[Bug fortran/36754] Compile-time bound-checking for allocatable arrays with known bonds

kargl at gcc dot gnu dot org gcc-bugzilla@gcc.gnu.org
Fri Apr 16 17:01:00 GMT 2010



------- Comment #11 from kargl at gcc dot gnu dot org  2010-04-16 17:00 -------
(In reply to comment #7)
> > I get an error at compile time with gfortran 4.2.3, 4.3.3, 4.4.0, and trunk
> > (intel-darwin9). Is this PR valid?
> 
> Good question - I get now the same error for comment 0, which makes the report
> INVALID<already implemented since a long time>.
> 
> I don't get an error for comment 4. I think it is in principle detectable, but
> difficult -> WONTFIX?
> 
> At run time, I get for comment 4 "(0/11)" instead of the proper "(1/12)", but I
> think the off-by-one bug is tracked elsewhere.
> 

The off-by-one bug is fixed.

laptop:kargl[286] gfc4x -o z -fcheck=bounds g.f90
laptop:kargl[287] ./z
At line 4 of file g.f90
Fortran runtime error: Array bound mismatch, size mismatch for dimension 1 of
array 'b' (1/12)

Can we close this PR?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36754



More information about the Gcc-bugs mailing list