comp.lang.ada
 help / color / mirror / Atom feed
From: Wes Groleau <wesgroleau@despammed.com>
Subject: Re: Common sense...
Date: Fri, 05 Jul 2002 16:34:48 -0500
Date: 2002-07-05T16:34:48-05:00	[thread overview]
Message-ID: <3D2610F8.F2756A89@despammed.com> (raw)
In-Reply-To: w7lV8.2183$7i3.136606@newsfep1-win.server.ntli.net


> > I'd say the iterators should be coded to prevent
> > more than one being on a particular node at any time.
> >
> > But then you'd have hassles with concurrency.
> 
> There are hassles with concurrency in allowing more than one iterator to use
> the same node.  What if the only iterator holding a node releases it at the

But you have even more hassles when you prevent it.

You have to ensure that the task holding the node
can't get blocked by the task waiting for the node.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau



  reply	other threads:[~2002-07-05 21:34 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-05 17:19 Common sense chris.danx
2002-07-05 17:28 ` Wes Groleau
2002-07-05 17:58   ` chris.danx
2002-07-05 21:34     ` Wes Groleau [this message]
2002-07-05 18:31 ` Jeffrey Carter
2002-07-05 18:54   ` chris.danx
2002-07-06  1:26     ` Jeffrey Carter
2002-07-06 10:28       ` chris.danx
2002-07-06 20:07         ` Jeffrey Carter
2002-07-06  6:52   ` Simon Wright
2002-07-05 20:16 ` chris.danx
2002-07-06  6:51 ` Simon Wright
2002-07-12 18:01   ` chris.danx
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox