Since last iteration: - PACKUSWD macro to support both sse4 and ssse3; - Instanciate SSE4 functions for WP; - Various side cleanups.
The first 3 patches I think don't have any caveat. However, the last one shows that we have issues in the current code: instead of having actual instances for every combination of operations and widths, more proxy functions could be used. It's a mind-numbing task which I don't think I'm up for. Looking at the object size, there might even be a speed benefit, though. Christophe Gisquet (5): x86: hevc_mc: split differently calls x86: hevc_mc: correct unneeded use of SSE4 code x86: hevc_mc: make WP proxy functions static x86: hevc_mc: convert to ssse3 x86: hevc_mc: reimplement SSE4 versions for WP libavcodec/x86/hevc_mc.asm | 98 ++++-- libavcodec/x86/hevcdsp.h | 54 ++-- libavcodec/x86/hevcdsp_init.c | 717 +++++++++++++++++++++++++++--------------- 3 files changed, 571 insertions(+), 298 deletions(-) -- 1.9.2.msysgit.0 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel