From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 01:31:00 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 29936106566B for ; Thu, 18 Nov 2010 01:31:00 +0000 (UTC) (envelope-from grarpamp@gmail.com) Received: from mail-px0-f182.google.com (mail-px0-f182.google.com [209.85.212.182]) by mx1.freebsd.org (Postfix) with ESMTP id 02F6E8FC16 for ; Thu, 18 Nov 2010 01:30:58 +0000 (UTC) Received: by pxi1 with SMTP id 1so719775pxi.13 for ; Wed, 17 Nov 2010 17:30:58 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:date:message-id :subject:from:to:content-type; bh=4aIj6L29M/ldNVqHYLfvA9YdnO5mkFcIOvApDNj32+o=; b=Camz6AzBJj3sodFhWXJRXhspH5iRAU1dan6CDMsprd7HZwZkDnwCNo6Ic2sSjFMCgW 8iv71/eDes7+GkcZ1JCtJ6GAQN3AMin07nuD5XzD4V3qp35A010YXidHYgx1xdSbZsqC j8KhYpy7AfmMFfZgSPGgpf1lWaR1vBikNlyuY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=hX164Tk1+hn9Hc/sYzCo5pJ2uc0Ld2Y4Mvu9OwzhoTXj/tbB4RBD7GZaEGCvCgZNX8 qFUUxN9PxnpHRzgO5b8Llv09878rEffQxGOm4uHlnP5B0wCprHVomKAnFNbBIWPWrAdw JOpQcMU8JfWA5RY4rnK4fQXRff1U2Sq1Fm/aI= MIME-Version: 1.0 Received: by 10.142.187.10 with SMTP id k10mr8892042wff.230.1290043858692; Wed, 17 Nov 2010 17:30:58 -0800 (PST) Received: by 10.142.178.2 with HTTP; Wed, 17 Nov 2010 17:30:58 -0800 (PST) Date: Wed, 17 Nov 2010 20:30:58 -0500 Message-ID: From: grarpamp To: freebsd-performance@freebsd.org Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Thu, 18 Nov 2010 05:25:22 +0000 Subject: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 01:31:00 -0000 Just documenting regarding interactive performance things. This one's from Linux. http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 11:23:19 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 853AA106566C; Thu, 18 Nov 2010 11:23:19 +0000 (UTC) (envelope-from ohartman@zedat.fu-berlin.de) Received: from outpost1.zedat.fu-berlin.de (outpost1.zedat.fu-berlin.de [130.133.4.66]) by mx1.freebsd.org (Postfix) with ESMTP id 3E73E8FC12; Thu, 18 Nov 2010 11:23:19 +0000 (UTC) Received: from inpost2.zedat.fu-berlin.de ([130.133.4.69]) by outpost1.zedat.fu-berlin.de (Exim 4.69) with esmtp (envelope-from ) id <1PJ2IA-0000Ju-93>; Thu, 18 Nov 2010 12:04:38 +0100 Received: from telesto.geoinf.fu-berlin.de ([130.133.86.198]) by inpost2.zedat.fu-berlin.de (Exim 4.69) with esmtpsa (envelope-from ) id <1PJ2IA-0000XK-4R>; Thu, 18 Nov 2010 12:04:38 +0100 Message-ID: <4CE50849.106@zedat.fu-berlin.de> Date: Thu, 18 Nov 2010 12:04:41 +0100 From: "O. Hartmann" Organization: Freie =?ISO-8859-1?Q?Universit=E4t_Berlin?= User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Thunderbird/3.1.6 MIME-Version: 1.0 To: freebsd-performance@freebsd.org References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 130.133.86.198 Cc: FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 11:23:19 -0000 On 11/18/10 02:30, grarpamp wrote: > Just documenting regarding interactive performance things. > This one's from Linux. > > http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 Well, it would be nice to have those improvements in FreeBSD, but I doubt this will make it in due time to FreeBSD's kernel. From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 13:06:50 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CC828106566B for ; Thu, 18 Nov 2010 13:06:50 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id 1E3E48FC0C for ; Thu, 18 Nov 2010 13:06:49 +0000 (UTC) Received: from odyssey.starpoint.kiev.ua (alpha-e.starpoint.kiev.ua [212.40.38.101]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id OAA23072; Thu, 18 Nov 2010 14:52:07 +0200 (EET) (envelope-from avg@freebsd.org) Message-ID: <4CE52177.3020306@freebsd.org> Date: Thu, 18 Nov 2010 14:52:07 +0200 From: Andriy Gapon User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: "O. Hartmann" References: <4CE50849.106@zedat.fu-berlin.de> In-Reply-To: <4CE50849.106@zedat.fu-berlin.de> X-Enigmail-Version: 1.1.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Thu, 18 Nov 2010 13:17:13 +0000 Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 13:06:50 -0000 on 18/11/2010 13:04 O. Hartmann said the following: > On 11/18/10 02:30, grarpamp wrote: >> Just documenting regarding interactive performance things. >> This one's from Linux. >> >> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 > > Well, > it would be nice to have those improvements in FreeBSD, but I doubt this will make > it in due time to FreeBSD's kernel. Well, I think that those improvements apply only to a very specific usage pattern and are greatly over-hyped. P.S. What is the due time, BTW? -- Andriy Gapon From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 18:45:33 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 04BDD106564A; Thu, 18 Nov 2010 18:45:33 +0000 (UTC) (envelope-from fullermd@over-yonder.net) Received: from thyme.infocus-llc.com (server.infocus-llc.com [206.156.254.44]) by mx1.freebsd.org (Postfix) with ESMTP id CADCE8FC0C; Thu, 18 Nov 2010 18:45:32 +0000 (UTC) Received: from draco.over-yonder.net (c-75-64-226-141.hsd1.ms.comcast.net [75.64.226.141]) (using TLSv1 with cipher ADH-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by thyme.infocus-llc.com (Postfix) with ESMTPSA id 003EC37B444; Thu, 18 Nov 2010 12:28:52 -0600 (CST) Received: by draco.over-yonder.net (Postfix, from userid 100) id 67BA761C42; Thu, 18 Nov 2010 12:28:52 -0600 (CST) Date: Thu, 18 Nov 2010 12:28:52 -0600 From: "Matthew D. Fuller" To: Alexander Best Message-ID: <20101118182852.GR63683@over-yonder.net> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118182324.GA36312@freebsd.org> X-Editor: vi X-OS: FreeBSD User-Agent: Mutt/1.5.20-fullermd.4 (2009-06-14) X-Virus-Scanned: clamav-milter 0.96.4 at thyme.infocus-llc.com X-Virus-Status: Clean Cc: "O. Hartmann" , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 18:45:33 -0000 On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of Alexander Best, and lo! it spake thus: > > judging from the videos the changes are having a huge impact imo. Well, my (admittedly limited, and certainly anecdotal) experience is that Linux's interactive response when under heavy load was always much worse than FreeBSD's. So maybe that's just them catching up to where we already are ;) -- Matthew Fuller (MF4839) | fullermd@over-yonder.net Systems/Network Administrator | http://www.over-yonder.net/~fullermd/ On the Internet, nobody can hear you scream. From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 18:56:35 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 1233) id ABCF7106566C; Thu, 18 Nov 2010 18:56:35 +0000 (UTC) Date: Thu, 18 Nov 2010 18:56:35 +0000 From: Alexander Best To: "Matthew D. Fuller" Message-ID: <20101118185635.GA43706@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118182852.GR63683@over-yonder.net> X-Mailman-Approved-At: Thu, 18 Nov 2010 19:03:48 +0000 Cc: "O. Hartmann" , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 18:56:35 -0000 On Thu Nov 18 10, Matthew D. Fuller wrote: > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > Alexander Best, and lo! it spake thus: > > > > judging from the videos the changes are having a huge impact imo. > > Well, my (admittedly limited, and certainly anecdotal) experience is > that Linux's interactive response when under heavy load was always > much worse than FreeBSD's. So maybe that's just them catching up to > where we already are ;) well...i tried playing back a 1080p vide files while doing `make -j64 buildkernel` and FreeBSD's interactivity seems far from perfect. it might be possible that linux'es interactivity was worse than freebsd's, but still this patch should be evaluated for freebsd. in this particular case it seems linux now does better than freebsd. cheers. alex > > > -- > Matthew Fuller (MF4839) | fullermd@over-yonder.net > Systems/Network Administrator | http://www.over-yonder.net/~fullermd/ > On the Internet, nobody can hear you scream. -- a13x From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 18:57:18 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CD9BF1065694; Thu, 18 Nov 2010 18:57:18 +0000 (UTC) (envelope-from rfarmer@predatorlabs.net) Received: from mail-qw0-f54.google.com (mail-qw0-f54.google.com [209.85.216.54]) by mx1.freebsd.org (Postfix) with ESMTP id 7606A8FC12; Thu, 18 Nov 2010 18:57:18 +0000 (UTC) Received: by qwd6 with SMTP id 6so64602qwd.13 for ; Thu, 18 Nov 2010 10:57:17 -0800 (PST) MIME-Version: 1.0 Received: by 10.224.196.67 with SMTP id ef3mr878967qab.160.1290105769592; Thu, 18 Nov 2010 10:42:49 -0800 (PST) Received: by 10.220.16.199 with HTTP; Thu, 18 Nov 2010 10:42:49 -0800 (PST) X-Originating-IP: [128.95.133.115] In-Reply-To: <39F4F32E-A30C-47F3-AD69-F7777A7E30A8@mac.com> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <39F4F32E-A30C-47F3-AD69-F7777A7E30A8@mac.com> Date: Thu, 18 Nov 2010 10:42:49 -0800 Message-ID: From: Rob Farmer To: freebsd-performance@freebsd.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Mailman-Approved-At: Thu, 18 Nov 2010 19:04:02 +0000 Cc: Alexander Best , FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 18:57:18 -0000 On Thu, Nov 18, 2010 at 10:39, Chuck Swiger wrote: > Frankly, I'm also turned off by the attempt to popup a full page ad in ad= dition to the rest of the advertising content which surrounds what is nomin= ally supposed to be the real content. =A0That doesn't mean there is anythin= g wrong with the patch or the notion of adjusting the scheduler, but I don'= t see any value added from these phoronix.com links. Most stuff on Phoronix is of dubious value, and they have outright lied about stuff in the past, in order to drum up advertising business (such as Steam on Linux, which Value has consistently said isn't happening). --=20 Rob Farmer From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 19:21:46 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2B2CF106566C; Thu, 18 Nov 2010 19:21:46 +0000 (UTC) (envelope-from lwindschuh@googlemail.com) Received: from mail-gx0-f182.google.com (mail-gx0-f182.google.com [209.85.161.182]) by mx1.freebsd.org (Postfix) with ESMTP id A2FC28FC0A; Thu, 18 Nov 2010 19:21:45 +0000 (UTC) Received: by gxk9 with SMTP id 9so2182912gxk.13 for ; Thu, 18 Nov 2010 11:21:44 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=n2z2pECZ4dz5Pikhase/txALObjsijI69ae9TSt7c/o=; b=VX3EyXddnk6u7Mz6Go7lq4XcC88RZuAt7xFtqz0FaOoD1SEt1QQOPNj4ynK0PdUlUY j5TX6BeFzXa1vggkiY5t0VfgRZf6W5H9KfoTbdUouv+gk9ccLYCEIHY0uXkvhqu4e+QL 6yCz+gCqIouoPy3yXi8FXnETPoMoCSK9aCoaE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=r7zFMpZlKPl0hSIJafNGmrlsJkGkvb4cL++NrTsYTvPpIqjVSMT291tIuoppJhCxr4 8PcL0Ij0c3sU+PMYq3nrLSR08JoDSeDl0dlrjSxDi44Okmq9gmpVop36Wkc3lJ9BboQo kJARhQgpCU2PKRmgMt16mpZmERoHnKflLcsrI= MIME-Version: 1.0 Received: by 10.42.240.195 with SMTP id lb3mr657505icb.114.1290106505346; Thu, 18 Nov 2010 10:55:05 -0800 (PST) Received: by 10.231.192.139 with HTTP; Thu, 18 Nov 2010 10:55:05 -0800 (PST) In-Reply-To: <4CE52177.3020306@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> Date: Thu, 18 Nov 2010 19:55:05 +0100 Message-ID: From: Lucius Windschuh To: Andriy Gapon Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Thu, 18 Nov 2010 19:37:44 +0000 Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 19:21:46 -0000 2010/11/18 Andriy Gapon : > [Grouping of processes into TTY groups] > > Well, I think that those improvements apply only to a very specific usage pattern > and are greatly over-hyped. But there are serious issue if you use FreeBSD as a desktop OS with SMP and SCHED_ULE, or? Because currently, my machine is barely usable if a compile job with parallelism is running. Movies stutter, Firefox hangs. And even nice -n 20 doesn't do the job in every case, as +20 seems not to be the idle priority anymore?!? And using "idprio 1 $cmd" as a workaround is, well, a kludge. I am not sure if TTY grouping is the right solution, if you look at potentially CPU-intensive GUI applications that all run on the same TTY (or no TTY at all? Same problem). Maybe, we could simply enhance the algorithm that decides if a task is interactive? That would also improve the described situation. Regards, Lucius From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 19:40:16 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 40C16106566B; Thu, 18 Nov 2010 19:40:16 +0000 (UTC) (envelope-from cswiger@mac.com) Received: from asmtpout024.mac.com (asmtpout024.mac.com [17.148.16.99]) by mx1.freebsd.org (Postfix) with ESMTP id 262E98FC12; Thu, 18 Nov 2010 19:40:16 +0000 (UTC) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; CHARSET=US-ASCII Received: from cswiger1.apple.com ([17.209.4.71]) by asmtp024.mac.com (Oracle Communications Messaging Exchange Server 7u4-18.01 64bit (built Jul 15 2010)) with ESMTPSA id <0LC300F0LFU76910@asmtp024.mac.com>; Thu, 18 Nov 2010 10:39:44 -0800 (PST) X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.2.15,1.0.148,0.0.0000 definitions=2010-11-18_08:2010-11-18, 2010-11-18, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 ipscore=0 suspectscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=6.0.2-1010190000 definitions=main-1011180150 From: Chuck Swiger In-reply-to: <20101118182324.GA36312@freebsd.org> Date: Thu, 18 Nov 2010 10:39:43 -0800 Message-id: <39F4F32E-A30C-47F3-AD69-F7777A7E30A8@mac.com> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> To: Alexander Best X-Mailer: Apple Mail (2.1082) Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: freebsd-performance@freebsd.org List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 19:40:16 -0000 On Nov 18, 2010, at 10:23 AM, Alexander Best wrote: > On Thu Nov 18 10, Andriy Gapon wrote: >> on 18/11/2010 13:04 O. Hartmann said the following: >>> On 11/18/10 02:30, grarpamp wrote: >>>> Just documenting regarding interactive performance things. >>>> This one's from Linux. >>>> >>>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 >>> >>> Well, >>> it would be nice to have those improvements in FreeBSD, but I doubt this will make >>> it in due time to FreeBSD's kernel. >> >> Well, I think that those improvements apply only to a very specific usage pattern >> and are greatly over-hyped. > > you think so? judging from the videos the changes are having a huge impact imo. > > cheers. > alex > > btw: i posted a similar thread on freebsd-backers@, but nobody seemed > interested in it. subject line is "sched_autogroup_enabled". I attempted to find reliable benchmarks or even an idea as to what they thought they were measuring, because a sentence like the following: Tests done by Mike show the maximum latency dropping by over ten times and the average latency of the desktop by about 60 times. ...isn't mathematically possible, I'm pretty sure. Frankly, I'm also turned off by the attempt to popup a full page ad in addition to the rest of the advertising content which surrounds what is nominally supposed to be the real content. That doesn't mean there is anything wrong with the patch or the notion of adjusting the scheduler, but I don't see any value added from these phoronix.com links. Regards, -- -Chuck From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 19:44:54 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 137571065693; Thu, 18 Nov 2010 19:44:54 +0000 (UTC) (envelope-from fjwcash@gmail.com) Received: from mail-gy0-f182.google.com (mail-gy0-f182.google.com [209.85.160.182]) by mx1.freebsd.org (Postfix) with ESMTP id 8480A8FC16; Thu, 18 Nov 2010 19:44:53 +0000 (UTC) Received: by gyg13 with SMTP id 13so2209392gyg.13 for ; Thu, 18 Nov 2010 11:44:53 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=1DooHdNPKrh28hk1XLwU+v4woJgksR0RKFN0Bki/nCY=; b=D2ouUGPS+KR4g0PuiRNaJ0KERqTr8ia+KxQxvRBxNq6rZc4iaELBVHfo1AtcuxAjw4 uFmGvN4KhXk0VKqeuKmiKc/ONZRQIkp6by/e2Yz+axrkZfPEkUUapyl6i1tXF4097jw3 9R7tSTbmUOLkX6DdmqjvDhV9fj/tHu5qAiOzU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=aCgDIQDcoFJ6ERhJwppKnKfxjwvsqpLIxJzSVpuEIFOUKFgbBKz2KjML2uJdmILE8U CR21OjW66uJkUEIbvd04fBoE4EoGYH+jnw1heSpmlRIdYCqf7W34y3GgwxlpiI8v9YvF EONxQ1pKod5u9UIR3/LVL8SWaFJO2VnTab/7I= MIME-Version: 1.0 Received: by 10.100.14.8 with SMTP id 8mr766406ann.65.1290107638130; Thu, 18 Nov 2010 11:13:58 -0800 (PST) Received: by 10.90.211.7 with HTTP; Thu, 18 Nov 2010 11:13:57 -0800 (PST) In-Reply-To: <20101118185635.GA43706@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> Date: Thu, 18 Nov 2010 11:13:57 -0800 Message-ID: From: Freddie Cash To: Alexander Best Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cc: FreeBSD Stable , Andriy Gapon , FreeBSD Current , "Matthew D. Fuller" , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 19:44:54 -0000 On Thu, Nov 18, 2010 at 10:56 AM, Alexander Best wrot= e: > On Thu Nov 18 10, Matthew D. Fuller wrote: >> On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of >> Alexander Best, and lo! it spake thus: >> > >> > judging from the videos the changes are having a huge impact imo. >> >> Well, my (admittedly limited, and certainly anecdotal) experience is >> that Linux's interactive response when under heavy load was always >> much worse than FreeBSD's. =C2=A0So maybe that's just them catching up t= o >> where we already are =C2=A0 ;) > > well...i tried playing back a 1080p vide files while doing > `make -j64 buildkernel` and FreeBSD's interactivity seems far from perfec= t. > > it might be possible that linux'es interactivity was worse than freebsd's= , > but still this patch should be evaluated for freebsd. in this particular = case > it seems linux now does better than freebsd. Maybe I'm just a lowly user and don't fully understand the issue, but isn't this the whole reason for having /usr/bin/nice installed? As in, if you don't want your make job to hog resources, then use nice to run it in the background. How does the work on the geom_sched (for I/O scheduling) play into this? Am I missing something fundamental to the issue in question? Also, does this really need to be cross-posted to -current, -hackers, and -performance? --=20 Freddie Cash fjwcash@gmail.com From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 20:20:23 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1320A106564A; Thu, 18 Nov 2010 20:20:23 +0000 (UTC) (envelope-from julian@freebsd.org) Received: from out-0.mx.aerioconnect.net (out-0-16.mx.aerioconnect.net [216.240.47.76]) by mx1.freebsd.org (Postfix) with ESMTP id E49368FC0C; Thu, 18 Nov 2010 20:20:22 +0000 (UTC) Received: from idiom.com (postfix@mx0.idiom.com [216.240.32.160]) by out-0.mx.aerioconnect.net (8.13.8/8.13.8) with ESMTP id oAIKKMB0024055; Thu, 18 Nov 2010 12:20:22 -0800 X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e Received: from julian-mac.elischer.org (h-67-100-89-137.snfccasy.static.covad.net [67.100.89.137]) by idiom.com (Postfix) with ESMTP id 31AEF2D601A; Thu, 18 Nov 2010 12:20:20 -0800 (PST) Message-ID: <4CE58A86.6010406@freebsd.org> Date: Thu, 18 Nov 2010 12:20:22 -0800 From: Julian Elischer User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X 10.4; en-US; rv:1.9.2.12) Gecko/20101027 Thunderbird/3.1.6 MIME-Version: 1.0 To: Lucius Windschuh References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.67 on 216.240.47.51 Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 20:20:23 -0000 On 11/18/10 10:55 AM, Lucius Windschuh wrote: > 2010/11/18 Andriy Gapon: >> [Grouping of processes into TTY groups] >> >> Well, I think that those improvements apply only to a very specific usage pattern >> and are greatly over-hyped. > But there are serious issue if you use FreeBSD as a desktop OS with > SMP and SCHED_ULE, or? > Because currently, my machine is barely usable if a compile job with > parallelism is running. Movies stutter, Firefox hangs. And even nice > -n 20 doesn't do the job in every case, as +20 seems not to be the > idle priority anymore?!? > And using "idprio 1 $cmd" as a workaround is, well, a kludge. > I am not sure if TTY grouping is the right solution, if you look at > potentially CPU-intensive GUI applications that all run on the same > TTY (or no TTY at all? Same problem). > Maybe, we could simply enhance the algorithm that decides if a task is > interactive? That would also improve the described situation. tty grouping is a variant of what we used to have at one stage which is a "kernel schedulable entity group".. KSEG the idea is that all items in a group share some characteristic and some amount of resources. We stripped the KSEG out of the picture because it really complicated the picture. > Regards, > > Lucius > _______________________________________________ > freebsd-current@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org" > From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 20:12:00 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 4704B106566C; Thu, 18 Nov 2010 20:12:00 +0000 (UTC) (envelope-from bruce@cran.org.uk) Received: from muon.cran.org.uk (muon.cran.org.uk [IPv6:2a01:348:0:15:5d59:5c40:0:1]) by mx1.freebsd.org (Postfix) with ESMTP id D5B7E8FC18; Thu, 18 Nov 2010 20:11:59 +0000 (UTC) Received: from muon.cran.org.uk (localhost [127.0.0.1]) by muon.cran.org.uk (Postfix) with ESMTP id 30F65E7203; Thu, 18 Nov 2010 20:11:59 +0000 (GMT) Received: from unknown (client-86-27-40-229.glfd.adsl.virginmedia.com [86.27.40.229]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by muon.cran.org.uk (Postfix) with ESMTPSA; Thu, 18 Nov 2010 20:11:58 +0000 (GMT) Date: Thu, 18 Nov 2010 20:11:58 +0000 From: Bruce Cran To: Lucius Windschuh Message-ID: <20101118201158.000029b5@unknown> In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> X-Mailer: Claws Mail 3.7.4cvs1 (GTK+ 2.16.0; i586-pc-mingw32msvc) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Thu, 18 Nov 2010 20:20:29 +0000 Cc: Alexander Best , freebsd-performance@freebsd.org, Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 20:12:00 -0000 [removed current and stable from Cc list] On Thu, 18 Nov 2010 19:55:05 +0100 Lucius Windschuh wrote: > Because currently, my machine is barely usable if a compile job with > parallelism is running. Movies stutter, Firefox hangs. And even nice > -n 20 doesn't do the job in every case, as +20 seems not to be the > idle priority anymore?!? Have you tried increasing kern.sched.preempt_thresh? According to http://groups.google.com/group/mailing.freebsd.stable/browse_thread/thread/05a39f816fd8acc6/82affa9f195b747d?lnk=raot&fwc=1&pli=1 a good value for desktop use would be 224. -- Bruce Cran From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 20:16:45 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7B4C2106564A; Thu, 18 Nov 2010 20:16:45 +0000 (UTC) (envelope-from bruce@cran.org.uk) Received: from muon.cran.org.uk (muon.cran.org.uk [IPv6:2a01:348:0:15:5d59:5c40:0:1]) by mx1.freebsd.org (Postfix) with ESMTP id 3FD818FC15; Thu, 18 Nov 2010 20:16:45 +0000 (UTC) Received: from muon.cran.org.uk (localhost [127.0.0.1]) by muon.cran.org.uk (Postfix) with ESMTP id 954A9E7203; Thu, 18 Nov 2010 20:16:44 +0000 (GMT) Received: from unknown (client-86-27-40-229.glfd.adsl.virginmedia.com [86.27.40.229]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by muon.cran.org.uk (Postfix) with ESMTPSA; Thu, 18 Nov 2010 20:16:43 +0000 (GMT) Date: Thu, 18 Nov 2010 20:16:44 +0000 From: Bruce Cran To: Lucius Windschuh Message-ID: <20101118201644.00004c3c@unknown> In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> X-Mailer: Claws Mail 3.7.4cvs1 (GTK+ 2.16.0; i586-pc-mingw32msvc) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Thu, 18 Nov 2010 20:30:43 +0000 Cc: Alexander Best , freebsd-performance@freebsd.org, Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 20:16:45 -0000 On Thu, 18 Nov 2010 19:55:05 +0100 Lucius Windschuh wrote: > Because currently, my machine is barely usable if a compile job with > parallelism is running. Movies stutter, Firefox hangs. And even nice > -n 20 doesn't do the job in every case, as +20 seems not to be the > idle priority anymore?!? If you're using UFS, I've found it to be quite a bottleneck when doing parallel IO: I even ran a "svn up" in one terminal and tried to login on another a couple of days ago only to find the motd took over 5 seconds to appear! That may be excessive since I was running a kernel with WITNESS and INVARIANTS, but I've found ZFS to be far better if you want good interactivity when reading/writing to disks. -- Bruce Cran From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 20:22:15 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 38B4C1065693; Thu, 18 Nov 2010 20:22:15 +0000 (UTC) (envelope-from ohartman@mail.zedat.fu-berlin.de) Received: from outpost1.zedat.fu-berlin.de (outpost1.zedat.fu-berlin.de [130.133.4.66]) by mx1.freebsd.org (Postfix) with ESMTP id E4C4E8FC13; Thu, 18 Nov 2010 20:22:14 +0000 (UTC) Received: from inpost2.zedat.fu-berlin.de ([130.133.4.69]) by outpost1.zedat.fu-berlin.de (Exim 4.69) with esmtp (envelope-from ) id <1PJAzl-0004gQ-Fu>; Thu, 18 Nov 2010 21:22:13 +0100 Received: from e178041009.adsl.alicedsl.de ([85.178.41.9] helo=thor.walstatt.dyndns.org) by inpost2.zedat.fu-berlin.de (Exim 4.69) with esmtpsa (envelope-from ) id <1PJAzl-0001Oq-90>; Thu, 18 Nov 2010 21:22:13 +0100 Message-ID: <4CE58AF4.5090304@mail.zedat.fu-berlin.de> Date: Thu, 18 Nov 2010 21:22:12 +0100 From: "O. Hartmann" User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: "Matthew D. Fuller" References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> In-Reply-To: <20101118182852.GR63683@over-yonder.net> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 85.178.41.9 X-Mailman-Approved-At: Thu, 18 Nov 2010 20:30:53 +0000 Cc: FreeBSD Stable , Andriy Gapon , FreeBSD Current , Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 20:22:15 -0000 On 11/18/10 19:28, Matthew D. Fuller wrote: > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > Alexander Best, and lo! it spake thus: >> >> judging from the videos the changes are having a huge impact imo. > > Well, my (admittedly limited, and certainly anecdotal) experience is > that Linux's interactive response when under heavy load was always > much worse than FreeBSD's. So maybe that's just them catching up to > where we already are ;) > > Wishful thinking? From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 20:30:17 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EA26A1065672; Thu, 18 Nov 2010 20:30:17 +0000 (UTC) (envelope-from ohartman@mail.zedat.fu-berlin.de) Received: from outpost1.zedat.fu-berlin.de (outpost1.zedat.fu-berlin.de [130.133.4.66]) by mx1.freebsd.org (Postfix) with ESMTP id A24428FC13; Thu, 18 Nov 2010 20:30:17 +0000 (UTC) Received: from inpost2.zedat.fu-berlin.de ([130.133.4.69]) by outpost1.zedat.fu-berlin.de (Exim 4.69) with esmtp (envelope-from ) id <1PJB7Y-0005kz-SL>; Thu, 18 Nov 2010 21:30:16 +0100 Received: from e178041009.adsl.alicedsl.de ([85.178.41.9] helo=thor.walstatt.dyndns.org) by inpost2.zedat.fu-berlin.de (Exim 4.69) with esmtpsa (envelope-from ) id <1PJB7Y-0001nv-Na>; Thu, 18 Nov 2010 21:30:16 +0100 Message-ID: <4CE58CD8.2000407@mail.zedat.fu-berlin.de> Date: Thu, 18 Nov 2010 21:30:16 +0100 From: "O. Hartmann" User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Lucius Windschuh References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 85.178.41.9 X-Mailman-Approved-At: Thu, 18 Nov 2010 21:09:52 +0000 Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 20:30:18 -0000 On 11/18/10 19:55, Lucius Windschuh wrote: > 2010/11/18 Andriy Gapon: >> [Grouping of processes into TTY groups] >> >> Well, I think that those improvements apply only to a very specific usage pattern >> and are greatly over-hyped. > > But there are serious issue if you use FreeBSD as a desktop OS with > SMP and SCHED_ULE, or? > Because currently, my machine is barely usable if a compile job with > parallelism is running. Movies stutter, Firefox hangs. And even nice > -n 20 doesn't do the job in every case, as +20 seems not to be the > idle priority anymore?!? > And using "idprio 1 $cmd" as a workaround is, well, a kludge. > I am not sure if TTY grouping is the right solution, if you look at > potentially CPU-intensive GUI applications that all run on the same > TTY (or no TTY at all? Same problem). > Maybe, we could simply enhance the algorithm that decides if a task is > interactive? That would also improve the described situation. > > Regards, > > Lucius Stuttering Response, being stuck for over 20 seconds also happens when I start updating the OS' sources via svn. This happens on all boxes, some of them do have 8 cores (ob two CPUs) and plenty of RAM. Heavy disk I/O, doesn't matter on UFS2 or ZFS, also brings boxes to stutter, those phenomena are most seen when you interact with the machine via X11 clients. I think it's hard to realize if a server only does console I/O, but console also seems to be stuck sometimes. It would be worth checking this with some 'benchmark'. X11 in its kind of oldish incarnation on FreeBSD seems to contribute most to those slowdowns, what so ever. From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 22:11:20 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A63121065670; Thu, 18 Nov 2010 22:11:20 +0000 (UTC) (envelope-from fullermd@over-yonder.net) Received: from thyme.infocus-llc.com (server.infocus-llc.com [206.156.254.44]) by mx1.freebsd.org (Postfix) with ESMTP id 7B7C78FC13; Thu, 18 Nov 2010 22:11:20 +0000 (UTC) Received: from draco.over-yonder.net (c-75-64-226-141.hsd1.ms.comcast.net [75.64.226.141]) (using TLSv1 with cipher ADH-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by thyme.infocus-llc.com (Postfix) with ESMTPSA id D38F337B454; Thu, 18 Nov 2010 16:11:19 -0600 (CST) Received: by draco.over-yonder.net (Postfix, from userid 100) id 3ACB861C42; Thu, 18 Nov 2010 16:11:19 -0600 (CST) Date: Thu, 18 Nov 2010 16:11:19 -0600 From: "Matthew D. Fuller" To: Alexander Best Message-ID: <20101118221119.GS63683@over-yonder.net> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118185635.GA43706@freebsd.org> X-Editor: vi X-OS: FreeBSD User-Agent: Mutt/1.5.20-fullermd.4 (2009-06-14) X-Virus-Scanned: clamav-milter 0.96.4 at thyme.infocus-llc.com X-Virus-Status: Clean Cc: freebsd-performance@freebsd.org, Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 22:11:20 -0000 [ Trim CC a bit] On Thu, Nov 18, 2010 at 06:56:35PM +0000 I heard the voice of Alexander Best, and lo! it spake thus: > On Thu Nov 18 10, Matthew D. Fuller wrote: > > > > Well, my (admittedly limited, and certainly anecdotal) experience > > is that Linux's interactive response when under heavy load was > > always much worse than FreeBSD's. > > well...i tried playing back a 1080p vide files while doing `make > -j64 buildkernel` and FreeBSD's interactivity seems far from > perfect. CPU soak against CPU soak isn't really what I meant :) I mean simple stuff like wandering around in a terminal, firing off top's and netstat's, loading up vi's and less's. When I've worked on Linux boxes that were nailed to the wall doing other stuff, that sort of thing got very laggy and inconsistent; similarly loaded FreeBSD boxes always felt much smoother. To be sure, things like competing disk access can be a different story. But I'm talking more CPU suds, and small scheduler tweaks point in that direction too. -- Matthew Fuller (MF4839) | fullermd@over-yonder.net Systems/Network Administrator | http://www.over-yonder.net/~fullermd/ On the Internet, nobody can hear you scream. From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 22:28:42 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EF154106566B; Thu, 18 Nov 2010 22:28:42 +0000 (UTC) (envelope-from kabaev@gmail.com) Received: from mail-qy0-f175.google.com (mail-qy0-f175.google.com [209.85.216.175]) by mx1.freebsd.org (Postfix) with ESMTP id 437A18FC0C; Thu, 18 Nov 2010 22:28:41 +0000 (UTC) Received: by qyk29 with SMTP id 29so451664qyk.13 for ; Thu, 18 Nov 2010 14:28:41 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:date:from:to:cc:subject :message-id:in-reply-to:references:x-mailer:mime-version :content-type; bh=G6gTXaaqCJpPu9PlygEv7NCUGM5qj47slKRFxtsO51o=; b=sTD0gjkmYL2VuBDq/OYTJrZMLTh6XnuEltxE1cO5xeiLFncW3q15CNSA9OVYE2mb7L C9yURePw/SfhKBHkS8n6TYnQNBcVdA0JWeHpZduYNiUm+XcOmJgLsq3IAZkpflR13+Et mfESpG/2wkrCZGVn+nBWqH91rA/Jb/nIilcxY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=date:from:to:cc:subject:message-id:in-reply-to:references:x-mailer :mime-version:content-type; b=WU/bTCIJtdnrZPa3Ab0ZO7nW/xgTW2BsDas+v7+Lum+TWrmqbQ9Yo2Fj29VQYzWqC9 OY3KevhjC95IwkfviMuMKlUnZ3xzbwblakk6zzwi2NIskCGn+vJMt1UjKlmeUoshU4QD YmjXwRJ0boFlkQkdhwSbTYzS5PGI/vegOdBDk= Received: by 10.229.236.193 with SMTP id kl1mr1055289qcb.37.1290117992914; Thu, 18 Nov 2010 14:06:32 -0800 (PST) Received: from kan.dnsalias.net (c-24-63-226-98.hsd1.ma.comcast.net [24.63.226.98]) by mx.google.com with ESMTPS id u2sm513935qcq.7.2010.11.18.14.06.30 (version=SSLv3 cipher=RC4-MD5); Thu, 18 Nov 2010 14:06:31 -0800 (PST) Date: Thu, 18 Nov 2010 17:06:23 -0500 From: Alexander Kabaev To: Alexander Best Message-ID: <20101118170623.7f9c14f3@kan.dnsalias.net> In-Reply-To: <20101118185635.GA43706@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> X-Mailer: Claws Mail 3.7.6 (GTK+ 2.20.1; amd64-portbld-freebsd9.0) Mime-Version: 1.0 Content-Type: multipart/signed; micalg=PGP-SHA1; boundary="Sig_/bizU4bG9u1fade+f3H9rMcX"; protocol="application/pgp-signature" X-Mailman-Approved-At: Thu, 18 Nov 2010 22:33:51 +0000 Cc: FreeBSD Stable , "Matthew D. Fuller" , Current , Andriy Gapon , FreeBSD, freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 22:28:43 -0000 --Sig_/bizU4bG9u1fade+f3H9rMcX Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Thu, 18 Nov 2010 18:56:35 +0000 Alexander Best wrote: > On Thu Nov 18 10, Matthew D. Fuller wrote: > > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > > Alexander Best, and lo! it spake thus: > > >=20 > > > judging from the videos the changes are having a huge impact imo. > >=20 > > Well, my (admittedly limited, and certainly anecdotal) experience is > > that Linux's interactive response when under heavy load was always > > much worse than FreeBSD's. So maybe that's just them catching up to > > where we already are ;) >=20 > well...i tried playing back a 1080p vide files while doing > `make -j64 buildkernel` and FreeBSD's interactivity seems far from > perfect. One thing that just begs to be asked: since when decoding 1080p became an interactive task? =20 --=20 Alexander Kabaev --Sig_/bizU4bG9u1fade+f3H9rMcX Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iD8DBQFM5aNlQ6z1jMm+XZYRAuAzAJ4yOUUZa/HC1AFxowzBaYuagv92PwCgjT5Y XQ8fvbCjXoZstEJhQ7LJAMA= =e9CV -----END PGP SIGNATURE----- --Sig_/bizU4bG9u1fade+f3H9rMcX-- From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 22:43:38 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id AD5F11065672 for ; Thu, 18 Nov 2010 22:43:38 +0000 (UTC) (envelope-from dieterbsd@engineer.com) Received: from imr-ma05.mx.aol.com (imr-ma05.mx.aol.com [64.12.100.31]) by mx1.freebsd.org (Postfix) with ESMTP id 6C9908FC0C for ; Thu, 18 Nov 2010 22:43:38 +0000 (UTC) Received: from imo-ma01.mx.aol.com (imo-ma01.mx.aol.com [64.12.78.136]) by imr-ma05.mx.aol.com (8.14.1/8.14.1) with ESMTP id oAIMJetc024954; Thu, 18 Nov 2010 17:19:40 -0500 Received: from dieterbsd@engineer.com by imo-ma01.mx.aol.com (mail_out_v42.9.) id n.c99.6e7764b2 (37109); Thu, 18 Nov 2010 17:19:36 -0500 (EST) Received: from smtprly-da02.mx.aol.com (smtprly-da02.mx.aol.com [205.188.249.145]) by cia-db08.mx.aol.com (v129.7) with ESMTP id MAILCIADB087-5bb94ce5a671dd; Thu, 18 Nov 2010 17:19:36 -0500 Received: from web-mmc-d05 (web-mmc-d05.sim.aol.com [205.188.103.95]) by smtprly-da02.mx.aol.com (v129.5) with ESMTP id MAILSMTPRLYDA028-5bb94ce5a671dd; Thu, 18 Nov 2010 17:19:29 -0500 To: freebsd-performance@freebsd.org Content-Transfer-Encoding: quoted-printable Date: Thu, 18 Nov 2010 17:19:28 -0500 X-AOL-IP: 67.206.163.196 X-MB-Message-Source: WebUI Received: from 67.206.163.196 by web-mmc-d05.sysops.aol.com (205.188.103.95) with HTTP (WebMailUI); Thu, 18 Nov 2010 17:19:28 -0500 MIME-Version: 1.0 From: dieterbsd@engineer.com X-MB-Message-Type: User Content-Type: text/plain; charset="us-ascii" X-Mailer: Mail.com Webmail 32843-STANDARD Message-Id: <8CD5585CD2199C8-12F8-3912@web-mmc-d05.sysops.aol.com> X-Spam-Flag: NO X-AOL-SENDER: dieterbsd@engineer.com Cc: "\"Lucius Windschuh , "\"Freddie Cash Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 22:43:38 -0000 Lucius writes: > And using "idprio 1 $cmd" as a workaround is, well, a kludge. You have to be root to use idprio, requiring a suid wrapper, which strikes me as kludgy. The worst part is that idprio and rtprio don't entirely work. An idprio process can still interfere with a rtprio process. :-( Rtprio is false advertising, it is *not* real time. Rtprio is really just pigprio. Freddie writes: > As in, if you don't want your make job to hog resources, then use nice > to run it in the background. Doesn't always work. Nice (and rtprio/idprio) only care about CPU usage, not I/O usage. Nice and friends do very little to throttle a process that is hogging I/O resources. Also: Some device drivers (SATA drivers, firewire, perhaps others) hog the CPU for obscene lengths of time, locking out even other device drivers. This causes DATA TO BE LOST. :-( :-( :-( Some SATA drivers still don't support NCQ, causing *horrible* logjams. NCQ also gives 10-13x the speed. ---switch to metadiscussion--- > Also, does this really need to be cross-posted to -current, -hackers, > and -performance? The question is why does cross posting result in receiving duplicates? (Usenet was much better for this type of discussion.) Chuck writes: > I'm also turned off by the attempt to popup a full page ad in > addition to the rest of the advertising content ... > I don't see any value added from these phoronix.com links I haven't tried the link in question, but I skim through phoronix.com using "links -g" and don't get buried in ads. Lately I've been skimming most sites (including phoronix) with images turned off and they load *SO* much faster. From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 22:59:43 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 1233) id 0D0E81065695; Thu, 18 Nov 2010 22:59:43 +0000 (UTC) Date: Thu, 18 Nov 2010 22:59:43 +0000 From: Alexander Best To: Alexander Kabaev Message-ID: <20101118225943.GB99684@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118170623.7f9c14f3@kan.dnsalias.net> X-Mailman-Approved-At: Thu, 18 Nov 2010 23:04:39 +0000 Cc: FreeBSD Stable , Andriy Gapon , FreeBSD Current , "Matthew D. Fuller" , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 22:59:43 -0000 On Thu Nov 18 10, Alexander Kabaev wrote: > On Thu, 18 Nov 2010 18:56:35 +0000 > Alexander Best wrote: > > > On Thu Nov 18 10, Matthew D. Fuller wrote: > > > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > > > Alexander Best, and lo! it spake thus: > > > > > > > > judging from the videos the changes are having a huge impact imo. > > > > > > Well, my (admittedly limited, and certainly anecdotal) experience is > > > that Linux's interactive response when under heavy load was always > > > much worse than FreeBSD's. So maybe that's just them catching up to > > > where we already are ;) > > > > well...i tried playing back a 1080p vide files while doing > > `make -j64 buildkernel` and FreeBSD's interactivity seems far from > > perfect. > > One thing that just begs to be asked: since when decoding 1080p became > an interactive task? well i did exactly what they did in the video. watch a 1080p video and move the output window around while compiling the kernel. cheers. alex > > -- > Alexander Kabaev -- a13x From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 23:12:38 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9EC0C106564A; Thu, 18 Nov 2010 23:12:38 +0000 (UTC) (envelope-from sgk@troutmask.apl.washington.edu) Received: from troutmask.apl.washington.edu (troutmask.apl.washington.edu [128.208.78.105]) by mx1.freebsd.org (Postfix) with ESMTP id 7A48C8FC12; Thu, 18 Nov 2010 23:12:38 +0000 (UTC) Received: from troutmask.apl.washington.edu (localhost.apl.washington.edu [127.0.0.1]) by troutmask.apl.washington.edu (8.14.4/8.14.4) with ESMTP id oAINCZUT012234; Thu, 18 Nov 2010 15:12:35 -0800 (PST) (envelope-from sgk@troutmask.apl.washington.edu) Received: (from sgk@localhost) by troutmask.apl.washington.edu (8.14.4/8.14.4/Submit) id oAINCZjX012233; Thu, 18 Nov 2010 15:12:35 -0800 (PST) (envelope-from sgk) Date: Thu, 18 Nov 2010 15:12:35 -0800 From: Steve Kargl To: Alexander Best Message-ID: <20101118231235.GA12137@troutmask.apl.washington.edu> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118225943.GB99684@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118225943.GB99684@freebsd.org> User-Agent: Mutt/1.4.2.3i X-Mailman-Approved-At: Thu, 18 Nov 2010 23:15:14 +0000 Cc: FreeBSD Stable , "Matthew D. Fuller" , FreeBSD Current , Andriy Gapon , freebsd-performance@freebsd.org, "O. Hartmann" , Alexander Kabaev Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 23:12:38 -0000 On Thu, Nov 18, 2010 at 10:59:43PM +0000, Alexander Best wrote: > > well i did exactly what they did in the video. watch a 1080p video and move > the output window around while compiling the kernel. > It is trivial to bring ULE to its knees. If you have N cores then all you need is N+1 cpu intensive task. The issue has been known for a few years. http://freebsd.monkey.org/freebsd-current/200807/msg00278.html http://www.mail-archive.com/freebsd-hackers@freebsd.org/msg65839.html -- Steve From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 23:23:09 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 81F1C1065674 for ; Thu, 18 Nov 2010 23:23:09 +0000 (UTC) (envelope-from lwindschuh@googlemail.com) Received: from mail-iw0-f182.google.com (mail-iw0-f182.google.com [209.85.214.182]) by mx1.freebsd.org (Postfix) with ESMTP id 3DB608FC15 for ; Thu, 18 Nov 2010 23:23:09 +0000 (UTC) Received: by iwn39 with SMTP id 39so4267271iwn.13 for ; Thu, 18 Nov 2010 15:23:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=JdL9K6Vb6tU/OGGx3Z/xAZNJ+29LV3aUp9ZLWM5xLYI=; b=mf5gCrQGIHU+4b6yLDpM7N+Tm6jn5o/884KmgXNYb7E4dFOl0HBf9FPvDOa06Z0a5B 1KjIIXdnXEagEnd6Lh4RIuldsFCX6/IamFHdB5KlCDvsMhKfB8bK5qfmcuNA6j92y6sw Qkxiw75GFc+rrHYMeN2BL/WdLyn5ZtnZJsZbo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=i9rzyOA+pUcZWxxbfGCk9dYj3zz2utelYum5elmva9wnzlyKeYDlzCpk9XC5R2c/q7 W0z1AdT4vpFtz3Pmg8YrC0vuDuqWHykYKHs8uXbDUdAOsa4PO0IZazxhgf3oiFXixQ8B a8FBxdSdRke+fkMf8ZXLyeOkKmME2lTbrID9g= MIME-Version: 1.0 Received: by 10.231.39.198 with SMTP id h6mr1529440ibe.21.1290122588650; Thu, 18 Nov 2010 15:23:08 -0800 (PST) Received: by 10.231.192.139 with HTTP; Thu, 18 Nov 2010 15:23:08 -0800 (PST) In-Reply-To: <20101118201644.00004c3c@unknown> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118201644.00004c3c@unknown> Date: Fri, 19 Nov 2010 00:23:08 +0100 Message-ID: From: Lucius Windschuh To: Bruce Cran Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Thu, 18 Nov 2010 23:31:17 +0000 Cc: freebsd-performance@freebsd.org Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 23:23:09 -0000 2010/11/18 Bruce Cran : >Have you tried increasing kern.sched.preempt_thresh? According to >http://groups.google.com/group/mailing.freebsd.stable/browse_thread/thread/05a39f816fd8acc6/82affa9f195b747d?lnk=raot&fwc=1&pli=1 >a good value for desktop use would be 224. Hmm, I though I tried this -- but this helps indeed. :-) The browser, movie player etc. behave much better when a "make -j4 buildworld" is running on my 2-core machine in the background. Thank you. 2010/11/18 Bruce Cran : > If you're using UFS, I've found it to be quite a bottleneck when > doing parallel IO: I even ran a "svn up" in one terminal and tried to > login on another a couple of days ago only to find the motd took over 5 > seconds to appear! That may be excessive since I was running a kernel > with WITNESS and INVARIANTS, but I've found ZFS to be far better if you > want good interactivity when reading/writing to disks. This is indeed another issue, which I also encountered, but explicitly left out since I don't blame the task scheduler for that. ;) Unfortunately, I don't know how much SCHED_ULE's inability to cope with more runnable threads than cores, as Steve mentioned, accounts to the problem I observe. Time to switch back to SCHED_4BSD? *sigh* Lucius From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 23:24:49 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 43767106566B; Thu, 18 Nov 2010 23:24:49 +0000 (UTC) (envelope-from dnebdal@gmail.com) Received: from mail-iw0-f182.google.com (mail-iw0-f182.google.com [209.85.214.182]) by mx1.freebsd.org (Postfix) with ESMTP id B3FDF8FC13; Thu, 18 Nov 2010 23:24:48 +0000 (UTC) Received: by iwn39 with SMTP id 39so4268663iwn.13 for ; Thu, 18 Nov 2010 15:24:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=RF7CCOjjlMdWa/4rvqXFs+lIae9Tv2SV06gC2Iwg5lw=; b=sg9Lue/o9Zt6bStriwYNlVqnQS2YrtrgKujqeiMr1Nkx+9Gm5XOyKRnTPAUPmC8+/+ u7LeESnuWOXHJgWbFDgZEGnxcOVQzTDWN89cyiRmIYuKoWtt5yHy0XOvs0887AYgM8bg cCG4fLaATSvJ2WRDRJgVFeFGYEGbw2SIP7em4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=fMlbe/aYao6kyPml/TI2MMaYrkjEeXAYlqYo6rXpVarqxEHJR5QiqLRIjkLBNo37yI gnYRMlLahOhJilq+0MPIIKXkIkg537iFPno/+ZVvRsAWd3XoFWIsrrte00ij69wZWLr9 Fb8VOwpOraNIIU2yUONOG4sos62SPreFwEawg= MIME-Version: 1.0 Received: by 10.231.34.3 with SMTP id j3mr1454614ibd.100.1290120931782; Thu, 18 Nov 2010 14:55:31 -0800 (PST) Received: by 10.231.58.202 with HTTP; Thu, 18 Nov 2010 14:55:31 -0800 (PST) In-Reply-To: <20101118170623.7f9c14f3@kan.dnsalias.net> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> Date: Fri, 19 Nov 2010 00:55:31 +0200 Message-ID: From: Daniel Nebdal To: FreeBSD Stable Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Mailman-Approved-At: Thu, 18 Nov 2010 23:31:28 +0000 Cc: FreeBSD@freebsd.org, Andriy Gapon , Current , "Matthew D. Fuller" , Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 23:24:49 -0000 On Fri, Nov 19, 2010 at 12:06 AM, Alexander Kabaev wrote= : > On Thu, 18 Nov 2010 18:56:35 +0000 > Alexander Best wrote: > >> On Thu Nov 18 10, Matthew D. Fuller wrote: >> > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of >> > Alexander Best, and lo! it spake thus: >> > > >> > > judging from the videos the changes are having a huge impact imo. >> > >> > Well, my (admittedly limited, and certainly anecdotal) experience is >> > that Linux's interactive response when under heavy load was always >> > much worse than FreeBSD's. =A0So maybe that's just them catching up to >> > where we already are =A0 ;) >> >> well...i tried playing back a 1080p vide files while doing >> `make -j64 buildkernel` and FreeBSD's interactivity seems far from >> perfect. > > One thing that just begs to be asked: since when decoding 1080p became > an interactive task? > Strictly speaking it isn't - but displaying it is a timing-sensitive task that isn't CPU- or I/O-bound, and scheduling-wise that probably makes it more like the "fast response when woken up" interactive tasks than a CPU-bound non-interactive process. Decoding it into another file on the disk is in the latter category, of course - but I don't think that's what he meant. :) More on topic - while this was a tiny patch for Linux, it seems like it would take more work for us, since I don't believe either of the schedulers handles task groups in the required way. The linux patch was just "create task groups automatically", since they already had some suitable logic for scheduling based on task groups in their CFS scheduler. We would have to (re-)add that first, which is non-trivial. -- Daniel Nebdal From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 23:43:52 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 564A7106564A; Thu, 18 Nov 2010 23:43:52 +0000 (UTC) (envelope-from julian@freebsd.org) Received: from out-0.mx.aerioconnect.net (out-0-19.mx.aerioconnect.net [216.240.47.79]) by mx1.freebsd.org (Postfix) with ESMTP id 28F718FC14; Thu, 18 Nov 2010 23:43:51 +0000 (UTC) Received: from idiom.com (postfix@mx0.idiom.com [216.240.32.160]) by out-0.mx.aerioconnect.net (8.13.8/8.13.8) with ESMTP id oAINho0W001757; Thu, 18 Nov 2010 15:43:50 -0800 X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e Received: from julian-mac.elischer.org (h-67-100-89-137.snfccasy.static.covad.net [67.100.89.137]) by idiom.com (Postfix) with ESMTP id 1B2192D601E; Thu, 18 Nov 2010 15:43:48 -0800 (PST) Message-ID: <4CE5BA37.20604@freebsd.org> Date: Thu, 18 Nov 2010 15:43:51 -0800 From: Julian Elischer User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X 10.4; en-US; rv:1.9.2.12) Gecko/20101027 Thunderbird/3.1.6 MIME-Version: 1.0 To: Alexander Best References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> In-Reply-To: <20101118233731.GA10392@freebsd.org> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.67 on 216.240.47.51 Cc: FreeBSD@freebsd.org, FreeBSD Stable , "O. Hartmann" , "Matthew D. Fuller" , Current , Andriy Gapon , freebsd-performance@freebsd.org, Daniel Nebdal Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 23:43:52 -0000 On 11/18/10 3:37 PM, Alexander Best wrote: > On Fri Nov 19 10, Daniel Nebdal wrote: >> On Fri, Nov 19, 2010 at 12:06 AM, Alexander Kabaev wrote: >>> On Thu, 18 Nov 2010 18:56:35 +0000 >>> Alexander Best wrote: >>> >>>> On Thu Nov 18 10, Matthew D. Fuller wrote: >>>>> On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of >>>>> Alexander Best, and lo! it spake thus: >>>>>> judging from the videos the changes are having a huge impact imo. >>>>> Well, my (admittedly limited, and certainly anecdotal) experience is >>>>> that Linux's interactive response when under heavy load was always >>>>> much worse than FreeBSD's. So maybe that's just them catching up to >>>>> where we already are ;) >>>> well...i tried playing back a 1080p vide files while doing >>>> `make -j64 buildkernel` and FreeBSD's interactivity seems far from >>>> perfect. >>> One thing that just begs to be asked: since when decoding 1080p became >>> an interactive task? >>> >> Strictly speaking it isn't - but displaying it is a timing-sensitive >> task that isn't CPU- or I/O-bound, and scheduling-wise that probably >> makes it more like the "fast response when woken up" interactive tasks >> than a CPU-bound non-interactive process. >> Decoding it into another file on the disk is in the latter category, >> of course - but I don't think that's what he meant. :) >> >> More on topic - while this was a tiny patch for Linux, it seems like >> it would take more work for us, since I don't believe either of the >> schedulers handles task groups in the required way. The linux patch >> was just "create task groups automatically", since they already had >> some suitable logic for scheduling based on task groups in their CFS >> scheduler. We would have to (re-)add that first, which is non-trivial. > personally i think freebsd would hugely benefit from a scheduler framework > such as geom/gsched, where it's easy to switch between various algorithms. > > that way it be much easier to try out new concepts without having to write a > completely new scheduler. we are part of the way there.. at least we did abstract the scheduler to the point where we have two completely different ones. you are welcome to develop a 'framework as you describe and plug it into the abstraction we already have. > cheers. > alex > >> >> -- >> Daniel Nebdal From owner-freebsd-performance@FreeBSD.ORG Thu Nov 18 23:37:31 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 1233) id 92FCC106566B; Thu, 18 Nov 2010 23:37:31 +0000 (UTC) Date: Thu, 18 Nov 2010 23:37:31 +0000 From: Alexander Best To: Daniel Nebdal Message-ID: <20101118233731.GA10392@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: X-Mailman-Approved-At: Thu, 18 Nov 2010 23:46:20 +0000 Cc: FreeBSD@freebsd.org, FreeBSD Stable , Andriy Gapon , Current , "Matthew D. Fuller" , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Nov 2010 23:37:31 -0000 On Fri Nov 19 10, Daniel Nebdal wrote: > On Fri, Nov 19, 2010 at 12:06 AM, Alexander Kabaev wrote: > > On Thu, 18 Nov 2010 18:56:35 +0000 > > Alexander Best wrote: > > > >> On Thu Nov 18 10, Matthew D. Fuller wrote: > >> > On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > >> > Alexander Best, and lo! it spake thus: > >> > > > >> > > judging from the videos the changes are having a huge impact imo. > >> > > >> > Well, my (admittedly limited, and certainly anecdotal) experience is > >> > that Linux's interactive response when under heavy load was always > >> > much worse than FreeBSD's.  So maybe that's just them catching up to > >> > where we already are   ;) > >> > >> well...i tried playing back a 1080p vide files while doing > >> `make -j64 buildkernel` and FreeBSD's interactivity seems far from > >> perfect. > > > > One thing that just begs to be asked: since when decoding 1080p became > > an interactive task? > > > > Strictly speaking it isn't - but displaying it is a timing-sensitive > task that isn't CPU- or I/O-bound, and scheduling-wise that probably > makes it more like the "fast response when woken up" interactive tasks > than a CPU-bound non-interactive process. > Decoding it into another file on the disk is in the latter category, > of course - but I don't think that's what he meant. :) > > More on topic - while this was a tiny patch for Linux, it seems like > it would take more work for us, since I don't believe either of the > schedulers handles task groups in the required way. The linux patch > was just "create task groups automatically", since they already had > some suitable logic for scheduling based on task groups in their CFS > scheduler. We would have to (re-)add that first, which is non-trivial. personally i think freebsd would hugely benefit from a scheduler framework such as geom/gsched, where it's easy to switch between various algorithms. that way it be much easier to try out new concepts without having to write a completely new scheduler. cheers. alex > > > -- > Daniel Nebdal -- a13x From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 00:00:00 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E5B761065672; Thu, 18 Nov 2010 23:59:59 +0000 (UTC) (envelope-from yanegomi@gmail.com) Received: from mail-ww0-f50.google.com (mail-ww0-f50.google.com [74.125.82.50]) by mx1.freebsd.org (Postfix) with ESMTP id 4A3928FC1E; Thu, 18 Nov 2010 23:59:58 +0000 (UTC) Received: by wwd20 with SMTP id 20so3880954wwd.31 for ; Thu, 18 Nov 2010 15:59:58 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:sender:received :in-reply-to:references:date:x-google-sender-auth:message-id:subject :from:to:cc:content-type:content-transfer-encoding; bh=/WOlykv8xFvgtJ2Af3FWvX3HKveRtvuImXmPM9HGXJY=; b=qrDlMAKEckzjOxsssbpZj6y+Ao583W3U87S+32aG+AoZo/66aZ8UGdaZ0bTjVXXOQX rQOyC2nGzNv8gaDk0WKbWz0m8kBzK6XtN20hDNgD3jXEa32curfXUc7DcNVQRGJLKR5B fZMfflcZlQ2jguWFJWRNwrpmpJAhgcY+sG5oI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=fpvWgY9WGFmdwWtDAxsxt3VhzhgmThiYdoktIXj/KOW0SekSTqwdciejI1MUjbqhsm tbQRoYpvUiIbppwsEK6jSViM9CT8opSEpx5UK6gduHMXo8tPLE6CtA2qmqi4NnXvr/qb ial56ef351rX/njIfoVv17nkJjMhr6nCA1FJ0= MIME-Version: 1.0 Received: by 10.216.46.200 with SMTP id r50mr180958web.45.1290123417909; Thu, 18 Nov 2010 15:36:57 -0800 (PST) Sender: yanegomi@gmail.com Received: by 10.216.198.27 with HTTP; Thu, 18 Nov 2010 15:36:57 -0800 (PST) In-Reply-To: <20101118231235.GA12137@troutmask.apl.washington.edu> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118225943.GB99684@freebsd.org> <20101118231235.GA12137@troutmask.apl.washington.edu> Date: Thu, 18 Nov 2010 15:36:57 -0800 X-Google-Sender-Auth: QQMUstOnBrl7mFCsjAfL84y5D2Q Message-ID: From: Garrett Cooper To: Steve Kargl Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Mailman-Approved-At: Fri, 19 Nov 2010 00:17:26 +0000 Cc: FreeBSD Stable , Andriy Gapon , FreeBSD Current , "Matthew D. Fuller" , Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 00:00:00 -0000 On Thu, Nov 18, 2010 at 3:12 PM, Steve Kargl wrote: > On Thu, Nov 18, 2010 at 10:59:43PM +0000, Alexander Best wrote: >> >> well i did exactly what they did in the video. watch a 1080p video and m= ove >> the output window around while compiling the kernel. >> > > It is trivial to bring ULE to its knees. =A0If you > have N cores then all you need is N+1 cpu intensive > task. =A0The issue has been known for a few years. I/O intensive tasks don't help either though. Things tend to get choppy whenever I'm doing something like update from svn and doing something a bit more interactive like use firefox (and scroll... for instance gmail seems to be a pig in this area -- heh), vlc is a little less painful, etc. I wonder if the issue isn't necessarily tasks but more or less locking. firefox uses a lot of threads and file based mutexes according to what I've seen with top and ps (please correct me if I'm wrong). firefox also has a tendency with the nvidia-driver (I know... blobs are bad) to produce funky artifacts on the screen (again, when scrolling on gmail, dealing with flash, etc) or certain bits don't redraw properly (text when scrolling ends up ghosting on multiple lines until I force a redraw with the cursor or by scrolling back over that section of text). I'm sure there are a lot of performance issues within FreeBSD and opensource desktop software that needs to be properly worked out on a case by case basis, so I think that writing everything off as awesome and working better with one magic patch is unlikely. Besides, Linux has a more complicated scheduler than FreeBSD does. Thanks, -Garrett From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 00:17:10 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 1233) id 169521065694; Fri, 19 Nov 2010 00:17:10 +0000 (UTC) Date: Fri, 19 Nov 2010 00:17:10 +0000 From: Alexander Best To: Julian Elischer Message-ID: <20101119001710.GA14641@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4CE5BA37.20604@freebsd.org> X-Mailman-Approved-At: Fri, 19 Nov 2010 00:26:51 +0000 Cc: FreeBSD@freebsd.org, FreeBSD Stable , "O. Hartmann" , "Matthew D. Fuller" , Current , Andriy Gapon , freebsd-performance@freebsd.org, Daniel Nebdal Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 00:17:10 -0000 On Thu Nov 18 10, Julian Elischer wrote: > On 11/18/10 3:37 PM, Alexander Best wrote: > >On Fri Nov 19 10, Daniel Nebdal wrote: > >>On Fri, Nov 19, 2010 at 12:06 AM, Alexander Kabaev > >>wrote: > >>>On Thu, 18 Nov 2010 18:56:35 +0000 > >>>Alexander Best wrote: > >>> > >>>>On Thu Nov 18 10, Matthew D. Fuller wrote: > >>>>>On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of > >>>>>Alexander Best, and lo! it spake thus: > >>>>>>judging from the videos the changes are having a huge impact imo. > >>>>>Well, my (admittedly limited, and certainly anecdotal) experience is > >>>>>that Linux's interactive response when under heavy load was always > >>>>>much worse than FreeBSD's. So maybe that's just them catching up to > >>>>>where we already are ;) > >>>>well...i tried playing back a 1080p vide files while doing > >>>>`make -j64 buildkernel` and FreeBSD's interactivity seems far from > >>>>perfect. > >>>One thing that just begs to be asked: since when decoding 1080p became > >>>an interactive task? > >>> > >>Strictly speaking it isn't - but displaying it is a timing-sensitive > >>task that isn't CPU- or I/O-bound, and scheduling-wise that probably > >>makes it more like the "fast response when woken up" interactive tasks > >>than a CPU-bound non-interactive process. > >>Decoding it into another file on the disk is in the latter category, > >>of course - but I don't think that's what he meant. :) > >> > >>More on topic - while this was a tiny patch for Linux, it seems like > >>it would take more work for us, since I don't believe either of the > >>schedulers handles task groups in the required way. The linux patch > >>was just "create task groups automatically", since they already had > >>some suitable logic for scheduling based on task groups in their CFS > >>scheduler. We would have to (re-)add that first, which is non-trivial. > >personally i think freebsd would hugely benefit from a scheduler framework > >such as geom/gsched, where it's easy to switch between various algorithms. > > > >that way it be much easier to try out new concepts without having to write > >a > >completely new scheduler. > > we are part of the way there.. > > at least we did abstract the scheduler to the point where > we have two completely different ones. > you are welcome to develop a 'framework as you describe and plug it into > the abstraction we already have. **** 17:49 @ arundel : also looking at the svn log shows that still a lot of \ commits happen to sched_4bsd. so it's defenately not being abbandoned. in \ fact there might be situations where it performs better than sched_ule. 17:50 @ arundel : i'm looking forward to a scheduler which looks sorta like \ geom and enables you to plugin addition plugins with different scheduling \ algorithms. :) 17:51 @ Genesys : Luigi Rizzo had a plugabble scheduler back in 4.* or \ thereabouts 17:51 @ Genesys : you could kldload new ones and switch to them on the fly 17:52 @ arundel : wow. that sounds cool. too bad it didn't make it into src \ tree. by now it's probably outdated and needs to be reworked quite a bit. **** does anybody know something about this? i'm sorry. i'd really love to contribute some code, but my programing skills are pretty scrappy. ;) it would probably take me 20 years to figure out the current sched code. cheers. alex > > >cheers. > >alex > > > >> > >>-- > >>Daniel Nebdal -- a13x From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 00:35:28 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 97DBE1065673 for ; Fri, 19 Nov 2010 00:35:28 +0000 (UTC) (envelope-from onemda@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 2C0658FC0A for ; Fri, 19 Nov 2010 00:35:27 +0000 (UTC) Received: by wyb35 with SMTP id 35so3065287wyb.13 for ; Thu, 18 Nov 2010 16:35:27 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:mime-version:received:in-reply-to :references:from:date:message-id:subject:to:cc:content-type; bh=I1SiLxkyEUtVIDzOHSFXQDWIuwtVGBwbn0xo4qGU+oA=; b=WLoEBOXZJLA0zLvphfqUr7R7VSf3M37mG3qqbA8GcdlHOQ5YzlXHN8OpVvOwl7Imw/ rvk4m8bV8L9dtXfBfnMRF6KVbsHkBORgL552zZxlAEhdD1ztpvTxmtqFmtv653FDepqy TioEqvdFwzTb+YRegPrrW9y98Ak0h8NSUI9qg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; b=Ffj/WED8pmupiXTHG+PI+mXDFGBOuhAIMW0efLvqTpQx4zbh3Et6oMXdaP4dh+kRfC gmUP7WxmHYgnCumED40nAt+hdfFV56xBrsYQnoGGDx5laVPB4SgbKQqsC/n33AHnUYR8 bQVgEQ0RWlcQkWk33XzoTH54z/kKHvYeihKqE= Received: by 10.216.238.130 with SMTP id a2mr232254wer.77.1290125107416; Thu, 18 Nov 2010 16:05:07 -0800 (PST) MIME-Version: 1.0 Received: by 10.216.234.82 with HTTP; Thu, 18 Nov 2010 16:04:47 -0800 (PST) In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118201644.00004c3c@unknown> From: Paul B Mahol Date: Fri, 19 Nov 2010 00:04:47 +0000 Message-ID: To: Lucius Windschuh Content-Type: text/plain; charset=ISO-8859-1 Cc: Bruce Cran , freebsd-performance@freebsd.org Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 00:35:28 -0000 On Thu, Nov 18, 2010 at 11:23 PM, Lucius Windschuh wrote: > 2010/11/18 Bruce Cran : >>Have you tried increasing kern.sched.preempt_thresh? According to >>http://groups.google.com/group/mailing.freebsd.stable/browse_thread/thread/05a39f816fd8acc6/82affa9f195b747d?lnk=raot&fwc=1&pli=1 >>a good value for desktop use would be 224. > > Hmm, I though I tried this -- but this helps indeed. :-) > The browser, movie player etc. behave much better when a "make -j4 > buildworld" is running on my 2-core machine in the background. Thank > you. > > 2010/11/18 Bruce Cran : >> If you're using UFS, I've found it to be quite a bottleneck when >> doing parallel IO: I even ran a "svn up" in one terminal and tried to >> login on another a couple of days ago only to find the motd took over 5 >> seconds to appear! That may be excessive since I was running a kernel >> with WITNESS and INVARIANTS, but I've found ZFS to be far better if you >> want good interactivity when reading/writing to disks. > > This is indeed another issue, which I also encountered, but explicitly > left out since I don't blame the task scheduler for that. ;) > > Unfortunately, I don't know how much SCHED_ULE's inability to cope > with more runnable threads than cores, as Steve mentioned, accounts to > the problem I observe. Time to switch back to SCHED_4BSD? *sigh* OT: Compare building kernel on tmpfs vs ufs on mdX vs ufs on usb stick and guess what is faster and does not cause non interactive mouse movement. From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 01:33:21 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8A1891065679 for ; Fri, 19 Nov 2010 01:33:21 +0000 (UTC) (envelope-from davidxu@freebsd.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id 5C5DA8FC2C; Fri, 19 Nov 2010 01:33:21 +0000 (UTC) Received: from xyf.my.dom (localhost [127.0.0.1]) by freefall.freebsd.org (8.14.4/8.14.4) with ESMTP id oAJ1XJqE089239; Fri, 19 Nov 2010 01:33:20 GMT (envelope-from davidxu@freebsd.org) Message-ID: <4CE64463.3050508@freebsd.org> Date: Fri, 19 Nov 2010 09:33:23 +0000 From: David Xu User-Agent: Thunderbird 2.0.0.24 (X11/20100630) MIME-Version: 1.0 To: Lucius Windschuh References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118201644.00004c3c@unknown> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Bruce Cran , freebsd-performance@freebsd.org Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 01:33:21 -0000 Lucius Windschuh wrote: > 2010/11/18 Bruce Cran : >> Have you tried increasing kern.sched.preempt_thresh? According to >> http://groups.google.com/group/mailing.freebsd.stable/browse_thread/thread/05a39f816fd8acc6/82affa9f195b747d?lnk=raot&fwc=1&pli=1 >> a good value for desktop use would be 224. > > Hmm, I though I tried this -- but this helps indeed. :-) > The browser, movie player etc. behave much better when a "make -j4 > buildworld" is running on my 2-core machine in the background. Thank > you. > > 2010/11/18 Bruce Cran : >> If you're using UFS, I've found it to be quite a bottleneck when >> doing parallel IO: I even ran a "svn up" in one terminal and tried to >> login on another a couple of days ago only to find the motd took over 5 >> seconds to appear! That may be excessive since I was running a kernel >> with WITNESS and INVARIANTS, but I've found ZFS to be far better if you >> want good interactivity when reading/writing to disks. > > This is indeed another issue, which I also encountered, but explicitly > left out since I don't blame the task scheduler for that. ;) > > Unfortunately, I don't know how much SCHED_ULE's inability to cope > with more runnable threads than cores, as Steve mentioned, accounts to > the problem I observe. Time to switch back to SCHED_4BSD? *sigh* > > Lucius Sometimes, I am thinking that our thread scheduler should be split into two layers, looks like Solaris did, the sched_ule really should only be responsible for CPU dispatching, it only cares where a thread should be dispatched based on CPU-affinity, each CPU's load, ... Another layer is how to calculate thread's priority for time-sharing thread, you can specify which priority algorithm to used, static or dynamic priority scheduling, ULE's algorithm or 4BSD. with cpuset, one even can bind all real-time processes to a specific cpu group, they needn't to be a superuser to run real-time thread. Regards, David Xu From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 01:05:19 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 041E9106566B; Fri, 19 Nov 2010 01:05:19 +0000 (UTC) (envelope-from dmagda@ee.ryerson.ca) Received: from eccles.ee.ryerson.ca (ee.ryerson.ca [141.117.1.2]) by mx1.freebsd.org (Postfix) with ESMTP id B19838FC17; Fri, 19 Nov 2010 01:05:18 +0000 (UTC) Received: from [10.0.1.3] (bas1-toronto09-1279534875.dsl.bell.ca [76.68.39.27]) (authenticated bits=0) by eccles.ee.ryerson.ca (8.14.4/8.14.4) with ESMTP id oAJ0owe8008179 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Thu, 18 Nov 2010 19:50:59 -0500 (EST) (envelope-from dmagda@ee.ryerson.ca) Message-Id: <38521DB9-B3AE-4F96-89C9-75A80D2E0B3E@ee.ryerson.ca> From: David Magda To: Julian Elischer In-Reply-To: <4CE5BA37.20604@freebsd.org> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v936) Date: Thu, 18 Nov 2010 19:50:58 -0500 References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> X-Mailer: Apple Mail (2.936) X-Mailman-Approved-At: Fri, 19 Nov 2010 03:34:48 +0000 Cc: freebsd-performance@freebsd.org, Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 01:05:19 -0000 On Nov 18, 2010, at 18:43, Julian Elischer wrote: > we are part of the way there.. > > at least we did abstract the scheduler to the point where we have > two completely different ones. you are welcome to develop a > 'framework as you describe and plug it into the abstraction we > already have. It may be something to suggest for the next round of Google's Summer of Code. Or perhaps part of a school project in operating systems work (master's level? eager-beaver bachelor's thesis?). Having a bit more flexibility in being able to make different components "pluggable" would help encourage the use of BSD in more research projects. And more people learning and hacking on BSD can't be a bad thing. :) From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 09:46:51 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6401B1065712; Fri, 19 Nov 2010 09:46:51 +0000 (UTC) (envelope-from bruce@cran.org.uk) Received: from muon.cran.org.uk (muon.cran.org.uk [IPv6:2a01:348:0:15:5d59:5c40:0:1]) by mx1.freebsd.org (Postfix) with ESMTP id DF91B8FC0A; Fri, 19 Nov 2010 09:46:50 +0000 (UTC) Received: from muon.cran.org.uk (localhost [127.0.0.1]) by muon.cran.org.uk (Postfix) with ESMTP id 3DF98E7211; Fri, 19 Nov 2010 09:46:50 +0000 (GMT) Received: from unknown (client-86-27-40-229.glfd.adsl.virginmedia.com [86.27.40.229]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by muon.cran.org.uk (Postfix) with ESMTPSA; Fri, 19 Nov 2010 09:46:49 +0000 (GMT) Date: Fri, 19 Nov 2010 09:46:52 +0000 From: Bruce Cran To: Andrew Reilly Message-ID: <20101119094652.00003652@unknown> In-Reply-To: <20101119044129.GA4063@johnny.reilly.home> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> X-Mailer: Claws Mail 3.7.4cvs1 (GTK+ 2.16.0; i586-pc-mingw32msvc) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" , Gapon , Andriy Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 09:46:51 -0000 [removed current@ and stable@ from the Cc list] On Fri, 19 Nov 2010 15:41:29 +1100 Andrew Reilly wrote: > On Linux. Have you ever seen those sorts of UI problems on FreeBSD? > I don't watch much video on my systems, but I haven't seen that. > FreeBSD has always been good at keeping user-interactive processes > responsive while compiles or what-not are going on in the background. I've definitely seen problems when running builds in an xterm. I've often resorted to canceling it and running it on a syscons console instead to improve performance. -- Bruce Cran From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 10:02:09 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A06761065697; Fri, 19 Nov 2010 10:02:09 +0000 (UTC) (envelope-from bruce@cran.org.uk) Received: from muon.cran.org.uk (muon.cran.org.uk [IPv6:2a01:348:0:15:5d59:5c40:0:1]) by mx1.freebsd.org (Postfix) with ESMTP id 5EC718FC15; Fri, 19 Nov 2010 10:02:09 +0000 (UTC) Received: from muon.cran.org.uk (localhost [127.0.0.1]) by muon.cran.org.uk (Postfix) with ESMTP id B22AAE7211; Fri, 19 Nov 2010 10:02:08 +0000 (GMT) Received: from unknown (client-86-27-40-229.glfd.adsl.virginmedia.com [86.27.40.229]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by muon.cran.org.uk (Postfix) with ESMTPSA; Fri, 19 Nov 2010 10:02:08 +0000 (GMT) Date: Fri, 19 Nov 2010 10:02:11 +0000 From: Bruce Cran To: Andriy Gapon Message-ID: <20101119100211.00002eb3@unknown> In-Reply-To: <4CE64879.2060802@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> <20101119094652.00003652@unknown> <4CE64879.2060802@freebsd.org> X-Mailer: Claws Mail 3.7.4cvs1 (GTK+ 2.16.0; i586-pc-mingw32msvc) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" , Andrew Reilly Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 10:02:09 -0000 On Fri, 19 Nov 2010 11:50:49 +0200 Andriy Gapon wrote: > So, what was it a problem with scheduler or with, e.g., "something X" > being too slow rendering glyphs? Who can tell... There are too many components involved when running the Linux benchmark of running a build to know the scheduler is at fault. However the slowdown did still exist when I minimized the window. I've at least found that the FreeBSD schedulers can scale much better than Linux: you can have a load of 100 or more before the system becomes unusable whereas people are amazed to see loads of more than 15 on Linux. -- Bruce Cran From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 10:19:44 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D32711065670 for ; Fri, 19 Nov 2010 10:19:44 +0000 (UTC) (envelope-from ohartman@zedat.fu-berlin.de) Received: from outpost1.zedat.fu-berlin.de (outpost1.zedat.fu-berlin.de [130.133.4.66]) by mx1.freebsd.org (Postfix) with ESMTP id 89C8F8FC14 for ; Fri, 19 Nov 2010 10:19:44 +0000 (UTC) Received: from inpost2.zedat.fu-berlin.de ([130.133.4.69]) by outpost1.zedat.fu-berlin.de (Exim 4.69) with esmtp (envelope-from ) id <1PJO4E-0007hl-L5>; Fri, 19 Nov 2010 11:19:42 +0100 Received: from telesto.geoinf.fu-berlin.de ([130.133.86.198]) by inpost2.zedat.fu-berlin.de (Exim 4.69) with esmtpsa (envelope-from ) id <1PJO4E-0006Mr-Fe>; Fri, 19 Nov 2010 11:19:42 +0100 Message-ID: <4CE64F42.1060006@zedat.fu-berlin.de> Date: Fri, 19 Nov 2010 11:19:46 +0100 From: "O. Hartmann" Organization: Freie =?ISO-8859-1?Q?Universit=E4t_Berlin?= User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Thunderbird/3.1.6 MIME-Version: 1.0 To: Bruce Cran References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> <20101119094652.00003652@unknown> In-Reply-To: <20101119094652.00003652@unknown> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: 130.133.86.198 Cc: Alexander Best , freebsd-performance@freebsd.org, Andriy Gapon , Andrew Reilly Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 10:19:44 -0000 On 11/19/10 10:46, Bruce Cran wrote: > [removed current@ and stable@ from the Cc list] > > On Fri, 19 Nov 2010 15:41:29 +1100 > Andrew Reilly wrote: > >> On Linux. Have you ever seen those sorts of UI problems on FreeBSD? >> I don't watch much video on my systems, but I haven't seen that. >> FreeBSD has always been good at keeping user-interactive processes >> responsive while compiles or what-not are going on in the background. > > I've definitely seen problems when running builds in an xterm. I've > often resorted to canceling it and running it on a syscons console > instead to improve performance. > A simple test: use X11, simply use windowmaker as the GUI (this is my configuration on most FreeBSD boxes around here in use for scientidic duties). The simply update your siurce tree via 'svn update' and warch responsivenes of your desktop. Or try start building world AND do something other, like building a big port like something from Qt4. I realize hangs and stuttering on the following FreeBSD OS' and hardware: FreeBSD 8.1-STABLE/amd64: box with 8 GB RAM, Intel Q6600 4-core CPU, UFS2 filesystem containing root. Another box 4 GB RAM, Intel E8400 2-core CPU, also UFS2 as the base filesystem. Both boxes use X11. The next box is our number crunching server, DELL Poweredge1950III, 16 GB RAM, two CPU XEON hw.model: Intel(R) Xeon(R) CPU E5420 @ 2.50GHz. No GUI. Base filesystem on which is build done: UFS2. These boxes have a significant 'stutter' and hang, when starting updating source tree or start building the base system, visible on each xterm or ssh session connected to the box (this is for the server which has no GUI). FreeBSD 9.0-CURRENT/amd64: Notebook, 4GB RAM, CPU: Intel Core i-5 with 2,4 GHz. Base filesystem UFS2, graphical device nVidia XM3100 with nVidia most reent 64bit BLOB driver for this kind of hardware. On the notebook the drop in performance and responsivenes is amazing, the box is 'dead' for a minute when doing 'svn update' for the source tree. I do not know whether those hangs are due to the topic of this thread, I doubt it is more due to a weakness of the I/O subsystem dealing with harddrives (I see hangs while on heavy load also on our ZFS homes). Since our institute uses several Linux flavors around here of several revisions I have some 'naive' comparisons. Even on heavy I/O load Linux does respond nicely. Our 8 core FreeBSD box runs some modelling software for astrodynamics, the software is 'serial', not yet parallelized, but even utilizing 6 out of 8 cores and push them under heavy load does not bother the system and the server responding is good. But any kind of heavy HD I/O, even if this job utilizes only one CPU (as far as I can guess), it gets worse. From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 07:05:17 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0F1B4106566B; Fri, 19 Nov 2010 07:05:17 +0000 (UTC) (envelope-from areilly@bigpond.net.au) Received: from nskntqsrv01p.mx.bigpond.com (nskntqsrv01p.mx.bigpond.com [61.9.168.231]) by mx1.freebsd.org (Postfix) with ESMTP id 495FB8FC08; Fri, 19 Nov 2010 07:05:15 +0000 (UTC) Received: from nskntotgx01p.mx.bigpond.com ([124.188.161.100]) by nskntmtas03p.mx.bigpond.com with ESMTP id <20101119044139.EVJD24865.nskntmtas03p.mx.bigpond.com@nskntotgx01p.mx.bigpond.com>; Fri, 19 Nov 2010 04:41:39 +0000 Received: from johnny.reilly.home ([124.188.161.100]) by nskntotgx01p.mx.bigpond.com with ESMTP id <20101119044139.GCRY25056.nskntotgx01p.mx.bigpond.com@johnny.reilly.home>; Fri, 19 Nov 2010 04:41:39 +0000 Date: Fri, 19 Nov 2010 15:41:29 +1100 From: Andrew Reilly To: Alexander Best Message-ID: <20101119044129.GA4063@johnny.reilly.home> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101118182324.GA36312@freebsd.org> User-Agent: Mutt/1.4.2.3i X-RPD-ScanID: Class unknown; VirusThreatLevel unknown, RefID str=0001.0A150202.4CE60003.00B0,ss=1,fgs=0 X-SIH-MSG-ID: rRg6ENH9TAD0zmQs0WyzOwJxyArnqyN48Z4QX81loRIGTUDCp8DeQ9rVMfpRv9GgxD9FJhqBNGUhaa/tTY3Rs9mK X-Mailman-Approved-At: Fri, 19 Nov 2010 11:44:37 +0000 Cc: "O. Hartmann" , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 07:05:17 -0000 On Thu, Nov 18, 2010 at 06:23:24PM +0000, Alexander Best wrote: > you think so? judging from the videos the changes are having a huge impact imo. On Linux. Have you ever seen those sorts of UI problems on FreeBSD? I don't watch much video on my systems, but I haven't seen that. FreeBSD has always been good at keeping user-interactive processes responsive while compiles or what-not are going on in the background. Cheers, -- Andrew From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 09:25:34 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 59EA61065673; Fri, 19 Nov 2010 09:25:34 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id E9F2B8FC1E; Fri, 19 Nov 2010 09:25:32 +0000 (UTC) Received: from porto.topspin.kiev.ua (porto-e.starpoint.kiev.ua [212.40.38.100]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id LAA07142; Fri, 19 Nov 2010 11:25:31 +0200 (EET) (envelope-from avg@freebsd.org) Received: from localhost.topspin.kiev.ua ([127.0.0.1]) by porto.topspin.kiev.ua with esmtp (Exim 4.34 (FreeBSD)) id 1PJNDm-000LXs-O4; Fri, 19 Nov 2010 11:25:30 +0200 Message-ID: <4CE64289.5030903@freebsd.org> Date: Fri, 19 Nov 2010 11:25:29 +0200 From: Andriy Gapon User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Alexander Best References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> In-Reply-To: <20101118185635.GA43706@freebsd.org> X-Enigmail-Version: 1.1.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 19 Nov 2010 11:57:34 +0000 Cc: "O. Hartmann" , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , "Matthew D. Fuller" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 09:25:34 -0000 on 18/11/2010 20:56 Alexander Best said the following: > On Thu Nov 18 10, Matthew D. Fuller wrote: >> On Thu, Nov 18, 2010 at 06:23:24PM +0000 I heard the voice of >> Alexander Best, and lo! it spake thus: >>> >>> judging from the videos the changes are having a huge impact imo. >> >> Well, my (admittedly limited, and certainly anecdotal) experience is >> that Linux's interactive response when under heavy load was always >> much worse than FreeBSD's. So maybe that's just them catching up to >> where we already are ;) > > well...i tried playing back a 1080p vide files while doing > `make -j64 buildkernel` and FreeBSD's interactivity seems far from perfect. > > it might be possible that linux'es interactivity was worse than freebsd's, > but still this patch should be evaluated for freebsd. in this particular case > it seems linux now does better than freebsd. You do realize that there are many more variables for such a test than just "1080p video" and "make -j64 buildkernel"? Let's not forget about hardware, video drivers, player capabilities, exact kind of the video (you know, 1080p alone doesn't tell much). Besides, someone might be interested in running -j64 on his 1,2,4-core desktop system, but it's definitely not me. I prefer to be reasonable. I am not saying that our scheduler (ULE) is perfect. I don't even say that it's better (in whatever comparison system) than Linux scheduler X. I say that I wouldn't spend my time improving system behavior in a scenario like this. I compile stuff very frequently (kernels, ports, world) while browsing web, reading email, doing various "desktop stuff", sometimes watching videos from the web (like e.g. trailers). On this machine/hardware I have never personally felt a need for improvements in the scheduler. And I run KDE4 with all bells and whistles enabled. YMMV. P.S. I don't discourage anyone from improving our scheduler, I even do encourage that. -- Andriy Gapon From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 09:37:18 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 285891065670; Fri, 19 Nov 2010 09:37:18 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id 827F78FC0A; Fri, 19 Nov 2010 09:37:15 +0000 (UTC) Received: from porto.topspin.kiev.ua (porto-e.starpoint.kiev.ua [212.40.38.100]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id LAA07284; Fri, 19 Nov 2010 11:37:13 +0200 (EET) (envelope-from avg@freebsd.org) Received: from localhost.topspin.kiev.ua ([127.0.0.1]) by porto.topspin.kiev.ua with esmtp (Exim 4.34 (FreeBSD)) id 1PJNP7-000LZD-99; Fri, 19 Nov 2010 11:37:13 +0200 Message-ID: <4CE64548.4020709@freebsd.org> Date: Fri, 19 Nov 2010 11:37:12 +0200 From: Andriy Gapon User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Daniel Nebdal References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> In-Reply-To: X-Enigmail-Version: 1.1.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 19 Nov 2010 11:57:42 +0000 Cc: FreeBSD@freebsd.org, FreeBSD Stable , "Matthew D. Fuller" , Current , Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 09:37:18 -0000 on 19/11/2010 00:55 Daniel Nebdal said the following: > On Fri, Nov 19, 2010 at 12:06 AM, Alexander Kabaev wrote: >> On Thu, 18 Nov 2010 18:56:35 +0000 >> Alexander Best wrote: >>> well...i tried playing back a 1080p vide files while doing >>> `make -j64 buildkernel` and FreeBSD's interactivity seems far from >>> perfect. >> >> One thing that just begs to be asked: since when decoding 1080p became >> an interactive task? >> > > Strictly speaking it isn't - but displaying it is a timing-sensitive > task that isn't CPU- or I/O-bound, and scheduling-wise that probably Well, I am not sure if I can agree about CPU-bound-ness. Depends on the exact video file, of course, but certain high-quality 1080p are very CPU intensive unless decoding is offloaded from the CPU. Depends on decoder code too. I had some videos that were CPU-bound on my Athlon II X2 250 with then-version of mplayer from ports. YMMV. -- Andriy Gapon From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 09:42:07 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3E73B106566C; Fri, 19 Nov 2010 09:42:07 +0000 (UTC) (envelope-from bruce@cran.org.uk) Received: from muon.cran.org.uk (muon.cran.org.uk [IPv6:2a01:348:0:15:5d59:5c40:0:1]) by mx1.freebsd.org (Postfix) with ESMTP id C2F4C8FC18; Fri, 19 Nov 2010 09:42:06 +0000 (UTC) Received: from muon.cran.org.uk (localhost [127.0.0.1]) by muon.cran.org.uk (Postfix) with ESMTP id AA582E60BE; Fri, 19 Nov 2010 09:42:05 +0000 (GMT) Received: from unknown (client-86-27-40-229.glfd.adsl.virginmedia.com [86.27.40.229]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by muon.cran.org.uk (Postfix) with ESMTPSA; Fri, 19 Nov 2010 09:42:04 +0000 (GMT) Date: Fri, 19 Nov 2010 09:42:07 +0000 From: Bruce Cran To: Alexander Best Message-ID: <20101119094207.00004cb4@unknown> In-Reply-To: <20101119001710.GA14641@freebsd.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> X-Mailer: Claws Mail 3.7.4cvs1 (GTK+ 2.16.0; i586-pc-mingw32msvc) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 19 Nov 2010 11:57:50 +0000 Cc: FreeBSD@freebsd.org, Stable , Daniel Nebdal , Julian Elischer , "Matthew D. Fuller" , Current , Gapon , FreeBSD, Andriy, freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 09:42:07 -0000 On Fri, 19 Nov 2010 00:17:10 +0000 Alexander Best wrote: > 17:51 @ Genesys : Luigi Rizzo had a plugabble scheduler back in 4.* > or \ thereabouts > 17:51 @ Genesys : you could kldload new ones and switch to them on > the fly 17:52 @ arundel : wow. that sounds cool. too bad it didn't > make it into src \ tree. by now it's probably outdated and needs to > be reworked quite a bit. **** > > does anybody know something about this? Google suggests that the work was a GSoC project in 2005 on a pluggable disk scheduler. -- Bruce Cran From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 09:51:00 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6CB09106564A; Fri, 19 Nov 2010 09:51:00 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id 7669E8FC22; Fri, 19 Nov 2010 09:50:58 +0000 (UTC) Received: from porto.topspin.kiev.ua (porto-e.starpoint.kiev.ua [212.40.38.100]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id LAA07442; Fri, 19 Nov 2010 11:50:50 +0200 (EET) (envelope-from avg@freebsd.org) Received: from localhost.topspin.kiev.ua ([127.0.0.1]) by porto.topspin.kiev.ua with esmtp (Exim 4.34 (FreeBSD)) id 1PJNcI-000Lam-CN; Fri, 19 Nov 2010 11:50:50 +0200 Message-ID: <4CE64879.2060802@freebsd.org> Date: Fri, 19 Nov 2010 11:50:49 +0200 From: Andriy Gapon User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Bruce Cran References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> <20101119094652.00003652@unknown> In-Reply-To: <20101119094652.00003652@unknown> X-Enigmail-Version: 1.1.2 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 19 Nov 2010 11:58:00 +0000 Cc: Alexander Best , freebsd-performance@freebsd.org, "O. Hartmann" , Andrew Reilly Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 09:51:00 -0000 on 19/11/2010 11:46 Bruce Cran said the following: > [removed current@ and stable@ from the Cc list] > > On Fri, 19 Nov 2010 15:41:29 +1100 > Andrew Reilly wrote: > >> On Linux. Have you ever seen those sorts of UI problems on FreeBSD? >> I don't watch much video on my systems, but I haven't seen that. >> FreeBSD has always been good at keeping user-interactive processes >> responsive while compiles or what-not are going on in the background. > > I've definitely seen problems when running builds in an xterm. I've > often resorted to canceling it and running it on a syscons console > instead to improve performance. > So, what was it a problem with scheduler or with, e.g., "something X" being too slow rendering glyphs? Who can tell... -- Andriy Gapon From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 11:29:17 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0F64B1065673; Fri, 19 Nov 2010 11:29:17 +0000 (UTC) (envelope-from alexander@leidinger.net) Received: from mail.ebusiness-leidinger.de (mail.ebusiness-leidinger.de [217.11.53.44]) by mx1.freebsd.org (Postfix) with ESMTP id A7C4E8FC15; Fri, 19 Nov 2010 11:29:16 +0000 (UTC) Received: from outgoing.leidinger.net (p57B3A2B7.dip.t-dialin.net [87.179.162.183]) by mail.ebusiness-leidinger.de (Postfix) with ESMTPSA id D582384400E; Fri, 19 Nov 2010 12:09:43 +0100 (CET) Received: from webmail.leidinger.net (unknown [IPv6:fd73:10c7:2053:1::2:102]) by outgoing.leidinger.net (Postfix) with ESMTP id 5D85F118D; Fri, 19 Nov 2010 12:09:40 +0100 (CET) Received: (from www@localhost) by webmail.leidinger.net (8.14.4/8.13.8/Submit) id oAJB8xkN002093; Fri, 19 Nov 2010 12:08:59 +0100 (CET) (envelope-from Alexander@Leidinger.net) Received: from pslux.ec.europa.eu (pslux.ec.europa.eu [158.169.9.14]) by webmail.leidinger.net (Horde Framework) with HTTP; Fri, 19 Nov 2010 12:08:59 +0100 Message-ID: <20101119120859.59361egrovtna9q8@webmail.leidinger.net> Date: Fri, 19 Nov 2010 12:08:59 +0100 From: Alexander Leidinger To: Alexander Best References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> In-Reply-To: <20101119001710.GA14641@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; DelSp="Yes"; format="flowed" Content-Disposition: inline Content-Transfer-Encoding: 7bit User-Agent: Dynamic Internet Messaging Program (DIMP) H3 (1.1.4) X-EBL-MailScanner-Information: Please contact the ISP for more information X-EBL-MailScanner-ID: D582384400E.A69F4 X-EBL-MailScanner: Found to be clean X-EBL-MailScanner-SpamCheck: not spam, spamhaus-ZEN, SpamAssassin (not cached, score=1.274, required 6, autolearn=disabled, RDNS_NONE 1.27) X-EBL-MailScanner-SpamScore: s X-EBL-MailScanner-From: alexander@leidinger.net X-EBL-MailScanner-Watermark: 1290769784.20671@yoypZWdpM06a5ccjyLEuLQ X-EBL-Spam-Status: No X-Mailman-Approved-At: Fri, 19 Nov 2010 11:58:09 +0000 Cc: FreeBSD@freebsd.org, Stable , Daniel Nebdal , Julian Elischer , "Matthew D. Fuller" , Current , Gapon , FreeBSD, Andriy, freebsd-performance@freebsd.org, "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 11:29:17 -0000 Quoting Alexander Best (from Fri, 19 Nov 2010 00:17:10 +0000): > 17:51 @ Genesys : Luigi Rizzo had a plugabble scheduler back in 4.* or \ > thereabouts > 17:51 @ Genesys : you could kldload new ones and switch to them on the fly > 17:52 @ arundel : wow. that sounds cool. too bad it didn't make it > into src \ > tree. by now it's probably outdated and needs to be reworked quite a bit. > **** > > does anybody know something about this? I'm aware of the I/O scheduling code (which is now available at least in -current), but I do not remember CPU scheduling code from Luigi. Are you sure Genesys didn't mix up something by accident? Bye, Alexander. -- Ferengi Rule of Acquisition #123: Even a blind man can recognize the glow of latinum. -- ST: Legends of the Ferengi http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 13:10:48 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 04A4E1065672; Fri, 19 Nov 2010 13:10:48 +0000 (UTC) (envelope-from emss.mail@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 4FB138FC19; Fri, 19 Nov 2010 13:10:46 +0000 (UTC) Received: by wyb35 with SMTP id 35so3639283wyb.13 for ; Fri, 19 Nov 2010 05:10:45 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:received :x-virus-scanned:received:received:to:cc:subject:from:in-reply-to :references:x-operating-system:date:message-id:user-agent :mime-version:content-type:content-transfer-encoding; bh=ARCthvL34kJuHzwdGbFdIq8QLq5CkruPFrVzrKEuh5k=; b=rnGIO8/OpL7RTzPFBtAqyf4K7ZjdfTPhu0Owq+lkQ3FqfvLE7HJMwG6w21kt2hZuvN PeBtzoji9w/zv/nyV2REr9f/p5FPnTahV2L9IdXXj3hkWHJxskjERMCSykvtPObqPPRE SEqKadyJIdxtOlqHaItRjcsaoULCtAd7u2Jys= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:x-virus-scanned:to:cc:subject:from:in-reply-to:references :x-operating-system:date:message-id:user-agent:mime-version :content-type:content-transfer-encoding; b=aMbxVjew7mpPIUERohgypaeDDkB+9jAymgYowlu1rif+BSwv5Vl4R3QOCnWJriKOGR MkJqEYPeNNRv/P7EPmurfsiDsKax39sqRhYHNWkQ7Lq/MBnbiyqsB66lKznIDTCeRcE/ EhRvnWr3KQrFNquvs3OoJ7a4HnpqcWeiHMMCI= Received: by 10.227.147.209 with SMTP id m17mr2215700wbv.51.1290170546015; Fri, 19 Nov 2010 04:42:26 -0800 (PST) Received: from srvbsdfenssv.interne.associated-bears.org (LCaen-151-92-21-48.w217-128.abo.wanadoo.fr [217.128.200.48]) by mx.google.com with ESMTPS id x12sm764804weq.18.2010.11.19.04.42.23 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 19 Nov 2010 04:42:25 -0800 (PST) Sender: Eric Masson Received: from srvbsdfenssv.interne.associated-bears.org (localhost [127.0.0.1]) by srvbsdfenssv.interne.associated-bears.org (Postfix) with ESMTP id DA83B1CF6F; Fri, 19 Nov 2010 13:42:21 +0100 (CET) X-Virus-Scanned: amavisd-new at interne.associated-bears.org Received: from srvbsdfenssv.interne.associated-bears.org ([127.0.0.1]) by srvbsdfenssv.interne.associated-bears.org (srvbsdfenssv.interne.associated-bears.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DJ8aqkcGtYVY; Fri, 19 Nov 2010 13:42:18 +0100 (CET) Received: by srvbsdfenssv.interne.associated-bears.org (Postfix, from userid 1001) id 8953D1CC5B; Fri, 19 Nov 2010 13:42:18 +0100 (CET) To: Bruce Cran From: Eric Masson In-Reply-To: <20101119094207.00004cb4@unknown> (Bruce Cran's message of "Fri, 19 Nov 2010 09:42:07 +0000") References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> <20101119094207.00004cb4@unknown> X-Operating-System: FreeBSD 8.1-RELEASE-p1 amd64 Date: Fri, 19 Nov 2010 13:42:18 +0100 Message-ID: <864obdeb85.fsf@srvbsdfenssv.interne.associated-bears.org> User-Agent: Gnus/5.1008 (Gnus v5.10.8) XEmacs/21.5-b28 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 8bit X-Mailman-Approved-At: Fri, 19 Nov 2010 13:19:24 +0000 Cc: Andriy@FreeBSD.ORG, FreeBSD@freebsd.org, Stable , Daniel Nebdal , "Matthew D. Fuller" , "O. Hartmann" , Current , Gapon , Alexander Best , freebsd-performance@freebsd.org, Julian Elischer Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 13:10:48 -0000 Bruce Cran writes: Hello, > Google suggests that the work was a GSoC project in 2005 on a pluggable > disk scheduler. It seems that something similar has found its way in DFlyBSD, dsched. Éric Masson -- manquerait plus que les groupes soient pollués. c'est beaucoup plus grave que des plages bretonnes à deux francs qui étaient déjà polluées par ces salopards de volatiles. dieu merci, il n'y en aura bientôt plus -+- tilt in http://www.le-gnu.net : Les oiseaux sont des cons. From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 13:14:17 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D4D85106564A; Fri, 19 Nov 2010 13:14:17 +0000 (UTC) (envelope-from avg@freebsd.org) Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id 6F0898FC0A; Fri, 19 Nov 2010 13:14:15 +0000 (UTC) Received: from odyssey.starpoint.kiev.ua (alpha-e.starpoint.kiev.ua [212.40.38.101]) by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id PAA13146; Fri, 19 Nov 2010 15:14:13 +0200 (EET) (envelope-from avg@freebsd.org) Message-ID: <4CE67825.4010700@freebsd.org> Date: Fri, 19 Nov 2010 15:14:13 +0200 From: Andriy Gapon User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: Julian Elischer References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <4CE58A86.6010406@freebsd.org> In-Reply-To: <4CE58A86.6010406@freebsd.org> X-Enigmail-Version: 1.1.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 19 Nov 2010 13:20:57 +0000 Cc: Lucius Windschuh , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 13:14:17 -0000 on 18/11/2010 22:20 Julian Elischer said the following: > tty grouping is a variant of what we used to have at one stage which is > a "kernel schedulable entity group".. KSEG Or rather, I think, a concrete application of a variant of that. > the idea is that all items in a group share some characteristic and some amount > of resources. > > We stripped the KSEG out of the picture because it really complicated the picture. Yes, unfortunately. One can think about a number of applications for hierarchical schedulable resources. Even one-level group scheduling could be a very useful subcase. BTW, http://www.mjmwired.net/kernel/Documentation/cgroups.txt -- Andriy Gapon From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 14:18:56 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 18BED106564A; Fri, 19 Nov 2010 14:18:56 +0000 (UTC) (envelope-from vince@unsane.co.uk) Received: from unsane.co.uk (unsane-pt.tunnel.tserv5.lon1.ipv6.he.net [IPv6:2001:470:1f08:110::2]) by mx1.freebsd.org (Postfix) with ESMTP id 21E508FC14; Fri, 19 Nov 2010 14:18:54 +0000 (UTC) Received: from vhoffman-macbooklocal.local (lon.namesco.net [195.7.254.102]) (authenticated bits=0) by unsane.co.uk (8.14.4/8.14.4) with ESMTP id oAJEIqJm059907 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO); Fri, 19 Nov 2010 14:18:52 GMT (envelope-from vince@unsane.co.uk) Message-ID: <4CE6874C.1060204@unsane.co.uk> Date: Fri, 19 Nov 2010 14:18:52 +0000 From: Vincent Hoffman User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101027 Thunderbird/3.1.6 MIME-Version: 1.0 To: Eric Masson References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> <20101119094207.00004cb4@unknown> <864obdeb85.fsf@srvbsdfenssv.interne.associated-bears.org> In-Reply-To: <864obdeb85.fsf@srvbsdfenssv.interne.associated-bears.org> X-Enigmail-Version: 1.1.1 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 8bit Cc: freebsd-performance@freebsd.org, Current , Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 14:18:56 -0000 On 19/11/2010 12:42, Eric Masson wrote: > Bruce Cran writes: > > Hello, > >> Google suggests that the work was a GSoC project in 2005 on a pluggable >> disk scheduler. > It seems that something similar has found its way in DFlyBSD, dsched. And indeed to FreeBSD, man gsched. Added sometime round April http://svn.freebsd.org/viewvc/base/head/sys/geom/sched/README?view=log Vince > Éric Masson > From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 14:48:04 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id AD57C1065674 for ; Fri, 19 Nov 2010 14:48:04 +0000 (UTC) (envelope-from kostikbel@gmail.com) Received: from mail.zoral.com.ua (mx0.zoral.com.ua [91.193.166.200]) by mx1.freebsd.org (Postfix) with ESMTP id 00F4C8FC08 for ; Fri, 19 Nov 2010 14:48:03 +0000 (UTC) Received: from deviant.kiev.zoral.com.ua (root@deviant.kiev.zoral.com.ua [10.1.1.148]) by mail.zoral.com.ua (8.14.2/8.14.2) with ESMTP id oAJE8OfC060679 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Fri, 19 Nov 2010 16:08:24 +0200 (EET) (envelope-from kostikbel@gmail.com) Received: from deviant.kiev.zoral.com.ua (kostik@localhost [127.0.0.1]) by deviant.kiev.zoral.com.ua (8.14.4/8.14.4) with ESMTP id oAJE8O6h032767; Fri, 19 Nov 2010 16:08:24 +0200 (EET) (envelope-from kostikbel@gmail.com) Received: (from kostik@localhost) by deviant.kiev.zoral.com.ua (8.14.4/8.14.4/Submit) id oAJE8Omt032766; Fri, 19 Nov 2010 16:08:24 +0200 (EET) (envelope-from kostikbel@gmail.com) X-Authentication-Warning: deviant.kiev.zoral.com.ua: kostik set sender to kostikbel@gmail.com using -f Date: Fri, 19 Nov 2010 16:08:24 +0200 From: Kostik Belousov To: Andriy Gapon Message-ID: <20101119140824.GO2392@deviant.kiev.zoral.com.ua> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> <20101119094652.00003652@unknown> <4CE64879.2060802@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="mcnwp324tjjFs4IZ" Content-Disposition: inline In-Reply-To: <4CE64879.2060802@freebsd.org> User-Agent: Mutt/1.4.2.3i X-Virus-Scanned: clamav-milter 0.95.2 at skuns.kiev.zoral.com.ua X-Virus-Status: Clean X-Spam-Status: No, score=-3.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00, DNS_FROM_OPENWHOIS autolearn=no version=3.2.5 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on skuns.kiev.zoral.com.ua Cc: freebsd-performance@freebsd.org Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 14:48:04 -0000 --mcnwp324tjjFs4IZ Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Nov 19, 2010 at 11:50:49AM +0200, Andriy Gapon wrote: > on 19/11/2010 11:46 Bruce Cran said the following: > > [removed current@ and stable@ from the Cc list] > >=20 > > On Fri, 19 Nov 2010 15:41:29 +1100 > > Andrew Reilly wrote: > >=20 > >> On Linux. Have you ever seen those sorts of UI problems on FreeBSD? > >> I don't watch much video on my systems, but I haven't seen that. > >> FreeBSD has always been good at keeping user-interactive processes > >> responsive while compiles or what-not are going on in the background. > >=20 > > I've definitely seen problems when running builds in an xterm. I've > > often resorted to canceling it and running it on a syscons console > > instead to improve performance. > >=20 >=20 > So, what was it a problem with scheduler or with, e.g., "something X" > being too slow rendering glyphs? Who can tell... Probably will pay a lot in negative karma by posting anything in the thread. But I can confirm your words, that tty->xterm->X server chain of output indeed significantly slows down the build processes. I usually never start build in the barebone xterm, always running screen under xterm. make -j 10 on 4 core/HTT cpu slows up to a half, from my unscientific impression, when run in the active screen window. Switching to other window in screen significantly speeds it up (note the prudent omission of any measured numbers). --mcnwp324tjjFs4IZ Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (FreeBSD) iEYEARECAAYFAkzmhNgACgkQC3+MBN1Mb4j6BwCg4JGYHhi3OwCkYR0ccv5ro5Ki 8dgAoOsRIy9Sj0SZl8QYOVKghzkcpEYn =QAjw -----END PGP SIGNATURE----- --mcnwp324tjjFs4IZ-- From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 14:44:54 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C994A106564A for ; Fri, 19 Nov 2010 14:44:54 +0000 (UTC) (envelope-from jdc@koitsu.dyndns.org) Received: from qmta04.emeryville.ca.mail.comcast.net (qmta04.emeryville.ca.mail.comcast.net [76.96.30.40]) by mx1.freebsd.org (Postfix) with ESMTP id A981A8FC0A for ; Fri, 19 Nov 2010 14:44:54 +0000 (UTC) Received: from omta14.emeryville.ca.mail.comcast.net ([76.96.30.60]) by qmta04.emeryville.ca.mail.comcast.net with comcast id Z1gz1f0051HpZEsA42ku4R; Fri, 19 Nov 2010 14:44:54 +0000 Received: from koitsu.dyndns.org ([98.248.41.155]) by omta14.emeryville.ca.mail.comcast.net with comcast id Z2ks1f00K3LrwQ28a2ktR8; Fri, 19 Nov 2010 14:44:53 +0000 Received: by icarus.home.lan (Postfix, from userid 1000) id B60DD9B427; Fri, 19 Nov 2010 06:44:52 -0800 (PST) Date: Fri, 19 Nov 2010 06:44:52 -0800 From: Jeremy Chadwick To: Vincent Hoffman Message-ID: <20101119144452.GA67750@icarus.home.lan> References: <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> <20101119094207.00004cb4@unknown> <864obdeb85.fsf@srvbsdfenssv.interne.associated-bears.org> <4CE6874C.1060204@unsane.co.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4CE6874C.1060204@unsane.co.uk> User-Agent: Mutt/1.5.21 (2010-09-15) X-Mailman-Approved-At: Fri, 19 Nov 2010 15:19:32 +0000 Cc: Eric Masson , freebsd-performance@freebsd.org, Current , Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 14:44:54 -0000 On Fri, Nov 19, 2010 at 02:18:52PM +0000, Vincent Hoffman wrote: > On 19/11/2010 12:42, Eric Masson wrote: > > Bruce Cran writes: > > > > Hello, > > > >> Google suggests that the work was a GSoC project in 2005 on a pluggable > >> disk scheduler. > > It seems that something similar has found its way in DFlyBSD, dsched. > And indeed to FreeBSD, man gsched. Added sometime round April > http://svn.freebsd.org/viewvc/base/head/sys/geom/sched/README?view=log It's been pointed out on the list a couple times, and I've sent mail to the authors about this, that gsched breaks (very, very badly) things like sysinstall, and does other strange things like leaves trailing periods at the end of its ".sched." labels. This appears to be by design, but I'm still left thinking "?!" It's hard to discern technical innards/workings of GEOM since the documentation is so poor (and reading the code doesn't help, especially with regards to libgeom). IMHO, the gsched "stuff", as a "layer", should probably be moved into the I/O framework by default, with the functionality *disabled* by default and tunables to adjust it. That's just how I feel about it. -- | Jeremy Chadwick jdc@parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 15:49:59 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E6BCD106564A; Fri, 19 Nov 2010 15:49:58 +0000 (UTC) (envelope-from taku@tackymt.homeip.net) Received: from basalt.tackymt.homeip.net (unknown [IPv6:2001:3e0:577:0:20d:61ff:fecc:2253]) by mx1.freebsd.org (Postfix) with ESMTP id 82A588FC08; Fri, 19 Nov 2010 15:49:58 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by basalt.tackymt.homeip.net (Postfix) with ESMTP id B141813324; Sat, 20 Nov 2010 00:49:57 +0900 (JST) X-Virus-Scanned: amavisd-new at tackymt.homeip.net Received: from localhost ([127.0.0.1]) by localhost (basalt.tackymt.homeip.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 41fp7la0-sfI; Sat, 20 Nov 2010 00:49:56 +0900 (JST) Received: from basalt.tackymt.homeip.net (basalt.tackymt.homeip.net [IPv6:2001:3e0:577:0:20d:61ff:fecc:2253]) by basalt.tackymt.homeip.net (Postfix) with ESMTP; Sat, 20 Nov 2010 00:49:56 +0900 (JST) Date: Sat, 20 Nov 2010 00:49:55 +0900 From: Taku YAMAMOTO To: "O. Hartmann" Message-Id: <20101120004955.68c8af6a.taku@tackymt.homeip.net> In-Reply-To: <4CE58CD8.2000407@mail.zedat.fu-berlin.de> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <4CE58CD8.2000407@mail.zedat.fu-berlin.de> X-Mailer: Sylpheed 3.0.3 (GTK+ 2.20.1; i386-portbld-freebsd9.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Lucius Windschuh , freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable , Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 15:49:59 -0000 On Thu, 18 Nov 2010 21:30:16 +0100 "O. Hartmann" wrote: > On 11/18/10 19:55, Lucius Windschuh wrote: > > 2010/11/18 Andriy Gapon: > >> [Grouping of processes into TTY groups] > >> > >> Well, I think that those improvements apply only to a very specific usage pattern > >> and are greatly over-hyped. > > > > But there are serious issue if you use FreeBSD as a desktop OS with > > SMP and SCHED_ULE, or? > > Because currently, my machine is barely usable if a compile job with > > parallelism is running. Movies stutter, Firefox hangs. And even nice > > -n 20 doesn't do the job in every case, as +20 seems not to be the > > idle priority anymore?!? > > And using "idprio 1 $cmd" as a workaround is, well, a kludge. > > I am not sure if TTY grouping is the right solution, if you look at > > potentially CPU-intensive GUI applications that all run on the same > > TTY (or no TTY at all? Same problem). > > Maybe, we could simply enhance the algorithm that decides if a task is > > interactive? That would also improve the described situation. > > > > Regards, > > > > Lucius > > Stuttering Response, being stuck for over 20 seconds also happens when I > start updating the OS' sources via svn. This happens on all boxes, some > of them do have 8 cores (ob two CPUs) and plenty of RAM. Heavy disk I/O, > doesn't matter on UFS2 or ZFS, also brings boxes to stutter, those > phenomena are most seen when you interact with the machine via X11 > clients. I think it's hard to realize if a server only does console I/O, > but console also seems to be stuck sometimes. It would be worth checking > this with some 'benchmark'. X11 in its kind of oldish incarnation on > FreeBSD seems to contribute most to those slowdowns, what so ever. I guess schedulers can hardly distinguish heavy disk I/Os from nanosleep()s and user-interactions; schedulers think both as voluntary sleep. To make the matters worse, the current implementation of SCHED_ULE reassigns ts_slice on sched_wakeup() no matter how short the sleep was. I have a dumb local hack to grant ts_slice proportional to the duration the waking thread slept rather than unconditionally reset to sched_slice. --- sys/kern/sched_ule.c.orig +++ sys/kern/sched_ule.c @@ -1928,12 +1928,16 @@ sched_wakeup(struct thread *td) u_int hzticks; hzticks = (ticks - slptick) << SCHED_TICK_SHIFT; + if (hzticks > SCHED_SLP_RUN_MAX) + hzticks = SCHED_SLP_RUN_MAX; ts->ts_slptime += hzticks; + /* Grant additional slices after we sleep. */ + ts->ts_slice += hzticks / tickincr; + if (ts->ts_slice > sched_slice) + ts->ts_slice = sched_slice; sched_interact_update(td); sched_pctcpu_update(ts); } - /* Reset the slice value after we sleep. */ - ts->ts_slice = sched_slice; sched_add(td, SRQ_BORING); } -- -|-__ YAMAMOTO, Taku | __ < - A chicken is an egg's way of producing more eggs. - From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 18:12:43 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 090EB106566B for ; Fri, 19 Nov 2010 18:12:43 +0000 (UTC) (envelope-from dieterbsd@engineer.com) Received: from imr-db03.mx.aol.com (imr-db03.mx.aol.com [205.188.91.97]) by mx1.freebsd.org (Postfix) with ESMTP id BCE878FC0C for ; Fri, 19 Nov 2010 18:12:42 +0000 (UTC) Received: from imo-da02.mx.aol.com (imo-da02.mx.aol.com [205.188.169.200]) by imr-db03.mx.aol.com (8.14.1/8.14.1) with ESMTP id oAJICevO005246 for ; Fri, 19 Nov 2010 13:12:40 -0500 Received: from dieterbsd@engineer.com by imo-da02.mx.aol.com (mail_out_v42.9.) id n.f4c.82607ab (45477) for ; Fri, 19 Nov 2010 13:12:34 -0500 (EST) Received: from smtprly-ma03.mx.aol.com (smtprly-ma03.mx.aol.com [64.12.207.142]) by cia-mc06.mx.aol.com (v129.7) with ESMTP id MAILCIAMC066-5c564ce6be0add; Fri, 19 Nov 2010 13:12:33 -0500 Received: from web-mmc-d05 (web-mmc-d05.sim.aol.com [205.188.103.95]) by smtprly-ma03.mx.aol.com (v129.5) with ESMTP id MAILSMTPRLYMA034-5c564ce6be0add; Fri, 19 Nov 2010 13:12:26 -0500 To: freebsd-performance@freebsd.org Content-Transfer-Encoding: quoted-printable Date: Fri, 19 Nov 2010 13:12:26 -0500 X-MB-Message-Source: WebUI X-AOL-IP: 67.206.164.39 X-MB-Message-Type: User MIME-Version: 1.0 From: dieterbsd@engineer.com Content-Type: text/plain; charset="us-ascii"; format=flowed X-Mailer: Mail.com Webmail 32945-STANDARD Received: from 67.206.164.39 by web-mmc-d05.sysops.aol.com (205.188.103.95) with HTTP (WebMailUI); Fri, 19 Nov 2010 13:12:26 -0500 Message-Id: <8CD562C7462116D-1204-3BE9@web-mmc-d05.sysops.aol.com> X-Spam-Flag: NO X-AOL-SENDER: dieterbsd@engineer.com Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 18:12:43 -0000 Alexander writes: > One thing that just begs to be asked: since when decoding 1080p became > an interactive task? Normally, decoding video would not be considered an interactive task, unless you are doing things like stepping through frame-by-frame. Playing video, and/or audio is a true real time task. The computer must perform the work at a specific rate, not faster, not slower. Recording video and/or audio is also a true real time task, If the machine/OS drops the ball you have a spoiled recording, and most of the time you can't try again. Bruce writes: > you > can have a load of 100 or more before the system becomes unusable > whereas people are amazed to see loads of more than 15 on Linux. The load average leaves much to be desired as a metric. I have generated an obscenely high lead average while retaining great responsiveness, and done a simple "cp /disk1/file1 /disk2/" resulting in the machine taking *minutes* to respond to the simplest command. Andriy writes: > Well, I am not sure if I can agree about CPU-bound-ness. > Depends on the exact video file, of course, but certain high-quality=20 1080p are > very CPU intensive unless decoding is offloaded from the CPU. =20 Depends on > decoder code too. I had some videos that were CPU-bound on my Athlon=20 II X2 250 > with then-version of mplayer from ports. The bitrate is more useful than saying "1080p". The speed of the CPU is important, the codec, the efficiency of the decoder, whether parts can=20 be offloaded to a GPU or hardware decoder, if the output is being scaled,=20 etc. If playing a video maxes out the CPU, then the video isn't going to be displayed correctly. From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 21:35:07 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id BADF1106566B; Fri, 19 Nov 2010 21:35:07 +0000 (UTC) (envelope-from oliver.pntr@gmail.com) Received: from mail-ww0-f42.google.com (mail-ww0-f42.google.com [74.125.82.42]) by mx1.freebsd.org (Postfix) with ESMTP id 045E68FC18; Fri, 19 Nov 2010 21:35:06 +0000 (UTC) Received: by wwb17 with SMTP id 17so761067wwb.1 for ; Fri, 19 Nov 2010 13:35:06 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=PsIHBlWEyPmN6HqKUL3brVtQee4UfxNPxql8LwGgqG0=; b=o3RDokiyqa8gOh1254/uN8hkdYQH1LoDRFWFpVIFjQhXkvNK3P4N0bP93X6xOnPJrw ZNwq0fR4XnJ1H2qJRxdDgoUAF33UpR6eysqJmP5jPlwoNPEJ+W2NKFQnaXzgzLDEvd/A IpohYac81miTFPnUD1MExXEbLyqZLINR888tg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=NTpl+pc0WxZCsTMPRIGoVbi3bhyY14lCm8lufs+O/SaYplA/tobR4A+HE8dTomzgzH 1Uxa2DUvb8UIX/oo1zNF9/MEbrishCFW9F01K7o5yXPd0Zzt3NyjtcRgM6qTIRITPb1y DAXcwukR3+ohIrg0buj27PbiWtJrVRLHAjYVE= MIME-Version: 1.0 Received: by 10.227.142.208 with SMTP id r16mr2878057wbu.140.1290201049361; Fri, 19 Nov 2010 13:10:49 -0800 (PST) Received: by 10.227.133.66 with HTTP; Fri, 19 Nov 2010 13:10:48 -0800 (PST) In-Reply-To: <4CE50849.106@zedat.fu-berlin.de> References: <4CE50849.106@zedat.fu-berlin.de> Date: Fri, 19 Nov 2010 22:10:48 +0100 Message-ID: From: Oliver Pinter To: "O. Hartmann" Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Fri, 19 Nov 2010 21:37:41 +0000 Cc: freebsd-performance@freebsd.org, FreeBSD Current , FreeBSD Stable Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 21:35:07 -0000 http://lkml.org/lkml/2010/11/16/392 On 11/18/10, O. Hartmann wrote: > On 11/18/10 02:30, grarpamp wrote: >> Just documenting regarding interactive performance things. >> This one's from Linux. >> >> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 > > Well, > it would be nice to have those improvements in FreeBSD, but I doubt this > will make it in due time to FreeBSD's kernel. > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" > From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 22:47:20 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C1FDD106566C; Fri, 19 Nov 2010 22:47:20 +0000 (UTC) (envelope-from yanegomi@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 08B818FC16; Fri, 19 Nov 2010 22:47:19 +0000 (UTC) Received: by wyb35 with SMTP id 35so4204311wyb.13 for ; Fri, 19 Nov 2010 14:47:19 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:sender:received :in-reply-to:references:date:x-google-sender-auth:message-id:subject :from:to:cc:content-type; bh=OKivmdL38kmUllIet02P14SJJSx6J4tTwf29wTSZyu0=; b=qv+BR7YYdKqvLC+3pbAo9AyERTfIbpoqX7YRxEkMNTw1F/VVLB83cG06psQIPQS0xQ N2x7G9lF3tYdZNhTC0h06G7t7KOo/uLL7qolTQNhvNtgItG8WT521pSoruF0op9/q4tD vDlPj+sD8v/GCaEQQx1vBOcDiUzSNHn1hSvcE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; b=iqROAnHc5dMouhtLYDBTThJ5Jpa5eD6/cnxCOvyzSVsUJBxvE3VKLKMumQvFFmPxvJ r/lIdUnzOsXZOaeCYn31C6OjEx8ngu1aJeLA3zML9dl3FGi42uNsiMR9lXdyw7+mQBbh G33tdjUvBkSs7NIiIwK/YJSv2xbbqWXr1NaF0= MIME-Version: 1.0 Received: by 10.216.50.134 with SMTP id z6mr3080976web.15.1290206834004; Fri, 19 Nov 2010 14:47:14 -0800 (PST) Sender: yanegomi@gmail.com Received: by 10.216.198.27 with HTTP; Fri, 19 Nov 2010 14:46:51 -0800 (PST) In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> Date: Fri, 19 Nov 2010 14:46:51 -0800 X-Google-Sender-Auth: UWBQV1PqdJN4_V7UOuWPfL6gjXY Message-ID: From: Garrett Cooper To: Oliver Pinter Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Fri, 19 Nov 2010 22:52:29 +0000 Cc: FreeBSD Stable , freebsd-performance@freebsd.org, FreeBSD Current , "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 22:47:21 -0000 On Fri, Nov 19, 2010 at 1:10 PM, Oliver Pinter wrote: > http://lkml.org/lkml/2010/11/16/392 > > On 11/18/10, O. Hartmann wrote: >> On 11/18/10 02:30, grarpamp wrote: >>> Just documenting regarding interactive performance things. >>> This one's from Linux. >>> >>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 >> >> Well, >> it would be nice to have those improvements in FreeBSD, but I doubt this >> will make it in due time to FreeBSD's kernel. And my one line fix: renice 10 `pidof firefox-bin` Instantly my system is snappier (and in fact my system got really laggy after applying the preempt sysctl that everyone recommended before)... Performance issue with firefox maybe :P? I don't see the point of adding an additional layer to complicate the system (and essentially slow it down) if all you're trying to do is better describe the nice'ing problem, unless this logic is what you want to do strictly for desktop users in PCBSD, etc who may not have the technical wherewithal to accomplish this task. Besides, the Linux kernel has different compile time profiles for different workloads, so maybe it just works better for them because they already have a means for describing that functionality, whereas FreeBSD is more generic. It would be nice to describe this in a document though so people could also decide how to tune the system for themselves and not deal with a patch like what's noted above by the penguin crowd because it will invariably fail under some workloads or conditions (I have yet to see a one-size-fits-all solution in this area). SCHED_ULE improvements though should be looked into if possible, because there are some potential items that could be done to cluster processes together better, maybe. Thanks, -Garrett From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 22:48:52 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 66C681065670; Fri, 19 Nov 2010 22:48:52 +0000 (UTC) (envelope-from yanegomi@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 9EFDE8FC21; Fri, 19 Nov 2010 22:48:51 +0000 (UTC) Received: by wyb35 with SMTP id 35so4205885wyb.13 for ; Fri, 19 Nov 2010 14:48:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:sender:received :in-reply-to:references:date:x-google-sender-auth:message-id:subject :from:to:cc:content-type; bh=OKivmdL38kmUllIet02P14SJJSx6J4tTwf29wTSZyu0=; b=CxOCWEM4tSuYbm8sJrVTwvR49GlQL1vNjymk+PQ1OMRJShd41sUnTM57KnuR2DqGEX nNzq2h/PzPSQIvDWy//va9OHFppkqd0wXeRsRNKlbiaOStDh57+WpfIcwNdsMViFXNmC BJSUa77I4UrBaJ63oz9kZsUFrGENHCOhYgJEI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; b=iqROAnHc5dMouhtLYDBTThJ5Jpa5eD6/cnxCOvyzSVsUJBxvE3VKLKMumQvFFmPxvJ r/lIdUnzOsXZOaeCYn31C6OjEx8ngu1aJeLA3zML9dl3FGi42uNsiMR9lXdyw7+mQBbh G33tdjUvBkSs7NIiIwK/YJSv2xbbqWXr1NaF0= MIME-Version: 1.0 Received: by 10.216.46.200 with SMTP id r50mr1631469web.45.1290206829686; Fri, 19 Nov 2010 14:47:09 -0800 (PST) Sender: yanegomi@gmail.com Received: by 10.216.198.27 with HTTP; Fri, 19 Nov 2010 14:46:51 -0800 (PST) In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> Date: Fri, 19 Nov 2010 14:46:51 -0800 X-Google-Sender-Auth: UWBQV1PqdJN4_V7UOuWPfL6gjXY Message-ID: From: Garrett Cooper To: Oliver Pinter Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Fri, 19 Nov 2010 23:12:44 +0000 Cc: FreeBSD Stable , freebsd-performance@freebsd.org, FreeBSD Current , "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 22:48:52 -0000 On Fri, Nov 19, 2010 at 1:10 PM, Oliver Pinter wrote: > http://lkml.org/lkml/2010/11/16/392 > > On 11/18/10, O. Hartmann wrote: >> On 11/18/10 02:30, grarpamp wrote: >>> Just documenting regarding interactive performance things. >>> This one's from Linux. >>> >>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 >> >> Well, >> it would be nice to have those improvements in FreeBSD, but I doubt this >> will make it in due time to FreeBSD's kernel. And my one line fix: renice 10 `pidof firefox-bin` Instantly my system is snappier (and in fact my system got really laggy after applying the preempt sysctl that everyone recommended before)... Performance issue with firefox maybe :P? I don't see the point of adding an additional layer to complicate the system (and essentially slow it down) if all you're trying to do is better describe the nice'ing problem, unless this logic is what you want to do strictly for desktop users in PCBSD, etc who may not have the technical wherewithal to accomplish this task. Besides, the Linux kernel has different compile time profiles for different workloads, so maybe it just works better for them because they already have a means for describing that functionality, whereas FreeBSD is more generic. It would be nice to describe this in a document though so people could also decide how to tune the system for themselves and not deal with a patch like what's noted above by the penguin crowd because it will invariably fail under some workloads or conditions (I have yet to see a one-size-fits-all solution in this area). SCHED_ULE improvements though should be looked into if possible, because there are some potential items that could be done to cluster processes together better, maybe. Thanks, -Garrett From owner-freebsd-performance@FreeBSD.ORG Fri Nov 19 23:04:09 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 811B2106566B for ; Fri, 19 Nov 2010 23:04:09 +0000 (UTC) (envelope-from dan@dan.emsphone.com) Received: from email2.allantgroup.com (email2.emsphone.com [199.67.51.116]) by mx1.freebsd.org (Postfix) with ESMTP id 212598FC17 for ; Fri, 19 Nov 2010 23:04:08 +0000 (UTC) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by email2.allantgroup.com (8.14.4/8.14.4) with ESMTP id oAJMouKX067270 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Fri, 19 Nov 2010 16:50:56 -0600 (CST) (envelope-from dan@dan.emsphone.com) Received: from dan.emsphone.com (smmsp@localhost [127.0.0.1]) by dan.emsphone.com (8.14.4/8.14.4) with ESMTP id oAJMotOo088726 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Fri, 19 Nov 2010 16:50:56 -0600 (CST) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.14.4/8.14.4/Submit) id oAJMdriZ056350; Fri, 19 Nov 2010 16:39:53 -0600 (CST) (envelope-from dan) Date: Fri, 19 Nov 2010 16:39:52 -0600 From: Dan Nelson To: Alexander Leidinger Message-ID: <20101119223952.GK57869@dan.emsphone.com> References: <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101118182852.GR63683@over-yonder.net> <20101118185635.GA43706@freebsd.org> <20101118170623.7f9c14f3@kan.dnsalias.net> <20101118233731.GA10392@freebsd.org> <4CE5BA37.20604@freebsd.org> <20101119001710.GA14641@freebsd.org> <20101119120859.59361egrovtna9q8@webmail.leidinger.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101119120859.59361egrovtna9q8@webmail.leidinger.net> X-OS: FreeBSD 8.1-STABLE User-Agent: Mutt/1.5.21 (2010-09-15) X-Virus-Scanned: clamav-milter 0.96.4 at email2.allantgroup.com X-Virus-Status: Clean X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.6 (email2.allantgroup.com [199.67.51.78]); Fri, 19 Nov 2010 16:50:56 -0600 (CST) X-Scanned-By: MIMEDefang 2.68 on 199.67.51.78 X-Mailman-Approved-At: Sat, 20 Nov 2010 02:43:52 +0000 Cc: Andriy@freebsd.org, FreeBSD@freebsd.org, Stable , "O. Hartmann" , "Matthew D. Fuller" , Current , Gapon , Alexander Best , freebsd-performance@freebsd.org, Daniel Nebdal Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Nov 2010 23:04:09 -0000 In the last episode (Nov 19), Alexander Leidinger said: > Quoting Alexander Best (from Fri, 19 Nov 2010 00:17:10 +0000): > > 17:51 @ Genesys : Luigi Rizzo had a plugabble scheduler back in 4.* or thereabouts > > 17:51 @ Genesys : you could kldload new ones and switch to them on the fly > > 17:52 @ arundel : wow. that sounds cool. too bad it didn't make it > > into src tree. by now it's probably outdated and needs to be reworked quite a bit. > > **** > > > > does anybody know something about this? > > I'm aware of the I/O scheduling code (which is now available at least > in -current), but I do not remember CPU scheduling code from Luigi. > Are you sure Genesys didn't mix up something by accident? I am rarely mixed up :) A quick search didn't bring up a direct reference, but here's a mention of it from Luigi: http://lists.freebsd.org/pipermail/freebsd-hackers/2004-November/008891.html -- Dan Nelson dnelson@allantgroup.com From owner-freebsd-performance@FreeBSD.ORG Sat Nov 20 01:28:04 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id ED826106566B; Sat, 20 Nov 2010 01:28:03 +0000 (UTC) (envelope-from oliver.pntr@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 0A4C68FC1B; Sat, 20 Nov 2010 01:28:02 +0000 (UTC) Received: by wyb35 with SMTP id 35so4319483wyb.13 for ; Fri, 19 Nov 2010 17:28:01 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=tdNpb1QEuoGB8Gal23sGmrl4q11vEcMsiq1Tq+aaUCs=; b=u0m0jsJkKbk+PawPHQtsBK7Wbjd0FFQaEuPHUItchgBNsl6hKhmrEG8rsmwxhA//R4 dLj5ED+6p2cyR5BgHZ9JFGgl4E87LUZ6AVRRML9y/Zgau7r2WIaSYoN1QQtPZuxzlbIn be08Us0ADuPSQ9ogf8YFdSFMRXcah8QFw0mu8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=vnwp9t8j3YW/jfdYkbfTcGhsfFoF60wwhFRslJqdXzCGm5aT+1edxgw31BEiuZEkzS OkD6T9ovRkjzyIJPvbkQGDJcRkBXdWfuykK3utWAMLp7fgk0sne15u+CbF0615D5PeUW ZFmtg6rj5aPUuuYb6IY4luI5A04j2NlJCYDwI= MIME-Version: 1.0 Received: by 10.227.157.133 with SMTP id b5mr3103975wbx.53.1290216480334; Fri, 19 Nov 2010 17:28:00 -0800 (PST) Received: by 10.227.133.66 with HTTP; Fri, 19 Nov 2010 17:27:59 -0800 (PST) In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> Date: Sat, 20 Nov 2010 02:27:59 +0100 Message-ID: From: Oliver Pinter To: Garrett Cooper Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Sat, 20 Nov 2010 03:04:50 +0000 Cc: FreeBSD Stable , freebsd-performance@freebsd.org, FreeBSD Current , "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Nov 2010 01:28:04 -0000 My desktop running 7-STABLE with 100Hz and NOPREEMPT (it's a 4core SMP system), I tested 8-STABLE, but that is not too responsive, the solution is: 100Hz NOPREEMPT + kern.sched.preempt_thresh=224 After this setting, the system is likely responsive as 7-STABLE. On 11/19/10, Garrett Cooper wrote: > On Fri, Nov 19, 2010 at 1:10 PM, Oliver Pinter > wrote: >> http://lkml.org/lkml/2010/11/16/392 >> >> On 11/18/10, O. Hartmann wrote: >>> On 11/18/10 02:30, grarpamp wrote: >>>> Just documenting regarding interactive performance things. >>>> This one's from Linux. >>>> >>>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 >>> >>> Well, >>> it would be nice to have those improvements in FreeBSD, but I doubt this >>> will make it in due time to FreeBSD's kernel. > > And my one line fix: > > renice 10 `pidof firefox-bin` > > Instantly my system is snappier (and in fact my system got really > laggy after applying the preempt sysctl that everyone recommended > before)... Performance issue with firefox maybe :P? I don't see the > point of adding an additional layer to complicate the system (and > essentially slow it down) if all you're trying to do is better > describe the nice'ing problem, unless this logic is what you want to > do strictly for desktop users in PCBSD, etc who may not have the > technical wherewithal to accomplish this task. > > Besides, the Linux kernel has different compile time profiles for > different workloads, so maybe it just works better for them because > they already have a means for describing that functionality, whereas > FreeBSD is more generic. > > It would be nice to describe this in a document though so people could > also decide how to tune the system for themselves and not deal with a > patch like what's noted above by the penguin crowd because it will > invariably fail under some workloads or conditions (I have yet to see > a one-size-fits-all solution in this area). > > > SCHED_ULE improvements though should be looked into if possible, > because there are some potential items that could be done to cluster > processes together better, maybe. > > > Thanks, > -Garrett > From owner-freebsd-performance@FreeBSD.ORG Sat Nov 20 01:49:58 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0CFD61065670; Sat, 20 Nov 2010 01:49:58 +0000 (UTC) (envelope-from yanegomi@gmail.com) Received: from mail-wy0-f182.google.com (mail-wy0-f182.google.com [74.125.82.182]) by mx1.freebsd.org (Postfix) with ESMTP id 45DA08FC18; Sat, 20 Nov 2010 01:49:56 +0000 (UTC) Received: by wyb35 with SMTP id 35so4331934wyb.13 for ; Fri, 19 Nov 2010 17:49:56 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:sender:received :in-reply-to:references:date:x-google-sender-auth:message-id:subject :from:to:cc:content-type; bh=wZO0ulBRbTTkGcgy2Q32ThZN1QGyz5faGrbY6qdrhk4=; b=uMvThrM186fUIob7w7ojR5UPi6e+LsRZBxYjxmecGe53nMLZV7NcvbmWSIbIn6/Q36 tDFf7pr/s2OZPFI+ReQ8Jej7NPYOA8mItz7EwRqGIjRRCcYaBifN4IetxHTWhiczPM9k kKuDEdoFgnFI/TvI9p+77w5/mkP3PcF+xq3nI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; b=QvOVQpuXF+ey5BdpdazZZWZEN5lEbfh7yHaXmQlED6WoTcrqpiYuW78hwoRMr2pThB TP9mMEO2wNMQzeiuUda4qpQPt/gulAT6pVzM/nyyNnwniq5k16Qh4aihLshY8EpR73/T EZvzUMGrY2qjvc90vH3hg3sRXrb27frZ5xJC4= MIME-Version: 1.0 Received: by 10.216.7.210 with SMTP id 60mr1774698wep.30.1290217795101; Fri, 19 Nov 2010 17:49:55 -0800 (PST) Sender: yanegomi@gmail.com Received: by 10.216.198.27 with HTTP; Fri, 19 Nov 2010 17:49:55 -0800 (PST) In-Reply-To: References: <4CE50849.106@zedat.fu-berlin.de> Date: Fri, 19 Nov 2010 17:49:55 -0800 X-Google-Sender-Auth: Hg2MQh2qNfgfohc6PYyqzyksUxo Message-ID: From: Garrett Cooper To: Oliver Pinter Content-Type: text/plain; charset=ISO-8859-1 X-Mailman-Approved-At: Sat, 20 Nov 2010 03:05:00 +0000 Cc: FreeBSD Stable , freebsd-performance@freebsd.org, FreeBSD Current , "O. Hartmann" Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Nov 2010 01:49:58 -0000 On Fri, Nov 19, 2010 at 5:27 PM, Oliver Pinter wrote: > My desktop running 7-STABLE with 100Hz and NOPREEMPT (it's a 4core SMP system), > I tested 8-STABLE, but that is not too responsive, the solution is: > 100Hz NOPREEMPT + kern.sched.preempt_thresh=224 > After this setting, the system is likely responsive as 7-STABLE. > > On 11/19/10, Garrett Cooper wrote: >> On Fri, Nov 19, 2010 at 1:10 PM, Oliver Pinter >> wrote: >>> http://lkml.org/lkml/2010/11/16/392 >>> >>> On 11/18/10, O. Hartmann wrote: >>>> On 11/18/10 02:30, grarpamp wrote: >>>>> Just documenting regarding interactive performance things. >>>>> This one's from Linux. >>>>> >>>>> http://www.phoronix.com/scan.php?page=article&item=linux_2637_video&num=1 >>>> >>>> Well, >>>> it would be nice to have those improvements in FreeBSD, but I doubt this >>>> will make it in due time to FreeBSD's kernel. >> >> And my one line fix: >> >> renice 10 `pidof firefox-bin` >> >> Instantly my system is snappier (and in fact my system got really >> laggy after applying the preempt sysctl that everyone recommended >> before)... Performance issue with firefox maybe :P? I don't see the >> point of adding an additional layer to complicate the system (and >> essentially slow it down) if all you're trying to do is better >> describe the nice'ing problem, unless this logic is what you want to >> do strictly for desktop users in PCBSD, etc who may not have the >> technical wherewithal to accomplish this task. >> >> Besides, the Linux kernel has different compile time profiles for >> different workloads, so maybe it just works better for them because >> they already have a means for describing that functionality, whereas >> FreeBSD is more generic. >> >> It would be nice to describe this in a document though so people could >> also decide how to tune the system for themselves and not deal with a >> patch like what's noted above by the penguin crowd because it will >> invariably fail under some workloads or conditions (I have yet to see >> a one-size-fits-all solution in this area). >> >> >> SCHED_ULE improvements though should be looked into if possible, >> because there are some potential items that could be done to cluster >> processes together better, maybe. >> Ugh. Looks like this was the only machine that I setup recently where I didn't set kern.hz... Ok, will retry after building and installing a whole new world *queue lame Disney music here* and kernel. Thanks for the obvious reminder... -Garrett From owner-freebsd-performance@FreeBSD.ORG Sat Nov 20 14:31:09 2010 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 04C6B106564A for ; Sat, 20 Nov 2010 14:31:09 +0000 (UTC) (envelope-from brde@optusnet.com.au) Received: from fallbackmx07.syd.optusnet.com.au (fallbackmx07.syd.optusnet.com.au [211.29.132.9]) by mx1.freebsd.org (Postfix) with ESMTP id 814408FC29 for ; Sat, 20 Nov 2010 14:31:08 +0000 (UTC) Received: from mail06.syd.optusnet.com.au (mail06.syd.optusnet.com.au [211.29.132.187]) by fallbackmx07.syd.optusnet.com.au (8.13.1/8.13.1) with ESMTP id oAKCcsXx025118 for ; Sat, 20 Nov 2010 23:38:54 +1100 Received: from c122-106-145-124.carlnfd1.nsw.optusnet.com.au (c122-106-145-124.carlnfd1.nsw.optusnet.com.au [122.106.145.124]) by mail06.syd.optusnet.com.au (8.13.1/8.13.1) with ESMTP id oAKCcoPU005149 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sat, 20 Nov 2010 23:38:51 +1100 Date: Sat, 20 Nov 2010 23:38:50 +1100 (EST) From: Bruce Evans X-X-Sender: bde@besplex.bde.org To: Kostik Belousov In-Reply-To: <20101119140824.GO2392@deviant.kiev.zoral.com.ua> Message-ID: <20101120223925.T25498@besplex.bde.org> References: <4CE50849.106@zedat.fu-berlin.de> <4CE52177.3020306@freebsd.org> <20101118182324.GA36312@freebsd.org> <20101119044129.GA4063@johnny.reilly.home> <20101119094652.00003652@unknown> <4CE64879.2060802@freebsd.org> <20101119140824.GO2392@deviant.kiev.zoral.com.ua> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-performance@freebsd.org, Andriy Gapon Subject: Re: TTY task group scheduling X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Nov 2010 14:31:09 -0000 On Fri, 19 Nov 2010, Kostik Belousov wrote: > On Fri, Nov 19, 2010 at 11:50:49AM +0200, Andriy Gapon wrote: >> on 19/11/2010 11:46 Bruce Cran said the following: >>> [removed current@ and stable@ from the Cc list] >>> >>> On Fri, 19 Nov 2010 15:41:29 +1100 >>> Andrew Reilly wrote: >>> >>>> On Linux. Have you ever seen those sorts of UI problems on FreeBSD? Not since FreeBSD-1 or earlier, but I don't run much bloatware. >>>> I don't watch much video on my systems, but I haven't seen that. >>>> FreeBSD has always been good at keeping user-interactive processes >>>> responsive while compiles or what-not are going on in the background. >>> >>> I've definitely seen problems when running builds in an xterm. I've >>> often resorted to canceling it and running it on a syscons console >>> instead to improve performance. >> >> So, what was it a problem with scheduler or with, e.g., "something X" >> being too slow rendering glyphs? Who can tell... > > Probably will pay a lot in negative karma by posting anything in the > thread. But I can confirm your words, that tty->xterm->X server chain > of output indeed significantly slows down the build processes. I just tried a kernel build with -j256 on a 1-core system to be unreasnable, and didn't see any sluggishness (and I notice programs taking > 10 msec to start up), but this was under my version of 5.2 with my version of SCHED_4BSD. > I usually never start build in the barebone xterm, always running screen > under xterm. make -j 10 on 4 core/HTT cpu slows up to a half, from my > unscientific impression, when run in the active screen window. Switching > to other window in screen significantly speeds it up (note the prudent > omission of any measured numbers). For me, make -s -j 256 on 1 core ran at the same speed in an xterm with another xterm watching it using top. Without -s it took 5% longer. The X output is apparently quite slow. But I rarely run X. Syscons output is much more efficient. make(1) has interesting problems determining when jobs finish. It used to wait 10 msec (?) and that gave a lot of dead time whan 10 msec became a long time for a process runtime. Maybe X is interfering with its current mechanism. During the make -j256, the load average went up to about 100 and most of the cc1's reached a low (numerically high) priority very quickly, especially on the second run when the load average was high to start (my version of the SCHED_4BSD may affect how fast or slow the priority ramps up). An interactive process competing with these cc1's has a very easy time getting scheduled to run provided it is not a bloated one that runs enough to gain a high priority itself. If it runs as much as the cc1's then it will become just one of 257 processes wanting to run and it takes a very unfair scheduler to do much better than run 1 every (default 100 msec) and thus take a default of 25.7 seconds to get back to the interactive one. At least old versions of SCHED_4BSD had significant bugs that often resulted in very unfair scheduling that happened to favour interactive processes but sometimes went the other way. The most interesting one is still there :-( : sched_exit_thread() adds the child td_estcpu to the parent td_estcpu. Combined with the inheritance of td_estcpu on fork(), this results in td_estcpu being exponential in the number of reaped children, except td_estcpu is clamped to a maximum, so td_estcpu quickly reaches the maximum td_estcpu (and td_priority quickly reaches the minimum (numerical maximum) user priority) after a few fork/exit/waits. The process doing the fork/waits is often a shell, and its interactivity becomes bad when its priority becomes low. Between about 1995 and 2000, this bug was much worse. Then there was no clamp, so td_estcpu was fully exponential in the number of children, except after about 32 doublings it overflowed to a negative value. But before it became negative, it became large, so its process gained the maximum priority and therefore found it hard to run enough to create more children. This still happens with the clamp, but "large" is now quite small and decays after a few seconds or minutes. Without the clamp, the decay took minutes or hours if not days. The doubling is fixed in my version by setting the parent td_estcpu to the maximum of the parent and child estcpu's on exit. This risks not inheriting enough (I now see a simple better method: add only the part of the child's td_estcpu that was actually due to child activity and is not just virtual cpu created on fork). The doubling was originally implemented to improve interactivity, and it "worked" bogusly by inibiting forks. E.g., for -j 256, it would probably stop make forking long before it created 256 jobs. Now with the clamp, make will just take longer to create the 256 jobs once it increased its td_estcpu to more than that of the first few jobs its started by reaping a few. Well, I tried this under -current, but only have SCHED_ULE handy to test (on a FreeBSD cluster machine). -j256 didn't seem to be enough to cause latency (even over the Pacific link). Interactivity remained perfect with -j1512. The only noticeable difference (apart from 8 cores) in top was that the load average barely reached 15 (instead of 100) with a couple of hundred sleeping processes. 8 cores might do that. Bruce