From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 06:33:12 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4D5DB16A401 for ; Mon, 24 Apr 2006 06:33:12 +0000 (UTC) (envelope-from mckusick@chez.mckusick.com) Received: from chez.mckusick.com (dsl081-247-049.sfo1.dsl.speakeasy.net [64.81.247.49]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0780D43D4C for ; Mon, 24 Apr 2006 06:33:11 +0000 (GMT) (envelope-from mckusick@chez.mckusick.com) Received: from chez.mckusick.com (localhost [127.0.0.1]) by chez.mckusick.com (8.13.6/8.13.1) with ESMTP id k3O6XUJ0042841 for ; Sun, 23 Apr 2006 23:33:30 -0700 (PDT) (envelope-from mckusick@chez.mckusick.com) Message-Id: <200604240633.k3O6XUJ0042841@chez.mckusick.com> To: arch@freebsd.org X-URL: http://WWW.McKusick.COM/ Date: Sun, 23 Apr 2006 23:33:30 -0700 From: Kirk McKusick Cc: Subject: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: Kirk McKusick List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 06:33:12 -0000 Anyone working on zero-copy sockets care to respond to this? http://developers.slashdot.org/developers/06/04/21/1536213.shtml Linus Torvalds made reference to some possible future extensions. This included vmsplice(), a system call since implemented by Jens Axboe "to basically do a 'write to the buffer', but using the reference counting and VM traversal to actually fill the buffer." Reviewing the implications of using such a system call lead to a comparison with FreeBSD's ZERO_COPY_SOCKET which uses COW (copy on write). Linus explained that while this may look good on specific benchmarks, it actually introduces extra overhead, "the thing is, the cost of marking things COW is not just the cost of the initial page table invalidate: it's also the cost of the fault eventually when you _do_ write to the page, even if at that point you decide that the page is no longer shared, and the fault can just mark the page writable again." He went on to explain, "The COW approach does generate some really nice benchmark numbers, because the way you benchmark this thing is that you never actually write to the user page in the first place, so you end up having a nice benchmark loop that has to do the TLB invalidate just the _first_ time, and never has to do any work ever again later on." Linus didn't pull any punches when he summarized: "I claim that Mach people (and apparently FreeBSD) are incompetent idiots. Playing games with VM is bad. memory copies are _also_ bad, but quite frankly, memory copies often have _less_ downside than VM games, and bigger caches will only continue to drive that point home." From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 06:43:53 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 42E0316A402 for ; Mon, 24 Apr 2006 06:43:53 +0000 (UTC) (envelope-from jmg@hydrogen.funkthat.com) Received: from hydrogen.funkthat.com (gate.funkthat.com [69.17.45.168]) by mx1.FreeBSD.org (Postfix) with ESMTP id D83E743D45 for ; Mon, 24 Apr 2006 06:43:52 +0000 (GMT) (envelope-from jmg@hydrogen.funkthat.com) Received: from hydrogen.funkthat.com (am7ucn9tjzutsgl2@localhost.funkthat.com [127.0.0.1]) by hydrogen.funkthat.com (8.13.4/8.13.3) with ESMTP id k3O6hqQa018150; Sun, 23 Apr 2006 23:43:52 -0700 (PDT) (envelope-from jmg@hydrogen.funkthat.com) Received: (from jmg@localhost) by hydrogen.funkthat.com (8.13.4/8.13.3/Submit) id k3O6hqCh018149; Sun, 23 Apr 2006 23:43:52 -0700 (PDT) (envelope-from jmg) Date: Sun, 23 Apr 2006 23:43:52 -0700 From: John-Mark Gurney To: Kirk McKusick Message-ID: <20060424064352.GA728@funkthat.com> Mail-Followup-To: Kirk McKusick , arch@freebsd.org References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200604240633.k3O6XUJ0042841@chez.mckusick.com> User-Agent: Mutt/1.4.2.1i X-Operating-System: FreeBSD 5.4-RELEASE-p6 i386 X-PGP-Fingerprint: B7 EC EF F8 AE ED A7 31 96 7A 22 B3 D8 56 36 F4 X-Files: The truth is out there X-URL: http://resnet.uoregon.edu/~gurney_j/ X-Resume: http://resnet.uoregon.edu/~gurney_j/resume.html Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: John-Mark Gurney List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 06:43:53 -0000 Kirk McKusick wrote this message on Sun, Apr 23, 2006 at 23:33 -0700: > Linus explained that while this may look good on specific benchmarks, > it actually introduces extra overhead, "the thing is, the cost of Has he benchmarked this to prove his point? And has he done it over realworld work loads, like Apache or another "standard" program instead of a microbenchmark designed especially to make COW look bad? As w/ all theories, w/o numbers, they are only theories till backed up w/ benchmarks. This isn't suppose to defend COW, but it is designed to ensure that people don't stop exploring just because someone says something... -- John-Mark Gurney Voice: +1 415 225 5579 "All that I will do, has been done, All that I have, has not." From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 06:55:27 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2E88316A404 for ; Mon, 24 Apr 2006 06:55:27 +0000 (UTC) (envelope-from julian@elischer.org) Received: from a50.ironport.com (a50.ironport.com [63.251.108.112]) by mx1.FreeBSD.org (Postfix) with ESMTP id E917A43D53 for ; Mon, 24 Apr 2006 06:55:26 +0000 (GMT) (envelope-from julian@elischer.org) Received: from unknown (HELO [192.168.2.4]) ([10.251.60.116]) by a50.ironport.com with ESMTP; 23 Apr 2006 23:55:26 -0700 Message-ID: <444C765B.7070803@elischer.org> Date: Sun, 23 Apr 2006 23:55:23 -0700 From: Julian Elischer User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7.12) Gecko/20050915 X-Accept-Language: en-us, en MIME-Version: 1.0 To: John-Mark Gurney References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> <20060424064352.GA728@funkthat.com> In-Reply-To: <20060424064352.GA728@funkthat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Kirk McKusick , arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 06:55:27 -0000 John-Mark Gurney wrote: >Kirk McKusick wrote this message on Sun, Apr 23, 2006 at 23:33 -0700: > > >>Linus explained that while this may look good on specific benchmarks, >>it actually introduces extra overhead, "the thing is, the cost of >> >> > >Has he benchmarked this to prove his point? And has he done it over >realworld work loads, like Apache or another "standard" program instead >of a microbenchmark designed especially to make COW look bad? > > Well no-one has even confirmed that freeBSD does all this "page flipping" etc. I doubt that Linus has looked inside the BSD kernels. He's probably just repeating what he's been told, and that's so accurate, right? I know that freeBSD developers have over the last few years also acknowledged that the speed of modern CPUs vs. memeory speeds makes it often less efficient to do certain optimisations than it used to be. >As w/ all theories, w/o numbers, they are only theories till backed up >w/ benchmarks. > >This isn't suppose to defend COW, but it is designed to ensure that >people don't stop exploring just because someone says something... > > > From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 07:06:23 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 32E5F16A402 for ; Mon, 24 Apr 2006 07:06:23 +0000 (UTC) (envelope-from phk@critter.freebsd.dk) Received: from phk.freebsd.dk (phk.freebsd.dk [130.225.244.222]) by mx1.FreeBSD.org (Postfix) with ESMTP id CB2BD43D46 for ; Mon, 24 Apr 2006 07:06:22 +0000 (GMT) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (unknown [192.168.48.2]) by phk.freebsd.dk (Postfix) with ESMTP id 0629A17313; Mon, 24 Apr 2006 07:06:20 +0000 (UTC) To: Kirk McKusick From: "Poul-Henning Kamp" In-Reply-To: Your message of "Sun, 23 Apr 2006 23:33:30 PDT." <200604240633.k3O6XUJ0042841@chez.mckusick.com> Date: Mon, 24 Apr 2006 09:06:19 +0200 Message-ID: <3942.1145862379@critter.freebsd.dk> Sender: phk@critter.freebsd.dk Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 07:06:23 -0000 In message <200604240633.k3O6XUJ0042841@chez.mckusick.com>, Kirk McKusick write s: >Anyone working on zero-copy sockets care to respond to this? I will strongly recommend that nobody does. First: _maybe_ relevant and competent benchmarks will _eventually_ vindicate one or the other approach, but in all likelyhood, the difference is a wash for all real-world, practical purposes. Second: nobody is going to convince anybody about anything on a topic where the fronts have been drawn up so sharp from the beginning. Time spent on tilting windmills is time not spent on the code. Third: Linus has internal project fights to fight, and this could quite likely be part of internal Linux bickering that got out of hand. If any kind of official response should be generated, it should be humorous and mostly non-insulting. Something like: The FreeBSD Project has taken Linus recent comments "ad notam". We usually hold Linus' technical competence and judgement in high regard, but if we are indeed "incompetent idiots" people should really not trust us on that point. would do fine. Poul-Henning -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 17:35:03 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 804B816A400 for ; Mon, 24 Apr 2006 17:35:03 +0000 (UTC) (envelope-from jfvogel@gmail.com) Received: from pproxy.gmail.com (pproxy.gmail.com [64.233.166.182]) by mx1.FreeBSD.org (Postfix) with ESMTP id AE04943D46 for ; Mon, 24 Apr 2006 17:35:02 +0000 (GMT) (envelope-from jfvogel@gmail.com) Received: by pproxy.gmail.com with SMTP id t32so1069627pyc for ; Mon, 24 Apr 2006 10:35:01 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=ccvw7rKHR5RPyOly/LIYMsfj5Z/Lv1EoCPvG/u6kw4QuJ2tuKEHs5yZVAygE5QWhjyACohFFMJq+bi+qJJZ4wx/J3g/LGBX8D6HMJB7dka2Sqol2iGFr/4K/8vGPo0bhp+ykzZ6+tZ1mpEOAejH6hD0QoHOCTATgPHSXyrfDVto= Received: by 10.35.113.12 with SMTP id q12mr1378214pym; Mon, 24 Apr 2006 10:35:01 -0700 (PDT) Received: by 10.35.17.16 with HTTP; Mon, 24 Apr 2006 10:35:01 -0700 (PDT) Message-ID: <2a41acea0604241035s325bf2c3x96baac71eff3e7db@mail.gmail.com> Date: Mon, 24 Apr 2006 10:35:01 -0700 From: "Jack Vogel" To: "Kirk McKusick" In-Reply-To: <200604240633.k3O6XUJ0042841@chez.mckusick.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 17:35:03 -0000 On 4/23/06, Kirk McKusick wrote: > Anyone working on zero-copy sockets care to respond to this? > > http://developers.slashdot.org/developers/06/04/21/1536213.shtml > > Linus Torvalds made reference to some possible future extensions. > This included vmsplice(), a system call since implemented by Jens > Axboe "to basically do a 'write to the buffer', but using the > reference counting and VM traversal to actually fill the buffer." > Reviewing the implications of using such a system call lead to a > comparison with FreeBSD's ZERO_COPY_SOCKET which uses COW (copy on > write). > > Linus explained that while this may look good on specific benchmarks, > it actually introduces extra overhead, "the thing is, the cost of > marking things COW is not just the cost of the initial page table > invalidate: it's also the cost of the fault eventually when you > _do_ write to the page, even if at that point you decide that the > page is no longer shared, and the fault can just mark the page > writable again." He went on to explain, "The COW approach does > generate some really nice benchmark numbers, because the way you > benchmark this thing is that you never actually write to the user > page in the first place, so you end up having a nice benchmark loop > that has to do the TLB invalidate just the _first_ time, and never > has to do any work ever again later on." Linus didn't pull any > punches when he summarized: > > "I claim that Mach people (and apparently FreeBSD) are incompetent > idiots. Playing games with VM is bad. memory copies are _also_ > bad, but quite frankly, memory copies often have _less_ downside > than VM games, and bigger caches will only continue to drive > that point home." I would have expected something a bit less adolescent out of Linus, or maybe its the reporter.... I don't think it deserves a reply however. Jack From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 19:13:10 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 504FA16A404 for ; Mon, 24 Apr 2006 19:13:10 +0000 (UTC) (envelope-from dave@dogwood.com) Received: from ms-smtp-03-eri0.socal.rr.com (ms-smtp-03.socal.rr.com [66.75.162.135]) by mx1.FreeBSD.org (Postfix) with ESMTP id D8AA043D48 for ; Mon, 24 Apr 2006 19:13:09 +0000 (GMT) (envelope-from dave@dogwood.com) Received: from white.dogwood.com (white.dogwood.com [66.91.140.178]) by ms-smtp-03-eri0.socal.rr.com (8.13.4/8.13.4) with ESMTP id k3OJD8uA027702; Mon, 24 Apr 2006 12:13:08 -0700 (PDT) Received: from white.dogwood.com (localhost.dogwood.com [127.0.0.1]) by white.dogwood.com (8.13.4/8.13.4) with ESMTP id k3OJD7EX068827; Mon, 24 Apr 2006 09:13:07 -1000 (HST) (envelope-from dave@white.dogwood.com) Received: (from dave@localhost) by white.dogwood.com (8.13.4/8.13.1/Submit) id k3OJD6S1068826; Mon, 24 Apr 2006 09:13:06 -1000 (HST) (envelope-from dave) From: Dave Cornejo Message-Id: <200604241913.k3OJD6S1068826@white.dogwood.com> In-Reply-To: <3942.1145862379@critter.freebsd.dk> To: Poul-Henning Kamp Date: Mon, 24 Apr 2006 09:13:06 -1000 (HST) X-Mailer: ELM [version 2.4ME+ PL121h (25)] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.0.2 (white.dogwood.com [127.0.0.1]); Mon, 24 Apr 2006 09:13:07 -1000 (HST) X-Virus-Scanned: Symantec AntiVirus Scan Engine Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 19:13:10 -0000 > In message <200604240633.k3O6XUJ0042841@chez.mckusick.com>, Kirk McKusick write > s: > >Anyone working on zero-copy sockets care to respond to this? > > I will strongly recommend that nobody does. > > First: _maybe_ relevant and competent benchmarks will _eventually_ > vindicate one or the other approach, but in all likelyhood, the > difference is a wash for all real-world, practical purposes. > > Second: nobody is going to convince anybody about anything on a > topic where the fronts have been drawn up so sharp from the beginning. > Time spent on tilting windmills is time not spent on the code. > > Third: Linus has internal project fights to fight, and this could > quite likely be part of internal Linux bickering that got out of > hand. > > If any kind of official response should be generated, it should > be humorous and mostly non-insulting. Something like: > > The FreeBSD Project has taken Linus recent comments "ad > notam". We usually hold Linus' technical competence and > judgement in high regard, but if we are indeed "incompetent > idiots" people should really not trust us on that point. > > would do fine. This one truly deserves an honored place in the forutne file > > Poul-Henning > > -- > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD committer | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by incompetence. > _______________________________________________ > freebsd-arch@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-arch > To unsubscribe, send any mail to "freebsd-arch-unsubscribe@freebsd.org" > From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 20:00:01 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9C33216A50B for ; Mon, 24 Apr 2006 20:00:01 +0000 (UTC) (envelope-from dmitry@atlantis.dp.ua) Received: from postman.atlantis.dp.ua (postman.atlantis.dp.ua [193.108.47.1]) by mx1.FreeBSD.org (Postfix) with ESMTP id 86CC643D6B for ; Mon, 24 Apr 2006 19:59:57 +0000 (GMT) (envelope-from dmitry@atlantis.dp.ua) Received: from smtp.atlantis.dp.ua (smtp.atlantis.dp.ua [193.108.46.231]) by postman.atlantis.dp.ua (8.13.1/8.13.1) with ESMTP id k3OJxp7Q099344; Mon, 24 Apr 2006 22:59:51 +0300 (EEST) (envelope-from dmitry@atlantis.dp.ua) Date: Mon, 24 Apr 2006 22:59:51 +0300 (EEST) From: Dmitry Pryanishnikov To: Jack Vogel In-Reply-To: <2a41acea0604241035s325bf2c3x96baac71eff3e7db@mail.gmail.com> Message-ID: <20060424225726.O33363@atlantis.atlantis.dp.ua> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> <2a41acea0604241035s325bf2c3x96baac71eff3e7db@mail.gmail.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: Kirk McKusick , arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 20:00:01 -0000 Hello! On Mon, 24 Apr 2006, Jack Vogel wrote: >> has to do any work ever again later on." Linus didn't pull any >> punches when he summarized: >> >> "I claim that Mach people (and apparently FreeBSD) are incompetent >> idiots. Playing games with VM is bad. memory copies are _also_ >> bad, but quite frankly, memory copies often have _less_ downside >> than VM games, and bigger caches will only continue to drive >> that point home." > > I would have expected something a bit less adolescent out of Linus, > or maybe its the reporter.... I don't think it deserves a reply however. Maybe, it's the first sign of his senile marasmus instead? ;))) Sincerely, Dmitry -- Atlantis ISP, System Administrator e-mail: dmitry@atlantis.dp.ua nic-hdl: LYNX-RIPE From owner-freebsd-arch@FreeBSD.ORG Mon Apr 24 23:24:26 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 958DC16A403 for ; Mon, 24 Apr 2006 23:24:26 +0000 (UTC) (envelope-from kabaev@gmail.com) Received: from wproxy.gmail.com (wproxy.gmail.com [64.233.184.226]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0EC5643D46 for ; Mon, 24 Apr 2006 23:24:25 +0000 (GMT) (envelope-from kabaev@gmail.com) Received: by wproxy.gmail.com with SMTP id i12so1119360wra for ; Mon, 24 Apr 2006 16:24:25 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:date:from:to:cc:subject:message-id:in-reply-to:references:x-mailer:mime-version:content-type; b=lGQlO+mrlxyLhx2eTxuVYJ4kvSYS/4PZHVsq3J+O2k9lnw+9tUGTrMYqtpayikODP7BIhJLNbxjkzw0WuEBqmE/pggLRs+uf3+TwcoR6b8mofr9yXsV0EWpQ0QFEd45W0uy2H3kQhUB36qU+Rr0s3UT5NeY5Gq5O80Z1Zlfc3ks= Received: by 10.54.62.6 with SMTP id k6mr1283942wra; Mon, 24 Apr 2006 16:24:25 -0700 (PDT) Received: from kan.dnsalias.net ( [24.63.93.195]) by mx.gmail.com with ESMTP id g7sm221446wra.2006.04.24.16.24.24; Mon, 24 Apr 2006 16:24:24 -0700 (PDT) Date: Mon, 24 Apr 2006 19:24:14 -0400 From: Alexander Kabaev To: Julian Elischer Message-ID: <20060424192414.0dbaa534@kan.dnsalias.net> In-Reply-To: <444C765B.7070803@elischer.org> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> <20060424064352.GA728@funkthat.com> <444C765B.7070803@elischer.org> X-Mailer: Sylpheed-Claws 2.0.0 (GTK+ 2.8.16; i386-portbld-freebsd7.0) Mime-Version: 1.0 Content-Type: multipart/signed; boundary=Sig_iTCP_znP1nl+B37C3il8aZO; protocol="application/pgp-signature"; micalg=PGP-SHA1 Cc: Kirk McKusick , arch@freebsd.org, John-Mark Gurney Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Apr 2006 23:24:26 -0000 --Sig_iTCP_znP1nl+B37C3il8aZO Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Sun, 23 Apr 2006 23:55:23 -0700 Julian Elischer wrote: > John-Mark Gurney wrote: >=20 > >Kirk McKusick wrote this message on Sun, Apr 23, 2006 at 23:33 -0700: > > =20 > > > >>Linus explained that while this may look good on specific > >>benchmarks, it actually introduces extra overhead, "the thing is, > >>the cost of > >> =20 > >> > > > >Has he benchmarked this to prove his point? And has he done it over > >realworld work loads, like Apache or another "standard" program > >instead of a microbenchmark designed especially to make COW look bad? > > =20 > > >=20 > Well no-one has even confirmed that freeBSD does all this > "page flipping" etc. I doubt that Linus has looked inside the BSD > kernels. He's probably just repeating what he's been told, and that's > so accurate, right? >=20 > I know that freeBSD developers have over the last few years also=20 > acknowledged that > the speed of modern CPUs vs. memeory speeds makes it often less > efficient to do certain optimisations than it used to be. >=20 >=20 >=20 >=20 > >As w/ all theories, w/o numbers, they are only theories till backed > >up w/ benchmarks. > > > >This isn't suppose to defend COW, but it is designed to ensure that > >people don't stop exploring just because someone says something... > > > > =20 > > > _______________________________________________ > freebsd-arch@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-arch > To unsubscribe, send any mail to > "freebsd-arch-unsubscribe@freebsd.org" The original zero-copy code was used to stream network data directly to an onboard SDRAM memory on an extension PCI card. Linus' notes about relative cost of memory-to-memory copies vs. TLB shootdowns and possible page access traps had little relevance there. -- Alexander Kabaev --=20 Alexander Kabaev --Sig_iTCP_znP1nl+B37C3il8aZO Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (FreeBSD) iD8DBQFETV4nQ6z1jMm+XZYRAmb0AJ9wJUar/mwUJc7XwoCbcfTNR5j6WgCgn6xu HNu/4H0KgHUFtuWcyiXutpw= =MBNK -----END PGP SIGNATURE----- --Sig_iTCP_znP1nl+B37C3il8aZO-- From owner-freebsd-arch@FreeBSD.ORG Tue Apr 25 02:39:30 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 70BD016A400 for ; Tue, 25 Apr 2006 02:39:30 +0000 (UTC) (envelope-from gnn@neville-neil.com) Received: from mrout2-b.corp.dcn.yahoo.com (mrout2-b.corp.dcn.yahoo.com [216.109.112.28]) by mx1.FreeBSD.org (Postfix) with ESMTP id 171D443D46 for ; Tue, 25 Apr 2006 02:39:30 +0000 (GMT) (envelope-from gnn@neville-neil.com) Received: from minion.local.neville-neil.com (proxy7.corp.yahoo.com [216.145.48.98]) by mrout2-b.corp.dcn.yahoo.com (8.13.6/8.13.4/y.out) with ESMTP id k3P2cuiK039308; Mon, 24 Apr 2006 19:38:56 -0700 (PDT) Date: Tue, 25 Apr 2006 10:22:54 +0900 Message-ID: From: gnn@freebsd.org To: Kirk McKusick In-Reply-To: <200604240633.k3O6XUJ0042841@chez.mckusick.com> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> User-Agent: Wanderlust/2.14.0 (Africa) SEMI/1.14.6 (Maruoka) FLIM/1.14.8 (=?ISO-8859-4?Q?Shij=F2?=) APEL/10.6 Emacs/22.0.50 (i386-apple-darwin8.5.1) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Apr 2006 02:39:30 -0000 At Sun, 23 Apr 2006 23:33:30 -0700, Kirk McKusick wrote: > > Anyone working on zero-copy sockets care to respond to this? > > http://developers.slashdot.org/developers/06/04/21/1536213.shtml > > Linus Torvalds made reference to some possible future extensions. > This included vmsplice(), a system call since implemented by Jens > Axboe "to basically do a 'write to the buffer', but using the > reference counting and VM traversal to actually fill the buffer." > Reviewing the implications of using such a system call lead to a > comparison with FreeBSD's ZERO_COPY_SOCKET which uses COW (copy on > write). > To be honest, what all this has made me think of is that we should have a "big board" of unsolved problems we'd like to look at. This certainly deserves a place, as it's an interesting question. As scientists and engineers we should be interested in such things, even if the original statement was poorly worded. Perhaps on a FreeBSD Feature Wiki? We could even propose a benchmark as a SoC project. Later, George From owner-freebsd-arch@FreeBSD.ORG Tue Apr 25 04:18:39 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4D2C416A400; Tue, 25 Apr 2006 04:18:39 +0000 (UTC) (envelope-from linimon@lonesome.com) Received: from mail.soaustin.net (mail.soaustin.net [207.200.4.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id 154BA43D45; Tue, 25 Apr 2006 04:18:39 +0000 (GMT) (envelope-from linimon@lonesome.com) Received: by mail.soaustin.net (Postfix, from userid 502) id 968866FD; Mon, 24 Apr 2006 23:18:38 -0500 (CDT) Date: Mon, 24 Apr 2006 23:18:38 -0500 To: gnn@freebsd.org Message-ID: <20060425041838.GC20621@soaustin.net> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.9i From: linimon@lonesome.com (Mark Linimon) Cc: Kirk McKusick , arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Apr 2006 04:18:39 -0000 On Tue, Apr 25, 2006 at 10:22:54AM +0900, gnn@freebsd.org wrote: > Perhaps on a FreeBSD Feature Wiki? I would hope the existing wiki could serve. > We could even propose a benchmark as a SoC project. This is the best idea I've heard in a loooong time. mcl From owner-freebsd-arch@FreeBSD.ORG Tue Apr 25 13:13:41 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A155616A401; Tue, 25 Apr 2006 13:13:41 +0000 (UTC) (envelope-from Alexander@Leidinger.net) Received: from www.ebusiness-leidinger.de (jojo.ms-net.de [84.16.236.246]) by mx1.FreeBSD.org (Postfix) with ESMTP id ECF0443D6A; Tue, 25 Apr 2006 13:13:38 +0000 (GMT) (envelope-from Alexander@Leidinger.net) Received: from Andro-Beta.Leidinger.net (p54A5F798.dip.t-dialin.net [84.165.247.152]) (authenticated bits=0) by www.ebusiness-leidinger.de (8.13.4/8.13.1) with ESMTP id k3PD7Msx022356; Tue, 25 Apr 2006 15:07:23 +0200 (CEST) (envelope-from Alexander@Leidinger.net) Received: from localhost (localhost [127.0.0.1]) by Andro-Beta.Leidinger.net (8.13.4/8.13.3) with ESMTP id k3PDDTxb030792; Tue, 25 Apr 2006 15:13:29 +0200 (CEST) (envelope-from Alexander@Leidinger.net) Received: from pslux.cec.eu.int (pslux.cec.eu.int [158.169.9.14]) by webmail.leidinger.net (Horde MIME library) with HTTP; Tue, 25 Apr 2006 15:13:28 +0200 Message-ID: <20060425151328.qj9qycqq8cs4wso4@netchild.homeip.net> X-Priority: 3 (Normal) Date: Tue, 25 Apr 2006 15:13:28 +0200 From: Alexander Leidinger To: gnn@freebsd.org References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; DelSp="Yes"; format="flowed" Content-Disposition: inline Content-Transfer-Encoding: 7bit User-Agent: Internet Messaging Program (IMP) H3 (4.1) / FreeBSD-4.11 X-Virus-Scanned: by amavisd-new Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Apr 2006 13:13:41 -0000 Quoting gnn@freebsd.org (from Tue, 25 Apr 2006 10:22:54 +0900): > To be honest, what all this has made me think of is that we should > have a "big board" of unsolved problems we'd like to look at. This Like "http://www.freebsd.org/projects/ideas/"? The only limitation for entries on this list is, that it has to be relevant to FreeBSD. > certainly deserves a place, as it's an interesting question. As > scientists and engineers we should be interested in such things, even > if the original statement was poorly worded. > > Perhaps on a FreeBSD Feature Wiki? We could even propose a benchmark > as a SoC project. Just write up an appropriate text for the ideas list and send it to joel@ and me. Bye, Alexander. -- http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 The world is coming to an end. Please log off. From owner-freebsd-arch@FreeBSD.ORG Wed Apr 26 15:26:09 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3E15916A401 for ; Wed, 26 Apr 2006 15:26:09 +0000 (UTC) (envelope-from Antigen_WEBMAIL@mimosasystems.com) Received: from webmail.mimosasystems.com (h-66-166-239-134.snvacaid.covad.net [66.166.239.134]) by mx1.FreeBSD.org (Postfix) with ESMTP id BEFF343D46 for ; Wed, 26 Apr 2006 15:26:05 +0000 (GMT) (envelope-from Antigen_WEBMAIL@mimosasystems.com) Received: from mail pickup service by webmail.mimosasystems.com with Microsoft SMTPSVC; Wed, 26 Apr 2006 08:26:02 -0700 From: Antigen_WEBMAIL To: freebsd-arch@freebsd.org MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="ANTIGEN-2268-14263-19791-7931-18382-13380" Message-ID: X-OriginalArrivalTime: 26 Apr 2006 15:26:02.0816 (UTC) FILETIME=[BA6F6400:01C66945] Date: 26 Apr 2006 08:26:02 -0700 Subject: Antigen forwarded attachment X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Apr 2006 15:26:09 -0000 --ANTIGEN-2268-14263-19791-7931-18382-13380 Content-Type: text/plain; charset=us-ascii The entire message "freebsd-arch Digest, Vol 157, Issue 5", originally sent to you by owner-freebsd-arch@freebsd.org (owner-freebsd-arch@freebsd.org), has been forwarded to you from the Antigen Quarantine area. This message may have been re-scanned by Antigen and handled according to the appropriate scan job's settings. <> --ANTIGEN-2268-14263-19791-7931-18382-13380 Content-Type: message/rfc822; name="Entire Message.eml" Received: from psmtp.com ([10.0.0.105]) by webmail.mimosasystems.com with Microsoft SMTPSVC(6.0.3790.1830); Fri, 21 Apr 2006 05:01:47 -0700 Received: from source ([216.136.204.119]) by exprod5mx133.postini.com ([64.18.4.10]) with SMTP; Fri, 21 Apr 2006 05:01:47 PDT Received: from hub.freebsd.org (hub.freebsd.org [216.136.204.18]) by mx2.freebsd.org (Postfix) with ESMTP id 5602564929; Fri, 21 Apr 2006 12:00:31 +0000 (GMT) (envelope-from owner-freebsd-arch@freebsd.org) Received: from hub.freebsd.org (localhost [127.0.0.1]) by hub.freebsd.org (Postfix) with ESMTP id BEBEB16A448; Fri, 21 Apr 2006 12:00:28 +0000 (UTC) (envelope-from owner-freebsd-arch@freebsd.org) From: freebsd-arch-request@freebsd.org Subject: freebsd-arch Digest, Vol 157, Issue 5 To: freebsd-arch@freebsd.org Reply-To: freebsd-arch@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: owner-freebsd-arch@freebsd.org Errors-To: owner-freebsd-arch@freebsd.org Message-Id: <20060421120028.BEBEB16A448@hub.freebsd.org> Date: Fri, 21 Apr 2006 12:00:28 +0000 (UTC) X-pstn-levels: (S:89.14508/99.90000 R:95.9108 P:94.8683 M:99.4056 C:99.5902 ) X-pstn-settings: 3 (1.0000:1.0000) s gt3 gt2 gt1 r p m c X-pstn-addresses: from [36/1] Return-Path: owner-freebsd-arch@freebsd.org X-OriginalArrivalTime: 21 Apr 2006 12:01:48.0006 (UTC) FILETIME=[5DEF0860:01C6653B] Send freebsd-arch mailing list submissions to freebsd-arch@freebsd.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.freebsd.org/mailman/listinfo/freebsd-arch or, via email, send a message with subject or body 'help' to freebsd-arch-request@freebsd.org You can reach the person managing the list at freebsd-arch-owner@freebsd.org When replying, please edit your Subject line so it is more specific than "Re: Contents of freebsd-arch digest..." Today's Topics: 1. Re: Add some more information in the ktrace(1)/kdump(1) output (David Kirchner) 2. Re: remove rid pointer (but not rid)... (M. Warner Losh) 3. Re: nawov news (Sylvana Edgecomb) ---------------------------------------------------------------------- Message: 1 Date: Thu, 20 Apr 2006 08:38:36 -0700 From: "David Kirchner" Subject: Re: Add some more information in the ktrace(1)/kdump(1) output To: "Peter Jeremy" Cc: arch@freebsd.org Message-ID: <35c231bf0604200838w224c810cue82beace0d63f18b@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 On 4/20/06, Peter Jeremy wrote: > Looks good. One improvement I'd suggest is to report both raw hex as well > as decoded output - eg the way a printf(9) %b format looks. This would > make the above look like: > > 32229 telnet CALL mmap(0,0x8000,0x3,0x1002,0xffffffff,0,0,0) > 32229 telnet CALL open(0x2807bc28,0,0x1b6) > 32229 telnet CALL socket(PF_INET,SOCK_DGRAM,0) Yeah, I could see a benefit to that. Easy to add. > >more data in the dump output. I'm thinking, specifically, adding > >KTR_STAT for stat() results and KTR_SOCKADDR for connect(), bind() > >arguments, and accept() results. > > Wonderful. IMHO, the lack of struct sockaddr decoding is the biggest > drawback of ktrace (though I admit I've never been sufficiently > annoyed at the lack of support to actually implement it). > > Some comments on the code approach (these are personal opinions - > feel free to ignore them): > - I find case statements easier to follow than very long "else if" > clauses. I do, too. I built on the existing if/else structure already there, but I wasn't sure of the best accepted style to use. > - Have you considered a semi-interpreted approach to allow more > generalised decoding (less special-cased code)? > [..] > /*004*/ { SYS_write , "write(dbd)d" , 3 , printargs_write , printret_write } , > > The string contains the syscall name, the argument types in > parenthesis (b - buffer, d - signed decimal, p - pointer, o - octal > etc) and a return type, the number of arguments and functions to print > the arguments and return values. (The return handling wouldn't be > relevant to ktrace). About 3/4 of the Tru64 syscalls can be handled > using the generic printargs_gen(). That does look great. I'll give it a shot. I was definitely getting concerned over the number of different functions inside each if(...SYS_foo){} statement. It's easy to make a mistake there. > Since ktrace doesn't need special handling for syscalls that have I/O > buffers or various structures (they are passed via different KTR_xxx > records), FreeBSD is even more amenable to generic argument processing. > I suspect that virtually all of the FreeBSD syscalls could be handled > in a similar manner if a %b option string and an enum decoding > structure (or two) could be passed via a similar sort of table. Yeah. That sounds good. The original patch should probably not be committed then, since it'll just get changed again Real Soon Now(tm). ------------------------------ Message: 2 Date: Thu, 20 Apr 2006 10:40:48 -0600 (MDT) From: "M. Warner Losh" Subject: Re: remove rid pointer (but not rid)... To: phk@phk.freebsd.dk Cc: arch@FreeBSD.ORG, gurney_j@resnet.uoregon.edu Message-ID: <20060420.104048.57443855.imp@bsdimp.com> Content-Type: Text/Plain; charset=us-ascii In message: <3731.1145513558@critter.freebsd.dk> "Poul-Henning Kamp" writes: : In message <20060419.165709.22017808.imp@bsdimp.com>, "M. Warner Losh" writes: : : >This will, of course, be a big undertaking in the tree. There are : >1216 instances of the string 'alloc_resource' Every single one of : >them will need to change in some way. : : Most of them could beneficially change to use bus_alloc_resources(), : thereby improving the code (there are far too many bugs in the error : handling) and at the same time cutting the 1216 in half or further : down. Agreed. I think I made that point a little later in my post. The big difficulty here is testing the driver after the changes to make sure that no unintended bugs have been introduced. Warner ------------------------------ Message: 3 Date: Thu, 20 Apr 2006 18:55:14 -0400 From: "Sylvana Edgecomb" Subject: Re: nawov news To: freebsd-arch@freebsd.org Message-ID: <000001c664cd$7c8e4650$5944a8c0@pow45> Content-Type: text/plain; charset="us-ascii" Dea r r Home Ow i ne q r , Your c p red x it doesn't matter to us ! If you O y WN real e g st f at f e and want I v MMED a IAT i E c y ash to sp k en f d ANY way you like, or simply wish to L g OWER your monthly p z aym z ents by a third or more, here are the dea s ls we have T t ODA u Y : $ 4 i 88 , 000 at a 3 , g 67% f z ixed - rat z e $ 3 r 72 , 000 at a 3 , s 90% v p ariabl z e - rat l e $ 49 w 2 , 000 at a 3 , 2 m 1% i c ntere b st - only $ 2 r 48 , 000 at a 3 , a 36% fi z xed - rat d e $ 1 i 98 , 000 at a 3 , a 55% v j ariable - ra d te Hurr i y, when these d x eaIs are gone, they are gone ! Don't worry about app x rov o al, your c g red b it will not d o isqualif j y you ! Vi m si u t our si h te Sincerely, Sylvana Edgecomb Ap z pro f val Manager ------------------------------ _______________________________________________ freebsd-arch@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-arch To unsubscribe, send any mail to "freebsd-arch-unsubscribe@freebsd.org" End of freebsd-arch Digest, Vol 157, Issue 5 ******************************************** --ANTIGEN-2268-14263-19791-7931-18382-13380-- From owner-freebsd-arch@FreeBSD.ORG Wed Apr 26 22:18:56 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 47C8516A400 for ; Wed, 26 Apr 2006 22:18:56 +0000 (UTC) (envelope-from ken@nargothrond.kdm.org) Received: from nargothrond.kdm.org (nargothrond.kdm.org [70.56.43.81]) by mx1.FreeBSD.org (Postfix) with ESMTP id BA28A43D45 for ; Wed, 26 Apr 2006 22:18:55 +0000 (GMT) (envelope-from ken@nargothrond.kdm.org) Received: from nargothrond.kdm.org (localhost [127.0.0.1]) by nargothrond.kdm.org (8.13.6/8.13.6) with ESMTP id k3QMIrgL094769; Wed, 26 Apr 2006 16:18:53 -0600 (MDT) (envelope-from ken@nargothrond.kdm.org) Received: (from ken@localhost) by nargothrond.kdm.org (8.13.6/8.13.6/Submit) id k3QMIqiR094768; Wed, 26 Apr 2006 16:18:52 -0600 (MDT) (envelope-from ken) Date: Wed, 26 Apr 2006 16:18:52 -0600 From: "Kenneth D. Merry" To: Alexander Kabaev Message-ID: <20060426221852.GA94673@nargothrond.kdm.org> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> <20060424064352.GA728@funkthat.com> <444C765B.7070803@elischer.org> <20060424192414.0dbaa534@kan.dnsalias.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20060424192414.0dbaa534@kan.dnsalias.net> User-Agent: Mutt/1.4.2i X-Virus-Scanned: ClamAV 0.88.1/1426/Wed Apr 26 12:03:01 2006 on nargothrond.kdm.org X-Virus-Status: Clean Cc: Kirk McKusick , arch@freebsd.org, John-Mark Gurney , Julian Elischer Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Apr 2006 22:18:56 -0000 On Mon, Apr 24, 2006 at 19:24:14 -0400, Alexander Kabaev wrote: > On Sun, 23 Apr 2006 23:55:23 -0700 > Julian Elischer wrote: > > > John-Mark Gurney wrote: > > > > >Kirk McKusick wrote this message on Sun, Apr 23, 2006 at 23:33 -0700: > > > > > > > > >>Linus explained that while this may look good on specific > > >>benchmarks, it actually introduces extra overhead, "the thing is, > > >>the cost of > > >> > > >> > > > > > >Has he benchmarked this to prove his point? And has he done it over > > >realworld work loads, like Apache or another "standard" program > > >instead of a microbenchmark designed especially to make COW look bad? > > > > > > > > > > Well no-one has even confirmed that freeBSD does all this > > "page flipping" etc. I doubt that Linus has looked inside the BSD > > kernels. He's probably just repeating what he's been told, and that's > > so accurate, right? > > > > I know that freeBSD developers have over the last few years also > > acknowledged that > > the speed of modern CPUs vs. memeory speeds makes it often less > > efficient to do certain optimisations than it used to be. > > > > > > > > > > >As w/ all theories, w/o numbers, they are only theories till backed > > >up w/ benchmarks. > > > > > >This isn't suppose to defend COW, but it is designed to ensure that > > >people don't stop exploring just because someone says something... > > > > > > > > > > > _______________________________________________ > > freebsd-arch@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-arch > > To unsubscribe, send any mail to > > "freebsd-arch-unsubscribe@freebsd.org" > The original zero-copy code was used to stream network data directly to > an onboard SDRAM memory on an extension PCI card. Linus' notes about > relative cost of memory-to-memory copies vs. TLB shootdowns and possible > page access traps had little relevance there. If you're talking about the code I wrote at Pluto/Avid, yeah, it was primarily to get around very poor CPU access speed to the external RAID cache we had on that box. The zero copy send model used async I/O semantics -- you'd get the completed buffers back when they were freed, so you knew you could reuse them in userland. The zero copy receive model relied on extensive modifications to the Alteon Tigon II firmware -- a whole lot more than the header splitting firmware that's in the FreeBSD tree now. I added 16 extra receive rings to the firmware to handle up to 16 incoming FTP connections. Packet headers were DMAed into normal kernel memory, and the payloads were DMAed into the PCI attached RAID cache. Since neither implementation was generally useful, Drew Gallatin and I cleaned up his zero copy sockets implementation for inclusion in FreeBSD. The receive semantics with respect to userland were similar to what I did at Pluto/Avid, but the send semantics are somewhat different, since they rely on COW instead of a separate notification that the buffer is done. Is any of that old codebase still in use? I figured it probably went away with the old hardware... Ken -- Kenneth Merry ken@FreeBSD.ORG From owner-freebsd-arch@FreeBSD.ORG Wed Apr 26 22:51:22 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B05E516A402 for ; Wed, 26 Apr 2006 22:51:22 +0000 (UTC) (envelope-from kabaev@gmail.com) Received: from xproxy.gmail.com (xproxy.gmail.com [66.249.82.200]) by mx1.FreeBSD.org (Postfix) with ESMTP id B219443D46 for ; Wed, 26 Apr 2006 22:51:21 +0000 (GMT) (envelope-from kabaev@gmail.com) Received: by xproxy.gmail.com with SMTP id s9so1067906wxc for ; Wed, 26 Apr 2006 15:51:20 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:date:from:to:cc:subject:message-id:in-reply-to:references:x-mailer:mime-version:content-type; b=sNql64S0fCvht4rEbVbF33uondwCZSb6iZee0DfUAJQ+FXdOCKVdQ7++fnC9Zw/TDqpJwYBb6ihohiL/yddvhDo+0QMqU7hHtRQrdgJT03xSD53o8FJndraQfG6yPV5KLN5CSM/B0uzQRkzxorcTzBcEpSOOuZF9cCqUCsDu5lI= Received: by 10.70.21.4 with SMTP id 4mr1494688wxu; Wed, 26 Apr 2006 15:51:20 -0700 (PDT) Received: from kan.dnsalias.net ( [24.63.93.195]) by mx.gmail.com with ESMTP id i15sm2091949wxd.2006.04.26.15.51.19; Wed, 26 Apr 2006 15:51:20 -0700 (PDT) Date: Wed, 26 Apr 2006 18:51:15 -0400 From: Alexander Kabaev To: "Kenneth D. Merry" Message-ID: <20060426185115.6dcd8d67@kan.dnsalias.net> In-Reply-To: <20060426221852.GA94673@nargothrond.kdm.org> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> <20060424064352.GA728@funkthat.com> <444C765B.7070803@elischer.org> <20060424192414.0dbaa534@kan.dnsalias.net> <20060426221852.GA94673@nargothrond.kdm.org> X-Mailer: Sylpheed-Claws 2.0.0 (GTK+ 2.8.16; i386-portbld-freebsd7.0) Mime-Version: 1.0 Content-Type: multipart/signed; boundary=Sig_fnO_L9XEQGoDAj0zS6txULs; protocol="application/pgp-signature"; micalg=PGP-SHA1 Cc: Kirk McKusick , arch@freebsd.org, John-Mark Gurney , Julian Elischer Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Apr 2006 22:51:22 -0000 --Sig_fnO_L9XEQGoDAj0zS6txULs Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Wed, 26 Apr 2006 16:18:52 -0600 "Kenneth D. Merry" wrote: > If you're talking about the code I wrote at Pluto/Avid, yeah, it was > primarily to get around very poor CPU access speed to the external > RAID cache we had on that box. Yes, I am talking about that code. >=20 > The zero copy send model used async I/O semantics -- you'd get the > completed buffers back when they were freed, so you knew you could > reuse them in userland. >=20 > The zero copy receive model relied on extensive modifications to the > Alteon Tigon II firmware -- a whole lot more than the header > splitting firmware that's in the FreeBSD tree now. I added 16 extra > receive rings to the firmware to handle up to 16 incoming FTP > connections. Packet headers were DMAed into normal kernel memory, > and the payloads were DMAed into the PCI attached RAID cache. >=20 > Since neither implementation was generally useful, Drew Gallatin and I > cleaned up his zero copy sockets implementation for inclusion in > FreeBSD. The receive semantics with respect to userland were similar > to what I did at Pluto/Avid, but the send semantics are somewhat > different, since they rely on COW instead of a separate notification > that the buffer is done. >=20 > Is any of that old codebase still in use? I figured it probably went > away with the old hardware... >=20 AirSpaces are still being used in the wild, and we still care and feed them occasionally. The same code is present in newer AirSpeed product, but is disabled by default because AirSpeed uses onboard em network adapters.=20 --=20 Alexander Kabaev --Sig_fnO_L9XEQGoDAj0zS6txULs Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (FreeBSD) iD8DBQFET/lnQ6z1jMm+XZYRAqDeAJ9hOof4Kx5i0e+DCxBaTFo6sug5uACfblcY xf7rry1Lce6VizNgRpSq7R0= =lo/0 -----END PGP SIGNATURE----- --Sig_fnO_L9XEQGoDAj0zS6txULs-- From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 13:54:54 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3C84316A400 for ; Fri, 28 Apr 2006 13:54:54 +0000 (UTC) (envelope-from jumper99@gmx.de) Received: from mail.gmx.net (mail.gmx.de [213.165.64.20]) by mx1.FreeBSD.org (Postfix) with SMTP id 6DAA143D45 for ; Fri, 28 Apr 2006 13:54:53 +0000 (GMT) (envelope-from jumper99@gmx.de) Received: (qmail invoked by alias); 28 Apr 2006 13:54:51 -0000 Received: from unknown (EHLO wsa096) [193.101.155.96] by mail.gmx.net (mp037) with SMTP; 28 Apr 2006 15:54:51 +0200 X-Authenticated: #682707 Message-ID: <00e101c66acb$526b8000$609b65c1@vpe.de> From: "Helmut Schneider" To: Date: Fri, 28 Apr 2006 15:54:50 +0200 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2869 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2869 X-Y-GMX-Trusted: 0 Subject: targ-cpu.h X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 13:54:54 -0000 Hi, please forgive my noob question. I have a new CPU (AMD64) and I successfully played around with cross compiling to i386. Now, a collegue of mine asked me if it is also possible to do this with ARM. But is seems that there are some things missing for that plattform: root@BSDHelmut:/usr/src# make TARGET_ARCH=arm buildworld [...] In file included from /usr/src/gnu/usr.bin/binutils/as/obj-format.h:3, from /usr/src/gnu/usr.bin/binutils/as/../../../../contrib/binutils/gas/config/te-freebsd.h:30, from /usr/src/gnu/usr.bin/binutils/as/targ-env.h:3, from /usr/src/gnu/usr.bin/binutils/as/../../../../contrib/binutils/gas/as.h:626, from /usr/src/gnu/usr.bin/binutils/as/../../../../contrib/binutils/gas/config/tc-arm.c:29: /usr/src/gnu/usr.bin/binutils/as/../../../../contrib/binutils/gas/config/obj-elf.h:42:22: targ-cpu.h: No such file or directory mkdep: compile failed *** Error code 1 root@BSDHelmut:/usr/src# root@BSDHelmut:/usr/src# find . | grep "targ-cpu.h" /usr/src/gnu/usr.bin/binutils/as/alpha-freebsd/targ-cpu.h /usr/src/gnu/usr.bin/binutils/as/amd64-freebsd/targ-cpu.h /usr/src/gnu/usr.bin/binutils/as/i386-freebsd/targ-cpu.h /usr/src/gnu/usr.bin/binutils/as/ia64-freebsd/targ-cpu.h /usr/src/gnu/usr.bin/binutils/as/powerpc-freebsd/targ-cpu.h /usr/src/gnu/usr.bin/binutils/as/sparc64-freebsd/targ-cpu.h root@BSDHelmut:/usr/src# Where do I get as/arm-freebsd/targ-cpu.h etc. from? Thanks, Helmut From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 14:12:38 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 003EE16A400 for ; Fri, 28 Apr 2006 14:12:37 +0000 (UTC) (envelope-from mlfbsd@dong.ci0.org) Received: from dong.ci0.org (cognet.ci0.org [80.65.224.102]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2C0AE43D48 for ; Fri, 28 Apr 2006 14:12:36 +0000 (GMT) (envelope-from mlfbsd@dong.ci0.org) Received: from dong.ci0.org (localhost.ci0.org [127.0.0.1]) by dong.ci0.org (8.13.6/8.13.4) with ESMTP id k3SEkCeo031839; Fri, 28 Apr 2006 16:46:13 +0200 (CEST) (envelope-from mlfbsd@dong.ci0.org) Received: (from mlfbsd@localhost) by dong.ci0.org (8.13.6/8.13.4/Submit) id k3SEkCUp031838; Fri, 28 Apr 2006 16:46:12 +0200 (CEST) (envelope-from mlfbsd) Date: Fri, 28 Apr 2006 16:46:12 +0200 From: Olivier Houchard To: Helmut Schneider Message-ID: <20060428144612.GA31797@ci0.org> References: <00e101c66acb$526b8000$609b65c1@vpe.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <00e101c66acb$526b8000$609b65c1@vpe.de> User-Agent: Mutt/1.4.1i Cc: freebsd-arch@freebsd.org Subject: Re: targ-cpu.h X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 14:12:38 -0000 On Fri, Apr 28, 2006 at 03:54:50PM +0200, Helmut Schneider wrote: > Hi, > > please forgive my noob question. I have a new CPU (AMD64) and I > successfully played around with cross compiling to i386. > Now, a collegue of mine asked me if it is also possible to do this with > ARM. But is seems that there are some things missing for that plattform: > > Where do I get as/arm-freebsd/targ-cpu.h etc. from? > > Thanks, Helmut > Hi Helmut, To be able to build a cross-compiler, you need to apply the following patches : http://people.FreeBSD.org/~cognet/contrib-arm.diff http://people.FreeBSD.org/~cognet/gnu-arm.diff Cheers, Olivier From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 15:14:14 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 03C1816A402 for ; Fri, 28 Apr 2006 15:14:14 +0000 (UTC) (envelope-from jumper99@gmx.de) Received: from mail.gmx.net (mail.gmx.net [213.165.64.20]) by mx1.FreeBSD.org (Postfix) with SMTP id 304CE43D46 for ; Fri, 28 Apr 2006 15:14:13 +0000 (GMT) (envelope-from jumper99@gmx.de) Received: (qmail invoked by alias); 28 Apr 2006 15:14:12 -0000 Received: from unknown (EHLO wsa096) [193.101.155.96] by mail.gmx.net (mp033) with SMTP; 28 Apr 2006 17:14:12 +0200 X-Authenticated: #682707 Message-ID: <013c01c66ad6$67a7e5c0$609b65c1@vpe.de> From: "Helmut Schneider" To: "Olivier Houchard" References: <00e101c66acb$526b8000$609b65c1@vpe.de> <20060428144612.GA31797@ci0.org> Date: Fri, 28 Apr 2006 17:14:10 +0200 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2869 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2869 X-Y-GMX-Trusted: 0 Cc: freebsd-arch@freebsd.org Subject: Re: targ-cpu.h X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 15:14:14 -0000 From: "Olivier Houchard" > On Fri, Apr 28, 2006 at 03:54:50PM +0200, Helmut Schneider wrote: >> Hi, >> >> please forgive my noob question. I have a new CPU (AMD64) and I >> successfully played around with cross compiling to i386. >> Now, a collegue of mine asked me if it is also possible to do this with >> ARM. But is seems that there are some things missing for that plattform: >> >> Where do I get as/arm-freebsd/targ-cpu.h etc. from? > > To be able to build a cross-compiler, you need to apply the following > patches : http://people.FreeBSD.org/~cognet/contrib-arm.diff > http://people.FreeBSD.org/~cognet/gnu-arm.diff Excellent! As it is still compiling I assume this is going to work. Thank you very much! From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 22:20:02 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 927D416A403; Fri, 28 Apr 2006 22:20:02 +0000 (UTC) (envelope-from cperciva@freebsd.org) Received: from pd2mo3so.prod.shaw.ca (shawidc-mo1.cg.shawcable.net [24.71.223.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2E6BD43D48; Fri, 28 Apr 2006 22:20:02 +0000 (GMT) (envelope-from cperciva@freebsd.org) Received: from pd4mr8so.prod.shaw.ca (pd4mr8so-qfe3.prod.shaw.ca [10.0.141.101]) by l-daemon (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00HP4E1DT700@l-daemon>; Fri, 28 Apr 2006 16:20:01 -0600 (MDT) Received: from pn2ml6so.prod.shaw.ca ([10.0.121.150]) by pd4mr8so.prod.shaw.ca (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00LMAE1DUYK0@pd4mr8so.prod.shaw.ca>; Fri, 28 Apr 2006 16:20:01 -0600 (MDT) Received: from [192.168.0.60] ([24.82.18.31]) by l-daemon (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00J8KE1COUY0@l-daemon>; Fri, 28 Apr 2006 16:20:01 -0600 (MDT) Date: Fri, 28 Apr 2006 15:20:00 -0700 From: Colin Percival In-reply-to: <002401c66b0a$44c230e0$01655050@jersey> To: freebsd-current@freebsd.org, freebsd-arch@freebsd.org Message-id: <44529510.6030704@freebsd.org> MIME-version: 1.0 Content-type: text/plain; charset=KOI8-R Content-transfer-encoding: 7bit X-Enigmail-Version: 0.94.0.0 References: <002401c66b0a$44c230e0$01655050@jersey> User-Agent: Thunderbird 1.5 (X11/20060416) Cc: Michael Bushkov Subject: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 22:20:02 -0000 Michael Bushkov wrote: > - getservXXX() functions now work through nsdispatch(). "files", "nis" > and "compat" nsswitch sources are implemeted. "sources" and > "sources_compat" databases can now be specified in nsswitch.conf. The > idea of "services_compat" database is the same as for "passwd_compat" > and "group_compat": if we find "+" in /etc/services, we'll substitute it > with information received from the "services_compat" database's source. Now that searching through a large /etc/services file is no longer a performance bottleneck, I intend to merge most of IANA's port assignment list into our /etc/services some time in mid-May. If anyone objects to this, please let me know now. Colin Percival From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 22:26:30 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3BF5B16A41A; Fri, 28 Apr 2006 22:26:30 +0000 (UTC) (envelope-from ceri@submonkey.net) Received: from shrike.submonkey.net (cpc2-cdif2-0-0-cust107.cdif.cable.ntl.com [81.104.168.108]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5367E43D5D; Fri, 28 Apr 2006 22:26:29 +0000 (GMT) (envelope-from ceri@submonkey.net) Received: from ceri by shrike.submonkey.net with local (Exim 4.61 (FreeBSD)) (envelope-from ) id 1FZbPv-000FVF-PU; Fri, 28 Apr 2006 23:26:27 +0100 Date: Fri, 28 Apr 2006 23:26:27 +0100 From: Ceri Davies To: Colin Percival Message-ID: <20060428222627.GI51777@submonkey.net> Mail-Followup-To: Ceri Davies , Colin Percival , freebsd-current@freebsd.org, freebsd-arch@freebsd.org, Michael Bushkov References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="84ND8YJRMFlzkrP4" Content-Disposition: inline In-Reply-To: <44529510.6030704@freebsd.org> X-PGP: finger ceri@FreeBSD.org User-Agent: Mutt/1.5.11 Sender: Ceri Davies Cc: freebsd-current@freebsd.org, Michael Bushkov , freebsd-arch@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 22:26:30 -0000 --84ND8YJRMFlzkrP4 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Apr 28, 2006 at 03:20:00PM -0700, Colin Percival wrote: > Michael Bushkov wrote: > > - getservXXX() functions now work through nsdispatch(). "files", "nis" > > and "compat" nsswitch sources are implemeted. "sources" and > > "sources_compat" databases can now be specified in nsswitch.conf. The > > idea of "services_compat" database is the same as for "passwd_compat" > > and "group_compat": if we find "+" in /etc/services, we'll substitute it > > with information received from the "services_compat" database's source. >=20 > Now that searching through a large /etc/services file is no longer a > performance bottleneck, I intend to merge most of IANA's port assignment > list into our /etc/services some time in mid-May. >=20 > If anyone objects to this, please let me know now. Would it be wise to wait until cached is enabled by default? Ceri --=20 That must be wonderful! I don't understand it at all. -- Moliere --84ND8YJRMFlzkrP4 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (FreeBSD) iD8DBQFEUpaTocfcwTS3JF8RAq8vAKC+ZWMZK30Tqqbhf5OHsT9TSU0+DACfYL4t ibE/M/trFfaJuUEnoH2STs8= =GiGV -----END PGP SIGNATURE----- --84ND8YJRMFlzkrP4-- From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 22:35:28 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 070AB16A403; Fri, 28 Apr 2006 22:35:28 +0000 (UTC) (envelope-from cperciva@freebsd.org) Received: from pd5mo2so.prod.shaw.ca (shawidc-mo1.cg.shawcable.net [24.71.223.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7728143D46; Fri, 28 Apr 2006 22:35:27 +0000 (GMT) (envelope-from cperciva@freebsd.org) Received: from pd4mr4so.prod.shaw.ca (pd4mr4so-qfe3.prod.shaw.ca [10.0.141.215]) by l-daemon (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00LFMER3JD90@l-daemon>; Fri, 28 Apr 2006 16:35:27 -0600 (MDT) Received: from pn2ml6so.prod.shaw.ca ([10.0.121.150]) by pd4mr4so.prod.shaw.ca (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00MW6ER3Y4N0@pd4mr4so.prod.shaw.ca>; Fri, 28 Apr 2006 16:35:27 -0600 (MDT) Received: from [192.168.0.60] ([24.82.18.31]) by l-daemon (Sun ONE Messaging Server 6.0 HotFix 1.01 (built Mar 15 2004)) with ESMTP id <0IYG00DT4ER1AWE0@l-daemon>; Fri, 28 Apr 2006 16:35:27 -0600 (MDT) Date: Fri, 28 Apr 2006 15:35:25 -0700 From: Colin Percival In-reply-to: <20060428222627.GI51777@submonkey.net> To: Ceri Davies , freebsd-current@freebsd.org, freebsd-arch@freebsd.org Message-id: <445298AD.7030400@freebsd.org> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 7bit X-Enigmail-Version: 0.94.0.0 References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> User-Agent: Thunderbird 1.5 (X11/20060416) Cc: Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 22:35:28 -0000 Ceri Davies wrote: > On Fri, Apr 28, 2006 at 03:20:00PM -0700, Colin Percival wrote: >> Now that searching through a large /etc/services file is no longer a >> performance bottleneck, I intend to merge most of IANA's port assignment >> list into our /etc/services some time in mid-May. >> >> If anyone objects to this, please let me know now. > > Would it be wise to wait until cached is enabled by default? My (perhaps mistaken) impression was that cached was going to be enabled by default soon in HEAD, and at very least long before 7.0-RELEASE. I wasn't planning on MFCing the /etc/services update until cached was enabled by default in RELENG_6. If there's a significant reason why people running HEAD would not want to turn cached on, I'll certainly reconsider my plans here. Colin Percival From owner-freebsd-arch@FreeBSD.ORG Fri Apr 28 22:49:51 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0765316A402; Fri, 28 Apr 2006 22:49:51 +0000 (UTC) (envelope-from keramida@ceid.upatras.gr) Received: from igloo.linux.gr (igloo.linux.gr [62.1.205.36]) by mx1.FreeBSD.org (Postfix) with ESMTP id 078E943D46; Fri, 28 Apr 2006 22:49:49 +0000 (GMT) (envelope-from keramida@ceid.upatras.gr) Received: from gothmog.pc (aris.bedc.ondsl.gr [62.103.39.226]) (authenticated bits=128) by igloo.linux.gr (8.13.6/8.13.6/Debian-1) with ESMTP id k3SMnLJf024040 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Sat, 29 Apr 2006 01:49:31 +0300 Received: from gothmog.pc (gothmog [127.0.0.1]) by gothmog.pc (8.13.6/8.13.6) with ESMTP id k3SMnOks010535; Sat, 29 Apr 2006 01:49:24 +0300 (EEST) (envelope-from keramida@ceid.upatras.gr) Received: (from giorgos@localhost) by gothmog.pc (8.13.6/8.13.6/Submit) id k3SMnG2U010534; Sat, 29 Apr 2006 01:49:16 +0300 (EEST) (envelope-from keramida@ceid.upatras.gr) Date: Sat, 29 Apr 2006 01:49:16 +0300 From: Giorgos Keramidas To: Colin Percival Message-ID: <20060428224916.GA10513@gothmog.pc> References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> <445298AD.7030400@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <445298AD.7030400@freebsd.org> X-Hellug-MailScanner: Found to be clean X-Hellug-MailScanner-SpamCheck: not spam, SpamAssassin (score=-3.393, required 5, autolearn=not spam, ALL_TRUSTED -1.80, AWL 0.81, BAYES_00 -2.60, DNS_FROM_RFC_ABUSE 0.20) X-Hellug-MailScanner-From: keramida@ceid.upatras.gr X-Spam-Status: No Cc: freebsd-arch@freebsd.org, Ceri Davies , freebsd-current@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Apr 2006 22:49:51 -0000 On 2006-04-28 15:35, Colin Percival wrote: > Ceri Davies wrote: > > Would it be wise to wait until cached is enabled by default? > > My (perhaps mistaken) impression was that cached was going to > be enabled by default soon in HEAD, and at very least long > before 7.0-RELEASE. I wasn't planning on MFCing the > /etc/services update until cached was enabled by default in > RELENG_6. > > If there's a significant reason why people running HEAD would > not want to turn cached on, I'll certainly reconsider my plans > here. I think turning it on by default will give it more visibility and testing on HEAD, so it's probably good to do so. Running CURRENT and having new features shouldn't be a very big surprise for most people I guess :) From owner-freebsd-arch@FreeBSD.ORG Sat Apr 29 04:03:44 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 511E416A402; Sat, 29 Apr 2006 04:03:44 +0000 (UTC) (envelope-from ume@mahoroba.org) Received: from ameno.mahoroba.org (gw4.mahoroba.org [218.45.22.175]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8843143D49; Sat, 29 Apr 2006 04:03:43 +0000 (GMT) (envelope-from ume@mahoroba.org) Received: from kasuga.mahoroba.org (IDENT:lUkiGB36NSbZujPq7cFXQMGEMidk+2VWthMkmfQ6sdH/JLpHUbZW121qJGDvb/st@kasuga-iwi.mahoroba.org [IPv6:3ffe:501:185b:8010:212:f0ff:fe52:6ac]) (user=ume mech=CRAM-MD5 bits=0) by ameno.mahoroba.org (8.13.6/8.13.6) with ESMTP/inet6 id k3T43c3B020782 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sat, 29 Apr 2006 13:03:38 +0900 (JST) (envelope-from ume@mahoroba.org) Date: Sat, 29 Apr 2006 13:03:38 +0900 Message-ID: From: Hajimu UMEMOTO To: Colin Percival In-Reply-To: <445298AD.7030400@freebsd.org> References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> <445298AD.7030400@freebsd.org> User-Agent: xcite1.38> Wanderlust/2.14.0 (Africa) SEMI/1.14.6 (Maruoka) FLIM/1.14.8 (=?ISO-8859-4?Q?Shij=F2?=) APEL/10.6 Emacs/22.0.50 (i386-unknown-freebsd6.1) MULE/5.0 (SAKAKI) X-Operating-System: FreeBSD 6.1-RC X-PGP-Key: http://www.imasy.or.jp/~ume/publickey.asc X-PGP-Fingerprint: 1F00 0B9E 2164 70FC 6DC5 BF5F 04E9 F086 BF90 71FE Organization: Internet Mutual Aid Society, YOKOHAMA MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII X-Greylist: Sender succeeded SMTP AUTH authentication, not delayed by milter-greylist-2.1.4 (ameno.mahoroba.org [IPv6:3ffe:501:185b:8010::1]); Sat, 29 Apr 2006 13:03:39 +0900 (JST) X-Virus-Scanned: by amavisd-new X-Virus-Status: Clean X-Spam-Status: No, score=-2.6 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.1 X-Spam-Checker-Version: SpamAssassin 3.1.1 (2006-03-10) on ameno.mahoroba.org Cc: freebsd-arch@freebsd.org, Ceri Davies , freebsd-current@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 29 Apr 2006 04:03:44 -0000 Hi, >>>>> On Fri, 28 Apr 2006 15:35:25 -0700 >>>>> Colin Percival said: cperciva> Ceri Davies wrote: > On Fri, Apr 28, 2006 at 03:20:00PM -0700, Colin Percival wrote: >> Now that searching through a large /etc/services file is no longer a >> performance bottleneck, I intend to merge most of IANA's port assignment >> list into our /etc/services some time in mid-May. >> >> If anyone objects to this, please let me know now. > > Would it be wise to wait until cached is enabled by default? cperciva> My (perhaps mistaken) impression was that cached was going to be cperciva> enabled by default soon in HEAD, and at very least long before cperciva> 7.0-RELEASE. I wasn't planning on MFCing the /etc/services update cperciva> until cached was enabled by default in RELENG_6. Though I committed it but not enabled by dafault, I think it is better to enable it by default at least for testing. Okay, I'll enable it by default. cperciva> If there's a significant reason why people running HEAD would not cperciva> want to turn cached on, I'll certainly reconsider my plans here. We need to change /etc/nsswitch.conf. /etc/nsswitch.conf is not installed, and generated by /etc/rc.d/nsswitch. So, mergemaster(8) doesn't care of it. I worry about this issue, bit. Sincerely, -- Hajimu UMEMOTO @ Internet Mutual Aid Society Yokohama, Japan ume@mahoroba.org ume@{,jp.}FreeBSD.org http://www.imasy.org/~ume/ From owner-freebsd-arch@FreeBSD.ORG Sat Apr 29 04:10:07 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B545216A401; Sat, 29 Apr 2006 04:10:07 +0000 (UTC) (envelope-from bushman@rsu.ru) Received: from mail.r61.net (mail.r61.net [195.208.245.249]) by mx1.FreeBSD.org (Postfix) with ESMTP id 55BB043D55; Sat, 29 Apr 2006 04:10:06 +0000 (GMT) (envelope-from bushman@rsu.ru) Received: from jersey (p232.mp119.aaanet.ru [80.80.119.232]) (authenticated bits=0) by mail.r61.net (8.13.6/8.13.6) with ESMTP id k3T49eCq044299 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Sat, 29 Apr 2006 08:09:57 +0400 (MSD) (envelope-from bushman@rsu.ru) Message-ID: <001801c66b42$d48d4740$e8775050@jersey> From: "Michael Bushkov" To: "Ceri Davies" , "Colin Percival" References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> Date: Sat, 29 Apr 2006 08:09:35 +0400 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-Virus-Scanned: ClamAV version 0.88.1, clamav-milter version 0.88.1 on asterix.r61.net X-Virus-Status: Clean Cc: freebsd-current@freebsd.org, freebsd-arch@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 29 Apr 2006 04:10:08 -0000 Hi! Colin Percival wrote: >> >> Now that searching through a large /etc/services file is no longer a >> performance bottleneck, I intend to merge most of IANA's port assignment >> list into our /etc/services some time in mid-May. >> >> If anyone objects to this, please let me know now. > > Would it be wise to wait until cached is enabled by default? Yes - and caching for services should be turned on in nsswitch.conf by default too. I think, that "perform-actual-lookups" mode should also be enabled for "services" by default in cached.conf - so that the cache for services information would be the same for all users. Besides, as services information changes extremely rarely, I'd also suggest putting greater TTL values for "services" in the cached.conf. With best regards, Michael Bushkov From owner-freebsd-arch@FreeBSD.ORG Sat Apr 29 09:31:06 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B1C3B16A401; Sat, 29 Apr 2006 09:31:06 +0000 (UTC) (envelope-from matteo@freebsd.org) Received: from vsmtp4.tin.it (vsmtp4.tin.it [212.216.176.224]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9773243D6D; Sat, 29 Apr 2006 09:31:01 +0000 (GMT) (envelope-from matteo@freebsd.org) Received: from kaiser.sig11.org (82.50.119.228) by vsmtp4.tin.it (7.2.072.1) id 44469C83008D6998; Sat, 29 Apr 2006 11:32:21 +0200 Received: by kaiser.sig11.org (Postfix, from userid 1000) id 3943260D3; Sat, 29 Apr 2006 11:31:00 +0200 (CEST) Date: Sat, 29 Apr 2006 11:31:00 +0200 From: Matteo Riondato To: Hajimu UMEMOTO Message-ID: <20060429093100.GD914@kaiser.sig11.org> Mail-Followup-To: Matteo Riondato , Hajimu UMEMOTO , Colin Percival , freebsd-arch@freebsd.org, Ceri Davies , freebsd-current@freebsd.org References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> <445298AD.7030400@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.11 Cc: freebsd-current@freebsd.org, Ceri Davies , Colin Percival , freebsd-arch@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 29 Apr 2006 09:31:06 -0000 On Sat, Apr 29, 2006 at 01:03:38PM +0900, Hajimu UMEMOTO wrote: > >>>>> Colin Percival said: > cperciva> If there's a significant reason why people running HEAD would not > cperciva> want to turn cached on, I'll certainly reconsider my plans here. > > We need to change /etc/nsswitch.conf. /etc/nsswitch.conf is not > installed, and generated by /etc/rc.d/nsswitch. So, mergemaster(8) > doesn't care of it. I worry about this issue, bit. An entry in UPDATING will probably be enough, IMHO. Something like: "cached $SHORT_DESCRIPTION_OF_CACHED is now turned on by default. To use it you need to update your /etc/nsswitch.conf . This can be done with the following commands: # rm /etc/nsswitch.conf # /etc/rc.d/nsswitch restart " Or something similar.. Thanks for your work. Best Regards -- Matteo Riondato FreeBSD Committer (http://www.freebsd.org) G.U.F.I. Staff Member (http://www.gufi.org) FreeSBIE Developer (http://www.freesbie.org) From owner-freebsd-arch@FreeBSD.ORG Sat Apr 29 10:02:41 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0447216A408; Sat, 29 Apr 2006 10:02:41 +0000 (UTC) (envelope-from ceri@submonkey.net) Received: from shrike.submonkey.net (cpc2-cdif2-0-0-cust107.cdif.cable.ntl.com [81.104.168.108]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3088443D53; Sat, 29 Apr 2006 10:02:39 +0000 (GMT) (envelope-from ceri@submonkey.net) Received: from ceri by shrike.submonkey.net with local (Exim 4.61 (FreeBSD)) (envelope-from ) id 1FZmHd-000751-5c; Sat, 29 Apr 2006 11:02:37 +0100 Date: Sat, 29 Apr 2006 11:02:37 +0100 From: Ceri Davies To: Michael Bushkov Message-ID: <20060429100236.GJ51777@submonkey.net> Mail-Followup-To: Ceri Davies , Michael Bushkov , Colin Percival , freebsd-current@freebsd.org, freebsd-arch@freebsd.org References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> <001801c66b42$d48d4740$e8775050@jersey> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="/QKKmeG/X/bPShih" Content-Disposition: inline In-Reply-To: <001801c66b42$d48d4740$e8775050@jersey> X-PGP: finger ceri@FreeBSD.org User-Agent: Mutt/1.5.11 Sender: Ceri Davies Cc: freebsd-current@freebsd.org, Colin Percival , freebsd-arch@freebsd.org Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 29 Apr 2006 10:02:41 -0000 --/QKKmeG/X/bPShih Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Apr 29, 2006 at 08:09:35AM +0400, Michael Bushkov wrote: > Hi! >=20 > Colin Percival wrote: > >> > >>Now that searching through a large /etc/services file is no longer a > >>performance bottleneck, I intend to merge most of IANA's port assignment > >>list into our /etc/services some time in mid-May. > >> > >>If anyone objects to this, please let me know now. > > > >Would it be wise to wait until cached is enabled by default? >=20 > Yes - and caching for services should be turned on in nsswitch.conf by=20 > default too. Good point. > I think, that "perform-actual-lookups" mode should also be enabled for=20 > "services" by default in cached.conf - so that the cache for services=20 > information would be the same for all users. So "perform-actual-lookups" essentially creates a system-wide cache? > Besides, as services=20 > information changes extremely rarely, I'd also suggest putting greater TT= L=20 > values for "services" in the cached.conf. Yes, the services cache could always be flushed by the superuser after changing the services data source. Ceri --=20 That must be wonderful! I don't understand it at all. -- Moliere --/QKKmeG/X/bPShih Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (FreeBSD) iD8DBQFEUzm8ocfcwTS3JF8RAusXAJ41L7hyi+MxQKGd1Lx+330/sLT7XACgoOBE PPT2KCnjKx1FkeNVfAZ480I= =InY5 -----END PGP SIGNATURE----- --/QKKmeG/X/bPShih-- From owner-freebsd-arch@FreeBSD.ORG Sat Apr 29 16:44:57 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5027116A410; Sat, 29 Apr 2006 16:44:57 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.FreeBSD.org (Postfix) with ESMTP id AD6ED43D46; Sat, 29 Apr 2006 16:44:56 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 62B4646B04; Sat, 29 Apr 2006 12:44:55 -0400 (EDT) Date: Sat, 29 Apr 2006 17:44:55 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Ceri Davies In-Reply-To: <20060429100236.GJ51777@submonkey.net> Message-ID: <20060429174129.H11416@fledge.watson.org> References: <002401c66b0a$44c230e0$01655050@jersey> <44529510.6030704@freebsd.org> <20060428222627.GI51777@submonkey.net> <001801c66b42$d48d4740$e8775050@jersey> <20060429100236.GJ51777@submonkey.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-arch@freebsd.org, freebsd-current@freebsd.org, Colin Percival , Michael Bushkov Subject: Re: [HEADS UP] upcoming /etc/services updating X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 29 Apr 2006 16:44:57 -0000 On Sat, 29 Apr 2006, Ceri Davies wrote: >> I think, that "perform-actual-lookups" mode should also be enabled for >> "services" by default in cached.conf - so that the cache for services >> information would be the same for all users. > > So "perform-actual-lookups" essentially creates a system-wide cache? > >> Besides, as services information changes extremely rarely, I'd also >> suggest putting greater TTL values for "services" in the cached.conf. > > Yes, the services cache could always be flushed by the superuser after > changing the services data source. Very few sites use distributed services databases (at least, that I've ever seen). Performing stat() on /etc/services to check the last modification date is pretty light-weight, and probably worth doing. What you don't want is someone modifying /etc/services, restarting the daemon immediately, and having it fail due to an added service not being found. Likewise for other local databases. BTW, since this is in the context of significantly increasing the size of the services database, have we: (1) Measured what impact adding the cache daemon for local databases has? Specifically, does adding the cache daemon for a database like /etc/services, /etc/passwd, etc, improve performance, or add overhead? (2) Looked at adding /etc/services.db, similar to the other compiled database pieces, in order to improve lookup times for very large tables. This change would be orthoganal to a cache daemon. Robert N M Watson