Subj : Re: pthread_mutex_lock throws an exception on receiving signal 15
To   : comp.programming.threads
From : pankaj-startup
Date : Thu Oct 13 2005 12:52 am

It seems problem is solved now.

In original code SIGTERM was blocked and then sigwait was called to
handle async-signals in synchronous manner.
for more details:
http://www.awprofessional.com/content/images/0201633922/sourcecode/sigwait.c


So there were not an issue of calling unsafe functions from the context
of  async signal-handler.

My signal-handler was cancelling 2 threads before calling exit(0).
I just replaced exit(0) by _exit(0). And problem appears to be
disappeared (process never dumped the core).

I think this resolution has following logical reason.
<snip>
In the child branch of a fork(), it is normally incorrect to use
exit(), because that can lead to stdio buffers being flushed twice, and
temporary files being unexpectedly removed. In C++ code the situation
is worse, because destructors for static objects may be run
incorrectly.
</snip>
http://www.erlenstar.demon.co.uk/unix/faq_2.html

dynamic loader must be calling fork/exec to create a process. In that
sense my process can be a child process. But if this is the case, then
it would be always dangeourous to call exit in C++ multithreaded
application.

.