e know all our
testing could be executing only one of the cases. This concerns me but I
am not yet sure what to do about it.
Thanks,
David
-
Also I tested it on submit repo, then it has execution error
(mach5-one-ysuenaga-JDK-8242428-20200624-0054-12034717) due to
dependency error. So
On 6/24/20 6:53 PM, Yasumasa Suenaga wrote:
On 2020/06/25 10:00, Chris Plummer wrote:
On 6/24/20 5:17 PM, Yasumasa Suenaga wrote:
On 2020/06/25 3:22, Chris Plummer wrote:
On 6/24/20 12:01 AM, Yasumasa Suenaga wrote:
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64C
No objections.
Thanks,
David
On 24/06/2020 8:57 am, Ioi Lam wrote:
I've updated the patch to include just the fix for class initialization:
http://cr.openjdk.java.net/~iklam/jdk16/8246019-avoid-PerfClassTraceTime.v02/
Hopefully this part is non-controversial. We are unlikely to make
call_
On 2020/06/25 10:00, Chris Plummer wrote:
On 6/24/20 5:17 PM, Yasumasa Suenaga wrote:
On 2020/06/25 3:22, Chris Plummer wrote:
On 6/24/20 12:01 AM, Yasumasa Suenaga wrote:
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64CFrame is used for pstack and what I've been l
On 6/24/20 5:17 PM, Yasumasa Suenaga wrote:
On 2020/06/25 3:22, Chris Plummer wrote:
On 6/24/20 12:01 AM, Yasumasa Suenaga wrote:
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64CFrame is used for pstack and what I've been
looking at has been jstack, and in particul
On 2020/06/25 3:22, Chris Plummer wrote:
On 6/24/20 12:01 AM, Yasumasa Suenaga wrote:
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64CFrame is used for pstack and what I've been looking at has been jstack, and in
particular AMD64CurrentFrameGuess, which does use "la
Hi Paul,
The change looks good to me.
Thanks!
--Daniil
On 6/22/20, 8:48 AM, "serviceability-dev on behalf of Hohensee, Paul"
wrote:
Thanks very much for review, Volker. I'll file a follow-up issue.
One more reviewer, please? :)
Paul
On 6/22/20, 8:10 AM, "serviceability-
On 6/24/20 12:44, Mandy Chung wrote:
On 6/24/20 12:26 PM, serguei.spit...@oracle.com wrote:
On 6/24/20 05:25, David Holmes wrote:
Ah! The test class SimpleAgent is what is not public. That seems a
bug in the test.
There are many such tests.
We can break some of the existing agents by reje
On 6/24/20 12:26 PM, serguei.spit...@oracle.com wrote:
On 6/24/20 05:25, David Holmes wrote:
Ah! The test class SimpleAgent is what is not public. That seems a
bug in the test.
There are many such tests.
We can break some of the existing agents by rejecting non-public agent
classes.
I'm i
On 6/24/20 05:25, David Holmes wrote:
On 24/06/2020 8:14 pm, Alan Bateman wrote:
On 24/06/2020 10:57, David Holmes wrote:
But you are ignoring my next statement. If we remove the
setAccessible(true) then the premain method will not be accessible
as Serguei reported.
Exception in thread "ma
On 6/24/20 12:01 AM, Yasumasa Suenaga wrote:
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64CFrame is used for pstack and what I've been
looking at has been jstack, and in particular AMD64CurrentFrameGuess,
which does use "last java frame". I'm not sure why LinuxAMD
On 6/23/20 11:04 PM, Chris Plummer wrote:
On 6/23/20 7:07 PM, Daniel D. Daugherty wrote:
Just one more comment on this part:
> L220: System.out.println("CurrentFrameGuess: choosing
interpreter frame: sp = " +
> L221: spFound + ", fpFound = "
+ fp + ", pc
On 24/06/2020 8:14 pm, Alan Bateman wrote:
On 24/06/2020 10:57, David Holmes wrote:
But you are ignoring my next statement. If we remove the
setAccessible(true) then the premain method will not be accessible as
Serguei reported.
Exception in thread "main" java.lang.IllegalAccessException: c
On 24/06/2020 10:57, David Holmes wrote:
But you are ignoring my next statement. If we remove the
setAccessible(true) then the premain method will not be accessible as
Serguei reported.
Exception in thread "main" java.lang.IllegalAccessException: class
sun.instrument.InstrumentationImpl
(
On 24/06/2020 7:43 pm, Alan Bateman wrote:
On 24/06/2020 10:26, David Holmes wrote:
If we call setAccessible(true) then canAccess will return true.
Sure but the bug fix will remove the setAccessible(true) so canAccess
will do what he wants without needing to catch the exception. This is of
c
On 24/06/2020 10:26, David Holmes wrote:
If we call setAccessible(true) then canAccess will return true.
Sure but the bug fix will remove the setAccessible(true) so canAccess
will do what he wants without needing to catch the exception. This is of
course all a side show to the important issue
On 24/06/2020 10:22, David Holmes wrote:
The other way around. The setAccessible has been there long before the
module system existed, to allow a non-public premain. As a side-effect
when the module system came along it also disabled some module access
check (I'm not sure exactly what).
This
On 24/06/2020 5:33 pm, Alan Bateman wrote:
On 24/06/2020 07:55, serguei.spit...@oracle.com wrote:
Thank you for the example.
Yes, I'm working on a helpful message and was thinking to use the
Reflection method:
IllegalAccessException newIllegalAccessException(Class currentClass,
On 24/06/2020 4:22 pm, Alan Bateman wrote:
On 24/06/2020 06:50, David Holmes wrote:
It sounds like the use of setAccessible was hiding the need to disable
some module related access checks.
This will have a much bigger compatibility problem if agents with a
public premain suddenly stop wo
On 24/06/2020 4:24 pm, serguei.spit...@oracle.com wrote:
Hi David,
On 6/23/20 22:50, David Holmes wrote:
Hi Serguei,
On 24/06/2020 3:37 pm, serguei.spit...@oracle.com wrote:
Hi Larry,
Thank you for looking at this!
On 6/23/20 21:32, Laurence Cable wrote:
should we not consider some form o
On 24/06/2020 07:55, serguei.spit...@oracle.com wrote:
Thank you for the example.
Yes, I'm working on a helpful message and was thinking to use the
Reflection method:
IllegalAccessException newIllegalAccessException(Class currentClass,
Class m
On 2020/06/24 15:32, Chris Plummer wrote:
Hi Yasumasa ,
I think LinuxAMD64CFrame is used for pstack and what I've been looking at has been jstack, and in
particular AMD64CurrentFrameGuess, which does use "last java frame". I'm not sure why
LinuxAMD64CFrame does not look at "last java frame". M
22 matches
Mail list logo
