Skip to content

SVE/SVE2 support

David Chen requested to merge david.chen/x264:SVE2-support into master

Update ASM code to support SVE/SVE2 instruction set for Aarch64 architectures

The aim of this merge request is to improve x264 performance by using SVE/SVE2 instruction set of Aarch64 architectures.

Only the parts of the code whose migration to SVE/SVE2 improves the performance were considered. For the rest, NEON instruction set is still being used.

It should be noted that only the ASM code has been migrated. Not the C/C++ intrinsics.

All unit and regression tests pass.

Edited by David Chen

Merge request reports

VideoLAN code repository instance