>> From: gxw <guxiwei...@loongson.cn>
>> Optimize vp8 loop filter with mmi, four functions optimized:
>> 1. ff_vp8_h_loop_filter8uv_mmi.
>> 2. ff_vp8_v_loop_filter8uv_mmi.
>> 3. ff_vp8_h_loop_filter16_mmi.
>> 4. ff_vp8_v_loop_filter16_mmi.
>> Vp8 decoding speed improved about 50%(from 73fps to 110fps,
>> Tested on loongson 3A3000).
>Sounds great.
>Again: Is this meant to only work on loongson or also on other
>mips cores?

The mmi optimization indicate loongson 64bit SIMD optimization, can only work 
on loongson platform.

>> -#if 0
>I believe this change and similar ones in the rest of the patch
>are welcome but they are not related to the speed-up that this
>patch is about and should be in a separate patch.

OK, I will recover other comment blocks except those in the function 
optimizatized by this patch. 
ffmpeg-devel mailing list

Reply via email to