-
Type:
Enhancement
-
Resolution: Unresolved
-
Priority:
P3
-
Affects Version/s: repo-valhalla
-
Component/s: hotspot
The current implementation of Arrays.fill intrinsic for flattened arrays is highly inefficient.
It's > 50x times slower than manual loop filling.
It's > 50x times slower than manual loop filling.
- relates to
-
JDK-8251971 [lworld] Improve performance of arraycopy intrinsics
-
- Open
-
-
JDK-8253160 [lworld] C2's superword optimization should vectorize flat inline type array accesses
-
- Open
-