From owner-freebsd-smp Sun Jul 8 3:52:26 2001 Delivered-To: freebsd-smp@freebsd.org Received: from klima.physik.uni-mainz.de (klima.Physik.Uni-Mainz.DE [134.93.180.162]) by hub.freebsd.org (Postfix) with ESMTP id 2E57037B406; Sun, 8 Jul 2001 03:52:20 -0700 (PDT) (envelope-from ohartman@klima.physik.uni-mainz.de) Received: from klima.Physik.Uni-Mainz.DE (Sturm@klima.Physik.Uni-Mainz.DE [134.93.180.162]) by klima.physik.uni-mainz.de (8.11.4/8.11.4) with ESMTP id f68AqGL51615; Sun, 8 Jul 2001 12:52:19 +0200 (CEST) (envelope-from ohartman@klima.physik.uni-mainz.de) Date: Sun, 8 Jul 2001 12:52:16 +0200 (CEST) From: "Hartmann, O." To: Cc: Subject: FBSD 4.3-STABLE and SMP problems Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hello. Usually many people join newsgroups, access newsgroups or mailinglists when the ran into trouble. well I did also this procedure. But I want to report that since saturday morning, after a new cvsupdate and the dare to switch back to SMP kernels on our production systems (servers at our institute) did not reveal any problems (the last time the SMP problem occured our machine did not survive one night!). Another problem occured with the Linuxulator the same time went away after reinstalling the linux_base stuff. No one knows why Linux emulation showed this misbehaviour ... -- MfG O. Hartmann ohartman@klima.physik.uni-mainz.de ---------------------------------------------------------------- IT-Administration des Institut fuer Physik der Atmosphaere (IPA) ---------------------------------------------------------------- Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinenraum) Tel: +496131/3924144 FAX: +496131/3923532 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Jul 8 5:32:17 2001 Delivered-To: freebsd-smp@freebsd.org Received: from onsult-meyers.com (dialup-51-73.dplanet.ch [212.35.51.73]) by hub.freebsd.org (Postfix) with ESMTP id A899D37B403; Sun, 8 Jul 2001 05:32:05 -0700 (PDT) (envelope-from a.l.meyers@consult-meyers.com) Received: from localhost (localhost [127.0.0.1]) by 0onsult-meyers.com (8.11.4/8.11.4) with ESMTP id f68Cdkv02715; Sun, 8 Jul 2001 14:39:52 +0200 (CEST) (envelope-from a.l.meyers@consult-meyers.com) Date: Sun, 8 Jul 2001 14:39:44 +0200 (CEST) From: "A. L. Meyers" To: "Hartmann, O." Cc: , Subject: Re: FBSD 4.3-STABLE and SMP problems In-Reply-To: Message-ID: <20010708143650.E1826-100000@consult-meyers.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sun, 8 Jul 2001, Hartmann, O. wrote: > Hello. > > Usually many people join newsgroups, access newsgroups or mailinglists when > the ran into trouble. well I did also this procedure. > > But I want to report that since saturday morning, after a new cvsupdate > and the dare to switch back to SMP kernels on our production systems > (servers at our institute) did not reveal any problems (the last time > the SMP problem occured our machine did not survive one night!). It's comforting to hear the problem appears to be solved. > > Another problem occured with the Linuxulator the same time went away > after reinstalling the linux_base stuff. No one knows why Linux > emulation showed this misbehaviour ... It's discomforting to hear you appear not to know why. We can just hope it will not recur. Lucien To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Jul 9 2:56:36 2001 Delivered-To: freebsd-smp@freebsd.org Received: from prg.traveller.cz (prg.traveller.cz [193.85.2.2]) by hub.freebsd.org (Postfix) with ESMTP id 2B46537B401 for ; Mon, 9 Jul 2001 02:56:32 -0700 (PDT) (envelope-from mime@traveller.cz) Received: from prg.traveller.cz (prg.traveller.cz [193.85.2.2]) by prg.traveller.cz (EUnet.1022902037-17/pukvis) with ESMTP id f699uT703400 for ; Mon, 9 Jul 2001 11:56:30 +0200 (CEST) Date: Mon, 9 Jul 2001 11:56:28 +0200 (CEST) From: Michal Mertl To: freebsd-smp@freebsd.org Subject: Is SMP BETA in STABLE? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org From what I read in welcome message from majordomo when subscribing to thie list it sounds like I shouldn't enable SMP even with STABLE branch for production servers. Is it correct? The reason I subscribed to this list is that we run some MP servers and I want to clarify one question and hear about problems (mainly on STABLE). The problem I have is with apache. Does it run on STABLE on more processors? It seems to me, that it never consumes more than 50% processor (and from systat I don't see much harddisk activity). I've heard about processes from the same process group run only one at a time but I've also heard that it's the case with libc_r threads only. I tested with apachebench from localhost, could this be the reason? Sorry if this question isn't technical enough but I don't know where else I should ask. -- Michal Mertl mime@traveller.cz To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Jul 9 8:46:50 2001 Delivered-To: freebsd-smp@freebsd.org Received: from sneakerz.org (sneakerz.org [216.33.66.254]) by hub.freebsd.org (Postfix) with ESMTP id CE83D37B401 for ; Mon, 9 Jul 2001 08:46:47 -0700 (PDT) (envelope-from bright@sneakerz.org) Received: by sneakerz.org (Postfix, from userid 1092) id 32D335D020; Mon, 9 Jul 2001 10:46:37 -0500 (CDT) Date: Mon, 9 Jul 2001 10:46:37 -0500 From: Alfred Perlstein To: Michal Mertl Cc: freebsd-smp@freebsd.org Subject: Re: Is SMP BETA in STABLE? Message-ID: <20010709104637.C1894@sneakerz.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: ; from mime@traveller.cz on Mon, Jul 09, 2001 at 11:56:28AM +0200 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org * Michal Mertl [010709 04:56] wrote: > >From what I read in welcome message from majordomo when subscribing to > thie list it sounds like I shouldn't enable SMP even with STABLE branch > for production servers. Is it correct? That is incorrect, some people recently complained about stability issues, but were unable to track it down, then mysteriously the problems disappeared later that week. SMP is usually rock solid in -stable, but it does have it's moments. :) > The problem I have is with apache. Does it run on STABLE on more > processors? It seems to me, that it never consumes more than 50% processor > (and from systat I don't see much harddisk activity). I've heard about > processes from the same process group run only one at a time but I've also > heard that it's the case with libc_r threads only. I tested with > apachebench from localhost, could this be the reason? > > Sorry if this question isn't technical enough but I don't know where else > I should ask. Apache is a single threaded program, each instance should only be able to take up one cpu on a mp system, so what you're seeing looks like it makes sense. -- -Alfred Perlstein [alfred@freebsd.org] Ok, who wrote this damn function called '??'? And why do my programs keep crashing in it? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Jul 9 10:39:39 2001 Delivered-To: freebsd-smp@freebsd.org Received: from tp.databus.com (p101-46.acedsl.com [160.79.101.46]) by hub.freebsd.org (Postfix) with ESMTP id 2229937B403 for ; Mon, 9 Jul 2001 10:39:37 -0700 (PDT) (envelope-from barney@tp.databus.com) Received: (from barney@localhost) by tp.databus.com (8.11.4/8.11.4) id f69Hd9432072; Mon, 9 Jul 2001 13:39:09 -0400 (EDT) (envelope-from barney) Date: Mon, 9 Jul 2001 13:39:04 -0400 From: Barney Wolff To: Alfred Perlstein Cc: Michal Mertl , freebsd-smp@FreeBSD.ORG Subject: Re: Is SMP BETA in STABLE? Message-ID: <20010709133904.A32043@tp.databus.com> References: <20010709104637.C1894@sneakerz.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010709104637.C1894@sneakerz.org>; from bright@sneakerz.org on Mon, Jul 09, 2001 at 10:46:37AM -0500 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Er, then what are the 8 httpd processes I see running on my machine? It may well be that Apache handles all requests from a single client in one process, to avoid unfairness, so you might try hitting it from multiple clients. Oh yes - I've been running smp on two machines since 4.0, with absolutely zero problems. But they're lightly loaded, except during make buildworld. Barney Wolff On Mon, Jul 09, 2001 at 10:46:37AM -0500, Alfred Perlstein wrote: > > Apache is a single threaded program, each instance should only be able > to take up one cpu on a mp system, so what you're seeing looks like > it makes sense. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Jul 9 12:44:47 2001 Delivered-To: freebsd-smp@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id 28C5F37B401 for ; Mon, 9 Jul 2001 12:44:43 -0700 (PDT) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.3/8.11.2) id f69JiW799011; Mon, 9 Jul 2001 12:44:32 -0700 (PDT) (envelope-from dillon) Date: Mon, 9 Jul 2001 12:44:32 -0700 (PDT) From: Matt Dillon Message-Id: <200107091944.f69JiW799011@earth.backplane.com> To: Barney Wolff Cc: Alfred Perlstein , Michal Mertl , freebsd-smp@FreeBSD.ORG Subject: Re: Is SMP BETA in STABLE? References: <20010709104637.C1894@sneakerz.org> <20010709133904.A32043@tp.databus.com> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org :Er, then what are the 8 httpd processes I see running on my machine? :It may well be that Apache handles all requests from a single client :in one process, to avoid unfairness, so you might try hitting it :from multiple clients. : :Oh yes - I've been running smp on two machines since 4.0, with :absolutely zero problems. But they're lightly loaded, except :during make buildworld. : :Barney Wolff : :On Mon, Jul 09, 2001 at 10:46:37AM -0500, Alfred Perlstein wrote: :> :> Apache is a single threaded program, each instance should only be able :> to take up one cpu on a mp system, so what you're seeing looks like :> it makes sense. On -stable, only one cpu can be running in the kernel at any given time. The cpu's can only run in parallel when running in userland (or one in the kernel and the other in userland). So if your web server is basically just copying bits, it will be running mostly in the kernel. If your web server is doing significant userland processing, you will see both cpu's in use - that is assuming that you have a significant enough web load. 'Single threaded' in Alfred's response simply meant that Apache can only handle one connection per fork. It still forks a number of processes and so can handle a number of connections in parallel. Apache serializes connection accept()s but that would have no bearing on what you are seeing. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Jul 10 1: 0:48 2001 Delivered-To: freebsd-smp@freebsd.org Received: from prg.traveller.cz (prg.traveller.cz [193.85.2.2]) by hub.freebsd.org (Postfix) with ESMTP id 729CD37B403 for ; Tue, 10 Jul 2001 01:00:44 -0700 (PDT) (envelope-from mime@traveller.cz) Received: from prg.traveller.cz (prg.traveller.cz [193.85.2.2]) by prg.traveller.cz (EUnet.1022902037-17/pukvis) with ESMTP id f6A80dL25481; Tue, 10 Jul 2001 10:00:39 +0200 (CEST) Date: Tue, 10 Jul 2001 10:00:39 +0200 (CEST) From: Michal Mertl To: Matt Dillon Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Is SMP BETA in STABLE? In-Reply-To: <200107091944.f69JiW799011@earth.backplane.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, 9 Jul 2001, Matt Dillon wrote: > > :Er, then what are the 8 httpd processes I see running on my machine? > :It may well be that Apache handles all requests from a single client > :in one process, to avoid unfairness, so you might try hitting it > :from multiple clients. > : > :Oh yes - I've been running smp on two machines since 4.0, with > :absolutely zero problems. But they're lightly loaded, except > :during make buildworld. > : > :Barney Wolff > : > :On Mon, Jul 09, 2001 at 10:46:37AM -0500, Alfred Perlstein wrote: > :> > :> Apache is a single threaded program, each instance should only be able > :> to take up one cpu on a mp system, so what you're seeing looks like > :> it makes sense. > > On -stable, only one cpu can be running in the kernel at any given time. > The cpu's can only run in parallel when running in userland (or one in the > kernel and the other in userland). So if your web server is basically > just copying bits, it will be running mostly in the kernel. If your > web server is doing significant userland processing, you will see both > cpu's in use - that is assuming that you have a significant enough > web load. > Thanks. That clarifies the thing. The consequence of this is that as I assume a lot of web server processing is quite simple this 'one kernel at a time' seems quite limiting for serving apache on stable. Do you know if the people telling SMP in Linux kernel (say 2.2 and 2.4) is better (I mean better performing) right? I won't switch to Linux anyway :-). The page I tested on was in perl - I bechmarked once with mod-perl and once without (got interesting results - on dual p3-600 with mod-perl I got 17 pages/second and without 1.6) and resulting html was about 1.5Kb. Both cases showed a little more than 50% cpu utilization. This could mean than (even with a bit more complicated page) the overhead of connection setup (may be mostly Apache's fault) and data transfer (even on loopback) is crucial. I'll do some more benchmarking because I thing it's not correct setup to benchmark on localhost. I'll also look at ApacheBench. It's implementation may quite easily affect the results too. This is getting quite off-topic, but I think that lot of people interested in FreeBSD on SMP machines do on their servers webserving so they might be a bit interested. -- Michal Mertl mime@traveller.cz To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Jul 10 10: 0:57 2001 Delivered-To: freebsd-smp@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id 770C537B401 for ; Tue, 10 Jul 2001 10:00:54 -0700 (PDT) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.3/8.11.2) id f6AH0dQ13994; Tue, 10 Jul 2001 10:00:39 -0700 (PDT) (envelope-from dillon) Date: Tue, 10 Jul 2001 10:00:39 -0700 (PDT) From: Matt Dillon Message-Id: <200107101700.f6AH0dQ13994@earth.backplane.com> To: Michal Mertl Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Is SMP BETA in STABLE? References: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org :> just copying bits, it will be running mostly in the kernel. If your :> web server is doing significant userland processing, you will see both :> cpu's in use - that is assuming that you have a significant enough :> web load. :> :Thanks. That clarifies the thing. : :The consequence of this is that as I assume a lot of web server processing :is quite simple this 'one kernel at a time' seems quite limiting for :serving apache on stable. Do you know if the people telling SMP in Linux :kernel (say 2.2 and 2.4) is better (I mean better performing) :right? I won't switch to Linux anyway :-). : :The page I tested on was in perl - I bechmarked once with mod-perl and :once without (got interesting results - on dual p3-600 with mod-perl I got :17 pages/second and without 1.6) and resulting html was about 1.5Kb. Both :cases showed a little more than 50% cpu utilization. This could mean than :(even with a bit more complicated page) the overhead of connection setup :(may be mostly Apache's fault) and data transfer (even on loopback) is :crucial. : :I'll do some more benchmarking because I thing it's not correct setup to :benchmark on localhost. I'll also look at ApacheBench. It's implementation :may quite easily affect the results too. : :This is getting quite off-topic, but I think that lot of people interested :in FreeBSD on SMP machines do on their servers webserving so they might be :a bit interested. : :-- :Michal Mertl :mime@traveller.cz Well, that sounds to me like the benchmark is flawed. Apache and FreeBSD kernel doing nothing but accepting connections and pushing data can do hundreds of connections per second. It sounds like your benchmark is serializing its requests to the web server, running just one at a time, resulting in only one [mod]perl running at any given moment, which of course would only be able to utilize a single cpu (one non-threaded process can only run on one cpu in any SMP environment). If it's perl it is probably cpu bound. I would imagine that with a proper test you would find both cpu's in full use. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Jul 10 13:57: 3 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mailhost-1.mtl-loo.activecom.net (mailhost-1.mtl-loo.activecom.net [64.135.16.23]) by hub.freebsd.org (Postfix) with ESMTP id EAE4837B401 for ; Tue, 10 Jul 2001 13:56:56 -0700 (PDT) (envelope-from woodrow@activecom.net) Received: from saturn.activecom.net (saturn.activecom.net [64.135.16.6]) by mailhost-1.mtl-loo.activecom.net (Active MailServer) with ESMTP id 803E78DC37 for ; Tue, 10 Jul 2001 16:56:50 -0400 (EDT) Date: Tue, 10 Jul 2001 16:10:53 -0400 (EDT) From: Joshua MacQueen To: Subject: BSD/OS cpu command Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Are there any plans to include anything similar to BSD/OS's cpu (8) command into 5.0? Just in case someone hasn't heard of it: http://www.bsdi.com/bsdi-man?proto=1.1&apropos=0&msection=ALL&query=cpu _________________ Testing? What's that? If it compiles, it is good, if it boots up, it is perfect. -- Linus Torvalds To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jul 12 23:57: 3 2001 Delivered-To: freebsd-smp@freebsd.org Received: from beppo.feral.com (beppo.feral.com [192.67.166.79]) by hub.freebsd.org (Postfix) with ESMTP id 6763737B401 for ; Thu, 12 Jul 2001 23:56:59 -0700 (PDT) (envelope-from mjacob@feral.com) Received: from beppo (mjacob@beppo [192.67.166.79]) by beppo.feral.com (8.11.3/8.11.3) with ESMTP id f6D6uwS68076 for ; Thu, 12 Jul 2001 23:56:59 -0700 (PDT) (envelope-from mjacob@feral.com) Date: Thu, 12 Jul 2001 23:56:58 -0700 (PDT) From: Matthew Jacob X-Sender: mjacob@beppo Reply-To: mjacob@feral.com To: smp@freebsd.org Subject: CPU disabled by loader... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I have a very simple patch for alpha that allows me to disable CPUS at loader time via the hints mechanism: Index: mp_machdep.c =================================================================== RCS file: /home/ncvs/src/sys/alpha/alpha/mp_machdep.c,v retrieving revision 1.22 diff -u -r1.22 mp_machdep.c --- mp_machdep.c 2001/06/29 11:10:25 1.22 +++ mp_machdep.c 2001/07/13 06:54:12 @@ -37,6 +37,7 @@ #include #include #include +#include #include #include @@ -324,7 +325,7 @@ } void -cpu_mp_start() +cpu_mp_start(void) { int i; @@ -335,6 +336,7 @@ ("mp_start() called on non-primary CPU")); all_cpus = 1 << boot_cpu_id; for (i = 0; i < hwrpb->rpb_pcs_cnt; i++) { + int dv; struct pcs *pcsp; if (i == boot_cpu_id) @@ -361,7 +363,12 @@ } continue; } - all_cpus |= 1 << i; + dv = 0; + if (resource_int_value("cpu", i, "disable", &dv) == 0 && dv) { + printf("CPU %d disabled by loader.\n", i); + continue; + } + all_cpus |= (1 << i); mp_ncpus++; } PCPU_SET(other_cpus, all_cpus & ~(1 << boot_cpu_id)); ------------------- This allows me to set hint.cpu.N.disable="1" (for N=0..M Cpus) at loader time or in my hints file. This can be desirable for any large SMP system- you may have a system that has a flakey CPU that you just want to keep out of the mix, but you don't want to pull the system apart. I *think* I see how I'd do i386- but I thought I'd ask about this here as a more general question of utility. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 13 0: 4:50 2001 Delivered-To: freebsd-smp@freebsd.org Received: from beppo.feral.com (beppo.feral.com [192.67.166.79]) by hub.freebsd.org (Postfix) with ESMTP id 7344F37B401; Fri, 13 Jul 2001 00:04:48 -0700 (PDT) (envelope-from mjacob@feral.com) Received: from beppo (mjacob@beppo [192.67.166.79]) by beppo.feral.com (8.11.3/8.11.3) with ESMTP id f6D74lS68156; Fri, 13 Jul 2001 00:04:47 -0700 (PDT) (envelope-from mjacob@feral.com) Date: Fri, 13 Jul 2001 00:04:47 -0700 (PDT) From: Matthew Jacob X-Sender: mjacob@beppo Reply-To: mjacob@feral.com To: smp@freebsd.org Cc: bmilekic@freebsd.org Subject: uh oh.. bad assumption here... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I just finally looked a bit more at the MB allocation code. The assumption being made here is that: #define NCPU_PRESENT mp_ncpus Yes, this is true. What is a problem here is that it is then assumed that this CPU ids are a dense array from 0..mp_ncpus. cpuid is then used as a per-cpu index later. That this is dense can never, in fact, be true on alpha, unless we make cpuid a kernel virtual that indexes to hardware CPUid. I found this out by disabling processors 1..2 of a 4-way 4100. I crashed in inet startup as soon as CPU3 tried to do m_gethdr. Oops. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 13 11: 9:26 2001 Delivered-To: freebsd-smp@freebsd.org Received: from kira.epconline.net (kira2.epconline.net [209.83.132.2]) by hub.freebsd.org (Postfix) with ESMTP id C4C5337B403 for ; Fri, 13 Jul 2001 11:09:24 -0700 (PDT) (envelope-from carock@kira.epconline.net) Received: from localhost (carock@localhost) by kira.epconline.net (8.11.4/8.11.4) with ESMTP id f6DI9On07170 for ; Fri, 13 Jul 2001 13:09:24 -0500 (CDT) Date: Fri, 13 Jul 2001 13:09:19 -0500 (CDT) From: Chuck Rock To: freebsd-smp@freebsd.org Subject: SMP on Compaq Proliant 1850R trouble.... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I'm trying to get SMP working on my dual p-3 450 Compaq. The GENERIC kernel boots, but when I recompile with SMP options enabled, and nothing else changed. It locks up right after the APIC line on the console. No keyboard, nothin... I have to turn off and on to reboot. Anyone have any tweaks or other helpful hints to get this working if possible? Thanks, Chuck Rock To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 13 11:45:57 2001 Delivered-To: freebsd-smp@freebsd.org Received: from mail.fpsn.net (mail.fpsn.net [63.224.69.57]) by hub.freebsd.org (Postfix) with ESMTP id 1B08D37B40B for ; Fri, 13 Jul 2001 11:45:51 -0700 (PDT) (envelope-from cfaber@fpsn.net) Received: from fpsn.net (control.fpsn.net [63.224.69.60]) by mail.fpsn.net (8.9.3/8.9.3) with ESMTP id MAA12765; Fri, 13 Jul 2001 12:45:46 -0600 (MDT) (envelope-from cfaber@fpsn.net) Message-ID: <3B4F4187.B67AC12B@fpsn.net> Date: Fri, 13 Jul 2001 12:44:23 -0600 From: Colin Faber Reply-To: cfaber@fpsn.net Organization: fpsn.net, Inc. X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: Chuck Rock Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP on Compaq Proliant 1850R trouble.... References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Make sure to set the compaq up to run in APIC mode, This is a configuration option in your soft bios that you get to via pressing `CTRL-A' (if I remember correctly) Chuck Rock wrote: > > I'm trying to get SMP working on my dual p-3 450 Compaq. > > The GENERIC kernel boots, but when I recompile with SMP options enabled, > and nothing else changed. It locks up right after the APIC line on the > console. > > No keyboard, nothin... I have to turn off and on to reboot. > > Anyone have any tweaks or other helpful hints to get this working if > possible? > > Thanks, > Chuck Rock > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 13 17:52:55 2001 Delivered-To: freebsd-smp@freebsd.org Received: from gate.soum.co.jp (gate.soum.co.jp [202.221.40.2]) by hub.freebsd.org (Postfix) with ESMTP id D300537B403 for ; Fri, 13 Jul 2001 17:52:40 -0700 (PDT) (envelope-from fujita@soum.co.jp) Received: from force.soum.co.jp (force.soum.co.jp [3ffe:501:80a:1:a00:20ff:fef0:4c9c]) by gate.soum.co.jp (8.11.4/8.11.4) with ESMTP id f6E0qcH92651 for ; Sat, 14 Jul 2001 09:52:38 +0900 (JST) (envelope-from fujita@soum.co.jp) Received: from vanilla.soum.co.jp (vanilla.soum.co.jp [3ffe:501:80a:1:204:76ff:fe6e:1f81]) by force.soum.co.jp (8.11.3/3.7W-2001031912) with ESMTP id f6E0qc124655; Sat, 14 Jul 2001 09:52:38 +0900 (JST) Received: from localhost (localhost [::1]) by vanilla.soum.co.jp (Postfix) with ESMTP id F37903F9D; Sat, 14 Jul 2001 09:52:32 +0900 (JST) To: freebsd-smp@FreeBSD.ORG Subject: SMP kernel hungs up on ASUS CUV4X-D From: FUJITA Kazutoshi X-Mailer: Mew version 1.94.1 on Emacs 20.7 / Mule 4.0 =?iso-2022-jp?B?KBskQjJWMWMbKEIp?= X-PGP-PublicKey: http://www.soum.co.jp/~fujita/fujita-GnuPG-publickey.txt X-PGP-FingerPrint: 9956 2ECE 7E7D B425 EC2D D49E FEBB 3C5F 2C34 1ECA Mime-Version: 1.0 Content-Type: Multipart/Mixed; boundary="--Next_Part(Sat_Jul_14_09:51:44_2001_809)--" Content-Transfer-Encoding: 7bit Message-Id: <20010714095232C.fujita@soum.co.jp> Date: Sat, 14 Jul 2001 09:52:32 +0900 X-Dispatcher: imput version 20000228(IM140) Lines: 258 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org ----Next_Part(Sat_Jul_14_09:51:44_2001_809)-- Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, smp people. I've got ASUS CUVX-D motherboard (VIA 694XDP Chipset) which has dual Socket 370. I tried 4.3-STABLE(cvsuped Jul 13) SMP kernel, but often hungs up. It does not accept keyboard, mouse and network input or connection. I have no messages in log(such as /var/log/messages). I also tried with 5.0-CURRENT, but the result was same. Any suggestions ? Regards, ----Next_Part(Sat_Jul_14_09:51:44_2001_809)-- Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Description: mptable -dmesg Content-Disposition: attachment; filename="mptable.dmesg" =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000f54d0 signature: '_MP_' length: 16 bytes version: 1.4 checksum: 0x7d mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f50c4 signature: 'PCMP' base table length: 276 version: 1.4 checksum: 0x8e OEM ID: 'OEM00000' Product ID: 'PROD00000000' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 26 local APIC address: 0xfee00000 extended table length: 124 extended table checksum: 167 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 3 0x11 BSP, usable 6 8 6 0x387fbff 0 0x11 AP, usable 6 8 6 0x387fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT conforms conforms 2 0 2 0 INT conforms conforms 2 1 2 1 INT conforms conforms 2 0 2 2 INT conforms conforms 2 3 2 3 INT conforms conforms 2 4 2 4 INT conforms conforms 2 6 2 6 INT conforms conforms 2 7 2 7 INT conforms conforms 2 8 2 8 INT conforms conforms 2 12 2 12 INT conforms conforms 2 14 2 14 INT conforms conforms 2 15 2 15 INT active-lo level 1 0:A 2 16 INT active-lo level 2 0 2 0 INT active-lo level 0 4:D 2 18 INT active-lo level 0 9:A 2 19 INT active-lo level 0 10:A 2 18 INT active-lo level 0 11:A 2 17 INT active-lo level 0 13:A 2 19 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 2 0 255 0 NMI active-hi edge 2 0 255 1 ------------------------------------------------------------------------------- MP Config Extended Table Entries: -- System Address Space bus ID: 0 address type: I/O address address base: 0x0 address range: 0x10000 -- System Address Space bus ID: 0 address type: memory address address base: 0x20000000 address range: 0xcfec0000 -- System Address Space bus ID: 0 address type: prefetch address address base: 0xefec0000 address range: 0xf8140000 -- System Address Space bus ID: 0 address type: memory address address base: 0xe8000000 address range: 0x18000000 -- System Address Space bus ID: 0 address type: memory address address base: 0xa0000 address range: 0x20000 -- Bus Heirarchy bus ID: 2 bus info: 0x01 parent bus ID: 0 -- Compatibility Bus Address bus ID: 0 address modifier: add predefined range: 0x00000000 -- Compatibility Bus Address bus ID: 0 address modifier: add predefined range: 0x00000001 ------------------------------------------------------------------------------- dmesg output: Copyright (c) 1992-2001 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 4.3-STABLE #0: Fri Jul 13 18:37:05 JST 2001 fujita@vanilla.soum.co.jp:/usr/obj/usr/src/sys/VANILLA Timecounter "i8254" frequency 1193182 Hz CPU: Pentium III/Pentium III Xeon/Celeron (1004.52-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x686 Stepping = 6 Features=0x387fbff real memory = 536870912 (524288K bytes) sio0: gdb debugging port avail memory = 518193152 (506048K bytes) Programming 24 pins in IOAPIC #0 IOAPIC #0 intpin 2 -> irq 0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00178011, at 0xfec00000 Preloaded elf kernel "kernel" at 0xc046a000. ccd0-3: Concatenated disk drivers netsmb_dev: loaded Pentium Pro MTRR support enabled md0: Malloc disk apm0: on motherboard apm: found APM BIOS v1.2, connected at v1.2 npx0: on motherboard npx0: INT 16 interface pcib0: on motherboard IOAPIC #0 intpin 18 -> irq 2 IOAPIC #0 intpin 19 -> irq 5 IOAPIC #0 intpin 17 -> irq 9 pci0: on pcib0 pcib2: at device 1.0 on pci0 IOAPIC #0 intpin 16 -> irq 10 pci1: on pcib2 pci1: at 0.0 irq 10 isab0: at device 4.0 on pci0 isa0: on isab0 atapci0: port 0xd800-0xd80f at device 4.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 uhci0: port 0xd400-0xd41f irq 2 at device 4.2 on pci0 usb0: on uhci0 usb0: USB revision 1.0 uhub0: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub0: 2 ports with 2 removable, self powered uhci1: port 0xd000-0xd01f irq 2 at device 4.3 on pci0 usb1: on uhci1 usb1: USB revision 1.0 uhub1: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub1: 2 ports with 2 removable, self powered sym0: <875> port 0xb800-0xb8ff mem 0xed000000-0xed000fff,0xed800000-0xed8000ff irq 5 at device 9.0 on pci0 sym0: Symbios NVRAM, ID 7, Fast-20, SE, parity checking pcm0: port 0xb400-0xb41f irq 2 at device 10.0 on pci0 vr0: port 0xa800-0xa8ff mem 0xec800000-0xec8000ff irq 9 at device 11.0 on pci0 vr0: Ethernet address: 00:90:99:22:e0:14 miibus0: on vr0 acphy0: on miibus0 acphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto xl0: <3Com 3c905B-TX Fast Etherlink XL> port 0xa400-0xa47f mem 0xec000000-0xec00007f irq 5 at device 13.0 on pci0 xl0: Ethernet address: 00:04:76:6e:1f:81 miibus1: on xl0 xlphy0: <3Com internal media interface> on miibus1 xlphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto pcib1: on motherboard pci2: on pcib1 orm0: