Hi, I've created a fix proposal for the issue I have reported in this bug: https://bugs.openjdk.java.net/browse/JDK-8150704
The webrev can be found here: http://cr.openjdk.java.net/~clanger/webrevs/8150704.1/ The Xalan parser would eventually run out of DTM IDs if xsl transformations involve lots of temporary result trees. Those are never released although they could. A testcase is included for this. I've also done some cleanups in the Xalan code and in the tests. Thanks in advance for looking at this :) Best regards Christoph
