Geek Answers Handbook

Why does not a runtime error occur when clearly writing the boundaries of an array?

I have a program that assigns an array out of bounds, and I expected a runtime error to be selected. However, no error occurs at all, and the program proceeds to write to undeclared memory. Is there a compiler option to protect against this? With the memory dump shown, it is clear that exceeding boundaries is real. Is there a way to declare variables or argument specifications to catch this? Obviously, this is a clear case, but when it is tasked with supporting thousands of lines of derived F77 code, it is not always clear (to me) if this could happen.

PROGRAM TEST_CODE
IMPLICIT NONE

INTEGER*4 :: R(5)           ! Array of 5

    CALL R_TEST(R, 10)

END PROGRAM

SUBROUTINE R_TEST(R, J)
IMPLICIT NONE

INTEGER*4, INTENT(INOUT) :: R(1)    ! Dummy is array of 1
INTEGER*4, INTENT(IN) :: J
INTEGER*4 :: K

DO K=J-5,J+5            ! K=5..15
    R(K) = K            ! No Runtime Error
END DO

END SUBROUTINE

The compiler is Intel Fortran 2011 XE, and yes, I use the byte spec INTEGER*4because I know what I get with it.

. Compiler options

Memory of <code> R </code> variable

+5
2

Intel . (. ) - :

$ifort -O0 -debug -traceback -check -ftrapuv TEST_CODE.f90

. Intel . F77 .. . , Fortran - ; , ! .

R (:) , , ; . ( F2008).

PROGRAM TEST_CODE
IMPLICIT NONE

INTEGER*4 :: R(5)           ! Array of 5

    CALL R_TEST(R, 10)
    print *,R

END PROGRAM

SUBROUTINE R_TEST(R, J)
IMPLICIT NONE

INTEGER*4, INTENT(INOUT) :: R(:)    ! Dummy is array of 1
INTEGER*4, INTENT(IN) :: J
INTEGER*4 :: K

DO K=J-5,J+5            ! K=5..15
    R(K) = K            ! No Runtime Error
END DO

END SUBROUTINE
+4

. gfortran 4.6 :

At line 18 of file test_code.f90
Fortran runtime error: Index '5' of dimension 1 of array 'r' above upper bound of 1

ifort XE 12.1.1.246 - .

EDIT: Intel: " , , , * 1." R (2), ifort .

, "1" , . , , , , , . . Fortran 90 , ( ).

, " ( ), ", .. ifort - , (1) (*), .

+3

More articles:


All Articles