|
[f2py] f2py and PGI again: msg#00003python.f2py.user
I am having a hard time to get code compiled using PGI 7.0 to work with f2py on linux_x86_64 and Python 2.5. I am aware of the "sys.setdlopenflags(258)" trick from <http://www.pgroup.com/userforum/viewtopic.php?t=345&postdays=0&postorder=asc&start=10&sid=84d954e1c5d2e4e7e87689253b9cb4de>, but this trick leads to further segfaults:: hoel@amur:20070327 cat aa20070327.f subroutine A() print *, "HALLO TEST" return end hoel@amur:20070327 f2py -c aa20070327.f ... hoel@amur:20070327 f2py -c aa20070327.f -m aa ... hoel@amur:20070327 python -c "import sys;sys.setdlopenflags(258);import untitled,aa;untitled.a();aa.a()" Segmentation fault (core dumped) hoel@amur:20070327 python -c "import untitled,aa;untitled.a();aa.a()" HALLO TEST HALLO TEST There is not PGI involved up to this point. g77 is used to compile the module, but without the "setdlopenflags" PGI compiled modules segfault when they access e.g. the buildin for "write", something like:: hoel@amur:20070327 f2py -c aa20070327.f -m aa --fcompiler=pg --f77exec=pgf90 ... hoel@amur:20070327 f2py -c aa20070327.f --fcompiler=pg --f77exec=pgf90 ... hoel@amur:20070327 python -c "import untitled,aa;untitled.a();aa.a()" HALLO TEST PGFIO-F-202/list-directed write/unit=6/conflicting specifiers. File name = stdout formatted, sequential access record = 1 In source file aa20070327.f, at line number 3 hoel@amur:20070327 python -c "import sys;sys.setdlopenflags(258);import untitled,aa;untitled.a();aa.a()" Segmentation fault (core dumped) How do I escape this catch-22? Thanks Berthold -- __ Address: G / \ L Germanischer Lloyd phone: +49-40-36149-7374 -+----+- Vorsetzen 35 P.O.Box 111606 fax : +49-40-36149-7320 \__/ D-20459 Hamburg D-20416 Hamburg |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | [f2py] OS X 10.4.9 Mac intel and gfortran - has anyone gotten this to work?: 00003, James Boyle |
|---|---|
| Next by Date: | [f2py] F2PY, Mac Intel, gfortran/ifort fixes?: 00003, Louis Wicker |
| Previous by Thread: | [f2py] OS X 10.4.9 Mac intel and gfortran - has anyone gotten this to work?i: 00003, James Boyle |
| Next by Thread: | [f2py] F2PY, Mac Intel, gfortran/ifort fixes?: 00003, Louis Wicker |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |