From owner-freebsd-scsi Sun Feb 9 07:23:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA20136 for freebsd-scsi-outgoing; Sun, 9 Feb 1997 07:23:59 -0800 (PST) Received: from sonyinet.sony.co.jp (sonyinet.sony.co.jp [202.238.80.17]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA20131 for ; Sun, 9 Feb 1997 07:23:56 -0800 (PST) Received: from sonygw.sony.co.jp ([43.0.1.249]) by sonyinet.sony.co.jp (8.6.10/3.4W4-96030410) with ESMTP id AAA25582; Mon, 10 Feb 1997 00:23:53 +0900 Received: from s64.saskg.semicon.sony.co.jp ([43.0.234.82]) by sonygw.sony.co.jp (8.6.12+2.4W/3.4W-97020512) with ESMTP id AAA10793; Mon, 10 Feb 1997 00:23:37 +0900 Received: from s64 by s64.saskg.semicon.sony.co.jp (8.7.4+2.6Wbeta6/3.4W5-Mimoric1.2-MX) id AAA11861; Mon, 10 Feb 1997 00:24:15 +0900 (JST) Message-Id: <199702091524.AAA11861@s64.saskg.semicon.sony.co.jp> To: rmike@sbox.tu-graz.ac.at cc: freebsd-scsi@freebsd.org, saeki@saskg.semicon.sony.co.jp, saeki@jp.freebsd.org Subject: Re: AHA2920, TMC1830 - need help - need PAO (pccard-test) source! In-reply-to: Your message of "Sat, 08 Feb 1997 10:41:55 JST." Date: Mon, 10 Feb 1997 00:24:15 +0900 From: Takashi Saeki Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi! rmike@sbox.tu-graz.ac.at wrote: > It would be nice, if someone can tell me, where I can find the source for > PAO, or which person I should contact! > You can get latest PAO source from ftp://jaz.jp.freebsd.org/~ftp/pub/FreeBSD-jp/PAO/, PAO-970210.tar.gz must be latest now. Person to contact is, I think, Atsushi Furuta . He ports the driver from NetBSD/pc98. Original auther of the driver are, Naofumi Honda and Kouichi Matsuda # They are member of the NetBSD/pc98 team, not FreeBSD people. Good luck! -------------------------------------------------------------------------- Takashi Saeki saeki@jp.freebsd.org -------------------------------------------------------------------------- From owner-freebsd-scsi Sun Feb 9 17:36:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA27115 for freebsd-scsi-outgoing; Sun, 9 Feb 1997 17:36:51 -0800 (PST) Received: from mail.futuresouth.com (mail.futuresouth.com [207.141.254.21]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA27104 for ; Sun, 9 Feb 1997 17:36:39 -0800 (PST) Received: from shell.futuresouth.com (shell.futuresouth.com [207.141.254.20]) by mail.futuresouth.com (8.8.5/8.8.5) with ESMTP id TAA06663; Sun, 9 Feb 1997 19:36:06 -0600 (CST) From: Tim Tsai Received: (from tim@localhost) by shell.futuresouth.com (8.8.3/8.8.3) id TAA16088; Sun, 9 Feb 1997 19:36:05 -0600 (CST) Message-Id: <199702100136.TAA16088@shell.futuresouth.com> Subject: Re: how to use tape changer ch0 To: langfod@dihelix.com (David Langford) Date: Sun, 9 Feb 1997 19:36:05 -0600 (CST) Cc: scsi@freebsd.org In-Reply-To: <199702082111.LAA10163@caliban.dihelix.com> from David Langford at "Feb 8, 97 11:09:54 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I have a Seagate/Conner/Archive 4586NP that I would like to > get working with Amanda. > Shows as ether st0 or ch0 but never both. It works great as st0 > if I feel like manully flipping through the tapes. I would like > to get it working for automated backups but I cant seem to figure > out how one would go about using the ch0 driver. I have the same setup. Try the following: chio move slot 0 drive 0 (to load tape) chio move drive 0 slot 0 (to eject) I don't have it working with Amanda yet but I see /dev/ch* and /dev/nrst* /dev/rst*, etc. I am running 2.1.6.1 BTW. Tim PS: Please let me know if you get Amanda working. I've got the tpchanger script working but it generates quite a few warning messages (mostly st0 timeouts, etc.). From owner-freebsd-scsi Mon Feb 10 02:16:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA23544 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 02:16:24 -0800 (PST) Received: from madoka.hal.t.u-tokyo.ac.jp (madoka.hal.t.u-tokyo.ac.jp [133.11.98.129]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA23520; Mon, 10 Feb 1997 02:16:11 -0800 (PST) Received: from hal.t.u-tokyo.ac.jp (mayu.hal.t.u-tokyo.ac.jp [133.11.98.131]) by madoka.hal.t.u-tokyo.ac.jp (8.8.5+2.7Wbeta5/3.5Wpl2/HALmailhost/97020422) with ESMTP id TAA06339; Mon, 10 Feb 1997 19:12:29 +0900 (JST) Message-Id: <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> To: asami@vader.cs.berkeley.edu (Satoshi Asami) Cc: questions@freebsd.org, scsi@freebsd.org Subject: Re: HELP!! ST32155W - Not detected during probing!! In-reply-to: Your message of "Fri, 07 Feb 1997 13:34:48 PST" References: <199702072134.NAA29174@vader.cs.berkeley.edu> Date: Mon, 10 Feb 1997 19:12:28 +0900 From: "" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi!, I solved the problem thanks to suggestions I received. Thank you!! The problem was that there was a clash of SCSI ID'ss. CDROM I use (Matsushita 8X) supposed to come with factory preset SCSI ID#6 and the Hard Disk at #0. What has happened was that Gateway people seemed to have changed the CDROM ID to #0. (I've NO idea why). Fortunately 'cos HD is SCAM compliant, 2940uw automatically detects the conflict and assigns HD #15. This works for Win95. But not for FreeBSD. So I changed the CDROM SCSI ID back to #6 and now everything is OK!! Best Regards, Ajith. ------------------------------------------------------------------------------ Ajith Pasqual - Dept of Info. & Comm. Eng., Univ. of Tokyo. Email:pasqual@hal.t.u-tokyo.ac.jp (WWW)http://www.hal.t.u-tokyo.ac.jp/~pasqual/ ------------------------------------------------------------------------------ From owner-freebsd-scsi Mon Feb 10 03:19:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA26018 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 03:19:25 -0800 (PST) Received: from shadows.aeon.net (shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA26007 for ; Mon, 10 Feb 1997 03:19:10 -0800 (PST) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.8.5/8.8.3) id NAA27160 for freebsd-scsi@freebsd.org; Mon, 10 Feb 1997 13:17:49 +0200 (EET) From: mika ruohotie Message-Id: <199702101117.NAA27160@shadows.aeon.net> Subject: weirdo scsi... To: freebsd-scsi@freebsd.org Date: Mon, 10 Feb 1997 13:17:49 +0200 (EET) X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk yeah, i know this is probably ooold news, i tried to look from the archives but didnt find exactly this one... Feb 10 10:05:21 pluto /kernel: sd1(ahc0:1:0): BUS DEVICE RESET message queued. Feb 10 10:05:21 pluto /kernel: Bus Device Reset Message Sent Feb 10 10:05:21 pluto /kernel: sd1(ahc0:1:0): Bus Device Reset delivered. 2 SCBs aborted Feb 10 10:05:21 pluto /kernel: sd1(ahc0:1:0): UNIT ATTENTION asc:29,3 Feb 10 10:05:21 pluto /kernel: , retries:4 that's probably nothing dangerous, but what is it? system is 2.1.5-RELEASE (soon 2.1.6-WHATEVER) and uses 3940UWD and four fujitsu FUJITSU M2952S-512 0124" type 0 fixed SCSI 2 drives, it should be ultra, not wide. mickey From owner-freebsd-scsi Mon Feb 10 05:53:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA03565 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 05:53:07 -0800 (PST) Received: from mercury.ukc.ac.uk (mercury.ukc.ac.uk [129.12.21.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id FAA03560 for ; Mon, 10 Feb 1997 05:53:03 -0800 (PST) Received: from crane.ukc.ac.uk by mercury.ukc.ac.uk with SMTP (PP); Mon, 10 Feb 1997 13:52:40 +0000 Received: from localhost by crane.ukc.ac.uk (SMI-8.6/UKC-2.14) id NAA25556; Mon, 10 Feb 1997 13:52:34 GMT Date: Mon, 10 Feb 1997 13:52:34 +0000 (GMT) From: "K.R.Marshall" X-Sender: krm2@crane To: Joerg Wunsch cc: scsi@freebsd.org Subject: Re: Nakamichi 4-disk changers In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 8 Feb 1997, J Wunsch wrote: > As K.R.Marshall wrote: > > > This server won't go online for a couple of weeks so if anyone wants me to > > test out CD-ROM changer code I'll be happy to oblige. > > I've got a Nakamichi 7-disk changer here for testing. The biggest > problem i found (after fixing the ``Logical unit is in the process of > becoming ready'' problem) is that the drives are likely to go > thrashing on concurrent access, since the kernel doesn't lock one > medium in a device for long enough, so the drives will finally be very > busy constantly swapping their media. The way I will be configuring it the swapping shouldn't be too much of a problem. This won't be a "full" networked server in he normal sense. It's more of a general repository for CDs away from the public - each CD will only ever be accessed by one machine at a time so I should be able to avoid heavy swapping by judiciously arranging the CDs in the changers. I wouldn't yet consider such a system for a full CD-ROM network, but hopefully FreeBSD will support such things in a better way at some point in the future..? I understand there are now NT-based systems which deal intelligently with the problems of changers (i.e. with cacheing etc..) Currently we just use normal stacks of CD-ROM drives and a large 9Gb hard disk with several packages copied to it for faster access (all hanging off a FreeBSD box, of course..). Keith. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ "Mummy was an asteroid, Daddy was a | Keith Marshall small, non-stick kitchen utensil..." | Computing Officer, Templeman Library - Quiet Sun, 1975 | University of Kent at Canterbury. From owner-freebsd-scsi Mon Feb 10 06:53:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA05625 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 06:53:58 -0800 (PST) Received: from mail.ruhrgebiet.individual.net ([141.39.224.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA05620 for ; Mon, 10 Feb 1997 06:53:55 -0800 (PST) Received: from robkaos.ruhr.de (admin@localhost) by mail.ruhrgebiet.individual.net (8.8.5/8.8.5) with UUCP id OAA19460 for freebsd.org!freebsd-scsi; Mon, 10 Feb 1997 14:52:10 +0100 (MET) Received: by robkaos.ruhr.de (/\oo/\ Smail3.1.29.1 #29.1) id ; Mon, 10 Feb 97 14:50 MET Message-Id: From: robsch@robkaos.ruhr.de (Robert Schien) Subject: Problem with 3701B CD-ROM To: freebsd-scsi@freebsd.org Date: Mon, 10 Feb 1997 14:50:41 +0100 (MET) X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Last week I upgraded my hardware from a DX486-50 to a P6-200 (256k). The motherboard is a P6NP5 from ASUS , the SCSI host adapter a SC-200. All works fine except the Toshiba 3701B CD-ROM drive which sometimes gives the following error messages: /kernel: cd0 at scbus0 target 2 lun 0: COMMAND FAILED (9 0) @f20ba800. /kernel: cd0 at scbus0 target 2 lun 0: extraneous data discarded. The errors disappear when the kernel option FAILSAFE is set. Are the Toshs known to have some SCSI problem? TIA Robert From owner-freebsd-scsi Mon Feb 10 08:19:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA09974 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 08:19:59 -0800 (PST) Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA09968 for ; Mon, 10 Feb 1997 08:19:46 -0800 (PST) Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id IAA15723; Mon, 10 Feb 1997 08:18:52 -0800 (PST) Message-ID: <32FF4A6C.2781E494@progroup.com> Date: Mon, 10 Feb 1997 08:18:52 -0800 From: Craig Shaver Organization: Productivity Group, Inc. X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386) MIME-Version: 1.0 To: scsi@freebsd.org Subject: Re: 2940 and ncr target problems re: HELP!! ST32155W - Not detected during probing!! References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi!, > > I solved the problem thanks to suggestions I received. Thank you!! > > The problem was that there was a clash of SCSI ID'ss. CDROM I use (Matsushita > 8X) supposed to come with factory preset SCSI ID#6 and the Hard Disk at > #0. What has happened was that Gateway people seemed to have changed the CDROM > ID to #0. (I've NO idea why). Fortunately 'cos HD is SCAM compliant, 2940uw > automatically detects the conflict and assigns HD #15. This works for > Win95. But not for FreeBSD. So I changed the CDROM SCSI ID back to #6 and now > everything is OK!! > > Best Regards, > Ajith. I believe this may point out a significant problem with the driver for the 2940W and most likely the ncr also. I have tried to use a wide drive on the ncr825 and 875 at a target id greater than 7. It would not work. I am not sure of the exact reason, but after seeing the problems Ajith had, I believe there is a bug in the drivers. I went through the ncr.c source for the 2.1.6 driver and saw in every instance, that the target id was being masked with the literal '7' which is '111'b. This only allows a total of 8 targets on the board. The wide controllers physically allow up to 16 targets. That would be a mask of '1111'b, or 15, or 017, or 0x0f. Of course target 7 is usually reserved for the target id of the controller hba itself. Am I wrong? Is anyone using a higher target id with these device drivers? -- Craig Shaver (craig@progroup.com) (415)390-0654 Productivity Group POB 60458 Sunnyvale, CA 94088 From owner-freebsd-scsi Mon Feb 10 13:52:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA27665 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 13:52:44 -0800 (PST) Received: from freebee.tu-graz.ac.at (root@freebee.tu-graz.ac.at [129.27.193.128]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA27608; Mon, 10 Feb 1997 13:52:13 -0800 (PST) Received: from dwarf.tu-graz.ac.at (isdn052.tu-graz.ac.at [129.27.240.52]) by freebee.tu-graz.ac.at (8.6.11/8.6.9) with ESMTP id WAA08696; Mon, 10 Feb 1997 22:52:06 +0100 Received: (from rmike@localhost) by dwarf.tu-graz.ac.at (8.7.5/8.7.3) id WAA00284; Mon, 10 Feb 1997 22:50:34 +0100 (MET) Date: Mon, 10 Feb 1997 22:50:34 +0100 (MET) From: Michael Ranner Reply-To: rmike@sbox.tu-graz.ac.at To: freebsd-scsi@freebsd.org, freebsd-hardware@freebsd.org, freebsd-hackers@freebsd.org Subject: Adaptec 2920 PCI driver (TMC18c30) - need persons to test! Message-ID: Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi there! This afternoon I have written the necessary code to get the Adaptec 2920 running under FreeBSD! It is based on parts of the latest PAO release and it's developed under FreeBSD 2.1.5. It was only necessary to write the PCI stuff! Now I need some persons to test the driver more accurate. Regards, /\/\ike /\/\ichael Ranner - rmike@sbox.tu-graz.ac.at http://www.sbox.tu.graz.ac.at/home/rmike/ --- end of message - non-sense follows --- ________ .' `. / \ |_____ _____| (_____><_____) \ /\ / \ oo / Grey-type ASCII encounter ... \ __ / `----' From owner-freebsd-scsi Mon Feb 10 15:21:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA01855 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 15:21:06 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA01849 for ; Mon, 10 Feb 1997 15:20:59 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA27642; Tue, 11 Feb 1997 00:20:43 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.6.9) id XAA20496; Mon, 10 Feb 1997 23:58:19 +0100 (MET) Message-ID: Date: Mon, 10 Feb 1997 23:58:19 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: K.R.Marshall@ukc.ac.uk (K.R.Marshall) Cc: scsi@freebsd.org Subject: Re: Nakamichi 4-disk changers References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from K.R.Marshall on Feb 10, 1997 13:52:34 +0000 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As K.R.Marshall wrote: > I wouldn't yet consider such a system for a full CD-ROM network, but > hopefully FreeBSD will support such things in a better way at some point > in the future..? Still seeking an idea how to do it best. The only thing we've got so far was throwing some ideas. Sure, we wouldn't mind if you submit some nice patches for it. :-) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Mon Feb 10 17:20:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA01171 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 17:20:52 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA01160 for ; Mon, 10 Feb 1997 17:20:46 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id CAA00204; Tue, 11 Feb 1997 02:20:42 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.6.9) id BAA21481; Tue, 11 Feb 1997 01:51:21 +0100 (MET) Message-ID: Date: Tue, 11 Feb 1997 01:51:20 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: robsch@robkaos.ruhr.de (Robert Schien) Cc: freebsd-scsi@freebsd.org Subject: Re: Problem with 3701B CD-ROM References: X-Mailer: Mutt 0.55-PL10 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Robert Schien on Feb 10, 1997 14:50:41 +0100 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Robert Schien wrote: > /kernel: cd0 at scbus0 target 2 lun 0: COMMAND FAILED (9 0) @f20ba800. > /kernel: cd0 at scbus0 target 2 lun 0: extraneous data discarded. > > The errors disappear when the kernel option FAILSAFE is set. That would mean they don't grok tagged queuing. When reviewing Mark Gourney's (sp?) various SCSI-related section 9 manpages, i learned that we support quirk entries for drives that don't grok tagged commands. Perhaps we should start populating scsiconf.c with entries for ``known rogues''. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Mon Feb 10 21:28:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA13731 for freebsd-scsi-outgoing; Mon, 10 Feb 1997 21:28:15 -0800 (PST) Received: from dfw-ix7.ix.netcom.com (dfw-ix7.ix.netcom.com [206.214.98.7]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA13726 for ; Mon, 10 Feb 1997 21:28:14 -0800 (PST) Received: (from smap@localhost) by dfw-ix7.ix.netcom.com (8.8.4/8.8.4) id XAA20349; Mon, 10 Feb 1997 23:27:06 -0600 (CST) Received: from wck-ca11-11.ix.netcom.com(204.31.231.171) by dfw-ix7.ix.netcom.com via smap (V1.3) id sma020306; Mon Feb 10 23:27:05 1997 Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id VAA24440; Mon, 10 Feb 1997 21:27:00 -0800 (PST) Date: Mon, 10 Feb 1997 21:27:00 -0800 (PST) Message-Id: <199702110527.VAA24440@silvia.HIP.Berkeley.EDU> To: craig@progroup.com CC: scsi@freebsd.org In-reply-to: <32FF4A6C.2781E494@progroup.com> (message from Craig Shaver on Mon, 10 Feb 1997 08:18:52 -0800) Subject: Re: 2940 and ncr target problems re: HELP!! ST32155W - Not detected during probing!! From: asami@vader.cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * Am I wrong? Is anyone using a higher target id with these device * drivers? I've connected up to 15 devices on an Adaptec 2940. Satoshi From owner-freebsd-scsi Tue Feb 11 09:35:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA06458 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 09:35:52 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA06451 for ; Tue, 11 Feb 1997 09:35:47 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr2-44.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA14925 (5.67b/IDA-1.5 for ); Tue, 11 Feb 1997 18:35:41 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id SAA21977; Tue, 11 Feb 1997 18:35:38 +0100 (CET) Message-Id: <19970211183534.ZZ52644@x14.mi.uni-koeln.de> Date: Tue, 11 Feb 1997 18:35:34 +0100 From: se@freebsd.org (Stefan Esser) To: craig@progroup.com (Craig Shaver) Cc: scsi@freebsd.org Subject: Re: 2940 and ncr target problems re: HELP!! ST32155W - Not detected during probing!! References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> X-Mailer: Mutt 0.60-PL0 Mime-Version: 1.0 In-Reply-To: <32FF4A6C.2781E494@progroup.com>; from Craig Shaver on Feb 10, 1997 08:18:52 -0800 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Feb 10, craig@progroup.com (Craig Shaver) wrote: > I believe this may point out a significant problem with the driver for > the > 2940W and most likely the ncr also. I have tried to use a wide drive on > the ncr825 and 875 at a target id greater than 7. It would not work. I > am not sure of the exact reason, but after seeing the problems Ajith > had, > I believe there is a bug in the drivers. > > I went through the ncr.c source for the 2.1.6 driver and saw in every > instance, that the target id was being masked with the literal '7' which > is '111'b. This only allows a total of 8 targets on the board. The > wide > controllers physically allow up to 16 targets. That would be a mask of > '1111'b, or 15, or 017, or 0x0f. Of course target 7 is usually reserved > for > the target id of the controller hba itself. This has been fixed in the NCR driver in -current and -stable for some time, but the changes had not made it into 2.1.6. Regards, STefan From owner-freebsd-scsi Tue Feb 11 10:20:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA08729 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 10:20:44 -0800 (PST) Received: from amazon.cs.umd.edu (amazon.cs.umd.edu [128.8.128.40]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA08724 for ; Tue, 11 Feb 1997 10:20:40 -0800 (PST) From: jlee@cs.umd.edu Received: by amazon.cs.umd.edu (8.8.5/UMIACS-0.9/04-05-88) id NAA00899; Tue, 11 Feb 1997 13:19:37 -0500 (EST) Message-Id: <199702111819.NAA00899@amazon.cs.umd.edu> Subject: Re: Adaptec 2940 data overruns continue To: gibbs@narnia.plutotech.com (Justin T. Gibbs) Date: Tue, 11 Feb 1997 13:19:36 -0500 (EST) Cc: freebsd-scsi@freebsd.org In-Reply-To: <199702111740.KAA04890@narnia.plutotech.com> from "Justin T. Gibbs" at Feb 11, 97 10:40:44 am MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > >I'm still getting heaps of: > > > > sd1: data overrun of 16777215 bytes detected. Forcing a retry. > > > >messages appearing. > > > >A week or so ago I was advised to get a later aic7xxxx.c; I'm now running > >3.0-CURRENT from just before the lite2 merge. Is this late enough? > > I just committed the change to fix this today. The patch at the end > of this mail should bring you up to date without having to deal with > Lite2 stuff. > > -- > Justin T. Gibbs I am curious whether this fix is applied to RELENG_2_2 too? I am still getting the overrun messages under 2.2-GAMMA. Thank you for your time and effort. J Lee (jlee@amazon.cs.umd.edu) From owner-freebsd-scsi Tue Feb 11 10:29:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA09282 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 10:29:13 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA09263 for ; Tue, 11 Feb 1997 10:29:09 -0800 (PST) Received: from narnia (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.5/8.7.3) with ESMTP id LAA05116; Tue, 11 Feb 1997 11:29:09 -0700 (MST) Message-Id: <199702111829.LAA05116@narnia.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: jlee@cs.umd.edu cc: freebsd-scsi@freebsd.org Subject: Re: Adaptec 2940 data overruns continue In-reply-to: Your message of "Tue, 11 Feb 1997 13:19:36 EST." <199702111819.NAA00899@amazon.cs.umd.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 11 Feb 1997 11:29:09 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I am curious whether this fix is applied to RELENG_2_2 too? >I am still getting the overrun messages under 2.2-GAMMA. It will be later today, hopefully. The patch I included will not apply cleanly to 2.2-GAMMA as there are other changes that also need to be incorperated to that branch. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Tue Feb 11 10:47:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA10087 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 10:47:26 -0800 (PST) Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA10076; Tue, 11 Feb 1997 10:47:13 -0800 (PST) Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id KAA05421; Tue, 11 Feb 1997 10:45:47 -0800 (PST) Message-ID: <3300BE5B.59E2B600@progroup.com> Date: Tue, 11 Feb 1997 10:45:47 -0800 From: Craig Shaver Organization: Productivity Group, Inc. X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386) MIME-Version: 1.0 To: Stefan Esser CC: scsi@freebsd.org Subject: Re: 2940 and ncr target problems re: HELP!! ST32155W - Not detected during probing!! References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> <19970211183534.ZZ52644@x14.mi.uni-koeln.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Stefan Esser wrote: > > On Feb 10, craig@progroup.com (Craig Shaver) wrote: > > I believe this may point out a significant problem with the driver for > > the del .... > > the target id of the controller hba itself. > > This has been fixed in the NCR driver in -current and -stable > for some time, but the changes had not made it into 2.1.6. > > Regards, STefan Will this be in 2.1.7R? I was going to try to apply the Tekram patches to the 2.1.7R sources, and see if I can build a kernel with the tekram driver. On another note, will I be able to boot using the GENERIC kernel against the Tekram controller? I am under the impression the base driver will recognize the controller as an 825. Thanks, -- Craig Shaver (craig@progroup.com) (415)390-0654 Productivity Group POB 60458 Sunnyvale, CA 94088 From owner-freebsd-scsi Tue Feb 11 11:25:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA12210 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 11:25:56 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA12199 for ; Tue, 11 Feb 1997 11:25:51 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr3-17.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA16414 (5.67b/IDA-1.5 for ); Tue, 11 Feb 1997 20:25:38 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id UAA22459; Tue, 11 Feb 1997 20:24:56 +0100 (CET) Message-Id: <19970211202335.YC36261@x14.mi.uni-koeln.de> Date: Tue, 11 Feb 1997 20:23:35 +0100 From: se@freebsd.org (Stefan Esser) To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Cc: robsch@robkaos.ruhr.de (Robert Schien), freebsd-scsi@freebsd.org Subject: Re: Problem with 3701B CD-ROM References: X-Mailer: Mutt 0.60-PL0 Mime-Version: 1.0 In-Reply-To: ; from J Wunsch on Feb 11, 1997 01:51:20 +0100 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Feb 11, j@uriah.heep.sax.de (J Wunsch) wrote: > As Robert Schien wrote: > > > /kernel: cd0 at scbus0 target 2 lun 0: COMMAND FAILED (9 0) @f20ba800. > > /kernel: cd0 at scbus0 target 2 lun 0: extraneous data discarded. > > > > The errors disappear when the kernel option FAILSAFE is set. > > That would mean they don't grok tagged queuing. Like just about any CDROM drive on the market today :) I've never had a single problem with my Toshiba 3601, and I think the error is caused by some application sending an invalid request. What program did access the CDROM drive at that time ? > When reviewing Mark Gourney's (sp?) various SCSI-related section 9 > manpages, i learned that we support quirk entries for drives that > don't grok tagged commands. Perhaps we should start populating > scsiconf.c with entries for ``known rogues''. Well, I did the "NEW_SCSICONF" changes back in September 1994, in part to move some of the code that dealt with buggy devices out of the NCR driver. No driver (writer) took notice, even after you made the new code the default (in December 1995). Justin Gibbs' new generic SCSI code will take care of tagged commands, and I suggest to just wait for him to complete his interfaces, before any driver is modified ... Regards, STefan From owner-freebsd-scsi Tue Feb 11 11:42:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA13377 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 11:42:00 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA13336; Tue, 11 Feb 1997 11:41:32 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr2-43.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA16618 (5.67b/IDA-1.5); Tue, 11 Feb 1997 20:41:16 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id UAA22537; Tue, 11 Feb 1997 20:40:33 +0100 (CET) Message-Id: <19970211203913.GS05631@x14.mi.uni-koeln.de> Date: Tue, 11 Feb 1997 20:39:13 +0100 From: se@freebsd.org (Stefan Esser) To: craig@progroup.com (Craig Shaver) Cc: se@freebsd.org (Stefan Esser), scsi@freebsd.org Subject: Re: 2940 and ncr target problems re: HELP!! ST32155W - Not detected during probing!! References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> <19970211183534.ZZ52644@x14.mi.uni-koeln.de> <3300BE5B.59E2B600@progroup.com> X-Mailer: Mutt 0.60-PL0 Mime-Version: 1.0 In-Reply-To: <3300BE5B.59E2B600@progroup.com>; from Craig Shaver on Feb 11, 1997 10:45:47 -0800 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Feb 11, craig@progroup.com (Craig Shaver) wrote: > > This has been fixed in the NCR driver in -current and -stable > > for some time, but the changes had not made it into 2.1.6. > Will this be in 2.1.7R? I was going to try to apply the Tekram patches Sure. The 2.1.7 release will be based on -stable. > to the 2.1.7R sources, and see if I can build a kernel with the tekram > driver. You may of course try the Tekram driver. I'd like to know how well it works for you. (If you can run some benchmarks, I'd love to see the results.) There might be some minor problems with the Tekram driver, which I'd have to check in the sources. > On another note, will I be able to boot using the GENERIC kernel > against the Tekram controller? I am under the impression the base > driver will recognize the controller as an 825. No, the controller will be fully recognized, but many of the new features of the 53c875 will not be taken advantage of. (There is no support for loading the NCR code into the 875's 4KB SRAM, for example, or no negotiation of Ultra-SCSI transfer rates.) I wanted to clean up some of the code before the 20MHz speed is enabled, but was held back by some other (higher priority) tasks ... But I got reports of the "generic" NCR driver giving significantly better throughput anyway, and I guess I know why this is the case :) Regards, STefan From owner-freebsd-scsi Tue Feb 11 12:14:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA15439 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 12:14:47 -0800 (PST) Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA15425; Tue, 11 Feb 1997 12:14:11 -0800 (PST) Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id MAA08084; Tue, 11 Feb 1997 12:13:13 -0800 (PST) Message-ID: <3300D2D9.3F54BC7E@progroup.com> Date: Tue, 11 Feb 1997 12:13:13 -0800 From: Craig Shaver Organization: Productivity Group, Inc. X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386) MIME-Version: 1.0 To: Stefan Esser CC: scsi@freebsd.org Subject: Re: ncr in stable ... 875 References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> <19970211183534.ZZ52644@x14.mi.uni-koeln.de> <3300BE5B.59E2B600@progroup.com> <19970211203913.GS05631@x14.mi.uni-koeln.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Stefan Esser wrote: > > > > You may of course try the Tekram driver. I'd like > to know how well it works for you. (If you can run > some benchmarks, I'd love to see the results.) I am running the Tekram now, on a 2.1.5R with the Tekram patches, and a minor patch from you. Would you like me to run bonnie on it? I have never run bonnie, but can give it a try. > > There might be some minor problems with the Tekram > driver, which I'd have to check in the sources. Ouch! Well I am grateful to the Tekram people for even putting a FreeBSD driver up in the first place. Even if your code is better, I would still want to encourage their future support of FreeBSD. > > > On another note, will I be able to boot using the GENERIC kernel > > against the Tekram controller? I am under the impression the base > > driver will recognize the controller as an 825. > > No, the controller will be fully recognized, but > many of the new features of the 53c875 will not > be taken advantage of. (There is no support for > loading the NCR code into the 875's 4KB SRAM, for > example, or no negotiation of Ultra-SCSI transfer > rates.) Ok, I gather that it support UW and W devices, but will not negotiate the setup. Sounds like it should work out of the box for me. > > I wanted to clean up some of the code before the > 20MHz speed is enabled, but was held back by some > other (higher priority) tasks ... Is it enabled by default? Will regular wide drives like the Q Atlas work? I want to use it with a 2G seagate UW, and 2 4G Atlas drives. They all work now. > > But I got reports of the "generic" NCR driver > giving significantly better throughput anyway, > and I guess I know why this is the case :) > > Regards, STefan Sounds good, thanks for the information and help... -- Craig Shaver (craig@progroup.com) (415)390-0654 Productivity Group POB 60458 Sunnyvale, CA 94088 From owner-freebsd-scsi Tue Feb 11 12:50:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA17435 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 12:50:51 -0800 (PST) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA17425; Tue, 11 Feb 1997 12:50:36 -0800 (PST) Received: from x14.mi.uni-koeln.de (annexr3-6.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA17288 (5.67b/IDA-1.5); Tue, 11 Feb 1997 21:50:25 +0100 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id VAA22817; Tue, 11 Feb 1997 21:49:42 +0100 (CET) Message-Id: <19970211214822.TV64263@x14.mi.uni-koeln.de> Date: Tue, 11 Feb 1997 21:48:22 +0100 From: se@freebsd.org (Stefan Esser) To: craig@progroup.com (Craig Shaver) Cc: se@freebsd.org (Stefan Esser), scsi@freebsd.org Subject: Re: ncr in stable ... 875 References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> <19970211183534.ZZ52644@x14.mi.uni-koeln.de> <3300BE5B.59E2B600@progroup.com> <19970211203913.GS05631@x14.mi.uni-koeln.de> <3300D2D9.3F54BC7E@progroup.com> X-Mailer: Mutt 0.60-PL0 Mime-Version: 1.0 In-Reply-To: <3300D2D9.3F54BC7E@progroup.com>; from Craig Shaver on Feb 11, 1997 12:13:13 -0800 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Feb 11, craig@progroup.com (Craig Shaver) wrote: > > You may of course try the Tekram driver. I'd like > > to know how well it works for you. (If you can run > > some benchmarks, I'd love to see the results.) > > I am running the Tekram now, on a 2.1.5R with the Tekram > patches, and a minor patch from you. Would you like me > to run bonnie on it? I have never run bonnie, but can > give it a try. Well, it's only really useful, if you are willing to run the benchmark with both drivers ... > > There might be some minor problems with the Tekram > > driver, which I'd have to check in the sources. > > Ouch! Well I am grateful to the Tekram people for even > putting a FreeBSD driver up in the first place. Even if > your code is better, I would still want to encourage their > future support of FreeBSD. Sure! I really appreciate, that a company cares for the free Unix market, and understands, that even if there are many more Win-3.1 or Win-95 systems than BSD or Linux, most of them will never see a SCSI device from close distance :) > > No, the controller will be fully recognized, but > > many of the new features of the 53c875 will not > > be taken advantage of. (There is no support for > > loading the NCR code into the 875's 4KB SRAM, for > > example, or no negotiation of Ultra-SCSI transfer > > rates.) > > Ok, I gather that it support UW and W devices, but > will not negotiate the setup. Sounds like it should > work out of the box for me. I wanted the GENERIC FreeBSD kernel to support all NCR 53c8xx based SCSI cards at least to the degree required to complete a kernel install. And this has been the case for many months, now. > > I wanted to clean up some of the code before the > > 20MHz speed is enabled, but was held back by some > > other (higher priority) tasks ... > > Is it enabled by default? Will regular wide drives > like the Q Atlas work? I want to use it with a 2G > seagate UW, and 2 4G Atlas drives. They all work now. The driver will only support 10MHz WIDE transfers, i.e. 20MB/s. I could provide you with diffs that make the driver negotiate 20MHz speed, but have not even tested them myself ... I have half-way completed the cleanup of the code, that deals with negotiation values. It will use "abstract" variables, instead of "register bits". Regards, STefan From owner-freebsd-scsi Tue Feb 11 13:39:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA20665 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 13:39:23 -0800 (PST) Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA20576; Tue, 11 Feb 1997 13:37:27 -0800 (PST) Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id NAA10726; Tue, 11 Feb 1997 13:36:00 -0800 (PST) Message-ID: <3300E640.ABD322C@ProGroup.com> Date: Tue, 11 Feb 1997 13:36:00 -0800 From: Craig Shaver Organization: Productivity Group, Inc. X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386) MIME-Version: 1.0 To: Stefan Esser CC: scsi@freebsd.org Subject: Re: ncr in stable ... 875 References: <199702072134.NAA29174@vader.cs.berkeley.edu> <199702101012.TAA06339@madoka.hal.t.u-tokyo.ac.jp> <32FF4A6C.2781E494@progroup.com> <19970211183534.ZZ52644@x14.mi.uni-koeln.de> <3300BE5B.59E2B600@progroup.com> <19970211203913.GS05631@x14.mi.uni-koeln.de> <3300D2D9.3F54BC7E@progroup.com> <19970211214822.TV64263@x14.mi.uni-koeln.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Stefan Esser wrote: > > On Feb 11, craig@progroup.com (Craig Shaver) wrote: > > > You may of course try the Tekram driver. I'd like > > > to know how well it works for you. (If you can run > > > some benchmarks, I'd love to see the results.) > > > > I am running the Tekram now, on a 2.1.5R with the Tekram > > patches, and a minor patch from you. Would you like me > > to run bonnie on it? I have never run bonnie, but can > > give it a try. > > Well, it's only really useful, if you are willing to > run the benchmark with both drivers ... I have several machines set up here. I can set up an older asus p55tp4xe p133 64MB to bench the 2.1.7R with your driver, patch in the tekram stuff and bench that, patch your driver for ultra negotiations, and test + bench that. Please send your ultra patch if it isn't any trouble. I will be getting another 2 UW disks, and can use them to test with. It would be nice to have a big stack of disks to test the speeds from multiple UW, but .... :) Would it be helpful to test with ccd? del ... > > > I wanted to clean up some of the code before the > > > 20MHz speed is enabled, but was held back by some > > > other (higher priority) tasks ... > > > > Is it enabled by default? Will regular wide drives > > like the Q Atlas work? I want to use it with a 2G > > seagate UW, and 2 4G Atlas drives. They all work now. > > The driver will only support 10MHz WIDE transfers, > i.e. 20MB/s. I could provide you with diffs that > make the driver negotiate 20MHz speed, but have not > even tested them myself ... I will use the stock 2.1.7R driver with my server, and set the seagate uw drive to 10mhz in the tekram setup. That should be ok for now. Yes, please send the diffs. I am going to use my other machine to stage the upgrade, and then do it on the server. I can run tests and bench- marks on the test machine while I am at it. > > I have half-way completed the cleanup of the code, > that deals with negotiation values. It will use > "abstract" variables, instead of "register bits". > > Regards, STefan Good, will you back port to the 2.1.7R stable? I am a *stable* kind of guy. :) Thanks, -- Craig Shaver (craig@progroup.com) (415)390-0654 Productivity Group POB 60458 Sunnyvale, CA 94088 From owner-freebsd-scsi Tue Feb 11 15:15:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA27731 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 15:15:01 -0800 (PST) Received: from iafnl.es.iaf.nl (uucp@iafnl.es.iaf.nl [195.108.17.20]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA27715; Tue, 11 Feb 1997 15:14:51 -0800 (PST) Received: by iafnl.es.iaf.nl with UUCP id AA26240 (5.67b/IDA-1.5); Wed, 12 Feb 1997 00:14:46 +0100 Received: (from wilko@localhost) by yedi.iaf.nl (8.7.5/8.6.12) id XAA02978; Tue, 11 Feb 1997 23:58:41 +0100 (MET) From: Wilko Bulte Message-Id: <199702112258.XAA02978@yedi.iaf.nl> Subject: Re: Problem with 3701B CD-ROM To: se@freebsd.org (Stefan Esser) Date: Tue, 11 Feb 1997 23:58:41 +0100 (MET) Cc: joerg_wunsch@uriah.heep.sax.de, robsch@robkaos.ruhr.de, freebsd-scsi@freebsd.org In-Reply-To: <19970211202335.YC36261@x14.mi.uni-koeln.de> from "Stefan Esser" at Feb 11, 97 08:23:35 pm X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Stefan Esser wrote... > On Feb 11, j@uriah.heep.sax.de (J Wunsch) wrote: > > As Robert Schien wrote: > > > > > /kernel: cd0 at scbus0 target 2 lun 0: COMMAND FAILED (9 0) @f20ba800. > > > /kernel: cd0 at scbus0 target 2 lun 0: extraneous data discarded. > > > > > > The errors disappear when the kernel option FAILSAFE is set. > > > > That would mean they don't grok tagged queuing. > > Like just about any CDROM drive on the market today :) Which is logical since TCQ only makes sense in devices that are not slow like molasses by design ;-) > I've never had a single problem with my Toshiba 3601, > and I think the error is caused by some application > sending an invalid request. All Toshiba's I have also work fine with every FreeBSD rev I threw at them. E.g. (ncr0:4:0): "TOSHIBA CD-ROM XM-5301TA 0925" type 5 removable SCSI 2 Wilko _ ____________________________________________________________________ | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda -------------------------------------------------------------------------- From owner-freebsd-scsi Tue Feb 11 20:24:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA21481 for freebsd-scsi-outgoing; Tue, 11 Feb 1997 20:24:38 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA21414; Tue, 11 Feb 1997 20:24:27 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by who.cdrom.com (8.7.5/8.6.11) with ESMTP id TAA04035 ; Tue, 11 Feb 1997 19:35:08 -0800 (PST) Received: from narnia (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.5/8.7.3) with ESMTP id UAA06394; Tue, 11 Feb 1997 20:34:36 -0700 (MST) Message-Id: <199702120334.UAA06394@narnia.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: Wilko Bulte cc: se@FreeBSD.ORG (Stefan Esser), joerg_wunsch@uriah.heep.sax.de, robsch@robkaos.ruhr.de, freebsd-scsi@FreeBSD.ORG Subject: Re: Problem with 3701B CD-ROM In-reply-to: Your message of "Tue, 11 Feb 1997 23:58:41 +0100." <199702112258.XAA02978@yedi.iaf.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 11 Feb 1997 20:34:36 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> > That would mean they don't grok tagged queuing. >> >> Like just about any CDROM drive on the market today :) > >Which is logical since TCQ only makes sense in devices that >are not slow like molasses by design ;-) This is far from the truth. Tagged queuing reduces the number and distance of seeks; a very winning proposition on a CDROM drive. >Wilko >_ ____________________________________________________________________ > | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands > |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda >-------------------------------------------------------------------------- > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Wed Feb 12 11:44:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA22291 for freebsd-scsi-outgoing; Wed, 12 Feb 1997 11:44:44 -0800 (PST) Received: from iafnl.es.iaf.nl (uucp@iafnl.es.iaf.nl [195.108.17.20]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA22223; Wed, 12 Feb 1997 11:44:00 -0800 (PST) Received: by iafnl.es.iaf.nl with UUCP id AA06051 (5.67b/IDA-1.5); Wed, 12 Feb 1997 20:43:59 +0100 Received: (from wilko@localhost) by yedi.iaf.nl (8.7.5/8.6.12) id UAA01229; Wed, 12 Feb 1997 20:16:54 +0100 (MET) From: Wilko Bulte Message-Id: <199702121916.UAA01229@yedi.iaf.nl> Subject: Re: Problem with 3701B CD-ROM To: gibbs@narnia.plutotech.com (Justin T. Gibbs) Date: Wed, 12 Feb 1997 20:16:53 +0100 (MET) Cc: se@FreeBSD.ORG, joerg_wunsch@uriah.heep.sax.de, robsch@robkaos.ruhr.de, freebsd-scsi@FreeBSD.ORG In-Reply-To: <199702120334.UAA06394@narnia.plutotech.com> from "Justin T. Gibbs" at Feb 11, 97 08:34:36 pm X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Justin T. Gibbs wrote... > >> > That would mean they don't grok tagged queuing. > >> > >> Like just about any CDROM drive on the market today :) > > > >Which is logical since TCQ only makes sense in devices that > >are not slow like molasses by design ;-) > > This is far from the truth. Tagged queuing reduces the number and > distance of seeks; a very winning proposition on a CDROM drive. Hm. IMHO CDROMs are speedwise only suitable to install software from (compared to about anything in magnetic disks that is). Beating them with random I/O is not my idea of fun in any case. Wilko _ ____________________________________________________________________ | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda -------------------------------------------------------------------------- From owner-freebsd-scsi Wed Feb 12 11:53:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA22690 for freebsd-scsi-outgoing; Wed, 12 Feb 1997 11:53:24 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA22682; Wed, 12 Feb 1997 11:53:08 -0800 (PST) Received: from narnia (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.5/8.7.3) with ESMTP id MAA08056; Wed, 12 Feb 1997 12:53:03 -0700 (MST) Message-Id: <199702121953.MAA08056@narnia.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: Wilko Bulte cc: se@FreeBSD.ORG, joerg_wunsch@uriah.heep.sax.de, robsch@robkaos.ruhr.de, freebsd-scsi@FreeBSD.ORG Subject: Re: Problem with 3701B CD-ROM In-reply-to: Your message of "Wed, 12 Feb 1997 20:16:53 +0100." <199702121916.UAA01229@yedi.iaf.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 12 Feb 1997 12:53:03 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Hm. IMHO CDROMs are speedwise only suitable to install software from >(compared to about anything in magnetic disks that is). Beating them >with random I/O is not my idea of fun in any case. Sure, but in a multi-user system, its fairly easy to get pathological seek behavior on a CDROM drive and tagged queueing would greatly increase throughput in this case. >Wilko >_ ____________________________________________________________________ > | / o / / _ Bulte email: wilko@yedi.iaf.nl - Arnhem, The Netherlands > |/|/ / / /( (_) Do, or do not. There is no 'try' - Yoda >-------------------------------------------------------------------------- > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Wed Feb 12 18:27:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA12511 for freebsd-scsi-outgoing; Wed, 12 Feb 1997 18:27:10 -0800 (PST) Received: from caliban.dihelix.com (caliban.mrtc.org [199.4.33.251]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA12502 for ; Wed, 12 Feb 1997 18:27:01 -0800 (PST) Received: (from langfod@localhost) by caliban.dihelix.com (8.8.4/8.8.3) id QAA21207; Wed, 12 Feb 1997 16:31:56 -1000 (HST) Message-Id: <199702130231.QAA21207@caliban.dihelix.com> Subject: Re: how to use tape changer ch0 In-Reply-To: <199702100136.TAA16088@shell.futuresouth.com> from Tim Tsai at "Feb 9, 97 07:36:05 pm" To: tim@futuresouth.com (Tim Tsai) Date: Wed, 12 Feb 1997 16:31:56 -1000 (HST) Cc: langfod@dihelix.com, scsi@freebsd.org From: "David Langford" X-blank-line: This space intentionaly left blank. X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Tim Tsai >> I have a Seagate/Conner/Archive 4586NP that I would like to >> get working with Amanda. > >> Shows as ether st0 or ch0 but never both. It works great as st0 >> if I feel like manully flipping through the tapes. I would like >> to get it working for automated backups but I cant seem to figure >> out how one would go about using the ch0 driver. > > I have the same setup. Try the following: > > chio move slot 0 drive 0 (to load tape) > chio move drive 0 slot 0 (to eject) > > I don't have it working with Amanda yet but I see /dev/ch* and >/dev/nrst* /dev/rst*, etc. > > I am running 2.1.6.1 BTW. Umm where do I find chio ? Running 2.2-GAMMA here. -David Langford langfod@dihelix.com From owner-freebsd-scsi Wed Feb 12 18:46:17 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA13378 for freebsd-scsi-outgoing; Wed, 12 Feb 1997 18:46:17 -0800 (PST) Received: from mail.futuresouth.com (mail.futuresouth.com [207.141.254.21]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA13373 for ; Wed, 12 Feb 1997 18:46:09 -0800 (PST) Received: from shell.futuresouth.com (shell.futuresouth.com [207.141.254.20]) by mail.futuresouth.com (8.8.5/8.8.5) with ESMTP id UAA15572; Wed, 12 Feb 1997 20:45:30 -0600 (CST) From: Tim Tsai Received: (from tim@localhost) by shell.futuresouth.com (8.8.3/8.8.3) id UAA21809; Wed, 12 Feb 1997 20:45:24 -0600 (CST) Message-Id: <199702130245.UAA21809@shell.futuresouth.com> Subject: Re: how to use tape changer ch0 To: langfod@dihelix.com (David Langford) Date: Wed, 12 Feb 1997 20:45:24 -0600 (CST) Cc: tim@mail.futuresouth.com, langfod@dihelix.com, scsi@freebsd.org In-Reply-To: <199702130231.QAA21207@caliban.dihelix.com> from David Langford at "Feb 12, 97 04:31:56 pm" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Umm where do I find chio ? Running 2.2-GAMMA here. http://www.freebsd.org/cgi/query-pr.cgi?pr=1201 It's not integrated with any system yet that I know of. Tim From owner-freebsd-scsi Thu Feb 13 04:55:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA08814 for freebsd-scsi-outgoing; Thu, 13 Feb 1997 04:55:33 -0800 (PST) Received: from hda.hda.com (ip13-max1-fitch.ziplink.net [199.232.245.13]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id EAA08809 for ; Thu, 13 Feb 1997 04:55:29 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.6.12/8.6.12) id HAA14059 for scsi@freebsd.org; Thu, 13 Feb 1997 07:50:16 -0500 From: Peter Dufault Message-Id: <199702131250.HAA14059@hda.hda.com> Subject: scsinew3.tgz To: scsi@freebsd.org Date: Thu, 13 Feb 1997 07:50:15 -0500 (EST) X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The test version of a new scsi(8) is now scsinew3.tgz. On freefall: ftp://freefall.freebsd.org/pub/dufault/scsinew3.tgz -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 From owner-freebsd-scsi Thu Feb 13 13:09:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA08559 for freebsd-scsi-outgoing; Thu, 13 Feb 1997 13:09:56 -0800 (PST) Received: from Kitten.mcs.com (Kitten.mcs.com [192.160.127.90]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA08552 for ; Thu, 13 Feb 1997 13:09:49 -0800 (PST) Received: from Mars.mcs.net (ljo@Mars.mcs.com [192.160.127.85]) by Kitten.mcs.com (8.8.5/8.8.2-biteme) with ESMTP id PAA08937; Thu, 13 Feb 1997 15:09:47 -0600 (CST) Received: (from ljo@localhost) by Mars.mcs.net (8.8.5/8.8.2) id PAA02682; Thu, 13 Feb 1997 15:09:46 -0600 (CST) From: Lars Jonas Olsson Message-Id: <199702132109.PAA02682@Mars.mcs.net> Subject: WD7000-FASST2 hangs in scsicmd To: scsi@freebsd.org Date: Thu, 13 Feb 1997 15:09:46 -0600 (CST) Cc: ljo@mcs.net X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm testing a NEC 8 speed CDROM with a Western Digital WD7000-FASST2 SCSI controller under 3.0 current from a few weeks ago. The system boots OK and probes controller and device OK. When I do "mount -t cd9660 /dev/cd0c /cdrom" the commands sleeps on scsicmd in scsi_base.c. When compiling kernel with SCSI debugging I get the following info. Any idea what's wrong? I guess there isn't that many users with WD7000s so perhaps the controller isn't supported by FreeBSD anymore? Jonas Feb 13 14:43:13 amcell1 /kernel: wds0 at 0x350-0x357 irq 15 drq 6 on isa Feb 13 14:43:14 amcell1 /kernel: wds0: firmware version 8.00 Feb 13 14:43:14 amcell1 /kernel: wds0: using 73728 bytes for dma buffer Feb 13 14:43:14 amcell1 /kernel: wds0 waiting for scsi devices to settle Feb 13 14:43:14 amcell1 /kernel: scbus0 at wds0 bus 0 Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): scsi_cmd Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): get_xs Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): returning Feb 13 14:43:14 amcell1 /kernel: xs(0xf10c2a00): flg(0x63)sc_link(0xf10c2a80)retr(0x2)timo(0x186a0)cmd(0xf10c2a58)len(0x6)data(0x0)len(0x0)res(0x0)err(0x0)bp(0x0)scbus0 target 3 lun 0: command: 0,0,0,0,0,0-[0 bytes] Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): back in cmd() Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): sc_err1,err = 0x1 Feb 13 14:43:14 amcell1 /kernel: code1 valid0 seg0 key0 ili0 eom0 fmark0 Feb 13 14:43:14 amcell1 /kernel: info: 0 0 0 0 followed by 0 extra bytes Feb 13 14:43:14 amcell1 /kernel: extra: Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): scsi_interpret_sense (no bp) returned 5 Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): free_xs Feb 13 14:43:14 amcell1 /kernel: probe0(wds0:3:0): scsi_cmd Feb 13 14:43:15 amcell1 /kernel: probe0(wds0:3:0): get_xs Feb 13 14:43:15 amcell1 /kernel: probe0(wds0:3:0): returning Feb 13 14:43:15 amcell1 /kernel: xs(0xf10c2a00): flg(0x423)sc_link(0xf10c2a80)retr(0x2)timo(0x186a0)cmd(0xf10c2a58)len(0x6)data(0xf10c2aac)len(0x2c)res(0x0)err(0x0)bp(0x0)scbus0 target 3 lun 0: command: 12,0,0,0,2c,0-[44 bytes] Feb 13 14:43:15 amcell1 /kernel: ------------------------------ Feb 13 14:43:15 amcell1 /kernel: 000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Feb 13 14:43:15 amcell1 /kernel: 016: 00 00 0 Feb 13 14:43:15 amcell1 /kernel: 0 00 00 00 00 00 00 00 00 00 00 00 00 00 Feb 13 14:43:15 amcell1 /kernel: 032: 00 00 00 00 00 00 00 00 00 00 00 00 Feb 13 14:43:15 amcell1 /kernel: ------------------------------ Feb 13 14:43:15 amcell1 /kernel: probe0(wds0:3:0): back in cmd() Feb 13 14:43:15 amcell1 /kernel: probe0(wds0:3:0): sc_err1,err = 0x0 Feb 13 14:43:15 amcell1 /kernel: probe0(wds0:3:0): free_xs Feb 13 14:43:15 amcell1 /kernel: scbus0 target 3 lun 0: type 5 removable SCSI 2 Feb 13 14:43:15 amcell1 /kernel: cd0(wds0:3:0): cd0attach: cd0 at scbus0 target 3 lun 0 Feb 13 14:43:15 amcell1 /kernel: cd0: CD-ROM cd0(wds0:3:0): scsi_cmd Feb 13 14:43:15 amcell1 /kernel: cd0(wds0:3:0): get_xs Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): returning Feb 13 14:43:16 amcell1 /kernel: xs(0xf10c2a00): flg(0x463)sc_link(0xf10c2a80)retr(0x1)timo(0x4e20)cmd(0xf10c2a58)len(0xa)data(0xf3325000)len(0x8)res(0x0)err(0x0)bp(0x0)cd0 at scbus0 target 3 lun 0: command: 25,0,0,0,0,0,0,0,0,0-[8 bytes] Feb 13 14:43:16 amcell1 /kernel: ------------------------------ Feb 13 14:43:16 amcell1 /kernel: 000: 00 00 00 00 00 00 00 00 Feb 13 14:43:16 amcell1 /kernel: ------------------------------ Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): back in cmd() Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): sc_err1,err = 0x0 Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): free_xs Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): calling private start() Feb 13 14:43:16 amcell1 /kernel: cd0(wds0:3:0): cdstart0 cd0(wds0:3:0): cd0: 325252 2048 byte blocks Feb 13 14:43:16 amcell1 /kernel: cd present [325252 x 2048 byte records] The following is after mount -t cd9660 /dev/cd0c /cdrom Feb 13 14:43:45 amcell1 /kernel: cd0(wds0:3:0): cd_open: dev=0x602 (unit 0,partition 2) Feb 13 14:43:45 amcell1 /kernel: cd0(wds0:3:0): scsi_cmd Feb 13 14:43:45 amcell1 /kernel: cd0(wds0:3:0): get_xs Feb 13 14:43:46 amcell1 /kernel: cd0(wds0:3:0): returning Feb 13 14:43:46 amcell1 /kernel: xs(0xf10c2a00): flg(0x60)sc_link(0xf10c2a80)retr(0x2)timo(0x186a0)cmd(0xf10c2a58)len(0x6)data(0x0)len(0x0)res(0x0)err(0x0)bp(0x0)cd0 at scbus0 target 3 lun 0: command: 0,0,0,0,0,0-[0 bytes] From owner-freebsd-scsi Thu Feb 13 22:22:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA10269 for freebsd-scsi-outgoing; Thu, 13 Feb 1997 22:22:53 -0800 (PST) Received: from sendero.i-connect.net ([206.190.144.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA10264; Thu, 13 Feb 1997 22:22:45 -0800 (PST) Received: (from shimon@localhost) by sendero.i-connect.net (8.8.5/8.8.4) id XAA14590; Thu, 13 Feb 1997 23:21:05 -0800 (PST) Message-ID: X-Mailer: XFMail 1.1-alpha [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-8 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Date: Thu, 13 Feb 1997 22:44:46 -0800 (PST) Organization: iConnect Corp. From: Simon Shapiro To: freebsd-hackers@freebsd.org, freebsd-scsi@freebsd.org Subject: Software Interrupts Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have dug into the software interrupts a bit and would like to confirm the best approach. Problem Description: I would like to trigger an interrupt, from within a device driver. The reason for doing so (may not be so good :-): I would like to de-couple queueing SCSI commands from the calling layer on one side and results analysis and error handling from the interrupt handler on the other side. Let me be more specific: when the SCSI code calls my driver's xxx_scsi_cmd entry point, I take the command, tweak it to the hardware vendor's content, put the request in a queue. At this point i would like to evoke an interrupt and return QUEUED_SUCCESSFULLY. The interrupt routine then will go to the queue and negotiate with the device all these wonderful inb, outb and such. The calling process is not bound by hardware delays. When a hardware interrupt, from the HBA, occurs, I want to be able to, in the interrupt service routine to just capture the hardware state (it takes only two inb's (can be reduced to one) one indirect dereferencing and a 24 bytes bcopy) into the request's CCB, move the request to the ``complete'' queue, schedule a software interrupt and schedule another software interrupt. This one will process however many requests there are in the ``completed'' queue. In this fasion, the interrupt routine will remain very short, consume very little time. The device we are interfacing to can give us SCSI command completion in less than 1ms. Much less in fact. Now, if you nice people know of a better way (than software interrupts) to do this, i will be happy to hear about it. Just as an intelectual excercise, please help me understand the software interrupts business. I have noticed that the Inet code that uses them, boils down to calling setbits (an inline defined in i386/include/cpufunc.h). Setbits takes the address of an unsigned and a u_int called bits. Different purposes in the kernel call setbits with different arguments. The argument is always the address of either ipending or of idelayed. These appear to be global bitmaps describing which interrupts are pending or delayed. What I cannot find is how to register a software interrupt. They appear, as if by magic. So, after all this, i am asking for a way to schedule two software interrupts. I really do not care what arguments will be passed to these functions, what value they return, or what. Just so they get invoked at some known priority, without fighting for CPU with the splbio guys. At the context of a true SMP machine, this scheme makes even more sense. But it may improve our responsiveness quite a bit, at some cost of increased overhead. the old game of latency vs. throughput. Thanx a million. Simon From owner-freebsd-scsi Fri Feb 14 16:37:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA15608 for freebsd-scsi-outgoing; Fri, 14 Feb 1997 16:37:50 -0800 (PST) Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA15600; Fri, 14 Feb 1997 16:37:47 -0800 (PST) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.8.5/8.8.4) with SMTP id QAA16425; Fri, 14 Feb 1997 16:29:25 -0800 (PST) Message-ID: <330502EF.1CFBAE39@whistle.com> Date: Fri, 14 Feb 1997 16:27:28 -0800 From: Julian Elischer Organization: Whistle Communications X-Mailer: Mozilla 3.0Gold (X11; I; FreeBSD 2.2-CURRENT i386) MIME-Version: 1.0 To: Simon Shapiro CC: freebsd-hackers@freebsd.org, freebsd-scsi@freebsd.org Subject: Re: Software Interrupts References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Simon Shapiro wrote: > > I have dug into the software interrupts a bit and would like to confirm the > best approach. > > Problem Description: > > I would like to trigger an interrupt, from within a device driver. The > reason for doing so (may not be so good :-): > I would like to de-couple queueing SCSI commands from the calling layer on > one side and results analysis and error handling from the interrupt handler > on the other side. There is already support for this.. this is what the filesystem code does.. check out the files in kern/vfs_bio.c especially look at biodone() if you fill ou the b_iodone entry in the buf struct then it will call whatever routine you want on completion.. check the output() strategy() code for sd.c and the adapter to understand how your routine may return after having queueud the data. > > Let me be more specific: when the SCSI code calls my driver's xxx_scsi_cmd > entry point, I take the command, tweak it to the hardware vendor's content, > put the request in a queue. At this point i would like to evoke an > interrupt and return QUEUED_SUCCESSFULLY. The interrupt routine then will > go to the queue and negotiate with the device all these wonderful inb, > outb and such. The calling process is not bound by hardware delays. > > When a hardware interrupt, from the HBA, occurs, I want to be able to, > in the interrupt service routine to just capture the hardware state (it > takes only two inb's (can be reduced to one) one indirect dereferencing > and a 24 bytes bcopy) into the request's CCB, move the request to the > ``complete'' queue, schedule a software interrupt and schedule another > software interrupt. This one will process however many requests there are > in the ``completed'' queue. In this fasion, the interrupt routine will > remain very short, consume very little time. The device we are interfacing > to can give us SCSI command completion in less than 1ms. Much less in > fact. Usually, what you do is: get the old command status.. load teh next command process the old status.. > > Now, if you nice people know of a better way (than software interrupts) > to do this, i will be happy to hear about it. > > Just as an intelectual excercise, please help me understand the software > interrupts business. > > I have noticed that the Inet code that uses them, boils down to calling > setbits (an inline defined in i386/include/cpufunc.h). Setbits takes > the address of an unsigned and a u_int called bits. > > Different purposes in the kernel call setbits with different arguments. > The argument is always the address of either ipending or of idelayed. > > These appear to be global bitmaps describing which interrupts are pending > or delayed. > > What I cannot find is how to register a software interrupt. They appear, > as if by magic. look for schednetisr() software interrupts are run if there is anything scheduked for them when the SPL is set back to 0. > > So, after all this, i am asking for a way to schedule two software > interrupts. I really do not care what arguments will be passed to these > functions, what value they return, or what. Just so they get invoked > at some known priority, without fighting for CPU with the splbio guys. unless you duplicate teh splnet cade, you will end up running at the networking spl.. which may or may not be what you wanted.. > > At the context of a true SMP machine, this scheme makes even more sense. > But it may improve our responsiveness quite a bit, at some cost of > increased overhead. the old game of latency vs. throughput. > > Thanx a million. > > Simon From owner-freebsd-scsi Sat Feb 15 17:58:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA05376 for freebsd-scsi-outgoing; Sat, 15 Feb 1997 17:58:32 -0800 (PST) Received: from sendero.i-connect.net ([206.190.144.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA05353; Sat, 15 Feb 1997 17:58:20 -0800 (PST) Received: (from shimon@localhost) by sendero.i-connect.net (8.8.5/8.8.4) id SAA22265; Sat, 15 Feb 1997 18:57:43 -0800 (PST) Message-ID: X-Mailer: XFMail 1.1-alpha [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-8 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Date: Sat, 15 Feb 1997 18:35:49 -0800 (PST) Organization: iConnect Corp. From: Simon Shapiro To: freebsd-scsi@freebsd.org, FreeBSD-hackers@freebsd.org Subject: Device Not Ready, 2nd Posting Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I added the worm drive to both 2.2-BETA and 3.0-before_the_explosion with the result that it tells me ``Device not configured'' from wormcontrol. It did ``work'' for a while, then we rebooted and now it is very persistant. The worm support was added by adding: device worm0 at scbus? to the config file. Then we also added: disk worm6 at scbus0 target 6 unit 0 Still the same thing. nm on the kernel reveals that the code is there: ... f0197e88 t _worm_rezero_unit f0197898 t _worm_size f0197a7c t _worm_strategy f01e8a0c d _worm_switch f01978e4 t _wormattach f0197814 t _wormclose f01e8ae0 d _wormdev_sys_init f01977c8 T _worminit f01977f4 t _wormioctl f0197830 t _wormminphys f01977d8 t _wormopen f0197914 t _wormstart f0197844 t _wormstrategy f01977b8 t _wormunit f01977b0 F worm.o So what gives? Simon