Agreed, I took causing problem too wide. Well, at least my sentence triggered fruitful discussion on dev-list and Tim suggested nice fix.
2006/11/25, Stefano Mazzocchi <[EMAIL PROTECTED]>:
Alexey Varlamov (JIRA) wrote: > [ http://issues.apache.org/jira/browse/HARMONY-2135?page=comments#action_12452379 ] > > Alexey Varlamov commented on HARMONY-2135: > ------------------------------------------ > > If the root cause of this is really as described HARMONY-2163, it sounds more like a bug in the DaCapo itself... uh? are you seriously suggesting that that when dacapo works on all other VMs and fails on harmony, it's its fault? I suggest otherwise: there is clearly something that that test triggers (in harmony's classloaders or somewhere else) that behaves differently from RI and the other compatible VMs. Let's find out what it is, create a test that reproduces it and fix it. -- Stefano.
