From: Georg Bauhaus <see.reply.to@maps.futureapps.de>
Subject: Re: GNAT's stack checking in Ubuntu 9.04 (and Shootout regex-dna)
Date: Tue, 11 Aug 2009 23:38:43 +0200
Date: 2009-08-11T23:38:44+02:00 [thread overview]
Message-ID: <4a81e4e4$0$31327$9b4e6d93@newsspool4.arcor-online.net> (raw)
In-Reply-To: <4564d943-9a6a-4112-9ae2-f379b8654669@18g2000yqa.googlegroups.com>
jonathan wrote:
> In fact all settings of Max_CPUs
> from 9 to 15 ran in 9 seconds. At Max_CPUs = 24
> it was 9.7 sec.
Thanks, Jonathan, for these comments.
Your findings make me think that 9 is "magical"
because there are 9 patterns: The algorithm is
such that it would try to arrange for 9 rendezvous
if there are 9 or more task objects available.
If this turns out to be true, I can simplify the
algorithm, getting rid of a clumsy loop that
distributes work to Max_CPUs task objects.
next prev parent reply other threads:[~2009-08-11 21:38 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-03 22:54 GNAT's stack checking in Ubuntu 9.04 (and Shootout regex-dna) Georg Bauhaus
2009-08-03 22:56 ` Georg Bauhaus
2009-08-04 7:50 ` Ludovic Brenta
2009-08-04 9:17 ` Georg Bauhaus
2009-08-04 9:58 ` Vadim Godunko
2009-08-04 10:44 ` Georg Bauhaus
2009-08-04 12:30 ` Vadim Godunko
2009-08-04 14:15 ` Georg Bauhaus
2009-08-04 22:20 ` Egil
2009-08-04 15:38 ` Robert A Duff
2009-08-04 11:59 ` Brian Drummond
2009-08-04 14:18 ` Georg Bauhaus
2009-08-09 19:13 ` Georg Bauhaus
2009-08-10 13:10 ` jonathan
2009-08-10 20:12 ` jonathan
2009-08-10 20:29 ` Ludovic Brenta
2009-08-10 23:34 ` Georg Bauhaus
2009-08-11 20:02 ` jonathan
2009-08-11 21:19 ` jonathan
2009-08-11 21:38 ` Georg Bauhaus [this message]
2009-08-11 0:27 ` Georg Bauhaus
2009-08-11 19:05 ` jonathan
2009-08-12 9:32 ` Ludovic Brenta
2009-08-12 16:37 ` jonathan
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox