[x264-devel] x86: Add asm for mbtree fixed point conversion

[Henrik Gramner]

x264 | branch: master | Henrik Gramner <henrik at gramner.com> | Fri Mar  4 17:53:08 2016 +0100| [c82c7374938f4342971adf8b2495c3a1bbe621c4] | committer: Henrik Gramner

x86: Add asm for mbtree fixed point conversion

The QP offsets of each macroblock are stored as floats internally and
converted to big-endian Q8.8 fixed point numbers when written to the 2-pass
stats file, and converted back to floats when read from the stats file.

Add SSSE3 and AVX2 implementations for conversions in both directions.

About 8x faster than C on Haswell.

> http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=c82c7374938f4342971adf8b2495c3a1bbe621c4
---

 common/mc.c           |   15 +++++++
 common/mc.h           |    3 +-
 common/x86/mc-a2.asm  |  113 +++++++++++++++++++++++++++++++++++++++++++++++++
 common/x86/mc-c.c     |   10 +++++
 encoder/ratecontrol.c |   10 +----
 tools/checkasm.c      |   54 +++++++++++++++++++++++
 6 files changed, 196 insertions(+), 9 deletions(-)

Diff:   http://git.videolan.org/gitweb.cgi/x264.git/?a=commitdiff;h=c82c7374938f4342971adf8b2495c3a1bbe621c4