On Thu, Sep 17, 2015 at 5:58 PM, Ronald S. Bultje <rsbul...@gmail.com> wrote: > The randomize_buffer() implementation assures that "most of the time", > we'll do a good mix of wide16/wide8/hev/regular/no filters for complete > code coverage. However, this is not mathematically assured because that > would make the code either much more complex, or much less random.
http://fate.ffmpeg.org/ runs the tests often enough that any issue should be caught (with the prng seed to reproduce it), so that's probably fine. > +static void check_loopfilter() (void) > + int E[2] = { 20, 28 }, I[2] = { 10, 16 }, H[2] = { 7, 11 }, F[2] = { 1, > 1 }; const + uint8_t *buf0, *buf1; [..] + buf0 = base0 + midoff_aligned; + buf1 = base1 + midoff_aligned; Could be declared and initialized at the same time. I unfortunately don't know enough about VP9 to comment on most of the logic or numerical values chosen, but I'd say go ahead and push if nobody objects. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
