(j3.2006) "Contiguous" doesn't quite do what I need

[dick.hendrickson at att.net]

-------------- Original message ----------------------
From: Dan Nagle <dannagle at verizon.net>
>
> Hello,
> 
> Van Snyder wrote:
> > On Fri, 2007-03-23 at 13:01 -0700, Van Snyder wrote:
> >> is 2x2, and I do an operation on (1:3,:) or (1:2,2:2,:), any benefit 
> > -------------------------------------------------->1:2
> 
> May I take it that this clinches the case
> that the proposed syntax is error prone?
> 
No, it does clinch the fact that the "new" F90 array syntax is error
prone, but it's a little late to change that ;).

Van wasn't talking about a dimension statement here, he was talking about
doing an operation on a slice of a previously dimensioned array.

More seriously, isn't the CONTIGUOUS attribute also error prone?  It
took us several tries to get it as right as we now have it in the
standard.  That means that contiguous has ramifications that don't
immediatly come to mind.  That means that non-experts will be surprised
when they use it.  And it's one of those odd errors that is highly
processor dependent and goes away when you add a print statement inside
a loop and "disable" the optimization that uses an erronous contiguous
declaration.

Dick Hendrickson
 


> -- 
> 
> Dan Nagle
> Purple Sage Computing Solutions, Inc.
> _______________________________________________
> J3 mailing list
> J3 at j3.scs.gmu.edu
> http://j3.scs.gmu.edu/mailman/listinfo/j3