Hi Dominique, hi all,

you are absolutely right, Dominique: I missed the part of pr60322_base_*. 

But this time it is there and furthermore does solve the allocate( mold=e) and
the loc(e) issue. 

Paul: I have simplified your patch by only checking whether the
arg_expr.ts.type == BT_CLASS. All tests showed, that this enough to produce the
correct code.

Bootstraps and regtests ok on x86_64-linux-gnu/F20. 

Comments, please!

Regards,
	Andre

On Wed, 25 Mar 2015 10:43:34 +0100
Dominique d'Humières <dominiq@lps.ens.fr> wrote:

> Hi Andre,
> 
> > Le 24 mars 2015 à 18:06, Andre Vehreschild <vehre@gmx.de> a écrit :
> > 
> > Hi all,
> > 
> > I have worked on the comments Mikael gave me. I am now checking for
> > class_pointer in the way he pointed out.
> > 
> > Furthermore did I *join the two parts* of the patch into this one, because
> > keeping both in sync was no benefit but only tedious and did not prove to be
> > reviewed faster.
> 
> Are you sure that you attached the right patch? It does not apply on a clean
> tree unless I apply the patch at
> 
> https://gcc.gnu.org/ml/fortran/2015-02/msg00105.html
> 
> with minor surgery for gcc/fortran/expr.c.
> 
> > Paul, Dominique: I have addressed the LOC issue that came up lately. Or
> > rather the patch addressed it already. I feel like this is not tested very
> > well, not the loc() call nor the sizeof() call as given in the 57305
> > second's download.
> 
> The ICE is fixed and the LOC issue seems fixed. 
> 
> > Unfortunately, is that download not runable. I would love to see a test
> > similar to that download, but couldn't come up with one, that satisfied me.
> > Given that the patch's review will last some days, I still have enough time
> > to come up with something beautiful which I will add then.
> 
> I have changed the test to
> 
> use iso_c_binding
> implicit none
> real, target :: e
> class(*), allocatable, target :: a(:)
> e = 1.0
> call add_element_poly(a,e)
> print *, size(a)
> call add_element_poly(a,e)
> print *, size(a)
> select type (a)
>   type is (real)
>     print *, a
> end select
> contains
>     subroutine add_element_poly(a,e)
>       use iso_c_binding
>       class(*),allocatable,intent(inout),target :: a(:)
>       class(*),intent(in),target :: e
>       class(*),allocatable,target :: tmp(:)
>       type(c_ptr) :: dummy
>       
>       interface
>         function memcpy(dest,src,n) bind(C,name="memcpy") result(res)
>           import
>           type(c_ptr) :: res
>           integer(c_intptr_t),value :: dest
>           integer(c_intptr_t),value :: src
>           integer(c_size_t),value :: n
>         end function
>       end interface
> 
>       if (.not.allocated(a)) then
>         allocate(a(1), source=e)
>       else
>         print *, size(a)
>         allocate(tmp(size(a)),source=a)
>         print *, size(a), size(tmp) + 1
>         print *, loc(a(1)),loc(tmp),sizeof(tmp)
>         deallocate(a)
> !        allocate(a(size(tmp)+1),mold=e)
>         allocate(a(size(tmp)+1),source=e)
>         print *, size(a), size(tmp)
>         dummy = memcpy(loc(a(1)),loc(tmp),sizeof(tmp))
>         dummy = memcpy(loc(a(size(tmp)+1)),loc(e),sizeof(e))
>       end if
>     end subroutine
> end
> 
> As pointed by Paul, I get a segfault at run time if I use the commented line,
> i.e. ‘mold’ instead of ‘source’.
> 
> > Bootstraps and regtests ok on x86_64-linux-gnu/F20.
> > 
> > Regards,
> > 	Andre
> 
> Thanks for your work.
> 
> Dominique
> 


-- 
Andre Vehreschild * Email: vehre ad gmx dot de