From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,2866100c9a2b8ce7
X-Google-Attributes: gid103376,public
From: bobduff@world.std.com (Robert A Duff)
Subject: Re: Free'ing extended types
Date: 1996/04/29
Message-ID: <Dqn9Dr.Kzy@world.std.com>#1/1
X-Deja-AN: 152111102
references: <3183AC75.335C@ehs.ericsson.se> <dewar.830716817@schonberg>
 <4xrat6vlse.fsf@lagrange.enst-bretagne.fr>
organization: The World Public Access UNIX, Brookline, MA
newsgroups: comp.lang.ada
Date: 1996-04-29T00:00:00+00:00
List-Id: <comp.lang.ada>


In article <4xrat6vlse.fsf@lagrange.enst-bretagne.fr>,
Laurent Guerby <Laurent.Guerby@enst-bretagne.fr> wrote:
>   If another Ada  compiler use different  pools for the  types in the
>example, as permitted by the  RM 13.11(17), then  the execution of the
>construct is erroneous. Then, of course, it can just work, don't work,
>crash your disk or nicely dies with Program_Error ;-).

Yes, but let's not get too alarmed, here.  If "type Acc1 is access all
T1'Class;" and "type Acc2 is access all T2'Class;", where T2 is a type
extension of T1 (both tagged, of course), then we know (from
accessibility rules) that these two are at the same nesting level, so I
can't imagine any *reason* for the implementation to choose two
different storage pools for Acc1 and Acc2.

In retrospect, I admit we should have *required* Acc1 and Acc2 to share
the same pool.  But if no compiler writers show up saying they don't,
why worry about it?  (And if they do, perhaps the ARG ought to make a
ruling on this point.)

- Bob