From owner-freebsd-fs Fri May 14 20:27:43 1999 Delivered-To: freebsd-fs@freebsd.org Received: from ohm.ingsala.unal.edu.co (ohm.ingsala.unal.edu.co [168.176.15.18]) by hub.freebsd.org (Postfix) with ESMTP id 7E31E14D69 for ; Fri, 14 May 1999 20:26:26 -0700 (PDT) (envelope-from pfgiffun@bachue.usc.unal.edu.co) Received: from bachue.usc.unal.edu.co (unalmodem01.usc.unal.edu.co [168.176.3.31]) by ohm.ingsala.unal.edu.co (8.8.5/8.8.5) with ESMTP id WAA02988 for ; Fri, 14 May 1999 22:28:16 -0500 (COT) Message-ID: <373CE936.356A29D6@bachue.usc.unal.edu.co> Date: Fri, 14 May 1999 22:25:44 -0500 From: "Pedro F. Giffuni" Reply-To: giffunip@asme.org Organization: U. Nacional de Colombia X-Mailer: Mozilla 4.08 [en] (X11; I; FreeBSD 3.1-RELEASE i386) MIME-Version: 1.0 To: freebsd-fs@FreeBSD.org Subject: FYI: SGI releases OpenVault Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org OpenVaultTM http://www.sgi.com/software/opensource/openvault/ OpenVault is a package of mediation software that helps storage management applications manage removable media. In a Storage Area Network (SAN), OpenVault arbitrates access to removable media devices by allocating & deallocating these devices to various applications. Following are some tertiary storage applications that can benefit from OpenVault: *Tape access, such as tar or cpio *Backup, to guard against data loss * Archive, for long-term storage of unused data * Hierarchical storage management, to automatically stage data to the most cost-effective medium *CD-ROM jukeboxes or information libraries *Broadcast libraries containing videotapes SGI is releasing the OpenVault source code under a public source license. This will bring key media management technology to other platforms running Linux, BSD, and other popular operating systems. The release is targeted at storage management software developers that are enhancing their software to manage removable media in Storage Area Networks. OpenVault solves the problems of traditional media management by providing middleware that raises the level of abstraction, enabling rapid deployment of removable media libraries, drives, systems, and client applications. *Manufacturers of robotic libraries and drives can develop a single OpenVault interface to support a multitude of system platforms *Software vendors can integrate new robotic libraries and drives quickly, eliminating product support delays *Computer system providers can offer a complete range of robotic libraries and storage applications when customers want them *Users and administrators gain access to removable media libraries and application sharing becomes possible _______________________ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed May 19 16:20:24 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cygnus.rush.net (cygnus.rush.net [209.45.245.133]) by hub.freebsd.org (Postfix) with ESMTP id 8FA1315164 for ; Wed, 19 May 1999 16:20:14 -0700 (PDT) (envelope-from bright@rush.net) Received: from localhost (bright@localhost) by cygnus.rush.net (8.9.3/8.9.3) with SMTP id SAA20285 for ; Wed, 19 May 1999 18:45:31 -0500 (EST) Date: Wed, 19 May 1999 18:45:29 -0500 (EST) From: Alfred Perlstein To: freebsd-fs@freebsd.org Subject: ufs/ffs interaction? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Does the code in src/sys/ufs/ufs/ layer itself on top of all filesystems? ie, UFS == naming scheme wereas msdosfs,ffs,ntfs are just called by the ufs code? for instance in src/sys/ufs/ufs/ufs_vnops.c line 1628: error = VOP_READ(ap->a_vp, uio, 0, ap->a_cred); depending on the filesystem associated with the vnode, we can call the msdosfs_read(ap) (line 545) function? thanks, -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed May 19 17: 7:45 1999 Delivered-To: freebsd-fs@freebsd.org Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (Postfix) with ESMTP id 719651520C for ; Wed, 19 May 1999 17:07:40 -0700 (PDT) (envelope-from tlambert@usr04.primenet.com) Received: (from daemon@localhost) by smtp02.primenet.com (8.8.8/8.8.8) id RAA20429; Wed, 19 May 1999 17:07:39 -0700 (MST) Received: from usr04.primenet.com(206.165.6.204) via SMTP by smtp02.primenet.com, id smtpd020340; Wed May 19 17:07:30 1999 Received: (from tlambert@localhost) by usr04.primenet.com (8.8.5/8.8.5) id RAA14098; Wed, 19 May 1999 17:07:26 -0700 (MST) From: Terry Lambert Message-Id: <199905200007.RAA14098@usr04.primenet.com> Subject: Re: ufs/ffs interaction? To: bright@rush.net (Alfred Perlstein) Date: Thu, 20 May 1999 00:07:26 +0000 (GMT) Cc: freebsd-fs@FreeBSD.ORG In-Reply-To: from "Alfred Perlstein" at May 19, 99 06:45:29 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Does the code in src/sys/ufs/ufs/ layer itself on top of all > filesystems? ie, UFS == naming scheme wereas msdosfs,ffs,ntfs > are just called by the ufs code? > > for instance in src/sys/ufs/ufs/ufs_vnops.c line 1628: > > error = VOP_READ(ap->a_vp, uio, 0, ap->a_cred); > > depending on the filesystem associated with the vnode, we can call the > > msdosfs_read(ap) (line 545) > > function? It's true that whatever FS is associated with the vnode will have it's VOP_READ vector called; this could be a different FS, or the the FS itself. The code path you took to get in to that point, however, is really the determining factor. The UFS code provides directory management for the MFS, FFS, LFS, and EXT2FS file systems. In general, the layout and flexname implementation that these FS's have is similar enough that they can share code. The implementation of naming, seperate from a flat numeric block namespace, or a flat numeric inode namespace representing a variable granularity block store, etc., is actually logically more seperate than the current relationship between the UFS and other code would have you believe. There are a number of layering issues where the interfaces are not sufficiently abstract to allow you to implement a totally seperate repacement for UFS, but these are rather trivial to clean up (but without something much more complicated that requires this cleanup, you are unlikely to be able to defend the changes against claims that you are merely editorializing, and that your changes are unnecessary). About three years ago, I did the cleanups and implemented a btree for storing directory contents (I replaced the UFS layer). That change itself speeds up directory lookups significantly, but is most valuable when you have file name globbing in the kernel instead of user space, since it allows you to push only matching data back across the user/kernel boundary. For a NetWare server (as an example), where the client will ask first for directories, then for files, this significantly reduces the work in the seperate request necessitated first and second pass over the data. For the first pass, you push back only directories, and for the second, you push back only files. This results in a factor of three speedup with a directory containing a 5:1 file:directory ratio, with an average of 20uS per user/kernel boundary crossing, and more if the globbing eliminates some files or directories. In any case, you should view the UFS layer as implementing directory management above several layers, and the fact that it is being directly referenced by the FFS code as a design compromise having to do with dynamic linking capabilities at the time it was implemented. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sat May 22 13:32:40 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cygnus.rush.net (cygnus.rush.net [209.45.245.133]) by hub.freebsd.org (Postfix) with ESMTP id 222F614D92 for ; Sat, 22 May 1999 13:32:36 -0700 (PDT) (envelope-from bright@rush.net) Received: from localhost (bright@localhost) by cygnus.rush.net (8.9.3/8.9.3) with SMTP id PAA02361 for ; Sat, 22 May 1999 15:51:28 -0500 (EST) Date: Sat, 22 May 1999 15:51:26 -0500 (EST) From: Alfred Perlstein To: fs@freebsd.org Subject: reference for filesystem with btree layout? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org anyone have a url to a discussion about such a creation? I don't much care for the sequential layout of directories in FFS and was interested in alternatives. also, specifically in FreeBSD, I don't see a reason why a filesystem can't take ownership of multiple block devices, is there a reason? thanks, -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 1:56:44 1999 Delivered-To: freebsd-fs@freebsd.org Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by hub.freebsd.org (Postfix) with ESMTP id 03B3B14D26 for ; Mon, 24 May 1999 01:56:38 -0700 (PDT) (envelope-from narvi@haldjas.folklore.ee) Received: from haldjas.folklore.ee (haldjas.folklore.ee [172.17.2.1] (may be forged)) by haldjas.folklore.ee (8.8.8/8.8.4) with SMTP id LAA13338; Mon, 24 May 1999 11:56:21 +0300 (EEST) Date: Mon, 24 May 1999 11:56:21 +0300 (EEST) From: Narvi To: Alfred Perlstein Cc: fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 22 May 1999, Alfred Perlstein wrote: > > anyone have a url to a discussion about such a creation? > No. > I don't much care for the sequential layout of directories in FFS > and was interested in alternatives. > Presently ffs uses ufs code for directory related operations. It should be possible (and not too hard at my reading of it back when I had a similar idea) to replace them. > also, specifically in FreeBSD, I don't see a reason why a filesystem can't > take ownership of multiple block devices, is there a reason? > What's the need? You can stripe the block devices with ccd (or concat with vinum) and get the same result. > thanks, > -Alfred > Sander There is no love, no good, no happiness and no future - all these are just illusions. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 2: 0:27 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cygnus.rush.net (cygnus.rush.net [209.45.245.133]) by hub.freebsd.org (Postfix) with ESMTP id 79F1314E07 for ; Mon, 24 May 1999 02:00:24 -0700 (PDT) (envelope-from bright@rush.net) Received: from localhost (bright@localhost) by cygnus.rush.net (8.9.3/8.9.3) with SMTP id EAA21034; Mon, 24 May 1999 04:19:37 -0500 (EST) Date: Mon, 24 May 1999 04:19:35 -0500 (EST) From: Alfred Perlstein To: Narvi Cc: fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 24 May 1999, Narvi wrote: > > On Sat, 22 May 1999, Alfred Perlstein wrote: > > > > > anyone have a url to a discussion about such a creation? > > > > No. > > > I don't much care for the sequential layout of directories in FFS > > and was interested in alternatives. > > > > Presently ffs uses ufs code for directory related operations. It should be > possible (and not too hard at my reading of it back when I had a similar > idea) to replace them. It would change the on disk structure otherwise there would be a lot of extra IO involved no? > > also, specifically in FreeBSD, I don't see a reason why a filesystem can't > > take ownership of multiple block devices, is there a reason? > > > > What's the need? You can stripe the block devices with ccd (or concat > with vinum) and get the same result. the point would be to at runtime add more backing block devices to expand or migrate the filesystem. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 2:18:38 1999 Delivered-To: freebsd-fs@freebsd.org Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by hub.freebsd.org (Postfix) with ESMTP id AAC28151C9 for ; Mon, 24 May 1999 02:18:33 -0700 (PDT) (envelope-from narvi@haldjas.folklore.ee) Received: from haldjas.folklore.ee (haldjas.folklore.ee [172.17.2.1] (may be forged)) by haldjas.folklore.ee (8.8.8/8.8.4) with SMTP id MAA13744; Mon, 24 May 1999 12:18:29 +0300 (EEST) Date: Mon, 24 May 1999 12:18:29 +0300 (EEST) From: Narvi To: Alfred Perlstein Cc: fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 24 May 1999, Alfred Perlstein wrote: > On Mon, 24 May 1999, Narvi wrote: > > > > > Presently ffs uses ufs code for directory related operations. It should be > > possible (and not too hard at my reading of it back when I had a similar > > idea) to replace them. > > It would change the on disk structure otherwise there would be a lot > of extra IO involved no? > The way I thought of it, it would have been a btree inside an "ordinary" directory file. The on disk directory structure would have two additional fields, right & left which point to the offsets of respective lower brances in the tree. The tree is kept Balanced. But I guess that is just one way. There will be slightly more I/O. > > > also, specifically in FreeBSD, I don't see a reason why a filesystem can't > > > take ownership of multiple block devices, is there a reason? > > > > > > > What's the need? You can stripe the block devices with ccd (or concat > > with vinum) and get the same result. > > the point would be to at runtime add more backing block devices > to expand or migrate the filesystem. > That would require more extensive re-engineering of the FS. > -Alfred > Sander There is no love, no good, no happiness and no future - all these are just illusions. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 4: 6:54 1999 Delivered-To: freebsd-fs@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.8.8]) by hub.freebsd.org (Postfix) with ESMTP id 74CBC14CB9 for ; Mon, 24 May 1999 04:06:48 -0700 (PDT) (envelope-from ticso@cicely8.cicely.de) Received: from cicely7.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.8.6/8.8.6) with ESMTP id NAA11030; Mon, 24 May 1999 13:05:17 +0200 (MET DST) Received: from cicely8.cicely.de (cicely8.cicely.de [10.1.2.10]) by cicely7.cicely.de (8.9.0/8.9.0) with ESMTP id NAA27060; Mon, 24 May 1999 13:06:04 +0200 (CEST) Received: (from ticso@localhost) by cicely8.cicely.de (8.9.3/8.9.2) id NAA54357; Mon, 24 May 1999 13:08:05 +0200 (CEST) (envelope-from ticso) Date: Mon, 24 May 1999 13:08:04 +0200 From: Bernd Walter To: Alfred Perlstein Cc: Narvi , fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? Message-ID: <19990524130804.A53899@cicely8.cicely.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.3i In-Reply-To: ; from Alfred Perlstein on Mon, May 24, 1999 at 04:19:35AM -0500 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, May 24, 1999 at 04:19:35AM -0500, Alfred Perlstein wrote: > On Mon, 24 May 1999, Narvi wrote: > > > > > On Sat, 22 May 1999, Alfred Perlstein wrote: > > > > > also, specifically in FreeBSD, I don't see a reason why a filesystem can't > > > take ownership of multiple block devices, is there a reason? > > > > > > > What's the need? You can stripe the block devices with ccd (or concat > > with vinum) and get the same result. > > the point would be to at runtime add more backing block devices > to expand or migrate the filesystem. I'm currently developing a growfs tool which is able to increase the size of a FFS filesystem. You simply increase the partitionsize and run growfs on it. Geting a big enough partition is still the job of vinum, ccd or big HDDs. At this moment my implementation only shows that it might be working some day. It still contains at least one principle problem, several bugs and some work on handling the previously last cylinder group correctly. After the programm is working the next step will be to get it and depending programs working without the need to umount the partition before increasing. Solaris 2.x with MD installed shows that it is possible to do. Decreasing a FS is generaly a problematic thing to do, because the need to move the filedata from the cut-away part. This will change the inode-numbers of that files and invalidades all references to it. Unlukily even NFS-Clients references inodes anyhow. What else do you expect from a multipartition FS? -- B.Walter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 4:12:16 1999 Delivered-To: freebsd-fs@freebsd.org Received: from phk.freebsd.dk (phk.freebsd.dk [212.242.40.153]) by hub.freebsd.org (Postfix) with ESMTP id B823514CB9 for ; Mon, 24 May 1999 04:12:13 -0700 (PDT) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (critter.freebsd.dk [212.242.40.131]) by phk.freebsd.dk (8.9.1/8.8.8) with ESMTP id NAA04664; Mon, 24 May 1999 13:12:11 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.9.3/8.9.2) with ESMTP id NAA12363; Mon, 24 May 1999 13:12:05 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Bernd Walter Cc: Alfred Perlstein , Narvi , fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-reply-to: Your message of "Mon, 24 May 1999 13:08:04 +0200." <19990524130804.A53899@cicely8.cicely.de> Date: Mon, 24 May 1999 13:12:05 +0200 Message-ID: <12361.927544325@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Multi spindle is very interesting for LFS use. -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." FreeBSD -- It will take a long time before progress goes too far! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 6:53: 0 1999 Delivered-To: freebsd-fs@freebsd.org Received: from sasami.jurai.net (sasami.jurai.net [63.67.141.99]) by hub.freebsd.org (Postfix) with ESMTP id E6F4D1537E for ; Mon, 24 May 1999 06:52:58 -0700 (PDT) (envelope-from winter@jurai.net) Received: from localhost (winter@localhost) by sasami.jurai.net (8.8.8/8.8.7) with ESMTP id JAA01498; Mon, 24 May 1999 09:49:52 -0400 (EDT) Date: Mon, 24 May 1999 09:49:52 -0400 (EDT) From: "Matthew N. Dodd" To: Poul-Henning Kamp Cc: fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-Reply-To: <12361.927544325@critter.freebsd.dk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 24 May 1999, Poul-Henning Kamp wrote: > Multi spindle is very interesting for LFS use. How about a filesystem that stores its meta-data in 2 places? This would make a hierarchical filesystem a little easier to do. -- | Matthew N. Dodd | 78 280Z | 75 164E | 84 245DL | FreeBSD/NetBSD/Sprite/VMS | | winter@jurai.net | This Space For Rent | ix86,sparc,m68k,pmax,vax | | http://www.jurai.net/~winter | Are you k-rad elite enough for my webpage? | To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 9: 1:43 1999 Delivered-To: freebsd-fs@freebsd.org Received: from sss00205.schwab.com (sss00205.schwab.com [162.93.15.188]) by hub.freebsd.org (Postfix) with ESMTP id 39C4115453 for ; Mon, 24 May 1999 09:01:40 -0700 (PDT) (envelope-from Andreas.Pleschutznig@Schwab.COM) Received: (from root@localhost) by fire sf id LAA13717 for ; Mon, 24 May 1999 11:57:56 -0400 (EDT) Received: from s0001asf.schwab.com(162.93.32.201) by sss00205 via smap (V2.1/2.1+anti-relay+anti-spam) id xma013700; Mon, 24 May 99 11:57:37 -0400 Received: (from root@localhost) by ihop sf id MAA21150 for fs@freebsd.org; Mon, 24 May 1999 12:01:16 -0400 (EDT) Received: ( Schwab Email ) by copymail sf with ESMTP id MAA21130 for ; Mon, 24 May 1999 12:01:16 -0400 (EDT) Received: by n2101pmx.cdc.schwab.com with Internet Mail Service (5.5.2448.0) id ; Mon, 24 May 1999 11:58:08 -0400 Message-ID: <11585F032846CF11867900805FE2A57706D615FC@N1002SMX.nt.schwab.com> From: "Pleschutznig, Andreas" To: "'fs@freebsd.org'" Subject: Journaled FS? Date: Mon, 24 May 1999 11:57:55 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Content-Type: multipart/mixed; boundary="----_=_NextPart_000_01BEA5FE.36E53FA4" Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_000_01BEA5FE.36E53FA4 Content-Type: text/plain; charset="iso-8859-1" Hi everyone, The other day I was wondering if there is already available, or somewhere in work a journaled filesystem like vxfs or IBM's AIX jfs? -- Andreas Pleschutznig Sr. Unix System Admin E-mail: Andreas.Pleschutznig@Schwab.com Phone: (415) 636 0493 Cell: (415) 850 7996 Brick: 260 Pager: http://www.nextel.com/paging/indivpage.html Warning: All e-mail sent to this address will be received by the Charles Schwab & Co., Inc. corporate e-mail system and is subject to archival and review by someone other than the recipient. ------_=_NextPart_000_01BEA5FE.36E53FA4 Content-Type: application/octet-stream; name="Andreas Pleschutznig (E-mail).vcf" Content-Disposition: attachment; filename="Andreas Pleschutznig (E-mail).vcf" BEGIN:VCARD VERSION:2.1 N:Pleschutznig;Andreas FN:Andreas Pleschutznig (E-mail) ORG:Schwab TITLE:Sr. UNIX System Admin TEL;WORK;VOICE:(415) 636-0493 TEL;HOME;VOICE:(925) 556-1485 TEL;CELL;VOICE:(415) 850-7996 TEL;WORK;FAX:(415) 636-0466 EMAIL;PREF;EX:/o=Schwab/ou=SFO/cn=Recipients/cn=95428 REV:19990514T181230Z END:VCARD ------_=_NextPart_000_01BEA5FE.36E53FA4-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 10:12:43 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20]) by hub.freebsd.org (Postfix) with ESMTP id DC55614BD4 for ; Mon, 24 May 1999 10:12:39 -0700 (PDT) (envelope-from ezk@shekel.mcl.cs.columbia.edu) Received: from shekel.mcl.cs.columbia.edu (shekel.mcl.cs.columbia.edu [128.59.18.15]) by cs.columbia.edu (8.9.1/8.9.1) with ESMTP id NAA23005; Mon, 24 May 1999 13:12:38 -0400 (EDT) Received: (from ezk@localhost) by shekel.mcl.cs.columbia.edu (8.9.1/8.9.1) id NAA22724; Mon, 24 May 1999 13:12:37 -0400 (EDT) Date: Mon, 24 May 1999 13:12:37 -0400 (EDT) Message-Id: <199905241712.NAA22724@shekel.mcl.cs.columbia.edu> X-Authentication-Warning: shekel.mcl.cs.columbia.edu: ezk set sender to ezk@shekel.mcl.cs.columbia.edu using -f From: Erez Zadok To: Alfred Perlstein Cc: fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-reply-to: Your message of "Sat, 22 May 1999 15:51:26 CDT." Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message , Alfred Perlstein writes: > > anyone have a url to a discussion about such a creation? Maybe this'll help. Look for a Usenix paper on XFS, SGI's f/s. It was published in '96 in their technical conf. proc. http://www.usenix.org/publications/library/proceedings/sd96/sweeney.html Check out Reiserfs, a f/s for linux that does that: http://www.idiom.com/~beverly/reiserfs.html You might also look at my Usenetfs paper, a simple stackable f/s that restructures large dirs: http://www.cs.columbia.edu/~ezk/research/ Erez. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 13:54:16 1999 Delivered-To: freebsd-fs@freebsd.org Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (Postfix) with ESMTP id C2E9B15399 for ; Mon, 24 May 1999 13:54:11 -0700 (PDT) (envelope-from grog@freebie.lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.9.1/8.9.0) with ESMTP id GAA04321; Tue, 25 May 1999 06:24:10 +0930 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.9.3/8.9.0) id GAA95321; Tue, 25 May 1999 06:24:08 +0930 (CST) Date: Tue, 25 May 1999 06:24:06 +0930 From: Greg Lehey To: Erez Zadok Cc: Alfred Perlstein , fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? Message-ID: <19990525062405.A95201@freebie.lemis.com> References: <199905241712.NAA22724@shekel.mcl.cs.columbia.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <199905241712.NAA22724@shekel.mcl.cs.columbia.edu>; from Erez Zadok on Mon, May 24, 1999 at 01:12:37PM -0400 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Monday, 24 May 1999 at 13:12:37 -0400, Erez Zadok wrote: > In message , Alfred Perlstein writes: >> >> anyone have a url to a discussion about such a creation? > > Maybe this'll help. > > Look for a Usenix paper on XFS, SGI's f/s. It was published in '96 in their > technical conf. proc. > http://www.usenix.org/publications/library/proceedings/sd96/sweeney.html More to the point is this one: http://www.sgi.com/newsroom/press_releases/1999/may/xfs.html SGI is releasing the source of XFS. This looks to be by far the most interesting one. Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 17:35:18 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cygnus.rush.net (cygnus.rush.net [209.45.245.133]) by hub.freebsd.org (Postfix) with ESMTP id 70E3714EC9 for ; Mon, 24 May 1999 17:35:15 -0700 (PDT) (envelope-from bright@rush.net) Received: from localhost (bright@localhost) by cygnus.rush.net (8.9.3/8.9.3) with SMTP id TAA16523 for ; Mon, 24 May 1999 19:54:51 -0500 (EST) Date: Mon, 24 May 1999 19:54:49 -0500 (EST) From: Alfred Perlstein To: fs@freebsd.org Subject: how are partial writes handled in FFS? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org considering a (dumb) process that writes to a file calling write() several times. The writes are done to a buffer, but let's say the buffer is uncached, what exactly happens? does: A) it fetch the block off disk the write the data into it? b) a block is allocated for the data and the partial write is done to it maintaining the offset of the write, the buffer is marked so that before it is sync'd back to disk (or any reading from a location not within the partial write) that a buffer must be allocated, and the on disk data must be read into the buffer, then merged into the partially written buffer. If it is B, or if sometimes B is a possiblity, can someone show me where the code that does this is located? thanks, -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon May 24 23:30:44 1999 Delivered-To: freebsd-fs@freebsd.org Received: from cs.columbia.edu (cs.columbia.edu [128.59.16.20]) by hub.freebsd.org (Postfix) with ESMTP id 2696115625 for ; Mon, 24 May 1999 23:30:41 -0700 (PDT) (envelope-from ezk@shekel.mcl.cs.columbia.edu) Received: from shekel.mcl.cs.columbia.edu (shekel.mcl.cs.columbia.edu [128.59.18.15]) by cs.columbia.edu (8.9.1/8.9.1) with ESMTP id CAA28821; Tue, 25 May 1999 02:30:39 -0400 (EDT) Received: (from ezk@localhost) by shekel.mcl.cs.columbia.edu (8.9.1/8.9.1) id CAA07446; Tue, 25 May 1999 02:30:39 -0400 (EDT) Date: Tue, 25 May 1999 02:30:39 -0400 (EDT) Message-Id: <199905250630.CAA07446@shekel.mcl.cs.columbia.edu> X-Authentication-Warning: shekel.mcl.cs.columbia.edu: ezk set sender to ezk@shekel.mcl.cs.columbia.edu using -f From: Erez Zadok To: Greg Lehey Cc: Erez Zadok , Alfred Perlstein , fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? In-reply-to: Your message of "Tue, 25 May 1999 06:24:06 +0930." <19990525062405.A95201@freebie.lemis.com> Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <19990525062405.A95201@freebie.lemis.com>, Greg Lehey writes: > On Monday, 24 May 1999 at 13:12:37 -0400, Erez Zadok wrote: > > In message , Alfred Perlstein writes: [...] > SGI is releasing the source of XFS. This looks to be by far the most > interesting one. I just returned from LinuxExpo and this was discussed there. It doesn't appear the sources will be available soon. Someone from SGI said at least a few months: they need to cleanup any encumberances and figure out the licensing. Nevertheless, the linux community was excited about it, but still going ahead with plans for ext3fs. > Greg Erez. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue May 25 0: 0:13 1999 Delivered-To: freebsd-fs@freebsd.org Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (Postfix) with ESMTP id F33B415518 for ; Mon, 24 May 1999 23:59:58 -0700 (PDT) (envelope-from grog@freebie.lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.9.1/8.9.0) with ESMTP id QAA06264; Tue, 25 May 1999 16:29:55 +0930 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.9.3/8.9.0) id QAA97048; Tue, 25 May 1999 16:29:54 +0930 (CST) Date: Tue, 25 May 1999 16:29:53 +0930 From: Greg Lehey To: Erez Zadok Cc: Alfred Perlstein , fs@FreeBSD.ORG Subject: Re: reference for filesystem with btree layout? Message-ID: <19990525162953.J95827@freebie.lemis.com> References: <19990525062405.A95201@freebie.lemis.com> <199905250630.CAA07446@shekel.mcl.cs.columbia.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <199905250630.CAA07446@shekel.mcl.cs.columbia.edu>; from Erez Zadok on Tue, May 25, 1999 at 02:30:39AM -0400 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tuesday, 25 May 1999 at 2:30:39 -0400, Erez Zadok wrote: > In message <19990525062405.A95201@freebie.lemis.com>, Greg Lehey writes: >> On Monday, 24 May 1999 at 13:12:37 -0400, Erez Zadok wrote: >>> In message , Alfred Perlstein writes: > [...] >> SGI is releasing the source of XFS. This looks to be by far the most >> interesting one. > > I just returned from LinuxExpo and this was discussed there. It doesn't > appear the sources will be available soon. Someone from SGI said at least a > few months: they need to cleanup any encumberances and figure out the > licensing. That sounds like a short period of time to me. > Nevertheless, the linux community was excited about it, but still > going ahead with plans for ext3fs. I think that's wise. There's always room for diversity. Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue May 25 0:58:39 1999 Delivered-To: freebsd-fs@freebsd.org Received: from peik.bol.sharif.ac.ir (unknown [194.225.42.33]) by hub.freebsd.org (Postfix) with ESMTP id 9AC021543A for ; Tue, 25 May 1999 00:57:56 -0700 (PDT) (envelope-from farshid@peik.bol.sharif.ac.ir) Received: from localhost (farshid@localhost) by peik.bol.sharif.ac.ir (8.8.7/8.8.7) with SMTP id MAA00657; Tue, 25 May 1999 12:31:09 +0430 Date: Tue, 25 May 1999 12:31:09 +0430 (IRST) From: Farshid Eslami To: "Pleschutznig, Andreas" Cc: "'fs@freebsd.org'" Subject: Re: Journaled FS? In-Reply-To: <11585F032846CF11867900805FE2A57706D615FC@N1002SMX.nt.schwab.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 24 May 1999, Pleschutznig, Andreas wrote: Have a look at http://www.linuxworld.com/linuxworld/lw-1999-05/lw-05-sgi.html as you may know, XFS is a 64-bit and journaling file system. --Farshid : Hi everyone, : : The other day I was wondering if there is already available, or somewhere in : work a journaled filesystem like vxfs or IBM's AIX jfs? : : : : -- : Andreas Pleschutznig : Sr. Unix System Admin : E-mail: Andreas.Pleschutznig@Schwab.com : : Phone: (415) 636 0493 Cell: (415) 850 7996 Brick: 260 : Pager: http://www.nextel.com/paging/indivpage.html : : : Warning: All e-mail sent to this address will be received by the Charles : Schwab & Co., Inc. corporate e-mail system and is subject to archival and : review by someone other than the recipient. : : Farshid Eslami farshidoo@yahoo.com ----------------------------------------------------------- Without passion man is a mere latent force and possibility, like the flint which awaits the shock of the iron before it can give forth its spark. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue May 25 1:58:40 1999 Delivered-To: freebsd-fs@freebsd.org Received: from peik.bol.sharif.ac.ir (unknown [194.225.42.33]) by hub.freebsd.org (Postfix) with ESMTP id 6C7D014C93 for ; Tue, 25 May 1999 01:58:07 -0700 (PDT) (envelope-from farshid@peik.bol.sharif.ac.ir) Received: from localhost (farshid@localhost) by peik.bol.sharif.ac.ir (8.8.7/8.8.7) with SMTP id NAA00849; Tue, 25 May 1999 13:29:27 +0430 Date: Tue, 25 May 1999 13:29:27 +0430 (IRST) From: Farshid Eslami To: Alfred Perlstein Cc: fs@FreeBSD.ORG Subject: Re: how are partial writes handled in FFS? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 24 May 1999, Alfred Perlstein wrote: I've not read the FFS code in detail, but as far as I know, if write is not block-aligned, the block should be read from disk and then the write syscall updates the necessary data in block. Otherwise, if the write is block-aligned, there is no need to read the block from disk because all of the block is going to be overwitten. So why should the data that will be overwirtten be read from disk? In this case, there is no need to pay the cost of a read from disk. A block should be assigned from buffer cache to write and write copies the data to the buffer. Again in this case, the content of the buffer is not needed to be zero or any thing else, because all the buffer is overwirtten by write syscall. These are for cached write. For uncached write, it's done through raw device. I don't know exactly how it handles this, but the policy is straight forward. One more point is that if the write in uncached, there is no need to copy data, which size is a multiple of block size, into a separate buffer. There write syscall can send out data directly from user space to device. --Farshid : : considering a (dumb) process that writes to a file calling write() : several times. The writes are done to a buffer, but let's say the : buffer is uncached, what exactly happens? : : does: : A) it fetch the block off disk the write the data into it? : b) a block is allocated for the data and the partial write : is done to it maintaining the offset of the write, the buffer : is marked so that before it : is sync'd back to disk (or any reading from a location not : within the partial write) that a buffer must be allocated, and : the on disk data must be read into the buffer, then merged into : the partially written buffer. : : If it is B, or if sometimes B is a possiblity, can someone show me : where the code that does this is located? : : thanks, : -Alfred : : : : To Unsubscribe: send mail to majordomo@FreeBSD.org : with "unsubscribe freebsd-fs" in the body of the message : Farshid Eslami farshidoo@yahoo.com ----------------------------------------------------------- Without passion man is a mere latent force and possibility, like the flint which awaits the shock of the iron before it can give forth its spark. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue May 25 10: 8: 5 1999 Delivered-To: freebsd-fs@freebsd.org Received: from news.kiev.sovam.com (news.kiev.sovam.com [194.186.143.16]) by hub.freebsd.org (Postfix) with ESMTP id CC01D15019 for ; Tue, 25 May 1999 10:07:56 -0700 (PDT) (envelope-from mail@news.kiev.sovam.com) Received: from mail by news.kiev.sovam.com with local (Exim 2.12 #1) id 10mKgF-0007uM-00 for freebsd-fs@freebsd.org; Tue, 25 May 1999 20:07:55 +0300 From: gw-bsd-fs@news.kiev.sovam.com To: freebsd-fs@freebsd.org Reply-To: Vladimir Litovka Subject: test, ignore this message Date: 25 May 1999 17:07:55 GMT Message-ID: <7ielda$tf8$1@news.kiev.sovam.com> X-Organization: Sovam Teleport Kiev (post does not reflect views of Sovam Teleport) X-Complaints-To: news@news.kiev.sovam.com X-Gated-By: news2list v1.0-b1, (c) Vladimir Litovka X-Gated-Date: Tue May 25 20:07:55 EEST 1999 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm sorry, I'm just testing my gateway to world -- doka To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue May 25 10:58:56 1999 Delivered-To: freebsd-fs@freebsd.org Received: from news.kiev.sovam.com (news.kiev.sovam.com [194.186.143.16]) by hub.freebsd.org (Postfix) with ESMTP id 9FC2914C9D for ; Tue, 25 May 1999 10:58:50 -0700 (PDT) (envelope-from mail@news.kiev.sovam.com) Received: from mail by news.kiev.sovam.com with local (Exim 2.12 #1) id 10mLTT-000825-00 for freebsd-fs@freebsd.org; Tue, 25 May 1999 20:58:47 +0300 From: gw-bsd-fs@news.kiev.sovam.com To: freebsd-fs@freebsd.org Reply-To: Vladimir Litovka Subject: test, ignore this message Date: 25 May 1999 17:58:47 GMT Message-ID: <7ieocn$u2j$1@news.kiev.sovam.com> X-Organization: Sovam Teleport Kiev (post does not reflect views of Sovam Teleport) X-Complaints-To: news@news.kiev.sovam.com X-Gated-By: news2list v1.0-b1, (c) Vladimir Litovka X-Gated-Date: Tue May 25 20:58:47 EEST 1999 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Do not respond, please I'm sorry, still can't get freebsd-fs in my mbox -- doka To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed Jun 2 14:25:26 1999 Delivered-To: freebsd-fs@freebsd.org Received: from sol (cs1-gw.cs.binghamton.edu [128.226.171.72]) by hub.freebsd.org (Postfix) with SMTP id 26F2D15948 for ; Wed, 2 Jun 1999 14:24:24 -0700 (PDT) (envelope-from zzhang@cs.binghamton.edu) Received: from localhost (zzhang@localhost) by sol (SMI-8.6/8.6.9) with SMTP id RAA26493 for ; Wed, 2 Jun 1999 17:13:02 -0400 Date: Wed, 2 Jun 1999 17:13:01 -0400 (EDT) From: Zhihui Zhang To: freebsd-fs@freebsd.org Subject: allocate file blocks contiguously Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In FFS, there is a parameter called maxcontig (default to 16) that determines the number of blocks we can allocate contiguously for a single file. What is its optimal value? I mean, if we allocate ALL the data blocks of a very big file contiguously, will its I/O performance be improved greatly? It seems to me this number may also be limited by system buffering capability (MAXPHYS?) and underlying hardware controller. Can anyone give me some hints on the choice of the value of maxcontig? Any help is appreciated. -------------------------------------------------- Zhihui Zhang. Please visit http://www.freebsd.org -------------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sat Jun 5 16:46:37 1999 Delivered-To: freebsd-fs@freebsd.org Received: from mail1.noc.netcom.net (mail1.noc.netcom.net [204.31.1.150]) by hub.freebsd.org (Postfix) with ESMTP id E51E6151FE for ; Sat, 5 Jun 1999 16:46:35 -0700 (PDT) (envelope-from kprater@industrysearch.com) Received: from webserver.industrysearch.com (webserver.industrysearch.com [207.95.128.10]) by mail1.noc.netcom.net (8.9.1/8.9.1/(NETCOM v2.00)) with ESMTP id QAA15958 for ; Sat, 5 Jun 1999 16:46:33 -0700 (PDT) Message-Id: <199906052346.QAA15958@mail1.noc.netcom.net> Received: from PB-S607 by webserver.industrysearch.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.0.1458.49) id L9MBWCW0; Sat, 5 Jun 1999 19:38:07 -0400 To: freebsd-fs@FreeBSD.ORG From: "Ken Prater, IndustrySearch.Com" Date: Sat, 5 Jun 99 19:47:45 +0000 Subject: "Link Your Web Site" To IndustrySearch.Com X-Mailer: WM - IndustrySearch.Com Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Increase traffic to your company's web site with a FREE Hyperlink to IndustrySearch.Com. Thousands of industrial purchasing agents, buyers, engineers and others searching for suppliers and services can locate your business easily with our USA Industrial Directory. You can visit IndustrySearch.Com at http://industrysearch.com "Link Your Web Site" to our USA Industrial Directory Data Base today! Visit IndustrySearch.Com at http://industrysearch.com and click on "Link Your Web Site" Thank you, K. Prater USA INDUSTRIAL DATA BASE MANAGEMENT To be removed from our mailing list, please click Reply and type "REMOVE" in the subject field To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message