Boost logo

Boost :

From: Peter Dimov (pdimov_at_[hidden])
Date: 2006-07-06 15:50:15


David Abrahams wrote:
> "Peter Dimov" <pdimov_at_[hidden]> writes:
>
>> Pedro Lamarão wrote:
>
>>> Shouldn't there be some sentry object then, with that
>>> "on_thread_exit()" thing called by the destructor?
>
>> If you allow an exception to escape from your thread function, the
>> default behavior is for the runtime to call terminate(), so a missing
>> on_thread_exit is the least of your worries. :-)
>
> Well, let's not forget that terminate() is a form of orderly shutdown.

It's easier to move the on_thread_exit call to a destructor than to argue
about the precise meaning of orderly shutdown, so that's what I've done. :-)

Index: thread.cpp
===================================================================
RCS file: /cvsroot/boost/boost/libs/thread/src/thread.cpp,v
retrieving revision 1.22
diff -u -r1.22 thread.cpp
--- thread.cpp 6 Jul 2006 13:45:13 -0000 1.22
+++ thread.cpp 6 Jul 2006 19:45:57 -0000
@@ -94,6 +94,18 @@
     bool m_started;
 };

+#if defined(BOOST_HAS_WINTHREADS)
+
+struct on_thread_exit_guard
+{
+ ~on_thread_exit_guard()
+ {
+ on_thread_exit();
+ }
+};
+
+#endif
+
 } // unnamed namespace

 extern "C" {
@@ -107,13 +119,16 @@
     {
 // try
         {
+#if defined(BOOST_HAS_WINTHREADS)
+
+ on_thread_exit_guard guard;
+
+#endif
+
             thread_param* p = static_cast<thread_param*>(param);
             boost::function0<void> threadfunc = p->m_threadfunc;
             p->started();
             threadfunc();
-#if defined(BOOST_HAS_WINTHREADS)
- on_thread_exit();
-#endif
         }
 // catch (...)
 // {


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk