-
Enhancement
-
Resolution: Fixed
-
P4
-
9
-
b12
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8063506 | 8u45 | Thomas Schatzl | P4 | Resolved | Fixed | b01 |
JDK-8040673 | 8u40 | Thomas Schatzl | P4 | Resolved | Fixed | b01 |
JDK-8070791 | emb-8u47 | Thomas Schatzl | P4 | Resolved | Fixed | team |
The remembered set iterator has not been adapted to this change, still using the hash table and collision list to iterate over the fine remembered set.
Change this, as iterating over the direct links is faster than going through the hash table: also micro benchmarks (doing nothing per card) show good improvements in overall iteration time.
- backported by
-
JDK-8040673 Use the "next" field to iterate over fine remembered instead of using the hash table
-
- Resolved
-
-
JDK-8063506 Use the "next" field to iterate over fine remembered instead of using the hash table
-
- Resolved
-
-
JDK-8070791 Use the "next" field to iterate over fine remembered instead of using the hash table
-
- Resolved
-
- relates to
-
JDK-7182260 G1: Fine grain RSet freeing bottleneck
-
- Resolved
-