Message11287

Sorry I haven't got back to you on where the exceptions occur, I haven't really had any time to look at it.

About the test suite having the potential for deadlock. I agree this is bad, however is this actually the case? With the increased Xss setting I have never observed these deadlocks, whereas before they would happen frequently (a bit vague I know). So while I can't explain why increasing Xss should effect this issue, it seems to be the case. Has anyone seen these deadlocks occur since Xss was increased?

On the JVM, I don't know if offering extra memory to allow catch or finally to complete is plausible, where would you stop, there is nothing stopping someone assigning huge new objects in a finally (although I think it would be considered bad practice). I think the issue here is caused because Jython catches Throwable and then attempts to continue, this is risky, but needed here to produce the correct behaviour a Python application would expect. I think it's best said here https://docs.oracle.com/javase/7/docs/api/java/lang/Error.html "An Error is a subclass of Throwable that indicates serious problems that a reasonable application should not try to catch" and "these errors are abnormal conditions that should never occur"

Finally, I don't think this should be a issue to hold back the 2.7.1 release, essentially this only occurs because tests are deliberately performing infinite recursions to see what happens. This should not be the case in any normal operation.