On Thu, Nov 26, 2015 at 6:52 AM, Derek Buitenhuis <derek.buitenh...@gmail.com> wrote: > On 11/26/2015 7:49 AM, Hendrik Leppkes wrote: >> I'm really not a fan of this change. configure is slow and complex >> enough as it is, and host tools don't need to be micro-optimized, >> since they are never run by a user. > > I agree. It's opening an endless can complexity for essentially no gain.
How does it open an "endless can of complexity"? There are not many host funcs needed. Gain of a couple of 0.1 ms startup across various codecs is not minimal. In that case, why even bother keeping the "hardcoded tables" option? The complexity of that and associated ifdefry (note: it is being done at every tablegen file, not just localized to a avutil/host_libm) is definitely at least as great as that of doing this change. > > We went down this path 2+ years ago (in Libav), and in the end it was not > done. > > - Derek > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
