Stupid FORTRAN Tricks (oddest floating point error I have seen)

From: David Warren <warren_at_atmos.washington.edu>
Date: Wed, 4 Jun 1997 15:32:48 -0700 (PDT)

I got this from one of my users:
>>>>>>
David:

These are the last lines of the mm5.print.out

                129990 1
                130200 1
                130410 1
 CALCULATING RADIATION
                130620 1
                130830 1
                131040 1
                131250 1
                131460 1
                131670 1
                131880 1
                132090 1
time: command terminated abnormally.

real 35274.4
user 15826.5
sys 16351.3


In the screen appear the following message:

244 jisao4% FATAL IEEE FLOATING POINT EMULATION ERROR:
unf: "UNEXPECTED 2nd execution calls underflow"
   pc: 0xfffffc00005a2628
  ins: 0x2003fdd4
  op1: 0x59fdb036
  op2: 0x0

[1] Exit -1 mm5.deck
>>>>>>

This is an MM5 weather model compiled with kf90 and the -fpe3 option (as the cray
input files seem to have real*4's that are too small). However, this is not your
standard underflow message. So, does anyone know what it means?
OS: OSF1 jisao4.atmos.washington.edu V4.0 564 alpha
FORTRAN90: 4.10


David Warren INTERNET: warren_at_atmos.washington.edu
(206) 543-0945 Fax: (206) 543-0308
University of Washington
Dept of Atmospheric Sciences, Box 351640
Seattle, WA 98195-1640
-------------------------------------------------------------------------------
DECUS E-PUBS Library Committee representative
SeaLUG DECUS Chair
Received on Thu Jun 05 1997 - 00:41:55 NZST

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:36 NZDT