This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: C Compiler benchmark: gcc 4.6.3 vs. Intel v11 and others


"willus.com" <willus.com@willus.com> writes:
>> For the math functions, this is normally more a libc feature, so you
>> might get very different results on different OS. Then again, by using
>> -ffast-math, you allow the math functions to return any random value,
>> so I can think of ways to make it even faster ;-)
>
> I use -ffast-math all the time and have always gotten virtually
> identical results to when I turn it off.  The speed difference is
> important for me.

Also, of course, "-ffast-math" is really shorthand for a bunch of more
detailed options; which are OK and which are not is often very
application-dependent.

One can easily fine-tune the result by either using the more detailed
options instead of -ffast-math, or perhaps by specifying -ffast-math and
then turning _off_ (with -fno-xxx) anything that's too risky....

-Miles

-- 
`There are more things in heaven and earth, Horatio,
 Than are dreamt of in your philosophy.'


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]