Re: [Boost-bugs] [Boost C++ Libraries] #12871: boost::posix_time::operator<< results in heap corruption

Subject: Re: [Boost-bugs] [Boost C++ Libraries] #12871: boost::posix_time::operator<< results in heap corruption
From: Boost C++ Libraries (noreply_at_[hidden])
Date: 2017-02-26 05:57:45


#12871: boost::posix_time::operator<< results in heap corruption
-------------------------------+-------------------------
  Reporter: nmmreddy11@… | Owner: az_sw_dude
      Type: Bugs | Status: new
 Milestone: To Be Determined | Component: date_time
   Version: Boost 1.59.0 | Severity: Showstopper
Resolution: | Keywords:
-------------------------------+-------------------------

Comment (by nmmreddy11@…):

 I see a similar incident reported 8 years ago.
 https://svn.boost.org/trac/boost/ticket/3369

 '''
 Issue:'''

 posix_time::operator<< crashes with double free under possible race
 conditions
 #13 0x08070aca in boost::posix_time::operator<< <char,
 std::char_traits<char> > (os=@0x80def94,p=@0x40bb32c8) at
 posix_time_io.hpp:53


 '''Reply:'''

 Is the streaming expression protected with a mutex lock or a similar
 synchronization mechanism? If not, that is the most probable source of the
 problem.


 '''Closure comments:'''

 We haven't seen this issue reappear, so perhaps it was a temporary lapse
 in synchronization after all. Can be closed.

 '''What should I do?'''

 Am I supposed to synchronize function calls to data time library?
 posix_time is not thread safe?

-- 
Ticket URL: <https://svn.boost.org/trac/boost/ticket/12871#comment:1>
Boost C++ Libraries <http://www.boost.org/>
Boost provides free peer-reviewed portable C++ source libraries.

This archive was generated by hypermail 2.1.7 : 2017-02-26 06:01:13 UTC