Re: Threadsafe singletons

From:

"Matthias Hofmann" <hofmann@anvil-soft.com>

Newsgroups:

comp.lang.c++.moderated,comp.programming.threads

Date:

31 Jul 2006 11:56:41 -0400

Message-ID:

<44ce1ead$0$18504$9b4e6d93@newsread2.arcor-online.net>

"David Barrett-Lennard" <davidbl@iinet.net.au> schrieb im Newsbeitrag
news:1154354445.414520.259340@75g2000cwc.googlegroups.com...

Matthias Hofmann wrote:

"David Barrett-Lennard" <davidbl@iinet.net.au> schrieb im Newsbeitrag
news:1154313829.569314.297530@p79g2000cwp.googlegroups.com...

It is assumed that no additional threads are created until after

main()

begins. Therefore before main() only one thread can call
GetInstance(). Even if other static initialization code causes
GetInstance() to be called there is no threading issue.

You mean *dynamic* initialization code, don't you?

Are these precise formal terms? I'm referring to the code that runs
when constructors of static objects execute before main() begins.

The terms "static initialization" and "dynamic initialization" are defined
in section 3.6.2/1 of the standard.

Furthermore
the lazy creation within GetInstance() ensures that the MySingleton
object is properly constructed before it is first used.

After main() is called, threads may be created that call
GetInstance().

Yes, *after* main() has been called. But what about calling
MySingleton::GetInstance() *before* main() is called? There is no

guarantee

that InitMySingleton's constructor will be called before that happens.

So what? The program will behave correctly. There is no racing
condition. MySingleton is created before it is first used
(irrespective of when GetInstance() is first called before main()
begins), as required.

My mistake. You are right.

--
Matthias Hofmann
Anvil-Soft, CEO
http://www.anvil-soft.com - The Creators of Toilet Tycoon
http://www.anvil-soft.de - Die Macher des Klomanagers

[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]