summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-06-17Merge "Change bitreader to use a larger window."John Koleszar
2010-06-17CHANGELOG: 0.9.1John Koleszar
Change-Id: Icca54b9d51becc49255193801762e1936a07aa2d
2010-06-16Update AUTHORSJohn Koleszar
Change-Id: I1c6a275278788dfdc630ed436d2c770acfcbd097
2010-06-16Generate AUTHORS file with a scriptJohn Koleszar
This information is in git, so it's better to use that as a source than updating this file manually. This script can be run manually at release time for now, or we can set up a cron job sometime in the future. Change-Id: I0344135ceb9c04ed14e2e2d939a93194e35973db
2010-06-16Merge "Avoid encoding garbage when ivfenc encounters an unsupported Y4M file."John Koleszar
2010-06-16Merge "gen_scalers: fix 64-bit integer promotion bug"John Koleszar
2010-06-16Avoid encoding garbage when ivfenc encounters an unsupported Y4M file.Tom Finegan
This change stops ivfenc from treating unsupported Y4M files as raw input. For example, if given an interlaced Y4M file, ivfenc treated the input as if it were raw data because the unsupported Y4M file case previously fell through without being handled. Change-Id: I06caa50f3448e6388741a77346daaebf77c277e1
2010-06-16gen_scalers: fix 64-bit integer promotion bugJohn Koleszar
i needs to be treated as signed to get the proper indexing on 64-bit platforms. This behavior was accidentally reverted when fixing an unsigned/signed comparison warning. Change-Id: Ic306d609bdc8de94c8f8ba29c6e45c736101a82e
2010-06-15Change bitreader to use a larger window.Timothy B. Terriberry
Change bitreading functions to use a larger window which is refilled less often. This makes it cheap enough to do bounds checking each time the window is refilled, which avoids the need to copy the input into a large circular buffer. This uses less memory and speeds up the total decode time by 1.6% on an ARM11, 2.8% on a Cortex A8, and 2.2% on x86-32, but less than 1% on x86-64. Inlining vp8dx_bool_decoder_fill() has a big penalty on x86-32, as does moving the refill loop to the front of vp8dx_decode_bool(). However, having the refill loop between computation of the split values and the branch in vp8_decode_mb_tokens() is a big win on ARM (presumably due to memory latency and code size: refilling after normalization duplicates the code in the DECODE_AND_BRANCH_IF_ZERO and DECODE_AND_LOOP_IF_ZERO cases. Unfortunately, refilling at the end of vp8dx_bool_decoder_fill() and at the beginning of each decode step in vp8_decode_mb_tokens() means the latter requires an extra refill at the end. Platform-specific versions could avoid the problem, but would require most of detokenize.c to be duplicated. Change-Id: I16c782a63376f2a15b78f8086d899b987204c1c7
2010-06-15Merge "More on "some XMM registers are non-volatile on windows x64 ABI""Yunqing Wang
2010-06-15More on "some XMM registers are non-volatile on windows x64 ABI"Yunqing Wang
Add same fix in subpixel_sse2.asm. Change-Id: Icfda6103cbf74ec43308e96961dd738aa823c14d
2010-06-14VisualStudio projects: asm tool updatesJames Zern
vs8 - pull yasm.rules [1] into the source tree to avoid need to install file into VC/VCProjectDefaults - reference same w/ToolFile & RelativePath - update arm branch to match vs7: - quote source file paths passed to yasm [1]: http://www.tortall.net/svn/yasm/trunk/yasm/Mkfiles/vc9/yasm.rules@2271 Change-Id: I52b801496340cd7b1d0023d12afbc04624ecefc3
2010-06-14vp8_cx_iface: set default cpu used to 0John Koleszar
Change-Id: I7b35f4717cdd204224112f72471b551617262417
2010-06-14Fix compiler warningsGuillermo Ballester Valor
Change-Id: I2a97f08cc3c7808ce5be39e910cc5147ecf03a1d
2010-06-14sse2 version of vp8_regular_quantize_bScott LaVarnway
Added sse2 version of vp8_regular_quantize_b which improved encode performance(for the clip used) by ~10% for 32 bit builds and ~3% for 64 bit builds. Also updated SHADOW_ARGS_TO_STACK to allow for more than 9 arguments. Change-Id: I62f78eabc8040b39f3ffdf21be175811e96b39af
2010-06-14Merge "Use local pointer to pbi->common."Paul Wilkins
2010-06-14Merge "platform autodetect: accept amd64 as a synonym for x86_64"John Koleszar
2010-06-14Remove useless 500 frame limitFabio Pedretti
Change-Id: Ib82de60cf32cf08844c3e2d88d7c587396f3892c
2010-06-14Merge "ivfenc: fix two-pass support of raw files"John Koleszar
2010-06-14Use public domain implementation for MD5 algorithmAndres Mejia
The RSA Data Security, Inc. implementation license bears a requirement similar to the old problematic BSD license with advertising clause. Change-Id: I877b71ff0548934b1c4fd87245696f53dedbdf26
2010-06-14ivfenc: fix two-pass support of raw filesJohn Koleszar
Commit 3245d46 "ivfenc: support reading/writing from a pipe" broke support for two pass encodes of raw files when done in two invocations of ivfenc. The raw image was only set up on pass 0, which was never hit when running with --pass=2 --passes=2. Change-Id: I6a9858be1a8998d5bd45331123b46b1baa05b379
2010-06-13Merge "Make this/next iiratio unsigned."John Koleszar
2010-06-13Merge "Tuning of baseline Rd equation to improve behavior at the"Paul Wilkins
2010-06-13Merge "Incorrect comment."Paul Wilkins
2010-06-12Make this/next iiratio unsigned.John Koleszar
This patch addresses issue #79, which is a regression since commit 28de670 "Fix RD bug." If the coded error value is zero, the iiratio calculation effectively multiplies by 1000000 by the DOUBLE_DIVIDE_CHECK macro. This can result in a value larger than INT_MAX, giving a negative ratio. Since the error values are conceptually unsigned (though they're stored in a double) this patch makes the iiratio values unsigned, which allows the clamping to work as expected.
2010-06-12Merge "require --enable-psnr to build ssim"John Koleszar
2010-06-11ivfenc: support reading/writing from a pipeJohn Koleszar
Use - for the filename to use stdin/stdout. Update to avoid opening the file multiple times. Change-Id: I356356fa16bb334d4b22abc531dc03c0d95917a3
2010-06-11Merge "Enable vp8_sad16x16x4d_sse3 in non-RTCD case"John Koleszar
2010-06-11Merge "Change preprocessor check to _WIN32"Yaowu Xu
2010-06-11Enable vp8_sad16x16x4d_sse3 in non-RTCD caseJohn Koleszar
Typo caused C version of 16x16x4 SAD to be called when built with --disable-runtime-cpu-detect. Change-Id: I0fe6fa67280b3a5f13acb3c8ed914f039aaaf316
2010-06-11require --enable-psnr to build ssimJohn Koleszar
ssim.c comiles in a huge (512M) amount of global scratch space. Allocating this data on the heap would be a better solution, but this file doesn't need to be built at all in most cases, so as a first pass, disable it except when doing opsnr.stt output (--enable-psnr). Change-Id: I320d812f6d652a12516a16b52295ebff20b5bd42
2010-06-11platform autodetect: accept amd64 as a synonym for x86_64John Koleszar
Thanks to James Cloos <cloos at jhcloos dot com> for the tip. Change-Id: If377cc084dd7c16a4f51191a2aa0d83e7117ebec
2010-06-11some XMM registers are non-volatile on windows x64 ABIMakoto Kato
XMM6 to XMM15 are non-volatile on Windows x64 ABI. We have to save these registers. Change-Id: I4676309f1350af25c8a35f0c81b1f0499ab99076
2010-06-11Incorrect comment.Paul Wilkins
(Thanks to Ronald S. Bultje)
2010-06-11Use local pointer to pbi->common.Paul Wilkins
2010-06-11Tuning of baseline Rd equation to improve behavior at thePaul Wilkins
low and high Q ends.
2010-06-11Change preprocessor check to _WIN32Frank Galligan
Change-Id: I841dc0b8ebb150ac998f4076c148d7bb187e4301
2010-06-11Merge "Improve vp8_sixtap_predict functions"Yunqing Wang
2010-06-10replace while(0) construct with if/elseJohn Koleszar
No good reason to be tricky here. I don't know why 'break' occurred to me as the natrual replacement for the 'return', but an if/else block is definitely clearer. Change-Id: I08a336307afeb0dc7efa494b37398f239f66c2cf
2010-06-10Fix new MV clamping scheme for chroma MVs.Timothy B. Terriberry
The new scheme introduced in I68d35a2f did not clamp chroma MVs in the SPLITMV case, and clamped them incorrectly (to the luma plane bounds) in every other case. Because chroma MVs are computed from the luma MVs before clamping occurs, they could still point outside of the frame buffer and cause crashes. This clamping happens outside of the MV prediction loop, and so should not affect bitstream decoding.
2010-06-10Remove reference to 'vpx Technologies'John Koleszar
Vestigial. Change-Id: Iffa9e6d5ba5199b136d7549890101da17c11e3c3
2010-06-10Fix MinGW toolchain detectionJohn Koleszar
Updated the comment in change I6bef2ab5, but missed adding the code to the commit. Change-Id: I14d300489b79730e3995175bfe5f9271b569abe3
2010-06-10Improve vp8_sixtap_predict functionsYunqing Wang
Restructure vp8_sixtap_predict functions to eliminate extra 5-line calculation while doing first-pass only. Also, combline functions to eliminate usage of intermediate buffer. This gives decoder a 3% performance gain on my test clips. Change-Id: I13de49638884d1a57d0855c63aea719316d08c1b
2010-06-10Merge "Detect toolchain based on gcc -dumpmachine"John Koleszar
2010-06-10Detect toolchain based on gcc -dumpmachinePhilip Jägenstedt
Using uname fails e.g. on a 64-bit machine with a 32-bit toolchain. The following gcc -dumpmachine strings have been verified: * 32-bit Linux gives i486-linux-gnu * 64-bit Linux gives x86_64-linux-gnu * Mac OS X 10.5 gives i686-apple-darwin9 * MinGW gives mingw32 *darwin8* and *bsd* can safely be assumed to be correct, but *cygwin* is a guess. Change-Id: I6bef2ab5e97cbd3410aa66b0c4f84d2231884b05
2010-06-10Merge "Adjust to avoid long line"Paul Wilkins
2010-06-10Adjust to avoid long linePaul Wilkins
2010-06-10Merge "Correct comment"Paul Wilkins
2010-06-10Merge "Fix RD bug."Paul Wilkins
2010-06-09Merge "Remove secondary mv clamping from decode stage"John Koleszar
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-10 19:21:09 +0000