From owner-freebsd-fs Sun Sep 16 7:19:42 2001 Delivered-To: freebsd-fs@freebsd.org Received: from mail.nipsi.de (dsl-213-023-032-013.arcor-ip.net [213.23.32.13]) by hub.freebsd.org (Postfix) with SMTP id 9BA9237B407 for ; Sun, 16 Sep 2001 07:19:30 -0700 (PDT) Received: (qmail 52259 invoked from network); 16 Sep 2001 14:19:05 -0000 Received: from unknown (HELO nipsi.de) (172.16.1.101) by nipsi with SMTP; 16 Sep 2001 14:19:05 -0000 Message-ID: <3BA4B507.CC70ECD4@nipsi.de> Date: Sun, 16 Sep 2001 16:19:51 +0200 From: Dennis Berger X-Mailer: Mozilla 4.75 [de] (Windows NT 5.0; U) X-Accept-Language: de MIME-Version: 1.0 To: freebsd-fs@freebsd.org Subject: Re: Porting a new filesystem to FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org >>Note that the JFS that IBM put out for Linux is the OS/2 >>JFS -- the only thing of real value it brings to the table, >>IMO, is the btree directory structure, which you can put >>into FFS fairly easily (less than a days work). So why nobody implemented it yet ? >>-- Terry >>To Unsubscribe: send mail to majordomo@FreeBSD.org >>with "unsubscribe freebsd-fs" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 14: 8:46 2001 Delivered-To: freebsd-fs@freebsd.org Received: from obsecurity.dyndns.org (adsl-63-207-60-6.dsl.lsan03.pacbell.net [63.207.60.6]) by hub.freebsd.org (Postfix) with ESMTP id 057BF37B401 for ; Sun, 16 Sep 2001 14:08:44 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 8D24466D20; Sun, 16 Sep 2001 14:08:43 -0700 (PDT) Date: Sun, 16 Sep 2001 14:08:43 -0700 From: Kris Kennaway To: Dennis Berger Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010916140843.A21982@xor.obsecurity.org> References: <3BA4B507.CC70ECD4@nipsi.de> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-md5; protocol="application/pgp-signature"; boundary="0OAP2g/MAC+5xKAE" Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3BA4B507.CC70ECD4@nipsi.de>; from Dennis.Berger@nipsi.de on Sun, Sep 16, 2001 at 04:19:51PM +0200 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org --0OAP2g/MAC+5xKAE Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Sun, Sep 16, 2001 at 04:19:51PM +0200, Dennis Berger wrote: > >>Note that the JFS that IBM put out for Linux is the OS/2 > >>JFS -- the only thing of real value it brings to the table, > >>IMO, is the btree directory structure, which you can put > >>into FFS fairly easily (less than a days work). > So why nobody implemented it yet ? Because a lot of people like to talk about how easy things are without doing the work and finding it to be harder. Kris --0OAP2g/MAC+5xKAE Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (FreeBSD) Comment: For info see http://www.gnupg.org iD8DBQE7pRTaWry0BWjoQKURAuqlAJsE6TVLC8idE1B7n/LxtUX5SlqONwCgsWVW zUMcsf5yPsKl51FBYEJJvCU= =Q7km -----END PGP SIGNATURE----- --0OAP2g/MAC+5xKAE-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 15:26:41 2001 Delivered-To: freebsd-fs@freebsd.org Received: from swan.mail.pas.earthlink.net (swan.mail.pas.earthlink.net [207.217.120.123]) by hub.freebsd.org (Postfix) with ESMTP id 3E7B537B40C for ; Sun, 16 Sep 2001 15:26:38 -0700 (PDT) Received: from mindspring.com (dialup-209.245.132.139.Dial1.SanJose1.Level3.net [209.245.132.139]) by swan.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id PAA13892; Sun, 16 Sep 2001 15:26:25 -0700 (PDT) Message-ID: <3BA5273F.A2131982@mindspring.com> Date: Sun, 16 Sep 2001 15:27:11 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Dennis Berger Cc: freebsd-fs@freebsd.org Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Dennis Berger wrote: > > >>Note that the JFS that IBM put out for Linux is the OS/2 > >>JFS -- the only thing of real value it brings to the table, > >>IMO, is the btree directory structure, which you can put > >>into FFS fairly easily (less than a days work). > So why nobody implemented it yet ? It breaks binary backward compatability. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 15:30:38 2001 Delivered-To: freebsd-fs@freebsd.org Received: from Aphex.NewGold.NET (aphex.newgold.net [209.42.222.44]) by hub.freebsd.org (Postfix) with ESMTP id E974937B405 for ; Sun, 16 Sep 2001 15:30:35 -0700 (PDT) Received: (from jmallett@localhost) by Aphex.NewGold.NET (8.11.3/8.11.3) id f8GMUAC68187; Sun, 16 Sep 2001 22:30:10 GMT (envelope-from jmallett) Date: Sun, 16 Sep 2001 22:30:09 +0000 From: Joseph Mallett To: Terry Lambert Cc: Dennis Berger , freebsd-fs@freebsd.org Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010916223009.A53663@NewGold.NET> References: <3BA4B507.CC70ECD4@nipsi.de> <3BA5273F.A2131982@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3BA5273F.A2131982@mindspring.com> User-Agent: Mutt/1.3.22.1i Organisation: New Gold Technology Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sun, Sep 16, 2001 at 03:27:11PM -0700, Terry Lambert wrote: > Dennis Berger wrote: > > > > >>Note that the JFS that IBM put out for Linux is the OS/2 > > >>JFS -- the only thing of real value it brings to the table, > > >>IMO, is the btree directory structure, which you can put > > >>into FFS fairly easily (less than a days work). > > So why nobody implemented it yet ? > > It breaks binary backward compatability. Then why has it not been implemented as a mount time or compile time option? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 15:49:10 2001 Delivered-To: freebsd-fs@freebsd.org Received: from harrier.mail.pas.earthlink.net (harrier.mail.pas.earthlink.net [207.217.121.12]) by hub.freebsd.org (Postfix) with ESMTP id DFB2B37B40C for ; Sun, 16 Sep 2001 15:49:06 -0700 (PDT) Received: from mindspring.com (dialup-209.245.132.139.Dial1.SanJose1.Level3.net [209.245.132.139]) by harrier.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id PAA18989; Sun, 16 Sep 2001 15:48:50 -0700 (PDT) Message-ID: <3BA52C79.E1E247F5@mindspring.com> Date: Sun, 16 Sep 2001 15:49:29 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Kris Kennaway Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Kris Kennaway wrote: > > On Sun, Sep 16, 2001 at 04:19:51PM +0200, Dennis Berger wrote: > > >>Note that the JFS that IBM put out for Linux is the OS/2 > > >>JFS -- the only thing of real value it brings to the table, > > >>IMO, is the btree directory structure, which you can put > > >>into FFS fairly easily (less than a days work). > > So why nobody implemented it yet ? > > Because a lot of people like to talk about how easy things are without > doing the work and finding it to be harder. That's not true. The reason is binary backward compatability, with a minor amount of politics. To guarantee backward compatability, there are several things that have to happen, in addition to the code changes to fsck for checking the new directories and the code changes to ufs to support them (both of which are trivial for anyone who has ever read Knuth's "Seminumerical Algorithms: Sorting and Searching", and is a halfway decent C coder): 1) FreeBSD's fsck needs to treat unknown file types (at least those within a certain range) as being valid files just like files or directories, but with an unknown type, rather than puking on them. 2) The ufs in libstand needs to be taught about the new directory type, in case / is one of them, so that it can still find the kernel to boot it. Unlike Linux, FreeBSD doesn't stomp the absolute block addresses, in order, of the kernel into the boot loader. 3) The FreeBSD VOP_READDIR code needs to be split into seperate "get directory block" and "externalize an already obtained directory block into the system neutral format" functions, and cookies need to go away. 4) The code that uses cookies needs to be changed; this os mostly the Linux ABI and the NFS server code. 5) FreeBSD's fsck needs to be updated to support upgrading existing directories to the new implementation. Without these changes being in the FreeBSD tree for at least one release cycle, the patches would never gain acceptance from anyone, except those people who are intimidated by pseudo-benchmarks intended to market a specific OS or FS as "better than FreeBSD". Note that without #2, the boot FS _must not_ use the new format, and without #3 and #4, the Linux software running weenies, who are the people who mostly want this, will not be able to run their code. #5 is just a "nice to have", and #1 is a "don't destroy my disk if I run the old fsck" option. #1 and #2 are what mostly drive the "one release cycle" requirement. Note that most of the changes in #1 and #3 will be seen as being "gratuitous: adds nothing" by most of the old guard, who want to have instant gratification on code changes that touch already working code in fundamental and potentially destabilizing ways, if they are unable/unwilling to look at the big picture. This kind of leaves anyone trying to do the code (unless they are a core team member) out in the cold, unless they are willing to campaign for it the same way Julian campaigned for the new KSE code (and he had the distinction of being one of the oldest members of the community, as he predates FreeBSD itself, and coded on it back when it was "386BSD+patchkit"). -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 15:58:13 2001 Delivered-To: freebsd-fs@freebsd.org Received: from harrier.mail.pas.earthlink.net (harrier.mail.pas.earthlink.net [207.217.121.12]) by hub.freebsd.org (Postfix) with ESMTP id 22DC037B401 for ; Sun, 16 Sep 2001 15:58:11 -0700 (PDT) Received: from mindspring.com (dialup-209.245.132.139.Dial1.SanJose1.Level3.net [209.245.132.139]) by harrier.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id PAA23028; Sun, 16 Sep 2001 15:58:00 -0700 (PDT) Message-ID: <3BA52EA7.13045ECE@mindspring.com> Date: Sun, 16 Sep 2001 15:58:47 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Joseph Mallett Cc: Dennis Berger , freebsd-fs@freebsd.org Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <3BA5273F.A2131982@mindspring.com> <20010916223009.A53663@NewGold.NET> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Joseph Mallett wrote: > > On Sun, Sep 16, 2001 at 03:27:11PM -0700, Terry Lambert wrote: > > Dennis Berger wrote: > > > > > > >>Note that the JFS that IBM put out for Linux is the OS/2 > > > >>JFS -- the only thing of real value it brings to the table, > > > >>IMO, is the btree directory structure, which you can put > > > >>into FFS fairly easily (less than a days work). > > > So why nobody implemented it yet ? > > > > It breaks binary backward compatability. > > Then why has it not been implemented as a mount time or compile time > option? You want patches? They apply (almost) cleanly to a vanilla FreeBSD 2.2.5. Just don't fsck this volume with a stock fsck, or every one of your directories will disappear in a puff of "unknown file type"... -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 17:19:27 2001 Delivered-To: freebsd-fs@freebsd.org Received: from mail.nipsi.de (dsl-213-023-032-073.arcor-ip.net [213.23.32.73]) by hub.freebsd.org (Postfix) with SMTP id CD80D37B409 for ; Sun, 16 Sep 2001 17:19:21 -0700 (PDT) Received: (qmail 59904 invoked from network); 17 Sep 2001 00:18:59 -0000 Received: from unknown (HELO nipsi.de) (172.16.1.101) by nipsi with SMTP; 17 Sep 2001 00:18:59 -0000 Message-ID: <3BA5419F.BF0C3E70@nipsi.de> Date: Mon, 17 Sep 2001 02:19:43 +0200 From: Dennis Berger X-Mailer: Mozilla 4.75 [de] (Windows NT 5.0; U) X-Accept-Language: de MIME-Version: 1.0 To: tlambert2@mindspring.com Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Then we maybe give the guys at sistina some support, cause they hold back the GFS-project for freebsd. A Projectleader told me if there will be more demand, they bump it to higher priority. GFS filesystem refer http://www.sistina.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 17:31:50 2001 Delivered-To: freebsd-fs@freebsd.org Received: from mailsrv.otenet.gr (mailsrv.otenet.gr [195.170.0.5]) by hub.freebsd.org (Postfix) with ESMTP id 5124B37B409 for ; Sun, 16 Sep 2001 17:31:47 -0700 (PDT) Received: from hades.hell.gr (patr530-a226.otenet.gr [212.205.215.226]) by mailsrv.otenet.gr (8.11.5/8.11.5) with ESMTP id f8H0VUd01230; Mon, 17 Sep 2001 03:31:31 +0300 (EEST) Received: (from charon@localhost) by hades.hell.gr (8.11.6/8.11.6) id f8H0VNm01138; Mon, 17 Sep 2001 03:31:23 +0300 (EEST) (envelope-from charon@labs.gr) Date: Mon, 17 Sep 2001 03:31:19 +0300 From: Giorgos Keramidas To: Joseph Mallett Cc: Terry Lambert , Dennis Berger , freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917033118.A1020@hades.hell.gr> References: <3BA4B507.CC70ECD4@nipsi.de> <3BA5273F.A2131982@mindspring.com> <20010916223009.A53663@NewGold.NET> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010916223009.A53663@NewGold.NET>; from jmallett@NewGold.NET on Sun, Sep 16, 2001 at 10:30:09PM +0000 X-PGP-Fingerprint: 3A 75 52 EB F1 58 56 0D - C5 B8 21 B6 1B 5E 4A C2 X-URL: http://students.ceid.upatras.gr/~keramida/index.html Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Joseph Mallett wrote: > On Sun, Sep 16, 2001 at 03:27:11PM -0700, Terry Lambert wrote: > > Dennis Berger wrote: > > > > > > >>Note that the JFS that IBM put out for Linux is the OS/2 > > > >>JFS -- the only thing of real value it brings to the table, > > > >>IMO, is the btree directory structure, which you can put > > > >>into FFS fairly easily (less than a days work). > > > So why nobody implemented it yet ? > > > > It breaks binary backward compatability. > > Then why has it not been implemented as a mount time or compile time > option? Simple questions also have simple answers. Because nobody has done it yet. There is this motto in the Linux camp that fits very nicely questions like this one: "Do you have any code to show to us?" Please guys. Stop wasting bandwidth to discuss what things would be a nice idea to do, and asking why nobody else has done or thought of this before... If you (or anyone else, this is not personal) think you can do something, then please, go ahead and do it. -giorgos To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 17:59:23 2001 Delivered-To: freebsd-fs@freebsd.org Received: from Aphex.NewGold.NET (aphex.newgold.net [209.42.222.44]) by hub.freebsd.org (Postfix) with ESMTP id 6C2FB37B403 for ; Sun, 16 Sep 2001 17:59:20 -0700 (PDT) Received: (from jmallett@localhost) by Aphex.NewGold.NET (8.11.3/8.11.3) id f8H0wn072220; Mon, 17 Sep 2001 00:58:49 GMT (envelope-from jmallett) Date: Mon, 17 Sep 2001 00:58:44 +0000 From: Joseph Mallett To: Giorgos Keramidas Cc: Terry Lambert , Dennis Berger , freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917005844.A62643@NewGold.NET> References: <3BA4B507.CC70ECD4@nipsi.de> <3BA5273F.A2131982@mindspring.com> <20010916223009.A53663@NewGold.NET> <20010917033118.A1020@hades.hell.gr> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20010917033118.A1020@hades.hell.gr> User-Agent: Mutt/1.3.22.1i Organisation: New Gold Technology Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, Sep 17, 2001 at 03:31:19AM +0300, Giorgos Keramidas wrote: > Joseph Mallett wrote: > > On Sun, Sep 16, 2001 at 03:27:11PM -0700, Terry Lambert wrote: > > > Dennis Berger wrote: > > > > > > > > >>Note that the JFS that IBM put out for Linux is the OS/2 > > > > >>JFS -- the only thing of real value it brings to the table, > > > > >>IMO, is the btree directory structure, which you can put > > > > >>into FFS fairly easily (less than a days work). > > > > So why nobody implemented it yet ? > > > > > > It breaks binary backward compatability. > > > > Then why has it not been implemented as a mount time or compile time > > option? > > Simple questions also have simple answers. > Because nobody has done it yet. > > There is this motto in the Linux camp that fits very nicely questions > like this one: "Do you have any code to show to us?" > > Please guys. Stop wasting bandwidth to discuss what things would be a > nice idea to do, and asking why nobody else has done or thought of > this before... If you (or anyone else, this is not personal) think > you can do something, then please, go ahead and do it. > Errr sorry, you seem to have misunderstood what -i- was talking about... I was under the impression such things may have already been done to some extent, and aside from it breaking binary compat, I wondered why nobody had made it a possible option, at the very least. I'd be glad to do the work to make it optional, if I had code that worked for FFS. I'm not interested in doing the code for FFS itself, but I'd be glad to do what I asked about, which is to make such code optional, as it would be likely to break binary compatability. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 18:45:26 2001 Delivered-To: freebsd-fs@freebsd.org Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120]) by hub.freebsd.org (Postfix) with ESMTP id 9C51C37B403 for ; Sun, 16 Sep 2001 18:45:22 -0700 (PDT) Received: from mindspring.com (dialup-209.245.138.236.Dial1.SanJose1.Level3.net [209.245.138.236]) by albatross.prod.itd.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id SAA22833; Sun, 16 Sep 2001 18:45:14 -0700 (PDT) Message-ID: <3BA555D8.D2C53387@mindspring.com> Date: Sun, 16 Sep 2001 18:46:00 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Dennis Berger Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Dennis Berger wrote: > > Then we maybe give the guys at sistina some support, cause they > hold back the GFS-project for freebsd. > A Projectleader told me if there will be more demand, they bump > it to higher priority. > > GFS filesystem refer http://www.sistina.com I already provided them with patches against their top level source tree which made all of the user space tools and utilities compile on FreeBSD. It too me all of about 2 hours to do all of them. They also informed me that one of their engineers has FreeBSD support code for some of the kernel parts, but has not yet committed it to their tree. My bigest problem with it right now is license, since a GPL means that FreeBSD could not use it as a boot FS, which makes the code useless to me. They have indicated that they are willing to work with the license, but there are still some problems with regard to contributions by third parties made under the license under which they distribute the Linux version: either those contributions will need to be ripped out and redone, or their authors will need to assign rights to Sistina. The kernel space stuff is more complicated; I haven't had experience with the shared hardware interfaces they use (I have had experience with similar interfaces in AIX, though), nor do I have the equipment available to do a driver for them using the Linux driver as a reference implementation; I would want the documentation, anyway, since I would be wary of the end result being GPL encumbered. I estimate from looking at the Linux driver that this would take no more than 3 days worth of work (24 man hours) to get done, for someone with experience. The FS implementation itself is rather trivial, if you know what you are doing, and have knowledge of the internals of the system you are doing the work on. What FreeBSD has done with the VOP_ABORTOP code (removed it) throws a wrench into things, but it is not a serious obstacle. Really, the GFS stuff wants someone to write media drivers with range locking entry points before the rest of the code can progress. If someone wants to tackle the drivers for the oddball hardware, and will do interoperability testing with their Linux version hitting the same storage at the same time, as long as it can be put on a floppy or other removable medium, I can do the FS itself after that (I'm not interested in using GFS on local hard drives for my own purposes, so unless there is a shared device, I don't care about it). But realize, I will NOT be developing in -current, but on the 4.x branch instead, if I do this, and the issue of license must be addressed first. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 18:55: 4 2001 Delivered-To: freebsd-fs@freebsd.org Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120]) by hub.freebsd.org (Postfix) with ESMTP id 982F637B407 for ; Sun, 16 Sep 2001 18:55:01 -0700 (PDT) Received: from mindspring.com (dialup-209.245.138.236.Dial1.SanJose1.Level3.net [209.245.138.236]) by albatross.prod.itd.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id SAA03829; Sun, 16 Sep 2001 18:54:31 -0700 (PDT) Message-ID: <3BA55805.5B6E1976@mindspring.com> Date: Sun, 16 Sep 2001 18:55:17 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Joseph Mallett Cc: Giorgos Keramidas , Dennis Berger , freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <3BA5273F.A2131982@mindspring.com> <20010916223009.A53663@NewGold.NET> <20010917033118.A1020@hades.hell.gr> <20010917005844.A62643@NewGold.NET> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Joseph Mallett wrote: > Errr sorry, you seem to have misunderstood what -i- was talking about... I > was under the impression such things may have already been done to some > extent, and aside from it breaking binary compat, I wondered why nobody > had made it a possible option, at the very least. > > I'd be glad to do the work to make it optional, if I had code that worked > for FFS. I'm not interested in doing the code for FFS itself, but I'd be > glad to do what I asked about, which is to make such code optional, as it > would be likely to break binary compatability. My old code will not make you happy, since you would need to disable soft updates to use it, or add in the dependencies and dependency resolution and rewind/forward code to handle them when they are flushed by the syncd clock. I can still give it to you as patches against 2.2.5 FreeBSD, however, if you want it, and have a 2.2.5 source tree checked out against which to apply it. My code is non-optional (i.e. you do not get the old directory format at all), and relies on you booting from an MFS image, since I did not do the bootstrap code. IMO, you would be better off starting from the 4.x code base, since it has Soft Updates integrated already, and you will need to deal with it. It is probably a two week task for a graduate student to do this code from scratch. If you have that level of capability, don't fear reading Knuth and writing code, since you will be working on a scratch disk anyway, and should expect to have to newfs it 40 or 50 times in the process (I use floppies, or the smallest partition possible for the FS type, when I do this kind of work). -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 19: 8:33 2001 Delivered-To: freebsd-fs@freebsd.org Received: from tomts5-srv.bellnexxia.net (tomts5.bellnexxia.net [209.226.175.25]) by hub.freebsd.org (Postfix) with ESMTP id 149A237B40B for ; Sun, 16 Sep 2001 19:08:30 -0700 (PDT) Received: from xena.gsicomp.on.ca ([65.93.38.74]) by tomts5-srv.bellnexxia.net (InterMail vM.4.01.03.16 201-229-121-116-20010115) with ESMTP id <20010917020829.NZAG16065.tomts5-srv.bellnexxia.net@xena.gsicomp.on.ca>; Sun, 16 Sep 2001 22:08:29 -0400 Received: from hermes (hermes.gsicomp.on.ca [192.168.0.18]) by xena.gsicomp.on.ca (8.11.1/8.11.1) with SMTP id f8H21gu55340; Sun, 16 Sep 2001 22:01:43 -0400 (EDT) (envelope-from matt@gsicomp.on.ca) Message-ID: <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> From: "Matthew Emmerton" To: Cc: References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> Subject: Re: Porting a new filesystem to FreeBSD Date: Sun, 16 Sep 2001 22:01:55 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > Dennis Berger wrote: > > > > Then we maybe give the guys at sistina some support, cause they > > hold back the GFS-project for freebsd. > > A Projectleader told me if there will be more demand, they bump > > it to higher priority. > > > > GFS filesystem refer http://www.sistina.com > > I already provided them with patches against their top level > source tree which made all of the user space tools and > utilities compile on FreeBSD. It too me all of about 2 hours > to do all of them. > > They also informed me that one of their engineers has FreeBSD > support code for some of the kernel parts, but has not yet > committed it to their tree. > > My bigest problem with it right now is license, since a GPL > means that FreeBSD could not use it as a boot FS, which makes > the code useless to me. I can see how FreeBSD would not ship GFS support in the GENERIC kernel (which is GPL-clean), but I don't see why the choice of licence would prevent anyone from using it as a boot FS. This is of great importance to me, since I'm working on porting JFS over to FreeBSD, which is GPL'ed code. Not being able to have a root device which is journaled significantly reduces the appear of having a journaled filesystem available. -- Matt Emmerton To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 19:48:45 2001 Delivered-To: freebsd-fs@freebsd.org Received: from harrier.mail.pas.earthlink.net (harrier.mail.pas.earthlink.net [207.217.121.12]) by hub.freebsd.org (Postfix) with ESMTP id 3173C37B406 for ; Sun, 16 Sep 2001 19:48:40 -0700 (PDT) Received: from mindspring.com (dialup-209.245.138.236.Dial1.SanJose1.Level3.net [209.245.138.236]) by harrier.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id TAA24304; Sun, 16 Sep 2001 19:48:34 -0700 (PDT) Message-ID: <3BA564AF.BF642E23@mindspring.com> Date: Sun, 16 Sep 2001 19:49:19 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Matthew Emmerton Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Matthew Emmerton wrote: > > My bigest problem with it right now is license, since a GPL > > means that FreeBSD could not use it as a boot FS, which makes > > the code useless to me. > > I can see how FreeBSD would not ship GFS support in the GENERIC kernel > (which is GPL-clean), but I don't see why the choice of licence would > prevent anyone from using it as a boot FS. I can not install a precompiled kernel that can boot GFS. I can not boot an emergency boot floppy which is capable of mounting and repairing a GFS on a corrupt hard disk. I can not distribute a product which relies upon GFS as the boot file system, since in doing that, I would have to distribute an illegal binary. Since I have generally been working on FreeBSD based embedded systems for the last four years of my career, and could not use GFS under the current license in this context, it makes doing the work rather unintersting to me. It also makes my employers uninterested in funding my coding time on such a project. > This is of great importance to me, since I'm working on porting > JFS over to FreeBSD, which is GPL'ed code. Not being able to > have a root device which is journaled significantly reduces the > appear of having a journaled filesystem available. Yes. I contacted the OS/2 JFS people within IBM while I was an IBM employee for a little over a year, following IBMs acquisition of Whistle Communications. They were uninterested in releasing the code under another license, meaning they were willing to shoot one of their divisions in the foot in order to achieve their own political and marketing goals. Also, realize that Journaling has the same drawbacks in the face of a hardware failure as does any FS, and that software can not tell the difference between a loss of power, a kernel panic, or a hardware failure, without specially designed equipment. So if you come back up after a failure, you have to do the full check anyway, for safety's sake. So the utility of journalling is primarily in the fast boot after a normal shutdown, or an abnormal shutdown where the cause of the shutdown is known to the kernel prior to the mount. This is fixable... follow along: Unfortunately, PCs do not have a standard "transient OS use only" area (one that is not sensitive to which OS is using it) or even an "OS use only" area (which would throw fits on a dual boot system, but would at least allow it to work single OS) in the CMOS layout/BIOS interface to opaque CMOS location. This needs to be dealt with via a BIOS extension, I believe, and until it is, we're going to have to treat any failure as if it were a panic as a result of a corrupt disk buffer, from either a driver, controller, or physical disk failure. If we are taking votes, I would sugest using a structure like: struct { u_int_8 OS_type; /* which OS*/ u_int_8 count; /* size, including prefix*/ u_int_8 data[1]; /* actually bigger*/ }; And impose a limit in the BIOS interface on "count" that prohibits any single OS from taking all of it just to screw with other OS vendors ability to use the machine (not mentioning any likely names here, just precluding the problem). I would further suggest: 1) The count of available bytes comes from the BIOS interface. 2) vendors be required to have at least two OSs worth of maximum count size of storage (i.e.: the limit is count/2). 3) Records fill from the end, back. This also keeps a vendor from writing two or more records worth, as they will overwrite their own data, but more importantly, it lets the BIOS vendor keep a "free reserve", and use up bytes, as needed, shortening the maximum, but then letting each OS keep operating over BIOS upgrades (in this case, being the first OS means having the last record, which is better). 4) OS_type be the same as the DOS partition type for the OS, so a single OS is not permitted to use two areas, and thus screw with other OS vendors ability to use the machine (angain, not mentioning likely names here). 5) Give me credit for the idea somewhere, in the bowels of their documentation. 8-). FreeBSD could then add a subsystem parameter to the panic() routine (it should do this anyway!), so that it could decide what to write to its CMOS area so that on reboot it could make some good assumptions... this also implies the ability to make it assume "worst case: corrupt CMOS" from a boot prompt, so we don't get stuck. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sun Sep 16 23:41: 7 2001 Delivered-To: freebsd-fs@freebsd.org Received: from ns.caldera.de (ns.caldera.de [212.34.180.1]) by hub.freebsd.org (Postfix) with ESMTP id 2EAD837B40A for ; Sun, 16 Sep 2001 23:41:02 -0700 (PDT) Received: (from hch@localhost) by ns.caldera.de (8.11.1/8.11.1) id f8H6eNb14369; Mon, 17 Sep 2001 08:40:23 +0200 Date: Mon, 17 Sep 2001 08:40:23 +0200 From: Christoph Hellwig To: Terry Lambert Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917084023.A13990@caldera.de> References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3BA555D8.D2C53387@mindspring.com>; from tlambert2@mindspring.com on Sun, Sep 16, 2001 at 06:46:00PM -0700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sun, Sep 16, 2001 at 06:46:00PM -0700, Terry Lambert wrote: > My bigest problem with it right now is license, since a GPL > means that FreeBSD could not use it as a boot FS, which makes > the code useless to me. They have indicated that they are > willing to work with the license, but there are still some > problems with regard to contributions by third parties made > under the license under which they distribute the Linux > version: either those contributions will need to be ripped > out and redone, or their authors will need to assign rights > to Sistina. Sistina has changed the License for GFS 4.2 to some own, propriterary one. They chooses to ignore the GPL problematic and thus might get into legal problems very soon. > The kernel space stuff is more complicated; I haven't had > experience with the shared hardware interfaces they use > (I have had experience with similar interfaces in AIX, though), > nor do I have the equipment available to do a driver for > them using the Linux driver as a reference implementation; > I would want the documentation, anyway, since I would be > wary of the end result being GPL encumbered. I estimate > from looking at the Linux driver that this would take no > more than 3 days worth of work (24 man hours) to get done, > for someone with experience. We (the OpenGFS project) have spend about ten times as much time just to fix the horrible implementation bugs in GFS, not to mention# that it also has a lot of design problems. > If someone wants to tackle the drivers for the oddball hardware, > and will do interoperability testing with their Linux version > hitting the same storage at the same time, as long as it can be > put on a floppy or other removable medium, I can do the FS > itself after that (I'm not interested in using GFS on local > hard drives for my own purposes, so unless there is a shared > device, I don't care about it). But realize, I will NOT be > developing in -current, but on the 4.x branch instead, if I do > this, and the issue of license must be addressed first. If you want to develop and freestanding filesystem driver (e.g. not trying to use butter ugly constructs to shared code between operating systems) feel free to join the OpenGFS project. (http://www.opengfs.org). Christoph -- Of course it doesn't work. We've performed a software upgrade. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 1: 7:22 2001 Delivered-To: freebsd-fs@freebsd.org Received: from gull.mail.pas.earthlink.net (gull.mail.pas.earthlink.net [207.217.121.85]) by hub.freebsd.org (Postfix) with ESMTP id BA67A37B401 for ; Mon, 17 Sep 2001 01:07:16 -0700 (PDT) Received: from mindspring.com (dialup-209.244.106.50.Dial1.SanJose1.Level3.net [209.244.106.50]) by gull.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id BAA23414; Mon, 17 Sep 2001 01:07:02 -0700 (PDT) Message-ID: <3BA5AF53.EE87658F@mindspring.com> Date: Mon, 17 Sep 2001 01:07:47 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Christoph Hellwig Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Christoph Hellwig wrote: > Sistina has changed the License for GFS 4.2 to some own, propriterary > one. They chooses to ignore the GPL problematic and thus might get > into legal problems very soon. I don't know what contributions, if any, were made under the terms of the old license. The new license requires that you license it from them for commercial gain, if you ship it in a product, and charge for it. I think that you could use it in BSD without a problem, from a first reading of it. The intellectual property claims may stop some contributors, if they wish to contribute pantented/patentable technology. The license grants for sources, for which there is a royalty free back-grant to the authors, for derivative works makes sense -- it makes more sense than the FSF approach, for which you must assign rights (as with the Sistina license), and for which your rights are only returned to you under the terms of the GPL, rather than as a perpetual royalty free rights to do with your own code as you wish (yes, I know the GPL does not require that, but it _is_ required for FSF controlled projects, if you want your code accepted by them). So from what I can see, the license is annoying from a commercial perspective, but would not stop a CDROM FreeBSD shipment, and would potentially not stop commercial use (they allow commercial internal use for everyone, and they allow commercial use for a fee for services incorporating the code to educational institutions). So you would still need to license for straight commercial use, or for use in a service for a fee, if you were not a non-profit edcucational institution. > We (the OpenGFS project) have spend about ten times as much time > just to fix the horrible implementation bugs in GFS, not to mention# > that it also has a lot of design problems. I'm only worried about the drivers for the weird hardware, and the interoperability issues. Frankly, I can write filesystems in my sleep... I can churn out that kind of code at about 2,200 lines a day -- ask Julian about my 22,000 line "fetchmail" replacement, which I churned out in a period of only two work weeks -- brainless (simple) code can be written at about the speed I can type. > If you want to develop and freestanding filesystem driver (e.g. > not trying to use butter ugly constructs to shared code between > operating systems) feel free to join the OpenGFS project. > (http://www.opengfs.org). Unfortunately, from the web page, it has the GPL issues which will preclude using it as a boot FS, or shipping a CDROM which would install to it as the root FS type, by way of user selection (preferred) or by default (annoying to commercial users). At least I can actually use the code... -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 1:48:59 2001 Delivered-To: freebsd-fs@freebsd.org Received: from ns.caldera.de (ns.caldera.de [212.34.180.1]) by hub.freebsd.org (Postfix) with ESMTP id B952C37B40C for ; Mon, 17 Sep 2001 01:48:54 -0700 (PDT) Received: (from hch@localhost) by ns.caldera.de (8.11.1/8.11.1) id f8H8mML25020; Mon, 17 Sep 2001 10:48:22 +0200 Date: Mon, 17 Sep 2001 10:48:22 +0200 From: Christoph Hellwig To: Terry Lambert Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917104822.B23758@caldera.de> References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3BA5AF53.EE87658F@mindspring.com>; from tlambert2@mindspring.com on Mon, Sep 17, 2001 at 01:07:47AM -0700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, Sep 17, 2001 at 01:07:47AM -0700, Terry Lambert wrote: > Christoph Hellwig wrote: > > Sistina has changed the License for GFS 4.2 to some own, propriterary > > one. They chooses to ignore the GPL problematic and thus might get > > into legal problems very soon. > > I don't know what contributions, if any, were made under the > terms of the old license. The problem are not contribution but other GPLed code they use. It gets a little offtopic, so if we want to continue that discussion, we should do it off-list. > The new license requires that you license it from them for > commercial gain, if you ship it in a product, and charge for > it. I think that you could use it in BSD without a problem, > from a first reading of it. You could. Not in a _free_ BSD though. > The license grants for sources, for which there is a royalty > free back-grant to the authors, for derivative works makes > sense -- it makes more sense than the FSF approach, for which > you must assign rights (as with the Sistina license), and for > which your rights are only returned to you under the terms of > the GPL, rather than as a perpetual royalty free rights to do > with your own code as you wish (yes, I know the GPL does not > require that, but it _is_ required for FSF controlled projects, > if you want your code accepted by them). The FSF appropeach is compaplity stupid, and with the minor glibc license change we already see that RMS absuses it. Again, we're not on -advocacy lists. > > We (the OpenGFS project) have spend about ten times as much time > > just to fix the horrible implementation bugs in GFS, not to mention# > > that it also has a lot of design problems. > > I'm only worried about the drivers for the weird hardware, and > the interoperability issues. There is no hardware driver. You just need a scsi subsystem that can handle 16 byte cmds if the host controller can handle it. (And of course a controller that actually implements this specific commands). > Unfortunately, from the web page, it has the GPL issues which will > preclude using it as a boot FS, or shipping a CDROM which would > install to it as the root FS type, by way of user selection > (preferred) or by default (annoying to commercial users). At least > I can actually use the code... The first yes, the second not. I didn't suggest that you use the code (once we're done with cleaning it up it won't be the same anyway), but implement a new kernel part, sharing the format and protocol with OpenGFS (and current Sistina GFS, but I suspect they will change their format as often as the LVM one). As you mentioned above that you write almost asleep it shouldn't be too difficult :) Christoph -- Of course it doesn't work. We've performed a software upgrade. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 2:10: 7 2001 Delivered-To: freebsd-fs@freebsd.org Received: from harrier.mail.pas.earthlink.net (harrier.mail.pas.earthlink.net [207.217.121.12]) by hub.freebsd.org (Postfix) with ESMTP id 3898937B403 for ; Mon, 17 Sep 2001 02:10:02 -0700 (PDT) Received: from mindspring.com (dialup-209.244.106.50.Dial1.SanJose1.Level3.net [209.244.106.50]) by harrier.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id CAA05764; Mon, 17 Sep 2001 02:09:52 -0700 (PDT) Message-ID: <3BA5BDBC.107DC520@mindspring.com> Date: Mon, 17 Sep 2001 02:09:16 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Christoph Hellwig Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <20010917104822.B23758@caldera.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Christoph Hellwig wrote: > > > We (the OpenGFS project) have spend about ten times as much time > > > just to fix the horrible implementation bugs in GFS, not to mention# > > > that it also has a lot of design problems. > > > > I'm only worried about the drivers for the weird hardware, and > > the interoperability issues. > > There is no hardware driver. You just need a scsi subsystem that can > handle 16 byte cmds if the host controller can handle it. (And of course > a controller that actually implements this specific commands). This disagrees with a discussion I had off list, as to the use and availableility of DLOCK, and the firmness of the SCSI III standard. The GFS people are using Matt Jacob's fiberchannel driver, and a network based distributed lock manager. I'm really not interested in GFS for it's logging properties alone, when run on a local disk (see my posting on crash recovery, also in this thread). > > Unfortunately, from the web page, it has the GPL issues which will > > preclude using it as a boot FS, or shipping a CDROM which would > > install to it as the root FS type, by way of user selection > > (preferred) or by default (annoying to commercial users). At least > > I can actually use the code... > > The first yes, the second not. Tell me how I can distribute a binary, where it is impossible for all the code to be under GPL because some of the code is under another license, yet statically link it with GPL'ed code? I _can not_ ship a CDROM which would install to it as the root FS type. Nor can I create a FreeBSD bootloader, since reading the kernel off the dosk would require linking ot with GFS code in the libstand -- something I can not do, because of the license conflict there, as well. I can't even do it as a user selection, unless I compile the kernel to an MFS locally -- and doing that is considered a "no no", or I would just distribute all GPL'ed code as ANDF code (using TenDRA), and then link it at install time, and totally igonre the GPL as an irrelevancy. > I didn't suggest that you use the code (once we're done with cleaning > it up it won't be the same anyway), but implement a new kernel part, > sharing the format and protocol with OpenGFS (and current Sistina > GFS, but I suspect they will change their format as often as the LVM > one). > As you mentioned above that you write almost asleep it shouldn't be > too difficult :) Do you have documentation for the layout and theory of operation, said documentation not itself being GPL'ed, but instead, fully public domain and/or published with perpetual rights granted for derivative works? I could code to that from scratch rather easily; on the other hand, it kind of makes no sense to have the code and not be able to use it -- if I did it, it would be the very first time that GPL'ed code had been successfully used as a reference implementation, and you probably could claim that it was the design documentation, not the code, which acted as the reference. 8-). If someone would write a fiberchannle driver, and tell me where I can buy cheap hardware, I could probably write a GFS-like FS from scratch; but it's not something I'm interested in taking on, if it means that there's nothing to run it on but local disks, when I get there (ugh!). -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 2:37: 1 2001 Delivered-To: freebsd-fs@freebsd.org Received: from ns.caldera.de (ns.caldera.de [212.34.180.1]) by hub.freebsd.org (Postfix) with ESMTP id 5F15437B401 for ; Mon, 17 Sep 2001 02:36:56 -0700 (PDT) Received: (from hch@localhost) by ns.caldera.de (8.11.1/8.11.1) id f8H9aRU28535; Mon, 17 Sep 2001 11:36:27 +0200 Date: Mon, 17 Sep 2001 11:36:27 +0200 From: Christoph Hellwig To: Terry Lambert Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917113627.A28298@caldera.de> References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <2 <3BA5BDBC.107DC520@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3BA5BDBC.107DC520@mindspring.com>; from tlambert2@mindspring.com on Mon, Sep 17, 2001 at 02:09:16AM -0700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, Sep 17, 2001 at 02:09:16AM -0700, Terry Lambert wrote: > Christoph Hellwig wrote: > > > > We (the OpenGFS project) have spend about ten times as much time > > > > just to fix the horrible implementation bugs in GFS, not to mention# > > > > that it also has a lot of design problems. > > > > > > I'm only worried about the drivers for the weird hardware, and > > > the interoperability issues. > > > > There is no hardware driver. You just need a scsi subsystem that can > > handle 16 byte cmds if the host controller can handle it. (And of course > > a controller that actually implements this specific commands). > > This disagrees with a discussion I had off list, as to the use > and availableility of DLOCK, and the firmness of the SCSI III > standard. The GFS people are using Matt Jacob's fiberchannel > driver, and a network based distributed lock manager. DLOCK was GFS 3.x. GFS 4.x and OpenGFS use dmep. > > > preclude using it as a boot FS, or shipping a CDROM which would > > > install to it as the root FS type, by way of user selection > > > (preferred) or by default (annoying to commercial users). At least > > > I can actually use the code... > > > > The first yes, the second not. > > Tell me how I can distribute a binary, where it is impossible for > all the code to be under GPL because some of the code is under > another license, yet statically link it with GPL'ed code? I never said that you can redistribute the binary. > I _can not_ ship a CDROM which would install to it as the root FS > type. Nor can I create a FreeBSD bootloader, since reading the > kernel off the dosk would require linking ot with GFS code in the > libstand -- something I can not do, because of the license conflict > there, as well. If libstand is only two-clause BSD on not four-clause it should be ok, at least there are a lot of this mixes. But I really hate this stupid license discussions, especially on technical lists. > I can't even do it as a user selection, unless I compile the kernel > to an MFS locally -- and doing that is considered a "no no", or I > would just distribute all GPL'ed code as ANDF code (using TenDRA), > and then link it at install time, and totally igonre the GPL as an > irrelevancy. You can link it withput problem - as long as you don't redistribute the linked product. > > I didn't suggest that you use the code (once we're done with cleaning > > it up it won't be the same anyway), but implement a new kernel part, > > sharing the format and protocol with OpenGFS (and current Sistina > > GFS, but I suspect they will change their format as often as the LVM > > one). > > As you mentioned above that you write almost asleep it shouldn't be > > too difficult :) > > Do you have documentation for the layout and theory of operation, > said documentation not itself being GPL'ed, but instead, fully > public domain and/or published with perpetual rights granted for > derivative works? Currently we don't. It would be a very very good thing as we are truly open then, compared to Sistina's version. > If someone would write a fiberchannle driver, and tell me where I > can buy cheap hardware, I could probably write a GFS-like FS from > scratch; but it's not something I'm interested in taking on, if it > means that there's nothing to run it on but local disks, when I get > there (ugh!). You can use OpenGFS without dmep-capable hardware as well, there's still the IP option. Christoph -- Of course it doesn't work. We've performed a software upgrade. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 2:50:49 2001 Delivered-To: freebsd-fs@freebsd.org Received: from robin.mail.pas.earthlink.net (robin.mail.pas.earthlink.net [207.217.120.65]) by hub.freebsd.org (Postfix) with ESMTP id 7C28E37B406 for ; Mon, 17 Sep 2001 02:50:44 -0700 (PDT) Received: from mindspring.com (dialup-209.244.106.50.Dial1.SanJose1.Level3.net [209.244.106.50]) by robin.mail.pas.earthlink.net (8.11.5/8.9.3) with ESMTP id f8H9oKD27346; Mon, 17 Sep 2001 02:50:21 -0700 (PDT) Message-ID: <3BA5C78B.FE14882@mindspring.com> Date: Mon, 17 Sep 2001 02:51:07 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Christoph Hellwig Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <2 <3BA5BDBC.107DC520@mindspring.com> <20010917113627.A28298@caldera.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Christoph Hellwig wrote: > > This disagrees with a discussion I had off list, as to the use > > and availableility of DLOCK, and the firmness of the SCSI III > > standard. The GFS people are using Matt Jacob's fiberchannel > > driver, and a network based distributed lock manager. > > DLOCK was GFS 3.x. GFS 4.x and OpenGFS use dmep. I think you are misunderstanding me. There is a SCSI III primitive in the drafts that permits multiple masters to range lock disk ranges to a a particular host. By doing this, it permits multiple writers, by guarantees that each host will not overlap a write. It also causes a writeback invalidation for other masters, in order to effect a distributed cache coherency protocol, without additiona daemons or hardware. > > > > preclude using it as a boot FS, or shipping a CDROM which would > > > > install to it as the root FS type, by way of user selection > > > > (preferred) or by default (annoying to commercial users). At least > > > > I can actually use the code... > > > > > > The first yes, the second not. > > > > Tell me how I can distribute a binary, where it is impossible for > > all the code to be under GPL because some of the code is under > > another license, yet statically link it with GPL'ed code? > > I never said that you can redistribute the binary. Can you explain "The first yes, the second not"? It's a bit cryptic, then... [ ... ] > But I really hate this stupid license discussions, especially on > technical lists. The problem is that you have proposed a technical solution which is politically impossible. You have to expect political reasons why it is impossible to result from the suggestion. Basically, you are calling for volunteers to work on something they won't be able to use, when it's done. > You can link it withput problem - as long as you don't redistribute the > linked product. I might as well use XFS, then, which is at least being ported to FreeBSD... > > If someone would write a fiberchannle driver, and tell me where I > > can buy cheap hardware, I could probably write a GFS-like FS from > > scratch; but it's not something I'm interested in taking on, if it > > means that there's nothing to run it on but local disks, when I get > > there (ugh!). > > You can use OpenGFS without dmep-capable hardware as well, there's > still the IP option. That's much less useful. I'd work on an FS that could do what GFS or XFS could do, without the encumberanbces, and assuming that it was all FS work, and not driver work for hardware I don't have yet. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 3: 9:44 2001 Delivered-To: freebsd-fs@freebsd.org Received: from ns.caldera.de (ns.caldera.de [212.34.180.1]) by hub.freebsd.org (Postfix) with ESMTP id 8F3E437B445 for ; Mon, 17 Sep 2001 03:09:34 -0700 (PDT) Received: (from hch@localhost) by ns.caldera.de (8.11.1/8.11.1) id f8HA94j32272; Mon, 17 Sep 2001 12:09:04 +0200 Date: Mon, 17 Sep 2001 12:09:04 +0200 From: Christoph Hellwig To: Terry Lambert Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010917120904.A31759@caldera.de> References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <2 <3BA5C78B.FE14882@mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <3BA5C78B.FE14882@mindspring.com>; from tlambert2@mindspring.com on Mon, Sep 17, 2001 at 02:51:07AM -0700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, Sep 17, 2001 at 02:51:07AM -0700, Terry Lambert wrote: > I think you are misunderstanding me. There is a SCSI III > primitive in the drafts that permits multiple masters to > range lock disk ranges to a a particular host. By doing > this, it permits multiple writers, by guarantees that each > host will not overlap a write. It also causes a writeback > invalidation for other masters, in order to effect a distributed > cache coherency protocol, without additiona daemons or hardware. Nice thing but irrelevant until SCSI III is actually in more widespread use. > > > Tell me how I can distribute a binary, where it is impossible for > > > all the code to be under GPL because some of the code is under > > > another license, yet statically link it with GPL'ed code? > > > > I never said that you can redistribute the binary. > > Can you explain "The first yes, the second not"? It's a bit > cryptic, then... You can redistribute a FreeBSD cdrom that links gfs into the kernel at install time. > The problem is that you have proposed a technical solution which > is politically impossible. You have to expect political reasons > why it is impossible to result from the suggestion. > > Basically, you are calling for volunteers to work on something > they won't be able to use, when it's done. I don't really care wether there is a FreeBSD (Open-)GFS version. If you want to create one I'm willing to cooperate wherever possible as I _really_ hate duplicated efforts. But I'm not calling for volunteers at all. > > You can link it withput problem - as long as you don't redistribute the > > linked product. > > I might as well use XFS, then, which is at least being ported to > FreeBSD... But XFS is not distributed filesystem. (So if the port is actually ongoing, could you point me to the source?) Christoph -- Of course it doesn't work. We've performed a software upgrade. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 4:51:43 2001 Delivered-To: freebsd-fs@freebsd.org Received: from robin.mail.pas.earthlink.net (robin.mail.pas.earthlink.net [207.217.120.65]) by hub.freebsd.org (Postfix) with ESMTP id 139B037B401 for ; Mon, 17 Sep 2001 04:51:40 -0700 (PDT) Received: from mindspring.com (dialup-209.244.105.125.Dial1.SanJose1.Level3.net [209.244.105.125]) by robin.mail.pas.earthlink.net (8.11.5/8.9.3) with ESMTP id f8HBpGD27534; Mon, 17 Sep 2001 04:51:17 -0700 (PDT) Message-ID: <3BA5E3E2.AB247A2C@mindspring.com> Date: Mon, 17 Sep 2001 04:52:02 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Christoph Hellwig Cc: Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <2 <3BA5C78B.FE14882@mindspring.com> <20010917120904.A31759@caldera.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Christoph Hellwig wrote: > Nice thing but irrelevant until SCSI III is actually in more > widespread use. Seagate supports it today. I'm 65% certain IBM does as well; the issue with SCSI III is that they haven't been able to make the optical parts fast enough (same as the issue with 10Gbit ethernet, which you can buy from HP today). > You can redistribute a FreeBSD cdrom that links gfs into the kernel > at install time. Everything in the GPL screams that you can't distribute code intended to specifically work with GPL code, which is not under the GPL. In other words, it's not legal to distribute a "binary weapon" type distribution to try to skirt the GPL. The FSF went to court on a crypto interface over this, and they won. I can find the specific cites, if you insist. > > Basically, you are calling for volunteers to work on something > > they won't be able to use, when it's done. > > I don't really care wether there is a FreeBSD (Open-)GFS version. > If you want to create one I'm willing to cooperate wherever possible > as I _really_ hate duplicated efforts. > > But I'm not calling for volunteers at all. You were proposing this as an alternative to GFS, in a response to a posting where I identified GFS' licensing problems. You have a competing GFS effort. > > I might as well use XFS, then, which is at least being ported to > > FreeBSD... > > But XFS is not distributed filesystem. (So if the port is actually > ongoing, could you point me to the source?) See the -hackers list archive, where it was discussed recently. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 5:13:28 2001 Delivered-To: freebsd-fs@freebsd.org Received: from the-village.bc.nu (lightning.swansea.linux.org.uk [194.168.151.1]) by hub.freebsd.org (Postfix) with ESMTP id 1FE9D37B415 for ; Mon, 17 Sep 2001 05:13:17 -0700 (PDT) Received: from alan by the-village.bc.nu with local (Exim 3.22 #1) id 15ixKo-0006zj-00; Mon, 17 Sep 2001 13:17:10 +0100 Subject: Re: Porting a new filesystem to FreeBSD To: tlambert2@mindspring.com Date: Mon, 17 Sep 2001 13:17:10 +0100 (BST) Cc: hch@ns.caldera.de (Christoph Hellwig), Dennis.Berger@nipsi.de (Dennis Berger), freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net In-Reply-To: <3BA5BDBC.107DC520@mindspring.com> from "Terry Lambert" at Sep 17, 2001 02:09:16 AM X-Mailer: ELM [version 2.5 PL6] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: From: Alan Cox Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > This disagrees with a discussion I had off list, as to the use > and availableility of DLOCK, and the firmness of the SCSI III > standard. The GFS people are using Matt Jacob's fiberchannel > driver, and a network based distributed lock manager. DLOCK is available on a suprisingly large number of FC drives, but DLOCK is actually useless if you want any kind of fault tolerance - and to an extent so is the very basis of GFS. > If someone would write a fiberchannle driver, and tell me where I > can buy cheap hardware, I could probably write a GFS-like FS from > scratch; but it's not something I'm interested in taking on, if it > means that there's nothing to run it on but local disks, when I get > there (ugh!). The issue with hardware is mostly getting hold of a magic adapter. QlogicFC 2100 controllers go or $US30 at times on ebay FC drives go for peanuts because nobody knows what to do with them 2nd hand But the magic "connect a single fc driver to fc" board you need to avoid using expensive fc disk array boxes are like gold dust To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 5:33:28 2001 Delivered-To: freebsd-fs@freebsd.org Received: from netbank.com.br (garrincha.netbank.com.br [200.203.199.88]) by hub.freebsd.org (Postfix) with ESMTP id 5915437B411 for ; Mon, 17 Sep 2001 05:33:24 -0700 (PDT) Received: from 1-024.ctame701-1.telepar.net.br (1-024.ctame701-1.telepar.net.br [200.181.137.24]) by netbank.com.br (Postfix) with ESMTP id E2DE646809; Mon, 17 Sep 2001 09:31:09 -0300 (BRST) Received: (from localhost user: 'riel', uid#500) by imladris.surriel.com with ESMTP id ; Mon, 17 Sep 2001 09:33:12 -0300 Date: Mon, 17 Sep 2001 09:33:11 -0300 (BRST) From: Rik van Riel X-X-Sender: To: Terry Lambert Cc: Christoph Hellwig , Dennis Berger , , Subject: Re: Porting a new filesystem to FreeBSD In-Reply-To: <3BA5C78B.FE14882@mindspring.com> Message-ID: X-spambait: aardvark@kernelnewbies.org X-spammeplease: aardvark@nl.linux.org MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, 17 Sep 2001, Terry Lambert wrote: > [ ... ] > > But I really hate this stupid license discussions, especially on > > technical lists. > > The problem is that you have proposed a technical solution which > is politically impossible. You have to expect political reasons > why it is impossible to result from the suggestion. Interesting that you have to be offended by this, since you seem to keep your "1 million tcp connections" stuff proprietary ;) If that means you won't be able to use opengfs or xfs in your product, that's bad luck for you. But please don't try to persuade people to not port something which could be useful for many other freebsd users... cheers, Rik -- IA64: a worthy successor to i860. http://www.surriel.com/ http://distro.conectiva.com/ Send all your spam to aardvark@nl.linux.org (spam digging piggy) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 8:20:35 2001 Delivered-To: freebsd-fs@freebsd.org Received: from beppo.feral.com (beppo.feral.com [192.67.166.79]) by hub.freebsd.org (Postfix) with ESMTP id 29B4F37B410 for ; Mon, 17 Sep 2001 08:20:32 -0700 (PDT) Received: from wonky.feral.com (wonky.feral.com [192.67.166.7]) by beppo.feral.com (8.11.3/8.11.3) with ESMTP id f8HFKJI24886; Mon, 17 Sep 2001 08:20:19 -0700 (PDT) (envelope-from mjacob@feral.com) Date: Mon, 17 Sep 2001 08:20:12 -0700 (PDT) From: Matthew Jacob Reply-To: To: Terry Lambert Cc: Christoph Hellwig , Dennis Berger , , Subject: Re: Porting a new filesystem to FreeBSD In-Reply-To: <3BA5E3E2.AB247A2C@mindspring.com> Message-ID: <20010917080710.R58734-100000@wonky.feral.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, 17 Sep 2001, Terry Lambert wrote: > Christoph Hellwig wrote: > > Nice thing but irrelevant until SCSI III is actually in more > > widespread use. > > Seagate supports it today. I'm 65% certain IBM does as well; > the issue with SCSI III is that they haven't been able to make > the optical parts fast enough (same as the issue with 10Gbit > ethernet, which you can buy from HP today). A slight clarification here. SCSI-3 is a set of specifications about various forms of the SCSI protocol. This has been pretty much put to bed some time back. Various pieces of it evolve. The commands for devices are in various subportions- DLOCK would probably be described in SPC-2 (SCSI Primary Commnds, 2) which is ahead of the nominal 'SCSI-3' set. The actual hardware transport for old fashioned copper wire SCSI is more formaly known as SPI (SCSI Parallel Interface). This is under the T10 part of NCITS (http://www.t10.org). Fibre Channel, at the signalling level, is covered under FC-PH and FC-PH-2, and is covered under the T11 part of NCITS (http://www.t11.org). The specs for each develop pretty nearly independently. The current 'just now shipping' parts for optical FC transport is 2Gb. Considering that *most* system I/O bus interfaces will practially saturate with this, worrying about 10Gb yet is premature. At any rate, locking, which this is the primary issue of, is not sensitive to data rate. It's more sensitive to media reliability and packet overhead. SCSI over FC vs. IP over FC or IP over GigEthernet has higher overhead (due to the nature of FC-SCSI exchanges vs. IP Exchanges), SCSI over FC has substantially less overhead than SCSI over SPI. One of things I've argued for is that DLOCK commands should piggy back with I/O commands. That is, you do a write of metadata along with the lock name and acquisition info- if you acquire the lock, the write succeeds- basically like the alpha's Store Locked Conditional instructions. Or you could use a single FC ELS frame to do the DLOCKs and not even have it part of SCSI at all. Or you could use a parallel port scoreboard interconnect (somebody made a cluster of Multias using this). There are a variety of locking mechanisms. > > > > I might as well use XFS, then, which is at least being ported to > > > FreeBSD... > > > > But XFS is not distributed filesystem. (So if the port is actually There's cXFS... -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 9:33:57 2001 Delivered-To: freebsd-fs@freebsd.org Received: from zipcode.az.mvista.com (fw-az.mvista.com [65.200.49.158]) by hub.freebsd.org (Postfix) with ESMTP id 15BEB37B41D for ; Mon, 17 Sep 2001 09:33:30 -0700 (PDT) Received: from mvista.com (IDENT:sdake@persist.az.mvista.com [10.50.1.216]) by zipcode.az.mvista.com (8.9.3/8.9.3) with ESMTP id JAA00368; Mon, 17 Sep 2001 09:41:50 -0700 Message-ID: <3BA62588.20005@mvista.com> Date: Mon, 17 Sep 2001 09:32:08 -0700 From: Steven Dake User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.3) Gecko/20010801 X-Accept-Language: en-us MIME-Version: 1.0 To: "Timothy D. Witham" Cc: freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <20010917084023.A13990@caldera.de> <3BA5AF53.EE87658F@mindspring.com> <2 <3BA5BDBC.107DC520@mindspring.com> <20010917113627.A28298@caldera.de> <3BA5C78B.FE14882@mindspring.com> <1000741723.15009.78.camel@wookie-laptop.pdx.osdl.net> Content-Type: multipart/alternative; boundary="------------020906060400000905040004" Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org --------------020906060400000905040004 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Timothy D. Witham wrote: >On Mon, 2001-09-17 at 02:51, Terry Lambert wrote: > >>Christoph Hellwig wrote: >> >>>>This disagrees with a discussion I had off list, as to the use >>>>and availableility of DLOCK, and the firmness of the SCSI III >>>>standard. The GFS people are using Matt Jacob's fiberchannel >>>>driver, and a network based distributed lock manager. >>>> >>>DLOCK was GFS 3.x. GFS 4.x and OpenGFS use dmep. >>> >>I think you are misunderstanding me. There is a SCSI III >>primitive in the drafts that permits multiple masters to >>range lock disk ranges to a a particular host. By doing >>this, it permits multiple writers, by guarantees that each >>host will not overlap a write. It also causes a writeback >>invalidation for other masters, in order to effect a distributed >>cache coherency protocol, without additiona daemons or hardware. >> > I have a very different and practical reason for not wanting to >see this used. And it has to do with the fact that it is a feature >that isn't used by the volume portion of the market. As drives can >be shipped to 99.999+% of the customers with this feature not working >correctly and they will be happy. This means that the vendor has >very little economic incentive to make this work correctly under all >conditions. The cost of implementing one fix would eat up way more >margin dollars than they would get out of the small numbers of drives >sold in these situations. > >Which puts you in the position of: > >GFS User: "Vendor this feature doesn't work on your drives like >it is documented.". >Vendor: "Let us check it out .... It works under our tests." >GFS User: "But it doesn't work for me." >Vendor: "Well we checked it out again and they way you are using > it causes the problem." >GFS User: "Fix it." >Vendor: "No, but we will allow you to return the drives for a refund." > > > And of course by using a SCSI features then you can only use >SCSI. > Interesting argument, however, if range locking were to become a differentiating factor because of some requirement of clustering storage, SCSI manufacturers would, eventually, follow the spec. There may be easier ways to solve this problem, though, then to persuade the SCSI vendors in to implementing the proper spec. --------------020906060400000905040004 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit

Timothy D. Witham wrote:
On Mon, 2001-09-17 at 02:51, Terry Lambert wrote:
Christoph Hellwig wrote:
This disagrees with a discussion I had off list, as to the use
and availableility of DLOCK, and the firmness of the SCSI III
standard.  The GFS people are using Matt Jacob's fiberchannel
driver, and a network based distributed lock manager.
DLOCK was GFS 3.x.  GFS 4.x and OpenGFS use dmep.
I think you are misunderstanding me.  There is a SCSI III
primitive in the drafts that permits multiple masters to
range lock disk ranges  to a a particular host.  By doing
this, it permits multiple writers, by guarantees that each
host will not overlap a write.  It also causes a writeback
invalidation for other masters, in order to effect a distributed
cache coherency protocol, without additiona daemons or hardware.

  I have a very different and practical reason for not wanting to
see this used. And it has to do with the fact that it is a feature
that isn't used by the volume portion of the market. As drives can
be shipped to 99.999+% of the customers with this feature not working
correctly and they will be happy.  This means that the vendor has
very little economic incentive to make this work correctly under all
conditions. The cost of implementing one fix would eat up way more
margin dollars than they would get out of the small numbers of drives
sold in these situations. 

Which puts you in the position of:

GFS User: "Vendor this feature doesn't work on your drives like 
it is documented.".
Vendor: "Let us check it out .... It works under our tests."
GFS User: "But it doesn't work for me."
Vendor: "Well we checked it out again and they way you are using
         it causes the problem."
GFS User: "Fix it.
"
Vendor: "No, but we will allow you to return the drives for a refund."


  And of course by using a SCSI features then you can only use
SCSI.
Interesting argument, however, if range locking were to become a differentiating factor because of some requirement of clustering storage, SCSI manufacturers would, eventually, follow the spec.  There may be easier ways to solve this problem, though, then to persuade the SCSI vendors in to implementing the proper spec.

<CUT>

--------------020906060400000905040004-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 15:38:27 2001 Delivered-To: freebsd-fs@freebsd.org Received: from tomts7-srv.bellnexxia.net (tomts7.bellnexxia.net [209.226.175.40]) by hub.freebsd.org (Postfix) with ESMTP id 43FD037B406 for ; Mon, 17 Sep 2001 15:38:16 -0700 (PDT) Received: from xena.gsicomp.on.ca ([65.93.38.74]) by tomts7-srv.bellnexxia.net (InterMail vM.4.01.03.16 201-229-121-116-20010115) with ESMTP id <20010917204229.GDV10041.tomts7-srv.bellnexxia.net@xena.gsicomp.on.ca>; Mon, 17 Sep 2001 16:42:29 -0400 Received: from hermes (hermes.gsicomp.on.ca [192.168.0.18]) by xena.gsicomp.on.ca (8.11.1/8.11.1) with SMTP id f8HKZiu57981; Mon, 17 Sep 2001 16:35:44 -0400 (EDT) (envelope-from matt@gsicomp.on.ca) Message-ID: <003b01c13fb8$6466b6f0$1200a8c0@gsicomp.on.ca> From: "Matthew Emmerton" To: Cc: References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> <3BA564AF.BF642E23@mindspring.com> Subject: Re: Porting a new filesystem to FreeBSD Date: Mon, 17 Sep 2001 16:36:12 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > Matthew Emmerton wrote: > > > My bigest problem with it right now is license, since a GPL > > > means that FreeBSD could not use it as a boot FS, which makes > > > the code useless to me. > > > > I can see how FreeBSD would not ship GFS support in the GENERIC kernel > > (which is GPL-clean), but I don't see why the choice of licence would > > prevent anyone from using it as a boot FS. > > I can not install a precompiled kernel that can boot GFS. > > I can not boot an emergency boot floppy which is capable > of mounting and repairing a GFS on a corrupt hard disk. > > I can not distribute a product which relies upon GFS as > the boot file system, since in doing that, I would have to > distribute an illegal binary. > > Since I have generally been working on FreeBSD based embedded > systems for the last four years of my career, and could not > use GFS under the current license in this context, it makes > doing the work rather unintersting to me. It also makes my > employers uninterested in funding my coding time on such a > project. This makes it totally different. The three scenarios you outline above are only problematic when you're trying to support a pre-packaged distribution that supports GFS out-of-the box, as the GPL licence "infects" the entire product. They do not apply to a sysadmin who chooses to switch a production system from UFS to GFS (or JFS) and support it internally. > > This is of great importance to me, since I'm working on porting > > JFS over to FreeBSD, which is GPL'ed code. Not being able to > > have a root device which is journaled significantly reduces the > > appear of having a journaled filesystem available. > > Yes. I contacted the OS/2 JFS people within IBM while I was an > IBM employee for a little over a year, following IBMs acquisition > of Whistle Communications. They were uninterested in releasing > the code under another license, meaning they were willing to shoot > one of their divisions in the foot in order to achieve their own > political and marketing goals. Agreed. With the amount of information I see flying around internally at IBM (where I am employed) and how fanatic they are about preventing GPL pollution with their Linux ports of products, I have to question why they chose to licence various things under the GPL. -- Matt Emmerton To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 16:13: 4 2001 Delivered-To: freebsd-fs@freebsd.org Received: from tao.org.uk (genius.tao.org.uk [212.135.162.51]) by hub.freebsd.org (Postfix) with ESMTP id 8862037B413 for ; Mon, 17 Sep 2001 16:12:59 -0700 (PDT) Received: by tao.org.uk (Postfix, from userid 100) id DB34F431; Tue, 18 Sep 2001 00:12:52 +0100 (BST) Date: Tue, 18 Sep 2001 00:12:52 +0100 From: Josef Karthauser To: Matthew Emmerton Cc: tlambert2@mindspring.com, freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD Message-ID: <20010918001252.R15600@tao.org.uk> References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> <3BA564AF.BF642E23@mindspring.com> <003b01c13fb8$6466b6f0$1200a8c0@gsicomp.on.ca> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-md5; protocol="application/pgp-signature"; boundary="JB7KW7Ey7eB5HOHs" Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <003b01c13fb8$6466b6f0$1200a8c0@gsicomp.on.ca>; from matt@gsicomp.on.ca on Mon, Sep 17, 2001 at 04:36:12PM -0400 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org --JB7KW7Ey7eB5HOHs Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Sep 17, 2001 at 04:36:12PM -0400, Matthew Emmerton wrote: > > Since I have generally been working on FreeBSD based embedded > > systems for the last four years of my career, and could not > > use GFS under the current license in this context, it makes > > doing the work rather unintersting to me. It also makes my > > employers uninterested in funding my coding time on such a > > project. >=20 > This makes it totally different. The three scenarios you outline above a= re > only problematic when you're trying to support a pre-packaged distribution > that supports GFS out-of-the box, as the GPL licence "infects" the entire > product. They do not apply to a sysadmin who chooses to switch a product= ion > system from UFS to GFS (or JFS) and support it internally. That's true, but unless the install tools support it, it makes it much more difficult for the sysadmin to get the root file system off the ground. Joe --JB7KW7Ey7eB5HOHs Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (FreeBSD) Comment: For info see http://www.gnupg.org iEYEARECAAYFAjumg3QACgkQXVIcjOaxUBYtzwCguDToFADq/SOuEgkFcRbxFkln XkIAniFpQtB2YIO70FAd1VHPYU7idrQR =C54n -----END PGP SIGNATURE----- --JB7KW7Ey7eB5HOHs-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Sep 17 22:53:45 2001 Delivered-To: freebsd-fs@freebsd.org Received: from gull.mail.pas.earthlink.net (gull.mail.pas.earthlink.net [207.217.121.85]) by hub.freebsd.org (Postfix) with ESMTP id 6A91537B40A for ; Mon, 17 Sep 2001 22:53:41 -0700 (PDT) Received: from mindspring.com (dialup-209.245.141.224.Dial1.SanJose1.Level3.net [209.245.141.224]) by gull.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id WAA00402; Mon, 17 Sep 2001 22:53:16 -0700 (PDT) Message-ID: <3BA6E17B.7DD12261@mindspring.com> Date: Mon, 17 Sep 2001 22:54:03 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Rik van Riel Cc: Christoph Hellwig , Dennis Berger , freebsd-fs@FreeBSD.ORG, opengfs-devel@lists.sourceforge.net Subject: Re: Porting a new filesystem to FreeBSD References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Rik van Riel wrote: > > The problem is that you have proposed a technical solution which > > is politically impossible. You have to expect political reasons > > why it is impossible to result from the suggestion. > > Interesting that you have to be offended by this, since > you seem to keep your "1 million tcp connections" stuff > proprietary ;) I posted the patch for the FreeBSD ucred stuff to the FreeBSD lists; it was incorporated in FreeBSD proper. I also posted the patches to allow tuning of open files and TCP vs. UDP sockets seperately, at boot time, but they were rejected, though no one else came up with a better way of achieving the same goal. The client side is somewhat a different matter. I posted the hash collision domain fixes for outbound connections to -current, as well. While it's unreasonable to do a million connections from a client, FreeBSD still has a number of bugs that mean you have to be very, very careful about how you code your client program to be able to do that level of connections. Effectively, you have to explicitly trigger the three "if" statements that get you around sharing the collision domain, even after the patches. No one has really bothered to look at the TCP code, or the problem would be obvious to them, the fix less so: I haven't bothered to fix it because I don't need that many client connections for my (mostly) server product. If anyone really cared, they'd bring in my patch, and then tackle the hash algorithm used to select ports, as well as the port reuse case when a port is in the "ALL" vs. "one or more single interfaces" hash collision case. The rest is a matter of tuning, and I have jumped down the throats of people who can't tune a system to save their lives (including those stupid benchmarks on the SPAM engine that showed FreeBSD to be "so poor"), that I can't count the number on both hands and both feet. People who think they can sysctl FreeBSD into a high performance box are deluded. > If that means you won't be able to use opengfs or xfs in > your product, that's bad luck for you. But please don't > try to persuade people to not port something which could > be useful for many other freebsd users... I am one of a handful of people capable of doing the work in a reasonably short period of time. I am merely stating my opinion on the utility of the work, to me. The other people I know who are in the same boat (at Zambeel and elsewhere) are also embedded systems people. If you can recruit some user space perl script kiddie to do the work for you, then don't let me dissuade them from doing the work, or from you accepting their code into your source repository. If on the other hand, you want the work done by someone with 20 years of UNIX kernel experience and over 10 puttering around in UNIX file systems, then you will need to come up with a more compelling argument than "other people can use it, but you can't -- please do the work anyway". Cheers, -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 0:41: 4 2001 Delivered-To: freebsd-fs@freebsd.org Received: from tomts4-srv.bellnexxia.net (tomts4.bellnexxia.net [209.226.175.10]) by hub.freebsd.org (Postfix) with ESMTP id 7C38837B403 for ; Tue, 18 Sep 2001 00:40:59 -0700 (PDT) Received: from xena.gsicomp.on.ca ([65.93.38.74]) by tomts7-srv.bellnexxia.net (InterMail vM.4.01.03.16 201-229-121-116-20010115) with ESMTP id <20010917204229.GDV10041.tomts7-srv.bellnexxia.net@xena.gsicomp.on.ca>; Mon, 17 Sep 2001 16:42:29 -0400 Received: from hermes (hermes.gsicomp.on.ca [192.168.0.18]) by xena.gsicomp.on.ca (8.11.1/8.11.1) with SMTP id f8HKZiu57981; Mon, 17 Sep 2001 16:35:44 -0400 (EDT) (envelope-from matt@gsicomp.on.ca) Message-ID: <003b01c13fb8$6466b6f0$1200a8c0@gsicomp.on.ca> From: "Matthew Emmerton" To: Cc: References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> <3BA564AF.BF642E23@mindspring.com> Subject: Re: Porting a new filesystem to FreeBSD Date: Mon, 17 Sep 2001 16:36:12 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > Matthew Emmerton wrote: > > > My bigest problem with it right now is license, since a GPL > > > means that FreeBSD could not use it as a boot FS, which makes > > > the code useless to me. > > > > I can see how FreeBSD would not ship GFS support in the GENERIC kernel > > (which is GPL-clean), but I don't see why the choice of licence would > > prevent anyone from using it as a boot FS. > > I can not install a precompiled kernel that can boot GFS. > > I can not boot an emergency boot floppy which is capable > of mounting and repairing a GFS on a corrupt hard disk. > > I can not distribute a product which relies upon GFS as > the boot file system, since in doing that, I would have to > distribute an illegal binary. > > Since I have generally been working on FreeBSD based embedded > systems for the last four years of my career, and could not > use GFS under the current license in this context, it makes > doing the work rather unintersting to me. It also makes my > employers uninterested in funding my coding time on such a > project. This makes it totally different. The three scenarios you outline above are only problematic when you're trying to support a pre-packaged distribution that supports GFS out-of-the box, as the GPL licence "infects" the entire product. They do not apply to a sysadmin who chooses to switch a production system from UFS to GFS (or JFS) and support it internally. > > This is of great importance to me, since I'm working on porting > > JFS over to FreeBSD, which is GPL'ed code. Not being able to > > have a root device which is journaled significantly reduces the > > appear of having a journaled filesystem available. > > Yes. I contacted the OS/2 JFS people within IBM while I was an > IBM employee for a little over a year, following IBMs acquisition > of Whistle Communications. They were uninterested in releasing > the code under another license, meaning they were willing to shoot > one of their divisions in the foot in order to achieve their own > political and marketing goals. Agreed. With the amount of information I see flying around internally at IBM (where I am employed) and how fanatic they are about preventing GPL pollution with their Linux ports of products, I have to question why they chose to licence various things under the GPL. -- Matt Emmerton To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 0:55: 8 2001 Delivered-To: freebsd-fs@freebsd.org Received: from hagbard.io.com (hagbard.io.com [199.170.88.13]) by hub.freebsd.org (Postfix) with ESMTP id 043E637B407 for ; Tue, 18 Sep 2001 00:55:06 -0700 (PDT) Received: from localhost (andrewl@localhost) by hagbard.io.com (8.11.2/8.11.2) with ESMTP id f8I7t5l08292 for ; Tue, 18 Sep 2001 02:55:05 -0500 X-Authentication-Warning: hagbard.io.com: andrewl owned process doing -bs Date: Tue, 18 Sep 2001 02:55:05 -0500 (CDT) From: "Andrew P. Lentvorski" To: Subject: Patches for NFS lockmanager Version 4 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Folks, I have made a set of changes to the -CURRENT rpc.lockd which enables it to pass the Connectathon 2001 regression suites when being run from a Solaris client to a FreeBSD server. The patches enable proper lock testing on file byte ranges. I'd like to get someone on the main FreeBSD team to look at it and get it committed (or at least give me a reason why they rejected it) before embarking on the more significant task of enabling some of the nastier cases of file range locking and unlocking. Also, I'd love to talk to someone more in the know about implementing NFS and NFS lock manager protocols. A better set of regression tests than the Connectathon suite would also be useful if they exist. Let me know who I should talk to and what I should do. Thanks, Andy Lentvorski To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 7:33:56 2001 Delivered-To: freebsd-fs@freebsd.org Received: from elvis.mu.org (elvis.mu.org [216.33.66.196]) by hub.freebsd.org (Postfix) with ESMTP id B37AC37B40F for ; Tue, 18 Sep 2001 07:33:47 -0700 (PDT) Received: by elvis.mu.org (Postfix, from userid 1192) id 6AEAE81D08; Tue, 18 Sep 2001 09:33:42 -0500 (CDT) Date: Tue, 18 Sep 2001 09:33:42 -0500 From: Alfred Perlstein To: "Andrew P. Lentvorski" Cc: freebsd-fs@freebsd.org Subject: Re: Patches for NFS lockmanager Version 4 Message-ID: <20010918093342.B61456@elvis.mu.org> 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 andrewl@io.com on Tue, Sep 18, 2001 at 02:55:05AM -0500 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org * Andrew P. Lentvorski [010918 02:55] wrote: > Folks, > > I have made a set of changes to the -CURRENT rpc.lockd which enables it to > pass the Connectathon 2001 regression suites when being run from a Solaris > client to a FreeBSD server. The patches enable proper lock testing on > file byte ranges. > > I'd like to get someone on the main FreeBSD team to look at it and get it > committed (or at least give me a reason why they rejected it) before > embarking on the more significant task of enabling some of the nastier > cases of file range locking and unlocking. > > Also, I'd love to talk to someone more in the know about implementing NFS > and NFS lock manager protocols. A better set of regression tests than the > Connectathon suite would also be useful if they exist. > > Let me know who I should talk to and what I should do. You can probably talk to either me (alfred), David Cross (crossd) or Ian Dowse (iedowse) about this. I'm interested to see what you have. -- -Alfred Perlstein [alfred@freebsd.org] 'Instead of asking why a piece of software is using "1970s technology," start asking why software is ignoring 30 years of accumulated wisdom.' To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 12:38:46 2001 Delivered-To: freebsd-fs@freebsd.org Received: from avocet.mail.pas.earthlink.net (avocet.mail.pas.earthlink.net [207.217.121.50]) by hub.freebsd.org (Postfix) with ESMTP id 9E21B37B410 for ; Tue, 18 Sep 2001 12:38:43 -0700 (PDT) Received: from dialup-209.247.143.42.dial1.sanjose1.level3.net ([209.247.143.42] helo=mindspring.com) by avocet.mail.pas.earthlink.net with esmtp (Exim 3.32 #2) id 15jQhc-0006Hl-00; Tue, 18 Sep 2001 12:38:40 -0700 Message-ID: <3BA7A2ED.70F39C08@mindspring.com> Date: Tue, 18 Sep 2001 12:39:25 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Matthew Emmerton Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Porting a new filesystem to FreeBSD References: <3BA4B507.CC70ECD4@nipsi.de> <20010916140843.A21982@xor.obsecurity.org> <3BA52C79.E1E247F5@mindspring.com> <3BA5419F.BF0C3E70@nipsi.de> <3BA555D8.D2C53387@mindspring.com> <004001c13f1c$bbcc3100$1200a8c0@gsicomp.on.ca> <3BA564AF.BF642E23@mindspring.com> <003b01c13fb8$6466b6f0$1200a8c0@gsicomp.on.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Matthew Emmerton wrote: [ ... ] > This makes it totally different. The three scenarios you outline above are > only problematic when you're trying to support a pre-packaged distribution > that supports GFS out-of-the box, as the GPL licence "infects" the entire > product. They do not apply to a sysadmin who chooses to switch a production > system from UFS to GFS (or JFS) and support it internally. I think you will find that most systems administrators will not be able to do the code necessary to make this work, and that it will take incenting a professional programmer to do the work. [ ... ] > Agreed. With the amount of information I see flying around internally at > IBM (where I am employed) and how fanatic they are about preventing GPL > pollution with their Linux ports of products, I have to question why they > chose to licence various things under the GPL. So you've gone through the week long class and seen the 18 page Louts FreeLance presentation on ow to treat GPL'ed code as hazardous waste that IBM forces their employees who will be working with Open Source code through? 8-). The reason is clearly that Linux is capable of grabbing free press exposure for IBM. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 17:10:44 2001 Delivered-To: freebsd-fs@freebsd.org Received: from hagbard.io.com (hagbard.io.com [199.170.88.13]) by hub.freebsd.org (Postfix) with ESMTP id A774A37B401; Tue, 18 Sep 2001 17:10:41 -0700 (PDT) Received: from localhost (andrewl@localhost) by hagbard.io.com (8.11.2/8.11.2) with ESMTP id f8J0AdU28252; Tue, 18 Sep 2001 19:10:39 -0500 X-Authentication-Warning: hagbard.io.com: andrewl owned process doing -bs Date: Tue, 18 Sep 2001 19:10:39 -0500 (CDT) From: "Andrew P. Lentvorski" To: , Cc: Subject: NFS partial-file lock testing patch submitted on PR bin/30661 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org All, I have submitted a patch to the rpc.lockd to do partial-file lock testing as PR bin/30661. I'd really like someone to look over it and either commit it or explain to me why it shouldn't be committed (ie. what I should change/did wrong) before I embark upon fixing the unlock semantics to do partial file unlocking (Partial file locking is the most painful part. I'm saving it for last.) Could someone let me know if new NFS stuff has propagated to the cvsup mirrors? I'll happily download the new stuff and bounce the Connectathon 2001 suites off it to see if anything breaks. Also, if anyone has any good regression suites for NFS lock testing, I'd love to try them. Thanks, Andy Lentvorski To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Sep 18 17:11:36 2001 Delivered-To: freebsd-fs@freebsd.org Received: from hagbard.io.com (hagbard.io.com [199.170.88.13]) by hub.freebsd.org (Postfix) with ESMTP id 099F237B407 for ; Tue, 18 Sep 2001 17:11:34 -0700 (PDT) Received: from localhost (andrewl@localhost) by hagbard.io.com (8.11.2/8.11.2) with ESMTP id f8J0BWH28259 for ; Tue, 18 Sep 2001 19:11:32 -0500 X-Authentication-Warning: hagbard.io.com: andrewl owned process doing -bs Date: Tue, 18 Sep 2001 19:11:32 -0500 (CDT) From: "Andrew P. Lentvorski" To: Subject: NFS partial-file lock testing patch submitted on PR bin/30661 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org All, I have submitted a patch to the rpc.lockd to do partial-file lock testing as PR bin/30661. I'd really like someone to look over it and either commit it or explain to me why it shouldn't be committed (ie. what I should change/did wrong) before I embark upon fixing the unlock semantics to do partial file unlocking (Partial file locking is the most painful part. I'm saving it for last.) Could someone let me know if new NFS stuff has propagated to the cvsup mirrors? I'll happily download the new stuff and bounce the Connectathon 2001 suites off it to see if anything breaks. Also, if anyone has any good regression suites for NFS lock testing, I'd love to try them. Thanks, Andy Lentvorski To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed Sep 19 3:40:36 2001 Delivered-To: freebsd-fs@freebsd.org Received: from gull.mail.pas.earthlink.net (gull.mail.pas.earthlink.net [207.217.121.85]) by hub.freebsd.org (Postfix) with ESMTP id 6DBFE37B40C for ; Wed, 19 Sep 2001 03:40:34 -0700 (PDT) Received: from mindspring.com (dialup-209.245.133.197.Dial1.SanJose1.Level3.net [209.245.133.197]) by gull.mail.pas.earthlink.net (EL-8_9_3_3/8.9.3) with ESMTP id DAA22628; Wed, 19 Sep 2001 03:40:25 -0700 (PDT) Message-ID: <3BA87648.EFBADE8E@mindspring.com> Date: Wed, 19 Sep 2001 03:41:12 -0700 From: Terry Lambert Reply-To: tlambert2@mindspring.com X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: "Andrew P. Lentvorski" Cc: freebsd-fs@freebsd.org Subject: Re: NFS partial-file lock testing patch submitted on PR bin/30661 References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org "Andrew P. Lentvorski" wrote: > > Also, if anyone has any good regression suites for NFS lock testing, I'd > love to try them. You may want to try running NIST PCTS (National Institute of Standards and Technology POSIX Conformance Test Suite). It has a small amount of lock testing code. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message