diff options
| author | Ronald S. Bultje <rsbultje@gmail.com> | 2010-10-05 22:06:18 +0000 |
|---|---|---|
| committer | Ronald S. Bultje <rsbultje@gmail.com> | 2010-10-05 22:06:18 +0000 |
| commit | dd68d4db430089bff70f62a38cb84e07d183a435 (patch) | |
| tree | 80651f97cce9f9c3073a9fb8cf0695849c8abf7b /libavcodec/x86/dsputil_mmx.c | |
| parent | 2f412421e8bb1927fcd866777dfedf223c9dfdfb (diff) | |
| download | ffmpeg-dd68d4db430089bff70f62a38cb84e07d183a435.tar.gz | |
MMX, MMX2, SSE2 and SSSE3 optimizations for pred16x16/8x8_plane H264 intra
prediction (plus some with different rounding for svq3/rv40). Speedup (for
SSSE3) about ~6-fold, 3.6% faster overall with cathedral sample.
Originally committed as revision 25361 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec/x86/dsputil_mmx.c')
| -rw-r--r-- | libavcodec/x86/dsputil_mmx.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c index 3f00762a86..f5fe0f16a9 100644 --- a/libavcodec/x86/dsputil_mmx.c +++ b/libavcodec/x86/dsputil_mmx.c @@ -48,6 +48,7 @@ DECLARE_ALIGNED(16, const xmm_reg, ff_pw_8 ) = {0x0008000800080008ULL, 0x00080 DECLARE_ALIGNED(16, const xmm_reg, ff_pw_9 ) = {0x0009000900090009ULL, 0x0009000900090009ULL}; DECLARE_ALIGNED(8, const uint64_t, ff_pw_15 ) = 0x000F000F000F000FULL; DECLARE_ALIGNED(16, const xmm_reg, ff_pw_16 ) = {0x0010001000100010ULL, 0x0010001000100010ULL}; +DECLARE_ALIGNED(16, const xmm_reg, ff_pw_17 ) = {0x0011001100110011ULL, 0x0011001100110011ULL}; DECLARE_ALIGNED(16, const xmm_reg, ff_pw_18 ) = {0x0012001200120012ULL, 0x0012001200120012ULL}; DECLARE_ALIGNED(8, const uint64_t, ff_pw_20 ) = 0x0014001400140014ULL; DECLARE_ALIGNED(16, const xmm_reg, ff_pw_27 ) = {0x001B001B001B001BULL, 0x001B001B001B001BULL}; |