diff options
author | Michael Niedermayer <michaelni@gmx.at> | 2011-11-12 02:50:25 +0100 |
---|---|---|
committer | Michael Niedermayer <michaelni@gmx.at> | 2011-11-12 02:50:25 +0100 |
commit | 29582df797745fa6c5eec22b007e4fd3a47e7dd9 (patch) | |
tree | 59c487e218e4f750b48151d548db9091236d096f /libavcodec/dsputil.h | |
parent | 6761b6b825c4aafff311a180a09c7013288480aa (diff) | |
parent | 29ae0565d98bb41e54fc74f74c330d3214825f47 (diff) | |
download | ffmpeg-29582df797745fa6c5eec22b007e4fd3a47e7dd9.tar.gz |
Merge remote-tracking branch 'qatar/master'
* qatar/master:
vble: remove vble_error_close
VBLE Decoder
tta: use an integer instead of a pointer to iterate output samples
shorten: do not modify samples pointer when interleaving
mpc7: only support stereo input.
dpcm: do not try to decode empty packets
dpcm: remove unneeded buf_size==0 check.
twinvq: add SSE/AVX optimized sum/difference stereo interleaving
vqf/twinvq: pass vqf COMM chunk info in extradata
vqf: do not set bits_per_coded_sample for TwinVQ.
twinvq: check for allocation failure in init_mdct_win()
swscale: add padding to conversion buffer.
rtpdec: Simplify finalize_packet
http: Handle proxy authentication
http: Print an error message for Authorization Required, too
AVOptions: don't return an invalid option when option list is empty
AIFF: add 'twos' FourCC for the mux/demuxer (big endian PCM audio)
Conflicts:
libavcodec/avcodec.h
libavcodec/tta.c
libavcodec/vble.c
libavcodec/version.h
libavutil/opt.c
libswscale/utils.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/dsputil.h')
-rw-r--r-- | libavcodec/dsputil.h | 17 |
1 files changed, 17 insertions, 0 deletions
diff --git a/libavcodec/dsputil.h b/libavcodec/dsputil.h index 92e081c411..a44c146af9 100644 --- a/libavcodec/dsputil.h +++ b/libavcodec/dsputil.h @@ -451,6 +451,23 @@ typedef struct DSPContext { */ void (*butterflies_float)(float *restrict v1, float *restrict v2, int len); + /** + * Calculate the sum and difference of two vectors of floats and interleave + * results into a separate output vector of floats, with each sum + * positioned before the corresponding difference. + * + * @param dst output vector + * constraints: 16-byte aligned + * @param src0 first input vector + * constraints: 32-byte aligned + * @param src1 second input vector + * constraints: 32-byte aligned + * @param len number of elements in the input + * constraints: multiple of 8 + */ + void (*butterflies_float_interleave)(float *dst, const float *src0, + const float *src1, int len); + /* (I)DCT */ void (*fdct)(DCTELEM *block/* align 16*/); void (*fdct248)(DCTELEM *block/* align 16*/); |