Further performance improvements and security fixes by
Vittorio Giovara, Luca Barbato and Diego Biurrun.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Signed-off-by: Diego Biurrun <diego@biurrun.de>
* commit '0e083d7e43805db1a978cb57bfa25fda62e8ff18':
build: Group general components separate from de/encoders in arch Makefiles
Conflicts:
libavcodec/arm/Makefile
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Results are from a Win64 build running on an AMD FX 6300
1121 decicycles in ttafilter_process_dec_c, 16777112 runs, 104 skips
522 decicycles in ff_ttafilter_process_dec_ssse3, 16777149 runs, 67 skips
477 decicycles in ff_ttafilter_process_dec_sse4, 16777156 runs, 60 skips
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
For the callable function (as opposed to the inline one):
C SSE SSE2 SSE4
Win32: 47 42 29 26
Win64: 30 33 25 23
The SSE version is neither compiled nor set for ARCH_X86_64, as the
inlinable function takes over.
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
benchmarked on sandybridge x86_64:
1358232 decicycles in flac_lpc_32_c
1244575 decicycles in flac_lpc_32_sse4, James Almer's patch
650045 decicycles in flac_lpc_32_sse4, this patch
I haven't tested the edgecases such as odd block lengths
odd block length tested-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Originally written by Ronald S. Bultje <rsbultje@gmail.com> and
Clément Bœsch <u@pkh.me>
Further contributions by:
Anton Khirnov <anton@khirnov.net>
Diego Biurrun <diego@biurrun.de>
Luca Barbato <lu_zero@gentoo.org>
Martin Storsjö <martin@martin.st>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
* commit '0338c396987c82b41d322630ea9712fe5f9561d6':
dsputil: Split off H.263 bits into their own H263DSPContext
Conflicts:
configure
libavcodec/mpegvideo.h
libavcodec/mpegvideo_enc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '1700b4e678ed329611a16b20d11e64b7abda4839':
x86: vp8dsp: Split loopfilter code into a separate file
Conflicts:
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '2ddb35b91131115c094d90e04031451023441b4d':
x86: dsputil: Separate ff_add_hfyu_median_prediction_cmov from dsputil_mmx
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '245b76a108585b6fb52eebc2626c472d6fa530dc':
x86: dsputil: Split inline assembly from init code
Conflicts:
libavcodec/x86/dsputil_mmx.c
Note, the author attribution is left in place and not removed
as it is in the merged commit.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'dc1b328d0df6e5ad5ff0ca4ae031e08466624f9c':
x86: hpeldsp: Only compile MMX hpeldsp code if MMX is enabled
Conflicts:
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '9b3a04d30691e85b77e63f75f5f26a93c3a000cd':
x86: Move duplicated put_pixels{8|16}_mmx functions into their own file
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '6b110d3a739c31602b59887ad65c67025df3f49d':
ppc: More consistent names for H.264 optimizations files
mpegaudiosp: More consistent names for ppc/x86 optimization files
Merged-by: Michael Niedermayer <michaelni@gmx.at>
This way, they can be shared between mpeg4qpel and h264qpel without
requiring either one to be compiled unconditionally.
Signed-off-by: Martin Storsjö <martin@martin.st>
This way, they can be shared between mpeg4qpel and h264qpel without
requiring either one to be compiled unconditionally.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* qatar/master:
x86: dsputil: Fix h263 loop filter link error in some configurations
Conflicts:
libavcodec/x86/dsputil.asm
Merged-by: Michael Niedermayer <michaelni@gmx.at>
This was caused by unconditionally referencing a conditionally compiled
table. Now the code is also compiled conditionally.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
* commit 'a846dccb29d2bb0798af1d47d06100eda9ca87cc':
h264chroma: x86: Fix building with yasm disabled
rv34: Drop now unnecessary dsputil dependencies
Conflicts:
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The sh4 optimizations are removed, because the code is
100% identical to the C code, so it is unlikely to
provide any real practical benefit.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
* commit '2e4bb99f4df7052b3e147ee898fcb4013a34d904':
vorbisdsp: convert x86 simd functions from inline asm to yasm.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Move some functions from dsputil. The idea is that videodsp contains
functions that are useful for a large and varied set of video decoders.
Currently, it contains emulated_edge_mc() and prefetch().
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
* qatar/master:
x86: vc1: call ff_vc1dsp_init_x86() under if (ARCH_X86)
x86: cavs: call ff_cavsdsp_init_x86() under if (ARCH_X86)
x86: call most of the x86 dsp init functions under if (ARCH_X86)
doc: support the new website layout
doc: remove a warning from filters.texi
doc: initial nut documentation
segment: drop global headers setting
lavu: fix typo in Makefile
Conflicts:
doc/Makefile
doc/filters.texi
doc/t2h.init
libavcodec/fmtconvert.c
libavcodec/proresdsp.c
libavcodec/x86/Makefile
libavcodec/x86/vc1dsp_mmx.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* qatar/master:
x86: dsputil: Only compile motion_est code when encoders are enabled
mem: fix typo in check for __ICC
fate: mp3: drop redundant CMP setting
rtp: Depacketization of JPEG (RFC 2435)
Rename ff_put_string to avpriv_put_string
mjpeg: Rename some symbols to avpriv_* instead of ff_*
yadif: cosmetics
Conflicts:
Changelog
libavcodec/mjpegenc.c
libavcodec/x86/Makefile
libavfilter/vf_yadif.c
libavformat/version.h
libavutil/mem.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* qatar/master:
build: export filtered -lz flag in config.mak
build: add separate setting for host linker
configure: probe_cc: use separate variable for linker output flag
x86: Always compile files with functions that are called unconditionally
x86: mpegvideoenc: fix linking with --disable-mmx
x86: mpegvideoenc: Do not abuse HAVE_ variables for template instantiation
Conflicts:
Makefile
configure
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'd488c3bcbaf7ddda42597e014deb661a7e9e2112':
configure: support Bitrig OS
yuv2rgb: handle line widths that are not a multiple of 4.
graph2dot: Use the fallback getopt implementation if needed
tools: Include io.h for open/read/write/close if unistd.h doesn't exist
testprogs: Remove unused includes
qt-faststart: Use other seek/tell functions on MSVC than on mingw
ismindex: Include direct.h for _mkdir on windows
sdp: Use static const char arrays instead of pointers to strings
x86: avcodec: Drop silly "_mmx" suffixes from filenames
x86: avcodec: Drop silly "_sse" suffixes from filenames
sdp: Include profile-level-id for H264
utvideoenc: use ff_huff_gen_len_table
huffman: add ff_huff_gen_len_table
cllc: simplify/fix swapped data buffer allocation.
rtpdec_h264: Don't set the pixel format
h264: Check that the codec isn't null before accessing it
audio_frame_queue: Define af_queue_log_state before using it
Conflicts:
libavcodec/audio_frame_queue.c
libavcodec/h264.c
libavcodec/huffman.h
libavcodec/huffyuv.c
libavcodec/utvideoenc.c
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* qatar/master:
vf_hqdn3d: Don't declare the loop variable within the for loop
huffyuv: update to current coding style
huffman: update to current coding style
rtsp: Free the rtpdec context properly
build: fft: x86: Drop unused YASM-OBJS-FFT- variable
Conflicts:
libavcodec/huffman.c
libavcodec/huffyuv.c
libavcodec/x86/Makefile
libavfilter/vf_hqdn3d.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'a1bcc76e6036e78f25cbb7323c145056cfca9d93': (21 commits)
cmdutils: fix a memleak when specifying an option twice.
x86: mpegvideo: more sensible names for optimization file and init function
x86: mpegvideoenc: Split optimizations off into a separate file
dnxhdenc: x86: more sensible names for optimization file and init function
svq1/svq3: Move common code out of SVQ1 decoder-specific file
dirac: add Comments and references to the standard
lavr: x86: optimized 6-channel flt to fltp conversion
lavr: x86: optimized 2-channel flt to fltp conversion
lavr: x86: optimized 6-channel flt to s16p conversion
lavr: x86: optimized 2-channel flt to s16p conversion
lavr: x86: optimized 6-channel s16 to fltp conversion
lavr: x86: optimized 2-channel s16 to fltp conversion
lavr: x86: optimized 6-channel s16 to s16p conversion
lavr: x86: optimized 2-channel s16 to s16p conversion
lavr: x86: optimized 2-channel fltp to flt conversion
lavr: x86: optimized 6-channel fltp to s16 conversion
lavr: x86: optimized 2-channel fltp to s16 conversion
lavr: x86: optimized 6-channel s16p to flt conversion
lavr: x86: optimized 2-channel s16p to flt conversion
lavr: x86: optimized 6-channel s16p to s16 conversion
...
Conflicts:
libavcodec/dirac.c
libavcodec/mpegvideo.h
libavcodec/x86/Makefile
Merged-by: Michael Niedermayer <michaelni@gmx.at>
