From Harri.Poyhonen@cs.Helsinki.FI Tue Mar 28 01:36:38 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Tue, 28 Mar 95 01:36:32 -0600; AA22110 Received: from kypros.Helsinki.FI by lucy.cs.wisc.edu; Tue, 28 Mar 95 01:36:28 -0600 Received: (hipoyhon@localhost) by kypros.Helsinki.FI (8.6.9/H48) id KAA22950 for exodus_all@cs.wisc.edu; Tue, 28 Mar 1995 10:36:23 +0300 From: Harri Poyhonen Message-Id: <199503280736.KAA22950@kypros.Helsinki.FI> Subject: compiling Exodus SM 3.1 on Linux? To: exodus_all@lucy.cs.wisc.edu Date: Tue, 28 Mar 1995 10:36:22 +0300 (EET DST) X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 715 Hi, I'm currently implementing ODMG-95 -based OODBMS in a university project. We are using Exodus as a platform and tried to compile the Exodus sm 3.1 in a Linux box using the Linux patch obtained from ftp.cs.wisc.edu (sm.3.1.linux.patch.tar.Z). Some problems occurred and subsequent reading of sm documents revelealed that the sm compilation doesn't work with GNU make. Question: how does one get sm 3.1 compiled under Linux? After all, GNU make is pretty much the standard make in Linux environment. Thanks, Harri Poyhonen -- a) REFETY ROWSETY ROWWSETY NONROW slice{860a} : weak REFETY ROWS ROWWSETY NONROW primary{81d}, sub symbol{31e}, ROWS leaving ROWSETY indexer{b,c,d,e}, bus symbol{31e}. ----------------------------------------------------------------------- REPLIES TO THIS MESSAGE WILL GO ONLY TO THE SENDER Requests to get on or off exodus_all should go to listproc@cs.wisc.edu. For more information on this list, including instructions on how to (un)subscribe and get digests, send mail to listproc@cs.wisc.edu with the message: information exodus_all For more information about listproc options, send the message: help ----------------------------------------------------------------------- From summersms@ornl.gov Tue Mar 28 13:45:37 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Tue, 28 Mar 95 13:45:31 -0600; AA11835 Received: from cosmail2.ctd.ornl.gov by lucy.cs.wisc.edu; Tue, 28 Mar 95 13:45:28 -0600 Received: from x7u (x7u.ed.ornl.gov [128.219.40.248]) by cosmail2.ctd.ornl.gov (8.6.10/8.6.9) with SMTP id OAA22401 for ; Tue, 28 Mar 1995 14:45:04 -0500 Date: Tue, 28 Mar 1995 14:45:04 -0500 Message-Id: <199503281945.OAA22401@cosmail2.ctd.ornl.gov> X-Sender: x7u@cosmail2.ctd.ornl.gov X-Mailer: Windows Eudora Version 2.0.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: exodus_all@lucy.cs.wisc.edu From: summersms@ornl.gov (Michael Summers) Subject: gcc-4.7 Where does one get gcc 2.4.5 (required to install exodus) ? Is the beta release of shore so imminent that I sould be avoiding using exodus/E? Mike Summers ----------------------------------------------------------------------- REPLIES TO THIS MESSAGE WILL GO ONLY TO THE SENDER Requests to get on or off exodus_all should go to listproc@cs.wisc.edu. For more information on this list, including instructions on how to (un)subscribe and get digests, send mail to listproc@cs.wisc.edu with the message: information exodus_all For more information about listproc options, send the message: help ----------------------------------------------------------------------- From zwilling@caseus.cs.wisc.edu Wed Mar 29 10:12:18 1995 Received: from caseus.cs.wisc.edu by sea.cs.wisc.edu; Wed, 29 Mar 95 10:12:11 -0600; AA08881 Date: Wed, 29 Mar 95 10:12:08 -0600 From: zwilling@caseus.cs.wisc.edu (Mike Zwilling) Message-Id: <9503291612.AA06685@caseus.cs.wisc.edu> Received: by caseus.cs.wisc.edu; Wed, 29 Mar 95 10:12:08 -0600 To: summersms@ornl.gov Subject: Re: gcc-4.7 Cc: exodus_all@caseus.cs.wisc.edu >Where does one get gcc 2.4.5 (required to install exodus) ? One thing you can use is the E compiler, eg++, since it will compile the SM. However, you can also use later versions of gcc (but you will have to recompile the SM since binaries generated from older versions are not compatible with those generated from current (2.6.*) ones). You might have to make some minor changes to get it to compile (I've done them, but don't remember what they are). Mail me if you have trouble. >Is the beta release of shore so imminent that I sould be avoiding >using exodus/E? The shore beta release will be out in april (mid-late most likely). Mike ----------------------------------------------------------------------- REPLIES TO THIS MESSAGE WILL GO ONLY TO THE SENDER Requests to get on or off exodus_all should go to listproc@cs.wisc.edu. For more information on this list, including instructions on how to (un)subscribe and get digests, send mail to listproc@cs.wisc.edu with the message: information exodus_all For more information about listproc options, send the message: help ----------------------------------------------------------------------- From pax@idt.unit.no Fri Mar 31 03:52:21 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Fri, 31 Mar 95 03:52:17 -0600; AA18460 Received: from pat.idt.unit.no by lucy.cs.wisc.edu; Fri, 31 Mar 95 03:52:11 -0600 Received: from parkalm.idt.unit.no by pat.idt.unit.no with SMTP id AA21494 (5.65c8/IDA-1.4.4); Fri, 31 Mar 1995 11:52:02 +0200 Received: by parkalm.idt.unit.no id AA09715 (5.0/idt-C-2.0 for idt); Fri, 31 Mar 1995 11:51:47 +0200 Date: Fri, 31 Mar 1995 11:51:45 +0200 (MET DST) From: Per Axel Aamot To: "Exodus Mailing List." Cc: Dan Schuh Subject: Transactions and locking in Exodus/E. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1063 Hello all! I am working (with a friend, Roy Paulsen) on a project where we are trying to implement an object oriented database for video. We are using the Exodus storage manager (version 3.1), and GNU E on top. As far as we know it is not possible to realise transactions on page level (collections ) with E, because when a program calls E_BeginTransaction() all other programs have to wait, even if it wants to access a different (persistent) collection. Is this correct, or have we misunderstood ? We have received information that indicates that it should be possible in Exodus to accomplish transactions on a page or object level. Is this true ? If so, why does not E support this ? Can we bypass this by integrating Exodus lowlevel calls in the E program or will this mean heavy hacking and lots of conflicts ? Replies ASAP will be greatly appreciated as we are in the final phase of our project (it's a small one :) ), and our solution depends on what is possible and what is not. Thanks a lot in advance. Per Axel Aamot (& Roy Paulsen). ----------------------------------------------------------------------- REPLIES TO THIS MESSAGE WILL GO ONLY TO THE SENDER Requests to get on or off exodus_all should go to listproc@cs.wisc.edu. For more information on this list, including instructions on how to (un)subscribe and get digests, send mail to listproc@cs.wisc.edu with the message: information exodus_all For more information about listproc options, send the message: help ----------------------------------------------------------------------- From zwilling@caseus.cs.wisc.edu Fri Mar 31 09:40:52 1995 Received: from caseus.cs.wisc.edu by sea.cs.wisc.edu; Fri, 31 Mar 95 09:40:50 -0600; AA04652 Date: Fri, 31 Mar 95 09:40:49 -0600 From: zwilling@caseus.cs.wisc.edu (Mike Zwilling) Message-Id: <9503311540.AA13283@caseus.cs.wisc.edu> Received: by caseus.cs.wisc.edu; Fri, 31 Mar 95 09:40:49 -0600 To: pax@idt.unit.no Subject: Re: Transactions and locking in Exodus/E. Cc: exodus_all@caseus.cs.wisc.edu >I am working (with a friend, Roy Paulsen) on a project where we are >trying to implement an object oriented database for video. We are >using the Exodus storage manager (version 3.1), and GNU E on top. As >far as we know it is not possible to realise transactions on page level >(collections ) with E, because when a program calls E_BeginTransaction() >all other programs have to wait, even if it wants to access a different >(persistent) collection. Is this correct, or have we misunderstood ? The SM (and therefore E) has page-level granularity locking. So, multiple clients can simultaneously access the same collection. Are you having some specific problem, or did you just misunderstand the documentation? Mike ----------------------------------------------------------------------- REPLIES TO THIS MESSAGE WILL GO ONLY TO THE SENDER Requests to get on or off exodus_all should go to listproc@cs.wisc.edu. For more information on this list, including instructions on how to (un)subscribe and get digests, send mail to listproc@cs.wisc.edu with the message: information exodus_all For more information about listproc options, send the message: help -----------------------------------------------------------------------