Re: yuvdeinterlace

"Scott Smith" <[email protected]>
Newsgroups gmane.comp.video.mjpeg.devel
Message-ID <[email protected]>
On 2/23/07, stefan <[email protected]> wrote:
> Could someone explain me, why it must be aligned to 16 bytes? 8 bytes I
> would understand (quad-word) but 16?!? Is this really necessary?

I can't for Altivec, but I can for sse...

it fits in a cache line better.  SSE has two load instructions, movaps
and movups.  movaps is for things aligned on 16 byte boundaries;
movups is for things not aligned.  movups is slower.  Internally, I
believe movups is (or was at one point) implemented as two separate
16-byte aligned reads, with the appropriate shifts.  Something about
the interface to the cache only accepting reads withing 16-byte
boundaries.

My guess is Altivec doesn't even have movups.
-- 
        Scott

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.