On 7/2/2017 11:45 AM, Scott Dorsey wrote:
> <VAXman- @SendSpamHere.ORG> wrote:
>> In article <oj8oc5$8s5$1 at panix2.panix.com>, kludge at panix.com (Scott Dorsey) writes:
>>> On 6/30/2017 5:06 AM, GerMarsh wrote:
>>>> I know that there have been a few discussions about
>>>> enhancing/replacing DCL, but I wonder if someone will fix the old
>>>> integer overflow which has been there since the year dot...
>>>>
>>>> $ int=%x7fffffff
>>>> $ sho sym int
>>>> INT = 2147483647 Hex = 7FFFFFFF Octal = 17777777777
>>>> $ int=int+1
>>>> $ sho sym int
>>>> INT = -2147483648 Hex = 80000000 Octal = 20000000000
>>>
>>> This is not a bug, it's a feature.
>>> If you don't want it to roll over, what DO you want it to do?
>>> Halt with an error?
>>
>> Halt and Catch Fire!
>>
>> In need of real math? Use Fortran!
>
> Fortran doesn't allow me to pitch an exception on integer overflow either,
> does it? I can do it on a float, but with integers I need to check explicitly
> before every operation.
I don't think the standard did/do.
But on VMS sure.
With the code posted a minute ago:
$ for/old/warn=all intovf.for /check=over
$ link intovf
$ run intovf
2147483647 7FFFFFFF 17777777777
%SYSTEM-F-HPARITH, high performance arithmetic trap, Imask=00000001,
Fmask=00000
000, summary=40, PC=00000000000300A8, PS=0000001B
-SYSTEM-F-INTOVF, arithmetic trap, integer overflow at
PC=00000000000300A8, PS=0
000001B
%TRACE-F-TRACEBACK, symbolic stack dump follows
image module routine line rel PC abs PC
INTOVF INTOVF INTOVF 5 00000000000000A8
00000000000300A8
0 FFFFFFFF8037FC44
FFFFFFFF8037FC44
%TRACE-I-END, end of TRACE stack dump
Arne