Error trapping for multiprocessor systems:
A primer
last update:
04/29/2003
Error trapping for multiprocessor systems: A primer
Table of contents
Credits and copyright
Introduction to error
trapping
Overview of this
primer
Valid IEEE floating-point
numbers
Ranges of valid
numbers
Types of execution
errors
Not a Number
(NaN), a representation of a number outside the allowable IEEE
range
Quiet
NaN (NAN)
Signaling NaN (NANS)
Infinity error
(INF) representing a number outside the allowable IEEE range
Single-precision
and double-precision NaNs and INFs
Sample code
for detecting NaNs and INFs
Two ways to handle
floating-point exceptions (FPEs)
Using the
fp_trap library routine at run time
Using
the -qsigtrap and -qflttrap options in compile
and load statements
Causes of
FPEs
Underflow
(UND)
Overflow
(OV)
Divide
by zero (ZERO)
Invalid
operand (INV)
Inexact
result (INEX)
Imprecise
(IMP)
Enable
(EN)
Tracing
SIGTRAP events
The -qflttrap
option for compile and load statements (SIGFPE)
The -qsigtrap
option for compile and load statements (SIGTRAP)
The
fp_trap library routine for trapping SIGFPE events at
run time
Simple Fortran run
script that produces an overflow
Fortran code
that produces overflow FPEs
Output sent
to stdout
Output displayed
by TotalView
Next page |
Table of contents -
TotalView primer
If you have questions about this document, please contact
SCD Customer Support.
You can also reach us by telephone 24 hours a day, seven days a week at
303-497-1278.
Additional contact methods:
consult1@ucar.edu
and during
business hours
in NCAR Mesa Lab Suite 39.
© Copyright 2003. University Corporation for Atmospheric
Research (UCAR). All Rights Reserved.
Address of this page:
http://www.scd.ucar.edu/docs/trap.error/index.html