https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125902
--- Comment #8 from Nicholas BREALEY <Nicholas.BREALEY at 3ds dot com> ---
Thank you for looking at this.
(In reply to Harald Anlauf from comment #5)
> Yes, there are some issues.
>
> Note:
>
> character(kind=c_char,len=*), intent(in) :: c
>
> should not translate to const CFI_cdesc_t* c but to const char *c,
> so that one seems correct to me.
No, I do not think that is correct.
character(kind=c_char,len=*), intent(in) :: c
should translate to a const CFI_cdesc_t* c.
My test program would not have worked if it was const char *c.
Section 18.3.7 "Interoperability of procedures and procedure interfaces" in
J3/23-007r1 says:
A Fortran procedure interface is interoperable with a C function prototype if
...
(5) any dummy argument without the VALUE attribute corresponds to a formal
parameter of the prototype that is of a pointer type, and either
...
the dummy argument is a nonallocatable nonpointer variable of type
CHARACTER with
assumed character length and the formal parameter is a pointer to
CFI_cdesc_t,
(J3/18-007r1 is the same)
NB
character(kind=c_char,len=1), dimension(*), intent(in) :: c
translates to const char *c.
>
> Removing the ALLOCATABLE attribute in
>
> real(c_double), dimension(:), allocatable, intent(inout) :: d
>
> "fixes" this particular one and produces CFI_cdesc_t *d.
