[FFmpeg-devel] [PATCH] swscale/output: Altivec-optimize yuv2plane1_8

[Lauri Kasanen]

On Sat, 17 Nov 2018 15:20:08 +0100
Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:

> 2018-11-17 9:09 GMT+01:00, Lauri Kasanen <cand at gmx.com>:
> > Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
> >> (This is less important atm, but I believe all functions currently
> >> in libswscale/ppc compile and run fine on - old - 32bit be hardware
> >> as your new function does.
> >> My completely inexperienced suspicion is that the instruction that
> >> you call "VSX" also exists on Altivec.)
> >
> > Ref
> > http://gcc.gnu.org/onlinedocs/gcc/PowerPC-AltiVec-Built-in-Functions-Available-on-ISA-2_002e06.html#PowerPC-AltiVec-Built-in-Functions-Available-on-ISA-2_002e06
> >
> > VSX functions such as vec_vsx_ld were added in ISA 2.06, aka POWER7.
> 
> The instruction vec_vsx_ld is currently only used for little-endian ppc
> which I thought did not exist before power7, am I wrong?

Looks like there were LE Powers like the 440 already in 1999:
https://lwn.net/Articles/408051/
datasheets.chipdb.org/IBM/PowerPC/440/PowerPC-440-Core.pdf

- Lauri