[Bug fortran/58009] New: Elements with same value in vector subscript in variable definition context

[Bug fortran/58009] New: Elements with same value in vector subscript in variable definition context

[tkoenig at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58009

            Bug ID: 58009
           Summary: Elements with same value in vector subscript in
                    variable definition context
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: enhancement
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tkoenig at gcc dot gnu.org

The following is illegal:

program main
  real, dimension(4) :: a,b
  read (*,*) a
  b([1,2,3,2]) = a
  print *,b
end

See F 2008, 6.5.3.3.2:

If a vector subscript has two or more elements with the same value, an array
section with that vector subscript
shall not appear in a variable definition context (16.6.7)

[Bug fortran/58009] Elements with same value in vector subscript in variable definition context

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58009

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-07-28
     Ever confirmed|0                           |1

--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> ... and it would be nice to diagnose it (with a hard
> error) whenever possible.

Indeed! I can think of an O(N**2) algorithm with a scalar temporary or an O(N)
one with an O(N) temporary. Do you have better ideas?

[Bug fortran/58009] Elements with same value in vector subscript in variable definition context

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58009

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |burnus at gcc dot gnu.org

--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> ---
(In reply to Thomas Koenig from comment #3)
> a([i,j,i]) = ...
> for which I cannot think of an algorithm which is O(n),
> so I guess it will have to be O(n**2).

Given that most code has either no or only very few vector subscripts, it
shouldn't matter too much.

Additionally, I expect that one either has only very few elements - or one uses
A([array]) which is not compile-time checkable.

The only case where one might have a lot of elements in the vector is in
generated code - but that looks as a very, very special case for which one does
not need to performance optimize, I'd guess.


Side note: I tried your example with NAG f95 5.1 (of 2007) - and it does not
detect the issue - neither at compile nor at runtime (-C=all).

[Bug fortran/58009] Elements with same value in vector subscript in variable definition context

[tkoenig at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58009

Thomas Koenig <tkoenig at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |tkoenig at gcc dot gnu.org

--- Comment #5 from Thomas Koenig <tkoenig at gcc dot gnu.org> ---
I have a patch.

[Bug fortran/58009] Elements with same value in vector subscript in variable definition context

[tkoenig at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58009

Thomas Koenig <tkoenig at gcc dot gnu.org> changed:

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

--- Comment #6 from Thomas Koenig <tkoenig at gcc dot gnu.org> ---
Author: tkoenig
Date: Sun Jul 28 21:10:23 2013
New Revision: 201294

URL: http://gcc.gnu.org/viewcvs?rev=201294&root=gcc&view=rev
Log:
2013-07-28  Thomas Koenig  <tkoenig@gcc.gnu.org>

    PR fortran/58009
    * expr.c (gfc_check_vardef_context):  Check for same values in
    vector expression subscripts.

2013-07-28  Thomas Koenig  <tkoenig@gcc.gnu.org>

    PR fortran/58009
    * gfortran.dg/vector_subsript_7.f90:  New test.


Added:
    trunk/gcc/testsuite/gfortran.dg/vector_subscript_7.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/expr.c
    trunk/gcc/testsuite/ChangeLog