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: 103376,9c3095e7116890a3
X-Google-Attributes: gid103376,public
From: holsti@cs.Helsinki.FI (Niklas Holsti)
Subject: Re: loop and block question
Date: 1997/03/16
Message-ID: <holsti.858505278@hydra>#1/1
X-Deja-AN: 225916443
Distribution: world
References: <1997Jan29.171423.3481@news.nbi.dk>
 <2svi8g5rzr.fsf@hp755d2.eurocontrol.fr> <E4wHKC.FB5@thomsoft.com>
 <32F3D3BA.4B88@ssf.fi> <E4yCBo.FrC@world.std.com>
Organization: University of Helsinki
Newsgroups: comp.lang.ada
Date: 1997-03-16T00:00:00+00:00
List-Id: <comp.lang.ada>


In <E4yCBo.FrC@world.std.com> bobduff@world.std.com (Robert A Duff) writes:

>In article <32F3D3BA.4B88@ssf.fi>, Niklas Holsti  <Niklas.Holsti@ssf.fi> wrote:
>>I found a surprising use for statement labels when automating the
>>timing measurements of an Ada program. A debugger script is used to
>>measure the time of a certain section of a subprogram. I delimited the
>>section to be measured with statement labels, thus:

>Sounds highly unreliable.  If there are no gotos, the compiler can
>freely move code across the labels, so any timing measurements are
>highly suspect.
 ...
>Highly questionable, IMHO, unless you looked at the machine code output,
>and verified that no code was moved past those labels.

Sure, I looked at the machine code output, but thanks for bringing out
the need to do so. For my uses, the timed code between the labels is
generally one or two subprogram calls.

>- Bob

Niklas Holsti                         
    Space Systems Finland Ltd, phone +358 9 6132 8625
    Keilaranta 8, FIN-02150 Espoo, Finland