[FFmpeg-devel] [PATCH 2/4] lavc/aarch64: Provide neon implementation of nsse8

[Martin Storsjö]

On Mon, 26 Sep 2022, Grzegorz Bernacki wrote:

> Add vectorized implementation of nsse8 function.
>
> Performance comparison tests are shown below.
> - nsse_1_c: 256.0
> - nsse_1_neon: 82.7
>
> Benchmarks and tests run with checkasm tool on AWS Graviton 3.
>
> Signed-off-by: Grzegorz Bernacki <gjb at semihalf.com>
> ---
> libavcodec/aarch64/me_cmp_init_aarch64.c | 15 ++++
> libavcodec/aarch64/me_cmp_neon.S         | 99 ++++++++++++++++++++++++
> 2 files changed, 114 insertions(+)

Looks reasonable to me, but do check to make sure there's no tabs.

// Martin