From owner-freebsd-arch Sun Oct 22 8:19: 8 2000 Delivered-To: freebsd-arch@freebsd.org Received: from point.osg.gov.bc.ca (point.osg.gov.bc.ca [142.32.102.44]) by hub.freebsd.org (Postfix) with ESMTP id 0BED037B4CF for ; Sun, 22 Oct 2000 08:19:05 -0700 (PDT) Received: (from daemon@localhost) by point.osg.gov.bc.ca (8.8.7/8.8.8) id IAA04881; Sun, 22 Oct 2000 08:19:02 -0700 Received: from passer.osg.gov.bc.ca(142.32.110.29) via SMTP by point.osg.gov.bc.ca, id smtpda04879; Sun Oct 22 08:18:43 2000 Received: (from uucp@localhost) by passer.osg.gov.bc.ca (8.11.0/8.9.1) id e9MFIgK18827; Sun, 22 Oct 2000 08:18:42 -0700 (PDT) Received: from cwsys9.cwsent.com(10.2.2.1), claiming to be "cwsys.cwsent.com" via SMTP by passer9.cwsent.com, id smtpdC18825; Sun Oct 22 08:18:31 2000 Received: (from uucp@localhost) by cwsys.cwsent.com (8.11.1/8.9.1) id e9MFIUX53822; Sun, 22 Oct 2000 08:18:30 -0700 (PDT) Message-Id: <200010221518.e9MFIUX53822@cwsys.cwsent.com> Received: from localhost.cwsent.com(127.0.0.1), claiming to be "cwsys" via SMTP by localhost.cwsent.com, id smtpdz53818; Sun Oct 22 08:18:03 2000 X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4 Reply-To: Cy Schubert - ITSD Open Systems Group From: Cy Schubert - ITSD Open Systems Group X-OS: FreeBSD 4.1.1-RELEASE X-Sender: cy To: Dag-Erling Smorgrav Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-reply-to: Your message of "22 Oct 2000 01:42:50 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Sun, 22 Oct 2000 08:18:02 -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message , Dag-Erling Smorgrav = writes: > What do people think about dispensing with the Pass# argument in fstab > and instead using the following algorithm for 'fsck -p': > = > 1) if the device does not begin with a slash, assume this is a > pseudo-filesystem, and skip this entry. > = > 2) if the mountpoint is "none", skip this entry. > = > 3) if the fs type is known, run the appropriate command (which can > be null, e.g. for cd9660), and skip to the next entry. > = > 4) if the fs type is unknown, but fsck_${fstype} exists, run it and > skip to the next entry. > = > 5) print a big fat "Unknown file system type" warning and skip to > the next entry. > = > As for which order to fsck file systems in, do / first, and everything > else in parallell afterwards (possibly with additional logic to try to > identify file systems that are on the same device and fsck them > sequentially to avoid thrashing) The additional logic to identify filesystems that are on the same = device would have to be a mandatory requirement, as fsck already avoids = thrashing by fscking filesystems on the same device serially. For that = matter specifying pass# 1 for / and 2 for all others already does what = you suggest -- except for the additional error checking. > = > I'm willing to write the code if people think it's a good idea. Other than making the fsck process more windows-like by reducing the = control the admin would have over the fsck process and giving people = with little or no UNIX experience an easier time managing FreeBSD = systems, I fail to see the benefit of this change -- the additional = error checking excepted. I would be in favour of the above logic you specify within the bounds = of a pass#. Then to implement the logic you've outlined above all one = would need to do is specify pass# 1 for each filesystem. Ideally to hasten boot, we could modify fsck to only execute pass# X = and have mount only mount filesystems of pass# X. This could be used = for example as follows: 1. fsck pass# 1 2. mount all filesystems of pass# 1. 3. fsck pass# 2 in background. 4. Bring up system services that don't need files in pass# 2 (while = fsck pass# 2 runs in the background). 5. Wait for fsck pass# 2 to complete. 6. mount all filesystems of pass# 2. 7. Bring up the rest of the services that do require pass# 2. While system services that don't need filesystems listed as pass# 2 are = being initiated, we are also fscking less critical filesystems in = parallel, hopefully reducing the time it requires to boot. This of course would require fsck and mount argument changes and use of = some kind of locking mechanism. To argue against my own arguments above, what I suggest can also be = creatively implemented using noauto and rc.local. Either way, taking = control away from the admin of a critical part of system initiation = doesn't sit right with me. Regards, Phone: (250)387-8437 Cy Schubert Fax: (250)387-5766 Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca Open Systems Group, ITSD, ISTA Province of BC To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 8:39:38 2000 Delivered-To: freebsd-arch@freebsd.org Received: from flood.ping.uio.no (flood.ping.uio.no [129.240.78.31]) by hub.freebsd.org (Postfix) with ESMTP id 66DD037B4C5 for ; Sun, 22 Oct 2000 08:39:35 -0700 (PDT) Received: (from des@localhost) by flood.ping.uio.no (8.9.3/8.9.3) id RAA35322; Sun, 22 Oct 2000 17:39:31 +0200 (CEST) (envelope-from des@ofug.org) X-URL: http://www.ofug.org/~des/ X-Disclaimer: The views expressed in this message do not necessarily coincide with those of any organisation or company with which I am or have been affiliated. To: Cy Schubert - ITSD Open Systems Group Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck References: <200010221518.e9MFIUX53822@cwsys.cwsent.com> From: Dag-Erling Smorgrav Date: 22 Oct 2000 17:39:30 +0200 In-Reply-To: Cy Schubert - ITSD Open Systems Group's message of "Sun, 22 Oct 2000 08:18:02 -0700" Message-ID: Lines: 14 User-Agent: Gnus/5.0802 (Gnus v5.8.2) Emacs/20.4 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Cy Schubert - ITSD Open Systems Group writes: > Other than making the fsck process more windows-like by reducing the > control the admin would have over the fsck process and giving people > with little or no UNIX experience an easier time managing FreeBSD > systems, I fail to see the benefit of this change -- the additional > error checking excepted. If you think obviating the need for Pass# is a Bad Thing, how about making it optional (defaulting to 1) so you can still set it manually on your systems, and the rest of us won't have to? DES -- Dag-Erling Smorgrav - des@ofug.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 8:48:22 2000 Delivered-To: freebsd-arch@freebsd.org Received: from point.osg.gov.bc.ca (point.osg.gov.bc.ca [142.32.102.44]) by hub.freebsd.org (Postfix) with ESMTP id 95A9937B4C5 for ; Sun, 22 Oct 2000 08:48:20 -0700 (PDT) Received: (from daemon@localhost) by point.osg.gov.bc.ca (8.8.7/8.8.8) id IAA05211; Sun, 22 Oct 2000 08:47:47 -0700 Received: from passer.osg.gov.bc.ca(142.32.110.29) via SMTP by point.osg.gov.bc.ca, id smtpda05209; Sun Oct 22 08:47:39 2000 Received: (from uucp@localhost) by passer.osg.gov.bc.ca (8.11.0/8.9.1) id e9MFldZ19409; Sun, 22 Oct 2000 08:47:39 -0700 (PDT) Received: from cwsys9.cwsent.com(10.2.2.1), claiming to be "cwsys.cwsent.com" via SMTP by passer9.cwsent.com, id smtpdr19383; Sun Oct 22 08:47:31 2000 Received: (from uucp@localhost) by cwsys.cwsent.com (8.11.1/8.9.1) id e9MFlUe56130; Sun, 22 Oct 2000 08:47:30 -0700 (PDT) Message-Id: <200010221547.e9MFlUe56130@cwsys.cwsent.com> Received: from localhost.cwsent.com(127.0.0.1), claiming to be "cwsys" via SMTP by localhost.cwsent.com, id smtpdp56126; Sun Oct 22 08:47:19 2000 Reply-To: Cy Schubert - ITSD Open Systems Group From: Cy Schubert - ITSD Open Systems Group X-OS: FreeBSD 4.1.1-RELEASE X-Mailer: nmh 1.0.4, Exmh 2.1.1 X-Sender: cy To: Dag-Erling Smorgrav Cc: Cy Schubert - ITSD Open Systems Group , arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-reply-to: Your message of "22 Oct 2000 17:39:30 +0200." Date: Sun, 22 Oct 2000 08:47:19 -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message , Dag-Erling Smorgrav writes: > Cy Schubert - ITSD Open Systems Group writes: > > Other than making the fsck process more windows-like by reducing the > > control the admin would have over the fsck process and giving people > > with little or no UNIX experience an easier time managing FreeBSD > > systems, I fail to see the benefit of this change -- the additional > > error checking excepted. > > If you think obviating the need for Pass# is a Bad Thing, how about > making it optional (defaulting to 1) so you can still set it manually > on your systems, and the rest of us won't have to? Cool. Regards, Phone: (250)387-8437 Cy Schubert Fax: (250)387-5766 Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca Open Systems Group, ITSD, ISTA Province of BC To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 10:37:15 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id 565E637B4E5 for ; Sun, 22 Oct 2000 10:37:12 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id TAA68058; Sun, 22 Oct 2000 19:37:10 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id TAA34013; Sun, 22 Oct 2000 19:37:10 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Sun, 22 Oct 2000 19:37:10 +0200 (CEST) From: Marius Bendiksen To: Dag-Erling Smorgrav Cc: arch@freebsd.org Subject: Re: Ideas concerning fsck In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > 2) if the mountpoint is "none", skip this entry. Please don't. =) It is entirely sensible to be able to fsck a filesystem upon boot without actually mounting it. Also, I would guess this is a good thing for auto-mounter as well. > 3) if the fs type is known, run the appropriate command (which can > be null, e.g. for cd9660), and skip to the next entry. > 4) if the fs type is unknown, but fsck_${fstype} exists, run it and > skip to the next entry. Why this distinction ? I would think you'd either want to go entirely with the approach in (4), or add a new file in /etc, say "fstypes" or "fscktab". > 5) print a big fat "Unknown file system type" warning and skip to > the next entry. Adding pass #0 currently suppresses anything like that. You'd want to be able to suppress the warning. > As for which order to fsck file systems in, do / first, and everything > else in parallell afterwards (possibly with additional logic to try to > identify file systems that are on the same device and fsck them > sequentially to avoid thrashing) The logic to avoid thrashing would be a must. Currently, this can be avoided by logic on the part of the admin, by using pass 1 where neccessary. As to doing / first, why? > I'm willing to write the code if people think it's a good idea. I think it needs to be worked out some more, and needs backward-compatibility, but apart from that, I'm for. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 13:50:41 2000 Delivered-To: freebsd-arch@freebsd.org Received: from aslan.scsiguy.com (aslan.scsiguy.com [63.229.232.106]) by hub.freebsd.org (Postfix) with ESMTP id C049037B4CF for ; Sun, 22 Oct 2000 13:50:39 -0700 (PDT) Received: from aslan (localhost [127.0.0.1]) by aslan.scsiguy.com (8.11.0/8.9.3) with ESMTP id e9MKoQw52873; Sun, 22 Oct 2000 14:50:27 -0600 (MDT) (envelope-from gibbs@aslan.scsiguy.com) Message-Id: <200010222050.e9MKoQw52873@aslan.scsiguy.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: Dag-Erling Smorgrav Cc: Cy Schubert - ITSD Open Systems Group , arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: Your message of "22 Oct 2000 17:39:30 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 22 Oct 2000 14:50:26 -0600 From: "Justin T. Gibbs" Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >If you think obviating the need for Pass# is a Bad Thing, how about >making it optional (defaulting to 1) so you can still set it manually >on your systems, and the rest of us won't have to? Having a default is fine, but removing this quite useful feature is not. You can actually increase the speed with which you boot by carefully tuning the passes on multi-filesystem per disk/multiple disk systems. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 18: 0:38 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id 90ED237B65E for ; Sun, 22 Oct 2000 18:00:31 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id DAA94848; Mon, 23 Oct 2000 03:00:25 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id DAA35559; Mon, 23 Oct 2000 03:00:24 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Mon, 23 Oct 2000 03:00:24 +0200 (CEST) From: Marius Bendiksen To: Dag-Erling Smorgrav Cc: arch@freebsd.org Subject: Re: Ideas concerning fsck In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > I would think you'd either want to go entirely with the approach in (4), > > or add a new file in /etc, say "fstypes" or "fscktab". > Yes, that might be a good idea. You might want to consult Adrian on this, in connection with his work with fsck wrappers. > Doing anything at all before you know you can trust your root > partition (where fsck itself is stored) is not a very good idea. And using /etc/fstab, stored on the same partition, is a better idea? I don't think we're going to get around using a potentially corrupt root filesystem in any case, hence I am of the mind that you would want to be able to specify odd orders, but that the system would by default place '/' as the first entry. This should not cost you more work, while still maintaining the option for someone to do something we've not thought of. As a side note, we would pretty much get around the problem of a potentially unclean root file system by going with Poul-Henning's idea concerning boot file systems, which ISTR AIX and others have been using with no trouble for a long time. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Sun Oct 22 18:19:52 2000 Delivered-To: freebsd-arch@freebsd.org Received: from beastie.mckusick.com (unknown [131.161.38.3]) by hub.freebsd.org (Postfix) with ESMTP id B215737B4CF for ; Sun, 22 Oct 2000 18:19:49 -0700 (PDT) Received: from beastie.mckusick.com (localhost [127.0.0.1]) by beastie.mckusick.com (8.9.3/8.9.3) with ESMTP id QAA06345; Sun, 22 Oct 2000 16:22:53 -0700 (PDT) (envelope-from mckusick@beastie.mckusick.com) Message-Id: <200010222322.QAA06345@beastie.mckusick.com> To: Dag-Erling Smorgrav Subject: Re: Ideas concerning fsck Cc: arch@FreeBSD.ORG In-Reply-To: Your message of "22 Oct 2000 01:42:50 +0200." Date: Sun, 22 Oct 2000 16:22:53 -0700 From: Kirk McKusick Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG From: Dag-Erling Smorgrav Date: 22 Oct 2000 01:42:50 +0200 What do people think about dispensing with the Pass# argument in fstab and instead using the following algorithm for 'fsck -p': 1) if the device does not begin with a slash, assume this is a pseudo-filesystem, and skip this entry. 2) if the mountpoint is "none", skip this entry. 3) if the fs type is known, run the appropriate command (which can be null, e.g. for cd9660), and skip to the next entry. 4) if the fs type is unknown, but fsck_${fstype} exists, run it and skip to the next entry. 5) print a big fat "Unknown file system type" warning and skip to the next entry. As for which order to fsck file systems in, do / first, and everything else in parallell afterwards (possibly with additional logic to try to identify file systems that are on the same device and fsck them sequentially to avoid thrashing) I'm willing to write the code if people think it's a good idea. DES -- Dag-Erling Smørgrav - des@thinksec.com The current suggested use of the pass # is to set the root filesystem to pass 1, and everything else that you want checked with pass 2. Fsck has a heuristic to decide which filesystems are on the same drive and will run them serially (see fsck/preen.c); filesystems on different spindles will be run in parallel. Running the root separately and by itself comes from the days when it was necessary to reboot after fixing errors on the root. As that is no longer necessary, this is no longer strictly necessary. It remains just so that you have the best chance of fixing your root filesystem before starting other things that may fail due to memory limitations or I/O errors. When the root is small, the extra delay is minimal. With folks starting to use a single monster root filesystem, the cost may not be acceptable. Having given this introduction to the current scheme, I do not see a major win to getting rid of the pass numbers. There are times that I do not want a filesystem checked - one that I normally mount read-only or noauto (I'll check it if I want to mount it), or one on which I am running with soft updates (I do not yet recommend this practise, but someone has to be the test subject :-). The system install script knows how to set up the pass numbers, so a novice does not need to know anything about them. My general take is that things are fine as they are. However, not to be one to stand in the face of progress, I would consider it an acceptable to add some new sentinal (say pass -1) to say that your proposed new heuristic should be used. Kirk McKusick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 4: 2:31 2000 Delivered-To: freebsd-arch@freebsd.org Received: from aes.thinksec.com (unknown [193.212.248.16]) by hub.freebsd.org (Postfix) with ESMTP id EA9A737B479 for ; Mon, 23 Oct 2000 04:02:28 -0700 (PDT) Received: (from des@localhost) by aes.thinksec.com (8.11.0/8.11.0) id e9MIRr346827; Sun, 22 Oct 2000 20:27:53 +0200 (CEST) (envelope-from des@thinksec.com) X-Authentication-Warning: aes.thinksec.com: des set sender to des@thinksec.com using -f X-URL: http://www.ofug.org/~des/ To: Marius Bendiksen Cc: arch@freebsd.org Subject: Re: Ideas concerning fsck References: From: Dag-Erling Smorgrav Date: 22 Oct 2000 20:27:52 +0200 In-Reply-To: Marius Bendiksen's message of "Sun, 22 Oct 2000 19:37:10 +0200 (CEST)" Message-ID: Lines: 36 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Marius Bendiksen writes: > > 2) if the mountpoint is "none", skip this entry. >=20 > Please don't. =3D) >=20 > It is entirely sensible to be able to fsck a filesystem upon boot without > actually mounting it. That's what noauto is for. The "none" mountpoint denotes a swap partition. > > 3) if the fs type is known, run the appropriate command (which can > > be null, e.g. for cd9660), and skip to the next entry. > > 4) if the fs type is unknown, but fsck_${fstype} exists, run it and > > skip to the next entry. > Why this distinction ? Because you may want to treat some systems specially (e.g. don't do anything for cd9660), so you only try to run fsck_${fstype} if you don't know of anything else to do. > I would think you'd either want to go entirely with the approach in (4), > or add a new file in /etc, say "fstypes" or "fscktab". Yes, that might be a good idea. > The logic to avoid thrashing would be a must. Currently, this can be > avoided by logic on the part of the admin, by using pass 1 where > neccessary. As to doing / first, why? Doing anything at all before you know you can trust your root partition (where fsck itself is stored) is not a very good idea. DES --=20 Dag-Erling Sm=F8rgrav - des@thinksec.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 5:19:36 2000 Delivered-To: freebsd-arch@freebsd.org Received: from flood.ping.uio.no (flood.ping.uio.no [129.240.78.31]) by hub.freebsd.org (Postfix) with ESMTP id 6A24D37B479 for ; Mon, 23 Oct 2000 05:19:33 -0700 (PDT) Received: (from des@localhost) by flood.ping.uio.no (8.9.3/8.9.3) id OAA39394; Mon, 23 Oct 2000 14:19:32 +0200 (CEST) (envelope-from des@ofug.org) X-URL: http://www.ofug.org/~des/ X-Disclaimer: The views expressed in this message do not necessarily coincide with those of any organisation or company with which I am or have been affiliated. To: Marius Bendiksen Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck References: From: Dag-Erling Smorgrav Date: 23 Oct 2000 14:19:31 +0200 In-Reply-To: Marius Bendiksen's message of "Mon, 23 Oct 2000 03:00:24 +0200 (CEST)" Message-ID: Lines: 17 User-Agent: Gnus/5.0802 (Gnus v5.8.2) Emacs/20.4 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Marius Bendiksen writes: > You might want to consult Adrian on this, in connection with his work with > fsck wrappers. I've spoken to him on IRC. > > Doing anything at all before you know you can trust your root > > partition (where fsck itself is stored) is not a very good idea. > And using /etc/fstab, stored on the same partition, is a better idea? No. However, we can't get around the need for fstab, but we can get around the need for some kind of fscktab which contains static information which practically noone will ever need to change. DES -- Dag-Erling Smorgrav - des@ofug.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 5:29:22 2000 Delivered-To: freebsd-arch@freebsd.org Received: from point.osg.gov.bc.ca (point.osg.gov.bc.ca [142.32.102.44]) by hub.freebsd.org (Postfix) with ESMTP id 72AE637B479 for ; Mon, 23 Oct 2000 05:29:20 -0700 (PDT) Received: (from daemon@localhost) by point.osg.gov.bc.ca (8.8.7/8.8.8) id FAA08594; Mon, 23 Oct 2000 05:28:00 -0700 Received: from passer.osg.gov.bc.ca(142.32.110.29) via SMTP by point.osg.gov.bc.ca, id smtpda08592; Mon Oct 23 05:27:59 2000 Received: (from uucp@localhost) by passer.osg.gov.bc.ca (8.11.0/8.9.1) id e9NCRxN23988; Mon, 23 Oct 2000 05:27:59 -0700 (PDT) Received: from cwsys9.cwsent.com(10.2.2.1), claiming to be "cwsys.cwsent.com" via SMTP by passer9.cwsent.com, id smtpds23984; Mon Oct 23 05:27:03 2000 Received: (from uucp@localhost) by cwsys.cwsent.com (8.11.1/8.9.1) id e9NCR1N92054; Mon, 23 Oct 2000 05:27:01 -0700 (PDT) Message-Id: <200010231227.e9NCR1N92054@cwsys.cwsent.com> Received: from localhost.cwsent.com(127.0.0.1), claiming to be "cwsys" via SMTP by localhost.cwsent.com, id smtpdq91847; Mon Oct 23 05:26:26 2000 X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4 Reply-To: Cy Schubert - ITSD Open Systems Group From: Cy Schubert - ITSD Open Systems Group X-OS: FreeBSD 4.1.1-RELEASE X-Sender: cy To: Marius Bendiksen Cc: Dag-Erling Smorgrav , arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-reply-to: Your message of "Sun, 22 Oct 2000 19:37:10 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 23 Oct 2000 05:26:22 -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message , Marius Bendiksen writes: > The logic to avoid thrashing would be a must. Currently, this can be > avoided by logic on the part of the admin, by using pass 1 where > neccessary. As to doing / first, why? Fsck currently avoids thrashing. Take a look at the first paragraph of the man page. Regards, Phone: (250)387-8437 Cy Schubert Fax: (250)387-5766 Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca Open Systems Group, ITSD, ISTA Province of BC To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 10:25:36 2000 Delivered-To: freebsd-arch@freebsd.org Received: from post.mail.nl.demon.net (post-10.mail.nl.demon.net [194.159.73.20]) by hub.freebsd.org (Postfix) with ESMTP id 8069037B479 for ; Mon, 23 Oct 2000 10:25:33 -0700 (PDT) Received: from [212.238.54.101] (helo=freebie.demon.nl) by post.mail.nl.demon.net with smtp (Exim 3.14 #2) id 13nlLk-0006of-00; Mon, 23 Oct 2000 17:25:30 +0000 Received: (from wkb@localhost) by freebie.demon.nl (8.11.0/8.11.0) id e9NHRx601135; Mon, 23 Oct 2000 19:27:59 +0200 (CEST) (envelope-from wkb) Date: Mon, 23 Oct 2000 19:27:59 +0200 From: Wilko Bulte To: Kirk McKusick Cc: Dag-Erling Smorgrav , arch@freebsd.org Subject: Re: Ideas concerning fsck Message-ID: <20001023192759.A1077@freebie.demon.nl> References: <200010222322.QAA06345@beastie.mckusick.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <200010222322.QAA06345@beastie.mckusick.com>; from mckusick@mckusick.com on Sun, Oct 22, 2000 at 04:22:53PM -0700 X-OS: FreeBSD 4.1-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sun, Oct 22, 2000 at 04:22:53PM -0700, Kirk McKusick wrote: > From: Dag-Erling Smorgrav > Date: 22 Oct 2000 01:42:50 +0200 > > What do people think about dispensing with the Pass# argument in fstab > and instead using the following algorithm for 'fsck -p': > > 1) if the device does not begin with a slash, assume this is a > pseudo-filesystem, and skip this entry. > > 2) if the mountpoint is "none", skip this entry. > > 3) if the fs type is known, run the appropriate command (which can > be null, e.g. for cd9660), and skip to the next entry. > > 4) if the fs type is unknown, but fsck_${fstype} exists, run it and > skip to the next entry. > > 5) print a big fat "Unknown file system type" warning and skip to > the next entry. > > As for which order to fsck file systems in, do / first, and everything > else in parallell afterwards (possibly with additional logic to try to > identify file systems that are on the same device and fsck them > sequentially to avoid thrashing) > > I'm willing to write the code if people think it's a good idea. > > DES > -- > Dag-Erling Smrgrav - des@thinksec.com > > The current suggested use of the pass # is to set the root filesystem > to pass 1, and everything else that you want checked with pass 2. Fsck > has a heuristic to decide which filesystems are on the same drive and > will run them serially (see fsck/preen.c); filesystems on different > spindles will be run in parallel. Running the root separately and by I could imagine that for multiple filesystems sharing one (hardware) RAID volume one might want to be able to override this heuristic. The inherent parallelism allowed by RAID would allow checking of multiple filesystems on the (apparantly) same spindle without speed degradation/trashing. W/ -- Wilko Bulte Arnhem, the Netherlands wilko@freebsd.org http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 15:41:53 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 818B137B479 for ; Mon, 23 Oct 2000 15:41:50 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9NMfmn83663; Mon, 23 Oct 2000 16:41:49 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id QAA10916; Mon, 23 Oct 2000 16:41:48 -0600 (MDT) Message-Id: <200010232241.QAA10916@harmony.village.org> To: Nick Sayer Cc: arch@freebsd.org Subject: Re: Old 4.0-R binaries remaining in 4.1.1-S In-reply-to: Your message of "Mon, 23 Oct 2000 15:21:30 PDT." <39F4B9EA.FACF4464@sftw.com> References: <39F4B9EA.FACF4464@sftw.com> <200010230328.XAA16342@world.std.com> <200010231857.MAA09601@harmony.village.org> Date: Mon, 23 Oct 2000 16:41:48 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <39F4B9EA.FACF4464@sftw.com> Nick Sayer writes: : I suggest that UPDATING tell folks do an ls -alt on /bin, /sbin, : /usr/bin, /usr/sbin, /usr/libexec, /usr/lib and /modules and look for : old stuff. In the case of /usr/lib, *.so.*.* should be moved to : /usr/lib/compat/aout and *.so.[0-9] should be moved to /usr/lib/compat. : All others can be deleted. Searching and deprecating old stuff in : /usr/share and /usr/include is... undefined. :-) I'd prefer that we have a mtree.obsolete that can be used to find and delete the files. NetBSD has this concept (I haven't looked at it, since they do it as part of the binary upgrade), which might make sense for us too. How do people feel about someone going off and doing this? Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 16: 9:24 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id D4EB037B4C5 for ; Mon, 23 Oct 2000 16:09:21 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9NN9Jn83753; Mon, 23 Oct 2000 17:09:20 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id RAA11156; Mon, 23 Oct 2000 17:09:19 -0600 (MDT) Message-Id: <200010232309.RAA11156@harmony.village.org> To: Anders Franzen Subject: Re: loadable drivers Cc: freebsd-arch@FreeBSD.ORG In-reply-to: Your message of "Wed, 18 Oct 2000 14:44:50 +0200." <39ED9B42.417E88FD@uab.ericsson.se> References: <39ED9B42.417E88FD@uab.ericsson.se> Date: Mon, 23 Oct 2000 17:09:19 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <39ED9B42.417E88FD@uab.ericsson.se> Anders Franzen writes: : Hi, I'm playing around with the linux pcmcia/cardbus drivers. What does this have to do with FreeBSD? : Since I crash quite often I would like to do it with modules. : The problem is that if I don't put any driver for the pcicl chip : the PCI bus driver hands it over to the chip 'device/bus' and I : wont see it when I attach the module to the PCI bus. : : Is there anyway of doing this or must certain drivers (bridges) be : linked : in the kernel? Sure. The cardbus drivers haven't been tested as loadable modules, but pccard and pcic work great as far as I can tell. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 16:11:46 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 64BAF37B479 for ; Mon, 23 Oct 2000 16:11:44 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9NNBhn83764; Mon, 23 Oct 2000 17:11:43 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id RAA11179; Mon, 23 Oct 2000 17:11:42 -0600 (MDT) Message-Id: <200010232311.RAA11179@harmony.village.org> To: "Justin T. Gibbs" Subject: Re: Ideas concerning fsck Cc: Dag-Erling Smorgrav , Cy Schubert - ITSD Open Systems Group , arch@FreeBSD.ORG In-reply-to: Your message of "Sun, 22 Oct 2000 14:50:26 MDT." <200010222050.e9MKoQw52873@aslan.scsiguy.com> References: <200010222050.e9MKoQw52873@aslan.scsiguy.com> Date: Mon, 23 Oct 2000 17:11:42 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <200010222050.e9MKoQw52873@aslan.scsiguy.com> "Justin T. Gibbs" writes: : >If you think obviating the need for Pass# is a Bad Thing, how about : >making it optional (defaulting to 1) so you can still set it manually : >on your systems, and the rest of us won't have to? : : Having a default is fine, but removing this quite useful feature is not. : You can actually increase the speed with which you boot by carefully : tuning the passes on multi-filesystem per disk/multiple disk systems. Especially since you have multiple scsi buses and multiple drives on them. You can keep things well balanced with pass numbers. Please don't remove them. I set them all the time. Mostly so we don't try to fsck all the file systems on the same spindle at the same time. It is faster to do them serially, in my experience, than in parallel due to disk seek times. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 16:26:49 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 3F71837B479 for ; Mon, 23 Oct 2000 16:26:47 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9NNQkn83855; Mon, 23 Oct 2000 17:26:46 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id RAA11369; Mon, 23 Oct 2000 17:26:45 -0600 (MDT) Message-Id: <200010232326.RAA11369@harmony.village.org> To: Jordan Hubbard Subject: Re: cvs commit: src/etc rc Cc: arch@freebsd.org In-reply-to: Your message of "Mon, 23 Oct 2000 16:23:38 PDT." <10649.972343418@winston.osd.bsdi.com> References: <10649.972343418@winston.osd.bsdi.com> Date: Mon, 23 Oct 2000 17:26:45 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG [[ Moved to arch since this isn't about a commit ]] In message <10649.972343418@winston.osd.bsdi.com> Jordan Hubbard writes: : > As I said at BSDcon, I'd love to see the entropy written by the kernel : > on shutdown to the end of swap space on shutdown and read in again : > when the system comes back. This has the advantage of always working : : What happens in cases where you have no swap space or it's a file? :) Then you have to arrange a fallback case. The driver would know this (the same way that the kernel knows about crash dumps) and not write things out. You'd have to have a fallback mechanism for these cases. These cases are edge for the most case, but an important edge case. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 16:57:15 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 0B06137B479 for ; Mon, 23 Oct 2000 16:57:12 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9NNvAn83952; Mon, 23 Oct 2000 17:57:11 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id RAA11628; Mon, 23 Oct 2000 17:57:10 -0600 (MDT) Message-Id: <200010232357.RAA11628@harmony.village.org> To: "John W. De Boskey" Subject: Re: cvs commit: src/etc rc Cc: arch@freebsd.org In-reply-to: Your message of "Mon, 23 Oct 2000 19:53:35 EDT." <20001023195335.A45719@bsdwins.com> References: <20001023195335.A45719@bsdwins.com> <10649.972343418@winston.osd.bsdi.com> Date: Mon, 23 Oct 2000 17:57:10 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG [[ moved to arch ]] In message <20001023195335.A45719@bsdwins.com> "John W. De Boskey" writes: : However, the idea(s) in general are good. To take his ideas : in a slightly different direction, What about reading data from : the kernel from random locations? Bootstrapping problems. You can't pick good random numbers until you have a good random state. Since many many people run GENERIC, you can't expect things to be too random. Before I did the mergemaster upgrade, I would always get "A cons is an object with feelings" fortune every single first login after reboot.... The diskless workstations are an interesting problem because the current scheme of writing to / exposes the random state to the entire network.... Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 17:43:40 2000 Delivered-To: freebsd-arch@freebsd.org Received: from bsdone.bsdwins.com (www.bsdwins.com [192.58.184.33]) by hub.freebsd.org (Postfix) with ESMTP id B5E7537B479 for ; Mon, 23 Oct 2000 17:43:38 -0700 (PDT) Received: (from jwd@localhost) by bsdone.bsdwins.com (8.11.0/8.11.0) id e9O0hac46178; Mon, 23 Oct 2000 20:43:36 -0400 (EDT) (envelope-from jwd) Date: Mon, 23 Oct 2000 20:43:36 -0400 From: "John W. De Boskey" To: Warner Losh Cc: arch@FreeBSD.org Subject: Re: cvs commit: src/etc rc Message-ID: <20001023204336.A46026@bsdwins.com> References: <20001023195335.A45719@bsdwins.com> <10649.972343418@winston.osd.bsdi.com> <20001023195335.A45719@bsdwins.com> <200010232357.RAA11628@harmony.village.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <200010232357.RAA11628@harmony.village.org>; from imp@village.org on Mon, Oct 23, 2000 at 05:57:10PM -0600 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG If you want the entropy to be saved into swap, it seems like we should add this type of functionality to rndcontrol. rndcontrol -w /dev/da0s1b # write entropy to end of partition rndcontrol -r /dev/da0s1b # attempt to read entropy from partition I also think we need to clean up the line(s) between the requirments for crypto strength randomness and normal randomness. * mkstemps() & friends don't really need crypto strength (or am I missing something. * mfs requires randomness for the FSIRAND code. If crypto strength is not available, will regular randomness do? Comments welcome. -John ----- Warner Losh's Original Message ----- > [[ moved to arch ]] > > In message <20001023195335.A45719@bsdwins.com> "John W. De Boskey" writes: > : However, the idea(s) in general are good. To take his ideas > : in a slightly different direction, What about reading data from > : the kernel from random locations? > > Bootstrapping problems. You can't pick good random numbers until you > have a good random state. Since many many people run GENERIC, you > can't expect things to be too random. Before I did the mergemaster > upgrade, I would always get "A cons is an object with feelings" > fortune every single first login after reboot.... > > The diskless workstations are an interesting problem because the > current scheme of writing to / exposes the random state to the entire > network.... > > Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 18:59:50 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 521C737B479; Mon, 23 Oct 2000 18:59:48 -0700 (PDT) Received: from billy-club.village.org (billy-club.village.org [10.0.0.3]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9O1xln84265; Mon, 23 Oct 2000 19:59:47 -0600 (MDT) (envelope-from imp@billy-club.village.org) Received: from billy-club.village.org (localhost [127.0.0.1]) by billy-club.village.org (8.11.1/8.8.3) with ESMTP id e9O1wUV04437; Mon, 23 Oct 2000 19:58:30 -0600 (MDT) Message-Id: <200010240158.e9O1wUV04437@billy-club.village.org> To: "John W. De Boskey" Subject: Re: cvs commit: src/etc rc Cc: arch@FreeBSD.org In-reply-to: Your message of "Mon, 23 Oct 2000 20:43:36 EDT." <20001023204336.A46026@bsdwins.com> References: <20001023204336.A46026@bsdwins.com> <20001023195335.A45719@bsdwins.com> <10649.972343418@winston.osd.bsdi.com> <20001023195335.A45719@bsdwins.com> <200010232357.RAA11628@harmony.village.org> Date: Mon, 23 Oct 2000 19:58:30 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <20001023204336.A46026@bsdwins.com> "John W. De Boskey" writes: : * mkstemps() & friends don't really need crypto strength (or am : I missing something. They do need it most of the time to make it harder to race. On some systems, this isn't an issue. : * mfs requires randomness for the FSIRAND code. If crypto strength : is not available, will regular randomness do? Crypto random is preferred here as well. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 19:11: 5 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 7936037B479; Mon, 23 Oct 2000 19:11:01 -0700 (PDT) Received: from billy-club.village.org (billy-club.village.org [10.0.0.3]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9O2B0n84301; Mon, 23 Oct 2000 20:11:00 -0600 (MDT) (envelope-from imp@billy-club.village.org) Received: from billy-club.village.org (localhost [127.0.0.1]) by billy-club.village.org (8.11.1/8.8.3) with ESMTP id e9O29hV04505; Mon, 23 Oct 2000 20:09:43 -0600 (MDT) Message-Id: <200010240209.e9O29hV04505@billy-club.village.org> To: "John W. De Boskey" Subject: Re: cvs commit: src/etc rc Cc: arch@FreeBSD.org In-reply-to: Your message of "Mon, 23 Oct 2000 20:43:36 EDT." <20001023204336.A46026@bsdwins.com> References: <20001023204336.A46026@bsdwins.com> <20001023195335.A45719@bsdwins.com> <10649.972343418@winston.osd.bsdi.com> <20001023195335.A45719@bsdwins.com> <200010232357.RAA11628@harmony.village.org> Date: Mon, 23 Oct 2000 20:09:43 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <20001023204336.A46026@bsdwins.com> "John W. De Boskey" writes: : If you want the entropy to be saved into swap, it seems : like we should add this type of functionality to rndcontrol. : : rndcontrol -w /dev/da0s1b # write entropy to end of partition : rndcontrol -r /dev/da0s1b # attempt to read entropy from partition [[ sorry to reply twice ]] The problem with having a program do it is that our shutdown proceedures won't always run that program. So you'd have to run that early in the life of a system (but not too early) to make sure that when the system shut down, the drive could dump there. rndcontrol -r would be usable early in the boot process, so we wouldn't need to have the driver automatically read from swap. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 23:18: 7 2000 Delivered-To: freebsd-arch@freebsd.org Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (Postfix) with ESMTP id E3A9637B479 for ; Mon, 23 Oct 2000 23:18:05 -0700 (PDT) Received: (from daemon@localhost) by smtp02.primenet.com (8.9.3/8.9.3) id XAA16295; Mon, 23 Oct 2000 23:14:30 -0700 (MST) Received: from usr01.primenet.com(206.165.6.201) via SMTP by smtp02.primenet.com, id smtpdAAA.maqOF; Mon Oct 23 23:14:17 2000 Received: (from tlambert@localhost) by usr01.primenet.com (8.8.5/8.8.5) id XAA08358; Mon, 23 Oct 2000 23:17:45 -0700 (MST) From: Terry Lambert Message-Id: <200010240617.XAA08358@usr01.primenet.com> Subject: Re: Ideas concerning fsck To: des@ofug.org (Dag-Erling Smorgrav) Date: Tue, 24 Oct 2000 06:17:45 +0000 (GMT) Cc: mbendiks@eunet.no (Marius Bendiksen), arch@FreeBSD.ORG In-Reply-To: from "Dag-Erling Smorgrav" at Oct 23, 2000 02:19:31 PM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > > Doing anything at all before you know you can trust your root > > > partition (where fsck itself is stored) is not a very good idea. > > And using /etc/fstab, stored on the same partition, is a better idea? > > No. However, we can't get around the need for fstab, but we can get > around the need for some kind of fscktab which contains static > information which practically noone will ever need to change. Frankly, the file system type can be determined from the magic number, and the mount point determined from the "last mounted on" field, for most FS types. Which basically means that there's no need for a /etc/fstab at all, for anything other than CDROMs, or DOS FSs. Passes and "noauto" could be handled via tunefs and a one byte flags field on the FS itself. Just so you know where going down this path in its entirety is likely to lead you... Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 23:41:15 2000 Delivered-To: freebsd-arch@freebsd.org Received: from c014.sfo.cp.net (c014-h001.c014.sfo.cp.net [209.228.12.65]) by hub.freebsd.org (Postfix) with SMTP id EA20937B479 for ; Mon, 23 Oct 2000 23:41:12 -0700 (PDT) Received: (cpmta 18527 invoked from network); 23 Oct 2000 23:41:12 -0700 Received: from d8c81e5f.dsl.flashcom.net (HELO quadrajet.flashcom.com) (216.200.30.95) by smtp.flashcom.net (209.228.12.65) with SMTP; 23 Oct 2000 23:41:12 -0700 X-Sent: 24 Oct 2000 06:41:12 GMT Received: (from guy@localhost) by quadrajet.flashcom.com (8.9.3/8.9.3) id XAA05114; Mon, 23 Oct 2000 23:41:09 -0700 (PDT) (envelope-from gharris) Date: Mon, 23 Oct 2000 23:41:09 -0700 From: Guy Harris To: Warner Losh Cc: Dag-Erling Smorgrav , Cy Schubert - ITSD Open Systems Group , arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck Message-ID: <20001023234109.Q442@quadrajet.flashcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > You can keep things well balanced with pass numbers. At least in the 3.4 version, the only uses of "fs_passno" in the "sbin/fsck" directory are in "main.c", where "docheck()" concludes that if "fs_passno" is 0, the file system doesn't have to be checked, and "preen.c", where "checkfstab()", during a preen, checks file systems with "fs_passno" of 1 sequentially in the first pass and checks file systems with "fs_passno" of 2 in parallel on the second pass. Perhaps I'm missing something, but it looks as if the pass number is a tri-state value: 0 don't check it at all 1 check it during the first pass 2 or higher check it during the second pass I think it's been that way for years (pre-Net-2). To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Mon Oct 23 23:45:25 2000 Delivered-To: freebsd-arch@freebsd.org Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (Postfix) with ESMTP id 126A737B479; Mon, 23 Oct 2000 23:45:23 -0700 (PDT) Received: (from daemon@localhost) by smtp02.primenet.com (8.9.3/8.9.3) id XAA22926; Mon, 23 Oct 2000 23:41:49 -0700 (MST) Received: from usr01.primenet.com(206.165.6.201) via SMTP by smtp02.primenet.com, id smtpdAAA8Na4PS; Mon Oct 23 23:41:37 2000 Received: (from tlambert@localhost) by usr01.primenet.com (8.8.5/8.8.5) id XAA09188; Mon, 23 Oct 2000 23:45:07 -0700 (MST) From: Terry Lambert Message-Id: <200010240645.XAA09188@usr01.primenet.com> Subject: Re: cvs commit: src/etc rc To: jwd@FreeBSD.ORG (John W. De Boskey) Date: Tue, 24 Oct 2000 06:45:07 +0000 (GMT) Cc: imp@village.org (Warner Losh), arch@FreeBSD.ORG In-Reply-To: <20001023204336.A46026@bsdwins.com> from "John W. De Boskey" at Oct 23, 2000 08:43:36 PM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > If you want the entropy to be saved into swap, it seems > like we should add this type of functionality to rndcontrol. > > rndcontrol -w /dev/da0s1b # write entropy to end of partition > rndcontrol -r /dev/da0s1b # attempt to read entropy from partition Please don't trash swap. I still have dreams of a "Suspend" mode that could be resumed from installation media, and get FreeBSD up and running on a new system in 30 seconds or less. It could also be used as a checkpoint/recovery mechanism to get a hot spare server up and running in about the same amount of time, assuming you can burn an image of the server you want to replicate to a CDROM, beforehand. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 3:45:50 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id ED5AA37B479 for ; Tue, 24 Oct 2000 03:45:47 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id MAA02164; Tue, 24 Oct 2000 12:45:42 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id MAA44964; Tue, 24 Oct 2000 12:45:41 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Tue, 24 Oct 2000 12:45:35 +0200 (CEST) From: Marius Bendiksen To: Dag-Erling Smorgrav Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > > Doing anything at all before you know you can trust your root > > > partition (where fsck itself is stored) is not a very good idea. > > And using /etc/fstab, stored on the same partition, is a better idea? > No. However, we can't get around the need for fstab, but we can get > around the need for some kind of fscktab which contains static > information which practically noone will ever need to change. We can get around the need for fstab by using the bootfs idea that AIX and others use (Poul-Henning has suggested this also in connection with DEVFS) to get less magic in the boot sequence. Also, if noone ever needs to change it, then it's even less likely to change, and thus be in an inconsistant state, than /etc/fstab. Hence, it is no real point of failure. While being able to employ a dynamic fsck would be a good thing, I am still concerned about removing the static version. I'd suggest adding a pass -1, like Kirk said, or somesuch. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 3:48:10 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id 4A1A237B479 for ; Tue, 24 Oct 2000 03:48:08 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id MAA02817; Tue, 24 Oct 2000 12:48:07 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id MAA44987; Tue, 24 Oct 2000 12:48:07 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Tue, 24 Oct 2000 12:48:07 +0200 (CEST) From: Marius Bendiksen To: Wilko Bulte Cc: arch@freebsd.org Subject: Re: Ideas concerning fsck In-Reply-To: <20001023192759.A1077@freebie.demon.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > The current suggested use of the pass # is to set the root filesystem > > to pass 1, and everything else that you want checked with pass 2. Fsck > > has a heuristic to decide which filesystems are on the same drive and > > will run them serially (see fsck/preen.c); filesystems on different > > spindles will be run in parallel. Running the root separately and by > I could imagine that for multiple filesystems sharing one (hardware) RAID > volume one might want to be able to override this heuristic. The inherent > parallelism allowed by RAID would allow checking of multiple filesystems > on the (apparantly) same spindle without speed degradation/trashing. Adding a pass 3 and beyond, or 3[a-z], would probably solve that. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5: 2:29 2000 Delivered-To: freebsd-arch@freebsd.org Received: from penguin-ext.wise.edt.ericsson.se (penguin-ext.wise.edt.ericsson.se [194.237.142.110]) by hub.freebsd.org (Postfix) with ESMTP id 1F55337B479 for ; Tue, 24 Oct 2000 05:02:26 -0700 (PDT) Received: from ms.uab.ericsson.se (ms.uab.ericsson.se [134.138.201.16]) by penguin.wise.edt.ericsson.se (8.11.0/8.10.1/WIREfire-1.3) with ESMTP id e9OC2PZ13209; Tue, 24 Oct 2000 14:02:25 +0200 (MEST) Received: from uab.ericsson.se (ulinpc62 [134.138.94.22]) by ms.uab.ericsson.se (8.10.0/8.10.0/uab-2.26) with ESMTP id e9OC2On23948; Tue, 24 Oct 2000 14:02:25 +0200 (MET DST) Message-ID: <39F57A50.5CA40232@uab.ericsson.se> Date: Tue, 24 Oct 2000 14:02:24 +0200 From: Anders Franzen X-Mailer: Mozilla 4.72 [en] (X11; I; FreeBSD 4.1-RELEASE i386) X-Accept-Language: en MIME-Version: 1.0 To: Warner Losh Cc: freebsd-arch@FreeBSD.ORG Subject: Re: loadable drivers References: <39ED9B42.417E88FD@uab.ericsson.se> <200010232309.RAA11156@harmony.village.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Warner Losh wrote: > In message <39ED9B42.417E88FD@uab.ericsson.se> Anders Franzen writes: > : Hi, I'm playing around with the linux pcmcia/cardbus drivers. > > What does this have to do with FreeBSD? > Not much except that I have successfully!! ported it to FreeBSD!!!!!. /Fra > > > : Since I crash quite often I would like to do it with modules. > : The problem is that if I don't put any driver for the pcicl chip > : the PCI bus driver hands it over to the chip 'device/bus' and I > : wont see it when I attach the module to the PCI bus. > : > : Is there anyway of doing this or must certain drivers (bridges) be > : linked > : in the kernel? > > Sure. The cardbus drivers haven't been tested as loadable modules, > but pccard and pcic work great as far as I can tell. > > Warner My question was: how can I take control over the 82365 from a loadable driver. Since if the driver is not part of the kernel, the pci bus configuration hands the chip over to a 'pseudo driver' chip. And when I load my driver I wont see the 82365 (TI 1225 in this case). To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5:45:24 2000 Delivered-To: freebsd-arch@freebsd.org Received: from erouter0.it-datacntr.louisville.edu (erouter0.it-datacntr.louisville.edu [136.165.1.36]) by hub.freebsd.org (Postfix) with ESMTP id 5F12037B479 for ; Tue, 24 Oct 2000 05:45:22 -0700 (PDT) Received: from osaka.louisville.edu (osaka.louisville.edu [136.165.1.114]) by erouter0.it-datacntr.louisville.edu (Postfix) with ESMTP id 6F618251A2; Tue, 24 Oct 2000 08:45:13 -0400 (EDT) Received: by osaka.louisville.edu (Postfix, from userid 15) id 12E511861D; Tue, 24 Oct 2000 08:45:10 -0400 (EDT) Date: Tue, 24 Oct 2000 08:45:09 -0400 From: Keith Stevenson To: Marius Bendiksen Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck Message-ID: <20001024084509.B10509@osaka.louisville.edu> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from mbendiks@eunet.no on Tue, Oct 24, 2000 at 12:45:35PM +0200 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Oct 24, 2000 at 12:45:35PM +0200, Marius Bendiksen wrote: > > We can get around the need for fstab by using the bootfs idea that AIX and > others use (Poul-Henning has suggested this also in connection with DEVFS) > to get less magic in the boot sequence. AIX does use fstab. IBM just cleverly renamed it "/etc/filesystems" and completely changed the format. This file is consulted as part of the boot process. The bootfs idea is pretty neat. I would like to see it implemented in FreeBSD, especially if it would make vinum disks usable for the / filesystem. Regards, --Keith Stevenson-- -- Keith Stevenson System Programmer - Data Center Services - University of Louisville keith.stevenson@louisville.edu GPG key fingerprint = 332D 97F0 6321 F00F 8EE7 2D44 00D8 F384 75BB 89AE To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5:57: 6 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id A3A6F37B479; Tue, 24 Oct 2000 05:57:04 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9OCv3n86339; Tue, 24 Oct 2000 06:57:03 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id GAA15087; Tue, 24 Oct 2000 06:57:03 -0600 (MDT) Message-Id: <200010241257.GAA15087@harmony.village.org> To: Terry Lambert Subject: Re: cvs commit: src/etc rc Cc: jwd@FreeBSD.ORG (John W. De Boskey), arch@FreeBSD.ORG In-reply-to: Your message of "Tue, 24 Oct 2000 06:45:07 -0000." <200010240645.XAA09188@usr01.primenet.com> References: <200010240645.XAA09188@usr01.primenet.com> Date: Tue, 24 Oct 2000 06:57:03 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <200010240645.XAA09188@usr01.primenet.com> Terry Lambert writes: : Please don't trash swap. I still have dreams of a "Suspend" : mode that could be resumed from installation media, and get : FreeBSD up and running on a new system in 30 seconds or less. Swap isn't trashed. Swap is written to as the last part of a complete shutdown. A hypothetical suspend shutdown wouldn't do this. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5:57:26 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id DDF1E37B4C5 for ; Tue, 24 Oct 2000 05:57:19 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id OAA33021; Tue, 24 Oct 2000 14:57:16 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id OAA45881; Tue, 24 Oct 2000 14:57:16 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Tue, 24 Oct 2000 14:57:16 +0200 (CEST) From: Marius Bendiksen To: Keith Stevenson Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: <20001024084509.B10509@osaka.louisville.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > We can get around the need for fstab by using the bootfs idea that AIX and > > others use (Poul-Henning has suggested this also in connection with DEVFS) > > to get less magic in the boot sequence. > AIX does use fstab. IBM just cleverly renamed it "/etc/filesystems" and > completely changed the format. This file is consulted as part of the boot > process. I made no claim wrt that. The reference to "less magic in the boot sequence" was related to discovering the boot device and such. And it would be possible for us to remove the need to rely on the root filesystem being uncorrupt. > The bootfs idea is pretty neat. I would like to see it implemented in > FreeBSD, especially if it would make vinum disks usable for the / filesystem. I use this whenever I do vinum based solutions, by having a minimal root filesystem as a memory filesystem, then starting vinum to be able to bring up the real root filesystem. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5:58:13 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id A38CE37B479 for ; Tue, 24 Oct 2000 05:58:11 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9OCw9n86355; Tue, 24 Oct 2000 06:58:09 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id GAA15121; Tue, 24 Oct 2000 06:58:09 -0600 (MDT) Message-Id: <200010241258.GAA15121@harmony.village.org> To: Anders Franzen Subject: Re: loadable drivers Cc: freebsd-arch@FreeBSD.ORG In-reply-to: Your message of "Tue, 24 Oct 2000 14:02:24 +0200." <39F57A50.5CA40232@uab.ericsson.se> References: <39F57A50.5CA40232@uab.ericsson.se> <39ED9B42.417E88FD@uab.ericsson.se> <200010232309.RAA11156@harmony.village.org> Date: Tue, 24 Oct 2000 06:58:08 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <39F57A50.5CA40232@uab.ericsson.se> Anders Franzen writes: : My question was: how can I take control over the 82365 from a loadable : driver. Since if the driver is not part of the kernel, the pci bus : configuration : hands the chip over to a 'pseudo driver' chip. And when I load my driver : I wont see the 82365 (TI 1225 in this case). No, it just works. When the cardbus bridge driver is loaded it will reprobe the device and it will just work. I just haven't written the makefile yet to make this possible with pccbb. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 5:59: 5 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 3371C37B479 for ; Tue, 24 Oct 2000 05:59:03 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9OCx2n86363; Tue, 24 Oct 2000 06:59:02 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id GAA15141; Tue, 24 Oct 2000 06:59:02 -0600 (MDT) Message-Id: <200010241259.GAA15141@harmony.village.org> To: Anders Franzen Subject: Re: loadable drivers Cc: freebsd-arch@FreeBSD.ORG In-reply-to: Your message of "Tue, 24 Oct 2000 14:02:24 +0200." <39F57A50.5CA40232@uab.ericsson.se> References: <39F57A50.5CA40232@uab.ericsson.se> <39ED9B42.417E88FD@uab.ericsson.se> <200010232309.RAA11156@harmony.village.org> Date: Tue, 24 Oct 2000 06:59:01 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <39F57A50.5CA40232@uab.ericsson.se> Anders Franzen writes: : Not much except that I have successfully!! ported it to FreeBSD!!!!!. Cool, but now that we have it supported natively in a non-GPL'd source base, the chances are good that it won't make it into the kernel. I hate to be the bearer of bad news. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 6: 1:59 2000 Delivered-To: freebsd-arch@freebsd.org Received: from erouter0.it-datacntr.louisville.edu (erouter0.it-datacntr.louisville.edu [136.165.1.36]) by hub.freebsd.org (Postfix) with ESMTP id 2AD0D37B479 for ; Tue, 24 Oct 2000 06:01:58 -0700 (PDT) Received: from osaka.louisville.edu (osaka.louisville.edu [136.165.1.114]) by erouter0.it-datacntr.louisville.edu (Postfix) with ESMTP id CDE2F24F78; Tue, 24 Oct 2000 09:01:57 -0400 (EDT) Received: by osaka.louisville.edu (Postfix, from userid 15) id AE6D21861D; Tue, 24 Oct 2000 09:01:55 -0400 (EDT) Date: Tue, 24 Oct 2000 09:01:53 -0400 From: Keith Stevenson To: Marius Bendiksen Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck Message-ID: <20001024090152.A10742@osaka.louisville.edu> References: <20001024084509.B10509@osaka.louisville.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from mbendiks@eunet.no on Tue, Oct 24, 2000 at 02:57:16PM +0200 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Oct 24, 2000 at 02:57:16PM +0200, Marius Bendiksen wrote: > > I made no claim wrt that. The reference to "less magic in the boot > sequence" was related to discovering the boot device and such. And > it would be possible for us to remove the need to rely on the root > filesystem being uncorrupt. Ok. I must have misunderstood. > > > The bootfs idea is pretty neat. I would like to see it implemented in > > FreeBSD, especially if it would make vinum disks usable for the / filesystem. > > I use this whenever I do vinum based solutions, by having a minimal root > filesystem as a memory filesystem, then starting vinum to be able to bring > up the real root filesystem. Is there documentation somewhere on how to do this? I've looked at the memory based filesystem stuff before, but was unable to figure out what steps needed to be performed in order to boot from it. Regards, --Keith Stevenson-- -- Keith Stevenson System Programmer - Data Center Services - University of Louisville keith.stevenson@louisville.edu GPG key fingerprint = 332D 97F0 6321 F00F 8EE7 2D44 00D8 F384 75BB 89AE To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7: 8:19 2000 Delivered-To: freebsd-arch@freebsd.org Received: from flood.ping.uio.no (flood.ping.uio.no [129.240.78.31]) by hub.freebsd.org (Postfix) with ESMTP id 69D9437B479 for ; Tue, 24 Oct 2000 07:08:17 -0700 (PDT) Received: (from des@localhost) by flood.ping.uio.no (8.9.3/8.9.3) id QAA44963; Tue, 24 Oct 2000 16:08:15 +0200 (CEST) (envelope-from des@ofug.org) X-URL: http://www.ofug.org/~des/ X-Disclaimer: The views expressed in this message do not necessarily coincide with those of any organisation or company with which I am or have been affiliated. To: Marius Bendiksen Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck References: From: Dag-Erling Smorgrav Date: 24 Oct 2000 16:08:15 +0200 In-Reply-To: Marius Bendiksen's message of "Tue, 24 Oct 2000 12:45:35 +0200 (CEST)" Message-ID: Lines: 11 User-Agent: Gnus/5.0802 (Gnus v5.8.2) Emacs/20.4 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Marius Bendiksen writes: > We can get around the need for fstab by using the bootfs idea that AIX and > others use (Poul-Henning has suggested this also in connection with DEVFS) > to get less magic in the boot sequence. I'm not familiar with the concept of bootfs - can somebody enlighten me? DES -- Dag-Erling Smorgrav - des@ofug.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7:17:32 2000 Delivered-To: freebsd-arch@freebsd.org Received: from critter.freebsd.dk (flutter.freebsd.dk [212.242.40.147]) by hub.freebsd.org (Postfix) with ESMTP id 75A8B37B4D7 for ; Tue, 24 Oct 2000 07:17:29 -0700 (PDT) Received: from critter (localhost [127.0.0.1]) by critter.freebsd.dk (8.11.1/8.9.3) with ESMTP id e9OEJMU19692; Tue, 24 Oct 2000 16:19:22 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Dag-Erling Smorgrav Cc: Marius Bendiksen , arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: Your message of "24 Oct 2000 16:08:15 +0200." Date: Tue, 24 Oct 2000 16:19:22 +0200 Message-ID: <19690.972397162@critter> From: Poul-Henning Kamp Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >I'm not familiar with the concept of bootfs - can somebody enlighten >me? The idea is to use a small RAM fs to get airborne after boot. AIX boots with what is in essence a fixit filesystem on a ramdisk, once it has located the real root filesystem and fsck'ed etc etc it has a magic syscall which means "throw away the rootfilesystem, mount this filesystem as root, (re)exec /sbin/init" The idea I have been toying with is to mount DEVFS as / and mount the "real" root on top of it with a unionmount. Amongst the things this would allow is a kind of "persistence" for /dev (should anybody still want that) and various trickery like mirror/raid initializations could be performed from userland before mounting the "real" root. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7:35:24 2000 Delivered-To: freebsd-arch@freebsd.org Received: from smtp01.primenet.com (smtp01.primenet.com [206.165.6.131]) by hub.freebsd.org (Postfix) with ESMTP id 64AB437B479 for ; Tue, 24 Oct 2000 07:35:22 -0700 (PDT) Received: (from daemon@localhost) by smtp01.primenet.com (8.9.3/8.9.3) id HAA19152; Tue, 24 Oct 2000 07:34:29 -0700 (MST) Received: from usr05.primenet.com(206.165.6.205) via SMTP by smtp01.primenet.com, id smtpdAAAKWaqyL; Tue Oct 24 07:34:22 2000 Received: (from tlambert@localhost) by usr05.primenet.com (8.8.5/8.8.5) id HAA28287; Tue, 24 Oct 2000 07:35:13 -0700 (MST) From: Terry Lambert Message-Id: <200010241435.HAA28287@usr05.primenet.com> Subject: Re: Ideas concerning fsck To: phk@critter.freebsd.dk (Poul-Henning Kamp) Date: Tue, 24 Oct 2000 14:34:56 +0000 (GMT) Cc: des@ofug.org (Dag-Erling Smorgrav), mbendiks@eunet.no (Marius Bendiksen), arch@FreeBSD.ORG In-Reply-To: <19690.972397162@critter> from "Poul-Henning Kamp" at Oct 24, 2000 04:19:22 PM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > The idea I have been toying with is to mount DEVFS as / and mount > the "real" root on top of it with a unionmount. > > Amongst the things this would allow is a kind of "persistence" for > /dev (should anybody still want that) and various trickery like > mirror/raid initializations could be performed from userland > before mounting the "real" root. Yes, this is absolutely the right way to do this. Another simplification is to mount into the mounted FS list, and then map into the namespace, as a seperate, second step. This would permit all of the mount point covering code to be moved to higher level common code in the mount system call. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7:39:21 2000 Delivered-To: freebsd-arch@freebsd.org Received: from shidahara1.planet.sci.kobe-u.ac.jp (shidahara1.planet.sci.kobe-u.ac.jp [133.30.50.200]) by hub.freebsd.org (Postfix) with ESMTP id C3D9937B4C5; Tue, 24 Oct 2000 07:39:16 -0700 (PDT) Received: from shidahara1.planet.sci.kobe-u.ac.jp (localhost [127.0.0.1]) by shidahara1.planet.sci.kobe-u.ac.jp (8.9.3/8.9.3) with ESMTP id XAA58776; Tue, 24 Oct 2000 23:38:12 +0900 (JST) (envelope-from takawata@shidahara1.planet.sci.kobe-u.ac.jp) Message-Id: <200010241438.XAA58776@shidahara1.planet.sci.kobe-u.ac.jp> To: imp@freebsd.org, arch@freebsd.org Subject: NEWCARD/OLDCARD compatibility layer Date: Tue, 24 Oct 2000 23:38:12 +0900 From: Takanori Watanabe Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi, I looked card_if.m and I think there is a problem:If there is a system that have neither OLDCARD or NEWCARD layer, module-ifyed driver that uses NEWCARD/OLDCARD compatibility layer mechanism cannot load because pccard_compat_probe/pccard_compat_attach is not exist. How about this change? And ,in child driver side, PCCARD_COMPAT_LAYER should be defined to use pccard_compat_* function before including "card_if.h". With this change kernel can be compiled.(I don't know wherther it collectly work or not.) Takanori Watanabe Public Key Key fingerprint = 2C 51 E2 78 2C E1 C5 2D 0F F1 20 A3 11 3A 62 2A --- card_if.m.org Wed Sep 20 14:24:23 2000 +++ card_if.m Tue Oct 24 23:21:01 2000 @@ -143,6 +143,34 @@ # # Drivers wishing to not retain OLDCARD compatibility needn't do this. # + +#These Function is defined in bus code + +METHOD int compatsubr_probe{ + device_t parent; + device_t dev; +}; + +METHOD int compatsubr_attach{ + device_t parent; + device_t dev; +}; + +HEADER{ + #ifdef PCCARD_COMPAT_LAYER + static int pccard_compat_probe(device_t); + static int pccard_compat_attach(device_t); + static int pccard_compat_probe(device_t dev) + { + return CARD_COMPATSUBR_PROBE(device_get_parent(dev), dev); + } + static int pccard_compat_attach(device_t dev) + { + return CARD_COMPATSUBR_ATTACH(device_get_parent(dev), dev); + } + #endif +}; + METHOD int compat_probe { device_t dev; } --- pccard.c~ Mon Oct 9 22:08:33 2000 +++ pccard.c Tue Oct 24 22:40:54 2000 @@ -563,14 +563,14 @@ * needs to grab devices while in the old they were assigned to the device by * the pccardd process. These symbols are exported to the upper layers. */ -int -pccard_compat_probe(device_t dev) +static int +pccard_compat_probe(device_t parent, device_t dev) { return (CARD_COMPAT_MATCH(dev)); } -int -pccard_compat_attach(device_t dev) +static int +pccard_compat_attach(device_t parent, device_t dev) { int err; @@ -811,7 +811,8 @@ DEVMETHOD(card_get_type, pccard_card_gettype), DEVMETHOD(card_attach_card, pccard_attach_card), DEVMETHOD(card_detach_card, pccard_detach_card), - + DEVMETHOD(card_compatsubr_probe, pccard_compat_probe), + DEVMETHOD(card_compatsubr_attach, pccard_compat_attach), { 0, 0 } }; --- pccardvar.h~ Mon Oct 9 22:08:33 2000 +++ pccardvar.h Tue Oct 24 22:59:07 2000 @@ -287,9 +287,12 @@ #define pccard_mem_unmap(pf, window) \ (pccard_chip_mem_unmap((pf)->sc->pct, (pf)->sc->pch, (window))) -/* compat layer */ -int pccard_compat_probe(device_t dev); -int pccard_compat_attach(device_t dev); +/* + * compat layer are defined in card_if.h statically.(A bit ugly.) + * + * int pccard_compat_probe(device_t dev); + * int pccard_compat_attach(device_t dev); + */ /* ivar interface */ enum { To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7:40:34 2000 Delivered-To: freebsd-arch@freebsd.org Received: from point.osg.gov.bc.ca (point.osg.gov.bc.ca [142.32.102.44]) by hub.freebsd.org (Postfix) with ESMTP id 0622A37B479 for ; Tue, 24 Oct 2000 07:40:31 -0700 (PDT) Received: (from daemon@localhost) by point.osg.gov.bc.ca (8.8.7/8.8.8) id HAA14046; Tue, 24 Oct 2000 07:40:05 -0700 Received: from passer.osg.gov.bc.ca(142.32.110.29) via SMTP by point.osg.gov.bc.ca, id smtpda14040; Tue Oct 24 07:39:51 2000 Received: (from uucp@localhost) by passer.osg.gov.bc.ca (8.11.0/8.9.1) id e9OEdo712272; Tue, 24 Oct 2000 07:39:50 -0700 (PDT) Received: from cwsys9.cwsent.com(10.2.2.1), claiming to be "cwsys.cwsent.com" via SMTP by passer9.cwsent.com, id smtpdd12269; Tue Oct 24 07:39:38 2000 Received: (from uucp@localhost) by cwsys.cwsent.com (8.11.1/8.9.1) id e9OEdbl30832; Tue, 24 Oct 2000 07:39:37 -0700 (PDT) Message-Id: <200010241439.e9OEdbl30832@cwsys.cwsent.com> Received: from localhost.cwsent.com(127.0.0.1), claiming to be "cwsys" via SMTP by localhost.cwsent.com, id smtpdE30827; Tue Oct 24 07:39:24 2000 X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4 Reply-To: Cy Schubert - ITSD Open Systems Group From: Cy Schubert - ITSD Open Systems Group X-OS: FreeBSD 4.1.1-RELEASE X-Sender: cy To: Warner Losh Cc: Nick Sayer , arch@FreeBSD.ORG Subject: Re: Old 4.0-R binaries remaining in 4.1.1-S In-reply-to: Your message of "Mon, 23 Oct 2000 16:41:48 MDT." <200010232241.QAA10916@harmony.village.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 24 Oct 2000 07:39:23 -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Regards, Phone: (250)387-8437 Cy Schubert Fax: (250)387-5766 Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca Open Systems Group, ITSD, ISTA Province of BC In message <200010232241.QAA10916@harmony.village.org>, Warner Losh writes: > In message <39F4B9EA.FACF4464@sftw.com> Nick Sayer writes: > : I suggest that UPDATING tell folks do an ls -alt on /bin, /sbin, > : /usr/bin, /usr/sbin, /usr/libexec, /usr/lib and /modules and look for > : old stuff. In the case of /usr/lib, *.so.*.* should be moved to > : /usr/lib/compat/aout and *.so.[0-9] should be moved to /usr/lib/compat. > : All others can be deleted. Searching and deprecating old stuff in > : /usr/share and /usr/include is... undefined. :-) > > I'd prefer that we have a mtree.obsolete that can be used to find and > delete the files. NetBSD has this concept (I haven't looked at it, > since they do it as part of the binary upgrade), which might make > sense for us too. > > How do people feel about someone going off and doing this? I like the idea, especially if mtree would have an interactive option. Blindly deleting files, especially if the to-be-deleted file is a local modification or a symlink to another file. An interactive option would allow those of us who wish to err on the side of caution. Regards, Phone: (250)387-8437 Cy Schubert Fax: (250)387-5766 Team Leader, Sun/DEC Team Internet: Cy.Schubert@osg.gov.bc.ca Open Systems Group, ITSD, ISTA Province of BC To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 7:43: 0 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail.interware.hu (mail.interware.hu [195.70.32.130]) by hub.freebsd.org (Postfix) with ESMTP id 6DC0037B479 for ; Tue, 24 Oct 2000 07:42:57 -0700 (PDT) Received: from kairo-01.budapest.interware.hu ([195.70.50.65] helo=elischer.org) by mail.interware.hu with esmtp (Exim 3.16 #1 (Debian)) id 13o5Ht-00047C-00; Tue, 24 Oct 2000 16:42:49 +0200 Message-ID: <39F59FE4.1FF84086@elischer.org> Date: Tue, 24 Oct 2000 07:42:44 -0700 From: Julian Elischer X-Mailer: Mozilla 4.7 [en] (X11; U; FreeBSD 5.0-CURRENT i386) X-Accept-Language: en MIME-Version: 1.0 To: Warner Losh Cc: Anders Franzen , freebsd-arch@FreeBSD.ORG Subject: Re: loadable drivers References: <39F57A50.5CA40232@uab.ericsson.se> <39ED9B42.417E88FD@uab.ericsson.se> <200010232309.RAA11156@harmony.village.org> <200010241259.GAA15141@harmony.village.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Warner Losh wrote: > > In message <39F57A50.5CA40232@uab.ericsson.se> Anders Franzen writes: > : Not much except that I have successfully!! ported it to FreeBSD!!!!!. > > Cool, but now that we have it supported natively in a non-GPL'd source > base, the chances are good that it won't make it into the kernel. I > hate to be the bearer of bad news. But I didn't read it like this.. I may have been dreaming but I read it as though he has a linux cardbus framework ported as a BSD application allowing the use of random Linux drivers (for which there may be no source). And that the native cardbus framwork is getting in the way.. of course I may have to cut back on the halucinogens! > > Warner > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-arch" in the body of the message -- __--_|\ Julian Elischer / \ julian@elischer.org ( OZ ) World tour 2000 ---> X_.---._/ presently in: Budapest v To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 9:26:31 2000 Delivered-To: freebsd-arch@freebsd.org Received: from warning.follo.net (warning.follo.net [195.204.136.30]) by hub.freebsd.org (Postfix) with ESMTP id 383BC37B65E; Tue, 24 Oct 2000 09:26:26 -0700 (PDT) Received: (from eivind@localhost) by warning.follo.net (8.9.3/8.9.3) id SAA38975; Tue, 24 Oct 2000 18:23:20 +0200 (CEST) Date: Tue, 24 Oct 2000 18:23:19 +0200 From: Eivind Eklund To: Kris Kennaway Cc: Warner Losh , Mark Murray , arch@FreeBSD.org Subject: Re: cvs commit: src/etc rc Message-ID: <20001024182319.C37278@warning.follo.net> References: <20001024124057.A4309@skriver.dk> <200010232046.e9NKkLR01463@grimreaper.grondar.za> <20001023081548.A41843@bsdwins.com> <200010232046.e9NKkLR01463@grimreaper.grondar.za> <200010232321.RAA11268@harmony.village.org> <20001024124057.A4309@skriver.dk> <200010241256.GAA15067@harmony.village.org> <20001024162445.A58246@warning.follo.net> <20001024083111.D58506@citusc17.usc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: <20001024083111.D58506@citusc17.usc.edu>; from kris@citusc.usc.edu on Tue, Oct 24, 2000 at 08:31:11AM -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG (Moving to -arch) On Tue, Oct 24, 2000 at 08:31:11AM -0700, Kris Kennaway wrote: > On Tue, Oct 24, 2000 at 04:24:45PM +0200, Eivind Eklund wrote: > > > Can't we just crypt the data with a strong cipher (or, preferably, two or > > three strong ciphers) and a key aquired by using random data from Yarrow > > before writing it out? That would not expose state, assuming we trust the > > How would you know the key when you want to read it back in after the > next reboot? You don't know the key. This is *entropy* - you don't need to recover it exactly :-) Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 10:32:10 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail.rpi.edu (mail.rpi.edu [128.113.100.7]) by hub.freebsd.org (Postfix) with ESMTP id CC25537B479 for ; Tue, 24 Oct 2000 10:32:07 -0700 (PDT) Received: from [128.113.24.47] (gilead.acs.rpi.edu [128.113.24.47]) by mail.rpi.edu (8.9.3/8.9.3) with ESMTP id NAA245948; Tue, 24 Oct 2000 13:31:51 -0400 Mime-Version: 1.0 X-Sender: drosih@mail.rpi.edu Message-Id: In-Reply-To: <200010232241.QAA10916@harmony.village.org> References: <39F4B9EA.FACF4464@sftw.com> <200010230328.XAA16342@world.std.com> <200010231857.MAA09601@harmony.village.org> <200010232241.QAA10916@harmony.village.org> Date: Tue, 24 Oct 2000 13:31:50 -0400 To: Warner Losh , Nick Sayer From: Garance A Drosihn Subject: Re: Old 4.0-R binaries remaining in 4.1.1-S Cc: arch@FreeBSD.ORG Content-Type: text/plain; charset="us-ascii" ; format="flowed" Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG At 4:41 PM -0600 10/23/00, Warner Losh wrote: >In message <39F4B9EA.FACF4464@sftw.com> Nick Sayer writes: >: I suggest that UPDATING tell folks do an ls -alt on /bin, /sbin, >: /usr/bin, /usr/sbin, /usr/libexec, /usr/lib and /modules and look for >: old stuff. In the case of /usr/lib, *.so.*.* should be moved to >: /usr/lib/compat/aout and *.so.[0-9] should be moved to /usr/lib/compat. >: All others can be deleted. Searching and deprecating old stuff in >: /usr/share and /usr/include is... undefined. :-) > >I'd prefer that we have a mtree.obsolete that can be used to find and >delete the files. NetBSD has this concept (I haven't looked at it, >since they do it as part of the binary upgrade), which might make >sense for us too. > >How do people feel about someone going off and doing this? Something to find the files would be nice. I'm a bit uneasy about it just deleting them. Perhaps move them into some "attic". That way, if it turns out the user still needs them for some reason, they can readily restore them. (and restore the EXACT SAME versions they had been using...) --- Garance Alistair Drosehn = gad@eclipse.acs.rpi.edu Senior Systems Programmer or drosih@rpi.edu Rensselaer Polytechnic Institute To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 11: 7:37 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 7F5D037B479 for ; Tue, 24 Oct 2000 11:07:31 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9OI7Tn87688; Tue, 24 Oct 2000 12:07:30 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id MAA17271; Tue, 24 Oct 2000 12:07:29 -0600 (MDT) Message-Id: <200010241807.MAA17271@harmony.village.org> To: Marcel Moolenaar Subject: Re: PLEASE REVIEW: buildworld fix for 3.x->RELENG_4 upgrades Cc: arch@freebsd.org In-reply-to: Your message of "Tue, 24 Oct 2000 13:57:59 EDT." <39F5CDA7.67288AD@cup.hp.com> References: <39F5CDA7.67288AD@cup.hp.com> <39F532B2.BDE29048@cup.hp.com> <39F5185C.62A92866@cup.hp.com> <39F4ADC9.920A9BB5@sftw.com> <39F39B57.8726A8BA@quack.kfu.com> <39F1A95B.86116571@quack.kfu.com> <20001022094114.A14613@dragon.nuxi.com> <200010230405.WAA04356@harmony.village.org> <200010232354.RAA11591@harmony.village.org> <200010240540.XAA13044@harmony.village.org> <200010241254.GAA15044@harmony.village.org> Date: Tue, 24 Oct 2000 12:07:29 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG [[ Moved to arch@ since this isn't a developers only issue ]] : So, either we need to install the boot stuff when we install the kernel : or put the device hints back into GENERIC (and removing the logic as : illustrated by the patch). I see the problem now. It is one of ordering. We could easily install a default device.hints and a corrected boot.conf file as part of the upgrade process, but we make it needlessly complex due to this check. This makes it harder to do the upgrade path. I see a couple of ways around this. One is to add a check to see if you are running 4.x or older and be helpful by doing an implicit copy of GENERIC.hints to /boot/device.hints as well as doing an install of the new boot loader in /boot (since really that what the checks you are proposing removing effectively check for). That is to say these things would be implied if there was no /boot/device.hint file pre-existing (we don't want to clobber it) (and that might be a better check than an OS version, come to think of it). Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 12:28:14 2000 Delivered-To: freebsd-arch@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 87A2737B479 for ; Tue, 24 Oct 2000 12:28:10 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.0/8.11.0) with ESMTP id e9OJS5n88171; Tue, 24 Oct 2000 13:28:05 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id NAA18224; Tue, 24 Oct 2000 13:28:04 -0600 (MDT) Message-Id: <200010241928.NAA18224@harmony.village.org> To: Takanori Watanabe Subject: Re: NEWCARD/OLDCARD compatibility layer Cc: arch@freebsd.org In-reply-to: Your message of "Tue, 24 Oct 2000 23:38:12 +0900." <200010241438.XAA58776@shidahara1.planet.sci.kobe-u.ac.jp> References: <200010241438.XAA58776@shidahara1.planet.sci.kobe-u.ac.jp> Date: Tue, 24 Oct 2000 13:28:04 -0600 From: Warner Losh Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <200010241438.XAA58776@shidahara1.planet.sci.kobe-u.ac.jp> Takanori Watanabe writes: : I looked card_if.m and I think there is a problem:If there is a system : that have neither OLDCARD or NEWCARD layer, module-ifyed driver that : uses NEWCARD/OLDCARD compatibility layer mechanism cannot load because : pccard_compat_probe/pccard_compat_attach is not exist. You are correct. I didn't think this would be a problem, but does impose ordering constraints to module loading which might be undesirable. However, I think that there are other issues with doing this module loading as well that are "silent" killers if done improperly. Reguardless of the other issues, I think this is a step in the right direction. : How about this change? : And ,in child driver side, PCCARD_COMPAT_LAYER should be defined : to use pccard_compat_* function before including "card_if.h". I see no harm in having them defnied all the time for now. I'll test this out when I get some time and report back. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 12:57:51 2000 Delivered-To: freebsd-arch@freebsd.org Received: from palrel1.hp.com (palrel1.hp.com [156.153.255.242]) by hub.freebsd.org (Postfix) with ESMTP id 8E6B637B479 for ; Tue, 24 Oct 2000 12:57:49 -0700 (PDT) Received: from adlmail.cup.hp.com (adlmail.cup.hp.com [15.0.100.30]) by palrel1.hp.com (Postfix) with ESMTP id 70CE8798; Tue, 24 Oct 2000 12:57:48 -0700 (PDT) Received: from cup.hp.com (gauss.cup.hp.com [15.28.97.152]) by adlmail.cup.hp.com (8.9.3 (PHNE_18546)/8.9.3 SMKit7.02) with ESMTP id MAA02324; Tue, 24 Oct 2000 12:57:48 -0700 (PDT) Message-ID: <39F5E9BC.2E7241D9@cup.hp.com> Date: Tue, 24 Oct 2000 15:57:48 -0400 From: Marcel Moolenaar Organization: Hewlett-Packard X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.2.12 i386) X-Accept-Language: en MIME-Version: 1.0 To: Warner Losh Cc: arch@freebsd.org Subject: Re: PLEASE REVIEW: buildworld fix for 3.x->RELENG_4 upgrades References: <39F5CDA7.67288AD@cup.hp.com> <39F532B2.BDE29048@cup.hp.com> <39F5185C.62A92866@cup.hp.com> <39F4ADC9.920A9BB5@sftw.com> <39F39B57.8726A8BA@quack.kfu.com> <39F1A95B.86116571@quack.kfu.com> <20001022094114.A14613@dragon.nuxi.com> <200010230405.WAA04356@harmony.village.org> <200010232354.RAA11591@harmony.village.org> <200010240540.XAA13044@harmony.village.org> <200010241254.GAA15044@harmony.village.org> <200010241807.MAA17271@harmony.village.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Warner Losh wrote: > > : So, either we need to install the boot stuff when we install the kernel > : or put the device hints back into GENERIC (and removing the logic as > : illustrated by the patch). > > I see the problem now. It is one of ordering. Yes. > We could easily > install a default device.hints and a corrected boot.conf file as part > of the upgrade process, but we make it needlessly complex due to this > check. This makes it harder to do the upgrade path. Our current upgrade path is not that fully fletched. Yes, ideally you would install the boot loaders before installing a new kernel before install world. It's all planned for the new and improved upgrade target. The problem in ordering is therefore twofold: 1. We need to install the bootloader before we install the kernel, 2. This issue would have been a non-issue if the change was made after we had new upgrade target :-) > I see a couple of ways around this. One is to add a check to see if > you are running 4.x or older and be helpful by doing an implicit copy > of GENERIC.hints to /boot/device.hints as well as doing an install of > the new boot loader in /boot (since really that what the checks you > are proposing removing effectively check for). To be 100% correct, I didn't propose to remove the checks. The patch was to demonstrate what was in the way of doing an upgrade from 4.x to -current. If device hints will not be compiled into the kernel, then we probably need to have these checks, but in a form that doesn't complain to the user, but resolves the problem automaticly as you say. > That is to say these > things would be implied if there was no /boot/device.hint file > pre-existing (we don't want to clobber it) (and that might be a better > check than an OS version, come to think of it). Correct. We can remove the checks in the current form and simply install the hints if not present, this automaticly implies we need to install a new bootloader configuration (does it?) -- Marcel Moolenaar mail: marcel@cup.hp.com / marcel@FreeBSD.org tel: (408) 447-4222 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Tue Oct 24 18:48:12 2000 Delivered-To: freebsd-arch@freebsd.org Received: from citusc17.usc.edu (citusc17.usc.edu [128.125.38.177]) by hub.freebsd.org (Postfix) with ESMTP id 5C32437B479; Tue, 24 Oct 2000 18:48:09 -0700 (PDT) Received: (from kris@localhost) by citusc17.usc.edu (8.9.3/8.9.3) id SAA59240; Tue, 24 Oct 2000 18:49:25 -0700 (PDT) Date: Tue, 24 Oct 2000 18:49:25 -0700 From: Kris Kennaway To: Eivind Eklund Cc: Kris Kennaway , Warner Losh , Mark Murray , arch@FreeBSD.org Subject: Re: cvs commit: src/etc rc Message-ID: <20001024184925.B59218@citusc17.usc.edu> References: <20001024124057.A4309@skriver.dk> <200010232046.e9NKkLR01463@grimreaper.grondar.za> <20001023081548.A41843@bsdwins.com> <200010232046.e9NKkLR01463@grimreaper.grondar.za> <200010232321.RAA11268@harmony.village.org> <20001024124057.A4309@skriver.dk> <200010241256.GAA15067@harmony.village.org> <20001024162445.A58246@warning.follo.net> <20001024083111.D58506@citusc17.usc.edu> <20001024182319.C37278@warning.follo.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20001024182319.C37278@warning.follo.net>; from eivind@FreeBSD.org on Tue, Oct 24, 2000 at 06:23:19PM +0200 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Oct 24, 2000 at 06:23:19PM +0200, Eivind Eklund wrote: > (Moving to -arch) > > On Tue, Oct 24, 2000 at 08:31:11AM -0700, Kris Kennaway wrote: > > On Tue, Oct 24, 2000 at 04:24:45PM +0200, Eivind Eklund wrote: > > > > > Can't we just crypt the data with a strong cipher (or, preferably, two or > > > three strong ciphers) and a key aquired by using random data from Yarrow > > > before writing it out? That would not expose state, assuming we trust the > > > > How would you know the key when you want to read it back in after the > > next reboot? > > You don't know the key. This is *entropy* - you don't need to recover it > exactly :-) This doesnt buy you anything, since the Yarrow algorithm already does this. Kris To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 0:37:59 2000 Delivered-To: freebsd-arch@freebsd.org Received: from citusc17.usc.edu (citusc17.usc.edu [128.125.38.177]) by hub.freebsd.org (Postfix) with ESMTP id 9D39137B479 for ; Wed, 25 Oct 2000 00:37:57 -0700 (PDT) Received: (from kris@localhost) by citusc17.usc.edu (8.11.1/8.11.1) id e9P7d6250352; Wed, 25 Oct 2000 00:39:06 -0700 (PDT) (envelope-from kris) Date: Wed, 25 Oct 2000 00:39:06 -0700 From: Kris Kennaway To: Warner Losh Cc: Nick Sayer , arch@FreeBSD.ORG Subject: Re: Old 4.0-R binaries remaining in 4.1.1-S Message-ID: <20001025003906.A50323@citusc17.usc.edu> References: <39F4B9EA.FACF4464@sftw.com> <200010230328.XAA16342@world.std.com> <200010231857.MAA09601@harmony.village.org> <39F4B9EA.FACF4464@sftw.com> <200010232241.QAA10916@harmony.village.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <200010232241.QAA10916@harmony.village.org>; from imp@village.org on Mon, Oct 23, 2000 at 04:41:48PM -0600 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, Oct 23, 2000 at 04:41:48PM -0600, Warner Losh wrote: > I'd prefer that we have a mtree.obsolete that can be used to find and > delete the files. NetBSD has this concept (I haven't looked at it, > since they do it as part of the binary upgrade), which might make > sense for us too. > > How do people feel about someone going off and doing this? Sounds like a good idea to me. Kris To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 3:17:41 2000 Delivered-To: freebsd-arch@freebsd.org Received: from warning.follo.net (warning.follo.net [195.204.136.30]) by hub.freebsd.org (Postfix) with ESMTP id 4EB6037B479 for ; Wed, 25 Oct 2000 03:17:37 -0700 (PDT) Received: (from eivind@localhost) by warning.follo.net (8.9.3/8.9.3) id MAA11658; Wed, 25 Oct 2000 12:15:26 +0200 (CEST) Date: Wed, 25 Oct 2000 12:15:26 +0200 From: Eivind Eklund To: Kris Kennaway Cc: Warner Losh , Mark Murray , arch@FreeBSD.org Subject: Re: cvs commit: src/etc rc Message-ID: <20001025121526.A11476@warning.follo.net> References: <200010232046.e9NKkLR01463@grimreaper.grondar.za> <20001023081548.A41843@bsdwins.com> <200010232046.e9NKkLR01463@grimreaper.grondar.za> <200010232321.RAA11268@harmony.village.org> <20001024124057.A4309@skriver.dk> <200010241256.GAA15067@harmony.village.org> <20001024162445.A58246@warning.follo.net> <20001024083111.D58506@citusc17.usc.edu> <20001024182319.C37278@warning.follo.net> <20001024184925.B59218@citusc17.usc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: <20001024184925.B59218@citusc17.usc.edu>; from kris@citusc.usc.edu on Tue, Oct 24, 2000 at 06:49:25PM -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Oct 24, 2000 at 06:49:25PM -0700, Kris Kennaway wrote: > On Tue, Oct 24, 2000 at 06:23:19PM +0200, Eivind Eklund wrote: > > On Tue, Oct 24, 2000 at 08:31:11AM -0700, Kris Kennaway wrote: > > > On Tue, Oct 24, 2000 at 04:24:45PM +0200, Eivind Eklund wrote: > > > > Can't we just crypt the data with a strong cipher (or, preferably, two or > > > > three strong ciphers) and a key aquired by using random data from Yarrow > > > > before writing it out? That would not expose state, assuming we trust the > > > > > > How would you know the key when you want to read it back in after the > > > next reboot? > > > > You don't know the key. This is *entropy* - you don't need to recover it > > exactly :-) > > This doesnt buy you anything, since the Yarrow algorithm already does this. I was responding to a mail from Warner regarding writing out the internal state of Yarrow (rather than some Yarrow output). I agree that writing out Yarrow output is a better solution, so please disregard the suggestion. Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 10: 6:18 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id F31FA37B4C5 for ; Wed, 25 Oct 2000 10:06:15 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id TAA26213; Wed, 25 Oct 2000 19:06:14 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id TAA54788; Wed, 25 Oct 2000 19:06:14 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Wed, 25 Oct 2000 19:06:14 +0200 (CEST) From: Marius Bendiksen To: Keith Stevenson Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: <20001024090152.A10742@osaka.louisville.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > Is there documentation somewhere on how to do this? I've looked at the > memory based filesystem stuff before, but was unable to figure out what steps > needed to be performed in order to boot from it. If you have a look at the release build process, you'll see that you basically need to newfs a vn device, stick the appropriate files on it, and have "load -t md_image /my-mfs-image" in your loader script. If you fail to bring along /etc/fstab, you will need to pass some flags to the kernel to let it know whence the root fs comes from, IIRC. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 10: 9:34 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id 557FF37B4D7 for ; Wed, 25 Oct 2000 10:09:31 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id TAA29531; Wed, 25 Oct 2000 19:09:30 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id TAA54796; Wed, 25 Oct 2000 19:09:30 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Wed, 25 Oct 2000 19:09:30 +0200 (CEST) From: Marius Bendiksen To: Dag-Erling Smorgrav Cc: arch@FreeBSD.ORG Subject: Re: Ideas concerning fsck In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > We can get around the need for fstab by using the bootfs idea that AIX and > > others use (Poul-Henning has suggested this also in connection with DEVFS) > > to get less magic in the boot sequence. > I'm not familiar with the concept of bootfs - can somebody enlighten > me? Certainly. A bootfs is basically a filesystem that the system can boot from, which will be in a safe state, because changes are not written back to disk. One example would be when you make an MFS image and boot from it, the other would be doing SCO-style stripped down fs'es which are mounted read-only. The advantage is that you can then hardwire rootvp to that filesystem, and have files stuck on it to intelligently discover how to bring up the system correctly, without doing lots of magic to discover what devices are available for use, and which one you want to boot from, etc. Also, as the state is always consistant, you have no problem with stuff like depending on the validity of the contents of a file, and you don't need to fsck / seperately from the rest of the volumes. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 10:59: 9 2000 Delivered-To: freebsd-arch@freebsd.org Received: from citusc17.usc.edu (citusc17.usc.edu [128.125.38.177]) by hub.freebsd.org (Postfix) with ESMTP id 4A0EB37B479; Wed, 25 Oct 2000 10:59:07 -0700 (PDT) Received: (from kris@localhost) by citusc17.usc.edu (8.11.1/8.11.1) id e9PHxfF51548; Wed, 25 Oct 2000 10:59:41 -0700 (PDT) (envelope-from kris) Date: Wed, 25 Oct 2000 10:59:41 -0700 From: Kris Kennaway To: Wilko Bulte Cc: "John W. De Boskey" , Warner Losh , obrien@freebsd.org, arch@freebsd.org Subject: Re: cvs commit: src/release/scripts dokern.sh Message-ID: <20001025105941.A51512@citusc17.usc.edu> References: <20001024140437.C19518@dragon.nuxi.com> <200010241848.LAA22095@freefall.freebsd.org> <200010242052.OAA19159@harmony.village.org> <20001024140437.C19518@dragon.nuxi.com> <200010242122.PAA19397@harmony.village.org> <20001024182832.A55202@bsdwins.com> <20001024190802.D59218@citusc17.usc.edu> <20001025195029.B6807@freebie.demon.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20001025195029.B6807@freebie.demon.nl>; from wkb@freebie.demon.nl on Wed, Oct 25, 2000 at 07:50:29PM +0200 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, Oct 25, 2000 at 07:50:29PM +0200, Wilko Bulte wrote: > > We need to add back random.ko either to the install kernel, or at > > least have it loaded from somewhere by the time packages are added. > > Or remove the possibility of adding packages during floppy-booted sysinstall > altogether. Why not have people install their packages crud while first-time > booted from the harddisk? I think a lot of people use this feature. Kris To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 11:46:49 2000 Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id 864A837B479; Wed, 25 Oct 2000 11:46:45 -0700 (PDT) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id UAA24293; Wed, 25 Oct 2000 20:46:39 +0200 (CEST) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id UAA56685; Wed, 25 Oct 2000 20:46:39 +0200 (CEST) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Wed, 25 Oct 2000 20:46:39 +0200 (CEST) From: Marius Bendiksen To: Kris Kennaway Cc: Wilko Bulte , "John W. De Boskey" , Warner Losh , obrien@freebsd.org, arch@freebsd.org Subject: Re: cvs commit: src/release/scripts dokern.sh In-Reply-To: <20001025105941.A51512@citusc17.usc.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > Or remove the possibility of adding packages during floppy-booted sysinstall > > altogether. Why not have people install their packages crud while first-time > > booted from the harddisk? > I think a lot of people use this feature. Indeed. I, for one, do this. Also, a number of people use the automated install scripts, et cetera. Note that this would also preclude the ability to do something like what was discussed in connection with inetd, namely to add security packages that change your configuration. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 12: 5:20 2000 Delivered-To: freebsd-arch@freebsd.org Received: from bsdone.bsdwins.com (www.bsdwins.com [192.58.184.33]) by hub.freebsd.org (Postfix) with ESMTP id 0420C37B4C5; Wed, 25 Oct 2000 12:05:18 -0700 (PDT) Received: (from jwd@localhost) by bsdone.bsdwins.com (8.11.0/8.11.0) id e9PJ3Es62991; Wed, 25 Oct 2000 15:03:14 -0400 (EDT) (envelope-from jwd) Date: Wed, 25 Oct 2000 15:03:14 -0400 From: "John W. De Boskey" To: Kris Kennaway Cc: Wilko Bulte , Warner Losh , obrien@FreeBSD.ORG, arch@FreeBSD.ORG Subject: Re: cvs commit: src/release/scripts dokern.sh Message-ID: <20001025150314.A62263@bsdwins.com> References: <20001024140437.C19518@dragon.nuxi.com> <200010241848.LAA22095@freefall.freebsd.org> <200010242052.OAA19159@harmony.village.org> <20001024140437.C19518@dragon.nuxi.com> <200010242122.PAA19397@harmony.village.org> <20001024182832.A55202@bsdwins.com> <20001024190802.D59218@citusc17.usc.edu> <20001025195029.B6807@freebie.demon.nl> <20001025105941.A51512@citusc17.usc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20001025105941.A51512@citusc17.usc.edu>; from kris@citusc17.usc.edu on Wed, Oct 25, 2000 at 10:59:41AM -0700 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG ----- Kris Kennaway's Original Message ----- > On Wed, Oct 25, 2000 at 07:50:29PM +0200, Wilko Bulte wrote: > > > > We need to add back random.ko either to the install kernel, or at > > > least have it loaded from somewhere by the time packages are added. > > > > Or remove the possibility of adding packages during floppy-booted sysinstall > > altogether. Why not have people install their packages crud while first-time > > booted from the harddisk? > > I think a lot of people use this feature. scripted installs... man sysinstall -John > Kris > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 13:40:35 2000 Delivered-To: freebsd-arch@freebsd.org Received: from wantadilla.lemis.com (wantadilla.lemis.com [192.109.197.80]) by hub.freebsd.org (Postfix) with ESMTP id BE34137B4C5; Wed, 25 Oct 2000 13:40:23 -0700 (PDT) Received: (from grog@localhost) by wantadilla.lemis.com (8.11.0/8.9.3) id e9PKaaO90464; Thu, 26 Oct 2000 06:06:36 +0930 (CST) (envelope-from grog) Date: Thu, 26 Oct 2000 06:06:36 +0930 From: Greg Lehey To: Kris Kennaway Cc: Wilko Bulte , "John W. De Boskey" , Warner Losh , obrien@FreeBSD.ORG, arch@FreeBSD.ORG Subject: Re: cvs commit: src/release/scripts dokern.sh Message-ID: <20001026060636.B90340@wantadilla.lemis.com> References: <20001024140437.C19518@dragon.nuxi.com> <200010241848.LAA22095@freefall.freebsd.org> <200010242052.OAA19159@harmony.village.org> <20001024140437.C19518@dragon.nuxi.com> <200010242122.PAA19397@harmony.village.org> <20001024182832.A55202@bsdwins.com> <20001024190802.D59218@citusc17.usc.edu> <20001025195029.B6807@freebie.demon.nl> <20001025105941.A51512@citusc17.usc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <20001025105941.A51512@citusc17.usc.edu>; from kris@citusc17.usc.edu on Wed, Oct 25, 2000 at 10:59:41AM -0700 Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-418-838-708 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wednesday, 25 October 2000 at 10:59:41 -0700, Kris Kennaway wrote: > On Wed, Oct 25, 2000 at 07:50:29PM +0200, Wilko Bulte wrote: > >>> We need to add back random.ko either to the install kernel, or at >>> least have it loaded from somewhere by the time packages are added. >> >> Or remove the possibility of adding packages during floppy-booted sysinstall >> altogether. Why not have people install their packages crud while first-time >> booted from the harddisk? > > I think a lot of people use this feature. It would be interesting to see how many people still do install from floppy. Once we have installed the base, though, we should have the modules available, even if they're in a different location. Am I missing something? Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Wed Oct 25 15: 7: 0 2000 Delivered-To: freebsd-arch@freebsd.org Received: from citusc17.usc.edu (citusc17.usc.edu [128.125.38.177]) by hub.freebsd.org (Postfix) with ESMTP id A525E37B4D7; Wed, 25 Oct 2000 15:06:57 -0700 (PDT) Received: (from kris@localhost) by citusc17.usc.edu (8.11.1/8.11.1) id e9PM7IS52188; Wed, 25 Oct 2000 15:07:18 -0700 (PDT) (envelope-from kris) Date: Wed, 25 Oct 2000 15:07:18 -0700 From: Kris Kennaway To: Greg Lehey Cc: Wilko Bulte , "John W. De Boskey" , Warner Losh , obrien@FreeBSD.ORG, arch@FreeBSD.ORG Subject: Re: cvs commit: src/release/scripts dokern.sh Message-ID: <20001025150718.A52170@citusc17.usc.edu> References: <20001024140437.C19518@dragon.nuxi.com> <200010241848.LAA22095@freefall.freebsd.org> <200010242052.OAA19159@harmony.village.org> <20001024140437.C19518@dragon.nuxi.com> <200010242122.PAA19397@harmony.village.org> <20001024182832.A55202@bsdwins.com> <20001024190802.D59218@citusc17.usc.edu> <20001025195029.B6807@freebie.demon.nl> <20001025105941.A51512@citusc17.usc.edu> <20001026060636.B90340@wantadilla.lemis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20001026060636.B90340@wantadilla.lemis.com>; from grog@lemis.com on Thu, Oct 26, 2000 at 06:06:36AM +0930 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, Oct 26, 2000 at 06:06:36AM +0930, Greg Lehey wrote: > It would be interesting to see how many people still do install from > floppy. Once we have installed the base, though, we should have the > modules available, even if they're in a different location. Am I > missing something? Perhaps. We're talking about the boot floppies, not a complete floppy-based install involving 50 floppies :-) Theres basically a choice between a floppy boot or CDROM boot, and I'd say a large percentage of installs use the former. Still, you have a point that after installation of the system the module is available on the hard disk and it can be kldloaded before adding any packages. If people are using sysinstall for post-install package addition (which I understand is also fairly popular) then it can do the same thing with the currently installed module. Kris To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Thu Oct 26 23:20: 9 2000 Delivered-To: freebsd-arch@freebsd.org Received: from lucifer.ninth-circle.org (lucifer.bart.nl [194.158.168.74]) by hub.freebsd.org (Postfix) with ESMTP id 2D55F37B4CF; Thu, 26 Oct 2000 23:20:04 -0700 (PDT) Received: (from asmodai@localhost) by lucifer.ninth-circle.org (8.11.0/8.11.0) id e9R6K2E55053; Fri, 27 Oct 2000 08:20:02 +0200 (CEST) (envelope-from asmodai) Date: Fri, 27 Oct 2000 08:20:02 +0200 From: Jeroen Ruigrok van der Werven To: John Baldwin Cc: arch@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/share/man/man9 spl.9 SPLASSERT.9 CONDSPLASSERT.9 Message-ID: <20001027082002.C54812@lucifer.bart.nl> Reply-To: arch@FreeBSD.org References: <200010262330.QAA93801@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <200010262330.QAA93801@freefall.freebsd.org>; from jhb@FreeBSD.org on Thu, Oct 26, 2000 at 04:30:03PM -0700 Organisation: VIA Net.Works The Netherlands Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG -On [20001027 01:35], John Baldwin (jhb@FreeBSD.org) wrote: > Modified files: > share/man/man9 spl.9 SPLASSERT.9 CONDSPLASSERT.9 > Log: > Add a note that the spl() API is deprecated in favor of mutexes. Which reminds me, please ignore my lack of mutex clue, don't we need apropriate MUTEXASSERT.9/CONDMUTEXASSERT.9 functions and subsequently manpages? Or do mutexes work so much differently from spls? So more background is appreciated, pointers, reply, whatever, so that I can expand this in the docs. -- Jeroen Ruigrok van der Werven Network- and systemadministrator VIA Net.Works The Netherlands BSD: Technical excellence at its best http://www.via-net-works.nl A friend is a single soul dwelling in two bodies... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Fri Oct 27 10:44:10 2000 Delivered-To: freebsd-arch@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id 2C05437B479 for ; Fri, 27 Oct 2000 10:44:08 -0700 (PDT) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id NAA18910; Fri, 27 Oct 2000 13:44:07 -0400 (EDT) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.0/8.9.1) id e9RHi7Z70989; Fri, 27 Oct 2000 13:44:07 -0400 (EDT) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Fri, 27 Oct 2000 13:44:07 -0400 (EDT) To: Anders Franzen Cc: freebsd-arch@FreeBSD.ORG Subject: Re: loadable drivers In-Reply-To: <39F57A50.5CA40232@uab.ericsson.se> References: <39ED9B42.417E88FD@uab.ericsson.se> <200010232309.RAA11156@harmony.village.org> <39F57A50.5CA40232@uab.ericsson.se> X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs Lucid Message-ID: <14841.48800.680090.502760@grasshopper.cs.duke.edu> Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Anders Franzen writes: > Warner Losh wrote: > > > In message <39ED9B42.417E88FD@uab.ericsson.se> Anders Franzen writes: > > : Hi, I'm playing around with the linux pcmcia/cardbus drivers. > > > > What does this have to do with FreeBSD? > > > Not much except that I have successfully!! ported it to FreeBSD!!!!!. > /Fra I'll bite -- exactly what have you ported to FreeBSD? Thanks, Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message From owner-freebsd-arch Fri Oct 27 11:25: 2 2000 Delivered-To: freebsd-arch@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 631CA37B479; Fri, 27 Oct 2000 11:24:58 -0700 (PDT) Received: from laptop.baldwin.cx (ether.osd.bsdi.com [204.216.28.196]) by pike.osd.bsdi.com (8.11.0/8.9.3) with ESMTP id e9RIORf85082; Fri, 27 Oct 2000 11:24:27 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.4.0 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20001027082002.C54812@lucifer.bart.nl> Date: Fri, 27 Oct 2000 11:25:17 -0700 (PDT) From: John Baldwin To: arch@FreeBSD.org Subject: Re: cvs commit: src/share/man/man9 spl.9 SPLASSERT.9 CONDSPLASSE Cc: cvs-all@FreeBSD.org Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On 27-Oct-00 Jeroen Ruigrok van der Werven wrote: > -On [20001027 01:35], John Baldwin (jhb@FreeBSD.org) wrote: >> Modified files: >> share/man/man9 spl.9 SPLASSERT.9 CONDSPLASSERT.9 >> Log: >> Add a note that the spl() API is deprecated in favor of mutexes. > > Which reminds me, please ignore my lack of mutex clue, don't we need > apropriate MUTEXASSERT.9/CONDMUTEXASSERT.9 functions and subsequently > manpages? My latest commit add a MLINK for mtx_assert.9 to mutex.9. :) With mutexes you just use the mtx_assert() function. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.baldwin.cx/~john/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message