JDK-8239347 : Refactor Symbol to make _length a standalone field again
- Type: Enhancement
- Component: hotspot
- Sub-Component: runtime
- Affected Version: 12,13,14
- Priority: P4
- Status: Resolved
- Resolution: Fixed
- Submitted: 2020-02-18
- Updated: 2020-02-27
- Resolved: 2020-02-20
The Version table provides details related to the release that this issue/RFE will be addressed.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
| JDK 15 |
|---|
15 b12Fixed  |
Related Reports
|
Relates :
|
Description
JDK-8207359 merged Symbol::_refcount and _length into a single field. We should refactor this to instead pack _refcount and _identity_hash, leaving _length as a standalone field. _length is used in a lot more places, so not having to call a method to extract the value using shifts results in a code size reduction and potential speed-up. Except for a path only used when dumping CDS archives, _identity_hash is only accessed in one place, and there in conjunction with an access to length. External tools like async-profiler[1] have been found to depend on Symbol::_length, which broke and required a fix in JDK 12. I haven't found any tools that care about the _identity_hash and _refcount fields, though. [1] https://github.com/jvm-profiling-tools/async-profiler
Comments
|