* qatar/master: (22 commits)
rv40dsp x86: use only one register, for both increment and loop counter
rv40dsp: implement prescaled versions for biweight.
avconv: use default channel layouts when they are unknown
avconv: parse channel layout string
nutdec: K&R formatting cosmetics
vda: Signal 4 byte NAL headers to the decoder regardless of what's in the extradata
mem: Consistently return NULL for av_malloc(0)
vf_overlay: implement poll_frame()
vf_scale: support named constants for sws flags.
lavc doxy: add all installed headers to doxy groups.
lavc doxy: add avfft to the main lavc group.
lavc doxy: add remaining avcodec.h functions to a misc doxygen group.
lavc doxy: add AVPicture functions to a doxy group.
lavc doxy: add resampling functions to a doxy group.
lavc doxy: replace \ with /
lavc doxy: add encoding functions to a doxy group.
lavc doxy: add decoding functions to a doxy group.
lavc doxy: fix formatting of AV_PKT_DATA_{PARAM_CHANGE,H263_MB_INFO}
lavc doxy: add AVPacket-related stuff to a separate doxy group.
lavc doxy: add core functions/definitions to a doxy group.
...
Conflicts:
ffmpeg.c
libavcodec/avcodec.h
libavcodec/vda.c
libavcodec/x86/rv40dsp.asm
libavfilter/vf_scale.c
libavformat/nutdec.c
libavutil/mem.c
tests/ref/acodec/pcm_s24daud
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Quite often, the original weights are multiple of 512. By prescaling them
by 1/512 when they are computed (once per frame), no intermediate shifting
is needed, and no prescaling on each call either.
The x86 code already used that trick.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Provide MMX, SSE2 and SSSE3 versions, with a fast-path when the weights are
multiples of 512 (which is often the case when the values round up nicely).
*_TIMER report for the 16x16 and 8x8 cases:
C:
9015 decicycles in 16, 524257 runs, 31 skips
2656 decicycles in 8, 524271 runs, 17 skips
MMX:
4156 decicycles in 16, 262090 runs, 54 skips
1206 decicycles in 8, 262131 runs, 13 skips
MMX on fast-path:
2760 decicycles in 16, 524222 runs, 66 skips
995 decicycles in 8, 524252 runs, 36 skips
SSE2:
2163 decicycles in 16, 262131 runs, 13 skips
832 decicycles in 8, 262137 runs, 7 skips
SSE2 with fast path:
1783 decicycles in 16, 524276 runs, 12 skips
711 decicycles in 8, 524283 runs, 5 skips
SSSE3:
2117 decicycles in 16, 262136 runs, 8 skips
814 decicycles in 8, 262143 runs, 1 skips
SSSE3 with fast path:
1315 decicycles in 16, 524285 runs, 3 skips
578 decicycles in 8, 524286 runs, 2 skips
This means around a 4% speedup for some sequences.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
