(j3.2006) (SC22WG5.3997) New summary of coarrays

[Bill Long]

John Reid wrote:
> WG5,
> 
> I have made a new version of my coarray summary, following the changes 
> made last month. Any comments before I put it on the server?
> 
> The changes are here:
> 
> page 4, para 2, second half.
> 
> page 8, para 3.
> 
> page 9, sentence 1 and footnote.
> 
> 11 title
> 
> page 14, para 3 (For this ...)
> 
> page 15, para 2.
> 
> page 16, last para.
> 
> page 17, para 2.
> 
> 12.4, para 2.


The penultimate sentence says that an actual argument with a lock 
component can only correspond to a dummy with intent(inout).  This is 
overly restrictive.  It is quite reasonable to pass a structure to a 
procedure with the dummy specified as intent(in), for example. Such a 
procedure might reference only other parts of the structure and never 
touch the lock. We don't want to disallow the advantages of intent(in) 
for a case like this. Instead of "must have intent(inout)", you want 
"must not have intent(out)".

Cheers,
Bill



> 
> 12.5, sentence 1 and much of the text after Fig 9
> 
> 12.7, para 2
> 
> 15.3
> 
> page 28, items 7,8.
> 
> Thanks,
> 
> John.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> J3 mailing list
> J3 at j3-fortran.org
> http://j3-fortran.org/mailman/listinfo/j3

-- 
Bill Long                                   longb at cray.com
Fortran Technical Support    &              voice: 651-605-9024
Bioinformatics Software Development         fax:   651-605-9142
Cray Inc., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120