On Mon, 31 Jan 2005, Mark Taylor wrote:
> Have you tried java -verbose:jni (and possibly -Xcheck:jni, not that
> I really understand what that does)? If you can narrow it down to one
> JNI call then I (or you) could take a Very Hard Look at said bit of
> JNIAST/AST.
Thanks Mark, I've tried those (and one or two other -X options) and
haven't got any insight. The problem really does move around depending on
the phase of the Moon, Microsoft share price etc. For instance this
morning my test data loaded once (got me all excited), next time it
crashed immediately after the first spectrum was loaded, and the time
after that it behaved what I've come to see as more normal (about 3/4 of
way through the list). Me, I just ran the same the command to get these
effects. I suspect Microsoft uses various sharing/caching schemes for
DLLs, so where and what you get in memory is a bit volatile.
Cheers,
Peter.
|