[FFmpeg-devel] [PATCH 2/2] x86: hevc_mc: convert to ssse3

Mickaël Raulet mraulet at gmail.com
Sat Aug 23 17:48:05 CEST 2014


For avx2 I have some to push to the trunk, I did merge it yesterday with
all recent changes. But I don t remember what those tables looks like.

For 10 and 12bits, ssse3 should slow down the decoding since it uses 4 more
instructions in the loop.

Le samedi 23 août 2014, Christophe Gisquet <christophe.gisquet at gmail.com> a
écrit :

> Hi,
>
> 2014-08-23 16:52 GMT+02:00 James Almer <jamrial at gmail.com <javascript:;>>:
> >> -QPEL_TABLE  8, 8, b, sse4
> >> -QPEL_TABLE 10, 4, w, sse4
> >> -QPEL_TABLE 12, 4, w, sse4
> >> +QPEL_TABLE  8, 8, b, ssse3
> >> +QPEL_TABLE 10, 4, w, ssse3
> >> +QPEL_TABLE 12, 4, w, ssse3
> >
> > Do these need to be duplicated? You could just remove the suffix and let
> > every version of the function use the same tables.
>
> What do you mean by duplicated? That tables for 10 and 12 are?
>
> Otherwise, I'm leaving the ssse3/opt parameter for 2 reasons:
> - try to not perform changes that aren't required for it to work
> - I don't know how the avx2 code would look but indeed, if it had to
> use other tables, it would then use new macro(s)
>
> Maybe your suggestions could be delayed after the code is converted to
> ssse3? Probably there wouldn't be any, but patch conflicts would be
> cumbersome.
>
> --
> Christophe
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org <javascript:;>
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>


More information about the ffmpeg-devel mailing list