[Patch, fortran] PR31213 , PR33888 and PR33998 - formal actual mapping problems

[Paul Richard Thomas]

Tobias,
>
>
>
> I get no failures for the Polyhedron benchmark (compile & run) nor for
> CP2K (compilation only).
>
That's great - thanks to you and Dominique.

If the character length of 'sv1' is set to a constant, does the error
in mapping_2.f90 go away?  Or does it remain if the length of
'my_string' is made constant.  If yes to both of these, then we have
uncovered another bug.  If no, this patch is at fault.

This departure from the original testcase came about because I made
mapping_2.f90 much more complicated at one stage, to test the patch.

Cheers

Paul


-- 
The knack of flying is learning how to throw yourself at the ground and miss.
       --Hitchhikers Guide to the Galaxy