Commits · 6df41d50d936c428e1e5239a2eded54ee13e9156 · VideoLAN / x264

21 Apr, 2008 1 commit
- --asm to allow testing of different versions of asm without recompile · 6df41d50
  Loren Merritt authored Apr 20, 2008
  
  6df41d50
17 Apr, 2008 2 commits
- copy left neighbor pixels directly from previous mb instead of main plane · 87132ed6
  Loren Merritt authored Apr 12, 2008
  
  87132ed6
- cacheline split workaround for mc_luma · 6eb54835
  Fiona Glaser authored Apr 09, 2008
  
  6eb54835
16 Apr, 2008 1 commit
- add "SECTION_RODATA" before "SECTION .text" to setup the fakegot label used in macho binaries. · c1e43f09
  Guillaume Poirier authored Apr 16, 2008
```
This fixes compilation with --enable-pic
Requires Yasm 0.7.0 or newer
Patch by Dave Lee % davelee P com A gmail P com %
```
  c1e43f09
13 Apr, 2008 1 commit
- more hpel fixes · 67813bbf
  Loren Merritt authored Apr 13, 2008
  
  67813bbf
12 Apr, 2008 2 commits
- update msvc projectfile · 0acaad1b
  Gabriel Bouvigne authored Apr 10, 2008
  
  0acaad1b
- r810 borked hpel_filter_sse2 on unaligned buffers · 25558461
  Loren Merritt authored Apr 11, 2008
  
  25558461
10 Apr, 2008 1 commit
- threads=auto on multicore now implies thread input, just like explicit thread numbers already did · 0f453853
  Loren Merritt authored Apr 10, 2008
  
  0f453853
09 Apr, 2008 6 commits
- dct4 sse2 · 539103a5
  Loren Merritt authored Apr 08, 2008
  
  539103a5
- faster x86_32 dct8 · 9168abfa
  Loren Merritt authored Apr 08, 2008
  
  9168abfa
- macros to deal with macros that permute their arguments · 56bf7565
  Loren Merritt authored Apr 07, 2008
  
  56bf7565
- mmx cachesplit sad of non-square sizes checked height instead of width · 32ef8652
  Loren Merritt authored Apr 07, 2008
  
  32ef8652
- sfence after nontemporal stores · afbcfdc2
  Loren Merritt authored Apr 04, 2008
  
  afbcfdc2
- simplify hpel filter asm (move control flow to C) and add sse2, ssse3 versions · 7bdaab60
  Loren Merritt authored Apr 02, 2008
  
  7bdaab60
04 Apr, 2008 1 commit
- more mmx/xmm macros (mova, movu, movh) · 29899d84
  Loren Merritt authored Apr 03, 2008
  
  29899d84
02 Apr, 2008 2 commits
- improve handling of cavlc dct coef overflows · 937b7925
  Fiona Glaser authored Apr 02, 2008
```
support large coefs in high profile, and clip to allowed range in baseline/main
```
  937b7925
- fix shared libs on MacOSX · bdfec13d
  Guillaume Poirier authored Mar 31, 2008
```
based on a patch by İsmail Dönmez
```
  bdfec13d
31 Mar, 2008 2 commits
- typo in r803 · 658b0586
  Loren Merritt authored Mar 31, 2008
  
  658b0586
- fix a crash on mp4 muxing with invalid params · def7e3aa
  Loren Merritt authored Mar 30, 2008
  
  def7e3aa
30 Mar, 2008 6 commits
- variance-based psy adaptive quantization · b59440f0
  Fiona Glaser authored Mar 30, 2008
```
new options: --aq-mode --aq-strength
AQ is enabled by default
```
  b59440f0
- fix naming of .dll on mingw · 8d8f3ea4
  Zuxy Meng authored Mar 29, 2008
  
  8d8f3ea4
- don't distinguish between mingw and cygwin · 7e3ef7ce
  Loren Merritt authored Mar 29, 2008
  
  7e3ef7ce
- remove a memset · 05e91fb1
  Fiona Glaser authored Mar 29, 2008
  
  05e91fb1
- typo. don't evaluate rd pskip when p16x16 found ref>0. · 6f44349a
  Loren Merritt authored Mar 29, 2008
  
  6f44349a
- r784 borked lossless dc zigzag · 27b73b3b
  Loren Merritt authored Mar 29, 2008
  
  27b73b3b
25 Mar, 2008 7 commits
- fix an arithmetic overflow that disabled SEA threshold after finding a mv with SAD < mvcost. · c1c00e6c
  Loren Merritt authored Mar 25, 2008
  
  c1c00e6c
- fix hpel_filter_altivec picked up by checkasm · 1c72e719
  Guillaume Poirier authored Mar 25, 2008
```
Patch by Manuel %maaanuuu A gmx.net % and Noboru Asai % noboru P asai A gmail P com %
```
  1c72e719
- faster residual · 66a0c19d
  Loren Merritt authored Mar 25, 2008
  
  66a0c19d
- nasm doesn't like align(nop) in structs · 41cd480c
  Loren Merritt authored Mar 24, 2008
  
  41cd480c
- reduce the size of some cabac arrays · 727377bf
  Loren Merritt authored Mar 24, 2008
  
  727377bf
- use cabac context transition table from trellis in normal residual coding too · c9e8cfed
  Fiona Glaser authored Mar 24, 2008
  
  c9e8cfed
- rearrange cabac struct to reduce code size · a3e11cbf
  Fiona Glaser authored Mar 24, 2008
  
  a3e11cbf
24 Mar, 2008 4 commits
- higher precision RD lambda · 9289e806
  Fiona Glaser authored Mar 24, 2008
```
improves quality at QP<=12.
```
  9289e806
- faster cabac_encode_ue_bypass · aaced086
  Loren Merritt authored Mar 24, 2008
  
  aaced086
- cabac asm. · 23e52ef3
  Loren Merritt authored Mar 23, 2008
```
mostly because gcc refuses to use cmov.
28% faster than c on core2, 11% on k8, 6% on p4.
```
  23e52ef3
- cosmetics in cabac · ecc95abd
  Loren Merritt authored Mar 23, 2008
  
  ecc95abd
23 Mar, 2008 1 commit
- inline cabac_size_decision · 9e7cfc35
  Loren Merritt authored Mar 22, 2008
  
  9e7cfc35
22 Mar, 2008 3 commits
- cosmetics in DECLARE_ALIGNED · 542027fa
  Loren Merritt authored Mar 22, 2008
  
  542027fa
- don't distinguish between luma4x4 and luma4x4ac · 52fb8334
  Loren Merritt authored Mar 22, 2008
  
  52fb8334
- faster lossless zigzag · b437d2d4
  Loren Merritt authored Mar 22, 2008
  
  b437d2d4