FFmpeg

Commit Graph

Author	SHA1	Message	Date
Ganesh Ajjanagadde	971d12b7f9	avutil/mathematics: speed up av_gcd by using Stein's binary GCD algorithm This uses Stein's binary GCD algorithm: https://en.wikipedia.org/wiki/Binary_GCD_algorithm to get a roughly 4x speedup over Euclidean GCD on standard architectures with a compiler intrinsic for ctzll, and a roughly 2x speedup otherwise. At the moment, the compiler intrinsic is used on GCC and Clang due to its easy availability. Quick note regarding overflow: yes, subtractions on int64_t can, but the llabs takes care of that. The llabs is also guaranteed to be safe, with no annoying INT64_MIN business since INT64_MIN being a power of 2, is shifted down before being sent to llabs. The binary GCD needs ff_ctzll, an extension of ff_ctz for long long (int64_t). On GCC, this is provided by a built-in. On Microsoft, there is a BitScanForward64 analog of BitScanForward that should work; but I can't confirm. Apparently it is not available on 32 bit builds; so this may or may not work correctly. On Intel, per the documentation there is only an intrinsic for _bit_scan_forward and people have posted on forums regarding _bit_scan_forward64, but often their documentation is woeful. Again, I don't have it, so I can't test. As such, to be safe, for now only the GCC/Clang intrinsic is added, the rest use a compiled version based on the De-Bruijn method of Leiserson et al: http://supertech.csail.mit.edu/papers/debruijn.pdf. Tested with FATE, sample benchmark (x86-64, GCC 5.2.0, Haswell) with a START_TIMER and STOP_TIMER in libavutil/rationsl.c, followed by a make fate. aac-am00_88.err: builtin: 714 decicycles in av_gcd, 4095 runs, 1 skips de-bruijn: 1440 decicycles in av_gcd, 4096 runs, 0 skips previous: 2889 decicycles in av_gcd, 4096 runs, 0 skips Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>	9 years ago
Vittorio Giovara	cdfe45ad37	lavu: Drop deprecated av_reverse function Deprecated in 10/2012.	9 years ago
Michael Niedermayer	666e29fe9e	avutil/mathematics/av_add_stable: Avoid av_cmp_q() call Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	11 years ago
Michael Niedermayer	e9add0d85b	av_add_stable: Add fast special case where step can be represented exactly Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	11 years ago
Michael Niedermayer	4956d0e5a6	avutil/mathematics/av_add_stable: check for the common case of inc=1 Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	11 years ago
Michael Niedermayer	5b7519fbaa	avutil/mathematics/av_add_stable: avoid unneeded variable Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	11 years ago
Luca Barbato	de69aedf99	mathematics: K&R formatting cosmetics	11 years ago
Michael Niedermayer	b317f9459f	avutil/mathematics: add av_add_stable() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	11 years ago
Anton Khirnov	94a417acc0	mathematics: remove asserts from av_rescale_rnd() It is a public function, it must not assert on its parameters.	11 years ago
Michael Niedermayer	740e740895	av_rescale: support passing MIN/MAX through Reviewed-by: Clément Bœsch <ubitux@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	12 years ago
Michael Niedermayer	8766ad9eb1	lavu: add av_rescale_delta() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	12 years ago
Michael Niedermayer	90d4b07063	mathemathics: update copyright years Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	12 years ago
Diego Biurrun	d5c62122a7	Move av_reverse table to libavcodec It is only used in that library.	12 years ago
Diego Biurrun	930c9d4373	avutil: Duplicate ff_log2_tab instead of sharing it across libs The table is so small that the space gain is not worth the performance overhead of cross-library access.	12 years ago
Diego Biurrun	9734b8ba56	Move avutil tables only used in libavcodec to libavcodec.	12 years ago
Michael Niedermayer	2f23a8ab17	libavutil/mathematics: use av_assert() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	13 years ago
Justin Ruggles	0b42a9388c	avutil: add av_rescale_q_rnd() to allow different rounding	13 years ago
Diego Biurrun	bb00b15f9e	avutil: Remove unused arbitrary precision integer code.	14 years ago
Michael Niedermayer	a18eff49c0	av_compare_ts: Improve speed when calculations fit in 64bit. about 110 cpu cycles before 60 cpu cycles afterwards. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	14 years ago
Mans Rullgard	2912e87a6c	Replace FFmpeg with Libav in licence headers Signed-off-by: Mans Rullgard <mans@mansr.com>	14 years ago
Eli Friedman	b7cdddcd1f	Silence "comparison of unsigned expression >= 0 is always true" warning. Patch by Eli Friedman, eli d friedman a gmail Originally committed as revision 24022 to svn://svn.ffmpeg.org/ffmpeg/trunk	15 years ago
Michael Niedermayer	65db0587a8	Add av_compare_mod() Originally committed as revision 23551 to svn://svn.ffmpeg.org/ffmpeg/trunk	15 years ago
Diego Biurrun	ba87f0801d	Remove explicit filename from Doxygen @file commands. 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	15 years ago
Måns Rullgård	2ed6f39944	Replace many includes of libavutil/common.h with what is actually needed This reduces the number of false dependencies on header files and speeds up compilation. Originally committed as revision 22407 to svn://svn.ffmpeg.org/ffmpeg/trunk	15 years ago
Michael Niedermayer	78b0182375	av_compare_ts() Originally committed as revision 21671 to svn://svn.ffmpeg.org/ffmpeg/trunk	15 years ago
Francesco Lavra	91cc5d3767	Move ff_reverse in libavcodec to av_reverse in libavutil. Patch by Francesco Lavra, francescolavra interfree it Originally committed as revision 20484 to svn://svn.ffmpeg.org/ffmpeg/trunk	15 years ago
Diego Biurrun	082dea8e40	Remove all remaining code that was disabled through the major version bump. Originally committed as revision 17903 to svn://svn.ffmpeg.org/ffmpeg/trunk	16 years ago
Diego Biurrun	bad5537e2c	Use full internal pathname in doxygen @file directives. 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	16 years ago
Diego Biurrun	89c9ff504b	spelling/grammar/consistency review part I Originally committed as revision 16840 to svn://svn.ffmpeg.org/ffmpeg/trunk	16 years ago
Aurelien Jacobs	eded5c4309	add a ff_gcd() function again, for compatibility with old libavcodec Originally committed as revision 16814 to svn://svn.ffmpeg.org/ffmpeg/trunk	16 years ago
Diego Biurrun	dfcb6b56f9	Directly #include a bunch of indirectly #included headers. Originally committed as revision 16748 to svn://svn.ffmpeg.org/ffmpeg/trunk	16 years ago
Aurelien Jacobs	9ce6c13879	export gcd function as av_gcd() Originally committed as revision 16653 to svn://svn.ffmpeg.org/ffmpeg/trunk	16 years ago
Michael Niedermayer	c448a09624	Faster ff_sqrt() Originally committed as revision 11586 to svn://svn.ffmpeg.org/ffmpeg/trunk	17 years ago
Diego Biurrun	f3635240b7	Fix a couple of 'return type defaults to int' and 'control reaches end of non-void function' warnings in test code. Originally committed as revision 11491 to svn://svn.ffmpeg.org/ffmpeg/trunk	17 years ago
Diego Biurrun	f0cb505aeb	Allow compilation of test programs when TEST is defined. Originally committed as revision 11473 to svn://svn.ffmpeg.org/ffmpeg/trunk	17 years ago
Diego Biurrun	f8a80fd69d	main() --> main(void) Originally committed as revision 11079 to svn://svn.ffmpeg.org/ffmpeg/trunk	17 years ago
Steve L'Homme	949b1a13bf	Replace most of the %lld and %llx by their (cleaner) PRI*64 counterparts. patch by Steve Lhomme, slhomme divxcorp com Originally committed as revision 6868 to svn://svn.ffmpeg.org/ffmpeg/trunk	18 years ago
Diego Biurrun	b78e7197a8	Change license headers to say 'FFmpeg' instead of 'this program/this library' and fix GPL/LGPL version mismatches. Originally committed as revision 6577 to svn://svn.ffmpeg.org/ffmpeg/trunk	18 years ago
Michael Niedermayer	fc861443d3	improve selftest Originally committed as revision 5424 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Michael Niedermayer	2bda41e5b2	extend range used for testing (larger range should work too but testing code overflows) Originally committed as revision 5334 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Michael Niedermayer	5c1cb3792d	<= vs. >= 10l bug Originally committed as revision 5333 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Måns Rullgård	d1c9b76287	add newline at end of file Originally committed as revision 5330 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Michael Niedermayer	fdb3a34156	avoid AVInteger usage in av_rescale disable integer.* (unused) libavutil.a 45k -> 32k Originally committed as revision 5311 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Diego Biurrun	5509bffa88	Update licensing information: The FSF changed postal address. Originally committed as revision 4842 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Diego Biurrun	115329f160	COSMETICS: Remove all trailing whitespace. Originally committed as revision 4749 to svn://svn.ffmpeg.org/ffmpeg/trunk	19 years ago
Alexander Strasser	c11c2bc20b	libavutil: Utility code from libavcodec moved to a separate library. Originally committed as revision 4489 to svn://svn.ffmpeg.org/ffmpeg/trunk	20 years ago

1 2

54 Commits (8e7571eacda57530b064f48bcc067c99f88a98ca)