Well, that and Solar Designer's crypt() in his password cracker, Mingo's
tcp checksumming, maybe an optimized strcpy()/memcpy() [I haven't actually
seen an implementation of that, tho], well actually it looks like anywhere
where people are looking to hand-optimize MMX has been useful.
HOWEVER - these cases would DEFINITELY not want to be using emulated
MMX. If a compiler is taught about MMX, it will most likely end up using
it in inner loops. MMX compilers should be just like a K6 or Pentium
compiler - something not to be used for distribution binaries unless making
an MMX specialized distribution.
Quoting Sylvain,
> I agree with you, but people may want to execute a Win$ binary (under Wine),
> that is poorly written and assumes an MMX is present. We have already seen
> such behaviour from certain "software" makers, to force you to buy high-end
> hardware.
Then let WINE and DOSEMU handle the problem in user-land.
David.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu