Re: [gfortran] patch to implement malloc and free intrinsics

[Brooks Moses <bmoses at stanford dot edu>]

FX Coudert wrote:
> This is the latest version of this patch. Regtested on i686-linux. 
> Approve if you think reasonnable, or propose another patch, but I won't 
> spend any more time on that. However, I'd like to stress that 
> implementing MALLOC and FREE is probably something we want done before 
> 4.1 branches, since most codes using Cray pointers use them.
[...]
> +@item @emph{Example}:
> +The following example demonstrates the use of @code{MALLOC} and
> +@code{FREE} with Cray pointers. This example is intended to run on
> +32-bit systems, where the default integer kind is suitable to store
> +pointers; on 64-bit systems, ptr_x would need to be declared as
> +@code{integer(kind=8)}.
> +
> +This example is not currently functionnal, as Cray pointer support is
> +still lacking.

I believe that this last sentence (in the MALLOC documentation) is now 
erroneous, yes?

I remember having spent some time critiquing the documentation when this 
was first proposed, and it all still looks good to me.  Regrettably, I 
don't have the knowledge to approve the actual code, though.

- Brooks