- developers-list - monetdb.org

[Monetdb-developers] MT_sleep_ns()?
by Jens Teubner 21 Sep '06

21 Sep '06

Hi all, in shredder.mx there is a call to MT_sleep_ns(). This came with the "new_update" merger. I cannot find, however, any definition of this function. The only thing I can find is MT_sleep_ms() in gdk_posix.mx. Is this a typo? If so, could anybody fix this? This seems to break compilation on Cygwin (at least as reported by our student). Regards Jens -- Jens Teubner Technische Universitaet Muenchen, Department of Informatics D-85748 Garching, Germany Tel: +49 89 289-17259 Fax: +49 89 289-17263 This email was written with 100% recycled electrons.

2 1

[Monetdb-developers] PF: back-merging the "new_updates" branch into the HEAD
by Stefan Manegold 20 Sep '06

20 Sep '06

Dear Pathfinder developers, you might have noticed, that Peter checked in the initial "raw" version of his new revised extended XML/XQuery update implementation into a "new_updates" branch of the pathfinder repository last Monday (Sep 11 2006). Since then, we spent a week on consolidating the code with respect to compilation and testing --- at least on our major development platform Fedora Core 4. We had decided to use a branch, first, so that the consolidation phase would not interfere with other development going on on the development trunk (HEAD). With most of the issues solved, we're now down to only 26 out of 1282 tests failing; see http://monetdb.cwi.nl/testing/projects/monetdb/new_updates.35/ for details. We consider this stable "enough" to merge the "new_updates" branch back into the development trunk (HEAD) A.S.A.P., i.e., preferably tomorrow, in order to (1) avoid further double-lane development and the resulting synchronization overhead/problems, (2) put the new code under the surveillance of automatic *multi-platform* testing, and (3) give more of us the opportunity to test the new code both as such and in combination with their own extensions. In particular, the following parts need to be tested (more) intensively - algebra version: the test-web XMark tests work fine, more hasn't been tested, yet - StandOff: known to fail, code requires adaption to comply with (new) update schema - Tijah: not tested at all, mainly because no tests are available in CVS, yet - the new document management features: hardly (if at all) tested, yet, tests need to be added Unless there are any severe objections by 2 pm (CEST) tomorrow (Tue, Sep 19 2006), we will perform the back-merge. Thank you very much for your cooperation. Kind regards, Stefan -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

1 3

Re: [Monetdb-developers] Monetdb-developers Digest, Vol 4, Issue 7
by Jim Foley 18 Sep '06

18 Sep '06

>------------------------------ > >Message: 2 >Date: Sat, 16 Sep 2006 08:37:24 +0200 >From: Fabian Groffen <Fabian.Groffen(a)cwi.nl> >Subject: Re: [Monetdb-developers] Compiled ok under openSUSE 10.1 (32 bit) but some errors in the tests > >Hi Jim, > >Thanks for your interest in MonetDB/SQL. > >On 15-09-2006 19:09:10 -0400, Jim Foley wrote: >> I freed up a linux server and got MonetDB 4.12.0 and the SQL client to >> compile and run under openSUSE 10.1 with a AMD 3800 64 bit processor. >> It seems to work fine but I am getting some errors on the test for >> both the DB and the SQL client. I understand from the archives that >> some errors are normal. Is there a list of errors on each platform? I >> can just copy off the test results before I compile it with as a 64 >> bit app and compare the results but I would prefer to clean up the >> errors if I can under the 32 bit app before I try it under 64 bits. I >> have a DBA background and I could just be being anal. ;-) >If you want to be sure, post the names of the tests failing. We don't have a list of failing tests. You can also compare >your output to that of our (SuSE) test machines on: >> http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB/.mTests03/ index.html >and http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests03/inde x.html > >I hope this helps. > >Regards > > > ------------------------------ Fabian, Thank you this is exactly what I was looking for. Jim Foley

1 0

[Monetdb-developers] Cygwin: implicit declaration of function closesocket
by Jens Teubner 18 Sep '06

18 Sep '06

Hi all, a student of ours is trying to compile the current CVS snapshot (the tarball from the website) on Windows using Cygwin. He's getting the error stream.mx:1471:implicit declaration of function closesocket. In the attached config.log, it seems like closesocket could be detected (i.e., successfully linked in). Is there maybe some #include missing in stream.mx to declare closesocket()? The attached config.log, by the way, is full of `gcc: unrecognized option `-no-undefined'' warnings. Is this supposed to be like this? Jens -- Jens Teubner Technische Universitaet Muenchen, Department of Informatics D-85748 Garching, Germany Tel: +49 89 289-17259 Fax: +49 89 289-17263 Young man, in mathematics you don't understand things, you just get used to them. -- John von Neumann

3 2

[Monetdb-developers] Your email requires verification verify#0JM82z4hQigRLCxtf4VWAdko5eyIZ1JJ
by monet＠billionairesforbush.com 18 Sep '06

18 Sep '06

The message you sent requires that you verify that you are a real live human being and not a spam source. To complete this verification, simply reply to this message and leave the subject line intact. The headers of the message sent from your address are show below: >From monetdb-developers(a)lists.sourceforge.net Mon Sep 18 06:11:29 2006 Received: from billiona by host.billionairesforbush.com with local-bsmtp (Exim 4.52) id 1GPG5y-0000ov-2R for monet(a)billionairesforbush.com; Mon, 18 Sep 2006 06:11:29 -0400 Received: from localhost by host.billionairesforbush.com with SpamAssassin (version 3.1.3); Mon, 18 Sep 2006 06:11:29 -0400 From: "monetdb-developers(a)lists.sourceforge.net" <monetdb-developers(a)lists.sourceforge.net> To: monet(a)billionairesforbush.com Subject: Best online adult toys store X-Spam-Flag: YES X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on host.billionairesforbush.com X-Spam-Level: ******************************* X-Spam-Status: Yes, score=31.5 required=5.0 tests=BAYES_60,HTML_MESSAGE, HTML_MIME_NO_HTML_TAG,HTML_NONELEMENT_70_80,INFO_TLD,MIME_HTML_ONLY, RCVD_IN_DSBL,RCVD_IN_NJABL_DUL,RCVD_IN_XBL,URIBL_AB_SURBL, URIBL_JP_SURBL,URIBL_OB_SURBL,URIBL_SC_SURBL,URIBL_WS_SURBL autolearn=spam version=3.1.3 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----------=_450E70D1.27AA11B4" Message-Id: <E1GPG5y-0000ov-2R(a)host.billionairesforbush.com> Date: Mon, 18 Sep 2006 06:11:29 -0400

1 0

[Monetdb-developers] Compiled ok under openSUSE 10.1 (32 bit) but some errors in the tests
by Jim Foley 16 Sep '06

16 Sep '06

I freed up a linux server and got MonetDB 4.12.0 and the SQL client to compile and run under openSUSE 10.1 with a AMD 3800 64 bit processor. It seems to work fine but I am getting some errors on the test for both the DB and the SQL client. I understand from the archives that some errors are normal. Is there a list of errors on each platform? I can just copy off the test results before I compile it with as a 64 bit app and compare the results but I would prefer to clean up the errors if I can under the 32 bit app before I try it under 64 bits. I have a DBA background and I could just be being anal. ;-) Thanks. Jim Foley

2 1

[Monetdb-developers] windows VM fragmentation
by p.a.boncz 11 Sep '06

11 Sep '06

Hi, Windows is indeed more vulnerable to VM fragmentation, but we still may take a look at the current behavior on 32-bits systems. Maybe things are worse than they need to be. Erwin, maybe you can try on your windows machine to type in the Mserver console: mem_printmap(); and report on the results? Windows gets some more breathing space when running on an Enterprise Edition OS, it then can use 3GB instead of 2GB. However, binaries that want to be able to use 3GB should be flagged to do so. This can be done using the EDITBIN.EXE utility from Microsoft (part of the VC++ dirstribution): EDITBIN /LARGEADDRESSAWARE /STACK:1048576,65536 /HEAP:1677216,65536 Mserver.exe (Type this all on one command line. It will modify some fields in the Mserver.exe binary) The above EDITBIN command also reduces the stack space to 1MB per thread and creates a heap of 16MB. The default stack space on windows is a huge 32MB and with all the threads that MonetDB/XQuery creates, you are easily 500MB poorer in terms of VM space available. [ Sjoerd: we should do this by default on our Win32 builds ] Note that the Enterprise Editions of Windows are very expensive. I do think though, that win32 programs on Windows-64bits edition should be able to use 3GB, and even 4GB, when running 32-bits compiled programs. So, if you posses an Athlon64 or modern Pentium4, you could switch to the 64-bits Windows XP Professional, which is not very expensive. The high VM pressure generated by MonetDB is a feature/design-error that at this stage is impossible to correct. When compiled in 64-bits (on a 64-bits OS, of course), though, it is a non-issue. There we can happily play with terabytes of VM. For the moment, however, as a MonetDB/XQuery user that wants to run 64-bits, your only option is Linux, as we do not provide a 64-bits Windows build yet. The problem is that Microsoft Visual C++ compiler cannot compile MonetDB/XQuery, as we use the full ANSI C-99 standard (which Microsoft does not support). The current MonetDB/XQuery windows binary is therefore compiled with the GNU gcc compiler (MingW), which is only available in 32-bit. We are trying to address this issue (well, Sjoerd is), and hope to provide a 64-bits Windows binary in the next major MonetDB/XQuery release. Best, Peter Boncz

3 2

[Monetdb-developers] Error with MonetDB 4.12.0 for Windows
by Jim Foley 08 Sep '06

08 Sep '06

I downloaded and installed the MonetDB 4.12.0 with SQL 2.12.0 from SourceForge using the link on the MonetDB website. http://prdownloads.sourceforge.net/monetdb/MonetDB-Installer-4.12.0.msi? download I installed it without error but when I try to start the database with sql_server I get the following error: C:\MonetDB>msqlserver.bat # Monet Database Server V4.12.0 # Copyright (c) 1993-2006, CWI. All rights reserved. # Compiled for i686-pc-win32/32bit with 32bit OIDs; dynamically linked. # Visit http://monetdb.cwi.nl/ for further information. !ERROR: Incompatible database version 020000, this server supports version 030001 !mvc_init: unable to create system tables !ERROR: interpret: no matching MIL operator to 'mapi_register(void)'. !MAYBE YOU MEAN: ! mapi_register(ptr) : void MonetDB> Laptop is a Dell C800 with 512M memory and Windows 2000 Professional, version 5.0.2195 Service Pack 4 Build 2195. I know this isn't the best platform to run MonetDB on but it will be a couple of weeks before I can free up a linux server so any help would be appreciated. Thanks. Jim Foley

3 3

[Monetdb-developers] MonetDB on UltraSPARC T1
by Wouter Alink (DT) 08 Sep '06

08 Sep '06

Hello, It took me some time, and others might be interested in it: Below my experiences with (successfully) compiling MonetDB on an UltraSPARC T1 running Solaris 10. Appearantly the UltraSPARC T1 with its many slow cores, is not the ultimate system for MonetDB (which likes fast CPU's the most), but someone has to try it... I configured with "--enable-bits=64 --enable-oid32 --enable-shared --disable-static". (using gcc) By patching iso/math_c99.h according to http://gcc.gnu.org/ml/gcc-patches/2005-05/msg02123.html as Fabian suggested in MonetDB-bug: 1304078, the first hurdle was taken. The warnings below could be worked around by deleting lines 512 t/m 516 from src/gdk/gdk_utils.mx gcc -m64 -DHAVE_CONFIG_H -I. -I/usr/users/wouter/scratch/MonetDB-4.12/src/gdk -I../.. -I/usr/users/wouter/scratch/MonetDB-4.12/src/gdk -I../common -I/usr/users/wouter/scratch/MonetDB-4.12/src/gdk/../common -DLIBGDK -O2 -std=c99 -D__EXTENSIONS__ -Wall -W -fno-strict-aliasing -O2 -fomit-frame-pointer -finline-functions -DHWCOUNTERS -DHW_SunOS -DHW_sun4v -Werror-implicit-function-declaration -Werror -Wno-unused-function -Wno-unused-label -Wno-uninitialized -D_REENTRANT -c gdk_utils.c -fPIC -DPIC -o .libs/libbat_la-gdk_utils.o /usr/users/wouter/scratch/MonetDB-4.12/src/gdk/gdk_utils.mx: In function `GDKmem_heapcheck': /usr/users/wouter/scratch/MonetDB-4.12/src/gdk/gdk_utils.mx:513: warning: comparison of unsigned expression < 0 is always false /usr/users/wouter/scratch/MonetDB-4.12/src/gdk/gdk_utils.mx:513: warning: comparison of unsigned expression < 0 is always false /usr/users/wouter/scratch/MonetDB-4.12/src/gdk/gdk_utils.mx:513: warning: comparison of unsigned expression < 0 is always false gmake[5]: *** [libbat_la-gdk_utils.lo] Error 1 gmake[5]: Leaving directory `/usr/users/wouter/scratch/MonetDB-4.12/SunOS/src/gdk' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/usr/users/wouter/scratch/MonetDB-4.12/SunOS/src/gdk' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/usr/users/wouter/scratch/MonetDB-4.12/SunOS/src' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/usr/users/wouter/scratch/MonetDB-4.12/SunOS/src' gmake[1]: *** [all-recursive] Error 1 gmake[1]: Leaving directory `/usr/users/wouter/scratch/MonetDB-4.12/SunOS' gmake: *** [all] Error 2 MonetDB then compiled successfully, without warnings. After setting: LD_LIBRARY_PATH=/usr/sfw/lib/64/:/lib/64/ and installing GNU diffutils, make check reported: 1 out of 313 could not be executed, 11 SIGNIFICANTLY different... Among which: two tests that fail due to 64bits vs 32oids in src/gdk:casts (it seems that this test does not take into account the possible difference between bit- and oid-size), 1 java example failed: output was written to err instead of std, two due to timeouts, two failed because of some additional linebrakes in tcp-output, nothing very serious it seems. I have the mTests results available if anyone is interested in those. Pathfinder compiled without problems and make check reported: 15 (out of 1187) could not be executed, 5 SIGNIFICANTLY different). Then the results of some simple queries (on a document with 1.4M elements and 2.3M attributes): -bash-3.00$ time echo "count(doc(\"patrick.xml\")//file)" | MapiClient -lxq 89017 real 0m0.110s user 0m0.021s sys 0m0.034s -bash-3.00$ time echo "count(doc(\"patrick.xml\")//file[@name])" | MapiClient -lxq 89017 real 0m8.255s user 0m0.021s sys 0m0.034s -bash-3.00$ The result of the last query is (in my eyes) dramatic (more than 4 times as slow as a Xeon 2.8). The next step will be to compile with the native compiler, which I haven't done yet, and neither do I know whether I will succeed. Greetings, Wouter

2 2

[Monetdb-developers] OS: Not enough space (again)
by Erwin Leonardi 08 Sep '06

08 Sep '06

Hi there, I tried to shred 1GB XML data into MonetDB. However, I faced the same problem as Klarinda. The error is as follows. -------------- Error --------------- ERROR = !ERROR: MT_mmap: MapViewOfFile(68c, 2, 0, 0, 436731904, 0) failed !OS: Not enough space !GDKmmap(436731904) fail => BBPtrim(enter) usage[mem=212972396,vm=803209216] ---------- End Error --------------- I use a PC with 1GB RAM, more than 5GB free hard disk space, and WinXP SP2 as OS. I have tried to set Windows' page file to 2GB. I read the previous postings regarding the same problem. I notice that Stefan has done similar experiment using his laptop (Pentium M (Dothan) 1.5 GHz, 2 GB RAM, 2.1 GB free disk space; running Fedora Core 5 Linux (Kernel 2.6.17)). My question is whether I need more than 1GB RAM if I want to shred 1GB XML data. Thanks Erwin Leonardi

2 1