Overflow Bug: Regression from Version 4.2.1

[Tim Prince]

Tom Browder wrote:
> I just filed bug number 38773 for an arithmetic overflow for the
> following program when compiled with recent versions of gfortran
> (4.3.1, 4.3.2, trunk at revision 143193):
> 
>       program overflow
>       integer i
>       parameter(i=2**31)
>       end
> 
> The error did not occur with version 4.2.1.
> 
> -Tom
Do you mean it doesn't report the overflow, where 4.2.1 and 4.4.0 do
report it?