From owner-freebsd-smp Mon Aug 3 09:02:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA14207 for freebsd-smp-outgoing; Mon, 3 Aug 1998 09:02:58 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from fleming.cs.strath.ac.uk (fleming.cs.strath.ac.uk [130.159.196.123]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA14146; Mon, 3 Aug 1998 09:02:32 -0700 (PDT) (envelope-from roger@cs.strath.ac.uk) Received: from cs.strath.ac.uk (posh.dmem.strath.ac.uk [130.159.202.3]) by fleming.cs.strath.ac.uk (8.8.8/8.8.8) with ESMTP id RAA14265 Mon, 3 Aug 1998 17:02:13 +0100 (BST) Message-ID: <35C5DEC4.BF4EF1D4@cs.strath.ac.uk> Date: Mon, 03 Aug 1998 17:01:08 +0100 From: Roger Hardiman Organization: Strathclyde Uni X-Mailer: Mozilla 4.05 [en] (X11; I; FreeBSD 3.0-980520-SNAP i386) MIME-Version: 1.0 To: freebsd-smp@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG CC: roger@cs.strath.ac.uk Subject: Kernel Threads. What is the status? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I was wondering what the status of Kernel Threads is. Is it a) On the wish list, but not started yet. b) Under casual development c) Under serious development I want to make maximum use of my dual PII box for image processing. I've used fork() and shared memory to share the processing over the 2 CPUs. I would like to use POSIX Pthreads. Linux has kernel threads and their Pthreads run over multiple CPUs. What is the state of FreeBSD? Thanks Roger Hardiman, Strathclyde Uni Telepresence Group To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Aug 8 10:18:48 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA03176 for freebsd-smp-outgoing; Sat, 8 Aug 1998 10:18:48 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from krygierpc.kph.uni-mainz.de (krygierpc.kph.uni-mainz.de [134.93.132.24]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id KAA03166 for ; Sat, 8 Aug 1998 10:18:40 -0700 (PDT) (envelope-from krygier@kph.uni-mainz.de) Received: from localhost (krygier@localhost) by krygierpc.kph.uni-mainz.de (8.8.8/8.8.8) with ESMTP id TAA00430 for ; Sat, 8 Aug 1998 19:18:22 +0200 (CEST) (envelope-from krygier@kph.uni-mainz.de) X-Authentication-Warning: krygierpc.kph.uni-mainz.de: krygier owned process doing -bs Date: Sat, 8 Aug 1998 19:18:22 +0200 (CEST) From: Klaus Werner Krygier To: freebsd-smp@FreeBSD.ORG Subject: system lock-up Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, here is a mail which I posted a few days ago to freebsd-current. I got very few answers which didn't help. Perhaps in this mailing list there is someone who can help me. Do these problems occur only on a few SMP systems? Are they specific for Gigabyte boards? Which other hardware allows a stable system (like my computer as it worked with a SNAP of May 1997 for about one year)? Klaus Werner Krygier ---------- Forwarded message ---------- Date: Thu, 6 Aug 1998 20:58:34 +0200 (CEST) From: Klaus Werner Krygier To: Philippe Regnauld Cc: freebsd-current@FreeBSD.ORG Subject: Re: system lock-up (SMP, AHC_TAGENABLE, softupdates) On Sun, 2 Aug 1998, Philippe Regnauld wrote: > System: GigaByte 586-DX, 2xP5-133, 48MB, on-board 7880 SCSI, > 4GB IBM DCAS-34330W. > Current as of CTM src-cur 3477 (< 48 hours). > > Softupdates enabled on _all_ filesystems: no problem, did > two make buildworlds/installworld in a row, as long > as AHC_TAGENABLE is off. > > If I enable AHC_TAGENABLE, and push the system a bit (lots > of FS activity) -> freeze. Last time it happened was with X, > so I'd have to do it again in text mode to write down the AHC > TAG messages... I have a very similar hardware setup on several machines (Gigabyte 586-DX, 2 x P5-200, 64 MB, on-board 7880 SCSI, several different disks (2GB, 4GB, 9GB)) and I have the same problems: If I do a lot of disk activity my computers freeze. Sometimes a SCSI bus reset seems to be initiated (which I can recogize on the loading activity of my CD changer). This problem occurs on standard 3.0-SNAP kernels without softupdates and almost SMP-GENERIC configuration. I tried several SNAPs, but the problem remained - I have the impression it has become worse with newe versions :-(. At last I damaged one of my disks. My last hope was CAM. Here (3.0CAM-19980712-SNAP) the system does not freeze completely but becomes very slow on disk access and is also not usable. One positive aspect: I get kernel messages. Every few seconds the following message is printed out for ever: Timedout SCB handled by another timeout Timedout SCB handled by another timeout last message repeated 4 times last message repeated 13 times last message repeated 30 times Timedout SCB handled by another timeout Timedout SCB handled by another timeout etc. With older 3.0-SNAPs (e.g. May 97) a SMP system was very stable on my first board and did not show such effects for almost a year. Because of other problems and to use new features and improvements I had to upgrade and have now very instable systems. Klaus Werner Krygier +-----------------------------------------------------------------------+ | Dr. Klaus Werner Krygier | Email: krygier@kph.uni-mainz.de | | Institut für Kernphysik | | | Johannes Gutenberg-Universität | Tel: +49-6131-39-2960 | | J.J.Becher-Weg 45 | +49-6131-39-5868 | | D-55099 Mainz | Fax: +49-6131-39-2964 | +-----------------------------------------------------------------------+ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" 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 Sat Aug 8 11:18:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA08673 for freebsd-smp-outgoing; Sat, 8 Aug 1998 11:18:58 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from sos.freebsd.dk (sos.freebsd.dk [212.242.40.180]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA08668 for ; Sat, 8 Aug 1998 11:18:55 -0700 (PDT) (envelope-from sos@sos.freebsd.dk) Received: (from sos@localhost) by sos.freebsd.dk (8.9.1/8.8.8) id UAA28524; Sat, 8 Aug 1998 20:17:23 +0200 (CEST) (envelope-from sos) Message-Id: <199808081817.UAA28524@sos.freebsd.dk> Subject: Re: system lock-up In-Reply-To: from Klaus Werner Krygier at "Aug 8, 98 07:18:22 pm" To: krygier@kph.uni-mainz.de (Klaus Werner Krygier) Date: Sat, 8 Aug 1998 20:17:23 +0200 (CEST) Cc: freebsd-smp@FreeBSD.ORG From: Søren Schmidt Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In reply to Klaus Werner Krygier who wrote: Hmm, I run two SMP systems here: 2*200Mhz P6 on TYAN S1662D NCR875 SCSI 4Gnarrow 96M 2*233Mhz P6 on TYAN S1668D EIDE UDMA 2*4.3G 1*11.5G 256M Both systems are rock stable, no bad symptoms whatsoever. Are you sure your SCSI cables/termination are OK ?? > Hello, > > here is a mail which I posted a few days ago to freebsd-current. I got > very few answers which didn't help. Perhaps in this mailing list there is > someone who can help me. > Do these problems occur only on a few SMP systems? > Are they specific for Gigabyte boards? > Which other hardware allows a stable system (like my computer as it worked > with a SNAP of May 1997 for about one year)? > > Klaus Werner Krygier > > ---------- Forwarded message ---------- > Date: Thu, 6 Aug 1998 20:58:34 +0200 (CEST) > From: Klaus Werner Krygier > To: Philippe Regnauld > Cc: freebsd-current@FreeBSD.ORG > Subject: Re: system lock-up (SMP, AHC_TAGENABLE, softupdates) > > On Sun, 2 Aug 1998, Philippe Regnauld wrote: > > > System: GigaByte 586-DX, 2xP5-133, 48MB, on-board 7880 SCSI, > > 4GB IBM DCAS-34330W. > > Current as of CTM src-cur 3477 (< 48 hours). > > > > Softupdates enabled on _all_ filesystems: no problem, did > > two make buildworlds/installworld in a row, as long > > as AHC_TAGENABLE is off. > > > > If I enable AHC_TAGENABLE, and push the system a bit (lots > > of FS activity) -> freeze. Last time it happened was with X, > > so I'd have to do it again in text mode to write down the AHC > > TAG messages... > > I have a very similar hardware setup on several machines (Gigabyte > 586-DX, 2 x P5-200, 64 MB, on-board 7880 SCSI, several different disks > (2GB, 4GB, 9GB)) and I have the same problems: > > If I do a lot of disk activity my computers freeze. Sometimes a SCSI > bus reset seems to be initiated (which I can recogize on the loading > activity of my CD changer). This problem occurs on standard 3.0-SNAP > kernels without softupdates and almost SMP-GENERIC configuration. I > tried several SNAPs, but the problem remained - I have the impression > it has become worse with newe versions :-(. At last I damaged one of > my disks. > > My last hope was CAM. Here (3.0CAM-19980712-SNAP) the system does not > freeze completely but becomes very slow on disk access and is also not > usable. One positive aspect: I get kernel messages. Every few seconds > the following message is printed out for ever: > > Timedout SCB handled by another timeout > Timedout SCB handled by another timeout > last message repeated 4 times > last message repeated 13 times > last message repeated 30 times > Timedout SCB handled by another timeout > Timedout SCB handled by another timeout > > etc. > > With older 3.0-SNAPs (e.g. May 97) a SMP system was very stable on my > first board and did not show such effects for almost a year. Because of > other problems and to use new features and improvements I had to upgrade > and have now very instable systems. > > Klaus Werner Krygier > > +-----------------------------------------------------------------------+ > | Dr. Klaus Werner Krygier | Email: krygier@kph.uni-mainz.de | > | Institut für Kernphysik | | > | Johannes Gutenberg-Universität | Tel: +49-6131-39-2960 | > | J.J.Becher-Weg 45 | +49-6131-39-5868 | > | D-55099 Mainz | Fax: +49-6131-39-2964 | > +-----------------------------------------------------------------------+ > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end? .. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Aug 8 12:02:44 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA13649 for freebsd-smp-outgoing; Sat, 8 Aug 1998 12:02:44 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from freebie.brann.org (doorman.brann.org [166.84.191.154]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA13643 for ; Sat, 8 Aug 1998 12:02:32 -0700 (PDT) (envelope-from jbrann@freebie.brann.org) Received: (from jbrann@localhost) by freebie.brann.org (8.9.1/8.8.7) id PAA00576; Sat, 8 Aug 1998 15:02:09 -0400 (EDT) (envelope-from jbrann) From: John Brann Message-Id: <199808081902.PAA00576@freebie.brann.org> Subject: Re: system lock-up In-Reply-To: from Klaus Werner Krygier at "Aug 8, 1998 7:18:22 pm" To: krygier@kph.uni-mainz.de (Klaus Werner Krygier) Date: Sat, 8 Aug 1998 15:02:09 -0400 (EDT) Cc: freebsd-smp@FreeBSD.ORG Organisation: Not while I'm at home X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I've just gone through the process of bringing my Toshiba equium up to SMP. (You'll find quite a few references to these machines in the list archives. the spec is 200MHz /256K P-Pro, Intel 440FX MB, on-board aic7880.) I also suffered from lockups due to the on-board Adaptec 7880 and the 4Gb Micropolis Tomahawk which come standard with these machines. I do not have / never have had softupdates enabled. I suffered from system freezes with no error messages, data overruns (VERY nasty - left the system unbootable) and timeouts. The situation was worse with AHC_TAGENABLE on, but I still got occasional lockups with it off. I'm afraid that the answers to my problems were all the ones you would expect. I found that the drive microcode was antiquated, and updated it (thanks to the lists) and careful scrutiny of the drive jumpers indicated that it had been installedaas the only device on the bus but without termination (!). The last problem I had came when I tried to install a second Micropolis drive in the same cage. Those things run HOT, and 2 mounted together internally caused really nasty problems after about an hour of running. The fact that the drive cage was almost too hot to touch when I opened the machine gave me a clue to that problem. After I had sorted out the microcode, termination and temperature, the drive now runs like a champ at 40M/s transfer rate with AHC_TAGENABLE on. make buildworld -j 8 had the machine running at a load >4 for more than an hour without any difficulties. The lists (scsi, stable and smp) all have references to this type of problem with Adaptec cards, but there doesn't seem to be any particular magic bullet solution, just the usual stuff. John Klaus Werner Krygier wrote... > > Hello, > > here is a mail which I posted a few days ago to freebsd-current. I got > very few answers which didn't help. Perhaps in this mailing list there is > someone who can help me. > Do these problems occur only on a few SMP systems? > Are they specific for Gigabyte boards? > Which other hardware allows a stable system (like my computer as it worked > with a SNAP of May 1997 for about one year)? > > Klaus Werner Krygier > > ---------- Forwarded message ---------- > Date: Thu, 6 Aug 1998 20:58:34 +0200 (CEST) > From: Klaus Werner Krygier > To: Philippe Regnauld > Cc: freebsd-current@FreeBSD.ORG > Subject: Re: system lock-up (SMP, AHC_TAGENABLE, softupdates) > > On Sun, 2 Aug 1998, Philippe Regnauld wrote: > > > System: GigaByte 586-DX, 2xP5-133, 48MB, on-board 7880 SCSI, > > 4GB IBM DCAS-34330W. > > Current as of CTM src-cur 3477 (< 48 hours). > > > > Softupdates enabled on _all_ filesystems: no problem, did > > two make buildworlds/installworld in a row, as long > > as AHC_TAGENABLE is off. > > > > If I enable AHC_TAGENABLE, and push the system a bit (lots > > of FS activity) -> freeze. Last time it happened was with X, > > so I'd have to do it again in text mode to write down the AHC > > TAG messages... > > I have a very similar hardware setup on several machines (Gigabyte > 586-DX, 2 x P5-200, 64 MB, on-board 7880 SCSI, several different disks > (2GB, 4GB, 9GB)) and I have the same problems: > > If I do a lot of disk activity my computers freeze. Sometimes a SCSI > bus reset seems to be initiated (which I can recogize on the loading > activity of my CD changer). This problem occurs on standard 3.0-SNAP > kernels without softupdates and almost SMP-GENERIC configuration. I > tried several SNAPs, but the problem remained - I have the impression > it has become worse with newe versions :-(. At last I damaged one of > my disks. > > My last hope was CAM. Here (3.0CAM-19980712-SNAP) the system does not > freeze completely but becomes very slow on disk access and is also not > usable. One positive aspect: I get kernel messages. Every few seconds > the following message is printed out for ever: > > Timedout SCB handled by another timeout > Timedout SCB handled by another timeout > last message repeated 4 times > last message repeated 13 times > last message repeated 30 times > Timedout SCB handled by another timeout > Timedout SCB handled by another timeout > > etc. > > With older 3.0-SNAPs (e.g. May 97) a SMP system was very stable on my > first board and did not show such effects for almost a year. Because of > other problems and to use new features and improvements I had to upgrade > and have now very instable systems. > > Klaus Werner Krygier > > +-----------------------------------------------------------------------+ > | Dr. Klaus Werner Krygier | Email: krygier@kph.uni-mainz.de | > | Institut für Kernphysik | | > | Johannes Gutenberg-Universität | Tel: +49-6131-39-2960 | > | J.J.Becher-Weg 45 | +49-6131-39-5868 | > | D-55099 Mainz | Fax: +49-6131-39-2964 | > +-----------------------------------------------------------------------+ > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > -- Unreal City, Under the brown fog of a winter dawn, finger jbrann@doorman.brann.org for pgp public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Aug 8 12:04:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA13787 for freebsd-smp-outgoing; Sat, 8 Aug 1998 12:04:15 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from sos.freebsd.dk (sos.freebsd.dk [212.242.40.180]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA13693 for ; Sat, 8 Aug 1998 12:03:52 -0700 (PDT) (envelope-from sos@sos.freebsd.dk) Received: (from sos@localhost) by sos.freebsd.dk (8.9.1/8.8.8) id VAA00319; Sat, 8 Aug 1998 21:02:43 +0200 (CEST) (envelope-from sos) Message-Id: <199808081902.VAA00319@sos.freebsd.dk> Subject: Re: system lock-up In-Reply-To: from Klaus Werner Krygier at "Aug 8, 98 08:39:47 pm" To: krygier@kph.uni-mainz.de (Klaus Werner Krygier) Date: Sat, 8 Aug 1998 21:02:43 +0200 (CEST) Cc: smp@FreeBSD.ORG From: Søren Schmidt Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In reply to Klaus Werner Krygier who wrote: > On Sat, 8 Aug 1998, Sxren Schmidt wrote: > > > In reply to Klaus Werner Krygier who wrote: > > > > Hmm, I run two SMP systems here: > > > > 2*200Mhz P6 on TYAN S1662D NCR875 SCSI 4Gnarrow 96M > > 2*233Mhz P6 on TYAN S1668D EIDE UDMA 2*4.3G 1*11.5G 256M > > > > Both systems are rock stable, no bad symptoms whatsoever. > > > > Are you sure your SCSI cables/termination are OK ?? > > Yes, without SMP everything works fine. And this happens on at least > three systems. Hmm, well when you run SMP there can be much more disk access going on so it could still be significant... Are all 3 system P5 based ?? It could be that we have gotten some P6 depended code in there... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end? .. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Aug 8 20:06:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA01538 for freebsd-smp-outgoing; Sat, 8 Aug 1998 20:06:58 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from johnny.syd.speednet.com.au (johnny.syd.speednet.com.au [202.135.206.16]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA01521 for ; Sat, 8 Aug 1998 20:06:54 -0700 (PDT) (envelope-from andyf@speednet.com.au) Received: from speednet.com.au (af.speednet.com.au [202.135.206.244]) by johnny.syd.speednet.com.au (8.8.8/8.8.8) with ESMTP id NAA23986 for ; Sun, 9 Aug 1998 13:06:33 +1000 (EST) (envelope-from andyf@speednet.com.au) Message-ID: <35CD1237.75F2C08A@speednet.com.au> Date: Sun, 09 Aug 1998 13:06:31 +1000 From: Andy Farkas Organization: Speed Internet Sevices X-Mailer: Mozilla 4.04 [en] (WinNT; I) MIME-Version: 1.0 To: freebsd-smp@FreeBSD.ORG Subject: Re: system lock-up References: <199808081817.UAA28524@sos.freebsd.dk> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by hub.freebsd.org id UAA01528 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Søren Schmidt wrote: > In reply to Klaus Werner Krygier ... > > Hmm, I run two SMP systems here: > > 2*200Mhz P6 on TYAN S1662D NCR875 SCSI 4Gnarrow 96M > 2*233Mhz P6 on TYAN S1668D EIDE UDMA 2*4.3G 1*11.5G 256M > > Both systems are rock stable, no bad symptoms whatsoever. > Which SNAP are you running? -- :{ andyf@speednet.com.au Andy Farkas System Administrator Speed Internet Services http://www.speednet.com.au/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message