Details
-
Sub-task
-
Resolution: Fixed
-
P4
-
21
-
b23
-
generic
-
generic
Backports
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8311607 | 17.0.9-oracle | Calvin Cheung | P4 | Resolved | Fixed | b02 |
JDK-8312363 | 17.0.9 | Goetz Lindenmaier | P4 | Resolved | Fixed | b01 |
Description
During the lengthy discussions and analysis of the following fix:
JDK-8305670 Performance regression in LockSupport.unpark with lots of idle threads
we concluded that a JavaThread* should be stored in the java.lang.Thread
object after the JavaThread* has been added to the main ThreadsList.
Publishing that JavaThread* early can lead to situations where the
JavaThread* can be accessed/used before it is protected by a
ThreadsListHandle which is not safe.
we concluded that a JavaThread* should be stored in the java.lang.Thread
object after the JavaThread* has been added to the main ThreadsList.
Publishing that JavaThread* early can lead to situations where the
JavaThread* can be accessed/used before it is protected by a
ThreadsListHandle which is not safe.
Attachments
Issue Links
- backported by
-
JDK-8311607 store a JavaThread* in the java.lang.Thread object after the JavaThread* is added to the main ThreadsList
-
- Resolved
-
-
JDK-8312363 store a JavaThread* in the java.lang.Thread object after the JavaThread* is added to the main ThreadsList
-
- Resolved
-
- blocks
-
JDK-8305670 Performance regression in LockSupport.unpark with lots of idle threads
-
- Resolved
-
- links to
-
Commit openjdk/jdk17u-dev/86e88878
-
Commit openjdk/jdk/ceca198e
-
Review openjdk/jdk17u-dev/1587
-
Review openjdk/jdk/13723
(2 links to)