Alan, What would be, the most suitable location for the test? I'm going to make sure this new test does not duplicate any existing one and, if not, will change the patch accordingly then.
Best regards, Alexander ----- Original Message ----- From: alan.bate...@oracle.com To: alexander.kulyakh...@oracle.com, serviceability-dev@openjdk.java.net Sent: Wednesday, April 6, 2016 12:51:44 PM GMT +03:00 Iraq Subject: Re: RFR:8153584:New jtreg test to verify PathSearchingVirutalMachine.bootClassPath() behaviour On 06/04/2016 10:46, Alexander Kulyakhtin wrote: > Hi, > > Could you, please, review, the following test-only changes (adding a new > test). > > CR: https://bugs.openjdk.java.net/browse/JDK-8153584 "New jtreg test to > verify PathSearchingVirutalMachine.bootClassPath() behaviour" > Webrev: http://cr.openjdk.java.net/~akulyakh/8153584/ > > The new test verifies that in the JDK9 > PathSearchingVirutalMachine.bootClassPath() always returns an empty string > list, regardless of the, now obsolete, sun.boot.class.path property. > In the JDK 8 bootClassPath() used to be verified by a closed test (therefore > the CR is closed). We are changing the test and moving it to the open as it > verifies the open, public API. > The tests for JDI are in the jdk repo, would it be saner to put additional tests there so that they aren't split between two repos? -Alan
