On 25 March 2016 at 17:01, Rob Savoye <r...@welcomehome.org> wrote: > On 03/25/2016 08:57 AM, Christophe Lyon wrote: > >> Well, as Yvan showed with his recent patches, we seem to be using >> DejaGnu in an unusual configuration, which uncovers corner case bugs. > > As I've mentioned before, we need to make our validation less unusual. :-) > >> I'm seeing "random" results in sanitizers tests, which is a relatively >> recent addition to GCC. >> And I know we are not the only ones having trouble with the >> sanitizers results. > > I agree with Ben, I think the problem is with the sanitizer tests, and > not necessarily something that has to be "fixed" in DejaGnu. >
At this point, I don't know where the problem is. We've been seeing random results mostly in the sanitizer tests for months. A possible explanation is that these tests wrongly expect a given order between stdout & stderr, which is not granted, and might be changed because of the high load we are putter on the tester machine. As I said, I'm still just trying to make sure that this is actually what is happening. And making DejaGnu enforce an order on stdout and stderr would certainly help diagnose badly written tests. > - rob - > _______________________________________________ DejaGnu mailing list DejaGnu@gnu.org https://lists.gnu.org/mailman/listinfo/dejagnu
