changelog

视频压缩编解码标准MPEG4商业级别的VC代码实现标准

2004-09-04 14:11:43 GMT	                                	patch-85

    Summary:
      Fixed function prototype/definition mismatch
    Revision:
      xvidcore--head--0.0--patch-85

    From sysKin:
     * Fixed function prototype/definition mismatch for some
       interpolation C functions.

    modified files:
     src/image/interpolate8x8.c


2004-09-04 14:08:13 GMT	                                	patch-84

    Summary:
      Fixed buffer termination logic in xvid_decraw.
    Revision:
      xvidcore--head--0.0--patch-84

    From ed.gomez:
     * Fixed main decoding loop condition to really match 
       the empty buffer and end of stream condition.
     * Removed the unwanted frame number limitation

    modified files:
     examples/xvid_decraw.c


2004-09-04 14:04:48 GMT	                                	patch-83

    Summary:
      Uninitialized user data usage.
    Revision:
      xvidcore--head--0.0--patch-83

    From ed.gomez:
     - Fixed user data parsing uninitialized data.

    modified files:
     src/bitstream/bitstream.c


2004-09-04 13:59:26 GMT	                                	patch-82

    Summary:
      Unitialized data in bvop ME
    Revision:
      xvidcore--head--0.0--patch-82

    From ed.gomez:
     * Fixed unitialized data usage during bvop ME.

    modified files:
     src/motion/estimation_bvop.c


2004-09-03 00:13:31 GMT	                                	patch-81

    Summary:
      Add VOL header saving in xvid_decraw
    Revision:
      xvidcore--head--0.0--patch-81

    From ed.gomez:
     * Added VOL header saving in xvid_decraw
    
    The little story:
       I was trying to cut some frames off of a big stream (150MB)
       with "xvid_decraw -m", and cat'ing the single frame stream
       files together. The reconstructed stream was rejected by all
       mpeg4 decoders because the vol header wasn't present. Thus
       the fix.

    modified files:
     examples/xvid_decraw.c


2004-08-30 23:22:35 GMT	                                	patch-80

    Summary:
      Complete previous API numbering change
    Revision:
      xvidcore--head--0.0--patch-80

    Complete previous API numbering change
    

    modified files:
     src/xvid.h


2004-08-29 11:53:05 GMT	                                	patch-79

    Summary:
      Merged stable tree
    Revision:
      xvidcore--head--0.0--patch-79

    Merged stable tree
    
    Patches applied:
    
     * [e-mail]--2004-1/xvidcore--stable--1.0--patch-54
       Marking 1.0.2
    
     * [e-mail]--2004-1/xvidcore--stable--1.0--patch-55
       Merged one important forgotten bugfix from head
    
     * [e-mail]--2004-1/xvidcore--stable--1.0--patch-56
       ChangeLog update
    

    modified files:
     ChangeLog-1.0 src/xvid.h

    new patches:
     [e-mail]--2004-1/xvidcore--stable--1.0--patch-54
     [e-mail]--2004-1/xvidcore--stable--1.0--patch-55
     [e-mail]--2004-1/xvidcore--stable--1.0--patch-56


2004-08-29 09:56:34 GMT	                                	patch-78

    Summary:
      Write ia32 symbols' size to elf output.
    Revision:
      xvidcore--head--0.0--patch-78

    From ed.gomez:
     - write symbols size to elf output, so the asm objects look
       really like any usual object file.

    modified files:
     src/bitstream/x86_asm/cbp_3dne.asm
     src/bitstream/x86_asm/cbp_mmx.asm
     src/bitstream/x86_asm/cbp_sse2.asm
     src/dct/x86_asm/fdct_mmx_ffmpeg.asm
     src/dct/x86_asm/fdct_mmx_skal.asm
     src/dct/x86_asm/fdct_sse2_skal.asm
     src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
     src/dct/x86_asm/idct_sse2_dmitry.asm
     src/dct/x86_asm/simple_idct_mmx.asm
     src/image/x86_asm/colorspace_mmx.inc
     src/image/x86_asm/colorspace_rgb_mmx.asm
     src/image/x86_asm/colorspace_yuv_mmx.asm
     src/image/x86_asm/colorspace_yuyv_mmx.asm
     src/image/x86_asm/interpolate8x8_3dn.asm
     src/image/x86_asm/interpolate8x8_3dne.asm
     src/image/x86_asm/interpolate8x8_mmx.asm
     src/image/x86_asm/interpolate8x8_xmm.asm
     src/image/x86_asm/postprocessing_mmx.asm
     src/image/x86_asm/postprocessing_sse2.asm
     src/image/x86_asm/qpel_mmx.asm
     src/image/x86_asm/reduced_mmx.asm
     src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm
     src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm
     src/motion/x86_asm/sad_xmm.asm
     src/quant/x86_asm/quantize_h263_3dne.asm
     src/quant/x86_asm/quantize_h263_mmx.asm
     src/quant/x86_asm/quantize_mpeg_mmx.asm
     src/quant/x86_asm/quantize_mpeg_xmm.asm
     src/utils/x86_asm/cpuid.asm
     src/utils/x86_asm/interlacing_mmx.asm
     src/utils/x86_asm/mem_transfer_3dne.asm
     src/utils/x86_asm/mem_transfer_mmx.asm


2004-08-28 13:00:56 GMT	                                	patch-77

    Summary:
      Thread safety problem in sse2 brightness control
    Revision:
      xvidcore--head--0.0--patch-77

    From ed.gomez:
     * CodingStyle for the sse2 image brightness file
     * Fixed thread safety problem/big error. Writing to a RO
       data segment is a no go ! and using global data segment
       is a no go either (use stack instead) !

    modified files:
     src/image/x86_asm/postprocessing_sse2.asm


2004-08-22 13:11:23 GMT	                                	patch-76

    Summary:
      Stable merge
    Revision:
      xvidcore--head--0.0--patch-76

    Stable merge
    
    Patches applied:
    
     * [e-mail]--2004-1/xvidcore--stable--1.0--patch-53
       Thread safety problem in idct C version
    

    modified files:
     src/bitstream/mbcoding.c src/dct/idct.c

    new patches:
     [e-mail]--2004-1/xvidcore--stable--1.0--patch-53


2004-08-22 11:48:23 GMT	                                	patch-75

    Summary:
      This tree isn't API 4.0 anymore, mark it as 4.1
    Revision:
      xvidcore--head--0.0--patch-75

    From ed.gomez:
     * The fields added to some structs make this lib isn't
       API 4.0 anymore, mark it as 4.1 because ABI compatibility
       is conserved.

    modified files:
     build/generic/configure.in


2004-08-22 11:41:22 GMT	                                	patch-74

    Summary:
      Functions qualified as such for elf format. 
    Revision:
      xvidcore--head--0.0--patch-74

    From ed.gomez:
     * Functions weren't marked as functions in ia32 asm files.
       Added support for the function qualifier for elf.

    modified files:
     build/generic/configure.in src/bitstream/x86_asm/cbp_3dne.asm
     src/bitstream/x86_asm/cbp_mmx.asm
     src/bitstream/x86_asm/cbp_sse2.asm
     src/dct/x86_asm/fdct_mmx_ffmpeg.asm
     src/dct/x86_asm/fdct_mmx_skal.asm
     src/dct/x86_asm/fdct_sse2_skal.asm
     src/dct/x86_asm/idct_3dne.asm src/dct/x86_asm/idct_mmx.asm
     src/dct/x86_asm/idct_sse2_dmitry.asm
     src/dct/x86_asm/simple_idct_mmx.asm
     src/image/x86_asm/colorspace_rgb_mmx.asm
     src/image/x86_asm/colorspace_yuv_mmx.asm
     src/image/x86_asm/colorspace_yuyv_mmx.asm
     src/image/x86_asm/interpolate8x8_3dn.asm
     src/image/x86_asm/interpolate8x8_3dne.asm
     src/image/x86_asm/interpolate8x8_mmx.asm
     src/image/x86_asm/interpolate8x8_xmm.asm
     src/image/x86_asm/postprocessing_mmx.asm
     src/image/x86_asm/postprocessing_sse2.asm
     src/image/x86_asm/qpel_mmx.asm
     src/image/x86_asm/reduced_mmx.asm
     src/motion/x86_asm/sad_3dn.asm src/motion/x86_asm/sad_3dne.asm
     src/motion/x86_asm/sad_mmx.asm src/motion/x86_asm/sad_sse2.asm
     src/motion/x86_asm/sad_xmm.asm
     src/quant/x86_asm/quantize_h263_3dne.asm
     src/quant/x86_asm/quantize_h263_mmx.asm
     src/quant/x86_asm/quantize_mpeg_mmx.asm
     src/quant/x86_asm/quantize_mpeg_xmm.asm
     src/utils/x86_asm/cpuid.asm
     src/utils/x86_asm/interlacing_mmx.asm
     src/utils/x86_asm/mem_transfer_3dne.asm
     src/utils/x86_asm/mem_transfer_mmx.asm


2004-08-21 17:04:57 GMT	                                	patch-73

    Summary:
      Added yasm support in configure.in
    Revision:
      xvidcore--head--0.0--patch-73

    From ed.gomez:
     * Added yasm configure.in support. It's my preferred ia32 assembly
       because it allows debugging/profiling of assembly code with oprofile.

    modified files:
     build/generic/configure.in


2004-08-21 11:47:31 GMT	                                	patch-72

    Summary:
      Merged fix from stable
    Revision:
      xvidcore--head--0.0--patch-72

    Merged fix from stable
    
    Patches applied:
    
     * [e-mail]--2004-1/xvidcore--stable--1.0--patch-52
       Stupid typo+error in fdct_xxx_skal macro generator.
    

    modified files:
     src/dct/x86_asm/fdct_mmx_skal.asm

    new patches:
     [e-mail]--2004-1/xvidcore--stable--1.0--patch-52


2004-08-16 22:32:46 GMT	                                	patch-71

    Summary:
      Decoder optimization (fixing regression)
    Revision:
      xvidcore--head--0.0--patch-71

    From ed.gomez:
     * With newly introduced vector checking, decoder became noticeably
       slower. This was caused by poorly written code (sorry sysKin :P)
        + unrolled loop
        + removed duplicated border computings
        + marks the function as __inline

    modified files:
     src/decoder.c


2004-08-15 11:42:20 GMT	                                	patch-70

    Summary:
      Out of bounds MVs clipping
    Revision:
      xvidcore--head--0.0--patch-70

    From sysKin:
     * Clip vectors that end up out of bounds.

    modified files:
     src/decoder.c


2004-08-10 22:30:09 GMT	                                	patch-69

    Summary:
      Fixed CBR plugin.
    Revision:
      xvidcore--head--0.0--patch-69

    From Foxer:
     * Do not set the return quantizer to the frame's quantizer
       (caused the crazy quant choices)
     * Allow quant1 to influence the sequence quality
     * Allow more than +- 1 quantizer variation if the desired
       quantizer is much higher than the previous.
     * Clamp the overflow influence to 1 unit of buffer, that
       should help cases where still motion scenes are followed
       by normal motion scenes... old code was reaching amazing
       high bitrate, with this modification it should keep
       smaller bitrate.

    modified files:
     src/plugins/plugin_single.c


2004-08-01 15:23:49 GMT	                                	patch-68

    Summary:
      error in dshow par array indexing
    Revision:
      xvidcore--head--0.0--patch-68

    error in dshow par array indexing
    

    modified files:
     dshow/src/CXvidDecoder.cpp


2004-08-01 13:38:36 GMT	                                	patch-67

    Summary:
      Faster bframe decoding (qpel this time)
    Revision:
      xvidcore--head--0.0--patch-67

    From ed.gomez:
     * Used the same trick as for halfpel bvops, merge backward
       interpolation and dst averaging steps.
    
    NB: i'm currently not able to say if it's a real speedup or not
        because my linux kernel uses a process scheduler gives great
        variance to results... so far i'm sure this isn't a slowdown
        neither for C nor ia32 SIMD.

    modified files:
     src/decoder.c src/image/qpel.c src/image/qpel.h


2004-08-01 11:24:07 GMT	                                	patch-66

    Summary:
      Unified qpel code path for all platforms
    Revision:
      xvidcore--head--0.0--patch-66

    From ed.gomez and skal:
     * Unified qpel code path for all platforms. Next step is
       to fully exploit this code path to speedup qpel bframe
       decoding
    
    NB: this makes also ports life easier as they would not port
        obsoleted function sets...

    modified files:
     src/decoder.c src/image/interpolate8x8.h src/image/qpel.c
     src/image/qpel.h src/motion/motion_comp.c


2004-07-31 15:08:19 GMT	                                	patch-65

    Summary:
      Faster bframe decoding.
    Revision:
      xvidcore--head--0.0--patch-65

    From ed.gomez and skal:
     * Faster direct/interpolated bvop blocks decoding for halfpel
       sequences. The trick is to compute and average directly with
       destination during one of the forward/backward interpolations.
       At this moment, this patch covers only halfpel decoding, the
       same trick is expected to be hacked for qpel.

    modified files:
     src/decoder.c src/image/interpolate8x8.c
     src/image/interpolate8x8.h
     src/image/x86_asm/interpolate8x8_mmx.asm
     src/image/x86_asm/interpolate8x8_xmm.asm src/xvid.c


2004-07-31 09:13:23 GMT	                                	patch-64

    Summary:
      Last bit for dshow gcc support
    Revision:
      xvidcore--head--0.0--patch-64

    From pete:
     * Last bits concerning gcc build support for dshow.

    removed files:
     dshow/dxpatch/.arch-ids/DXVCSDK-9.0-gcc.patch.id
     dshow/dxpatch/.arch-ids/DXVCSDK-9.0-gcc.txt.id
     dshow/dxpatch/DXVCSDK-9.0-gcc.patch
     dshow/dxpatch/DXVCSDK-9.0-gcc.txt

    modified files:
     dshow/src/Configure.cpp


2004-07-27 21:10:02 GMT	                                	patch-63

    Summary:
      Better cross compilation support for dshow.
    Revision:
      xvidcore--head--0.0--patch-63

    From pete and ed.gomez:
     * gcc 3.4.1 is even more pedantic, ::GUID was breaking it, use
       struct _GUID instead.
     * Some uppercase/lowercase mixing in MS headers.
     * More documentation footage for the braves !
    
    NB: with all this, dshow should compile, but it is not guaranted
        to work ! There's even a patch for Configure.cpp that could be
        required... postponed to a later patch.

    modified files:
     dshow/dxpatch/dx90sdk-update-gcc.patch