[MPlayer-dev-eng] optimized colorspace for ALTIVEC enabled platforms

Michael Niedermayer michaelni at gmx.at
Thu Mar 11 18:23:55 CET 2004


Hi

On Tuesday 09 March 2004 06:21, Marc Hoffman wrote:
> On Mar 9, 2004, at 12:06 AM, D Richard Felker III wrote:
> > On Mon, Mar 08, 2004 at 10:21:55PM -0500, Marc Hoffman wrote:
> >> I would like to submit my patch for postproc.
> >>
> >>
> >>
> >>
> >>
> >>
> >> /*
> >>   marc.hoffman at analog.com    March 8, 2004
> >>
> >>   Altivec Acceleration for Color Space Conversion revision 0.1
> >>
> >>   convert I420 YV12 to RGB in various formats,
> >>     it rejects images that are not in 420 formats
> >>     it rejects images that don't have widths of multiples of 16
> >>     it rejects images that don't have heights of multiples of 2
> >>   reject defers to C simulation codes.
> >
> > It shouldn't reject the whole image. Instead just convert the largest
> > region it can (simple by playing around with width/height/stride) and
> > then call the C version to process the small strips that remain.
> >
> > Rich
> >
> > _______________________________________________
> > MPlayer-dev-eng mailing list
> > MPlayer-dev-eng at mplayerhq.hu
> > http://mplayerhq.hu/mailman/listinfo/mplayer-dev-eng
>
> Ok I will work on the next version.  I'm thinking i can just pad the
> last block threw the loop and use the same code filling the vector
> register with zeros.  I will spend sometime on this.
>
> Can you put this under CVS while I work out the loose ends?
no, it uses constant coefficients so it will probably break 
brightness/contrast/saturation adjustment and it only checks for YUV 4:2:2 
and accepts all others, i also belive it uses an incorrect range for luma and 
chroma, u should use the coefficients from the SwsContext, and if u think 
they are wrong or u need them in a different format then fix/extend 
sws_setColorspaceDetails()

[...]
-- 
Michael
level[i]= get_vlc(); i+=get_vlc();		(violates patent EP0266049)
median(mv[y-1][x], mv[y][x-1], mv[y+1][x+1]);	(violates patent #5,905,535)
buf[i]= qp - buf[i-1];				(violates patent #?)
for more examples, see http://mplayerhq.hu/~michael/patent.html
stop it, see http://petition.eurolinux.org & http://petition.ffii.org/eubsa/en




More information about the MPlayer-dev-eng mailing list