From owner-freebsd-questions@FreeBSD.ORG Thu May 28 15:17:42 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id BA1C8106566B for ; Thu, 28 May 2009 15:17:42 +0000 (UTC) (envelope-from cpghost@cordula.ws) Received: from fw.farid-hajji.net (fw.farid-hajji.net [213.146.115.42]) by mx1.freebsd.org (Postfix) with ESMTP id 4BCC98FC21 for ; Thu, 28 May 2009 15:17:42 +0000 (UTC) (envelope-from cpghost@cordula.ws) Received: from phenom.cordula.ws (phenom [192.168.254.60]) by fw.farid-hajji.net (Postfix) with ESMTP id C7B5E34D84; Thu, 28 May 2009 17:17:38 +0200 (CEST) Date: Thu, 28 May 2009 17:17:38 +0200 From: cpghost To: Wojciech Puchar Message-ID: <20090528151738.GC1259@phenom.cordula.ws> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.19 (2009-01-05) Cc: freebsd-questions@freebsd.org Subject: Re: superpages? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 May 2009 15:17:43 -0000 On Thu, May 28, 2009 at 02:50:16PM +0200, Wojciech Puchar wrote: > maybe not new news but i just found this: > > http://www.h-online.com/open/FreeBSD-7-2-released-now-with-Superpages--/news/113204 > > It says about pages 4KB and 4MB and that it's done > automatically. > > Two questions: > > 1) is it on all architectures including amd64? As amd64 supports 4KB, 2MB > and 1GB pages it sounds inconsistent with the above. > > 2) how does this "automatic" selection work. By just having program with > large continous data space (like squid proxy) will it put that data on 2MB > pages. The following excerpt from: http://www.freebsd.org/releases/7.2R/relnotes-detailed.html may be helpful: [amd64, i386] The FreeBSD virtual memory subsystem now supports fully transparent use of superpages for application memory; application memory pages are dynamically promoted to or demoted from superpages without any modification to application code. This change offers the benefit of large page sizes such as improved virtual memory efficiency and reduced TLB (translation lookaside buffer) misses without downsides like application changes and virtual memory inflexibility. This is disabled by default and can be enabled by setting a loader tunable vm.pmap.pg_ps_enabled to 1. > if it's true i would be enough reason to upgrade to 7.2 on 2 computers. -cpghost. -- Cordula's Web. http://www.cordula.ws/