Skip to content

arm64: cdef: NEON optimized cdef filter function

Martin Storsjö requested to merge mstorsjo/dav1d:arm64-cdef into master
Speedup vs C code:     Cortex A53    A72    A73
cdef_filter_4x4_8bpc_neon:   4.62   4.48   4.76
cdef_filter_4x8_8bpc_neon:   4.82   4.80   5.08
cdef_filter_8x8_8bpc_neon:   5.29   5.33   5.79

I'll make the cdef_dir function separately in a separate MR afterwards, but this is the main cdef function showing up in profiles.

Edited by Martin Storsjö

Merge request reports