(j3.2006) parameterized derived types
Bill Long
longb
Thu Dec 10 18:42:23 EST 2015
I appreciate that users ask for ?slower, more accurate? versions of functions for various reasons. If they later complain that then are getting different answers, then the response is ?that?s what you asked for, wasn?t it?? This seems like something the standard should not get involved with. Maybe just a documentation notation reminding readers that the ?slower, more accurate? functions will return slightly different answer from the default functions. If they didn?t, there would be no point to the ?slower? bit.
Cheers,
Bill
On Dec 9, 2015, at 8:22 PM, Robert Corbett <robert.corbett at oracle.com> wrote:
> An additional problem I have with my implementation is that a user can choose to
> link with a variety of math libraries. For one of the platforms supported by
> Oracle Solaris Studio Fortran provides a general-purpose math library (which is
> the default math library), a fast math library, and a high accuracy math
> library. The compiler uses the general-purpose math library to evaluate
> intrinsic functions in constant expressions. Functions in the other two
> libraries produce different results for the same actual argument values in some
> cases. Over time, the functions in the libraries tend to change, which could be
> a problem for programs that dynamically link with the math libraries.
>
> I could defer evaluation of specification expressions that reference the
> intrinsic transcendental functions until run-time, but that will preclude static
> allocation in some cases.
>
> Robert Corbett
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3
Bill Long longb at cray.com
Fortran Technical Support & voice: 651-605-9024
Bioinformatics Software Development fax: 651-605-9142
Cray Inc./ Cray Plaza, Suite 210/ 380 Jackson St./ St. Paul, MN 55101
More information about the J3
mailing list