From owner-freebsd-smp Sun Mar 22 00:10:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA25063 for freebsd-smp-outgoing; Sun, 22 Mar 1998 00:10:51 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA25050 for ; Sun, 22 Mar 1998 00:10:46 -0800 (PST) (envelope-from jhay@zibbi.mikom.csir.co.za) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.8/8.8.7) id KAA17674; Sun, 22 Mar 1998 10:09:47 +0200 (SAT) From: John Hay Message-Id: <199803220809.KAA17674@zibbi.mikom.csir.co.za> Subject: Re: is it just me, or... In-Reply-To: <199803220759.PAA10674@spinner.netplex.com.au> from Peter Wemm at "Mar 22, 98 03:59:17 pm" To: peter@netplex.com.au (Peter Wemm) Date: Sun, 22 Mar 1998 10:09:47 +0200 (SAT) Cc: smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > > It is just me or has the smp kernel taken a nose dive over the last month > > > > > or so? > > Damn.. rc5des just took a dive on the machine while running on a UP > kernel. Damn! Damn! Damn! That narrows that part down to either the rest > of the system being ELF or a hardware problem... > Be carefull, at least some versions of rc5des had a bug in their PPro/PII core that made them coredump when they switched back from des to rc5. It was reported on their mailing list a while back and I then switched back to an older rc5 only version. I haven't looked to see if they have a fix yet. John -- John Hay -- John.Hay@mikom.csir.co.za To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 00:22:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA27008 for freebsd-smp-outgoing; Sun, 22 Mar 1998 00:22:03 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA26967 for ; Sun, 22 Mar 1998 00:21:59 -0800 (PST) (envelope-from jhay@zibbi.mikom.csir.co.za) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.8/8.8.7) id KAA17845; Sun, 22 Mar 1998 10:20:49 +0200 (SAT) From: John Hay Message-Id: <199803220820.KAA17845@zibbi.mikom.csir.co.za> Subject: Re: is it just me, or... In-Reply-To: <199803220756.PAA10655@spinner.netplex.com.au> from Peter Wemm at "Mar 22, 98 03:56:35 pm" To: peter@netplex.com.au (Peter Wemm) Date: Sun, 22 Mar 1998 10:20:49 +0200 (SAT) Cc: smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > > It is just me or has the smp kernel taken a nose dive over the last month > > > > > or so? > > > ... > > I do most of my testing, verification and day-to-day usage with SMP systems. > > It is only an after-thought that I check out UP systems. If there is anythin > g > > that you can point to where performance has decreased, let me know. > ... > Anyway, I got a couple of replies saying that they were having no trouble > at all. One even fired up a rc564/rc5des key searcher (since that was one > of the easiest ways I could provoke problems), but nothing unexpected > happened. > > So, it must be something I'm doing. Is *anybody* running parallel rc564 or > rc5des under SMP? (the rc5 cores are (apparently) not reentrant, so you > need two or more copies running at once to get max throughput, even in the > multithread version) I'm normally running two rc564s on my Dual PII-266 and also do two simultaneous "make release"es (offset with an hour so that cvs don't grind the disks almost to a halt and also so that the /dev/vn0s don't clash) every night with no problems. The machine has an onboard 2940UW with two disks attached. > > The only other things I could think of: > - the machine having trouble is a pure-ELF system, even though the rc5 > key search problems are from a.out executables. > - I had VM86 enabled (being compiled out now) > - I had USER_LDT enabled (going too) > - I had a sound driver enabled (going too). > - I have AHC_TAGENABLE, AHC_SCBPAGING_ENABLE and AHC_ALLOW_MEMIO (works > fine under UP). > I have the AHC_* defines in, but not VM86 or USER_LDT. John -- John Hay -- John.Hay@mikom.csir.co.za To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 00:34:32 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA00106 for freebsd-smp-outgoing; Sun, 22 Mar 1998 00:34:32 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from critter.freebsd.dk (critter.freebsd.dk [195.8.129.14]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA29921; Sun, 22 Mar 1998 00:33:54 -0800 (PST) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.8.7/8.8.5) with ESMTP id JAA21674; Sun, 22 Mar 1998 09:33:28 +0100 (CET) To: sos@FreeBSD.ORG cc: peter@netplex.com.au (Peter Wemm), smp@FreeBSD.ORG Subject: Re: is it just me, or... In-reply-to: Your message of "Sat, 21 Mar 1998 23:55:12 +0100." <199803212255.XAA26219@sos.freebsd.dk> Date: Sun, 22 Mar 1998 09:33:27 +0100 Message-ID: <21672.890555607@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <199803212255.XAA26219@sos.freebsd.dk>, Søren Schmidt writes: >In reply to Peter Wemm who wrote: >> It is just me or has the smp kernel taken a nose dive over the last month >> or so? > >I havn't had any problems until the latest breakage John provided :) > >I run SMP on a TYAN s1662d with 128M and two Maxtor EIDE disks, no SCSI, >so thats at least one difference. >The system behaves very nicely, even under very heavy load, interactive >response (here under X) is still exelent, no long delays or anything... I see the same thing with two rc564 processes running... -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." "Drink MONO-tonic, it goes down but it will NEVER come back up!" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 00:42:29 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA02342 for freebsd-smp-outgoing; Sun, 22 Mar 1998 00:42:29 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA02308 for ; Sun, 22 Mar 1998 00:42:24 -0800 (PST) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.8.8/8.8.8/Spinner) with ESMTP id QAA11326; Sun, 22 Mar 1998 16:41:48 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199803220841.QAA11326@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: John Hay cc: smp@FreeBSD.ORG Subject: Re: is it just me, or... In-reply-to: Your message of "Sat, 22 Mar 1998 10:09:47 +0200." <199803220809.KAA17674@zibbi.mikom.csir.co.za> Date: Sun, 22 Mar 1998 16:41:47 +0800 From: Peter Wemm Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org John Hay wrote: > > > > > It is just me or has the smp kernel taken a nose dive over the last m onth > > > > > > > or so? > > > > Damn.. rc5des just took a dive on the machine while running on a UP > > kernel. Damn! Damn! Damn! That narrows that part down to either the rest > > of the system being ELF or a hardware problem... > > > > Be carefull, at least some versions of rc5des had a bug in their PPro/PII > core that made them coredump when they switched back from des to rc5. It > was reported on their mailing list a while back and I then switched back > to an older rc5 only version. I haven't looked to see if they have a fix > yet. Ahhh shit!! This was it.. If I get it to the stage where it coredumps, and switch to non-ppro cpu code, it picks up the checkpoint and merrily submits the completed key. Damn!!!!!! I wish the distributed.net folks would update the freebsd binaries because this has lost them a whole chunk of keyspace. (not to mention the 16 byte "current version" of the non-threaded binary on the server, the .gz compression must be pretty good to get it from 700-odd K down to 16 bytes.) > John Cheers, -Peter -- Peter Wemm Netplex Consulting To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 01:01:12 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA05522 for freebsd-smp-outgoing; Sun, 22 Mar 1998 01:01:12 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from critter.freebsd.dk (critter.freebsd.dk [195.8.129.14]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA05511; Sun, 22 Mar 1998 01:01:05 -0800 (PST) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.8.7/8.8.5) with ESMTP id JAA21972; Sun, 22 Mar 1998 09:58:51 +0100 (CET) To: Peter Wemm cc: "John S. Dyson" , sos@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: is it just me, or... In-reply-to: Your message of "Sun, 22 Mar 1998 15:56:35 +0800." <199803220756.PAA10655@spinner.netplex.com.au> Date: Sun, 22 Mar 1998 09:58:51 +0100 Message-ID: <21970.890557131@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <199803220756.PAA10655@spinner.netplex.com.au>, Peter Wemm writes: >So, it must be something I'm doing. Is *anybody* running parallel rc564 or >rc5des under SMP? (the rc5 cores are (apparently) not reentrant, so you >need two or more copies running at once to get max throughput, even in the >multithread version) Yes I am, on a PR400FX mb. But this is not close to current, I was just about to get current on it when John hit... >I guess the other thing to do is to recompile the system for a.out again. >I suspect that there are problems with ELF executable startup.. I have nothing even aproaching ELF on it, so that is a wrong tree to bark at... -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." "Drink MONO-tonic, it goes down but it will NEVER come back up!" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 02:09:26 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id CAA10709 for freebsd-smp-outgoing; Sun, 22 Mar 1998 02:09:26 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA10699 for ; Sun, 22 Mar 1998 02:09:17 -0800 (PST) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.8.8/8.8.8/Spinner) with ESMTP id SAA12360; Sun, 22 Mar 1998 18:08:37 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199803221008.SAA12360@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Peter Wemm cc: John Hay , smp@FreeBSD.ORG Subject: Re: is it just me, or... In-reply-to: Your message of "Sun, 22 Mar 1998 16:41:47 +0800." <199803220841.QAA11326@spinner.netplex.com.au> Date: Sun, 22 Mar 1998 18:08:36 +0800 From: Peter Wemm Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Peter Wemm wrote: > John Hay wrote: > > > > > > It is just me or has the smp kernel taken a nose dive over the last m > onth > > > > > > > > > or so? > > > > > > Damn.. rc5des just took a dive on the machine while running on a UP > > > kernel. Damn! Damn! Damn! That narrows that part down to either the res t > > > of the system being ELF or a hardware problem... > > > > > > > Be carefull, at least some versions of rc5des had a bug in their PPro/PII > > core that made them coredump when they switched back from des to rc5. It > > was reported on their mailing list a while back and I then switched back > > to an older rc5 only version. I haven't looked to see if they have a fix > > yet. > > Ahhh shit!! This was it.. If I get it to the stage where it coredumps, > and switch to non-ppro cpu code, it picks up the checkpoint and merrily > submits the completed key. Damn!!!!!! I wish the distributed.net folks > would update the freebsd binaries because this has lost them a whole chunk > of keyspace. Well, that's solved the coredump problem anyway.. But it gets more bizzare still......... If I go into single user, do a 'fsck -p' then press ctrl-D, I can run rc5des however I like, and do not get response problems at all from anything (including compiling). If I do not go into single user (ie: don't type -s at the boot prompt), and if I run either 1 or 2 rc5des's, I get a 100% repeatable networking (and often disk) lockup with "fxp0 device timeout"s and ahc0 resets and timeouts etc. Compiling a kernel also brings the system to it's knees. What I did was back out my 2nd of march elf changes just before trying this. This difference of behavior between running singleuser or not is the exact same problem that I had with /usr/libexec/getty on this machine. I suspect the elf change is a false alarm, but I'm going to try this again with the current elf image loader just to make sure... When I think back over the last few days when I've been paying particular attention to the problem, I think it's quite likely that crashes and dropping into single user mode are the variable. Hmm... Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 09:10:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA22240 for freebsd-smp-outgoing; Sun, 22 Mar 1998 09:10:51 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from icicle.winternet.com (adm@icicle.winternet.com [198.174.169.13]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA22232; Sun, 22 Mar 1998 09:10:43 -0800 (PST) (envelope-from mestery@mail.winternet.com) Received: (from adm@localhost) by icicle.winternet.com (8.8.8/8.8.8) id LAA22379; Sun, 22 Mar 1998 11:10:41 -0600 (CST) Received: from tundra.winternet.com(198.174.169.11) by icicle.winternet.com via smap (V2.0) id xma022342; Sun, 22 Mar 98 11:10:25 -0600 Received: from localhost (mestery@localhost) by tundra.winternet.com (8.8.7/8.8.4) with SMTP id LAA10893; Sun, 22 Mar 1998 11:10:25 -0600 (CST) X-Authentication-Warning: tundra.winternet.com: mestery owned process doing -bs Date: Sun, 22 Mar 1998 11:10:24 -0600 (CST) From: Kyle Mestery To: Søren Schmidt cc: Peter Wemm , smp@FreeBSD.ORG Subject: Re: is it just me, or... In-Reply-To: <199803212255.XAA26219@sos.freebsd.dk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by hub.freebsd.org id JAA22233 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 21 Mar 1998, Søren Schmidt wrote: > In reply to Peter Wemm who wrote: > > It is just me or has the smp kernel taken a nose dive over the last month > > or so? > > I havn't had any problems until the latest breakage John provided :) > > I run SMP on a TYAN s1662d with 128M and two Maxtor EIDE disks, no SCSI, > so thats at least one difference. > The system behaves very nicely, even under very heavy load, interactive > response (here under X) is still exelent, no long delays or anything... > I also am having no problems at all. I amrunning a kernel from March 13th, and have had no problems at all. Interactive response is fine. System is a Tyan tomcat II, dual P120's, EIDE and SCSI disks. -- Kyle Mestery StorageTek's Network Systems Group "Keep honking, I'm reloading." "Lottery: A tax on people who are bad at math." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 13:04:40 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA03938 for freebsd-smp-outgoing; Sun, 22 Mar 1998 13:04:40 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA03740; Sun, 22 Mar 1998 13:03:46 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id QAA02388; Sun, 22 Mar 1998 16:03:29 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199803222103.QAA02388@dyson.iquest.net> Subject: Re: is it just me, or... In-Reply-To: <199803220756.PAA10655@spinner.netplex.com.au> from Peter Wemm at "Mar 22, 98 03:56:35 pm" To: peter@netplex.com.au (Peter Wemm) Date: Sun, 22 Mar 1998 16:03:29 -0500 (EST) Cc: toor@dyson.iquest.net, sos@FreeBSD.ORG, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > "John S. Dyson" wrote: > > So, it must be something I'm doing. Is *anybody* running parallel rc564 or > rc5des under SMP? (the rc5 cores are (apparently) not reentrant, so you > need two or more copies running at once to get max throughput, even in the > multithread version) > I have noticed some interesting scheduling anomolies. I haven't looked at it, due to others likely working the issues. John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 13:11:29 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA04554 for freebsd-smp-outgoing; Sun, 22 Mar 1998 13:11:29 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA04536; Sun, 22 Mar 1998 13:11:03 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id QAA02437; Sun, 22 Mar 1998 16:10:59 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199803222110.QAA02437@dyson.iquest.net> Subject: Re: is it just me, or... In-Reply-To: <199803220756.PAA10655@spinner.netplex.com.au> from Peter Wemm at "Mar 22, 98 03:56:35 pm" To: peter@netplex.com.au (Peter Wemm) Date: Sun, 22 Mar 1998 16:10:59 -0500 (EST) Cc: toor@dyson.iquest.net, sos@FreeBSD.ORG, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > "John S. Dyson" wrote: > > > In reply to Peter Wemm who wrote: > > > > It is just me or has the smp kernel taken a nose dive over the last month > > > > > or so? > > > > > > I havn't had any problems until the latest breakage John provided :) > > > > > > I run SMP on a TYAN s1662d with 128M and two Maxtor EIDE disks, no SCSI, > > > so thats at least one difference. > > > The system behaves very nicely, even under very heavy load, interactive > > > response (here under X) is still exelent, no long delays or anything... > > > > > I do most of my testing, verification and day-to-day usage with SMP systems. > > It is only an after-thought that I check out UP systems. If there is anythin > g > > that you can point to where performance has decreased, let me know. > > Sorry if this sounded like an accusation, it was meant as a fact finding > excercise to see if just I had a problem or whether everybody else was too > and had gone to UP or something. > I *never* consider statements like that as "accusations", and agree that it is reasonable for a fact finding exercise. John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Mar 22 22:06:39 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA02403 for freebsd-smp-outgoing; Sun, 22 Mar 1998 22:06:39 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from gratis.grondar.za (gratis.grondar.za [196.7.18.133]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA02389 for ; Sun, 22 Mar 1998 22:06:23 -0800 (PST) (envelope-from mark@grondar.za) Received: from greenpeace.grondar.za (rSWSmhdwLrd5BAvRcKPeS+TZcHnkZQ+O@greenpeace.grondar.za [196.7.18.132]) by gratis.grondar.za (8.8.8/8.8.8) with ESMTP id IAA17373; Mon, 23 Mar 1998 08:06:19 +0200 (SAST) (envelope-from mark@grondar.za) Received: from grondar.za (2HQaMRctMtLh+fNN+w666fOh59Ochtdw@localhost [127.0.0.1]) by greenpeace.grondar.za (8.8.8/8.8.8) with ESMTP id IAA03658; Mon, 23 Mar 1998 08:06:12 +0200 (SAST) (envelope-from mark@grondar.za) Message-Id: <199803230606.IAA03658@greenpeace.grondar.za> X-Mailer: exmh version 2.0.2 2/24/98 To: Peter Wemm cc: smp@FreeBSD.ORG Subject: Re: is it just me, or... Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 23 Mar 1998 08:06:11 +0200 From: Mark Murray Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Peter Wemm wrote: I get the same errors as belo if I doo "too much" (Building xemacs20 seems to do it). > sd0: SCB 0x1 - timed out in message in phase, SCSISIGI == 0xe6 > SEQADDR = 0x168 SCSISEQ = 0x12 SSTAT0 = 0x7 SSTAT1 = 0x3 > sd0: abort message in message buffer > ahc0:A:0: Missed busfree. > sd0: SCB 0x1 - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0xb6 > SEQADDR = 0x4 SCSISEQ = 0x5a SSTAT0 = 0x7 SSTAT1 = 0x13 > sd0: no longer in timeout > ahc0: Issued Channel A Bus Reset. 1 SCBs aborted > sd1: SCB 0x0 - timed out in message out phase, SCSISIGI == 0xb6 > SEQADDR = 0xab SCSISEQ = 0x12 SSTAT0 = 0x7 SSTAT1 = 0x3 > Ordered Tag queued > sd1: no longer in timeout > Ordered Tag sent > sd0: SCB 0x1 - timed out in message in phase, SCSISIGI == 0xe6 > SEQADDR = 0x168 SCSISEQ = 0x12 SSTAT0 = 0x7 SSTAT1 = 0x3 > sd0: abort message in message buffer > ahc0:A:0: Missed busfree. > sd0: SCB 0x1 - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0xb6 > SEQADDR = 0x5 SCSISEQ = 0x5a SSTAT0 = 0x7 SSTAT1 = 0x13 > sd0: no longer in timeout > ahc0: Issued Channel A Bus Reset. 1 SCBs aborted > sd1: SCB 0x2 - timed out in message out phase, SCSISIGI == 0xb6 > SEQADDR = 0xa2 SCSISEQ = 0x12 SSTAT0 = 0x7 SSTAT1 = 0x3 > Ordered Tag queued > sd0: UNIT ATTE Followed bay a hard hang or a reboot. M -- Mark Murray Join the anti-SPAM movement: http://www.cauce.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 24 00:00:25 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA13699 for freebsd-smp-outgoing; Tue, 24 Mar 1998 00:00:25 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA13527; Mon, 23 Mar 1998 23:58:55 -0800 (PST) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.8.8/8.8.8/Spinner) with ESMTP id PAA16611; Tue, 24 Mar 1998 15:58:08 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199803240758.PAA16611@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: "John S. Dyson" cc: sos@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: is it just me, or... In-reply-to: Your message of "Sun, 22 Mar 1998 16:03:29 EST." <199803222103.QAA02388@dyson.iquest.net> Date: Tue, 24 Mar 1998 15:58:07 +0800 From: Peter Wemm Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org "John S. Dyson" wrote: > > "John S. Dyson" wrote: > > > > So, it must be something I'm doing. Is *anybody* running parallel rc564 or > > rc5des under SMP? (the rc5 cores are (apparently) not reentrant, so you > > need two or more copies running at once to get max throughput, even in the > > multithread version) > > > I have noticed some interesting scheduling anomolies. I haven't looked at it , > due to others likely working the issues. > > John I modified my /sbin/init so that it did three forks and exec's of /bin/sync prior to running /etc/ rc. My machine now boots into multi-user mode, *with* a pair of rc5 search processes running from /usr/local/etc/rc.d. Having these started from rc has caused lockups for ages, unless I went into single user during boot. It seems that a couple of fork/exec/exit type operations prior to running / etc/rc has the same effect as going into singleuser and back out. I don't know if the three forks is significant, I suspect that one would be enough and will try it shortly. I guess this means that there is a real problem with the first(?) user-mode fork/exec within init... Something (still!) must be wrong with the bootstrap or init process creation.. This shows up as scheduler wierdness, lost interrupts, timeouts and lockups. (Hmm,, I'd better build an a.out /sbin/init as well, to make sure it's not a problem with a static ELF /sbin/init on smp). Cheers, -Peter -- Peter Wemm Netplex Consulting To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 27 16:56:37 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA28872 for freebsd-smp-outgoing; Fri, 27 Mar 1998 16:56:37 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from graduate.hku.hk (graduate.hku.hk [147.8.2.86]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA28864 for ; Fri, 27 Mar 1998 16:56:31 -0800 (PST) (envelope-from a9421787@graduate.hku.hk) Received: from localhost (a9421787@localhost) by graduate.hku.hk (8.8.7/8.8.7) with SMTP id JAA17797 for ; Sat, 28 Mar 1998 09:11:01 +0800 (HKT) Date: Sat, 28 Mar 1998 09:11:01 +0800 (HKT) From: Wong Kwan Po To: freebsd-smp@FreeBSD.ORG Subject: Kernel Locking Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello! I have a question concerning the mechanism of kernel locking. Is FreeBSD currently using giant locking for ensuring kernel integrity? If so, does it imply that there is no way to handle two or more incoming interrupt requests at the same time? TIA. Best regards, Raymond To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 27 17:39:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA04935 for freebsd-smp-outgoing; Fri, 27 Mar 1998 17:39:15 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from picnic.mat.net (picnic.mat.net [206.246.122.117]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA04929 for ; Fri, 27 Mar 1998 17:39:13 -0800 (PST) (envelope-from chuckr@glue.umd.edu) Received: from localhost (chuckr@localhost) by picnic.mat.net (8.8.8/8.8.5) with SMTP id UAA06353; Fri, 27 Mar 1998 20:37:44 -0500 (EST) Date: Fri, 27 Mar 1998 20:37:44 -0500 (EST) From: Chuck Robey X-Sender: chuckr@localhost To: Wong Kwan Po cc: freebsd-smp@FreeBSD.ORG Subject: Re: Kernel Locking In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 28 Mar 1998, Wong Kwan Po wrote: > Hello! > > I have a question concerning the mechanism of kernel locking. > > Is FreeBSD currently using giant locking for ensuring kernel integrity? If > so, does it imply that there is no way to handle two or more incoming > interrupt requests at the same time? No, you misunderstand the difference between the giant kernel lock that the SMP uses so that only one processor is in the kernel at once, and the interrupt masking level, which can be set to various different priorities. The interrupt level setting adjusts things so that a high priority interrupt CAN interrupt a low priority interrupt. In a uniprocessor machine, the kernel lock doesn't do anything at all. SMP shouldn't in itself use a giant kernel lock either, that open space for hackers to improve. > > TIA. > > Best regards, > Raymond > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > > ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@glue.umd.edu | communications topic, C programming, and Unix. 213 Lakeside Drive Apt T-1 | Greenbelt, MD 20770 | I run Journey2 and picnic, both FreeBSD (301) 220-2114 | version 3.0 current -- and great FUN! ----------------------------+----------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 28 21:11:31 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA00790 for freebsd-smp-outgoing; Sat, 28 Mar 1998 21:11:31 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from friley585.res.iastate.edu (friley585.res.iastate.edu [129.186.167.85]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA00771; Sat, 28 Mar 1998 21:11:28 -0800 (PST) (envelope-from ccsanady@friley585.res.iastate.edu) Received: from friley585.res.iastate.edu (loopback [127.0.0.1]) by friley585.res.iastate.edu (8.8.8/8.8.8) with ESMTP id XAA00367; Sat, 28 Mar 1998 23:11:21 -0600 (CST) (envelope-from ccsanady@friley585.res.iastate.edu) Message-Id: <199803290511.XAA00367@friley585.res.iastate.edu> X-Mailer: exmh version 2.0.1 12/23/97 Reply-to: ccsanady@iastate.edu To: freebsd-smp@FreeBSD.ORG cc: freebsd-scsi@FreeBSD.ORG Subject: Signal 11's.. (and SCB problems) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 28 Mar 1998 23:11:21 -0600 From: Chris Csanady Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org While building a world, I recently noticed that I was getting random signal 11's. I was under the impression that this was fixed earlier, is this correct? I suppose this is probably related to the SCB problems, but I can't say. Is anyone else seeing similar problems? Regardless, SMP is still miserable here. I seem to have similar SCB hangs (that I mentioned earlier) running UP as well, although they are fairly rare. Chris Csanady To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 28 21:24:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA01823 for freebsd-smp-outgoing; Sat, 28 Mar 1998 21:24:51 -0800 (PST) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA01805; Sat, 28 Mar 1998 21:24:47 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id AAA24173; Sun, 29 Mar 1998 00:24:42 -0500 (EST) (envelope-from toor) Message-Id: <199803290524.AAA24173@dyson.iquest.net> Subject: Re: Signal 11's.. (and SCB problems) In-Reply-To: <199803290511.XAA00367@friley585.res.iastate.edu> from Chris Csanady at "Mar 28, 98 11:11:21 pm" To: ccsanady@iastate.edu Date: Sun, 29 Mar 1998 00:24:42 -0500 (EST) Cc: freebsd-smp@FreeBSD.ORG, freebsd-scsi@FreeBSD.ORG From: "John S. Dyson" Reply-To: dyson@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Chris Csanady said: > While building a world, I recently noticed that I was getting random > signal 11's. I was under the impression that this was fixed earlier, > is this correct? > > I suppose this is probably related to the SCB problems, but I can't > say. Is anyone else seeing similar problems? > > Regardless, SMP is still miserable here. I seem to have similar SCB > hangs (that I mentioned earlier) running UP as well, although they > are fairly rare. > I have been working on some improvements, and have noticed that it is possible (with only minor mods) to cause sig-11's. It is too early to say whether or not I am reproducing what you are seeing, and I currently doubt it. Just FYI, I am listening (lurking), and not assuming that things are perfect on the VM and VFS fronts. -- John | Never try to teach a pig to sing, dyson@freebsd.org | it just makes you look stupid, jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message