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

2022-12-04 Thread Junxian Zhu
Compile with generic mid_pred function written in the old way on the mips platform, and objdump its .o file, it will show like this, with more instructions and branch jump. ```code mid_pred(int, int, ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.or

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

2022-12-01 Thread Michael Niedermayer
On Thu, Dec 01, 2022 at 02:42:20PM +0800, Junxian Zhu wrote: > From: Junxian Zhu > > 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,

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

2022-11-30 Thread Junxian Zhu
From: Junxian Zhu 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