Mirror of https://git.ffmpeg.org/ffmpeg.git https://ffmpeg.org/

Go to file

Ben Avison 42c1cc35b7 armv6: Accelerate ff_imdct_half for general case (mdct_bits != 6) The previous implementation targeted DTS Coherent Acoustics, which only requires mdct_bits == 6. This relatively small size lent itself to unrolling the loops a small number of times, and encoding offsets calculated at assembly time within the load/store instructions of each iteration. In the more general case (codecs such as AAC and AC3) much larger arrays are used - mdct_bits == [8, 9, 11]. The old method does not scale for these cases, so more integer registers are used with non-unrolled versions of the loops (and with some stack spillage). The postrotation filter loop is still unrolled by a factor of 2 to permit the double-buffering of some VFP registers to facilitate overlap of neighbouring iterations. I benchmarked the result by measuring the number of gperftools samples that hit anywhere in the AAC decoder (starting from aac_decode_frame()) or specifically in ff_imdct_half_c / ff_imdct_half_vfp, for the same example AAC stream: Before After Mean StdDev Mean StdDev Confidence Change aac_decode_frame 2368.1 35.8 2117.2 35.3 100.0% +11.8% ff_imdct_half_* 457.5 22.4 251.2 16.2 100.0% +82.1% Signed-off-by: Michael Niedermayer <michaelni@gmx.at>		11 years ago
compat	…
doc	fix spelling errors	11 years ago
libavcodec	armv6: Accelerate ff_imdct_half for general case (mdct_bits != 6)	11 years ago
libavdevice	avdevice/pulse_audio_dec: reimplement using the non simple API	11 years ago
libavfilter	fix spelling errors	11 years ago
libavformat	avformat/asfdec: dvrms timestamps are pts not dts	11 years ago
libavresample	…
libavutil	fix spelling errors	11 years ago
libpostproc	…
libswresample	fix spelling errors	11 years ago
libswscale	swscale/x86/rgb2rgb_template: fix 1 byte overread in yuyvtoyuv420 and uyvytoyuv420	11 years ago
presets	presets: remove moldering iPod presets	11 years ago
tests	avcodec/roqvideo: use JPEG color range	11 years ago
tools	tools/yuvcmp: add HAVE_UNISTD_H around #include <unistd.h>	11 years ago
.gitignore	gitignore: add examples/transcoding	11 years ago
COPYING.GPLv2	…
COPYING.GPLv3	…
COPYING.LGPLv2.1	…
COPYING.LGPLv3	…
CREDITS	…
Changelog	avformat: Add image3 demuxers with format autodetection	11 years ago
INSTALL.md	…
LICENSE.md	…
MAINTAINERS	avfilter: add hqx filter (hq2x, hq3x, hq4x)	11 years ago
Makefile	…
README.md	…
RELEASE	…
arch.mak	…
cmdutils.c	cmdutils: ignore sws_flags if swscale is not enabled	11 years ago
cmdutils.h	…
cmdutils_common_opts.h	…
cmdutils_opencl.c	cmdutils_opencl: Use av_malloc_array()	11 years ago
common.mak	…
configure	configure: fix alphabetical order of previous commit	11 years ago
ffmpeg.c	ffmpeg: fix integer overflows with sub->*display_time	11 years ago
ffmpeg.h	ffmpeg: support setting AVStream.discard	11 years ago
ffmpeg_dxva2.c	…
ffmpeg_filter.c	ffmpeg: Fix choose_pixel_fmt() so it uses the correct encoding context	11 years ago
ffmpeg_opt.c	ffmpeg_opt: route -ab to -b:a	11 years ago
ffmpeg_vda.c	…
ffmpeg_vdpau.c	…
ffplay.c	ffplay: increase subtitle que size	11 years ago
ffprobe.c	ffprobe: add color_range and color_space to stream output	11 years ago
ffserver.c	ffserver: add HAVE_UNISTD_H around #include <unistd.h>	11 years ago
library.mak	…
version.sh	…

README.md

FFmpeg README

Documentation

Read the documentation in the doc/ directory in git.

You can also view it online at http://ffmpeg.org/documentation.html

Licensing

See the LICENSE file.

Build and Install

See the INSTALL file.