The test results show that changing mf2 to m1 in ff_avg_pixels8x8_rvv in vc1, or changing mf2/mf4 to m1 in vsetvlstatic8 in vp8, results in a 10-20% performance decrease on both k230 and banana_f3.
I think we should just continue using it as is... Rémi Denis-Courmont <[email protected]> 于2024年5月10日周五 23:34写道： > Le perjantaina 10. toukokuuta 2024, 11.22.53 EEST flow gg a écrit : > > Hi, I got BananaPi F3, made some fixes, updated in reply > > So... Does it benefit from halving the logical multiplier to process > fixed-sized > block as compared to C908, or can we stick to the same code regardless of > vector sizes? > > Also beware that K60 cores have in-order pipelines, so data dependencies > will > probably hurt more than on C908. > > -- > Rémi Denis-Courmont > http://www.remlab.net/ > > > > _______________________________________________ > ffmpeg-devel mailing list > [email protected] > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > [email protected] with subject "unsubscribe". > _______________________________________________ ffmpeg-devel mailing list [email protected] https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
