libwebp: update to 0.5.0

This commit imports libwebp 0.5.0, including AUTHORS, COPYING, ChangeLog,
NEWS, PATENTS, README and src directories. In src, only includes header
and source files.

The patches required to build it in Qt will follow in separate
commit(s).

Change-Id: I96b4961ba63c75cc7fbab158c36a0f403f254c14
Reviewed-by: aavit <eirik.aavitsland@theqtcompany.com>

1 files changed, 784 insertions, 121 deletions

diff --git a/src/3rdparty/libwebp/ChangeLog b/src/3rdparty/libwebp/ChangeLog
index 0c362ad..2f8def2 100644
--- a/src/3rdparty/libwebp/ChangeLog
+++ b/src/3rdparty/libwebp/ChangeLog
@@ -1,125 +1,788 @@
-46e18c0 vwebp: fix incorrect clipping w/NO_BLEND
-fcfde90 update issue tracker url
-8c3fb33 update AUTHORS
-808d4a6 update NEWS
-6286404 bump version to 0.4.4
-b8b314a doc/webp-container-spec: update repo browser link
-c3953e3 fix typo: constitutes -> constitute
-cd377e2 Use __has_builtin to check clang support
-e2e8980 wicdec: fix alpha detection w/64bpp BGRA/RGBA
-5c3fe77 iosbuild: fix linking with Xcode 7 / iOS SDK 9
-f9f5498 VP8LAllocateHistogramSet: align histogram[] entries
-3026db2 Loosen the buffer size checks for Y/U/V/A too.
-d089362 loosen the padding check on buffer size
-53d22c5 dec_neon: add whitespace around stringizing operator
-8bcc4d4 dsp/mips: add whitespace around stringizing operator
-d49c44f Container spec: clarify ordering of ALPH chunk.
-382de22 msvc: fix pointer type warning in BitsLog2Floor
-84ecd9d FlattenSimilarBlocks should only be tried when blending is possible.
-f55ebbb backport rescaler fix
-2ff633c fix mips2 build target
-326b5fb update ChangeLog (tag: v0.4.3, origin/0.4.3, 0.4.3)
-a661e50 Disable NEON code on Native Client
-fcd94e9 update ChangeLog (tag: v0.4.3-rc1)
-569fe57 update NEWS
-bd852f5 bump version to 0.4.3
-2d58b64 WebPPictureRescale: add a note about 0 width/height
-a0d8ca5 examples/Android.mk: add webpmux_example target
-34b1d29 Android.mk: add webpmux target
-7561988 Android.mk: add webpdemux target
-a987576 Android.mk: add webpdecoder{,_static} targets
-a6d4859 Android.mk: split source lists per-directory
-77544d5 fix iOS arm64 build with Xcode 6.3
-6dea157 doc/webp-container-spec: note MSB order for chunk diagrams
-f7cd57b doc/webp-container-spec: cosmetics
-1d6b250 vwebp: clear canvas at the beginning of each loop
-f97b3f8 webp-container-spec: clarify background clear on loop
-4ba83c1 vwebp: remove unnecessary static Help() prototype
-d34e8e3 vwebp/animation: display last frame on end-of-loop
-bbbc524 dec/vp8: clear 'dither_' on skipped blocks
-0339fa2 lossless_neon: enable subtract green for aarch64
-5a0c220 Regression fix for lossless decoding
-6e3a31d wicdec: (msvs) quiet some /analyze warnings
-b49a578 dwebp/WritePNG: mark png variables volatile
-0a4391a dwebp: include setjmp.h w/WEBP_HAVE_PNG
-90f1ec5 dwebp: correct sign in format strings
-b61ce86 VP8LEncodeStream: add an assert
-df1081b dsp/cpu: (msvs) add include for __cpuidex
-39aa055 dsp/cpu: (msvs) avoid immintrin.h on _M_ARM
-f814f42 dsp/cpu: add include for _xgetbv() w/MSVS
-8508ab9 cpu: fix AVX2 detection for gcc/clang targets
-5769623 fix handling of zero-sized partition #0 corner case
-b2e71a9 make the 'last_cpuinfo_used' variable names unique
-1273e84 add -Wformat-nonliteral and -Wformat-security
-3ae78eb multi-thread fix: lock each entry points with a static var
-5c1eeda webp-container-spec: remove references to fragments
-c5ceea4 enc_neon: fix building with non-Xcode clang (iOS)
-d0859d6 iosbuild: add x64_64 simulator support
-046732c WebPEncode: Support encoding same pic twice (even if modified)
-4426f50 webp/types.h: use inline for clang++/-std=c++11
-e297fc7 gif2webp: Use the default hint instead of WEBP_HINT_GRAPH.
-855fe43 Makefile.vc: add a 'legacy' RTLIBCFG option
-b7eb6d5 gif2webp: Support GIF_DISPOSE_RESTORE_PREVIOUS
-5691bdd gif2webp: Handle frames with odd offsets + disposal to background.
-8301da1 stopwatch.h: fix includes
-6a2209a update ChangeLog (tag: v0.4.2, origin/0.4.2, 0.4.2)
-36cad6a bit_reader.h: cosmetics: fix a typo
-e2ecae6 enc_mips32: workaround gcc-4.9 bug
-243e68d update ChangeLog (tag: v0.4.2-rc2)
-eec5f5f enc/vp8enci.h: update version number
-0c1b98d update NEWS
-69b0fc9 update AUTHORS
-857578a bump version to 0.4.2
-9129deb restore encode API compatibility
-f17b95e AssignSegments: quiet -Warray-bounds warning
-9c56c8a enc_neon: initialize vectors w/vdup_n_u32
-a008902 iosbuild: cleanup
-cc6de53 iosbuild: output autoconf req. on failure
-740d765 iosbuild: make iOS 6 the minimum requirement
-403023f iobuild.sh: only install .h files in Headers
-b65727b Premultiply with alpha during U/V downsampling
-8de0deb gif2webp: Background color correction
-f8b7d94 Amend the lossless spec according to issue #205, #206 and #224
-9102a7b Add a WebPExtractAlpha function to dsp
-e407b5d webpmux: simplify InitializeConfig()
-3e70e64 webpmux: fix indent
-be38f1a webpmux: fix exit status on numeric value parse error
-94dadcb webpmux: fix loop_count range check
-40b3a61 examples: warn on invalid numeric parameters
-b7d209a gif2webp: Handle frames with missing  graphic control extension
-bf0eb74 configure: simplify libpng-config invocation
-3740f7d Rectify bug in lossless incremental decoding.
-3ab0a37 make VP8LSetBitPos() set br->eos_ flag
-2e4312b Lossless decoding: fix eos_ flag condition
-e6609ac fix erroneous dec->status_ setting
-5692eae add a fallback to ALPHA_NO_COMPRESSION
-6ecd5bf ExUtilReadFromStdin: (windows) open stdin in bin mode
-4206ac6 webpmux: (windows) open stdout in binary mode
-d40e885 cwebp: (windows) open stdout in binary mode
-4aaf463 example_util: add ExUtilSetBinaryMode
-4c82ff7 webpmux man page: Clarify some title, descriptions and examples
-23d4fb3 dsp/lossless: workaround gcc-4.9 bug on arm
-5af7719 dsp.h: collect gcc/clang version test macros
-90d1124 enc_neon: enable QuantizeBlock for aarch64
-ee78e78 SmartRGBYUV: fix odd-width problem with pixel replication
-c9ac204 fix some MSVC64 warning about float conversion
-f4497a1 cpu: check for _MSC_VER before using msvc inline asm
-e2159fd faster RGB->YUV conversion function (~7% speedup)
-21abaa0 Add smart RGB->YUV conversion option -pre 4
-1a161e2 configure: add work around for gcc-4.9 aarch64 bug
-55b10de MIPS: mips32r2: added optimization for BSwap32
-76d2192 Update PATENTS to reflect s/VP8/WebM/g
-29a9db1 MIPS: detect mips32r6 and disable mips32r1 code
-245c4a6 Correctly use the AC_CANONICAL_* macros
-40aa8b6 cosmetics
-2ddcca5 cosmetics: remove some extraneous 'extern's
-f40dd7c vp8enci.h: cosmetics: fix '*' placement
-4610c9c bit_writer: cosmetics: rename kFlush() -> Flush()
-fc3c175 dsp: detect mips64 & disable mips32 code
-c1a7955 cwebp.1: restore quality description
-57a7e73 correct alpha_dithering_strength ABI check
-6c83157 correct WebPMemoryWriterClear ABI check
+7e7b6cc faster rgb565/rgb4444/argb output
+4c7f565 update NEWS
+1f62b6b update AUTHORS
+e224fdc update mailmap
+7110050 bump version to 0.5.0
+230a685 README: update help text, repo link
+d48e427 Merge "demux: accept raw bitstreams"
+99a01f4 Merge "Unify some entropy functions."
+4b025f1 Merge "configure: disable asserts by default"
+92cbddf Merge "fix PrintBlockInfo()"
+ca509a3 Unify some entropy functions.
+367bf90 fix PrintBlockInfo()
+b0547ff move back common constants for lossless_enc*.c into the .h
+fb4c783 lossless: simpler alpha cleanup preprocessing
+ba7f4b6 Merge "anim_diff: add brief description of options"
+47ddd5a Move some codec logic out of ./dsp .
+b4106c4 anim_diff: add brief description of options
+357f455 yuv_sse2: fix 32-bit visual studio build
+b9d80fa configure: disable asserts by default
+7badd3d cosmetic fix: sizeof(type) -> sizeof(*var)
+80ce27d Speed up 24-bit packing / unpacking in YUV / RGB conversions.
+68eebcb remove a TODO about rotation
+2dee296 remove few obsolete TODO about aligned loads in SSE2
+e0c0bb3 remove TODO about unused ref_lf_delta[]
+9cf1cc2 remove few TODO:   * 256 -> RD_DISTO_MULT   * don't use TDisto for UV mode picking
+7918964 Merge changes from topic 'demux-fragment-cleanup'
+47399f9 demux: remove GetFragment()
+d3cfb79 demux: remove dead fragment related TODO
+ab714b8 demux, Frame: remove is_fragment_ field
+b105921 yuv_sse2, cosmetics: fix indent
+466c92e demux,WebPIterator: remove fragment_num/num_fragments
+11714ff demux: remove WebPDemuxSelectFragment
+c0f7cc4 fix for bug #280: UMR in next->bits
+578beeb Merge "enc/Makefile.am: add missing headers"
+1a819f0 makefile.unix: make visibility=hidden the default
+d4f9c2e enc/Makefile.am: add missing headers
+846caff configure: check for -fvisibility=hidden
+3f3ea2c demux: accept raw bitstreams
+d6dad5d man cwebp: add precision about exactness of the 'lossless' mode
+46bb1e3 Merge "gifdec: remove utils.h include"
+2b882e9 Merge "Makefile.vc: define WEBP_HAVE_GIF for gifdec.c"
+892b923 Merge "man/*, AUTHORS: clarify origin of the tool"
+e5687a1 Merge "fix optimized build with -mcmodel=medium"
+e56e685 Makefile.vc: define WEBP_HAVE_GIF for gifdec.c
+4077d94 gifdec: remove utils.h include
+b5e30da man/*, AUTHORS: clarify origin of the tool
+b275e59 fix optimized build with -mcmodel=medium
+64da45a cosmetics, cwebp: fix indent
+038a060 Merge "add disto-based refinement for UV mode (if method = 1 or 2)"
+2835089 Provide an SSE2 implementation of CombinedShannonEntropy.
+e6c9351 add disto-based refinement for UV mode (if method = 1 or 2)
+04507dc Merge "fix undefined behaviour during shift, using a cast"
+793c526 Merge "wicdec: add support for reading from stdin"
+d3d1639 Optimize the heap usage in HistogramCombineGreedy.
+202a710 fix undefined behaviour during shift, using a cast
+14d27a4 improve method #2 by merging DistoRefine() and  SimpleQuantize()
+cb1ce99 Merge "10% faster table-less SSE2/NEON version of YUV->RGB conversion"
+ac761a3 10% faster table-less SSE2/NEON version of YUV->RGB conversion
+79fcf29 wicdec: add support for reading from stdin
+015f173 Merge "cwebp: add support for stdin input"
+a9947c3 cwebp: add support for stdin input
+7eb01ff Merge "Improved alpha cleanup for the webp encoder when prediction transform is used."
+fb8c910 Merge "introduce WebPMemToUint32 and WebPUint32ToMem for memory access"
+bd91af2 Merge "bit_reader: remove aarch64 BITS TODO"
+6c702b8 Speed up hash chain initialization using memset.
+4c60f63 make ReadPNG and ReadJPEG take a filename instead of a FILE
+464ed10 bit_reader: remove aarch64 BITS TODO
+d478e58 Merge "configure: update issue tracker"
+6938111 Improved alpha cleanup for the webp encoder when prediction transform is used.
+2c08aac introduce WebPMemToUint32 and WebPUint32ToMem for memory access
+010ca3d Fix FindMatchLength with non-aligned buffers.
+a90e1e3 README: add prerequisites for an autoconf build
+458f086 configure: update issue tracker
+3391459 vwebp: work around the transparent background with GLUT bug
+e4a7eed cosmetics: fix indent
+0837512 Merge "Make a separate case for low_effort in CopyImageWithPrediction"
+aa2eb2d Merge "cosmetics: fix indent"
+b7551e9 cosmetics: fix indent
+5bda52d Make a separate case for low_effort in CopyImageWithPrediction
+66fa598 Merge "configure: fix intrinsics build w/older gcc"
+5ae220b backward_references.c: Fixed compiler warning
+1556da0 Merge "configure: restore 2 warnings"
+71a17e5 configure: restore 2 warnings
+9eeabc0 configure: fix intrinsics build w/older gcc
+363babe Merge "fix some warning about unaligned 32b reads"
+a141178 Optimization in hash chain comparison for 64 bit Arrays were compared 32 bits at a time, it is now done 64 bits at a time. Overall encoding speed-up is only of 0.2% on @skal's small PNG corpus. It is of 3% on my initial 1.3 Mp desktop screenshot image.
+829bd14 Combine Huffman cost and bit entropy into one loop
+a7a954c Merge "lossless: make prediction in encoder work per scanline"
+61b605b Merge "fix of undefined multiply (int32 overflow)"
+239421c lossless: make prediction in encoder work per scanline
+f5ca40e fix of undefined multiply (int32 overflow)
+5cd2ef4 Merge changes from topic 'win-threading-compat'
+76ce918 Makefile.vc: enable WEBP_USE_THREAD for windows phone
+d2afe97 thread: use CreateThread for windows phone
+0fd0e12 thread: use WaitForSingleObjectEx if available
+63fadc9 thread: use InitializeCriticalSectionEx if available
+110ad58 thread: use native windows cond var if available
+912c9fd dec/webp: use GetLE(24|32) from utils
+f169448 utils/GetLE32: correct uint32 promotion
+158763d Merge "always call WebPInitSamplers(), don't try to be smart"
+3770f3b Merge "cleanup the YFIX/TFIX difference by removing some code and #define"
+a40f60a Merge "3% speed improvement for lossless webp encoder for low effort mode:"
+ed1c2bc always call WebPInitSamplers(), don't try to be smart
+b8c44f1 3% speed improvement for lossless webp encoder for low effort mode:
+997e103 cleanup the YFIX/TFIX difference by removing some code and #define
+d73d1c8 Merge "Make discarding invisible RGB values (cleanup alpha) the default."
+1f9be97 Make discarding invisible RGB values (cleanup alpha) the default.
+f240117 Make dwebp listen more to the -quiet flag
+b37b017 fix for issue #275: don't compare to out-of-bound pointers
+21735e0 speed-up trivial one-symbol decoding case for lossless
+397863b Refactor CopyPlane() and CopyPixels() methods: put them in utils.
+6ecd72f Re-enable encoding of alpha plane with color cache for next release.
+1f7148a Merge "remove unused fields from WebPDecoderOptions and WebPBitstreamFeatures"
+6ae395f Merge "use ExReadFile() for ReadYUV()"
+8076a00 gitignore list: add anim_diff.
+1c1702d use ExReadFile() for ReadYUV()
+775d3a3 remove unused fields from WebPDecoderOptions and WebPBitstreamFeatures
+c13245c AnimEncoder: Add a GetError() method.
+688b265 AnimDecoder API: Add a GetDemuxer() method.
+1aa4e3d WebPAnimDecoder: add an option to enable multi-threaded decoding.
+3584abc AnimDecoder: option to decode to common color modes.
+afd5a62 Merge "mux.h does NOT need to include encode.h"
+8550d44 Merge "migrate anim_diff tool from C++ to C89"
+96201e5 migrate anim_diff tool from C++ to C89
+945cfa3 mux.h does NOT need to include encode.h
+8da07e8 Merge "~2x faster SSE2 RGB24toY, BGR24toY, ARGBToY|UV"
+bfd3fc0 ~2x faster SSE2 RGB24toY, BGR24toY, ARGBToY|UV
+0243242 man/cwebp.1, cosmetics: escape '-'s
+96f5b42 man/cwebp: group lossy-only options
+52fdbdf extract some RGB24 to Luma conversion function from enc/ to dsp/
+ab8c230 add missing \n
+8304179 sync NEWS with 0.4.4
+5bd04a0 sync versions with 0.4.4
+8f1fcc1 Merge "Move ARGB->YUV functions from dec/vp8l.c to dsp/yuv.c"
+25bf2ce fix some warning about unaligned 32b reads
+922268f s/TIFF/WebP
+fa8927e Move ARGB->YUV functions from dec/vp8l.c to dsp/yuv.c
+9b37359 Merge "for ReadXXXX() image-readers, use the value of pic->use_argb"
+f7c507a Merge "remove unnecessary #include "yuv.h""
+7861578 for ReadXXXX() image-readers, use the value of pic->use_argb
+14e4043 remove unnecessary #include "yuv.h"
+469ba2c vwebp: fix incorrect clipping w/NO_BLEND
+4b9186b update issue tracker url (master)
+d64d376 change WEBP_ALIGN_CST value to 31
+f717b82 vp8l.c, cosmetics: fix indent after 95509f9
+927ccdc Merge "fix alignment of allocated memory in AllocateTransformBuffer"
+fea94b2 fix alignment of allocated memory in AllocateTransformBuffer
+5aa8d61 Merge "MIPS: rescaler code synced with C implementation"
+e7fb267 MIPS: rescaler code synced with C implementation
+93c86ed Merge "format_constants.h: MKFOURCC, correct cast"
+5d791d2 format_constants.h: MKFOURCC, correct cast
+65726cd dsp/lossless: Average2, make a constant unsigned
+d26d9de Use __has_builtin to check clang support
+12ec204 moved ALIGN_CST into util/utils.h and renamed WEBP_ALIGN_xxx
+a264083 Merge "rescaler: ~20% faster SSE2 implementation for lossless ImportRowExpand"
+3fb600d Merge "wicdec: fix alpha detection w/64bpp BGRA/RGBA"
+67c547f rescaler: ~20% faster SSE2 implementation for lossless ImportRowExpand
+99e3f81 Merge "large re-organization of the delta-palettization code"
+95509f9 large re-organization of the delta-palettization code
+74fb458 fix for weird msvc warning message
+ae49ad8 Merge "SSE2 implementation of ImportRowShrink"
+932fd4d SSE2 implementation of ImportRowShrink
+badfcba wicdec: fix alpha detection w/64bpp BGRA/RGBA
+35cafa6 Merge "iosbuild: fix linking with Xcode 7 / iOS SDK 9"
+b0c9d8a label rename: NO_CHANGE -> NoChange
+b4e731c neon-implementation for rescaler code
+db1321a iosbuild: fix linking with Xcode 7 / iOS SDK 9
+6dfa5e3 rescaler: better handling of the fxy_scale=0 special case.
+55c0529 Revert "rescaler: better handling of the fxy_scale=0 special case."
+9f226bf rescaler: better handling of the fxy_scale=0 special case.
+f7b8f90 delta_palettization.*: add copyright
+c1e1b71 Changed delta palette to compress better
+0dd2826 Merge "Add delta_palettization feature to WebP"
+48f66b6 Add delta_palettization feature to WebP
+27933e2 anim_encoder: drop a frame if it has same pixels as the prev frame.
+df9f6ec Merge "webpmux/DisplayInfo: send non-error output to stdout"
+8af4993 Merge "rescaler_mips_dsp_r2: cosmetics, fix indent"
+2b9d249 Merge "rescaler: cosmetics, join two lines"
+cc020a8 webpmux/DisplayInfo: send non-error output to stdout
+a288e74 configure: add -Wshorten-64-to-32
+c4c3cf2 pngdec: fix type conversion warnings
+bef8e97 webpmux: fix type conversion warning
+5a84460 rescaler_mips_dsp_r2: cosmetics, fix indent
+acde0aa rescaler: cosmetics, join two lines
+306ce4f rescaler: move the 1x1 or 2x1 handling one level up
+cced974 remove _mm_set_epi64x(), which is too specific
+56668c9 fix warnings about uint64_t -> uint32_t conversion
+76a7dc3 rescaler: add some SSE2 code
+1df1d0e rescaler: harmonize function protos
+9ba1894 rescaler: simplify ImportRow logic
+5ff0079 fix rescaler vertical interpolation
+cd82440 VP8LAllocateHistogramSet: align histogram[] entries
+a406b1d Merge "fix memory over-allocation in lossless rescaler init"
+0fde33e add missing const in VP8InitFrame signature
+ac7d5e8 fix memory over-allocation in lossless rescaler init
+017f8cc Loosen the buffer size checks for Y/U/V/A too.
+15ca501 loosen the padding check on buffer size
+d623a87 dec_neon: add whitespace around stringizing operator
+29377d5 dsp/mips: cosmetics: add whitespace around XSTR macro
+eebaf97 dsp/mips: add whitespace around stringizing operator
+d39dc8f Create a WebPAnimDecoder API.
+03fb752 gif2webp: print output file size
+14efabb Android: limit use of cpufeatures
+7b83adb preparatory cosmetics for Rescaler code fix and clean-up
+77fb41c dec/vp8l/DecodeAlphaData: remove redundant cast
+90fcfcd Insert less hash chain entries from the beginnings of long copies.
+bd55604 SSE2: add yuv444 converters, re-using yuv_sse2.c
+41a5d99 add a -quiet option to 'dwebp'
+80ab3ed Merge "README: update dwebp help output after 1e595fe"
+32b71b2 README: update dwebp help output after 1e595fe
+3ec1182 use the DispatchAlpha() call from dsp
+c5f0062 incorporate bzero() into WebPRescalerInit() instead of call site
+3ebcdd4 remove duplicate "#include <stdlib.h>"
+1e595fe dwebp: add -resize as a synonym for -scale
+24a9693 dec: allow 0 as a scaling dimension
+b918724 utils/rescaler: add WebPRescalerGetScaledDimensions
+923e8ed Merge "update NEWS"
+020fd09 Merge "WebPPictureDistortion: support ARGB format for 'pic' when computing distortion."
+6a5292f update NEWS
+56a2e9f WebPPictureDistortion: support ARGB format for 'pic' when computing distortion.
+0ae582e configure: test and add -Wunreachable-code
+c2f9dc0 bit_writer: convert VP8L macro values to immediates
+b969f88 Reduce magic in palette reordering
+acb297e anim_diff: add a -raw_comparison flag
+155c1b2 Merge changes I76f4d6fe,I45434639
+717e4d5 mips32/mipsDSPr2: function ImportRow rebased
+7df9389 fix rescaling bug (uninitialized read, see bug #254).
+5cdcd56 lossless_enc_neon: add VP8LTransformColor
+a53c336 lossless_neon: add VP8LTransformColorInverse
+99131e7 Merge changes I9fb25a89,Ibc648e9e
+c455676 simplify the main loop for downscaling
+2a010f9 lossless_neon: remove predictors 5-13
+ca221bb ll_enc_neon: enable VP8LSubtractGreenFromBlueAndRed
+585d93d Container spec: clarify ordering of ALPH chunk.
+01d61fd lossless: ~20 % speedup
+f722c8f lossless: Speed up ComputeCacheEntropy by 40 %
+1ceecdc add a VP8LColorCacheSet() method for color cache
+17eb609 lossless: Allow copying from prev row in rle-mode.
+f3a7a5b lossless: bit writer optimization
+d97b9ff Merge changes from topic 'lossless-enc-improvements'
+0250dfc msvc: fix pointer type warning in BitsLog2Floor
+52931fd lossless: combine the Huffman code with extra bits
+c4855ca lossless: Inlining add literal
+8e9c94d lossless: simplify HashChainFindCopy heuristics
+888429f lossless: 0.5 % compression density improvement
+7b23b19 lossless: Add zeroes into the predicted histograms.
+85b44d8 lossless: encoding, don't compute unnecessary histo
+d92453f lossless: Remove about 25 % of the speed degradation
+2cce031 Faster alpha coding for webp
+5e75642 lossless: rle mode not to accept lengths smaller than 4.
+84326e4 lossless: Less code for the entropy selection
+16ab951 lossless: 0.37 % compression density improvement
+822f113 add WebPFree() to the API
+0ae2c2e SSE2/SSE41: optimize SSE_16xN loops
+39216e5 cosmetics: fix indent after 32462a07
+559e54c Merge "SSE2: slightly faster FTransformWHT"
+8ef9a63 SSE2: slightly faster FTransformWHT
+f27f773 lossless_neon: enable VP8LAddGreenToBlueAndRed
+36e9c4b SSE2: minor cosmetrics on in-loop filter code
+4741fac dsp/lossless_*sse2: remove some unnecessary inlines
+1819965 fix warning ("left shift of negative value") using a cast
+7017001 SSE2: speed-up some lossless-encoding functions
+abcb012 Merge "SSE2: slightly faster (~5%) AddGreenToBlueAndRed()"
+2df5bd3 Merge "Speedup to HuffmanCostCombinedCount"
+9e356d6 SSE2: slightly faster (~5%) AddGreenToBlueAndRed()
+fc6c75a SSE2: 53% faster TransformColor[Inverse]
+49073da SSE2: 46% speed-up of TransformColor[Inverse]
+32462a0 Speedup to HuffmanCostCombinedCount
+f3d687e SSE4.1 implementation of some lossless encoding functions
+bfc300c SSE4.1 implementation of some alpha-processing functions
+7f9c98f Merge "sse2 in-loop: simplify SignedShift8b() a bit"
+ef314a5 dec_sse2/GetNotHEV: micro optimization
+a729cff sse2 in-loop: simplify SignedShift8b() a bit
+422ec9f simplify Load8x4() a bit
+8df238e Merge "remove some duplicate FlipSign()"
+751506c remove some duplicate FlipSign()
+65ef5af Merge "lossless: 0.13% compression density gain"
+2beef2f lossless: 0.13% compression density gain
+3033f24 lossless: 0.06 % compression density improvement
+64960da dec_neon: add VE8uv / VE16
+14dbd87 dec_neon: add HE8uv / HE16
+ac76801 introduce FTransform2 to perform two transforms at a time.
+aa6065a dec_neon: use vld1_dup(mem) rather than vdup(mem[0])
+8b63ac7 Merge "dec_neon: add TM16"
+f51be09 Merge "dec_neon/TrueMotion: simply left border load"
+dc48196 dec_neon: add TM16
+ea95b30 dec_neon/TrueMotion: simply left border load
+f262d61 speed-up SetResidualSSE2
+bf46d0a fix mips2 build target
+929a0fd enc_sse2/TTransform: simplify abs calculation
+17dbd05 enc_sse2/CollectHistogram: simplify abs calculation
+a6c1593 dec_neon: add DC16 intra predictors
+03b4f50 Makefile.vc: add anim_diff build support.
+1b98987 Merge changes I9cd84125,Iee7e387f,I7548be72
+acd7b5a Introduce a test tool anim_diff.
+f274a96 dsp/enc_sse2: add luma4 intra predictors
+040b11b dsp/enc_sse2: add chroma intra predictors
+aee021b dsp/enc_sse2: add luma16 intra predictors
+9e00a49 makefile.unix: remove superclean target
+cefc9c0 makefile.unix: clean up after extras target
+4c9af02 dec_neon: add DC8uvNoTopLeft
+dd55b87 Merge "doc/webp-container-spec: update repo browser link"
+f048696 doc/webp-container-spec: update repo browser link
+9287761 Merge "GetResidualCostSSE2: simplify abs calculation"
+0e00936 dsp/cpu.c(x86): check maximum supported cpuid feature
+b243a4b GetResidualCostSSE2: simplify abs calculation
+6d4602b Merge "fix typo: constitutes -> constitute"
+5fe1fe3 fix typo: constitutes -> constitute
+b83bd7c Merge "populate 'libwebpextras' with: import gray, rgb565 and rgb4444 functions"
+b0114a3 Merge "histogram.h: cosmetics: remove unnecessary includes"
+feab45e gifdec: Move inclusion of webp/config.h to header.
+dbba67d histogram.h: cosmetics: remove unnecessary includes
+e978fec Merge "VP8LBitReader: fix remaining ubsan error with large shifts"
+d6fe588 Merge "ReconstructRow: move some one-time inits out of the main loop"
+a21d647 ReconstructRow: move some one-time inits out of the main loop
+7a01c3c VP8LBitReader: fix remaining ubsan error with large shifts
+7fa67c9 change GetPixPairHash64() return type to uint32_t
+ec1fb9f Merge "dsp/enc.c: cosmetics: move DST() def closer to use"
+7073bfb Merge "split 64-mult hashing into two 32-bit multiplies"
+0768b25 dsp/enc.c: cosmetics: move DST() def closer to use
+6a48b8f Merge "fix MSVC size_t->int conversion warning"
+1db07cd Merge "anim_encode: cosmetics: fix alignment"
+e28271a anim_encode: cosmetics: fix alignment
+7fe357b split 64-mult hashing into two 32-bit multiplies
+af74c14 populate 'libwebpextras' with: import gray, rgb565 and rgb4444 functions
+6121413 remove VP8Residual::cost unused field
+e254482 fix MSVC size_t->int conversion warning
+b69a6c3 vwebp: don't redefine snprintf with VS2015+
+0ac29c5 AnimEncoder API: Consistent use of trailing underscores in struct.
+d484555 AnimEncoder API: Use timestamp instead of duration as input to Add().
+9904e36 dsp/dec_sse2: DC8uv / DC8uvNoLeft speedup
+7df2049 dsp/dec_sse2: DC16 / DC16NoLeft speedup
+8e515df Merge "makefile.unix: add some missing headers"
+db12250 cosmetics: vp8enci.h: break long line
+bf516a8 makefile.unix: add some missing headers
+b44eda3 dsp: add DSP_INIT_STUB
+03e76e9 clarify the comment about double-setting the status in SetError()
+9fecdd7 remove unused EmitRGB()
+43f010d move ReconstructRow to top
+82d9802 add a dec/common.h header to collect common enc/dec #defines
+5d4744a Merge "enc_sse41: add Disto4x4 / Disto16x16"
+e38886a mux.h: Bump up ABI version
+46305ca configure: add --disable-<avx2|sse4.1|sse2>
+2fc8b65 CPPFLAGS->CFLAGS for detecting sse4.1 in preprocessor
+1a338fb enc_sse41: add Disto4x4 / Disto16x16
+9405550 encoding SSE4.1 stub for StoreHistogram + Quantize + SSE_16xN
+c64659e remove duplicate variables after the lossless{_enc}.c split
+67ba7c7 enc_sse2: call local FTransform in CollectHistogram
+1824979 dsp: s/VP8LSetHistogramData/VP8SetHistogramData/
+ede5e15 cosmetics: dsp/lossless.h: reorder prototypes
+553051f dsp/lossless: split enc/dec functions
+9064adc Merge "conditionally add -msse4.1 in Makefile.unix"
+cecf509 dsp/yuv*.c: rework WEBP_USE_<arch> ifdef
+6584d39 dsp/upsampling*.c: rework WEBP_USE_<arch> ifdef
+8080942 dsp/rescaler*.c: rework WEBP_USE_<arch> ifdef
+1d93dde dsp/lossless*.c: rework WEBP_USE_<arch> ifdef
+73805ff dsp/filters*.c: rework WEBP_USE_<arch> ifdef
+fbdcef2 dsp/enc*.c: rework WEBP_USE_<arch> ifdef
+66de69c dsp/dec*.c: rework WEBP_USE_<arch> ifdef
+48e4ffd dsp/cost*.c: rework WEBP_USE_<arch> ifdef
+29fd6f9 dsp/argb*.c: rework WEBP_USE_<arch> ifdef
+80ff381 dsp/alpha*.c: rework WEBP_USE_<arch> ifdef
+bf09cf1 conditionally add -msse4.1 in Makefile.unix
+e9570dd stub for SSE4.1 support.
+4a95384 Merge "dsp: add sse4.1 detection"
+cabf4bd dsp: add sse4.1 detection
+4ecba1a thread.h: rename interface param
+b8d706c Merge "sync versions with 0.4.3"
+ae64a71 Merge "add shell for libwebpextras"
+92a5da9 sync versions with 0.4.3
+9d4e2d1 Merge "~30% faster smart-yuv (-pre 4) with early-out criterion"
+b1bdbba ~30% faster smart-yuv (-pre 4) with early-out criterion
+7efb974 Merge "Disable NEON code on Native Client"
+ac4f578 Disable NEON code on Native Client
+0873f85 AnimEncoder API: Support input frames in YUV(A) format.
+5c176d2 add shell for libwebpextras
+44bd956 fix signature for VP8RecordCoeffTokens()
+c9b8ea0 small cosmetics on TokenBuffer.
+76394c0 Merge "MIPS: dspr2: added optimization for TrueMotion"
+0f77369 WebPPictureRescale: add a note about 0 width/height
+241bb5d MIPS: dspr2: added optimization for TrueMotion
+6cef0e4 examples/Android.mk: add webpmux_example target
+53c16ff Android.mk: add webpmux target
+21852a0 Android.mk: add webpdemux target
+8697a3b Android.mk: add webpdecoder{,_static} targets
+4a67049 Android.mk: split source lists per-directory
+b5e7942 MIPS: dspr2: Added optimization for some convert functions
+0f595db MIPS: dspr2: Added optimization for some convert functions
+8a218b4 MIPS: [mips32|dspr2]: GetResidualCost rebased
+ef98750 Speedup method StoreImageToBitMask by 5%.
+602a00f fix iOS arm64 build with Xcode 6.3
+2382050 1-2% faster encoding by removing an indirection in GetResidualCost()
+eddb7e7 MIPS: dspr2: added otpimization for DC8uv, DC8uvNoTop and DC8uvNoLeft
+73ba291 MIPS: dspr2: added optimization for functions RD4 and LD4
+c7129da Merge "4-5% faster encoding using SSE2 for GetResidualCost"
+94380d0 MIPS: dspr2: added optimizaton for functions VE4 and DC4
+2a40709 4-5% faster encoding using SSE2 for GetResidualCost
+17e1986 Merge "MIPS: dspr2: added optimization for simple filtering functions"
+3ec404c Merge "dsp: normalize WEBP_TSAN_IGNORE_FUNCTION usage"
+b969f5d dsp: normalize WEBP_TSAN_IGNORE_FUNCTION usage
+d7b8e71 MIPS: dspr2: added optimization for simple filtering functions
+235f774 Merge "MIPS: dspr2: Added optimization for function VP8LTransformColorInverse_C"
+42a8a62 MIPS: dspr2: Added optimization for function VP8LTransformColorInverse_C
+b442bef Merge "ApplyFiltersAndEncode: only copy lossless stats"
+b510fbf doc/webp-container-spec: note MSB order for chunk diagrams
+9bc0f92 ApplyFiltersAndEncode: only copy lossless stats
+3030f11 Merge "dsp/mips: add some missing TSan annotations"
+dfcf459 Merge "MIPS: dspr2: Added optimization for function VP8LAddGreenToBlueAndRed_C"
+55c75a2 dsp/mips: add some missing TSan annotations
+2cb879f MIPS: dspr2: Added optimization for function VP8LAddGreenToBlueAndRed_C
+e155601 move some cost tables from enc/ to dsp/
+c3a0316 Merge "picture_csp: fix build w/USE_GAMMA_COMPRESSION undefined"
+39537d7 Merge "VP8LDspInitMIPSdspR2: add missing TSan annotation"
+1dd419c picture_csp: fix build w/USE_GAMMA_COMPRESSION undefined
+43fd354 VP8LDspInitMIPSdspR2: add missing TSan annotation
+c7233df Merge "VP8LDspInit: remove memcpy"
+0ec4da9 picture_csp::InitGammaTables*: add missing TSan annotations
+35579a4 VP8LDspInit: remove memcpy
+97f6aff VP8YUVInit: add missing TSan annotation
+f9016d6 dsp/enc::InitTables: add missing TSan annotation
+e3d9771 VP8EncDspCostInit*: add missing TSan annotations
+d97c143 Merge "doc/webp-container-spec: cosmetics"
+309b790 MIPS: mips32: Added optimization for function SetResidualCoeffs
+a987fae MIPS: dspr2: added optimization for function GetResidualCost
+e7d3df2 doc/webp-container-spec: cosmetics
+be6635e Merge "VP8TBufferClear: remove some misleading const's"
+02971e7 Merge "VP8EmitTokens: remove unnecessary param void cast"
+3b77e5a VP8TBufferClear: remove some misleading const's
+aa139c8 VP8EmitTokens: remove unnecessary param void cast
+c24d8f1 cosmetics: upsampling_sse2: add const to some casts
+1829c42 cosmetics: lossless_sse2: add const to some casts
+183168f cosmetics: enc_sse2: add const to some casts
+860badc cosmetics: dec_sse2: add const to some casts
+0254db9 cosmetics: argb_sse2: add const to some casts
+1aadf85 cosmetics: alpha_processing_sse2: add const to some casts
+1579de3 vwebp: clear canvas at the beginning of each loop
+4b9fa5d Merge "webp-container-spec: clarify background clear on loop"
+4c82284 Updated the near-lossless level mapping.
+5603947 webp-container-spec: clarify background clear on loop
+19f0ba0 Implement true-motion prediction in SSE2
+774d4cb make VP8PredLuma16[] array non-const
+d7eabb8 Merge "MIPS: dspr2: Added optimization for function CollectHistogram"
+fe42739 Use integers for kmin/kmax for simplicity.
+b9df35f AnimEncode API: kmax=0 should imply all keyframes.
+6ce296d MIPS: dspr2: Added optimization for function CollectHistogram
+2c906c4 vwebp: remove unnecessary static Help() prototype
+be0fd1d Merge "dec/vp8: clear 'dither_' on skipped blocks"
+e96170f Merge "vwebp/animation: display last frame on end-of-loop"
+0f017b5 vwebp/animation: display last frame on end-of-loop
+c86b40c enc/near_lossless.c: fix alignment
+66935fb dec/vp8: clear 'dither_' on skipped blocks
+b7de794 Merge "lossless_neon: enable subtract green for aarch64"
+77724f7 SSE2 version of GradientUnfilter
+416e1ce lossless_neon: enable subtract green for aarch64
+72831f6 Speedup AnalyzeAndInit for low effort compression.
+a659748 Speedup Analyze methods for lossless compression.
+98c8138 Enable Near-lossless feature.
+c6b2454 AnimEncoder API: Fix for kmax=1 and default kmin case.
+022d2f8 add SSE2 variants for alpha filtering functions
+2db15a9 Temporarily disable encoding of alpha plane with color cache.
+1d575cc Merge "Lossless decoding: Remove an unnecessary if condition."
+cafa1d8 Merge "Simplify backward refs calculation for low-effort."
+7afdaf8 Alpha coding: reorganize the filter/unfiltering code
+4d6d728 Simplify backward refs calculation for low-effort.
+ec0d1be Cleaup Near-lossless code.
+9814ddb Remove the post-transform near-lossless heuristic.
+4509e32 Lossless decoding: Remove an unnecessary if condition.
+f2ebc4a Merge "Regression fix for lossless decoding"
+783a8cd Regression fix for lossless decoding
+9a062b8 AnimEncoder: Bugfix for kmin = 1 and kmax = 2.
+0f027a7 simplify smart RGB->YUV conversion code
+0d5b334 BackwardReferencesHashChainFollowChosenPath: remove unused variable
+f480d1a Fix to near lossless artefacts on palettized images.
+d4615d0 Merge changes Ia1686828,I399fda40
+cb4a18a rename HashChainInit into HashChainReset
+f079e48 use uint16_t for chosen_path[]
+da09121 MIPS: dspr2: Added optimization for function FTransformWHT
+b8c2013 Merge "wicdec: (msvs) quiet some /analyze warnings"
+9b228b5 wicdec: (msvs) quiet some /analyze warnings
+daeb276 Merge "MIPS: dspr2: Added optimization for MultARGBRow function"
+cc08742 Merge "dsp/cpu: (msvs) add include for __cpuidex"
+4a82aab Merge changes I87544e92,I0bb6cda5
+7a19139 dwebp/WritePNG: mark png variables volatile
+775dfad dwebp: include setjmp.h w/WEBP_HAVE_PNG
+47d26be dwebp: correct sign in format strings
+f0e0677 VP8LEncodeStream: add an assert
+c5f7747 VP8LColorCacheCopy: promote an int before shifting
+0de5f33 dsp/cpu: (msvs) add include for __cpuidex
+7d850f7 MIPS: dspr2: Added optimization for MultARGBRow function
+5487529 MIPS: dspr2: added optimization for function QuantizeBlock
+4fbe9cf dsp/cpu: (msvs) avoid immintrin.h on _M_ARM
+3fd5903 simplify/reorganize arguments for CollectColorBlueTransforms
+b9e356b Disable costly TraceBackwards for method=0.
+a7e7caa MIPS: dspr2: added optimization for function TransformColorRed
+2cb3918 Merge "MIPS: dspr2: added optimization for function TransformColorBlue"
+279e661 Merge "dsp/cpu: add include for _xgetbv() w/MSVS"
+b6c0428 dsp/cpu: add include for _xgetbv() w/MSVS
+d1c4ffa gif2webp: Move GIF decoding related code to a support library.
+07c3955 Merge "AnimEncoder API: Add info in README.mux"
+7b16197 MIPS: dspr2: added optimization for function TransformColorBlue
+d7c4b02 cpu: fix AVX2 detection for gcc/clang targets
+9d29946 AnimEncoder API: Add info in README.mux
+d581ba4 follow-up: clean up WebPRescalerXXX dsp function
+f8740f0 dsp: s/USE_INTRINSICS/WEBP_USE_INTRINSICS/
+ce73abe Merge "introduce a separate WebPRescalerDspInit to initialize pointers"
+ab66bec introduce a separate WebPRescalerDspInit to initialize pointers
+205c7f2 fix handling of zero-sized partition #0 corner case
+cbcdd5f Merge "move rescaler functions to rescaler* files in src/dsp/"
+bf586e8 Merge changes I230b3532,Idf3057a7
+6dc79dc Merge "anim_encode: fix type conversion warnings"
+11fce25 Merge "dec_neon: remove returns from void functions"
+c4e63f9 Makefile.vc: add gif2webp target
+4f43d38 enable NEON for Windows ARM builds
+3f6615a Makefile.vc: add rudimentary Windows ARM support
+e7c5954 dec_neon: remove returns from void functions
+f79c163 anim_encode: fix type conversion warnings
+0f54f1e Remove gif2webp_util which is no longer needed.
+cbcbedd move rescaler functions to rescaler* files in src/dsp/
+ac79ed1 webpmux: remove experimental fragment handling
+e8694d4 mux: remove experimental FRGM parsing
+9e92b6e AnimEncoder API: Optimize single-frame animated images
+abbae27 Merge "Move over gif2webp to the new AnimEncoder API."
+a28c4b3 MIPS: move WORK_AROUND_GCC define to appropriate place
+012d2c6 MIPS: dspr2: added optimization for functions SSEAxB
+67720c8 Move over gif2webp to the new AnimEncoder API.
+9241ecf MIPS: dspr2: added optimization for function Average
+9422211 Merge "Tune BackwardReferencesLz77 for low_effort (m=0)."
+df40057 Merge "Speedup VP8LGetHistoImageSymbols for low effort (m=0) mode."
+ea08466 Tune BackwardReferencesLz77 for low_effort (m=0).
+b0b973c Speedup VP8LGetHistoImageSymbols for low effort (m=0) mode.
+c6d3292 argb_sse2: cosmetics
+67f601c make the 'last_cpuinfo_used' variable names unique
+b948986 AnimEncoder API: Init method for default options.
+856f8ec Merge "AnimEncoder API: Remove AnimEncoderFrameOptions."
+c537514 Merge "AnimEncoder API: GenerateCandidates bugfix."
+dc0ce03 Merge "AnimEncoder API: Compute change rectangle for first frame too."
+f00b639 Merge "AnimEncoder API: In Assemble(), always set animation parameters."
+29ed796 Merge "AnimEncoder lib cleanup: prev to prev canvas not needed."
+9f0dd6e Merge "WebPAnimEncoder API: Header and implementation"
+5e56bbe AnimEncoder API: Remove AnimEncoderFrameOptions.
+b902c3e AnimEncoder API: GenerateCandidates bugfix.
+ef3c39b AnimEncoder API: Compute change rectangle for first frame too.
+eec423a AnimEncoder API: In Assemble(), always set animation parameters.
+ae1c046 AnimEncoder lib cleanup: prev to prev canvas not needed.
+4b997ae WebPAnimEncoder API: Header and implementation
+72208be move argb_*.o build target to encoder list
+9592053 Merge "multi-thread fix: lock each entry points with a static var"
+4c1b300 Merge "SSE2 implementation of VP8PackARGB"
+fbcc200 Merge "add -Wformat-nonliteral and -Wformat-security"
+80d950d add -Wformat-nonliteral and -Wformat-security
+04c20e7 Merge "MIPS: dspr2: added optimization for function Intra4Preds"
+a437694 multi-thread fix: lock each entry points with a static var
+ca7f60d SSE2 implementation of VP8PackARGB
+72d573f simplify the PackARGB signature
+4e2589f demux: restore strict fragment flag check
+4ba8e07 Merge "webp-container-spec: remove references to fragments"
+e752f0a Merge "demux: remove experimental FRGM parsing"
+f8abb11 Merge changes I109ec4d9,I73fe7743
+ae2188a MIPS: dspr2: added optimization for function Intra4Preds
+1f4b864 move VP8EncDspARGBInit() call closer to where it's needed
+14108d7 dec_neon: add DC8uvNoTop / DC8uvNoLeft
+d8340da dec_neon: add DC8uv
+a66e66c webp-container-spec: remove references to fragments
+7ce8788 MIPS: dspr2: added optimization for function MakeARGB32
+012e623 demux: remove experimental FRGM parsing
+87c3d53 method=0: Don't evaluate any predictor
+6f4fcb9 Merge "MIPS: dspr2: added optimization for function ImportRow"
+2428445 replace unneeded calls to HistogramCopy() by swaps
+bdf7b40 MIPS: dspr2: added optimization for function ImportRow
+e66a922 Merge "MIPS: dspr2: added optimization for function ExportRowC"
+c279fec MIPS: dspr2: added optimization for function ExportRowC
+31a9cf6 Speedup WebP lossless compression for low effort (m=0) mode with following: - Disable Cross-Color transform. - Evaluate predictors #11 (paeth), #12 and #13 only.
+9275d91 MIPS: dspr2: added optimization for function TrueMotion
+26106d6 Merge "enc_neon: fix building with non-Xcode clang (iOS)"
+1c4e3ef unroll the kBands[] indirection to remove a dereference in GetCoeffs()
+a3946b8 enc_neon: fix building with non-Xcode clang (iOS)
+8ed9c00 Merge "simplify the Histogram struct, to only store max_value and last_nz"
+bad7757 simplify the Histogram struct, to only store max_value and last_nz
+3cca0dc MIPS: dspr2: Added optimization for DCMode function
+37e395f MIPS: fix functions to use generic BPS istead of hardcoded value
+9475bef PickBestUV: fix VP8Copy16x8 invocation
+441f273 Merge changes I55f8da52,Id73a1e96
+4a279a6 cosmetics: add some missing != NULL comparisons
+66ad372 factorize BPS definition in dsp.h and add VP8Copy16x8
+432e5b5 make ALIGN_xxx naming consistent
+5760604 encoder: switch BPS to 32 instead of 16
+1b66bbe MIPS: dspr2: added optimization for function TransformColor_C
+c6d0f9e histogram: cosmetics
+f399d30 Merge changes I6eac17e5,I32d2b514
+9de9074 dec_neon: add TM8uv
+8e517ec bit_reader/kVP8NewRange: range_t -> uint8_t
+e185713 dsp: initialize VP8PredChroma8 in VP8DspInit()
+e0c809a Move Entropy methods to lossless.c
+a96ccf8 iosbuild: add x64_64 simulator support
+a0df551 Remove handling for WEBP_HINT_GRAPH
+413dfc0 Move static method definition before its usage.
+0f23566 Update BackwardRefsWithLocalCache.
+d69e36e Remove TODOs from lossless encoder code.
+fdaac8e Optmize VP8LGetBackwardReferences LZ77 references.
+2f0e2ba MIPS: dspr2: added optimization for function Select
+a3e79a4 Merge "WebPEncode: Support encoding same pic twice (even if modified)"
+e4f4ddd WebPEncode: Support encoding same pic twice (even if modified)
+cbc3fbb Merge "Updated VP8LGetBackwardReferences and color cache."
+95a9bd8 Updated VP8LGetBackwardReferences and color cache.
+54f2c14 MIPS: dspr2: added optimization for function FTransform
+aa42f42 MIPS: dspr2: Added optimization for function VP8LSubtractGreenFromBlueAndRed
+11a25f7 Merge "FlattenSimilarBlocks should only be tried when blending is possible."
+5cccdad FlattenSimilarBlocks should only be tried when blending is possible.
+95ca44a MIPS: dspr2: added optimization for Disto4x4
+4171b67 backward_references.c: reindent after c8581b0
+c8581b0 Optimize BackwardReferences for RLE encoding.
+5798eee MIPS: dspr2: unfilters bugfix (Ie7b7387478a6b5c3f08691628ae00f059cf6d899)
+4167a3f Optimize backwardreferences
+d18554c Merge "webp/types.h: use inline for clang++/-std=c++11"
+7489b0e gif2webp: Add '-min-size' option to get best compression.
+77bdddf Speed up BackwardReferences
+6638710 webp/types.h: use inline for clang++/-std=c++11
+abf0420 Enable entropy based merge histo for (q<100)
+572022a filters_mips_dsp_r2.c: disable unfilters
+a28e21b MIPS: dspr2: Added optimization for function ClampedAddSubtractFull
+18d5a1e MIPS: dspr2: added optimization for function ClampedAddSubtractHalf
+829a8c1 MIPS: dspr2: added optimization for ITransform
+c94ed49 gif2webp: Use the default hint instead of WEBP_HINT_GRAPH.
+653ace5 Increase the MAX_COLOR_CACHE_BITS from 9 to 10.
+919220c Change the logic adjusting the Histogram bits.
+53b096c Merge "Fix bug in VP8LCalculateEstimateForCacheSize."
+e912bd5 Fix bug in VP8LCalculateEstimateForCacheSize.
+541d783 Merge "dec_neon: add RD4 intra predictor"
+f8cd067 Merge "Makefile.vc: add a 'legacy' RTLIBCFG option"
+22881c9 dec_neon: add RD4 intra predictor
+613d281 update NEWS
+1304eb3 Merge "dec_neon: DC4: use pair-wise adds for top row"
+34c20c0 Makefile.vc: add a 'legacy' RTLIBCFG option
+7083006 Merge "dsp/dec_{neon,sse2}: VE4: normalize variable names"
+0db9031 dsp/dec_{neon,sse2}: VE4: normalize variable names
+b5bc153 dec_neon: DC4: use pair-wise adds for top row
+5b90d8f Unify the API between VP8BitWriter and VP8LBitWriter
+f7ada56 Merge changes I2e06907b,Ia9ed4ca6,I782282ff
+5beb6bf Merge "dec_neon: add VE4 intra predictor"
+eba6ce0 dec_neon: add DC4 intra predictor
+79abfbd dec_neon: add TM4 intra predictor
+fe395f0 dec_neon: add LD4 intra predictor
+32de385 dec_neon: add VE4 intra predictor
+72395ba Merge "Modify CostModel to allocate optimal memory."
+65e5eb8 gif2webp: Support GIF_DISPOSE_RESTORE_PREVIOUS
+e4c829e gif2webp: Handle frames with odd offsets + disposal to background.
+c2b5a03 Modify CostModel to allocate optimal memory.
+b7a33d7 implement VE4/HE4/RD4/... in SSE2
+97c76f1 make VP8PredLuma4[] non-const and initialize array in VP8DspInit()
+0ea8c6c Merge "PrintReg: output to stderr"
+d7ff2f9 Merge "stopwatch.h: fix includes"
+f85ec71 PrintReg: output to stderr
+54edbf6 stopwatch.h: fix includes
+139142e Optimize BackwardReferenceHashChainFollowPath.
+5f36b68 enc/backward_references.c: fix indent
+e0e9960 Merge "sync version numbers to 0.4.2 release"
+64ac514 sync version numbers to 0.4.2 release
+c24f895 Simplify and speedup Backward refs computation.
+d1c359e fix shared object build with -fvisibility=hidden
+a4c3a31 WEBP_TSAN_IGNORE_FUNCTION: fix gcc compat warning
+f358eeb add code for testing random incremental decoding in dwebp
+8024729 mark some init function as being safe for thread_sanitizer.
+79b5bdb bit_reader.h: cosmetics: fix a typo
+6c67368 Improved near-lossless mode.
+0ce27e7 enc_mips32: workaround gcc-4.9 bug
+aca1b98 enc/vp8l.c: fix indent
+ca00502 Evaluate non-palette compression for palette image
+c8a87bb AssignSegments: quiet -Warray-bounds warning
+32f67e3 Merge "enc_neon: initialize vectors w/vdup_n_u32"
+fabc65d 1-3% faster encoding optimizing SSE_NxN functions
+7534d71 enc_neon: initialize vectors w/vdup_n_u32
+5f81391 Merge "Fix return code of EncodeImageInternal()"
+e321abe Fix return code of EncodeImageInternal()
+f82cb06 optimize palette ordering
+f545fee don't set the alpha value for histogram index image
+2d9b0a4 add WebPDispatchAlphaToGreen() to dsp
+1bd4c2a Merge "Change Entropy based Histogram Combine heuristic."
+e295b8f Merge "iosbuild: cleanup"
+1be4e76 Merge "iosbuild: output autoconf req. on failure"
+d5e498d Change Entropy based Histogram Combine heuristic.
+47a2d8e fix MSVC float->int conversion warning
+041956f iosbuild: cleanup
+767eb40 iosbuild: output autoconf req. on failure
+35ad48b HistoHeapInit: correct positions allocation size
+45d9635 lossless: entropy clustering for high qualities.
+dc37df8 fix type warning for VS9_x64
+9f7d9e6 iosbuild: make iOS 6 the minimum requirement
+fdd6528 Remove unused VP8LDecoder member variable
+ea3bba5 Merge "rewrite Disto4x4 in enc_neon.c with intrinsic"
+f060dfc add lossless incremental decoding support
+ab70794 rewrite Disto4x4 in enc_neon.c with intrinsic
+d447163 MIPS: dspr2: added optimization for function FilterLoop24
+2aef54d Merge "prepare VP8LDecodeImage for incremental decode"
+aed0f5a Merge "MIPS: dspr2: added optimization for function FilterLoop26"
+2863068 prepare VP8LDecodeImage for incremental decode
+248f3ae remove br->error_ field
+49e1504 MIPS: dspr2: added optimization for function FilterLoop26
+38128cb iobuild.sh: only install .h files in Headers
+c792d41 Premultiply with alpha during U/V downsampling
+0cc811d gif2webp: Background color correction
+d7167ff Amend the lossless spec according to issue #205, #206 and #224
+b901416 Record the lossless size stats.
+cddd334 Add a WebPExtractAlpha function to dsp
+0716a98 fix indent after I0204949917836f74c0eb4ba5a7f4052a4797833b
+f9ced95 Optimize lossless decoding for trivial(ARB) codes.
+924fcfd Merge "webpmux: simplify InitializeConfig()"
+c0a462c webpmux: simplify InitializeConfig()
+6986bb5 webpmux: fix indent
+f89e169 webpmux: fix exit status on numeric value parse error
+2172cb6 Merge "webpmux: fix loop_count range check"
+e3b343e Merge "examples: warn on invalid numeric parameters"
+0e23c48 webpmux: fix loop_count range check
+6208338 Merge "fix loop bug in DispatchAlpha()"
+d51f3e4 gif2webp: Handle frames with missing  graphic control extension
+690b491 fix loop bug in DispatchAlpha()
+96d43a8 examples: warn on invalid numeric parameters
+3101f53 MIPS: dspr2: added optimization for TransformOne
+a6bb9b1 SSE2 for inverse Mult(ARGB)Row and ApplyAlphaMultiply
+d84a8ff Remove default initialization of decoder status.
+be70b86 configure: simplify libpng-config invocation
+e0a9932 Rectify bug in lossless incremental decoding.
+e2502a9 MIPS: dspr2: added optimization for TransformAC3
+24e1072 MIPS: dspr2: added optimization for TransformDC
+c0e84df Merge "Slightly faster lossless decoding (1%)"
+8dd28bb Slightly faster lossless decoding (1%)
+f010359 MIPS: dspr2: added optimization for ColorIndexInverseTransforms
+d3242ae make VP8LSetBitPos() set br->eos_ flag
+a9decb5 Lossless decoding: fix eos_ flag condition
+3fea6a2 fix erroneous dec->status_ setting
+80b8099 MIPS: dspr2: add some specific mips code to commit I2c3f2b12f8df15b785fad5a9c56316e954ae0c53
+e564062 Merge "further refine the COPY_PATTERN optim for DecodeAlpha"
+854509f enc/histogram.c: reindent after f4059d0
+3442196 Merge "~3-5% faster encoding optimizing PickBestIntra*()"
+865069c further refine the COPY_PATTERN optim for DecodeAlpha
+a595622 added C-level optimization for DecodeAlphaData function
+187d379 add a fallback to ALPHA_NO_COMPRESSION
+a48a2d7 ~3-5% faster encoding optimizing PickBestIntra*()
+a614019 ExUtilReadFromStdin: (windows) open stdin in bin mode
+e80eab1 webpmux: (windows) open stdout in binary mode
+e9bfb11 cwebp: (windows) open stdout in binary mode
+5927e15 example_util: add ExUtilSetBinaryMode
+30f3b75 webpmux man page: Clarify some title, descriptions and examples
+77d4c7e address cosmetic comments from patch #71380
+f75dfbf Speed up Huffman decoding for lossless
+637b388 dsp/lossless: workaround gcc-4.9 bug on arm
+8323a90 dsp.h: collect gcc/clang version test macros
+e6c4b52 move static initialization of WebPYUV444Converters[] to the Init function.
+49911d4 Merge "fix indentation"
+f4059d0 Code cleanup for HistogramRemap.
+e632b09 fix indentation
+f5c04d6 Merge "add a DispatchAlpha() for SSE2 that handles 8 pixels at a time"
+fc98edd add a DispatchAlpha() for SSE2 that handles 8 pixels at a time
+73d361d introduce VP8EncQuantize2Blocks to quantize two blocks at a time
+0b21c30 MIPS: dspr2: added optimization for EmitAlphaRGB
+953acd5 enc_neon: enable QuantizeBlock for aarch64
+f4ae143 MIPS: mips32: code rebase
+5697715 MIPS: dspr2: added optimizations for VP8YuvTo*
+2523aa7 SmartRGBYUV: fix odd-width problem with pixel replication
+ee52dc4 fix some MSVC64 warning about float conversion
+3fca851 cpu: check for _MSC_VER before using msvc inline asm
+e2a83d7 faster RGB->YUV conversion function (~7% speedup)
+de2d03e Merge "Add smart RGB->YUV conversion option -pre 4"
+3fc4c53 Add smart RGB->YUV conversion option -pre 4
+b4dc406 MIPS: dspr2: added optimization for (un)filters
+137e609 Merge "configure: add work around for gcc-4.9 aarch64 bug"
+b61c9ce MIPS: dspr2: Optimization of some simple point-sampling functions
+e2b8cec configure: add work around for gcc-4.9 aarch64 bug
+98c5410 MIPS: mips32r2: added optimization for BSwap32
+dab702b Update PATENTS to reflect s/VP8/WebM/g
+b564f7c Merge "MIPS: detect mips32r6 and disable mips32r1 code"
+b7e5a5c MIPS: detect mips32r6 and disable mips32r1 code
+63c2fc0 Correctly use the AC_CANONICAL_* macros
+bb07022 Merge "cosmetics"
+e300c9d cosmetics
+0e519ee Merge "cosmetics: remove some extraneous 'extern's"
+3ef0f08 Merge "vp8enci.h: cosmetics: fix '*' placement"
+4c6dde3 bit_writer: cosmetics: rename kFlush() -> Flush()
+f7b4c48 cosmetics: remove some extraneous 'extern's
+b47fb00 vp8enci.h: cosmetics: fix '*' placement
+b5a36cc add -near_lossless [0..100] experimental option
+0524d9e dsp: detect mips64 & disable mips32 code
+d3485d9 cwebp.1: fix quality description placement
+29a9fe2 Merge tag 'v0.4.1'
 8af2771 update ChangeLog (tag: v0.4.1, origin/0.4.1, 0.4.1)
+e09e9ff Record & log the image pre-processing time.
 f59c0b4 iosbuild.sh: specify optimization flags
 8d34ea3 update ChangeLog (tag: v0.4.1-rc1)
 dbc3da6 makefile.unix: add vwebp.1 to the dist target