Please review this additional cleanup of use of TRAPS in hotspot runtime code.
The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and Windows and
Mach5 tiers 3-5 on Linux x64.
Thanks, Harold
-
Commit messages:
- 8264711: More runtime TRAPS cleanups
Changes: https://g
On Mon, 5 Apr 2021 17:57:13 GMT, Harold Seigel wrote:
> Please review this additional cleanup of use of TRAPS in hotspot runtime
> code. The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and
> Windows and Mach5 tiers 3-5 on Linux x64.
>
> Thanks, Harold
Looks good Harold. One m
On Mon, 5 Apr 2021 17:57:13 GMT, Harold Seigel wrote:
> Please review this additional cleanup of use of TRAPS in hotspot runtime
> code. The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and
> Windows and Mach5 tiers 3-5 on Linux x64.
>
> Thanks, Harold
Hi Harold,
I looked at t
On Mon, 15 Mar 2021 02:01:12 GMT, Yasumasa Suenaga wrote:
> NPE was thrown when I set server name prefix for debugd as following:
>
> $ jhsdb -J-Dsun.jvm.hotspot.rmi.serverNamePrefix=test debugd --pid 781
> Attaching to process ID 781 and starting RMI services, please wait...
> Error attaching t
> Please review this additional cleanup of use of TRAPS in hotspot runtime
> code. The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and
> Windows and Mach5 tiers 3-5 on Linux x64.
>
> Thanks, Harold
Harold Seigel has updated the pull request incrementally with one additional
com
On Mon, 5 Apr 2021 19:11:49 GMT, Patricio Chilano Mateo
wrote:
>> Harold Seigel has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Undo change to ObjectSynchronizer::jni_exit()
>
> Hi Harold,
>
> I looked at the changes to synchronization
On Mon, 5 Apr 2021 01:45:15 GMT, Yasumasa Suenaga wrote:
> [JDK-8263670](https://bugs.openjdk.java.net/browse/JDK-8263670) has
> introduced `SATestUtils::validateSADebugDPrivileges` to check privileges for
> on OS X.
>
> https://github.com/openjdk/jdk/blob/a209ed01bafb7721d6b733d1c4bd3f1776463
On Mon, 5 Apr 2021 00:27:35 GMT, Yasumasa Suenaga wrote:
>> src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/SALauncher.java line 89:
>>
>>> 87: if (canConnectToRemote) {
>>> 88: System.out.println(" or jhsdb " + mode + "
>>> --connect debugserver");
>>> 89:
On Mon, 5 Apr 2021 00:25:38 GMT, Yasumasa Suenaga wrote:
>> `jhsdb debugd` will start RMI registry by default, but we want to prevent it
>> due to port confliction in some cases. We can control it with
>> `sun.jvm.hotspot.rmi.startRegistry` system property. However we have no way
>> to set it
[moving to serviceability-dev]
Hi,
I'm not sure if Address hashcodes are even used by SA, and if they are,
I doubt this slightly improved hash would make a noticeable difference.
However, if you want to pursue this change just to get started with
making OpenJDK contributions, I'm ok with that
On Mon, 5 Apr 2021 20:27:53 GMT, Harold Seigel wrote:
>> Please review this additional cleanup of use of TRAPS in hotspot runtime
>> code. The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and
>> Windows and Mach5 tiers 3-5 on Linux x64.
>>
>> Thanks, Harold
>
> Harold Seigel has
On Mon, 5 Apr 2021 20:24:31 GMT, Harold Seigel wrote:
> Thanks Lois and Patricio for reviewing the change!
> I removed my bogus change to ObjectSynchronizer::jni_exit() which also made
> calls consistent in jfrJavaSupport.cpp.
> Harold
Thanks Harold! Fix looks good.
-
PR: https://g
On Mon, 5 Apr 2021 20:27:53 GMT, Harold Seigel wrote:
>> Please review this additional cleanup of use of TRAPS in hotspot runtime
>> code. The changes were tested with Mach5 tiers 1-2 on Linux, Mac OS, and
>> Windows and Mach5 tiers 3-5 on Linux x64.
>>
>> Thanks, Harold
>
> Harold Seigel has
On 6/04/2021 5:14 am, Patricio Chilano Mateo wrote:
src/hotspot/share/prims/jni.cpp line 2738:
2736:
2737: Handle obj(THREAD, JNIHandles::resolve_non_null(jobj));
2738: ObjectSynchronizer::jni_exit(THREAD->as_Java_thread(), obj());
Here we would return JNI_ERR if we throw IMSE from jni_ex
On Mon, 5 Apr 2021 01:45:15 GMT, Yasumasa Suenaga wrote:
> [JDK-8263670](https://bugs.openjdk.java.net/browse/JDK-8263670) has
> introduced `SATestUtils::validateSADebugDPrivileges` to check privileges for
> on OS X.
>
> https://github.com/openjdk/jdk/blob/a209ed01bafb7721d6b733d1c4bd3f1776463
On Mon, 15 Mar 2021 02:01:12 GMT, Yasumasa Suenaga wrote:
> NPE was thrown when I set server name prefix for debugd as following:
>
> $ jhsdb -J-Dsun.jvm.hotspot.rmi.serverNamePrefix=test debugd --pid 781
> Attaching to process ID 781 and starting RMI services, please wait...
> Error attaching t
On Mon, 5 Apr 2021 20:57:34 GMT, Chris Plummer wrote:
> On the other hand, maybe you could go with just the id@debugserver:1234
> example and leave the other 3 off. The syntax does clearly show the id and
> port are optional.
Agree, I left `id@debugserver:1234` only in new commit.
---
On Mon, 5 Apr 2021 20:50:30 GMT, Chris Plummer wrote:
>> Yasumasa Suenaga has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Fix comments
>
> test/hotspot/jtreg/serviceability/sa/sadebugd/DebugdRmidTest.java line 51:
>
>> 49: JDKTo
> `jhsdb debugd` will start RMI registry by default, but we want to prevent it
> due to port confliction in some cases. We can control it with
> `sun.jvm.hotspot.rmi.startRegistry` system property. However we have no way
> to set it excepting system property. jhsdb should provide the way to set
On Tue, 30 Mar 2021 12:02:40 GMT, Hui Shi wrote:
> …ue to large TLAB size
>
> serviceability/jvmti/HeapMonitor tests intermittently fail when using
> PS/Serial GC, original test has implicit assumptions on TLAB size and depends
> on allocate fix amount of objects to consume TLAB and trigger ob
On Tue, 6 Apr 2021 00:56:08 GMT, Yasumasa Suenaga wrote:
>> test/hotspot/jtreg/serviceability/sa/sadebugd/DebugdRmidTest.java line 51:
>>
>>> 49: JDKToolLauncher rmidLauncher =
>>> JDKToolLauncher.createUsingTestJDK("rmid");
>>> 50:
>>> rmidLauncher.addToolArg("-J-Dsun.rmi.acti
On Tue, 6 Apr 2021 01:42:45 GMT, Chris Plummer wrote:
>> They are needed.
>>
>> If we don't give `execPolicy=none`, we can see warning on console.
>> SA code would run on rmid, so we need to add module.
>
> I was actually just referring to `--add-modules`, but github added a few
> lines before
On Tue, 6 Apr 2021 02:46:40 GMT, Yasumasa Suenaga wrote:
>> I was actually just referring to `--add-modules`, but github added a few
>> lines before the one I selected.
>>
>> I guess I'm not fully understanding what `rmid` is for in this context, and
>> how it relates to the `rmiregistry` comm
On Tue, 6 Apr 2021 03:16:29 GMT, Chris Plummer wrote:
>> SA has RMI remote object, it will be handled in RMI registry.
>> We can see following exception without `--add-modules`.
>>
>> $ jhsdb debugd --pid 40339 --disableregistry --hostname localhost
>> --registryport 1098
>> Attaching to proces
On Tue, 6 Apr 2021 03:32:25 GMT, Yasumasa Suenaga wrote:
>> Both of the above are using `--disableregistry`. Is that what you meant to
>> do? I would think that you would not want that on the first one.
>
> Sorry, the correct commands are as follows:
>
> * console 1 (start RMI registry)
> *
25 matches
Mail list logo
