Here is an excerpt of the stack on the numpy svn of wednesday :<br><br>#0 0x40000402 in __kernel_vsyscall ()<br>#1 0x00b8e382 in sem_post@GLIBC_2.0 () from /lib/libpthread.so.0<br>#2 0x080fe5b7 in PyThread_release_lock (lock=0x80d0fb2) at Python/thread_pthread.h:374
<br>#3 0x080d0fb2 in PyEval_SaveThread () at Python/ceval.c:299<br>#4 0x4064ec7a in dotblas_innerproduct (dummy=0x0, args=0x960522c) at numpy/core/blasdot/_dotblas.c:835<br>#5 0x0811ecf1 in PyCFunction_Call (func=0x960522c, arg=0x80cfec7, kw=0x942068c) at Objects/methodobject.c:73
<br>#6 0x080cfec7 in call_function (pp_stack=0x0, oparg=1) at Python/ceval.c:3564<br>#7 0x080cb19a in PyEval_EvalFrameEx. () at Python/ceval.c:2267<br>#8 0x080ca848 in PyEval_EvalCodeEx. () at Python/ceval.c:2831<br>#9 0x0811de5b in function_call (func=0x0, arg=0x960516c, kw=0x965b148) at Objects/funcobject.c:517
<br>#10 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x960516c, kw=0x8061a66) at Objects/abstract.c:1860<br>#11 0x08061a66 in instancemethod_call (func=0x95f16ac, arg=0x94f5824, kw=0x805bdb6) at Objects/classobject.c:2497<br>
#12 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x95f16ac, kw=0x80a86e9) at Objects/abstract.c:1860<br>#13 0x080a86e9 in slot_tp_call (self=0xbf975c78, args=0x0, kwds=0x805bdb6) at Objects/typeobject.c:4633<br>#14 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x95f16ac, kw=0x80d017b) at Objects/abstract.c:1860
<br>#15 0x080d017b in do_call (func=0x0, pp_stack=0x1, na=1, nk=154955456) at Python/ceval.c:3775<br>#16 0x080cfd56 in call_function (pp_stack=0x0, oparg=1) at Python/ceval.c:3587<br>#17 0x080cb19a in PyEval_EvalFrameEx. () at Python/ceval.c:2267
<br>#18 0x080ca848 in PyEval_EvalCodeEx. () at Python/ceval.c:2831<br>#19 0x0811de5b in function_call (func=0x0, arg=0x960520c, kw=0x30) at Objects/funcobject.c:517<br>#20 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x960520c, kw=0x8061a66) at Objects/abstract.c:1860
<br>#21 0x08061a66 in instancemethod_call (func=0x95ecbac, arg=0x94f57d4, kw=0x805bdb6) at Objects/classobject.c:2497<br>#22 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x95ecbac, kw=0x80a86e9) at Objects/abstract.c:1860<br>
#23 0x080a86e9 in slot_tp_call (self=0xbf97609c, args=0x0, kwds=0x805bdb6) at Objects/typeobject.c:4633<br>#24 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x95ecbac, kw=0x80d017b) at Objects/abstract.c:1860<br>#25 0x080d017b in do_call (func=0x0, pp_stack=0x1, na=1, nk=156957888) at Python/ceval.c:3775
<br>#26 0x080cfd56 in call_function (pp_stack=0x0, oparg=1) at Python/ceval.c:3587<br>#27 0x080cb19a in PyEval_EvalFrameEx. () at Python/ceval.c:2267<br>#28 0x080ca848 in PyEval_EvalCodeEx. () at Python/ceval.c:2831<br>#29 0x0811de5b in function_call (func=0x0, arg=0x960518c, kw=0x9548a1) at Objects/funcobject.c:517
<br>#30 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x960518c, kw=0x8061a66) at Objects/abstract.c:1860<br>#31 0x08061a66 in instancemethod_call (func=0x960526c, arg=0x95fc874, kw=0x805bdb6) at Objects/classobject.c:2497<br>
#32 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x960526c, kw=0x80a86e9) at Objects/abstract.c:1860<br>#33 0x080a86e9 in slot_tp_call (self=0xbf9764c0, args=0x0, kwds=0x805bdb6) at Objects/typeobject.c:4633<br>#34 0x0805bdb6 in PyObject_Call (func=0x0, arg=0x960526c, kw=0x80d017b) at Objects/abstract.c:1860
<br>#35 0x080d017b in do_call (func=0x0, pp_stack=0x1, na=1, nk=157640136) at Python/ceval.c:3775<br>#36 0x080cfd56 in call_function (pp_stack=0x0, oparg=1) at Python/ceval.c:3587<br>#37 0x080cb19a in PyEval_EvalFrameEx. () at Python/ceval.c:2267
<br>#38 0x080ca848 in PyEval_EvalCodeEx. () at Python/ceval.c:2831<br>#39 0x0811de5b in function_call (func=0x0, arg=0x95f260c, kw=0x0) at Objects/funcobject.c:517<br>#40 0x0805bdb6 in PyObject_Call (func=0x9609934, arg=0x95f260c, kw=0x8061a66) at Objects/abstract.c:1860
<br><br>Seems that the bug could be somewhere in the handling of the dot blas module ?<br><br>Matthieu<br><br><div><span class="gmail_quote">2007/10/15, Travis E. Oliphant <<a href="mailto:oliphant@enthought.com">oliphant@enthought.com
</a>>:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Matthieu Brucher wrote:<br>><br>> The problem is that there is a data-type reference counting error some
<br>> where that is attempting to deallocate the built-in data-type 'l'<br>><br>><br>><br>> That's what I supposed, but I couldn't find the reason why it wanted<br>> to do this<br>>
<br>><br>> It's not really a Python error but a logging. The code won't let you<br>> deallocate the built-ins, but it will tell you that something<br>> tried to.<br>><br>> Reference counting on data-types is easy to get wrong
<br>> (particularly with<br>> Pyrex extension modules) because most calls consume a reference to the<br>> data-type (if they return an object that contains a reference to the<br>> data-type).<br>
><br>> It is a bug, and it would be nice to figure it out, but that would<br>> require the code that caused it.<br>><br>><br>> I've updated my numpy version to the latest svn, the behaviour seems
<br>> to be different (more warnings), I'll try to give more information<br>> about the error, but giving the whole code will not be simple (it uses<br>> a big data file that seems to trigger the error as with other data
<br>> files, the error didn't show up :()<br>><br><br>There are two types of errors that can occur with reference counting on<br>data-types.<br><br>1) There are too many DECREF's --- this gets us to the error quickly and
<br>is usually easy to reproduce<br>2) There are too many INCREF's (the reference count keeps going up until<br>the internal counter wraps around to 0 and deallocation is attempted)<br>--- this error is harder to reproduce and usually takes a while before
<br>it happens in the code.<br><br><br>-Travis<br><br>_______________________________________________<br>Numpy-discussion mailing list<br><a href="mailto:Numpy-discussion@scipy.org">Numpy-discussion@scipy.org</a><br><a href="http://projects.scipy.org/mailman/listinfo/numpy-discussion">
http://projects.scipy.org/mailman/listinfo/numpy-discussion</a><br></blockquote></div><br>