cc: "Jordan K. Hubbard" , ports@freebsd.org, current@freebsd.org, stable@freebsd.org From: Poul-Henning Kamp Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 20:55:03 BST." Date: Sat, 02 Aug 1997 22:49:03 +0200 Message-ID: <2297.870554943@critter.dk.tfs.com> Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Possibly, a somewhat simplistic attitude, but certainly something >to consider, no? >-- >Ade Lovett, Demon Internet Ltd. And it has been considered, and as far as I know fell because what nobody saw any chance of "minimal system" being definable so that any significant majority would agree to it. -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. Power and ignorance is a disgusting cocktail. From owner-freebsd-stable Sat Aug 2 14:12:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA14300 for stable-outgoing; Sat, 2 Aug 1997 14:12:09 -0700 (PDT) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA14256; Sat, 2 Aug 1997 14:11:48 -0700 (PDT) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id QAA18564; Sat, 2 Aug 1997 16:10:07 -0500 (EST) Date: Sat, 2 Aug 1997 16:10:07 -0500 (EST) From: John Fieber To: "Jordan K. Hubbard" cc: dmaddox@scsn.net, David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <15692.870549801@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, 2 Aug 1997, Jordan K. Hubbard wrote: > > In any case, I see none of this bloatist v. antibloatist propaganda > > as cogent here. Tcl should not be part of the base system because it > > It's imminently cogent - this is NOT just a technical issue, it's > an emotional one, and if you think that all software decisions are > made on purely technical merits then I have a certain tower in Paris > which I could make you a _great_ deal on. ;-) Ah, yes. There are many snakes in this pit. One I have not seen recently in the discussion in the inherent problem of incorporating into the base system a substantial component that is on a fundamentally different development schedule than the rest of the OS. For things with a relatively long update cycle, such as gcc, this isn't a huge problem, but for more rapidly developing items, like tcl, users stand a good chance of wanting an update between FreeBSD releases. Our only easy to use interim update mechanism is the ports collection. -john From owner-freebsd-stable Sat Aug 2 14:25:24 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA14993 for stable-outgoing; Sat, 2 Aug 1997 14:25:24 -0700 (PDT) Received: from ns2.brasserie.ca (root@[142.154.146.34]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA14975 for ; Sat, 2 Aug 1997 14:25:06 -0700 (PDT) Received: from vulcan.brasserie.ca ([142.154.146.35]) by ns2.brasserie.ca (8.7.5/8.7.3) with ESMTP id RAA02312 for ; Sat, 2 Aug 1997 17:39:09 -0400 (EDT) Message-ID: <33E3A66A.A4F07A3E@brasserie.ca> Date: Sat, 02 Aug 1997 17:28:10 -0400 From: BL Reply-To: binh@brasserie.ca Organization: Brasserie Internet X-Mailer: Mozilla 4.01 [en] (Win95; I) MIME-Version: 1.0 To: freebsd-stable@FreeBSD.ORG Subject: subscribe freebsd-stable X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk subscribe freebsd-stable From owner-freebsd-stable Sat Aug 2 14:27:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA15106 for stable-outgoing; Sat, 2 Aug 1997 14:27:11 -0700 (PDT) Received: from hwcn.org (main.hwcn.org [199.212.94.65]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA15068; Sat, 2 Aug 1997 14:27:02 -0700 (PDT) Received: from james.freenet.hamilton.on.ca (ac199@james.hwcn.org [199.212.94.66]) by hwcn.org (8.8.6/8.8.6) with ESMTP id RAA15744; Sat, 2 Aug 1997 17:27:34 -0400 (EDT) Received: from localhost (ac199@localhost) by james.freenet.hamilton.on.ca (8.8.6/8.8.6) with SMTP id RAA25135; Sat, 2 Aug 1997 17:27:51 -0400 (EDT) X-Authentication-Warning: james.freenet.hamilton.on.ca: ac199 owned process doing -bs Date: Sat, 2 Aug 1997 17:27:48 -0400 (EDT) From: Tim Vanderhoek X-Sender: ac199@james.freenet.hamilton.on.ca Reply-To: hoek@hwcn.org To: "Jordan K. Hubbard" cc: David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <15386.870546758@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, 2 Aug 1997, Jordan K. Hubbard wrote: > Once you start with TCL, it will *not* stop there - I can only > assure you of that. Perl will follow immediately behind, as > will much other stuff (yes perl fans, there are many out there > who consider your favorite utility language an evil, bloated > monster which should not be bundled with FreeBSD at all). What > we have now is a rough state of Meta-ports!! Their time has come! Not only could meta-packages answer bloatist concerns that "an installed FreeBSD system is useless unless one adds 1001 extra packages", but they could provide increased flexibility. I believe bsd.port.mk could support meta-packages right now, but I think some extra mods to it could make the whole thing work quite well indeed. [meta-ports, for those who have forgotten, are just normal ports, but whose only purpose is to build other ports.] -- Outnumbered? Maybe. Outspoken? Never! tIM...HOEk From owner-freebsd-stable Sat Aug 2 14:39:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA15644 for stable-outgoing; Sat, 2 Aug 1997 14:39:19 -0700 (PDT) Received: from unique.usn.blaze.net.au (root@unique.usn.blaze.net.au [203.17.53.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA15631; Sat, 2 Aug 1997 14:38:57 -0700 (PDT) Received: from unique.usn.blaze.net.au (davidn@local [127.0.0.1]) by unique.usn.blaze.net.au (8.8.6/8.8.5) with ESMTP id HAA14498; Sun, 3 Aug 1997 07:33:40 +1000 (EST) Message-Id: <199708022133.HAA14498@unique.usn.blaze.net.au> To: "Jordan K. Hubbard" cc: Michael Smith , asami@cs.berkeley.edu (Satoshi Asami), andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 11:32:38 MST." <15386.870546758@time.cdrom.com> X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Date: Sun, 03 Aug 1997 07:33:39 +1000 From: David Nugent Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> I have seen in the FreeBSD project since my involvement. I can count the >> number of such incidents I've witnessed in the last couple of years on one >> hand, so it's not like the project is infested with stupidity. It was >> very ill-considered, and Satoshi's position here is critical. That he >> apparently got no say in this is incredible, to say the least. It seems >> obvious to an outsider that there are some very fundamental communication >> problems within the core team. Seeing this fixed is even more critical >> than where tcl or perl happen to reside. > >What most people here don't understand is that there was very fierce >debate about this behind the scenes for some time before it was done, >and what started as a TCL debate blew up into a whole "should FreeBSD >have everything unbundled, from the compilers to perl, or should it >bundle all the high level tools so that other system tools can be >written which depend on them?" sort of fracas. One man's useful tool >is another man's wasteful, unnecessary bloat, it seems, and rather [..] Yes, yes, yes. But all completely besides the point. I don't care one way or another about "bloat". It is irrelevent to the point I was making. "Bloat" is subjective, as you point out, but it has nothing to do with how tools are updated and integrated into the system. Tools which are imported into the base system are *necessarily* more static, because of the very real concerns that have caused problems here. They won't change as often, they can't change as often, else we should by rights be carefully following perl development IN THE SOURCE TREE rather than having not only an obsolete version of perl, one completely unsupported and with seriously security bugs, and one which hardly anyone uses anyway. Tcl, until this apparently unpopular import, was at least in a better position than that. I've said before that perl4 must die. I don't particularly care whether it is imported into the source tree or left as a port. It is stale, it is old, it is obsolete and completely unsupported. It has bugs, limitations and security holes. It must go. Period. My only concern about importing a current version into the source tree is the same as tcl - what guarantees are there that this will be kept up to date? The ports system, in spite of its faults and shortfalls, *works*, and works extremely well, mainly because of the efforts of porters and Satoshi's management. But I hardly need to point this out to you of all people. The ports system integrates third party software into the base system seemlessly, provides upgrade paths and has people actively supporting (the most popular packages, anyway) and updating their favourite apps. Sure, there are mistakes there, like the recent apache vs. apache current debate, but these are far more easily handled in the less expensive and more low-maintenance system that ports has been designed to handle. Again, this has nothing to do with any "bloatist" point of view, nor is it anti-perl or anti-tcl either - I use both daily and enhance systems constantly with both of them. BUT: How is tcl essential to a working FreeBSD environment? How is perl, for that matter, other than a few scattered scripts which can and do work quite happily with current versions of perl (and can easily be replaced with /bin/sh scripts anyway)? Why are we nailing ourselves to versions of third party software which are prone to become stale and unmaintained? Why can't the ports system be used for them? >Once you start with TCL, it will *not* stop there - I can only assure >you of that. Perl will follow immediately behind, as will much other >stuff (yes perl fans, there are many out there who consider your >favorite utility language an evil, bloated monster which should not be >bundled with FreeBSD at all). What we have now is a rough state of Surely, you're not suggesting that just because something is in not in the FreeBSD base system that it is "evil"? Oh come on, this is such a childish attitude I can't believe I'm hearing it. The question is not how large a package is, it is how useful it is in the base distribution - how dependant upon it a FreeBSD system is in installation and setup, and how integrated it is into the system. Neither perl nor tcl have any grip in the system where they could be considered so essential that FreeBSD will not run without them. It is more *appropriate* that they be maintained, integrated and - ESPECIALLY - kept current via the ports system. This is a major PLUS for users, not the opposite as you seem to be assuming. perl4 in the tree is living proof of exactly where we end up with the approach that is being taken. >equilibrium between the two sides (who are fundamentally at odds as to >what constitutes a reasonable bundling policy) and while TCL may be >causing some grief, I think the bloatists are content with that state >of affairs and nothing else is on the bundling horizon that I can see. >Nuke TCL and you will swing the balance in the other direction, with a >lot more than just TCL biting the dust as a result. Maybe that's not >such a bad thing, but just so you understand how much of a "linchpin" >issue this one is. No, I understand the issue well enough. But I think it is irrevelent. Having "current" versions of things in ports and "old" things in the tree is not only annoying, it is downright ridiculous. The opposite is almost as bad. I would quickly add that any generalisation here is dangerous - you have to consider the benefits of things being integrated into the source tree vs. the costs of possibly becoming stale. It is far easier to keep our cvs "hands" of a third party package and integrate it via the ports system. Diffs between a distribution and the "FreeBSD version" are absolutely plain without even having a source repository around. Things like bind, gcc, gdb etc. are a completely different ballgame. So, no, I don't see this as any sort of weighted argument. You can cry about bloat all day, and I simple don't care. That point of view misses the issue entirely. tcl is a major headache in terms of multiple version operability. Perl is likewise (do you do much perl debugging?). I'd live silently with these problems if the benefits outweighed the cost in terms of hassle, but to me, it just isn't worth it. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-freebsd-stable Sat Aug 2 14:44:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA15971 for stable-outgoing; Sat, 2 Aug 1997 14:44:13 -0700 (PDT) Received: from unique.usn.blaze.net.au (root@unique.usn.blaze.net.au [203.17.53.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA15926; Sat, 2 Aug 1997 14:43:36 -0700 (PDT) Received: from unique.usn.blaze.net.au (davidn@local [127.0.0.1]) by unique.usn.blaze.net.au (8.8.6/8.8.5) with ESMTP id HAA14589; Sun, 3 Aug 1997 07:40:47 +1000 (EST) Message-Id: <199708022140.HAA14589@unique.usn.blaze.net.au> To: Poul-Henning Kamp cc: Michael Smith , asami@cs.berkeley.edu (Satoshi Asami), andreas@klemm.gtn.com, ports@freebsd.org, current@freebsd.org, stable@freebsd.org Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 20:05:18 +0200." <1773.870545118@critter.dk.tfs.com> X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Date: Sun, 03 Aug 1997 07:40:47 +1000 From: David Nugent Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >In message <199708021720.DAA00921@labs.usn.blaze.net.au>, David Nugent writes: >>Just out of curiosity, why exactly was tcl8.0 *beta* added to the base >>distribution? > >Because -current is "not yet release code" and tcl8.0b2 is "almost but >not quite release code". > >That's why. Well, sure, but This isn't why I asked. What features in tcl8.0 are required for a running FreeBSD system? Why is it essential that tcl be present in the base system AT ALL? What benefits are there in tcl being present here instead of being built by the ports/packages system? >It sounds to me like a lot of people should seriously reconsider if >they ought to run -current :-( Most of the systems I maintain are 2.2. However, I do software development on -current. I already mentioned that I'm quite prepared to live with any inconveniences (and there are plenty of those) absolutely without complaint. This has nothing to do with the issues I'm raising. This change appears to be entirely gratuitous. Please correct me if I am wrong. I'd seriously like to know what the benefits are of nailing ourselves to tcl8.0 in the FreeBSD base system will yield, or in fact what benefits having tcl7.5 is already giving 2.2 systems. Why is it worth the cost, hassle and upset, particularly with respect to the ports system. What compelling reasons are there or having it there as compared with the much more cheaply maintained ports system? Enquiring minds want to know. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-freebsd-stable Sat Aug 2 15:10:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA17438 for stable-outgoing; Sat, 2 Aug 1997 15:10:13 -0700 (PDT) Received: from unique.usn.blaze.net.au (root@unique.usn.blaze.net.au [203.17.53.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA17423; Sat, 2 Aug 1997 15:10:04 -0700 (PDT) Received: from unique.usn.blaze.net.au (davidn@local [127.0.0.1]) by unique.usn.blaze.net.au (8.8.6/8.8.5) with ESMTP id IAA15046; Sun, 3 Aug 1997 08:08:19 +1000 (EST) Message-Id: <199708022208.IAA15046@unique.usn.blaze.net.au> To: John Fieber cc: dmaddox@scsn.net, Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 16:10:07 EST." X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Date: Sun, 03 Aug 1997 08:08:18 +1000 From: David Nugent Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Ah, yes. There are many snakes in this pit. One I have not seen > recently in the discussion in the inherent problem of > incorporating into the base system a substantial component that > is on a fundamentally different development schedule than the > rest of the OS. For things with a relatively long update cycle, > such as gcc, this isn't a huge problem, but for more rapidly > developing items, like tcl, users stand a good chance of wanting > an update between FreeBSD releases. Our only easy to use interim > update mechanism is the ports collection. Absolutely. This is much more succinct and clear than my own argument, and it is THE fundamental and most important issue at stake here. Well said. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-freebsd-stable Sat Aug 2 15:13:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA17711 for stable-outgoing; Sat, 2 Aug 1997 15:13:10 -0700 (PDT) Received: from unique.usn.blaze.net.au (root@unique.usn.blaze.net.au [203.17.53.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA17705; Sat, 2 Aug 1997 15:13:04 -0700 (PDT) Received: from unique.usn.blaze.net.au (davidn@local [127.0.0.1]) by unique.usn.blaze.net.au (8.8.6/8.8.5) with ESMTP id IAA15131; Sun, 3 Aug 1997 08:12:36 +1000 (EST) Message-Id: <199708022212.IAA15131@unique.usn.blaze.net.au> To: hoek@hwcn.org cc: Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-reply-to: Your message of "Sat, 02 Aug 1997 17:27:48 -0400." X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Date: Sun, 03 Aug 1997 08:12:36 +1000 From: David Nugent Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Meta-ports!! Their time has come! >~ >[meta-ports, for those who have forgotten, are just normal >ports, but whose only purpose is to build other ports.] You lost me. :-) How is the structure any different from the current *_DEPENDS system? Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-freebsd-stable Sat Aug 2 15:53:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA19723 for stable-outgoing; Sat, 2 Aug 1997 15:53:37 -0700 (PDT) Received: from sendero.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA19717 for ; Sat, 2 Aug 1997 15:53:34 -0700 (PDT) Received: (from shimon@localhost) by sendero.i-connect.net (8.8.6/8.8.5) id PAA13079 for freebsd-stable@freebsd.org; Sat, 2 Aug 1997 15:53:49 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-8 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Date: Sat, 02 Aug 1997 15:53:49 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: freebsd-stable@freebsd.org Subject: Failure (minor) in make world as of today Sender: owner-freebsd-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Y'all, I am getting this when making world on RELENG_2_2, as of today: install -c -o bin -g bin -m 444 /usr/src/2.2/src/lib/libtcl/../../contrib/tcl/library/[a-z]* /usr/libdata/tcl install: /usr/src/2.2/src/lib/libtcl/../../contrib/tcl/library/http1.0: Inappropriate file type or format *** Error code 71 (continuing) It (obviously) repeats itself in two other places. Simon From owner-freebsd-stable Sat Aug 2 15:55:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA19851 for stable-outgoing; Sat, 2 Aug 1997 15:55:06 -0700 (PDT) Received: from hwcn.org (main.hwcn.org [199.212.94.65]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA19810; Sat, 2 Aug 1997 15:54:53 -0700 (PDT) Received: from james.freenet.hamilton.on.ca (ac199@james.hwcn.org [199.212.94.66]) by hwcn.org (8.8.6/8.8.6) with ESMTP id SAA22918; Sat, 2 Aug 1997 18:55:26 -0400 (EDT) Received: from localhost (ac199@localhost) by james.freenet.hamilton.on.ca (8.8.6/8.8.6) with SMTP id SAA03479; Sat, 2 Aug 1997 18:55:41 -0400 (EDT) X-Authentication-Warning: james.freenet.hamilton.on.ca: ac199 owned process doing -bs Date: Sat, 2 Aug 1997 18:55:41 -0400 (EDT) From: Tim Vanderhoek X-Sender: ac199@james.freenet.hamilton.on.ca Reply-To: hoek@hwcn.org To: David Nugent cc: hoek@hwcn.org, Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <199708022212.IAA15131@unique.usn.blaze.net.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, 3 Aug 1997, David Nugent wrote: > >Meta-ports!! Their time has come! > >~ > >[meta-ports, for those who have forgotten, are just normal > >ports, but whose only purpose is to build other ports.] > > You lost me. :-) > > How is the structure any different from the current *_DEPENDS > system? A normal port, eg. xonix-1.4, or tvp-0.9.7, will depend on other pieces of the port, but those depends are merely a side-effect ie. they are needed to compile xonix or tvp. However, a meta-port would consist of little more than a list of dependancies. Its purpose would not be to install a single piece of software (eg. tvp or xonix), but to install a whole list of software. To better support such meta-packages, bsd.port.mk could be expanded a little. A variable such as META, when turned on, could merge all the PLISTS into one big one, and force ``make package'' to include all the dependancy's binaries in the package (not currently done). Sysinstall could give these meta ports a special place in its world, making it obvious that one is expected to install them to get a "complete" system. This would have the large advantage that users of older releases could update their system by installing a meta-package or two. If the idea proves succesful, it could be possible to upgrade a whole system and kernel just by typing "pkg_add". :) Of course, this puts a lot of power in Satoshi's hands, and I think that may scare a certain few people. ;) (Also, there is precedent for installing parts of the "base system" through a port/package. Consider XFree86). -- Outnumbered? Maybe. Outspoken? Never! tIM...HOEk From owner-freebsd-stable Sat Aug 2 18:46:09 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA27009 for stable-outgoing; Sat, 2 Aug 1997 18:46:09 -0700 (PDT) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA26958; Sat, 2 Aug 1997 18:45:52 -0700 (PDT) Received: from nexgen.hiwaay.net by fly.HiWAAY.net; (8.8.6/1.1.8.2/21Sep95-1003PM) id UAA12727; Sat, 2 Aug 1997 20:45:04 -0500 (CDT) Received: from nexgen (localhost [127.0.0.1]) by nexgen.hiwaay.net (8.8.6/8.8.4) with ESMTP id UAA04956; Sat, 2 Aug 1997 20:44:57 -0500 (CDT) Message-Id: <199708030144.UAA04956@nexgen.hiwaay.net> X-Mailer: exmh version 2.0zeta 7/24/97 To: ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG From: dkelly@hiwaay.net Subject: Re: ports-current/packages-current discontinued Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 02 Aug 1997 20:42:00 -0500 Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I trimmed the individual addresses from this reply but wonder if it really should be going to (3) FreeBSD lists? Maybe its time to consider a change in a bit different direction. Is anyone familiar with the inst system SGI uses? Its almsost unmanagable from an ASCII terminal but with the graphic tool its simple enough for my boss. Every component SGI provides is registered in their installation history. Some are not optional. Others are. And the inst tool knows what is required for what, and when a newer version is installed it knows what to delete. Try to install something without the prerequisites and it'll tell you exactly what is missing, and if its available you'll be offered the chance to install it too. Otherwise you may be swapping CDs until you find it. Needless to say, this would make it very easy for someone to delete all of TCL or perl or whatever from a base FreeBSD system if we had it. Or to install it. It would be handy for me right now as I've had some system freezes, possibly always in X (one time it rebooted in the middle of the night for no reason, one time in 2 years) and I'm about to make XFree86-3.3. Would be nice to be able to cleanly remove the prior version from underneath first. Maybe in the short term utilities such as TCL, perl, and x11 should be registered in /var/db/pkg even if they are part of the base installation so they could be removed later? It means other components which require these items would also have to be registered as a package. One hitch in the FreeBSD package system that SGI doesn't have is the concept of "upgrade". If I replace version 2.8 of something with version 2.9, to cleanly do it now I have to remove 2.8. But if other packages depend on it, then I can't delete it without deleting everything that depends on it. Of course there is always the problem of packages that work with 2.8 and not 2.9... Supposedly SGI tests for this before releasing the new one. -- David Kelly N4HHE, dkelly@hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. From owner-freebsd-stable Sat Aug 2 20:41:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA02055 for stable-outgoing; Sat, 2 Aug 1997 20:41:01 -0700 (PDT) Received: from earth.mat.net (root@earth.mat.net [206.246.122.2]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id UAA02047; Sat, 2 Aug 1997 20:40:58 -0700 (PDT) Received: from Journey2.mat.net (journey2.mat.net [206.246.122.116]) by earth.mat.net (8.6.12/8.6.12) with SMTP id XAA09522; Sat, 2 Aug 1997 23:36:25 -0400 Date: Sat, 2 Aug 1997 23:36:41 -0400 (EDT) From: Chuck Robey X-Sender: chuckr@Journey2.mat.net To: "Jordan K. Hubbard" cc: dmaddox@scsn.net, David Nugent , Michael Smith , Satoshi Asami , andreas@klemm.gtn.com, ports@FreeBSD.ORG, current@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ports-current/packages-current discontinued In-Reply-To: <15692.870549801@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, 2 Aug 1997, Jordan K. Hubbard wrote: > > Most of this argument is just silly. Even the most vehement anti- > > bloatists don't consider perl to be "an evil, bloated monster"; they just > > consider _it's inclusion in the base distribution_ to be AEBM. While you > > I don't see the difference from the POV of this discussion so this > paragraph of yours doesn't really parse for me. > > > In any case, I see none of this bloatist v. antibloatist propaganda > > as cogent here. Tcl should not be part of the base system because it > > It's imminently cogent - this is NOT just a technical issue, it's > an emotional one, and if you think that all software decisions are > made on purely technical merits then I have a certain tower in Paris > which I could make you a _great_ deal on. ;-) Just want one clarification. Most of you who've talked with me know I am not one of the anti-bloatists (I like _big_ disks), so I want to make sure that having tcl out of the tree is not singularly associated with being an anti-bloatist. Everyone knows that tcl and tk are like siamese twins, bodily joined, and only separated at great risk. Don't tell me that we can't have tk in the tree, I know it requires X, and I'm not suggesting that. I'm saying that tcl requires tk, and if we can't have tk, we mustn't have tcl. All the other BSD environments are very friendly to tcl applications, so why does FreeBSD have to have such a tcl unfriendly environment? If we didn't have tcl ports, the complaint level would be even higher. It just seems like a scene from Catch-22. Why is FreeBSD a hard system to build tcl apps on? Because we like tcl ... does that make sense? This argument is not really centered on being bloatist, at least not totally. I would fight taking perl out of the kernel (I want perl5.004 brought in) but I'll be pleased to see tcl make an exit. Might a compromise be made, let tcl go away, in exchange for updating perl? ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 213 Lakeside Drive Apt T-1 | Greenbelt, MD 20770 | I run Journey2 and picnic, both FreeBSD (301) 220-2114 | version 3.0 current -- and great FUN! ----------------------------+-----------------------------------------------