[Ffmpeg-cvslog] r6231 - trunk/libavcodec/vp56.c
Tue Sep 12 12:01:24 CEST 2006
Guillaume POIRIER wrote:
> On 9/12/06, Luca Barbato <lu_zero at gentoo.org> wrote:
>> Guillaume POIRIER wrote:
>> > Hi,
>> > On 9/12/06, aurel <subversion at mplayerhq.hu> wrote:
>> >> don't use dsp.put_pixels_tab with unaligned ptr (fix artifacts
>> >> on altivec)
>> > Isn't there a way to make sure the arrays passed to put_pixelsXY are
>> > aligned?
>> > Even though the MMX version of put_pixel doesn't care, it would still
>> > be faster...
>> > As a side note, I imagine this has been discussed in another thread,
>> > but it's not so easy to find where because there's no link to it in
>> > the commit msg. :-P
>> > Guillaume
>> The problem is the stride that isn't 16 multiple, put_pixel should
>> handle unaligned load just fine.
> Should? As in: a version that handles non-16-multiple strides has to
> be written, right?
it is just a matter of putting the vec_lvsl in the loop instead of
outside as it is now. Not sure if it could be interleaved in a way it
won't change much the performance.
More information about the ffmpeg-cvslog