[x264-devel] commit: add AltiVec implementation of x264_pixel_var_16x16 and x264_pixel_var_8x8 (Guillaume Poirier )

[Guillaume POIRIER]

Hi,

On Mon, Jan 26, 2009 at 6:09 PM, Jason Garrett-Glaser
<darkshikari at gmail.com> wrote:
> On Mon, Jan 26, 2009 at 5:49 AM, Guillaume POIRIER
> <gpoirier at mplayerhq.hu> wrote:

> You have to actually set chroma_plane_size to the correct size, you
> know.  You've changed the stride, so you have to change your math to
> calculate the stride:
>
> chroma_plane_size = (frame->i_stride[1] * ( frame->i_lines[1] +
> 2*i_padv )); should do the job, I think.

Thanks, and sorry for the annoyance. I must admit I've been a bit lazy
and didn't spend enough time to figure this out myself.

This seems to be the right way to compute the chroma plane size. This
allows to encode my sample with the same MD5, so I assume it's
correct.

Attached patch has the chroma plane mod16 alignment, and the AltiVec
optimizations that Loren suggested.

Thanks a lot for all you guys help, it's much appreciated.

Cheers,

Guillaume
-- 
Only a very small fraction of our DNA does anything; the rest is all
comments and ifdefs.

P. J. O'Rourke  - "You can't get rid of poverty by giving people money."
-------------- next part --------------
A non-text attachment was scrubbed...
Name: variance_optimization.3.diff
Type: application/octet-stream
Size: 4200 bytes
Desc: not available
Url : http://mailman.videolan.org/pipermail/x264-devel/attachments/20090126/f59171ea/attachment.obj