[tahoe-dev] [pycryptopp] #19: Segmentation fault in HashMultipleBlocks

pycryptopp trac at allmydata.org
Fri Sep 11 14:12:34 PDT 2009


#19: Segmentation fault in HashMultipleBlocks
---------------------+------------------------------------------------------
Reporter:  francois  |           Owner:  francois
    Type:  defect    |          Status:  new     
Priority:  major     |         Version:  0.5.1   
Keywords:            |   Launchpad_bug:          
---------------------+------------------------------------------------------

Comment(by zooko):

 Oh, something else that we should do: the fix
 (http://allmydata.org/trac/pycryptopp/changeset/669 ) should fix it only
 if the problem was that {{{sizeo(size_t) != sizeof(Py_ssize_t)}}} or
 {{{sizeof(int) != sizeof(Py_ssize_t)}}}.  So everyone involved (including
 you, François) please run the following C code and post your results:

 {{{
 #include "Python.h"
 #include <stdio.h>
 int main(int argc, char**argv) {
   printf('%d, %d, %d", sizeof(int), sizeof(size_t), sizeof(Py_ssize_t));
   return 0;
 }
 }}}

 If all three of the values are equal on your system, then Nathan's patch
 cannot be the fix to the problem on your system.  Therefore, I believe
 that these values will differ on all affected systems.  (However,
 François's upgrade of the OS may have changed these values on his system.)

-- 
Ticket URL: <http://allmydata.org/trac/pycryptopp/ticket/19#comment:8>
pycryptopp <http://allmydata.org/trac/pycryptopp>
Python bindings for the Crypto++ library


More information about the tahoe-dev mailing list