Bugs item #1876422, was opened at 2008-01-21 13:25 Message generated for change (Comment added) made by stmane You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1876422&group_id=56967
Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: PFtijah Group: Pathfinder 0.20 Status: Open Resolution: None Priority: 5 Private: No Submitted By: Ying Zhang (yingying) Assigned to: Jan Flokstra (jflokstra) Summary: PF: could not load module pftijah on Darwin 9 Initial Comment: Hello, starting Mserver with module pathfinder results in following errors. The same problem occurs with both stable brach and dev branch. $ Mserver --dbinit="module(pathfinder);" # MonetDB Server v4.20.1 # based on GDK v1.20.1 # Copyright (c) 1993-2007, CWI. All rights reserved. # Compiled for i686-apple-darwin9.1.0/32bit with 32bit OIDs; dynamically linked. # Visit http://monetdb.cwi.nl/ for further information. !ERROR: pathfinder: when trying to load module 'pftijah': !ERROR: DL_open: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib,10) -> Error: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib, 10): Symbol not found: _tj_extend_plane !ERROR: Referenced from: /Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib !ERROR: Expected in: flat namespace !ERROR: . !ERROR: TBL_openmod: DL_open: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib, 10): Symbol not found: _tj_extend_plane !ERROR: Referenced from: /Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib !ERROR: Expected in: flat namespace !ERROR: . !ERROR: TBL_loadmod: module load _pftijah cancelled. !ERROR: moduleClient: module(_pftijah) load error. MonetDB>exit(); !ERROR: interpret: unknown variable 'pf_wal'. !ERROR: interpret_params: lock_set(param 1): evaluation error. MonetDB>^D I'm using Mac OS 10.5 on Intel Duo core. $ uname -a Darwin leopard.local 9.1.0 Darwin Kernel Version 9.1.0: Wed Oct 31 17:46:22 PDT 2007; root:xnu-1228.0.2~1/RELEASE_I386 i386 The following configure options are used: --enable-debug --enable-assert --disable-optimize --enable-strict Regards, Jennie ---------------------------------------------------------------------- >Comment By: Stefan Manegold (stmane) Date: 2008-01-22 23:18 Message: Logged In: YES user_id=572415 Originator: NO I declared INLINEd function tj_extend_plane() static (just like all other INLINEd functions in this file). Please check, whether it helps ... ---------------------------------------------------------------------- Comment By: Stefan Manegold (stmane) Date: 2008-01-21 16:36 Message: Logged In: YES user_id=572415 Originator: NO There are many more INLINEd functions in the MonetDB code base (that seem to work fine?); hence, comparing this one to all the other ones --- consider the way it is declared, how it's called, how the code is compiled, how the object/library is built, how the object/library is linked-in where it is called, how the shared library (module) is loaded, etc. --- might shed some light on this problem... ---------------------------------------------------------------------- Comment By: Fabian (mr-meltdown) Date: 2008-01-21 15:35 Message: Logged In: YES user_id=963970 Originator: NO The linker from Xcode 3.1 (only available for Leopard) is completely different than the linker from the last Xcode for Tiger. On Tiger a prelease of this linker was available as ld64, which is distinct from ld, whereas on Leopard ld == ld64. This ld64 linker is more strict, and more precise on where what is wrong. I'd have to look into it into detail to figure out what it is, and how to fix it. ---------------------------------------------------------------------- Comment By: Ying Zhang (yingying) Date: 2008-01-21 15:29 Message: Logged In: YES user_id=341633 Originator: YES I see that pftijah can be at least be loaded on darwin 8, so this might be a darwin 9 specific problem... ---------------------------------------------------------------------- Comment By: Jan Flokstra (jflokstra) Date: 2008-01-21 15:28 Message: Logged In: YES user_id=1054297 Originator: NO This is a very strange error. The linker tries to find a symbol which should not be there because the function is INLINE. $PFROOT/modules/pftijah/serialize_pftijah.mx: INLINE oid tj_extend_plane(struct tijahContextStruct *tjctx) { oid base = tjctx->tijahPre - tjctx->dbat_collPre.seqbase; if ( base >= tjctx->dbat_collPre.raw_max ) { if ( dbat_extend(&tjctx->dbat_collPre,base, 0) < 0 ) return oid_nil; /* IMPORTANT: the size of the two bats is synchronized by the use * of the forced size (last) parameter of dbat_extend */ if ( dbat_extend(&tjctx->dbat_collSize,base,tjctx->dbat_collPre.raw_max) < 0 ) return oid_nil; } return tjctx->tijahPre++; } My first reaction there must be an error in the Darwin INLINE function handling. Maybe you could replace INLINE by static to see if this works. If so the INLINE function handling in Darwin has a bug. ---------------------------------------------------------------------- Comment By: Fabian (mr-meltdown) Date: 2008-01-21 14:55 Message: Logged In: YES user_id=963970 Originator: NO side-note: we should try not to use flat namespace ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1876422&group_id=56967 ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Monetdb-bugs mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
