(j3.2006) a question on c_f_pointer call

[Aleksandar Donev]

On 08/25/10 16:31, Bill Long wrote:
> If the information needed is stored somewhere in the descriptor, then 
> it should not get lost across procedure boundaries.  [Indeed, Tobias 
> is asking that this information also be included in the C descriptor 
> of the TR.]
Yes, I understand that this is what Jim is referring to. But I am not 
convinced it is possible to carry such information consistently across 
all procedure calls. What if I changed to:

call point_to(p1,p2(11:2:-1),10)

It should still be legal. Does the descriptor that is passed for the 
dummy t (this is the target) really carry information in it that it is a 
valid "deallocation target".  When do you actually calculate that (if I 
changed 11:2:-1 to 11:3:-1 it would no longer be valid to deallocate it).

I am not saying you are wrong, and certainly not saying that there is no 
confusion as to what errors the processor must detect correctly and 
report. But I am still not convinced that c_f_pointer does anything more 
or less than can already be done using external procedures that do 
pointer assignments to make a pointer point to some target.

Thanks,
Aleks

-- 
Aleksandar Donev, Assistant Professor of Mathematics
Courant Institute of Mathematical Sciences
Office: 909 Warren Weaver Hall, New York University
E-mail: donev at courant.nyu.edu
Phone: (212) 992-7315; Fax: (212) 995-4121
Mailing address: 251 Mercer St, New York, NY 10012
Web: http://cims.nyu.edu/~donev