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,f009ea55b3656834
X-Google-Attributes: gid103376,public
From: bobduff@world.std.com (Robert A Duff)
Subject: Re: Q:finalization or task terminate first?
Date: 1997/03/05
Message-ID: <E6K333.r3@world.std.com>#1/1
X-Deja-AN: 223196398
References: <331CF2F4.6B79@bix.com>
Organization: The World Public Access UNIX, Brookline, MA
Newsgroups: comp.lang.ada
Date: 1997-03-05T00:00:00+00:00
List-Id: <comp.lang.ada>


In article <331CF2F4.6B79@bix.com>, Tom Moran  <tmoran@bix.com> wrote:
>Given a package body with a declaration of an object descended from
>Ada.finalization.controlled, and also a task with a select-terminate
>alternative, when the main program that with's this package reaches its
>end, which will happen first: the Finalize call or the task termination?

Task termination happens first.  Then finalization.  See 7.6.1(4).
In this case, the "master" is the environment task.

This makes sense: you can't finalize things while tasks are still
meddling with them.

See also 10.2(26) and 9.8(20), and note that finalization is
abort-deferred.  And note that all Finalize procedures are library-level
procedures.  These paragraphs ensure that there's no need to await tasks
twice for the same master (that is, once finalization starts for a given
master, all tasks dependent on that master are dead and no more will be
created).

- Bob