jvmti spec says no events should be generated after the VM_DEATH.
http://java.sun.com/javase/6/docs/platform/jvmti/jvmti.html#VMDeath
It says: "No event will occur of after this event".
I suspect we are posting vm death event too soon. I think we should
post vm death after vm thread is destro
If that's true, then you do indeed need a mutex. The OrderAccess
methods only guarantee
visibility ordering. They don't guarantee exclusive access to shared
variables.
Paul
Hiroshi Yamauchi wrote:
Dan,
On Fri, Feb 13, 2009 at 11:39 AM, Daniel D. Daugherty
wrote:
I'm leaning toward ma
They looks related but it's hard for me to tell because I cannot see
the test cases :)
On Fri, Feb 13, 2009 at 2:13 PM, Daniel D. Daugherty
wrote:
> Hiroshi,
>
> I found a couple of good candidates:
>
> 6357005 4/3 'SingleStep' event fails assertion in
> 'src/share/vm/prims/jvmtiE
Dan,
On Fri, Feb 13, 2009 at 11:39 AM, Daniel D. Daugherty
wrote:
> I'm leaning toward making the two fields volatile and adding the
> appropriate OrderAccess::XXX calls. I vaguely remember an existing
> bug for events posting late. I'm going to see if I can find it.
Making the fields volatile
Hiroshi,
I found a couple of good candidates:
6357005 4/3 'SingleStep' event fails assertion in
'src/share/vm/prims/jvmtiEventController.cpp, 590'
6648438 4/4 src/share/vm/prims/jvmtiEnv.cpp:457 assert(phase
== JVMTI_PHASE_LIVE,"sanity check")
I suspect that
Hiroshi,
I'm leaning toward making the two fields volatile and adding the
appropriate OrderAccess::XXX calls. I vaguely remember an existing
bug for events posting late. I'm going to see if I can find it.
Dan
Hiroshi Yamauchi wrote:
Hi Dan,
On Wed, Feb 11, 2009 at 12:43 PM, Daniel D. Daugher
Hi Dan,
On Wed, Feb 11, 2009 at 12:43 PM, Daniel D. Daugherty <
daniel.daughe...@sun.com> wrote:
> I'll chime in on parts of this thread below.
>
>
>
> Tim Bell wrote:
>
>> I don't know the precise answer to this question, so I am forwarding it to
>> the Serviceability list to get a wider audienc
Changeset: 6ada6122dd4f
Author:mcimadamore
Date: 2009-02-13 11:57 +
URL: http://hg.openjdk.java.net/jdk7/tl/langtools/rev/6ada6122dd4f
6769027: Source line should be displayed immediately after the first diagnostic
line
Summary: Added support for customizing diagnostic output v
