Re: [FFmpeg-devel] [PATCH v3 1/2][GSoC 2024] libavcodec/x86/vvc: Add AVX2 DMVR SAD functions for VVC

[Stone Chen <chen.stonechen@gmail.com>]

On Sat, May 18, 2024 at 11:33 AM Ronald S. Bultje <rsbultje@gmail.com>
wrote:

> Hi,
>
> On Tue, May 14, 2024 at 4:40 PM Stone Chen <chen.stonechen@gmail.com>
> wrote:
>
>> +    vvc_sad_8:
>> +        .loop_height:
>> +        movu              xm0, [src1q]
>> +        movu              xm1, [src2q]
>> +        MIN_MAX_SAD       xm2, xm0, xm1
>> +        vpmovzxwd          m1, xm1
>> +        vpaddd             m3, m1
>>
> [..]
>
>> +    vvc_sad_16_128:
>> +        .loop_height:
>>
> [..]
>
>> +        .loop_width:
>> +            movu              xm0, [src1q]
>> +            movu              xm1, [src2q]
>> +            MIN_MAX_SAD       xm2, xm0, xm1
>> +            vpmovzxwd          m1, xm1
>> +            vpaddd             m3, m1
>>
>
Hi Ronald,


> Wouldn't it be more efficient if the main loops did a full register worth
> at a time?
>
> vpbroadcastd m4, [pw_1]
> loop:
> movu m0, [src1q]
> movu m1, [src2q]
> MIN_MAX_SAD m2, m0, m1
> pmaddwd m1, m4
> paddd m3, m1
>
> (And then for w8, load 2 rows per iteration using movu xmN, [row0] and
> vinserti128 mN, [row1], 1.)
>
> Ronald
>

Thank you, I didn't know about the pmaddwd instruction, using it is
definitely more efficient!

Stone
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".