[x264-devel] x86inc: AVX symmetry optimization

[Loren Merritt]

x264 | branch: master | Loren Merritt <pengvado at akuvian.org> | Sun Oct 23 23:15:11 2011 +0000| [4185ee883b04d9cee57a64fdebd153830b7b27ba] | committer: Jason Garrett-Glaser

x86inc: AVX symmetry optimization
3-arg AVX ops with a memory arg can only have it in src2,
whereas SSE emulation of 3-arg prefers to have it in src1 (i.e. the move).
So, if the op is symmetric and the wrong one is memory, swap them.
Eliminates redundant moves in some cases when using 3-operand without AVX with memory arguments.
Also fix movss and movsd in some cases, and flag shufps correctly as float.

> http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=4185ee883b04d9cee57a64fdebd153830b7b27ba
---

 common/x86/x86inc.asm |  365 +++++++++++++++++++++++++++---------------------
 1 files changed, 205 insertions(+), 160 deletions(-)

Diff:   http://git.videolan.org/gitweb.cgi/x264.git/?a=commitdiff;h=4185ee883b04d9cee57a64fdebd153830b7b27ba