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=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,fa81b07d7c3d2d7d
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2003-06-01 22:35:06 PST
Path: 
 archiver1.google.com!news1.google.com!newsfeed.stanford.edu!headwall.stanford.edu!newshub.sdsu.edu!canoe.uoregon.edu!logbridge.uoregon.edu!newsfeed.cs.utexas.edu!geraldo.cc.utexas.edu!not-for-mail
From: "Bobby D. Bryant" <bdbryant@mail.utexas.edu>
Newsgroups: comp.lang.ada
Subject: Re: Random numbers in tasks
Date: Sun, 01 Jun 2003 23:32:56 -0600
Organization: dis-
Message-ID: <pan.2003.06.02.05.32.53.764840@mail.utexas.edu>
References: <bbb8o5$2711$1@ulysses.noc.ntua.gr>
 <z5dCa.29385$s23.26499@fe07.atl2.webusenet.com>
 <pan.2003.06.01.03.42.32.191321@mail.utexas.edu>
 <bbdmsr$brr$1@ulysses.noc.ntua.gr> <3EDAC482.40604@attbi.com>
NNTP-Posting-Host: dial-81-13.ots.utexas.edu
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: geraldo.cc.utexas.edu 1054532005 29238 128.83.219.29 (2 Jun 2003
 05:33:25 GMT)
X-Complaints-To: abuse@utexas.edu
NNTP-Posting-Date: Mon, 2 Jun 2003 05:33:25 +0000 (UTC)
User-Agent: Pan/0.14.0 (I'm Being Nibbled to Death by Cats!)
Xref: archiver1.google.com comp.lang.ada:38306
Date: 2003-06-01T23:32:56-06:00
List-Id: <comp.lang.ada>

On Mon, 02 Jun 2003 03:29:33 +0000, Robert I. Eachus wrote:

> Papastefanos Serafeim wrote:
>> There is a small problem in that solution, I cannot just declare the
>> generator in the task type: There is a function in a different package
>> that is used to return a random number in a given range and that
>> function is used at other places too...
> 
> There is magic in Ada for especially this purpose.  It is in Annex C, so
> it is possible that some compilers don't support it.  Look at C.7.2 The
> Package Task_Attributes.  Here is an example and output using GNAT.
> Notice that this version uses a fixed seed for each different task.  If
> you want different seeds for every run modify the body of
> My_Float_Random.Gen_Source.  But notice that just using a call to get
> seeds based on the clock may not be sufficient, unless you put a delay
> in.

FWIW, I have a GA solver with similar requirements, and I just have the
tasks rendezvous with a 'controller' task that dispenses individual seeds
for them to initialize on.  After that they can run independently.

-- 
Bobby Bryant
Austin, Texas