Re: sse{2,3,4.2}, altivec, neon, ...
On Wed, Aug 09, 2017 at 11:56:32PM +0200, Adam Borowski wrote:
>...
> Thus, it'd be nice to have a way to automatically detect such cases, but I
> don't know of a anything better than testing manually. Which is extra unfun
> as any pre-sse2 (or pre-sse3) machine is crummy to the extreme, making
> unaccelerated qemu significantly faster than real silicon. And neither is
> fast enough to not make such testing a pain in the posterior.
You cannot reliably do this, e.g. the bad instructions might be used
only in some part of the program/library that the testing doesn't hit.
One nice part about including it in lintian is that this gives the
maintainer a standard way to indicate that the package had been
confirmed to only select such instructions after autodetection of
the cpu features by adding a lintian override.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
Reply to:
- References:
- sse{2,3,4.2}, altivec, neon, ...
- From: Adam Borowski <kilobyte@angband.pl>
- Re: sse{2,3,4.2}, altivec, neon, ...
- From: Adrian Bunk <bunk@debian.org>
- Re: sse{2,3,4.2}, altivec, neon, ...
- From: "A. Maitland Bottoms" <bottoms@debian.org>
- Re: sse{2,3,4.2}, altivec, neon, ...
- From: Adam Borowski <kilobyte@angband.pl>