[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-serviceability-dev
Subject:    Re: 8218727: vmTestbase/nsk/jvmti/scenarios/events/EM04/em04t001/TestDescription.java
From:       Daniil Titov <daniil.x.titov () oracle ! com>
Date:       2019-03-27 19:42:19
Message-ID: 36D019B1-0876-4E25-BC33-CD0654AF0616 () oracle ! com
[Download RAW message or body]

Thank you, Serguei and Gary, for reviewing this change!

--Daniil


On 3/27/19, 5:20 AM, "Gary Adams" <gary.adams@oracle.com> wrote:

    Looks OK to me.
    
    On 3/27/19, 3:59 AM, serguei.spitsyn@oracle.com wrote:
    > Hi Daniil,
    >
    > The fix looks good.
    >
    > Thanks,
    > Serguei
    >
    >
    > On 3/26/19 20:32, Daniil Titov wrote:
    >> Please review the change that fixes the test when running with Graal on.
    >>
    >> The problem here is the lack of synchronization when accessing the 
    >> internal data structure (the list of events) in the agent procedure 
    >> and in the callbacks the test enables. That results in the agent proc 
    >> starts removing elements from the list of the events while the 
    >> callback handlers are still enabled, assuming that no new events 
    >> could be generated,  but in case of Graal it is not true. This causes 
    >> to the crash when the callback iterates over the list of events and 
    >> tries to access the element that was just removed by the agent 
    >> procedure.
    >>
    >> Webrev: http://cr.openjdk.java.net/~dtitov/8218727/webrev.01/
    >> Bug: https://bugs.openjdk.java.net/browse/JDK-8218727
    >>
    >> Thanks!
    >>
    >> --Daniil
    >>
    >>
    >>
    >
    
    


[prev in list] [next in list] [prev in thread] [next in thread]