From owner-freebsd-hackers@FreeBSD.ORG Thu Jun 21 15:26:24 2007 Return-Path: X-Original-To: freebsd-hackers@freebsd.org Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 760DB16A468 for ; Thu, 21 Jun 2007 15:26:24 +0000 (UTC) (envelope-from jdp@polstra.com) Received: from rock.polstra.com (rock.polstra.com [64.119.0.113]) by mx1.freebsd.org (Postfix) with ESMTP id 55A4B13C465 for ; Thu, 21 Jun 2007 15:26:24 +0000 (UTC) (envelope-from jdp@polstra.com) Received: from [10.0.0.64] (adsl-sj-11-62.rockisland.net [64.119.11.62]) (authenticated bits=0) by rock.polstra.com (8.13.8/8.13.8) with ESMTP id l5LFQMq9097717 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 21 Jun 2007 08:26:23 -0700 (PDT) (envelope-from jdp@polstra.com) Message-ID: <467A989E.5@polstra.com> Date: Thu, 21 Jun 2007 08:26:22 -0700 From: John Polstra User-Agent: Thunderbird 1.5.0.12 (Macintosh/20070509) MIME-Version: 1.0 To: Julian Elischer References: <467787EF.9060009@elischer.org> <46797825.10900@polstra.com> <46799032.5060009@polstra.com> <467999C9.9000402@elischer.org> <4679D081.7070600@polstra.com> <4679D79E.5030200@elischer.org> In-Reply-To: <4679D79E.5030200@elischer.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Greylist: Sender succeeded SMTP AUTH authentication, not delayed by milter-greylist-3.0 (rock.polstra.com [64.119.0.113]); Thu, 21 Jun 2007 08:26:23 -0700 (PDT) Cc: freebsd-hackers@freebsd.org Subject: Re: in-kernel tcp server X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Jun 2007 15:26:24 -0000 Julian Elischer wrote: > but if you did find some old ksocket based code sitting around, > i'd love to try it in -current and work on the bottlenecks.. I'm sure I don't have it any more, unfortunately. It was six years old, and I just moved into a smaller house and threw out a half dozen old computers as well as my ancient backup tapes. > I'll certainly look at what I can do about the queue items. > I may make a per-cpu cache of them. That would probably help a lot. Each webserver or webclient is tied to one network interface, and I get the best performance when there is one CPU core per interface. I'm not using CPU affinity yet, but I'll probably put that in before long. John