[Scipy-tickets] [SciPy] #1233: Weired test failure in airye function depending on test context in scipy 0.8

SciPy Trac scipy-tickets@scipy....
Sat Jul 17 15:33:33 CDT 2010


#1233: Weired test failure in airye function depending on test context in scipy
0.8
---------------------------+------------------------------------------------
 Reporter:  cjgohlke       |       Owner:  pv   
     Type:  defect         |      Status:  new  
 Priority:  normal         |   Milestone:  0.8.0
Component:  scipy.special  |     Version:  0.7.0
 Keywords:                 |  
---------------------------+------------------------------------------------

Comment(by pv):

 That is quite strange, since in `amos/zairy.f` we have:
 {{{
 C***FIRST EXECUTABLE STATEMENT  ZAIRY
       IERR = 0
       NZ=0
 }}}
 and similarly in all other AMOS routines. Unless sizeof(int) !=
 sizeof(FORTRAN_INTEGER) on your platform, it's a bit difficult to
 understand why that helps.

 One scenario could be that on your platform, `ifort` assumes that INTEGER
 is a 64-bit number; if the C integers are 32-bit, this could easily lead
 to various problems. Or some other problem in the Fortran bindings.

 Such problems are perhaps revealed by building and linking this:
 {{{
 #include <stdio.h>
 int main() {
     int a = 0xffffffff;
     foo_(&a);
     return 0;
 }
 }}}
 {{{
       subroutine foo(j)
       integer j
       j = 0
       end
 }}}

 Another possibility is a bug in the F2Py code when a pointer does not fit
 into long.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1233#comment:6>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list