On Wed, Oct 08, 2008 at 11:07:05PM +0200, Bob van Loosen wrote: >Larry Hall (Cygwin) wrote: >>Bob van Loosen wrote: >>>Christopher Faylor wrote: >>>>On Wed, Oct 08, 2008 at 02:45:55AM +0200, Bob van Loosen wrote: >>>> >>>>>Christopher Faylor wrote: >>>>>>This should be fixed in the next snapshot. If you are looking for a >>>>>>workaround, specifically set the sigev_notify_attributes to >>>>>>PTHREAD_CREATE_DETACHED. >>>>>> >>>>>Thanks, that fixed it. Shouldn't PTHREAD_CREATE_DETACHED be default? >>>>>That's what the opengroup.org site says. >>>> >>>>Maybe you need to reinvestigate the meaning of "fixed" and >>>>"workaround"? >>>From my point of view, the workaround fixed the memory leak. >> >>So if the workaround was the fix in your mind, why the follow-up >>question? I think the point that's being missed here is that there was >>a way Chris found for you to move beyond the problem even without his >>fix. But there was a problem that needed fixing so that the next >>person doesn't run into the same thing. If that makes sense to you, >>then you've gotten all you can from this thread. > >I was only curious why using a pthread_attr_t with the default >attributes makes a difference.
It was a bug. The bug was fixed in the next snapshot. Second base. cgf -- 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/
