summaryrefslogtreecommitdiffstats
path: root/libavcodec/fft.c
Commit message (Collapse)AuthorAgeFilesLines
* Optimization of AC3 floating point decoder for MIPSNedeljko Babic2012-09-051-0/+1
| | | | | | | | | | FFT in MIPS implementation is working iteratively instead of "recursively" calling functions for smaller FFT sizes. Some of DSP and format convert utils functions are also optimized. Signed-off-by: Nedeljko Babic <nbabic@mips.com> Reviewed-by: Vitor Sessak <vitor1001@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* Merge remote-tracking branch 'qatar/master'Michael Niedermayer2012-01-291-1/+0
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * qatar/master: aacenc: Fix LONG_START windowing. aacenc: Fix a bug where deinterleaved samples were stored in the wrong place. avplay: use the correct array size for stride. lavc: extend doxy for avcodec_alloc_context3(). APIchanges: mention avcodec_alloc_context()/2/3 avcodec_align_dimensions2: set only 4 linesizes, not AV_NUM_DATA_POINTERS. aacsbr: ARM NEON optimised sbrdsp functions aacsbr: align some arrays aacsbr: move some simdable loops to function pointers cosmetics: Remove extra newlines at EOF Conflicts: libavcodec/utils.c libavfilter/formats.c libavutil/mem.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * cosmetics: Remove extra newlines at EOFAlex Converse2012-01-271-1/+0
| |
* | Merge remote branch 'qatar/master'Michael Niedermayer2011-04-271-5/+48
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * qatar/master: (23 commits) ac3enc: correct the flipped sign in the ac3_fixed encoder Eliminate pointless '#if 1' statements without matching '#else'. Add AVX FFT implementation. Increase alignment of av_malloc() as needed by AVX ASM. Update x86inc.asm from x264 to allow AVX emulation using SSE and MMX. mjpeg: Detect overreads in mjpeg_decode_scan() and error out. documentation: extend documentation for ffmpeg -aspect option APIChanges: update commit hashes for recent additions. lavc: deprecate FF_*_TYPE macros in favor of AV_PICTURE_TYPE_* enums aac: add headers needed for log2f() lavc: remove FF_API_MB_Q cruft lavc: remove FF_API_RATE_EMU cruft lavc: remove FF_API_HURRY_UP cruft pad: make the filter parametric vsrc_movie: add key_frame and pict_type. vsrc_movie: fix leak in request_frame() lavfi: add key_frame and pict_type to AVFilterBufferRefVideo. vsrc_buffer: add sample_aspect_ratio fields to arguments. lavfi: add fieldorder filter scale: make the filter parametric ... Conflicts: Changelog doc/filters.texi ffmpeg.c libavcodec/ac3dec.h libavcodec/dsputil.c libavfilter/avfilter.h libavfilter/vf_scale.c libavfilter/vf_yadif.c libavfilter/vsrc_buffer.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * Add AVX FFT implementation.Vitor Sessak2011-04-261-5/+48
| | | | | | | | Signed-off-by: Reinhard Tartler <siretart@tauware.de>
* | Merge remote branch 'qatar/master'Michael Niedermayer2011-04-041-0/+1
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * qatar/master: fate: fix partial run when no samples path is specified ARM: NEON fixed-point forward MDCT ARM: NEON fixed-point FFT lavf: bump minor version and add an APIChanges entry for avio changes avio: simplify url_open_dyn_buf_internal by using avio_alloc_context() avio: make url_fdopen internal. avio: make url_open_dyn_packet_buf internal. avio: avio_ prefix for url_close_dyn_buf avio: avio_ prefix for url_open_dyn_buf avio: introduce an AVIOContext.seekable field ac3enc: use generic fixed-point mdct lavfi: add fade filter Change yadif to not use out of picture lines. lavc: deprecate AVCodecContext.antialias_algo lavc: mark mb_qmin/mb_qmax for removal on next major bump. Conflicts: doc/filters.texi libavcodec/ac3enc_fixed.h libavcodec/ac3enc_float.h libavfilter/Makefile libavfilter/allfilters.c libavfilter/vf_fade.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * ARM: NEON fixed-point FFTMans Rullgard2011-04-031-0/+1
| | | | | | | | Signed-off-by: Mans Rullgard <mans@mansr.com>
* | Merge remote branch 'qatar/master'Michael Niedermayer2011-04-031-0/+3
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | | | | | * qatar/master: Fixed-point MDCT with 32-bit unscaled output lavc: deprecate rate_emu lavc: mark hurry_up for removal on next major bump parser: mark av_parser_parse() for removal on next major bump lavc: add missing audioconvert includes jvdec: don't use deprecated CODEC_TYPE_*/PKT_FLAG_KEY Conflicts: libavcodec/h264.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * Fixed-point MDCT with 32-bit unscaled outputMans Rullgard2011-04-021-0/+3
| | | | | | | | Signed-off-by: Mans Rullgard <mans@mansr.com>
* | Merge remote branch 'qatar/master'Michael Niedermayer2011-04-021-9/+4
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | * qatar/master: ac3enc: ARM optimised ac3_compute_matissa_size ac3: armv6 optimised bit_alloc_calc_bap fate: simplify fft test rules avio: document avio_alloc_context. lavf: make compute_chapters_end less picky. sierravmd: fix Indeo3 videos FFT: simplify fft8() fate: add fixed-point fft/mdct tests Fixed-point support in fft-test ape: check that number of seektable entries is equal to number of frames Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * FFT: simplify fft8()Mans Rullgard2011-04-011-9/+4
| | | | | | | | | | | | | | Part of the fft8() function corresponds to the BUTTERFLIES macro, so use it. Signed-off-by: Mans Rullgard <mans@mansr.com>
* | Merge remote-tracking branch 'newdev/master'Michael Niedermayer2011-04-011-23/+30
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | * newdev/master: mpegts: propagate avio EOF in read_packet() configure: Initial support for --target-os=symbian Fixed-point FFT and MDCT Include dependencies for test programs ac3enc: simplify sym_quant() flvdec: read index stored in the 'keyframes' tag. mov: Add support for zero-sized stsc runs. Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * Fixed-point FFT and MDCTMans Rullgard2011-03-311-23/+30
| |
| * Replace FFmpeg with Libav in licence headersMans Rullgard2011-03-191-4/+4
| | | | | | | | Signed-off-by: Mans Rullgard <mans@mansr.com>
| * FFT: factor a shuffle out of the inner loop and merge it into fft_permute.Loren Merritt2011-02-131-2/+7
| | | | | | | | | | | | 6% faster SSE FFT on Conroe, 2.5% on Penryn. Signed-off-by: Janne Grunau <janne-ffmpeg@jannau.net>
| * Mark C base versions of FFT functions static to fft.cDiego Elio Pettenò2011-01-261-2/+5
| | | | | | | | Signed-off-by: Mans Rullgard <mans@mansr.com>
* | FFT: factor a shuffle out of the inner loop and merge it into fft_permute.Loren Merritt2011-02-141-2/+7
| | | | | | | | | | | | | | 6% faster SSE FFT on Conroe, 2.5% on Penryn. Signed-off-by: Janne Grunau <janne-ffmpeg@jannau.net> (cherry picked from commit e6b1ed693ae4098e6b9eabf938fc31ec0b09b120)
* | Mark C base versions of FFT functions static to fft.cDiego Elio Pettenò2011-01-281-2/+5
|/ | | | | Signed-off-by: Mans Rullgard <mans@mansr.com> (cherry picked from commit 7767d8d361ca104d74c8390b72db21aa4d7fc322)
* IndentMåns Rullgård2010-07-011-8/+8
| | | | Originally committed as revision 23958 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove vestiges of radix-2 FFTMåns Rullgård2010-07-011-77/+5
| | | | | | Patch (mostly) by Loren Merritt Originally committed as revision 23957 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove explicit filename from Doxygen @file commands.Diego Biurrun2010-04-201-1/+1
| | | | | | | | 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
* Replace many includes of libavutil/common.h with what is actually neededMåns Rullgård2010-03-091-0/+2
| | | | | | | 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
* Move FFT parts from dsputil.h to fft.hMåns Rullgård2010-03-061-1/+2
| | | | Originally committed as revision 22235 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix build with --disable-mdctMåns Rullgård2010-02-111-0/+2
| | | | Originally committed as revision 21764 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Add ff_init_ff_cos_tabs function and use it in rdft.c to ensure that theReimar Döffinger2009-11-051-9/+15
| | | | | | | necessary ff_cos_tabs tables are initialized. Fixes issue 1507 (QDM2 broken since r20237 without hardcoded tables). Originally committed as revision 20464 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Pad ff_cos_tabs and ff_sin_tabs so that index n points to the table for n bits.Reimar Döffinger2009-11-041-1/+2
| | | | | | | While this "wastes" up to 2x32 bytes it makes the code slightly simpler and less confusing. Originally committed as revision 20449 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move/add COSTABLE/SINTABLE macros to dsputil to add extern definitionsReimar Döffinger2009-10-151-24/+15
| | | | | | for ff_cos_* and ff_sin_* without introducing too much code duplication. Originally committed as revision 20243 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use hardcoded instead of runtime-calculated ff_cos_* tables ifReimar Döffinger2009-10-141-13/+26
| | | | | | | --enable-hardcoded-tables was used. Due to the size, the code for the tables is generated at compile time. Originally committed as revision 20232 to svn://svn.ffmpeg.org/ffmpeg/trunk
* fft_dispatch array should be constReimar Döffinger2009-10-131-1/+1
| | | | Originally committed as revision 20223 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move per-arch fft init bits into the corresponding subdirsMåns Rullgård2009-09-151-34/+5
| | | | Originally committed as revision 19864 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: 10l: fix large FFTsMåns Rullgård2009-09-141-4/+1
| | | | Originally committed as revision 19846 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: NEON optimised MDCTMåns Rullgård2009-09-111-0/+1
| | | | Originally committed as revision 19819 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Prepare for optimised forward MDCT implementationsMåns Rullgård2009-09-111-0/+1
| | | | | | | | | This adds a function pointer for forward MDCT to FFTContext and initialises it with the existing C function. ff_calc_mdct() is changed to an inline function calling the selected version as done for other fft/mdct functions. Originally committed as revision 19818 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: NEON optimised FFT and MDCTMåns Rullgård2009-09-101-1/+9
| | | | | | | | Vorbis and AC3 ~3x faster. Parts by Naotoshi Nojiri, naonoj gmail Originally committed as revision 19806 to svn://svn.ffmpeg.org/ffmpeg/trunk
* The ff_cos_tabs table itself is constant, too, so mark it as such.Reimar Döffinger2009-09-061-1/+1
| | | | Originally committed as revision 19784 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove long unused ALTIVEC_USE_REFERENCE_C_CODE ifdef; all other referencesDavid Conrad2009-03-161-1/+1
| | | | | | to it were removed in r6606 Originally committed as revision 17997 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Only enable SSE/3DNOW optimizations when they have been enabled at compilation.Diego Biurrun2009-02-191-3/+3
| | | | | | Fixes shared build with SSE/3DNOW disabled but MMX and yasm enabled. Originally committed as revision 17451 to svn://svn.ffmpeg.org/ffmpeg/trunk
* prettyprinting whitespace cosmeticsDiego Biurrun2009-02-191-10/+10
| | | | Originally committed as revision 17443 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use full internal pathname in doxygen @file directives.Diego Biurrun2009-02-011-1/+1
| | | | | | | 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
* Add the rdft family of transforms (fft/ifft of an all real sequence) to dsputil.Alex Converse2009-01-301-1/+1
| | | | Originally committed as revision 16864 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark ff_fft_end av_cold.Alex Converse2009-01-291-1/+1
| | | | Originally committed as revision 16851 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark ff_fft_init with av_cold.Alex Converse2009-01-221-1/+1
| | | | Originally committed as revision 16722 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Clean up FFT related comments.Alex Converse2009-01-221-12/+0
| | | | Originally committed as revision 16721 to svn://svn.ffmpeg.org/ffmpeg/trunk
* cosmetics: Remove pointless period after copyright statement non-sentences.Diego Biurrun2009-01-191-1/+1
| | | | Originally committed as revision 16684 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Change semantic of CONFIG_*, HAVE_* and ARCH_*.Aurelien Jacobs2009-01-131-4/+4
| | | | | | They are now always defined to either 0 or 1. Originally committed as revision 16590 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove duplicated MM_* macros for CPU capabilities from dsputil.h.Dominik Mierzejewski2008-11-031-4/+4
| | | | | | | Add missing one for FF_MM_ALTIVEC to avcodec.h. Rename all the occurences of MM_* to the corresponding FF_MM_*. Originally committed as revision 15770 to svn://svn.ffmpeg.org/ffmpeg/trunk
* mdct wrapper function to match fftLoren Merritt2008-08-121-2/+2
| | | | Originally committed as revision 14703 to svn://svn.ffmpeg.org/ffmpeg/trunk
* optimize imdct_half:Loren Merritt2008-08-121-0/+2
| | | | | | | | remove tmp buffer. skip fft reinterleave pass, leaving data in a format more convenient for simd. merge post-rotate with post-reorder. Originally committed as revision 14700 to svn://svn.ffmpeg.org/ffmpeg/trunk
* indentLoren Merritt2008-08-121-8/+7
| | | | Originally committed as revision 14699 to svn://svn.ffmpeg.org/ffmpeg/trunk
* split-radix FFTLoren Merritt2008-08-121-126/+245
| | | | | | c is 1.9x faster than previous c (on various x86 cpus), sse is 1.6x faster than previous sse. Originally committed as revision 14698 to svn://svn.ffmpeg.org/ffmpeg/trunk
OpenPOWER on IntegriCloud