Andreas Rheinhardt 2570f5d307 avcodec/x86/vvc/of: Avoid scalar log2
Instead convert the integers to floats and inspect the exponent.

Old benchmarks:
apply_bdof_8_8x16_c:                                  3295.2 ( 1.00x)
apply_bdof_8_8x16_avx2:                                312.7 (10.54x)
apply_bdof_8_16x8_c:                                  3269.1 ( 1.00x)
apply_bdof_8_16x8_avx2:                                203.6 (16.05x)
apply_bdof_8_16x16_c:                                 6584.8 ( 1.00x)
apply_bdof_8_16x16_avx2:                               413.6 (15.92x)
apply_bdof_10_8x16_c:                                 3313.9 ( 1.00x)
apply_bdof_10_8x16_avx2:                               321.5 (10.31x)
apply_bdof_10_16x8_c:                                 3306.5 ( 1.00x)
apply_bdof_10_16x8_avx2:                               200.4 (16.50x)
apply_bdof_10_16x16_c:                                6659.7 ( 1.00x)
apply_bdof_10_16x16_avx2:                              402.4 (16.55x)
apply_bdof_12_8x16_c:                                 3305.7 ( 1.00x)
apply_bdof_12_8x16_avx2:                               321.8 (10.27x)
apply_bdof_12_16x8_c:                                 3258.1 ( 1.00x)
apply_bdof_12_16x8_avx2:                               198.6 (16.41x)
apply_bdof_12_16x16_c:                                6600.2 ( 1.00x)
apply_bdof_12_16x16_avx2:                              392.6 (16.81x)

New benchmarks:
apply_bdof_8_8x16_c:                                  3269.9 ( 1.00x)
apply_bdof_8_8x16_avx2:                                266.5 (12.27x)
apply_bdof_8_16x8_c:                                  3252.9 ( 1.00x)
apply_bdof_8_16x8_avx2:                                182.6 (17.81x)
apply_bdof_8_16x16_c:                                 6596.7 ( 1.00x)
apply_bdof_8_16x16_avx2:                               362.7 (18.19x)
apply_bdof_10_8x16_c:                                 3351.3 ( 1.00x)
apply_bdof_10_8x16_avx2:                               269.0 (12.46x)
apply_bdof_10_16x8_c:                                 3329.1 ( 1.00x)
apply_bdof_10_16x8_avx2:                               174.5 (19.08x)
apply_bdof_10_16x16_c:                                6654.3 ( 1.00x)
apply_bdof_10_16x16_avx2:                              357.8 (18.60x)
apply_bdof_12_8x16_c:                                 3274.1 ( 1.00x)
apply_bdof_12_8x16_avx2:                               276.0 (11.86x)
apply_bdof_12_16x8_c:                                 3263.5 ( 1.00x)
apply_bdof_12_16x8_avx2:                               176.8 (18.46x)
apply_bdof_12_16x16_c:                                6576.4 ( 1.00x)
apply_bdof_12_16x16_avx2:                              357.8 (18.38x)

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2026-03-30 13:51:53 +02:00
2026-03-14 21:36:58 +00:00
2025-08-08 21:51:15 +00:00
2026-02-22 04:39:22 +00:00
2025-06-23 14:48:40 +02:00
2025-08-03 13:48:47 +02:00
2025-05-07 15:35:47 +02:00
2025-08-14 08:42:29 -04:00

FFmpeg README

FFmpeg is a collection of libraries and tools to process multimedia content such as audio, video, subtitles and related metadata.

Libraries

  • libavcodec provides implementation of a wider range of codecs.
  • libavformat implements streaming protocols, container formats and basic I/O access.
  • libavutil includes hashers, decompressors and miscellaneous utility functions.
  • libavfilter provides means to alter decoded audio and video through a directed graph of connected filters.
  • libavdevice provides an abstraction to access capture and playback devices.
  • libswresample implements audio mixing and resampling routines.
  • libswscale implements color conversion and scaling routines.

Tools

  • ffmpeg is a command line toolbox to manipulate, convert and stream multimedia content.
  • ffplay is a minimalistic multimedia player.
  • ffprobe is a simple analysis tool to inspect multimedia content.
  • Additional small tools such as aviocat, ismindex and qt-faststart.

Documentation

The offline documentation is available in the doc/ directory.

The online documentation is available in the main website and in the wiki.

Examples

Coding examples are available in the doc/examples directory.

License

FFmpeg codebase is mainly LGPL-licensed with optional components licensed under GPL. Please refer to the LICENSE file for detailed information.

Contributing

Patches should be submitted to the ffmpeg-devel mailing list using git format-patch or git send-email. Github pull requests should be avoided because they are not part of our review process and will be ignored.

Description
No description provided
Readme 579 MiB
Languages
C 89.4%
Assembly 8.3%
Makefile 1.3%
C++ 0.3%
GLSL 0.2%
Other 0.3%