From: Dan K. <dk...@ix...> - 2002-02-02 05:33:20
|
I'm trying to track down a bug that kills pthreads programs on an embedded system. What are the best stress programs for threads? We're currently looking at running lots of copies of sh-pth_str01 sh-pth_str03 sh-pth_str02 concurrently in pan. The tests run on our embedded system; we'll try running them from pan next. Our existing stress test is able to reproduce the problem if you run six copies in parallel for ten or so hours; the source is at http://www.ugcs.caltech.edu/~diffusor/glibc-psem/psem-test2.c Another test that looks interesting is http://marc.theaimsgroup.com/?l=linux-kernel&m=99446582214095&w=2 If I manage to find a test that reproduces my problem quickly, would it be a candidate for inclusion in the LTP? - Dan |