Christopher Faylor wrote:
It is not necessarily a non-problem. If you can reproduce this consistently, then it is a problem, either in make or cygwin.
It is not consistent. (is it ever? :-)
Make -j10 will probably be more useful, since running a threaded make is usually what triggers the issue for me. I've also observed that the likelihood of creating what I call an "orphaned" process (when the parent returns but the child doesn't receive the proper SIG) happens to increase significantly as the average load increases. While this is most prevalent in Make, I have seen it happen with shell scripts that call multiple, concurrent sub-shells. Another observation is that this behavior seems to happen more frequently on Win9X boxes then on Win2k. What this means, I have no idea, but Chuck is right, it isn't consistent at all. I'll try to grab an strace the next time I find a situation which does this.
Cheers, Nicholas
-- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
