Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-30 Thread serguei.spit...@oracle.com
On 3/30/16 01:50, Dmitry Samersoff wrote: Serguei, Thank you for review! These tests are not part of any testing process. I plan to review and fix/delete it as necessary under a separate CR (JDK-8146200) Ok, reviewed. Thanks, Serguei -Dmitry On 2016-03-30 11:28, serguei.spit...@oracle.

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-30 Thread Dmitry Samersoff
Serguei, Thank you for review! These tests are not part of any testing process. I plan to review and fix/delete it as necessary under a separate CR (JDK-8146200) -Dmitry On 2016-03-30 11:28, serguei.spit...@oracle.com wrote: > Hi Dmitry, > > Looks good in general. > > The following files in t

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-30 Thread serguei.spit...@oracle.com
Hi Dmitry, Looks good in general. The following files in the jdk.hotspot.agent/test folder may need to be updated too: src/jdk.hotspot.agent/test/jdi/runjdb.sh src/jdk.hotspot.agent/test/jdi/runsa.sh Thanks, Serguei On 3/29/16 02:41, Alan Bateman wrote: On 29/03/2016 09:54, Dmitry Sa

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-29 Thread Alan Bateman
On 29/03/2016 09:54, Dmitry Samersoff wrote: Alan, done (in-place) http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ -Dmitry Looks okay to me. At some point (and nothing to do with this change) is that we should add type parameters to the other methods like getPath, classPath

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-29 Thread Dmitry Samersoff
Alan, done (in-place) http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ -Dmitry On 2016-03-29 11:21, Alan Bateman wrote: > > > On 29/03/2016 09:15, Dmitry Samersoff wrote: >> Everybody, >> >> http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ >> >> Webrev updated (in-pl

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-29 Thread Alan Bateman
On 29/03/2016 09:15, Dmitry Samersoff wrote: Everybody, http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ Webrev updated (in-place, press shift-reload). VirtualMachineImpl.bootClassPath() returns Collections.emptyList() rather than throw an exception. This looks okay except for

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-29 Thread Dmitry Samersoff
Everybody, http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ Webrev updated (in-place, press shift-reload). VirtualMachineImpl.bootClassPath() returns Collections.emptyList() rather than throw an exception. -Dmitry On 2016-03-28 21:15, Dmitry Samersoff wrote: > Everybody, > > Ple

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread serguei.spit...@oracle.com
Ok, thanks! Serguei On 3/28/16 14:08, Mandy Chung wrote: This fix can go in jdk9/dev as the module system has already been integrated in jdk9. It doesn’t depend on anything in jigsaw/jake. Mandy On Mar 28, 2016, at 1:54 PM, serguei.spit...@oracle.com wrote: Hi Dmitry, This bug is for Jigs

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Mandy Chung
This fix can go in jdk9/dev as the module system has already been integrated in jdk9. It doesn’t depend on anything in jigsaw/jake. Mandy > On Mar 28, 2016, at 1:54 PM, serguei.spit...@oracle.com wrote: > > Hi Dmitry, > > This bug is for Jigsaw M4. > Should it be fixed in the Jigsaw repositor

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread serguei.spit...@oracle.com
Hi Dmitry, This bug is for Jigsaw M4. Should it be fixed in the Jigsaw repository then? My understanding is that the M4 will be integrated into the jdk9-dev the same was as it was for M3. Thanks, Serguei On 3/28/16 11:15, Dmitry Samersoff wrote: Everybody, Please review: http://cr.openjdk

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Dmitry Samersoff
Alan, > Returning an empty List should be fine. OK. I'll change it. -Dmitry On 2016-03-28 22:45, Alan Bateman wrote: > > > On 28/03/2016 20:35, Dmitry Samersoff wrote: >> Alan, >> >>> Shouldn't VirtualMachineImpl.bootClassPath return emptyList() rather >>> than throw an implementation specif

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Alan Bateman
On 28/03/2016 20:35, Dmitry Samersoff wrote: Alan, Shouldn't VirtualMachineImpl.bootClassPath return emptyList() rather than throw an implementation specific exception? This function never used anymore. So I would prefer an exception to make any attempt to call it fails immediately. Would R

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Dmitry Samersoff
Alan, > Shouldn't VirtualMachineImpl.bootClassPath return emptyList() rather > than throw an implementation specific exception? This function never used anymore. So I would prefer an exception to make any attempt to call it fails immediately. Would RuntimeException better? -Dmitry On 2016-03-2

Re: RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Alan Bateman
On 28/03/2016 19:15, Dmitry Samersoff wrote: Everybody, Please review: http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ All references to boot class path in SA code is removed. VirtualMachineImpl.bootClassPath() is the part of sun.com.jdi.PathSearchingVirtualMachine interface,

RFR(S) JDK-8150973 Hotspot agent use of sun.boot.class.path needs to be updated for Jigsaw

2016-03-28 Thread Dmitry Samersoff
Everybody, Please review: http://cr.openjdk.java.net/~dsamersoff/JDK-8150973/webrev.02/ All references to boot class path in SA code is removed. VirtualMachineImpl.bootClassPath() is the part of sun.com.jdi.PathSearchingVirtualMachine interface, cleanup of this interface requires much more effo