[Scipy-tickets] [SciPy] #410: scipy.ndimage.sum crashes Python when index is dtype int64

SciPy scipy-tickets@scipy....
Sat May 5 02:54:49 CDT 2007


#410: scipy.ndimage.sum crashes Python when index is dtype int64
-------------------------------------------+--------------------------------
 Reporter:  socha                          |        Owner:  somebody
     Type:  defect                         |       Status:  new     
 Priority:  normal                         |    Milestone:          
Component:  scipy.ndimage                  |      Version:  0.5.2   
 Severity:  normal                         |   Resolution:          
 Keywords:  ndimage sum crash index int64  |  
-------------------------------------------+--------------------------------
Comment (by stefan):

 Thanks for the report.  Here's the problem narrowed down a bit:

 {{{
 ==13930== 1 errors in context 1 of 8:
 ==13930== Invalid read of size 4
 ==13930==    at 0x742C089: NA_elements (nd_image.h:97)
 ==13930==    by 0x742C34D: NI_ObjectToLongSequenceAndLength
 (nd_image.c:104)
 ==13930==    by 0x742C413: _NI_GetIndices (nd_image.c:744)
 ==13930==    by 0x742FDDC: Py_Statistics (nd_image.c:920)
 ==13930==    by 0x80C8B3B: PyEval_EvalFrameEx (in /usr/bin/python2.5)
 ==13930==    by 0x80C9D64: PyEval_EvalCodeEx (in /usr/bin/python2.5)
 ==13930==    by 0x80C8228: PyEval_EvalFrameEx (in /usr/bin/python2.5)
 ==13930==    by 0x80C9D64: PyEval_EvalCodeEx (in /usr/bin/python2.5)
 ==13930==    by 0x80C9DD6: PyEval_EvalCode (in /usr/bin/python2.5)
 ==13930==    by 0x80E91D7: PyRun_FileExFlags (in /usr/bin/python2.5)
 ==13930==  Address 0xC is not stack'd, malloc'd or (recently) free'd
 }}}

-- 
Ticket URL: <http://projects.scipy.org/scipy/scipy/ticket/410#comment:1>
SciPy <http://www.scipy.org/>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list