* commit '50078c1c8070dd8d1c329e8117ff30ec72489039':
libavutil: move FFALIGN macro from common.h to macros.h
Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
Include macros.h explicitly in common.h so that external code using
FFALIGN does not break. It was already implicitly included through
version.h. Include macros.h in lls.h and internal.h for FFALIGN.
lls.h was including common.h only for FFALIGN and internal.h was
missing the include for FFALIGN. `make checkheaders` did not catch it
because it's an internal header.
This is a trivial rewrite of the loops that results in better
prefetching and associated cache efficiency. Essentially, the problem is
that modern prefetching logic is based on finite state Markov memory, a reasonable
assumption that is used elsewhere in CPU's in for instance branch
predictors.
Surrounding loops all iterate forward through the array, making the
predictor think of prefetching in the forward direction, but the
intermediate loop is unnecessarily in the backward direction.
Speedup is nontrivial. Benchmarks obtained by 10^6 iterations within
solve_lls, with START/STOP_TIMER. File is tests/data/fate/flac-16-lpc-cholesky.err.
Hardware: x86-64, Haswell, GNU/Linux.
new:
17291 decicycles in solve_lls, 2096706 runs, 446 skips
17255 decicycles in solve_lls, 4193657 runs, 647 skips
17231 decicycles in solve_lls, 8384997 runs, 3611 skips
17189 decicycles in solve_lls,16771010 runs, 6206 skips
17132 decicycles in solve_lls,33544757 runs, 9675 skips
17092 decicycles in solve_lls,67092404 runs, 16460 skips
17058 decicycles in solve_lls,134188213 runs, 29515 skips
old:
18009 decicycles in solve_lls, 2096665 runs, 487 skips
17805 decicycles in solve_lls, 4193320 runs, 984 skips
17779 decicycles in solve_lls, 8386855 runs, 1753 skips
18289 decicycles in solve_lls,16774280 runs, 2936 skips
18158 decicycles in solve_lls,33548104 runs, 6328 skips
18420 decicycles in solve_lls,67091793 runs, 17071 skips
18310 decicycles in solve_lls,134187219 runs, 30509 skips
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* commit '502ab21af0ca68f76d6112722c46d2f35c004053':
x86: lpc: simd av_update_lls
The versions are bumped due to changes in lls.h which is used across
libraries affecting intra library ABI
(This version bump also covers changes to lls.h in the immedeatly previous
commits)
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '41578f70cf8aec8e7565fba1ca7e07f3dc46c3d2':
lpc: use function pointers, in preparation for asm
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '9d4da474f5f40b019cb4cb931c8499deee586174':
lls: move to the private namespace
Conflicts:
libavutil/version.h
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* qatar/master: (28 commits)
mp3enc: write a xing frame containing number of frames in the file
lavf: update AVStream.nb_frames when muxing.
ffmpeg: remove unused variables from InputStream.
doc: update ffmpeg -ar and -ac documentation to reflect reality.
ffmpeg: remove pointless if (nb_input_files)
ffmpeg: merge input_files_ts_offset into input_files.
ffmpeg: merge input_codecs into input_streams.
ffmpeg: drop AV prefixes from struct names.
ffmpeg: deprecate loop_input and loop_output options
gif: add loop private option.
img2: add loop private option.
AVOptions: in av_opt_find() don't return named constants unless unit is specified.
x11grab: replace undocumented nomouse hackery with a private option.
dict: extend documentation.
lls: whitespace cosmetics
docs: Use proper markup for a literal command line option
docs: Remove a remark that isn't relevant any longer
docs: Explain how to regenerate import libraries with MSVC tools
docs: Mention that libraries for MSVC can be built with a cross compiler
docs: Remove old docs that mention setting up a build environment with lib.exe
...
Conflicts:
doc/ffmpeg.texi
doc/general.texi
ffmpeg.c
libavcodec/Makefile
libavcodec/dnxhddata.c
libavformat/mp3enc.c
libavformat/utils.c
libavutil/Makefile
tests/copycooker.sh
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Passing an explicit filename to this command is only necessary if the
documentation in the @file block refers to a file different from the
one the block resides in.
Originally committed as revision 22921 to svn://svn.ffmpeg.org/ffmpeg/trunk
Otherwise doxygen complains about ambiguous filenames when files exist
under the same name in different subdirectories.
Originally committed as revision 16912 to svn://svn.ffmpeg.org/ffmpeg/trunk
avoiding casts would be even less readable, but other suggestions are welcome.
lls.c:56: warning: initialization from incompatible pointer type
lls.c:57: warning: initialization from incompatible pointer type
Originally committed as revision 11697 to svn://svn.ffmpeg.org/ffmpeg/trunk
