Re: [FFmpeg-devel] [PATCH v3 2/9] lavc/vp9dsp: R-V mc copy - Rémi Denis-Courmont

From: "Rémi Denis-Courmont" <remi@remlab.net>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH v3 2/9] lavc/vp9dsp: R-V mc copy
Date: Tue, 14 May 2024 20:02:26 +0300
Message-ID: <27673988.B1Vlz7fvLz@basile.remlab.net> (raw)
In-Reply-To: <CAEa-L+vAnu3-q7aBKQ_CpPejsqYFuHfV00taMDTrJ=3H15w07w@mail.gmail.com>

Le tiistaina 14. toukokuuta 2024, 7.44.55 EEST flow gg a écrit :
> I am locally using:
>     if (bpp == 8 && (flags & AV_CPU_FLAG_RVI)) {
> this performs better on k230/banana_f3 than C.
> For email, refer to [FFmpeg-devel] [PATCH 2/2] lavc/vp8dsp: restrict RVI
> optimisations and change it to
>     if (bpp == 8 && (flags & AV_CPU_FLAG_RV_MISALIGNED)) {
> So no output, but I think the same modification should be made here?

I just can't get any benchmarks out of checkasm. Even if I comment out the 
MISALIGNED flag check, this is not reporting anything. I tested with only patch 
1/9 and 2/9, not the following. I don't know why.

-- 
雷米‧德尼-库尔蒙
http://www.remlab.net/

_______________________________________________
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".