Currently the Vector "expand" API with subword types is not implemented on AArch64, resulting in large performance gap with other architectures. Although it does not have direct SIMD instructions to implement it efficiently, we can try to find a way to implement it with other instructions to improve the performance .
-
Xiaohong Gong
-
Xiaohong Gong
- Votes:
-
0 Vote for this issue
- Watchers:
-
1 Start watching this issue
- Created:
- Updated: