[FFmpeg-devel] [PATCH v3 0/7] sse4 flac lpc encoder

James Darnley james.darnley at gmail.com
Fri Feb 14 12:40:05 CET 2014


Patch 1:
    Made the changes discussed in the previous thread: now using inc/dec,
    removed the compare, and using one register less on x86_64.

Patch 2:
    No change.

Patch 3:
    No change.

Patch 4:
    No change.

Patch 5:
    No change.

Patch 6:
    Incorporated the changes made to patch 1 and corrected the number of xmm
    registers used.  Also added the changes discussed last time to the MIN/MAX
    macros to allow 3-operand instructions.

Patch 7:
    No change.

Does anyone have any final comments?  Spelling?  Grammar?  Layout?  Cosmetics?
Gross objections?  Otherwise I consider these to be finished and ready to be
committed.

James Darnley (7):
  lavc/flacenc: add sse4 version of the lpc encoder
  lavc/flacenc: partially unroll loop in flac_enc_lpc_16_sse4
  lavc/flacenc: increase buffer sizes in FlacSubFrame struct
  cosmetic alignment
  flacdsp_lpc_template: add comment to explain the CONFIG_SMALL code
  lavc/flacenc: add sse4 version of the 32-bit lpc encoder
  lavc/flacenc: partially unroll loop in flac_enc_lpc_32

 LICENSE                           |    1 +
 libavcodec/flacdsp_lpc_template.c |   18 +++
 libavcodec/flacenc.c              |    2 +-
 libavcodec/x86/Makefile           |    3 +
 libavcodec/x86/flac_dsp_gpl.asm   |  211 +++++++++++++++++++++++++++++++++++++
 libavcodec/x86/flacdsp_init.c     |    9 ++
 6 files changed, 243 insertions(+), 1 deletions(-)
 create mode 100644 libavcodec/x86/flac_dsp_gpl.asm

-- 
1.7.9



More information about the ffmpeg-devel mailing list