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: Questions about the ATTRS field of builtins.def


On Thu, 31 Jul 2003, Kaveh R. Ghazi wrote:
> Yes, I'd guess that the place where this fourth ATTR case would be
> appropriate is for rounding functions that ignore the mode and always
> round a particular way. (?)  Anywhere else?

I'm guessing that fmod and drem that specify the rounding mode to be
used in the calculation might fall under this category.  Probably best
to assume they're both ATTR_MATHFN_FPROUNDING_ERRNO until a real IEEE
expert can cast an authorative opinion.  Are fmod and drem independent
of the current FP rounding mode?

Roger
--


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