On Tue, 15 Feb 2022 06:10:57 GMT, Chris Plummer wrote:
> The changes look good. Have you verified that after your most recent changes
> you get the same sized hprof file as with your initial changes?
Thank you for the review.
I verified number of classes (HPROF_LOAD_CLASS and HPROF_GC_CLASS_DUM
On Tue, 15 Feb 2022 06:36:36 GMT, Chris Plummer wrote:
>> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 100:
>>
>>> 98: struct bag* deleted = deletedSignatures;
>>> 99: deletedSignatures = NULL;
>>> 100: debugMonitorExit(classTrackLock);
>>
>> This looks risky as the cri
> There are scenarios that JDWP agent can deadlock on `classTrackLock` monitor.
> Following is the scenario in bug report.
>
> **Java Thread**
> - loads a class and post `JVMTI_EVENT_CLASS_PREPARE` event
> - JDWP event callback handler calls `classTrack_processUnloads()` to handle
> the event
On Mon, 14 Feb 2022 23:26:17 GMT, David Holmes wrote:
>> Zhengyu Gu has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> David and Chris' comments
>
> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 215:
>
>> 213: classTrack_activa
On Tue, 15 Feb 2022 06:31:46 GMT, Chris Plummer wrote:
>> Zhengyu Gu has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> David and Chris' comments
>
> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 65:
>
>> 63: * handler may acq
On Tue, 15 Feb 2022 17:48:43 GMT, Zhengyu Gu wrote:
>> There are scenarios that JDWP agent can deadlock on `classTrackLock`
>> monitor. Following is the scenario in bug report.
>>
>> **Java Thread**
>> - loads a class and post `JVMTI_EVENT_CLASS_PREPARE` event
>> - JDWP event callback handle
On Tue, 15 Feb 2022 17:50:19 GMT, Zhengyu Gu wrote:
>> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 215:
>>
>>> 213: classTrack_activate(JNIEnv *env)
>>> 214: {
>>> 215: struct bag* new_bag = bagCreateBag(sizeof(char*), 1000);
>>
>> I don't think there can be any race during ac
Mostly mechanical changes; `git diff --color-moved=zebra` probably helps to see
the actual (small) change, such as dropping the `virtual` modifier, introducing
small helper methods to avoid naming collision, etc.
Test: tier1-6
-
Commit messages:
- flatten-cardgen
Changes: https:/
> There are scenarios that JDWP agent can deadlock on `classTrackLock` monitor.
> Following is the scenario in bug report.
>
> **Java Thread**
> - loads a class and post `JVMTI_EVENT_CLASS_PREPARE` event
> - JDWP event callback handler calls `classTrack_processUnloads()` to handle
> the event
On Tue, 15 Feb 2022 20:02:50 GMT, Chris Plummer wrote:
>> Zhengyu Gu has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> David and Chris' comments
>
> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 111:
>
>> 109: }
>> 110:
> There are scenarios that JDWP agent can deadlock on `classTrackLock` monitor.
> Following is the scenario in bug report.
>
> **Java Thread**
> - loads a class and post `JVMTI_EVENT_CLASS_PREPARE` event
> - JDWP event callback handler calls `classTrack_processUnloads()` to handle
> the event
On Tue, 15 Feb 2022 20:08:19 GMT, Chris Plummer wrote:
>> Zhengyu Gu has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> David and Chris' comments
>
> src/jdk.jdwp.agent/share/native/libjdwp/classTrack.c line 112:
>
>> 110: debugMonit
On Wed, 16 Feb 2022 00:17:46 GMT, Zhengyu Gu wrote:
> I am not sure if it is possible, but checking bagSize(deletedSignatures) == 0
> seems to race against classTrack_reset() where it does not take handlerLock
> lock.
I had thought of that too, but I think the way `classTrack_reset()` is calle
On Wed, 16 Feb 2022 00:53:27 GMT, Chris Plummer wrote:
>> I am not sure if it is possible, but checking `bagSize(deletedSignatures) ==
>> 0` seems to race against `classTrack_reset()` where it does not take
>> `handlerLock` lock.
>
>> I am not sure if it is possible, but checking bagSize(delete
On Wed, 16 Feb 2022 01:17:12 GMT, Zhengyu Gu wrote:
>>> I am not sure if it is possible, but checking bagSize(deletedSignatures) ==
>>> 0 seems to race against classTrack_reset() where it does not take
>>> handlerLock lock.
>>
>> I had thought of that too, but I think the way `classTrack_reset
15 matches
Mail list logo
