> Even without Zvbb's widening shift, widening multiplication is probably
faster
here.
Updated, it has indeed gotten faster.
于2024年10月29日周二 00:44写道：
> From: sunyuechi
>
> k230
> banana_f3
> put_chroma_pixels_8_4x4_c:
From: sunyuechi
k230
banana_f3
put_chroma_pixels_8_4x4_c: 63.5 ( 1.00x)59.2 (
1.00x)
put_chroma_pixels_8_4x4_rvv_i32:26.5 ( 2.39x)27.8 (
2.14x)
put_chroma_pixels_8_8
Le perjantaina 11. lokakuuta 2024, 13.38.42 EEST u...@foxmail.com a écrit :
> From: sunyuechi
>
> k230
> banana_f3 put_chroma_pixels_8_4x4_c: 61.5 (
> 1.00x)69.5 ( 1.00x) put_chroma_pixels_8_4x4_r
From: sunyuechi
k230
banana_f3
put_chroma_pixels_8_4x4_c: 61.5 ( 1.00x)69.5 (
1.00x)
put_chroma_pixels_8_4x4_rvv_i32:33.8 ( 1.82x)38.2 (
1.82x)
put_chroma_pixels_8_8
