[FFmpeg-devel] [PATCH 0/3] showcqt x86 optimization using intrinsic

Henrik Gramner henrik at gramner.com
Thu Mar 10 19:39:13 CET 2016


On Thu, Mar 10, 2016 at 12:01 PM, Ismail Donmez <ismail at i10z.com> wrote:
> On Thu, Mar 10, 2016 at 12:04 PM, wm4 <nfxjfg at googlemail.com> wrote:
>> We generally don't accept intrinsic in ffmpeg.
>
> Given this policy has roots from gcc 2.x times, it might be a good
> idea to discuss it again in the context of gcc5 and clang 3.8 and
> later.

I don't really see any advantage of adding intrinsics to FFMpeg so I'm
against it.

Standardizing the x86 assembly in one "language" is preferable for
multiple reasons, especially since yasm/nasm + x86inc is more
powerful/flexible (better macro support etc.) than using intrinsics.

If you know how to write intrinsics it shouldn't be that much of a
learning curve to become familiar with yasm/nasm + x86inc either, and
you can always drop by on IRC to ask questions.


More information about the ffmpeg-devel mailing list