[FFmpeg-devel] [PATCH] avutil/x86/intmath: remove inline asm implementations for clip functions

[Niklas Haas]

On Tue, 03 Jun 2025 18:22:30 +0200 Andreas Rheinhardt <andreas.rheinhardt at outlook.com> wrote:
> Niklas Haas:
> > On Mon, 02 Jun 2025 15:41:33 -0300 James Almer <jamrial at gmail.com> wrote:
> >> GCC/Clang is smart enough to emit minss/maxss the same way as these functions.
> >> The only theoretical benefit was in x86_32, where x87 floats are used, but the
> >> penalty of making the clipping opaque to the compiler's scheduler plus moving
> >> values from mmx regs to xmm and back will offset any potential speedup.
> >> x86_32 builds targetting anything made in the last two decades and a half
> >> should use -msse -mfp=sse anyway.
> >
> > As mention in the another thread, x87 FPU usage causes non-bitexact results in
> > swscale. Should we at this point consider setting -mfpu=sse by default for
> > x86_32 builds?
> >
> What about CPUs without SSE?

Have been dropped from all major distros, as far as I can tell, so I'm not sure
what userbase we would be alienating by not supporting them out of the box.

If you have a valid use case for combining FFmpeg git master builds from 2025
with obsolete hardware 1999, you can just disable the flag and live with the
fact that your output will not be bitexact.

>
> - Andreas
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".