Hi! > > The following patch fixes the problem but I'm not very knowledgeable > > on what fork13 does so this patch might not be a good approach. Can > > someone more knowledgeable on it let me know if this is a valid > > approach? > > Strange indeed. I could see the same errors in the log as well. I'll > look closer on this hopefully today.
Okay, here is the cause: The '-c 2' parameter of the test causes it to fork at the test very beginning (in the TEST_PAUSE macro that evaluates to call to usc_global_test_hook()). So two instances of the test are running. However the ltp-pan, as it executes the test, remebers pid of the test it has started (in this case first instance of the fork13 testcase) and then waits until this program exits, reports the test success/failure accordingly to it's return value etc... And also kills all lefover processes forked from the testcase (in this case, most likely the second instance of the test running and it's child). Moreover even if the second instance of the test finishes fine, it's return value is ignored. So the -c option is broken by design and should be removed, maybe even from LTP as whole. -- Cyril Hrubis chru...@suse.cz ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Ltp-list mailing list Ltp-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ltp-list
