diff options
author | Janne Grunau <janne-libav@jannau.net> | 2014-01-13 01:06:20 +0100 |
---|---|---|
committer | Janne Grunau <janne-libav@jannau.net> | 2014-03-09 00:45:33 +0100 |
commit | dfe224f377be3e45758c69d881ca7874b82d647a (patch) | |
tree | 1dc58e82ae81e068777ddf81e045eb42c8e1ceea /libavcodec/cabac.h | |
parent | 4506a854a4d846692ba71daeeff661dc214c8fa2 (diff) | |
download | ffmpeg-dfe224f377be3e45758c69d881ca7874b82d647a.tar.gz |
aarch64: get_cabac inline asm
Based on the x86 branchless get_cabac asm. get_cabac_noinline() gets
approximately 20% faster (no cycle counts available) compared to clang
from Xcode 5.1 beta5. More than 6% faster overall. A part of the overall
speedup might be explained by additional inlining of get_cabac().
Diffstat (limited to 'libavcodec/cabac.h')
-rw-r--r-- | libavcodec/cabac.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/libavcodec/cabac.h b/libavcodec/cabac.h index 7401bc96a9..426f338e34 100644 --- a/libavcodec/cabac.h +++ b/libavcodec/cabac.h @@ -31,6 +31,7 @@ #include "put_bits.h" +extern uint8_t ff_h264_cabac_tables[512 + 4*2*64 + 4*64 + 63]; #define H264_NORM_SHIFT_OFFSET 0 #define H264_LPS_RANGE_OFFSET 512 #define H264_MLPS_STATE_OFFSET 1024 |