Merge pull request #2059 from sijchen/mt52
1, refactor to remove similar function (WriteSliceToFrameBs is removed)
Merge pull request #2054 from HaiboZhu/Bugfix_pps_update_logic
Fix the pps update logic error
Merge pull request #2051 from huili2/sh_check_slice_type
add slice_type check for num_ref_frame
Merge pull request #2049 from ruil2/rc6
Merge pull request #2047 from huili2/T1_error_handling
bugfix for T1 error decoder dealing
Merge pull request #2039 from shihuade/NewSHA1TableV1.3
Merge pull request #2040 from mstorsjo/fix-warnings
Avoid warnings about comparison between signed and unsigned
Merge pull request #2038 from HaiboZhu/Bugfix_disable_redundant_picture
Merge pull request #2035 from HaiboZhu/Bugfix_width_height_level_limit_check
Merge pull request #2034 from HaiboZhu/BugFix_CheckAccessUnitBoundaryExt
syn/unify with local SHA1 script and update SHA1 tables
Merge pull request #2032 from GuangweiWang/sub8x8asm
Merge pull request #2033 from HaiboZhu/Bugfix_FMO_judgement_error
Merge pull request #2028 from sijchen/mt42
Disable redundant picture to avoid the memory overflow
Merge pull request #2031 from mstorsjo/astyle-tabs
Fix the bug that decoder not return when level limit occur
Fix max slice group number judgement error in FMO
Fix logic errors in CheckAccessUnitBoundaryExt
Merge pull request #2029 from mstorsjo/remove-tabs
Make astyle convert tabs to spaces
Remove unnecessary tabs in the middle of lines
Merge pull request #2027 from sijchen/dbp2
Add new x86 assembly functions to support sub8x8 mode
Merge pull request #2023 from ethanhugg/refcount
Merge pull request #2024 from huili2/submbtype_modif
save memory usage and improve error return logic under VLCOVERFLOW
use func pointer for deblocking to save judgements at each slice
Merge pull request #2026 from sijchen/fix_rc
Add new ARM AArch64 assembly functions to support sub8x8 mode
let qp clip to min/max qp available in all rc settings, so as to avoid qp exceed 51
add new AArch32 asm functions to support sub8x8 mode
modify sub_mb_type to be faster, and fix a function nam typo
Change GMP code to RefCount instead of deleting on enc/decode complete.
Merge pull request #2020 from brion/emscripten
Merge pull request #2022 from sijchen/fix_slc
Merge pull request #2021 from sijchen/mt32
put pSliceBsBuffer to thread buffer rather than per slice, so as to save memory usage
Allow compilation to JavaScript via emscripten
fix a first-mb judgement and enablings a test
Merge pull request #2015 from huili2/sub8_final
Merge pull request #2013 from sijchen/mem10
add sub8x8 mode decision functions
refactor for future simplification
Merge pull request #2010 from sijchen/mt2
correct the termination of error handling under multi-thread
Merge pull request #2011 from mstorsjo/fix-arflags
Merge pull request #2004 from sijchen/ps55
Merge pull request #2003 from sijchen/rc_trace2
Merge pull request #2009 from ruil2/rc3
Merge pull request #1997 from deadNightTiger/deterministic-build
Don't concatenate $@ to AR_OPTS
use the external qp setting for screen content
Move -D flag for ar to platform-linux.mk
Enable deterministic mode for ar
first version of post-frame skipping for rate control (not enabled)
Merge pull request #1999 from bmiklautz/build
Make the static library directory configurable
Merge pull request #1993 from huili2/dec_mem_stat
Merge pull request #1995 from sijchen/ps22
Merge pull request #1994 from sijchen/mt1
Merge pull request #1992 from sijchen/ps11
Merge pull request #1988 from sijchen/ps4
Merge pull request #1989 from sijchen/memc
remove the core-1 setting for multi-thread, if application has concern of using all thread, it can be done by setting specific value to iMultipleThreadIdc
remove one always-on macro to save code complexity, and use core-1 for multi-threading in real-time
Merge pull request #1984 from mstorsjo/aarch64-fix
Fix the NEON register restoring in McHorVer30WidthEq8_AArch64_neon
remove build error if MEMORY_CHECK is on
add a frame-skip setting option
Merge pull request #1979 from GuangweiWang/lumaMc
Merge pull request #1983 from mstorsjo/remove-tabs
make two un-error ending of encoding go to a same exit and save the duplicated calling of UpdateSpatialPictures and ForceCodingIDR
move the post-encoding exchanging in preprocessing to more reasonable function
Convert tabs to spaces before comments
Remove tabs from vertically aligned function declarations and typedefs
Remove tabs in struct and class definitions
Remove tabs where a simple space is just enough
Convert tabs to spaces in vertically aligned code
Convert tabs to spaces in defines
Remove tabs from commented out code
Change indentation to use tabs instead of spaces in a few test project files
Run astyle on a newly added function
Convert tabs to spaces within comments
bugfix:save neon register v8-v15
Merge pull request #1981 from HaiboZhu/Debug_mosaic_T30
Merge pull request #1980 from ruil2/rc_update
Merge pull request #1978 from sijchen/doc
Merge pull request #1975 from sijchen/rc_tr