Mail Archives: cygwin/2010/03/21/01:46:04
scott:
>>> According to the internets... the C3 purports to be a i686-class
>>> processor
>> Without CMOV or out-of-order execution, that seems a bit of a scam.
> [snip]
>> For gcc, "i686" implies the presence of the CMOV instruction, because,
>> well, CMOV was introduced with the Intel 686 (aka Pentium Pro).
>
> What's in a name? GCC can define it as march=hamburger but that doesn't
> make it so, it's just a name of convenience/convention.
Good luck with trying to get GCC to change their definition of "i686".
> I assume
> Intel's specs/contracts define what is/can be called an i686, so if they
> say it's optional then the C3 is still a legit i686 even if CMOV support
> is more common.
Where do you get that idea from that CMOV is optional? Yes, there's a
CPUID feature bit representing CMOV, but that's always set on the
i686, its descendants, and compatible processors. Pre-Nehemiah C3s are
not fully 686-compatible, simple as that.
>> Btw, that was almost fifteen years ago.
>
> Well, if ya wanna make a fanless i686 (ish) CPU that runs on five (5)
> watts or less, you gotta leave a few things out.
Maybe so, but you can't then go out and demand that everyone else
doesn't use those features.
> This just means my legacy app is stuck on a 60watt beigebox instead
C3-Nehemiah? C7? Atom?
Andy
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
- Raw text -