On 4/11/12 4:06 PM, Norman Clerman wrote:
> You are welcome.
>
> Yes, I agree. My best experience, by far, has been with the NAG
> compiler. (I don't have access to either the IBM or Cray compilers, and
> I haven't tried Pathscale or Sun.) But the NAG compiler is very robust.
>
The NAG compiler is well known for excellent standard-conformance.
With the Cray compiler ...
First code: compiles with no messages. (no main, so no execute).
Second code: Compiles and runs:
> ./a.out
OKAY CORRECT ONE
>
Third code: Compiles and runs:
> ./a.out
>
> [Neil wrote:]
> Ouch! I'm bewildered that so many compilers still can't get
> the Fortran 90 standard right; makes me pessimistic about
> what they're doing with 2003 and 2008 :(
>
Neither 2003 nor 2008 is simple to implement, but at least some vendors
to take the task seriously.
Cheers,
Bill
--
Bill Long [log in to unmask]
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
|