From owner-cvs-all@FreeBSD.ORG Wed Jul 25 00:44:01 2007 Return-Path: Delivered-To: cvs-all@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8872216A41F; Wed, 25 Jul 2007 00:44:01 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from webaccess-cl.virtdom.com (webaccess-cl.virtdom.com [216.240.101.25]) by mx1.freebsd.org (Postfix) with ESMTP id 599B813C467; Wed, 25 Jul 2007 00:44:01 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from [192.168.1.101] (c-71-231-138-78.hsd1.or.comcast.net [71.231.138.78]) (authenticated bits=0) by webaccess-cl.virtdom.com (8.13.6/8.13.6) with ESMTP id l6P0hqW6016923 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Tue, 24 Jul 2007 20:43:53 -0400 (EDT) (envelope-from jroberson@chesapeake.net) Date: Tue, 24 Jul 2007 17:46:50 -0700 (PDT) From: Jeff Roberson X-X-Sender: jroberson@10.0.0.1 To: Jeremie Le Hen In-Reply-To: <20070724153608.GH96643@obiwan.tataz.chchile.org> Message-ID: <20070724174504.I561@10.0.0.1> References: <200707032126.l63LQ7ea027929@repoman.freebsd.org> <20070703142714.F552@10.0.0.1> <20070723201642.GC96643@obiwan.tataz.chchile.org> <46A54A0B.4000109@elischer.org> <20070724153608.GH96643@obiwan.tataz.chchile.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: cvs-src@FreeBSD.org, src-committers@FreeBSD.org, Julian Elischer , cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/kern kern_descrip.c X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Jul 2007 00:44:01 -0000 On Tue, 24 Jul 2007, Jeremie Le Hen wrote: > On Mon, Jul 23, 2007 at 05:38:35PM -0700, Julian Elischer wrote: >> I think that is the wrong question.. >> the question is "why do we drop off after 8?" >> which I'm sure Jeff is already working on. > > Actually since the workbench has been run on an 8-core amd64 and that > both Linux and FreeBSD drop here, I thought it was natural to get the > best performance with 8 threads... Am I wrong? Well there seem to be some significant periods of waittime that are purely within mysql. I do not understand these very well but I do not think that all of the dropoff after 8 threads is our fault. There is, however, some considerable contention reported. My select patch was created to address this. Attilio's lockmgr work should fix the next biggest bottleneck. After that it looks like filedescriptor locking needs some more attention since there are so many threads with a shared fd set. Jeff > > Regards, > -- > Jeremie Le Hen > < jeremie at le-hen dot org >< ttz at chchile dot org > >