Re: [gentoo-dev] MPlayer cpu USE flags zaniness

[Michael Marineau]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chris White wrote:
> My God.. I love this part:
>
>    Ok, so I had some users complain about mmx2 being showed as enabled
> in mplayer despite the lack of mmxext in /proc/cpuinfo.  Therefore I
> decided to get upstream help.  Here's what we got...:
>
> mmx2 = sse
> sse = mmx2
>
> but in mplayer's configure script:
>
> mmx2 = sse's integer logic
> sse = sse's floating point stuff
>
> so mmx2 is NOT a flag for mmx2 support.. it's a flag for integer portion
> of sse...
>
> ok, so at this point I'm confused as to how to approach this.  What
> about AMD?  Do they use mmxext for stuff like that.. or sse2 extensions
> (as someone mentioned).  My plan is to have sse USE flag enable mmx2
> only (since it's "the best") and disable sse (since it's the same thing
> but "worse").  However, that only takes care of intel.. what about AMD
> processors though?  They deal with all this stuff differently.
> Recomendations are welcome at this point.. I'm going to get some asprin...
>
What's the point of a use flag if you purposfully use the sse useflag to
do something different?  this sort of thing would make more sense:

sse = sse's integer logic
sse = sse's floating point stuff
mmx2 = mmx2

then if both mmx2 and sse are set, then just use mmx2 and disable sse as
both are hardly needed.  but I'm a little confused, if the mplayer
config script uses mmx2 to enaple part of sse, is there something else
that enables real mmx2?

- --
Michael Marineau
marineam@×××××××××.edu
Oregon State University
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFCRlmYiP+LossGzjARAjVBAKDGgcN9z/ccAuevOZUWzIeroGvrxACdEPYk
02ta3WyorgPwhHqg6ANzIUw=
=VrPu
-----END PGP SIGNATURE-----
--
gentoo-dev@g.o mailing list