Details
-
Bug
-
Status: Resolved
-
P3
-
Resolution: Fixed
-
15, 16
-
b30
Backports
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8248592 | 16 | Markus Grönlund | P3 | Resolved | Fixed | b04 |
JDK-8250126 | 15.0.2 | Markus Grönlund | P3 | Resolved | Fixed | b01 |
JDK-8250425 | 15.0.1 | Markus Grönlund | P3 | Resolved | Fixed | b03 |
Description
Before
With
This has caused the access time to become a function of the number of buffers, indirectly becoming a function of the number of concurrent threads.
On systems with a high number of parallel threads, this becomes problematic.
We should address both of these aspects by using two mspaces instead of a single global JfrCheckpointMspace. One mspace is to be specialized for threads and one is to be specialized for the global access. This is similar to the layout of JfrStorage.
Attachments
Issue Links
- backported by
-
JDK-8248592 Poor scalability in JfrCheckpointManager when using many threads after JDK-8242088
-
- Resolved
-
-
JDK-8250126 Poor scalability in JfrCheckpointManager when using many threads after JDK-8242088
-
- Resolved
-
-
JDK-8250425 Poor scalability in JfrCheckpointManager when using many threads after JDK-8242088
-
- Resolved
-
- relates to
-
JDK-8242088 Replace mutually exclusive lists with concurrent alternatives
-
- Resolved
-
-
JDK-8234595 JfrBuffer::reinitialize failed "assert(!lease()) failed: invariant"
-
- Resolved
-
-
JDK-8247965 Two JFR tests failing in Loom repo
-
- Resolved
-