This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: GCC 3.3.5: -march=i586 does not use all pentium FPU instructions
On Fri, Feb 11, 2005 at 05:14:52PM -0000, Dave Korn wrote:
> > -----Original Message-----
> > From: 'Jonathan Wakely'
> > Sent: 11 February 2005 17:06
>
> > Debating the difference between Pentium, pentium, i586 etc.
> > is off-topic
> > and that is what I was drawing attention to - GCC doesn't
> > distinguish so
> > the mail I replied to was unhelpful in making the distinction.
>
> Apart from the fact that it is a unique enough occasion to have seen Sam post
> something technically accurate that it would be worthy of comment even if it was
> discussing butter-churning techniques, I think that if we refer back to the OP's
> question:
>
> > I was wondering why the above gcc parameter does not enable the use of the
> > fst/fld opcodes for pentium processors, while -march=i686 does. The Intel
> > manuals specifically say that they can be used across all pentium processors.
>
> it is entirely appropriate to point out that -march=i586 tunes the output to be
> suitable for a larger group of processors than just the intel ones. Therefore
> there may be additional constraints that the OP could not have deduced from
> reading the specs for the Intel manuals.
True - I stand corrected. Maybe I reacted too quickly assuming the
butter recipe was nonsense ;)
> Didn't the Cyrix series always have pretty poor FP support compared to the
> others?
Thanks for the correction and extra info,
jon
--
"Anybody who hates dogs and loves whiskey can't be all bad."
- W.C. Fields