Re: Help with Atomic_Components and whole array assignment

[Robert Dewar <robert_dewar@my-deja.com>]

In article <95ctln$ff437$2@ID-25716.news.dfncis.de>,
  "Nick Roberts" <nickroberts@callnetuk.com> wrote:
> If it's any consolation Tuck

I doubt Tuck needs consolation at all, let alone from Nick :-)
:-)

> my interpretation of Volatile (nothing to do
> with Atomic) would always have been (and still is) that:
>
>    for i in A'Range loop
>       A(i) := 0;
>    end loop;
>
> where Volatile (or Volatile_Components) applied to A, would
> generate A'Length separate copies into memory, in whatever
> machine code it was.

Fine, but you cannot just state "your interpretation" without
backing it up with evidence from the RM.

> My interpretation of Atomic would be simply to avoid
> generating code that might bugger up multi-task access to the
> object Atomicked

But as you surely know, all Atomic objects are also Volatile,
so I don't see your point

> I propose a representation attribute, Storage_Operations

<<proposal snipped>>

This seems ill-conceived and far too complex. I don't see what
it would buy in terms of portable code. I think the paragraph
of implementation advice I suggested would be adequate to
achieve everything this more complex proposal does in practice.


Sent via Deja.com
http://www.deja.com/