| Integer overflow not reported [message #94494] |
Tue, 13 June 2017 07:58  |
Markus Schmassmann
Messages: 129
Registered: April 2016
|
Senior Member |
|
|
Hello,
Why are integer overflows not reported, although according the
documentation of CHECK_MATH [1] they should be detected?
Anything I can do to activate such warnings?
(other than changing to Windows)
An example is below.
-- Markus
IDL> rsun=696000000
IDL> help, rsun, rsun^2, long64(rsun)^2, long64(rsun)^2 mod 2ll^31
RSUN LONG = 696000000
<Expression> LONG = 327417856
<Expression> LONG64 = 484416000000000000
<Expression> LONG64 = 327417856
IDL> help, 1/0
<Expression> INT = 1
% Program caused arithmetic error: Integer divide by 0
% Detected at $MAIN$
IDL> !version
{
"ARCH": "x86_64",
"OS": "linux",
"OS_FAMILY": "unix",
"OS_NAME": "linux",
"RELEASE": "8.5",
"BUILD_DATE": "Jul 7 2015",
"MEMORY_BITS": 64,
"FILE_OFFSET_BITS": 64
}
[1] http://www.harrisgeospatial.com/docs/CHECK_MATH.html
|
|
|
|
| Re: Integer overflow not reported [message #94496 is a reply to message #94494] |
Tue, 13 June 2017 10:17  |
wlandsman
Messages: 743
Registered: June 2000
|
Senior Member |
|
|
On Tuesday, June 13, 2017 at 10:58:25 AM UTC-4, Markus Schmassmann wrote:
> Hello,
>
> Why are integer overflows not reported, although according the
> documentation of CHECK_MATH [1] they should be detected?
The documentation on math errors says that integer overflow is not detected on all hardware.
I agree that nondetection of the overflow of long (32bit) integers is more insidious than nondetection of the overflow of short integers -- since we are now trained to always default to long integers. ---Wayne
http://www.harrisgeospatial.com/docs/Math_Errors.html
The detection of math errors, such as division by zero, overflow, and attempting to take the logarithm of a negative number, is hardware and operating system dependent. Some systems trap more errors than other systems. On systems that implement the IEEE floating-point standard, IDL substitutes the special floating-point values NaN and Infinity when it detects a floating point math error. (See Special Floating-Point Values.) Integer overflow and underflow is not detected. Integer divide by zero is detected on all platforms.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
