The following testsuite program fails when added to the libstdc++ testsuite: #include <iostream> int main() { for(int i=0;i<1000000;i++) std::cout << "C" << std::endl; } Is there some reason for this?
Are you sure that this is not a dejagnu bug? Or are you sure that you just don't get a timeout? Also anything this long will most likely get a timeout on slow machines so, I really doubt that we want this, also one thing, we will be filling up the long and with oldder machines with that much space left on their hd, I don't think we want this either.
I'm fairly sure it's not a timeout (It takes <5 seconds to run on my computer). It might be a dejagnu bug. Unfortunatly I find it very hard to find out exactly what is causing this bug. While I agree that allowing program's output to run out of control is a bad idea, the problem with the current cutoff system is that it seems quite unreliable. Around the mark at which the bug occurs (round about 200,000-300,000 characters on my computer) the test will sometimes run and sometimes fail. This kind of unpredictable behaviour led me to a merry afternoon of debugging (I was writing a test which outputed a large quantity of state information for testing reasons) as it randomly appeared and disappeared. I do think that the behaviour could be more reliable, and at least better documented.
Does your version of expect include the patch in bug 12096? If not, and if using a version with that patch fixes your problem, then this would be a duplicate of that bug.
Yep, that fixed it. Marking it as a dup. *** This bug has been marked as a duplicate of 12096 ***