Anthony,

Ticket 2309 already existed in Trac.  I updated it slightly and added my patch.

See: https://svn.boost.org/trac/boost/ticket/2309#comment:3

Is this likely to be in 1.43?  What can I do to further help?

Thanks,
SB

On Thu, Feb 4, 2010 at 1:02 AM, Anthony Williams <anthony@justsoftwaresolutions.co.uk> wrote:
On 03/02/10 23:13, Scott Bailey wrote:
Here is the setup:

MyCommon.so
- this lib contains some basic classes, one of which (baseTRx) has a
boost::thread.
- this lib includes boost_system and boost_thread using the
--whole-archive flag (linux) and equivalent for darwin and win
     - this was done to minimize the number of files the customer would
receive, among other reasons...
MyTRx.so
- this lib contains a class derived from baseTRx (myTRx).
- links to MyCommon
MyEXE
- this contains main.
- links to MyCommon
- loads MyTRx at run time.
- creates an instance od myTRx.
- signals MyEXE (via interupt()) when its time to cleanly shut down

Using GCC 4.2 w/ -fvisibility=hidden, thread_interrupted isn't being
caught in myTRx since it's being thrown from a different so w/ hidden
visibility.

OK, makes sense.


The fix for this is:
    #define BOOST_THREAD_DECL __attribute__ ((visibility("default")))
when gcc has visibility.


Am I correct?  If so, I'll write the defect and work on the patch....

If that works, write it up and I'll patch boost.thread

Anthony
--
Author of C++ Concurrency in Action     http://www.stdthread.co.uk/book/
just::thread C++0x thread library             http://www.stdthread.co.uk
Just Software Solutions Ltd       http://www.justsoftwaresolutions.co.uk
15 Carrallack Mews, St Just, Cornwall, TR19 7UL, UK. Company No. 5478976
_______________________________________________
threads-devel mailing list
threads-devel@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/threads-devel



--
Scott Bailey