[libav-commits] x86: Avoid movs on BUTTERFLYPS when in AVX mode

Module: libav Branch: master Commit: 4a301706fd682d3ad43b4239542e6f586ae29811 Author:Vitor Sessak Committer: Janne Grunau Date: Fri May 11 22:32:08 2012 +0200 x86: Avoid movs on BUTTERFLYPS when in AVX mode Signed-off-by: Janne Grunau --- libavutil/x86/x86util.asm | 13

[libav-commits] x86: use new schema for ASM macros

Module: libav Branch: master Commit: bac0729d9e1dcd4efc63637c7832c8bb013d4284 Author:Vitor Sessak Committer: Janne Grunau Date: Sun May 27 17:43:56 2012 + x86: use new schema for ASM macros Signed-off-by: Janne Grunau --- libavcodec/x86/fft.c | 12 +++--- libavcodec

[libav-commits] x86: use more standard construct for setting ASM functions in FFT code

Module: libav Branch: master Commit: fcc456b8293b829d868f55778d76f8da3d54e846 Author:Vitor Sessak Committer: Diego Biurrun Date: Sat May 12 08:59:00 2012 +0200 x86: use more standard construct for setting ASM functions in FFT code Signed-off-by: Diego Biurrun --- libavcodec/x86

[libav-commits] amrnbdec: check frame size before decoding.

Module: libav Branch: release/0.8 Commit: 11f3173e1bae135eb18a10b0060a5dd4b9fdcc74 Author:Vitor Sessak Committer: Reinhard Tartler Date: Wed Feb 29 22:09:10 2012 +0100 amrnbdec: check frame size before decoding. Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind CC:

[libav-commits] nellymoserdec: Saner and faster IMDCT windowing

Module: libav Branch: master Commit: 5cd1337f5db49b4537d83ae2a8115a90d77e3a28 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Wed Feb 29 19:39:16 2012 +0100 nellymoserdec: Saner and faster IMDCT windowing Signed-off-by: Ronald S. Bultje --- libavcodec/nellymoserdec.c | 16

[libav-commits] amrnbdec: check frame size before decoding.

Module: libav Branch: master Commit: 882abda5a26ffb8e3d1c5852dfa7cdad0a291d2d Author:Vitor Sessak Committer: Ronald S. Bultje Date: Wed Feb 29 22:09:10 2012 +0100 amrnbdec: check frame size before decoding. Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind CC:

[libav-commits] mpegaudiodec: use DSPUtil.butterflies_float().

Module: libav Branch: master Commit: 9b4cd586119a6c50907abc35279eb2af63f278a2 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Wed Feb 29 21:06:13 2012 +0100 mpegaudiodec: use DSPUtil.butterflies_float(). Signed-off-by: Ronald S. Bultje --- libavcodec/mpegaudiodec.c |7

[libav-commits] mpegaudiodec: move imdct and windowing function to mpegaudiodsp

Module: libav Branch: master Commit: 6dfcf53092aba9f1ef31629e11515df5752327db Author:Vitor Sessak Committer: Ronald S. Bultje Date: Wed Jan 4 21:32:47 2012 +0100 mpegaudiodec: move imdct and windowing function to mpegaudiodsp Signed-off-by: Ronald S. Bultje --- libavcodec

[libav-commits] mpegaudiodec: optimized iMDCT transform

Module: libav Branch: master Commit: 39df0c434c76aa6a6decccb969dfa51468440823 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Thu Jan 5 20:26:33 2012 +0100 mpegaudiodec: optimized iMDCT transform Signed-off-by: Ronald S. Bultje --- libavcodec/mpegaudiodec.c |2

[libav-commits] mpegaudiodec: change imdct window arrangment for better pointer alignment

Module: libav Branch: master Commit: 06677d0dd9d3c656665566e585e9ca6de815f247 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Wed Jan 4 21:43:47 2012 +0100 mpegaudiodec: change imdct window arrangment for better pointer alignment Signed-off-by: Ronald S. Bultje

[libav-commits] mpegaudiodec: interleave iMDCT buffer to simplify future SIMD implementations

Module: libav Branch: master Commit: 421c99a4a7c116fc2d4e7a6c866c2209852ef581 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Tue Jan 3 21:40:57 2012 +0100 mpegaudiodec: interleave iMDCT buffer to simplify future SIMD implementations Signed-off-by: Ronald S. Bultje

[libav-commits] mpegaudiodec: Use clearer pointer math

Module: libav Branch: master Commit: 96219141e2b2a71fde4dca5505325d21f7cc10b7 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Tue Jan 3 21:25:59 2012 +0100 mpegaudiodec: Use clearer pointer math Signed-off-by: Ronald S. Bultje --- libavcodec/mpegaudiodec.c |2 +- 1

[libav-commits] Plug some memory leaks in the VP6 decoder

Module: libav Branch: release/0.5 Commit: 280590e3380d8649e5b191c653b0f974626c8eeb Author:Vitor Sessak Committer: Reinhard Tartler Date: Wed Mar 3 17:24:32 2010 + Plug some memory leaks in the VP6 decoder Originally committed as revision 22172 to svn://svn.ffmpeg.org/ffmpeg

[libav-commits] x86inc: Flag shufps as an floating-point instruction for the AVX emulation code.

Module: libav Branch: master Commit: 6b6ee58249b233585f1f7d89aec4fe58f65b5ed7 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Sun Nov 27 15:41:56 2011 +0100 x86inc: Flag shufps as an floating-point instruction for the AVX emulation code. Without this, code like "shufps m

[libav-commits] dct32: Add SSE2 ASM optimizations

Module: libav Branch: master Commit: 18b131de0473a3110c63966cd7c6cd2ab118d401 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Sat Jul 30 18:39:25 2011 +0200 dct32: Add SSE2 ASM optimizations Signed-off-by: Ronald S. Bultje --- libavcodec/x86/dct32_sse.asm | 39

[libav-commits] Fix dct32() compilation with --disable-yasm

Module: libav Branch: master Commit: 165c7c420d611bfa16d999f2033619c542961926 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Sun May 22 12:04:33 2011 +0200 Fix dct32() compilation with --disable-yasm Signed-off-by: Ronald S. Bultje --- libavcodec/x86/fft.c |2 ++ 1

[libav-commits] dct32: Add AVX implementation of 32-point DCT

Module: libav Branch: master Commit: 6204feb160c843320f6001d7e2bb2361c82b90ca Author:Vitor Sessak Committer: Reinhard Tartler Date: Sat May 14 14:17:15 2011 +0200 dct32: Add AVX implementation of 32-point DCT --- libavcodec/mpegaudiodec.c|4 +- libavcodec/x86/dct32_sse.asm

[libav-commits] dct32: port SSE 32-point DCT to YASM

Module: libav Branch: master Commit: 3758eb0eb96217c6968d4748757f96aeecfb Author:Vitor Sessak Committer: Reinhard Tartler Date: Tue May 17 18:26:01 2011 +0200 dct32: port SSE 32-point DCT to YASM --- libavcodec/x86/Makefile |3 +- libavcodec/x86/dct32_sse.asm | 289

[libav-commits] dct32: Change pass 6 permutation to allow for AVX implementation

Module: libav Branch: master Commit: 4e653b98c888a922ee192c6c8f914dde6ea2dc40 Author:Vitor Sessak Committer: Reinhard Tartler Date: Sat May 14 14:16:30 2011 +0200 dct32: Change pass 6 permutation to allow for AVX implementation --- libavcodec/x86/dct32_sse.asm | 98

[libav-commits] qdm2: Use floating point synthesis filter.

Module: libav Branch: master Commit: 984ece7503597d30e6f3bdeb67e337ea1616f880 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Thu May 19 21:33:27 2011 +0200 qdm2: Use floating point synthesis filter. This avoid needlessly convertion from floating point to fixed point and back

[libav-commits] lavf/utils: fix ff_interleave_compare_dts corner case.

Module: libav Branch: master Commit: ecc297308ff3cb8b20359df44108ac299b22bdf1 Author:Vitor Sessak Committer: Ronald S. Bultje Date: Sat May 7 22:48:29 2011 +0200 lavf/utils: fix ff_interleave_compare_dts corner case. This should fix behavior introduced by commit

[libav-commits] Add AVX FFT implementation.

Module: libav Branch: master Commit: 9d35fa520e3b27f7dd9fe12c433eb596f1271515 Author:Vitor Sessak Committer: Reinhard Tartler Date: Mon Apr 25 11:39:01 2011 +0200 Add AVX FFT implementation. Signed-off-by: Reinhard Tartler --- Changelog |2 +- libavcodec

[libav-commits] Increase alignment of av_malloc() as needed by AVX ASM.

Module: libav Branch: master Commit: 13dfce3d44f99a2d7df71aba8ae003d58db726f7 Author:Vitor Sessak Committer: Reinhard Tartler Date: Sat Apr 23 19:24:31 2011 +0200 Increase alignment of av_malloc() as needed by AVX ASM. Signed-off-by: Reinhard Tartler --- libavutil/mem.c | 16

[libav-commits] Update x86inc.asm from x264 to allow AVX emulation using SSE and MMX.

Module: libav Branch: master Commit: 33cbfa6fa37d9aea235ce7d3c389d8eec279205f Author:Vitor Sessak Committer: Reinhard Tartler Date: Sat Apr 23 19:24:06 2011 +0200 Update x86inc.asm from x264 to allow AVX emulation using SSE and MMX. Signed-off-by: Reinhard Tartler --- libavcodec

[libav-commits] lavf/utils.c: Order packets with identical PTS by stream index.

Module: libav Branch: master Commit: 96573c0d7605672d69b42ae1dcf18764ce47c71a Author:Vitor Sessak Committer: Ronald S. Bultje Date: Mon Apr 11 16:20:36 2011 -0400 lavf/utils.c: Order packets with identical PTS by stream index. This allows for more reproducible results when using