[x264-devel] [patch] Microoptimize assembler code (2/n)

[Josef Zlomek]

Hi,

another microoptimization for assembler code is to parallelize
the code for computing absolute value of MMX register.
When the instructions for abs of one reg are interleaved
with instructions for abs of second reg, the dependencies have
smaller impact so the code is faster.

The patch is located at http://zlomek.jikos.cz/x264/x264-abs.patch

Josef Zlomek

-- 
This is the x264-devel mailing-list
To unsubscribe, go to: http://developers.videolan.org/lists.html