<AWT Dev> <AWT dev> Review request for JDK-8165555: [macosx] VM crashes on second attempt to execute JCK interactive tests that use Robot (single JVM, agent)
manajit.halder at oracle.com
Tue Sep 27 08:51:21 UTC 2016
Thanks for the comment. “self.javaToMacKeyMap is retaining the reference.
Please review the modified code:
> On 23-Sep-2016, at 10:24 pm, Sergey Bylokhov <Sergey.Bylokhov at oracle.com> wrote:
> On 21.09.16 16:01, Manajit Halder wrote:
>> Hi Sergey,
>> Thanks for the comment.
>> Access to "instance" is not broken. The problem is with the dictionary
>> variable "javaToMacKeyMap" within the "instance" reference.
>> The dictionary is not getting initialised for the second time when
>> singleton method is called for the second time. The dictionary is
>> getting initialised during the first time singleton method is called and
>> hence crash was observed. Tried adding self.javaToMacKeyMap
>> but doesn’t solve the problem.
> I still suggest you to check "self.javaToMacKeyMap" at line 48 this should call generated setter which will retain the reference.
>> I propose two solutions to this problem:
>> Solution 1:
>> Reinitialise the dictionary every-time the singleton method is called.
>> dictionary is initialised multiple times (every time singleton method is
>> Solution 2:
>> Make the dictionary static.
>> Still dictionary need to be initialised multiple times.
>> No singleton method, just two static methods, one method to initialise
>> the dictionary and other to get the key form the dictionary.
>> Not prepared. Will be prepared if second solution accepted.
>> Test case:
>> There are no test cases. Problem can be reproduced by
>> executing following JCK test:
>> java -jar JCK-runtime-9/lib/jtjck.jar -mode:single -k:interactive
>>> On 20-Sep-2016, at 4:42 am, Sergey Bylokhov
>>> <Sergey.Bylokhov at oracle.com <mailto:Sergey.Bylokhov at oracle.com> <mailto:Sergey.Bylokhov at oracle.com <mailto:Sergey.Bylokhov at oracle.com>>> wrote:
>>> Hi, Manajit.
>>> It seems that after the fix "(CRobotKeyCodeMapping *) sharedInstance"
>>> returns the new object per invocation, so it is not really
>>> sharedInstance. I am not sure I understand what is wrong in the
>>> current code, from the my point of view this is a correct singleton.
>>> It it true that the problem is in access to broken "instance" and not
>>> to "javaToMacKeyMap" inside the "instance"? If not then
>>> "javaToMacKeyMap" should be changed to "self.javaToMacKeyMap".
>>> Do you have a test case to reproduce the bug?
>>> On 19.09.16 15:26, Manajit Halder wrote:
>>>> Hi All,
>>>> Kindly review the fix for JDK9.
>>>> https://bugs.openjdk.java.net/browse/JDK-8165555 <https://bugs.openjdk.java.net/browse/JDK-8165555>
>>>> http://cr.openjdk.java.net/~mhalder/8165555/webrev.00/ <http://cr.openjdk.java.net/~mhalder/8165555/webrev.00/>
>>>> [macosx] VM crashes on second attempt to execute JCK interactive tests
>>>> that use Robot (single JVM, agent)
>>>> While executing the JCK test for the second time the robot was getting
>>>> initialised once again and old instance of CRobotKeyCodeMapping was not
>>>> Crash was observed while trying to access invalid instance of
>>>> A new instance of CRobotKeyCodeMapping is created when robot is
>>> Best regards, Sergey.
> Best regards, Sergey.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the awt-dev