Issue 1662529: [2.5 regression?] failure to import the ORBit extension (original) (raw)

Issue1662529

Created on 2007-02-17 21:38 by doko, last changed 2022-04-11 14:56 by admin. This issue is now closed.

Messages (3)
msg31286 - (view)	Author: Matthias Klose (doko) *	Date: 2007-02-17 21:38
seen with 2.5 SVN 20070216, with an interpreter built with --enable-pydebug, succeeds with a 2.4.4 interpreter built with --enable-pydebug: pyorbit is version 2.14.1. $ python-dbg -c 'import ORBit' [26750 refs] Fatal Python error: ../Objects/tupleobject.c:169 object at 0x8186bdc has negative ref count -1 Aborted (core dumped) (gdb) bt #0 0xb7f89410 in __kernel_vsyscall () #1 0xb7e11df0 in raise () from /lib/tls/i686/cmov/libc.so.6 #2 0xb7e13641 in abort () from /lib/tls/i686/cmov/libc.so.6 #3 0x081174f8 in Py_FatalError ( msg=0xbfd0ef78 "../Objects/tupleobject.c:169 object at 0x8186bdc has negative ref count -1") at ../Python/pythonrun.c:1555 #4 0x08093305 in _Py_NegativeRefcount (fname=0x816f958 "../Objects/tupleobject.c", lineno=169, op=0x8186bdc) at ../Objects/object.c:193 #5 0x080add27 in tupledealloc (op=0xb7a1bd1c) at ../Objects/tupleobject.c:169 #6 0x080973a6 in _Py_Dealloc (op=0xb7a1bd1c) at ../Objects/object.c:1928 #7 0x0814d037 in func_dealloc (op=0xb7a1d1c4) at ../Objects/funcobject.c:451 #8 0x080973a6 in _Py_Dealloc (op=0xb7a1d1c4) at ../Objects/object.c:1928 #9 0x0808eb99 in dict_dealloc (mp=0xb7a1c494) at ../Objects/dictobject.c:819 #10 0x080973a6 in _Py_Dealloc (op=0xb7a1c494) at ../Objects/object.c:1928 #11 0x0806459e in class_dealloc (op=0xb7a0bdb4) at ../Objects/classobject.c:184 #12 0x080973a6 in _Py_Dealloc (op=0xb7a0bdb4) at ../Objects/object.c:1928 #13 0x080add3b in tupledealloc (op=0xb7a22edc) at ../Objects/tupleobject.c:169 #14 0x080973a6 in _Py_Dealloc (op=0xb7a22edc) at ../Objects/object.c:1928 #15 0x08064540 in class_dealloc (op=0xb7a2446c) at ../Objects/classobject.c:183 #16 0x080973a6 in _Py_Dealloc (op=0xb7a2446c) at ../Objects/object.c:1928 #17 0x080add3b in tupledealloc (op=0xb7a243dc) at ../Objects/tupleobject.c:169 #18 0x080b13fe in subtype_dealloc (self=0xb7a243dc) at ../Objects/typeobject.c:709 #19 0x080973a6 in _Py_Dealloc (op=0xb7a243dc) at ../Objects/object.c:1928 #20 0x0808eb99 in dict_dealloc (mp=0xb7dcf214) at ../Objects/dictobject.c:819 #21 0x080973a6 in _Py_Dealloc (op=0xb7dcf214) at ../Objects/object.c:1928 #22 0x08112acb in PyInterpreterState_Clear (interp=0x81b8020) at ../Python/pystate.c:107 #23 0x081145be in Py_Finalize () at ../Python/pythonrun.c:441 #24 0x08059eb0 in Py_Main (argc=3, argv=0xbfd0f5b4) at ../Modules/main.c:545 #25 0x08058da6 in main (argc=Cannot access memory at address 0x155c ) at ../Modules/python.c:23
msg31287 - (view)	Author: Collin Winter (collinwinter) *	Date: 2007-03-10 03:51
The error message and backtrace here don't prove that it's a regression in Python: it could easily be a problem with an extension module. Do you get this error if you compile Python 2.5 without --with-pydebug? Is "ORBit" provided by the project at http://sourceforge.net/projects/orbit-python/ or the one at http://ftp.gnome.org/pub/GNOME/sources/pyorbit/? If the former, the project's SF page says active development stopped in 2004, so there's no way it could support Python 2.5.
msg31288 - (view)	Author: Neal Norwitz (nnorwitz) *	Date: 2007-03-27 06:18
These problems are almost always issues in the third party modules. I'm closing this on the basis that the problem is likely in ORBit. If you have more information that points to the core as the likely cause, please reopen this report. Without more info, we can't help with this problem. Note another problem could be that Python 2.5 does not allow the memory APIs to be mixed as they used to. This could also lead to a problem like this (though I think it often leads to a crash earlier IIRC.)