From owner-svn-src-all@FreeBSD.ORG Wed Jun 25 15:19:01 2014 Return-Path: Delivered-To: svn-src-all@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 32916656; Wed, 25 Jun 2014 15:19:01 +0000 (UTC) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id 19D582318; Wed, 25 Jun 2014 15:19:00 +0000 (UTC) Received: from Alfreds-MacBook-Pro-9.local (c-76-21-10-192.hsd1.ca.comcast.net [76.21.10.192]) by elvis.mu.org (Postfix) with ESMTPSA id 0E40E1A3C19; Wed, 25 Jun 2014 08:19:00 -0700 (PDT) Message-ID: <53AAE864.7010001@freebsd.org> Date: Wed, 25 Jun 2014 08:19:00 -0700 From: Alfred Perlstein User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 MIME-Version: 1.0 To: Andriy Gapon , attilio@FreeBSD.org, =?UTF-8?B?Um9nZXI=?= =?UTF-8?B?IFBhdSBNb25uw6k=?= Subject: Re: svn commit: r267858 - in head/sys/dev: virtio/balloon xen/balloon References: <201406250951.s5P9p8YR017159@svn.freebsd.org> <53AACEAB.3090702@FreeBSD.org> <53AAD356.20202@FreeBSD.org> In-Reply-To: <53AAD356.20202@FreeBSD.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Alan Cox , "svn-src-head@freebsd.org" , "svn-src-all@freebsd.org" , "src-committers@freebsd.org" X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Jun 2014 15:19:01 -0000 On 6/25/14 6:49 AM, Andriy Gapon wrote: > On 25/06/2014 17:44, Attilio Rao wrote: >> Why? If VM needs more wired memory I assume that we can tune up the >> default value of max_wired? >> >> I think that however your case makes an interesting point: if we want >> to make unmanaged pages as inherently wired, we likely need a little >> bit higher max_wired value. When I completed a patch for this, pho@ >> couldn't reproduce any similar issue even with stress-testing (and >> also, the places to allocate unmanaged pages and not requesting >> VM_ALLOC_WIRED were very little, almost 0, with the exception of >> vm_page_alloc_contig() calls) but I think it is a valid proposition. >> >> However I would still like to have more control on kernel-specific >> wired memory for processes. I'm for example thinking to ARC vs. buffer >> cache, where I expect the wired memory consumption to be much bigger >> for the former case. > My humble opinion is that userland page wiring should be tuned via > resource limits and that vm.max_wired could be retired altogether. > Kernel wiring ignores the knob anyway. > I think the goal of the limit as it stands would not to let the total amount of wired pages reach a bad point due to a userland program pushing it over the limit. Basically userland wants to know how many pages the kernel has wired in order to avoid asking for too many pages and making the system implode.