Simone Giannecchini a écrit :
> I should have not touched Resample at all lately, All I did yesterday
> after a fresh checkout of gt-trunk was to remove two packages from
> spike, hence this should not jeopardize the tests. On suggestion.
> Every time you see an EXCEPTION_ACCESS_VIOLATION, please, send the log
> file so that I can inspect and ask the JAI guys directly. This way we
> have fixed some bus very quickly more than once.
In the particular case of this bug, it is not necessary to get the log or send
anything to the JAI guys since this bug is well know. It is me who modified
Resample2D. As I told in my previous mail, this bug is know since a long time.
I
already reported it to JAI guys 3 years ago (I'm the author of this bug report):
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4906854
and I verified that this bug is fixed in JAI 1.1.3 (even if the above bug
report
doesn't said anything about 1.1.3). Note that this is a Windows only bug. JAI
1.1.2 work well on Unix.
I wrote Registry.allowNativeAcceleration(...) in the coverage module a while
ago
as a workaround, but it leads to a hugly hack in Resample2D which hurt
performances. I removed the hack since it is not needed anymore with JAI 1.1.3.
If we want to reintroduce the hack, I could do so but we should do it in a way
that affect only JAI 1.1.2 on Windows platform.
Martin
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
