[FFmpeg-devel] [PATCH v2] avcodec/mathops: Optimize generic mid_pred function

[James Almer]

On 3/5/2023 11:38 PM, Junxian Zhu wrote:
> From: Junxian Zhu <zhujunxian at oss.cipunited.com>
> 
> Rewrite mid_pred function in generic mathops.h, reduce branch jump to improve performance. And because nowadays new version compiler can compile enough short asmbbely code as handwritting in these function, so remove specified optimized mips inline asmbbely mathops.h.
> 
> Signed-off-by: Junxian Zhu <zhujunxian at oss.cipunited.com>
> ---
>   libavcodec/mathops.h      | 20 ++++--------
>   libavcodec/mips/mathops.h | 67 ---------------------------------------
>   2 files changed, 6 insertions(+), 81 deletions(-)
>   delete mode 100644 libavcodec/mips/mathops.h
> 
> diff --git a/libavcodec/mathops.h b/libavcodec/mathops.h
> index c89054d6ed..526ffe0eec 100644
> --- a/libavcodec/mathops.h
> +++ b/libavcodec/mathops.h
> @@ -41,8 +41,6 @@ extern const uint8_t ff_zigzag_scan[16+1];
>   #   include "arm/mathops.h"
>   #elif ARCH_AVR32
>   #   include "avr32/mathops.h"
> -#elif ARCH_MIPS
> -#   include "mips/mathops.h"
>   #elif ARCH_PPC
>   #   include "ppc/mathops.h"
>   #elif ARCH_X86
> @@ -98,18 +96,12 @@ static av_always_inline unsigned UMULH(unsigned a, unsigned b){
>   #define mid_pred mid_pred
>   static inline av_const int mid_pred(int a, int b, int c)
>   {
> -    if(a>b){
> -        if(c>b){
> -            if(c>a) b=a;
> -            else    b=c;
> -        }
> -    }else{
> -        if(b>c){
> -            if(c>a) b=c;
> -            else    b=a;
> -        }
> -    }
> -    return b;
> +    int t0,t1,t2,t3;
> +    t0 = (a > b) ? b : a ;
> +    t1 = (a > b) ? a : b ;
> +    t2 = (t0 > c) ? t0 : c;
> +    t3 = (t1 > t2) ? t2 : t1;
> +    return t3;

int t0 = FFMIN(a, b);
int t1 = FFMAX(a, b);
int t2 = FFMAX(t0, c);
return FFMIN(t1, t2);