[Patch, fortran] PR57522 - [F03] ASSOCIATE construct creates array descriptor with incorrect stride for derived type array component

[Richard Biener]

On February 8, 2014 5:08:52 PM GMT+01:00, Paul Richard Thomas <paul.richard.thomas@gmail.com> wrote:
>Dear All,
>
>I am aware that we are in stage 4 but this wrong-code-on-valid PR is
>confined to a corner of a corner and so I am certain that it can be
>safely applied - OK, Richie?

Ok with me.

Richard.

>This patch follows the same route as has been used for pointer array
>assignment to components of derived type arrays by rolling out the
>auxilliary 'span' variable for the associate-name to give it the
>correct extent.  Once the array descriptor reform has been completed,
>this fix should be removed. All such code can be found by grepping on
>'subref_array', so the excision will be swift and painless :-)
>
>Note that this is not needed for SELECT_TYPE since it is inaccessible
>there: "component to the right of an array reference may not have
>ALLOCATABLE or POINTER attribute"  && "class component must be
>ALLOCATABLE or POINTER" = false for any selector that I have been able
>to construct.
>
>Bootstraps and regtests on FC17/x86_64 - OK for trunk and 4.8?
>
>Cheers
>
>Paul
>
>2014-02-08  Paul Thomas  <pault@gcc.gnu.org>
>
>    PR fortran/57522
>    * resolve.c (resolve_assoc_var): Set the subref_array_pointer
>    attribute for the 'associate-name' if necessary.
>    * trans-stmt.c (trans_associate_var): If the 'associate-name'
>    is a subref_array_pointer, assign the element size of the
>    associate variable to 'span'.
>
>2014-02-08  Paul Thomas  <pault@gcc.gnu.org>
>
>    PR fortran/57522
>    * gfortran.dg/associated_target_5.f03 : New test