[Bug fortran/44604] New: Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[Bug fortran/44604] New: Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[burnus at gcc dot gnu dot org]

Reported by Michael Briggs at
http://gcc.gnu.org/ml/fortran/2010-06/msg00205.html

Checking fails for a zero-initialized pointers/allocatable scalar
variable/array element passed to a dummy argument with VALUE attribute.

For the following program, it fails for -fcheck=all:

program TestSmall
  implicit none
 integer, allocatable :: ptrVar
  allocate ( ptrVar )
  ptrVar = 0  ! if changed to 1, the problem goes away
  call test(ptrVar)
contains
  subroutine test(n)
     integer, VALUE :: n
  end subroutine test
end program TestSmall


-- 
           Summary: Wrong run-time checks with VALUE dummies and
                    pointer/allocatable actuals
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Keywords: diagnostic
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: burnus at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[burnus at gcc dot gnu dot org]

------- Comment #1 from burnus at gcc dot gnu dot org  2010-06-21 07:13 -------
Seemingly the allocated/associated check

              cond = fold_build2 (EQ_EXPR, boolean_type_node, parmse.expr,
                                  fold_convert (TREE_TYPE (parmse.expr),
                                                null_pointer_node));
in trans-expr.c's gfc_conv_procedure_call accesses the value of the variable
and not its location as parmse.expr is already dereferenced via gfc_conv_expr,
called by

              else if (fsym && fsym->attr.value)
                {
       //       [...]
                    gfc_conv_expr (&parmse, e);
                }


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[burnus at gcc dot gnu dot org]

------- Comment #2 from burnus at gcc dot gnu dot org  2010-06-21 07:24 -------
Reminder: Same problem occurs when calling with %VAL(actual_arg).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[dominiq at lps dot ens dot fr]

------- Comment #3 from dominiq at lps dot ens dot fr  2010-06-21 08:55 -------
Is it not a duplicate of pr35203?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[burnus at gcc dot gnu dot org]

------- Comment #4 from burnus at gcc dot gnu dot org  2010-06-21 09:13 -------
(In reply to comment #3)
> Is it not a duplicate of pr35203?

No. In this bug (PR 44604) one is in the caller and there one can distinguish
between address and value of a variable and the fix is relatively simple
(though requires some thinking and re-shoveling).

By contrast, PR 35203 requires a special handling of OPTIONAL + VALUE dummies,
where both the caller and the callee need to be involved by passing additional
information and making use of them.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|---                         |FIXED

--- Comment #7 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Works for me also for 4.9.2 and trunk (5.0 r218434). No answer for almost a
year and a half. Closing as FIXED.

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING

--- Comment #6 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Waiting.

[Bug fortran/44604] Wrong run-time checks with VALUE dummies and pointer/allocatable actuals

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44604

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-06-11
     Ever confirmed|0                           |1

--- Comment #5 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Works for me on 4.6.4, 4.7.3, 4.8.1, and trunk. Confirmed for 4.5.4. Apparently
fixed between revisions 179850 and 179999. This PR could probably be closed as
fixed.