From owner-freebsd-scsi@FreeBSD.ORG Sun Jun 6 09:30:23 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 69E8B16A4CF for ; Sun, 6 Jun 2004 09:30:23 -0700 (PDT) Received: from sys.heron.com.pl (sys.heron.com.pl [195.117.24.6]) by mx1.FreeBSD.org (Postfix) with ESMTP id C0AE543D2F for ; Sun, 6 Jun 2004 09:30:22 -0700 (PDT) (envelope-from robert@heron.pl) Received: from szef ([195.205.27.42] helo=robert) by sys.heron.com.pl with asmtp (Exim 4.23) id 1BX0Ym-0003TU-0p for freebsd-scsi@freebsd.org; Sun, 06 Jun 2004 18:31:48 +0200 Message-ID: <07e901c44be2$d4c68a90$2a1bcdc3@robert> From: "Robert Heron" To: Date: Sun, 6 Jun 2004 18:25:05 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Subject: Problem with AIC7902 and FreeBSD 4.7, 4.8, 4.9 and 4.10 X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 06 Jun 2004 16:30:23 -0000 Hi, I have an Intel mainboard SE7501HG2 with integrated AIC-7902W controller. When I run FreeBSD 4.7R on this motherboard I get: -------------------------------------------------------- ahd0: port 0x2c00-0x2cff,0x2800-0x28ff mem 0xfe9e0000-0xfe9e1fff irq 9 at device 4.0 o n pci3 ahd0: WARNING - Failed chip reset! Trying to initialize anyway. ahd0: Primary Auto-Term Sensing failed! Using Defaults. ahd0: Secondary Auto-Term Sensing failed! Using Defaults. ahd0: Unable to set termination settings! ahd0: No SCB space found device_probe_and_attach: ahd0 attach returned 12 ahd0: port 0x2400-0x24ff,0x2000-0x20ff mem 0xfe9f0000-0xfe9f1fff irq 9 at device 4.1 o n pci3 ahd0: WARNING - Failed chip reset! Trying to initialize anyway. ahd0: Primary Auto-Term Sensing failed! Using Defaults. ahd0: Secondary Auto-Term Sensing failed! Using Defaults. ahd0: Unable to set termination settings! ahd0: No SCB space found device_probe_and_attach: ahd0 attach returned 12 ----------------------------------------------------- and after booting, FreeBSD 4.7 does not see any SCSI disks. When I boot FreeBSD 4.8 / 4.9 / 4.10 from installation CD-ROM on the same motherboard, the ahd0 and ahd1 devices are detected correctly but the system hangs up during further device detection. The last message I get just before hanging up is: ---------- pci0: (vendor=0x8086, dev=0x2483) at 31.3 irq 0 ---------- When I disable AIC-7902 controller in the BIOS, then FreeBSD starts normally and does not hang up. I don't know how to fix this problem. I need to run FreeBSD 4.7 on this motherboard and probably FreeBSD 4.10 in the future, but the problem makes the motherboard's AIC-7902 controller unusable with FreeBSD :( Maybe some patch to the aic7xxx driver is available for FreeBSD 4.7 and the laters ? Best regards, Robert Heron From owner-freebsd-scsi@FreeBSD.ORG Sun Jun 6 11:03:15 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0129016A4CE for ; Sun, 6 Jun 2004 11:03:15 -0700 (PDT) Received: from mail.sandvine.com (sandvine.com [199.243.201.138]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8542243D1F for ; Sun, 6 Jun 2004 11:03:14 -0700 (PDT) (envelope-from don@sandvine.com) Received: by mail.sandvine.com with Internet Mail Service (5.5.2657.72) id ; Sun, 6 Jun 2004 14:03:10 -0400 Message-ID: From: Don Bowman To: 'Robert Heron' , freebsd-scsi@freebsd.org Date: Sun, 6 Jun 2004 14:03:00 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain; charset="iso-8859-2" Subject: RE: Problem with AIC7902 and FreeBSD 4.7, 4.8, 4.9 and 4.10 X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 06 Jun 2004 18:03:15 -0000 From: Robert Heron [mailto:robert@heron.pl] > Hi, > > I have an Intel mainboard SE7501HG2 with integrated AIC-7902W > controller. > When I run FreeBSD 4.7R on this motherboard I get: ... > > > When I boot FreeBSD 4.8 / 4.9 / 4.10 from installation CD-ROM > on the same > motherboard, the ahd0 and ahd1 devices are detected correctly > but the system > hangs up during further device detection. The last message I > get just before > hanging up is: > > ---------- > pci0: (vendor=0x8086, dev=0x2483) at 31.3 irq 0 > ---------- > Sounds like an IRQ assignment problem in the BIOS? The unknown card above is an intel device, its probably a normal message for a device with no driver. you might try boot -v to see where you are hanging on the 4.10. There have been some ahd driver updates since 4.7 that are reasonably important, you might consider merging them to the 4.7 branch if you need them, or running 4.10. From owner-freebsd-scsi@FreeBSD.ORG Mon Jun 7 08:15:56 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1EBFC16A4CE for ; Mon, 7 Jun 2004 08:15:56 +0000 (GMT) Received: from ns.env.cz (ns.env.cz [193.179.186.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id 57B7243D31 for ; Mon, 7 Jun 2004 08:15:54 +0000 (GMT) (envelope-from Jan_Rybensky@env.cz) Received: by ns.env.cz (Postfix, from userid 62001) id 7A1388270F; Mon, 7 Jun 2004 10:15:26 +0200 (CEST) Received: from fw.env.cz (fw.env.cz [193.179.186.2]) by ns.env.cz (Postfix) with ESMTP id 3937B82492 for ; Mon, 7 Jun 2004 10:15:26 +0200 (CEST) Received: from dns1.env.cz (dns1.env.cz [10.1.2.2]) by fw.env.cz (Postfix) with ESMTP id 2DB5C2F41D for ; Mon, 7 Jun 2004 10:15:26 +0200 (CEST) Received: from fireball.env.cz (fireball.env.cz [10.1.1.145]) by dns1.env.cz (8.12.11/8.12.11) with ESMTP id i578FNSd011785 for ; Mon, 7 Jun 2004 10:15:24 +0200 (CEST) (envelope-from Jan_Rybensky@env.cz) Received: from www.env.cz (charon.env.cz [10.0.1.9]) by fireball.env.cz (Postfix) with ESMTP id CC8A77238A for ; Mon, 7 Jun 2004 10:15:23 +0200 (CEST) MIME-Version: 1.0 Sensitivity: To: freebsd-scsi@freebsd.org X-Mailer: Lotus Notes Release 5.0.10 March 22, 2002 Message-ID: From: Jan_Rybensky@env.cz Date: Mon, 7 Jun 2004 10:26:51 +0200 X-MIMETrack: Serialize by Router on Undead/ENV/CZ(Release 6.5.1|January 21, 2004) at 07.06.2004 10:15:22, Serialize complete at 07.06.2004 10:15:22 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: Re: mpt driver patch X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 07 Jun 2004 08:15:56 -0000 With this patch, transfers from/to mirrored volumes runs at full speed,=20 but re-mirroring on background of FreeBSD takes days (although transfer=20 rates from/to mirrored volume during re-mirroring are OK). da0 at mpt0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: 320.000MB/s transfers (160.000MHz, offset 127, 16bit), Tagged=20 Queueing Enabled da0: 70001MB (143364061 512 byte sectors: 255H 63S/T 8924C) -bash-2.05b$ ls -la bigfile -rw-r--r-- 1 jry jry 252157013 May 14 10:12 bigfile -bash-2.05b$ ( cp bigfile bigfile.1 & ); iostat -d da0 1 da0 KB/t tps MB/s 0.00 0 0.00 63.90 923 57.62 63.93 958 59.84 63.95 962 60.10 63.95 956 59.72 39.17 29 1.10 0.00 0 0.00 Is possible to obtain sources of the solaris driver? And where? JRY=20 Odesl=E1no k=FDm: owner-freebsd-scsi@freebsd.org Komu: freebsd-scsi@freebsd.org Kopie:=20 P=F8edm=ECt: Re: mpt driver patch At 02:07 PM 06/04/2004, Jan=5FRybensky@env.cz wrote: >Hi All, > >as well as the others I've encoutered serious performace problems with IM >volumes on LSI 1030 controller. It's because the mpt driver is unable to >detect hidden disks of the IM volumes and initialize them. That disks >operates at default SCSI speed and slows the whole volume. >This patch is far from perfect (not performs full DV), but it solves the >performance issues. It seems to be stable, but it's tested only on IBM >version of controller (on-board controllers of xSeries 335, 345). > >There is other issue I'm unable to solve: although all discs operating at >proper speeds, the resync of mirrored disks is still very slow. The >controller probably needs additional initializations, but I'm unable to >trace them in linux driver. Can anybody help? Any suggestions are >welcomed. > >Here is diff against 4.9-REL mpt sources (it works well on 4.10-REL too) >and a tarball of the whole mpt directory from /sys/dev with patched=20 driver >sources is here: http://mpt-freebsd.env.cz. > >Many thanks for any feedback. > > Jan Rybensky I have seen this exact same behavior under 5.2.1 ...I have the LSI 20320R card installed. Using normal setup - this card works well and the speeds are as expected: da0 at mpt0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-3 device da0: 320.000MB/s transfers (160.000MHz, offset 63, 16bit), Tagged Queueing Enabled da0: 35003MB (71687372 512 byte sectors: 255H 63S/T 4462C) however, like you pointed out - the mirroring is atrocious. It took over 5hrs to fully sync a mirror (RAID-1). Then when I rebooted, the=20 performance was less than 2MB/sec on any transfers to/from the machine. As long as the mirror is not active, the card+drives seem to work just=20 fine. FWIW, setting up a mirror on this same card+drives under solaris x86,=20 works perfect and resync's in just over 2hrs. -- J.D. Bronson Aurora Health Care // Information Services // Milwaukee, WI USA Office: 414.978.8282 // Email: jd@aurora.org // Pager: 414.314.8282 =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F freebsd-scsi@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-scsi To unsubscribe, send any mail to "freebsd-scsi-unsubscribe@freebsd.org" From owner-freebsd-scsi@FreeBSD.ORG Mon Jun 7 11:20:52 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3463216A4CE for ; Mon, 7 Jun 2004 11:20:52 +0000 (GMT) Received: from shadow.wixb.com (shadow.wixb.com [65.43.82.173]) by mx1.FreeBSD.org (Postfix) with ESMTP id B8E1F43D46 for ; Mon, 7 Jun 2004 11:20:51 +0000 (GMT) (envelope-from jbronson@wixb.com) Received: from dakota.wixb.com (shadow.wixb.com [10.43.82.173]) i57BKoJq000600 for ; Mon, 7 Jun 2004 06:20:50 -0500 (CDT) Organization: Aurora Health Care, Milwaukee WI USA Message-Id: <6.1.1.1.2.20040607061955.00a8c310@localhost> Date: Mon, 07 Jun 2004 06:20:57 -0500 To: freebsd-scsi@freebsd.org From: "J.D. Bronson" In-Reply-To: References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Miltered: at shadow.wixb.com with ID 40C44F92.000 by j-chkmail Subject: Re: mpt driver patch X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 07 Jun 2004 11:20:52 -0000 At 03:26 AM 06/07/2004, Jan_Rybensky@env.cz wrote: >With this patch, transfers from/to mirrored volumes runs at full speed, >but re-mirroring on background of FreeBSD takes days (although transfer >rates from/to mirrored volume during re-mirroring are OK). > >da0 at mpt0 bus 0 target 0 lun 0 >da0: Fixed Direct Access SCSI-2 device >da0: 320.000MB/s transfers (160.000MHz, offset 127, 16bit), Tagged >Queueing Enabled >da0: 70001MB (143364061 512 byte sectors: 255H 63S/T 8924C) > >-bash-2.05b$ ls -la bigfile >-rw-r--r-- 1 jry jry 252157013 May 14 10:12 bigfile >-bash-2.05b$ ( cp bigfile bigfile.1 & ); iostat -d da0 1 >da0 >KB/t tps MB/s >0.00 0 0.00 >63.90 923 57.62 >63.93 958 59.84 >63.95 962 60.10 >63.95 956 59.72 >39.17 29 1.10 >0.00 0 0.00 > >Is possible to obtain sources of the solaris driver? And where? > > JRY LSI provided the driver to sun. So I would doubt that there will be any way to gain sources. In addition, it seems that Sun and LSI had a falling out and support will be missing - as it almost already is. This LSI card seems most happy only in Linux or Windows. -- J.D. Bronson Aurora Health Care // Information Services // Milwaukee, WI USA Office: 414.978.8282 // Email: jd@aurora.org // Pager: 414.314.8282 From owner-freebsd-scsi@FreeBSD.ORG Wed Jun 9 21:19:45 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2BAB416A4D1 for ; Wed, 9 Jun 2004 21:19:45 +0000 (GMT) Received: from zaphod.nitro.dk (port324.ds1-khk.adsl.cybercity.dk [212.242.113.79]) by mx1.FreeBSD.org (Postfix) with ESMTP id 72E4043D2F for ; Wed, 9 Jun 2004 21:19:44 +0000 (GMT) (envelope-from simon@zaphod.nitro.dk) Received: by zaphod.nitro.dk (Postfix, from userid 3000) id D61C8119A4; Wed, 9 Jun 2004 23:19:40 +0200 (CEST) Date: Wed, 9 Jun 2004 23:19:40 +0200 From: "Simon L. Nielsen" To: Guido Kollerie Message-ID: <20040609211940.GA829@zaphod.nitro.dk> References: <20040604085316.GA77766@kollerie.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="LZvS9be/3tNcYl/X" Content-Disposition: inline In-Reply-To: <20040604085316.GA77766@kollerie.com> User-Agent: Mutt/1.5.6i cc: freebsd-scsi@freebsd.org Subject: Re: raidutil: Engine connect failed: Open X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 09 Jun 2004 21:19:45 -0000 --LZvS9be/3tNcYl/X Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 2004.06.04 10:53:16 +0200, Guido Kollerie wrote: > I have an ADAPTEC 2400A installed in my server. Every night I run > the following job from cron: >=20 > /usr/local/dpt/raidutil -L all >=20 > and have its output emailed to me. This seemed to worked fine for > quite a while, however last night I received the following output > from this job: >=20 > Engine connect failed: Open >=20 > What's causing this and how can I resolve it? Check if you are running out of SysV shared memory... I have to stop PostgreSQL to run raidutil, since I haven't gotten around to bumping SysV memory in my kernel. --=20 Simon L. Nielsen FreeBSD Documentation Team --LZvS9be/3tNcYl/X Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAx37sh9pcDSc1mlERAjzLAJ9Jkef/MStK9cKLwdRHX/jFKr61pACfdwgN bqxGTa7tbnLypXeSkmsIDMM= =7neT -----END PGP SIGNATURE----- --LZvS9be/3tNcYl/X-- From owner-freebsd-scsi@FreeBSD.ORG Thu Jun 10 06:52:20 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id ED3AE16A4CE; Thu, 10 Jun 2004 06:52:20 +0000 (GMT) Received: from kollerie.com (148-198.bbned.dsl.internl.net [217.149.198.148]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9FA9A43D2F; Thu, 10 Jun 2004 06:52:20 +0000 (GMT) (envelope-from gkoller@kollerie.com) Received: by kollerie.com (Postfix, from userid 1001) id E75942F8FE; Thu, 10 Jun 2004 08:51:33 +0200 (CEST) Date: Thu, 10 Jun 2004 08:51:33 +0200 From: Guido Kollerie To: "Simon L. Nielsen" Message-ID: <20040610065133.GA14542@kollerie.com> References: <20040604085316.GA77766@kollerie.com> <20040609211940.GA829@zaphod.nitro.dk> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="IS0zKkzwUGydFO0o" Content-Disposition: inline In-Reply-To: <20040609211940.GA829@zaphod.nitro.dk> User-Agent: Mutt/1.5.6i cc: freebsd-scsi@freebsd.org Subject: Re: raidutil: Engine connect failed: Open X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Jun 2004 06:52:21 -0000 --IS0zKkzwUGydFO0o Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jun 09, 2004 at 11:19:40PM +0200, Simon L. Nielsen wrote: > On 2004.06.04 10:53:16 +0200, Guido Kollerie wrote: Simon, > > I have an ADAPTEC 2400A installed in my server. Every night I run > > the following job from cron: > >=20 > > /usr/local/dpt/raidutil -L all > >=20 > > and have its output emailed to me. This seemed to worked fine for > > quite a while, however last night I received the following output > > from this job: > >=20 > > Engine connect failed: Open > >=20 > > What's causing this and how can I resolve it? >=20 > Check if you are running out of SysV shared memory... I have to stop > PostgreSQL to run raidutil, since I haven't gotten around to bumping > SysV memory in my kernel. That did the trick. I happened to have installed PostgreSQL recently but never made the link with raidutil failing. Stopping PostgreSQL allows raidutil to run without any problems! Thank you very much! --=20 Guido --IS0zKkzwUGydFO0o Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iEYEARECAAYFAkDIBPUACgkQC7vo05+GQSyRlACfQO30qHqVisN3cdhGKfmQ0D/H OJkAn1+5qORcqXctGm8X0G6g/BWqXIP7 =M2Kv -----END PGP SIGNATURE----- --IS0zKkzwUGydFO0o-- From owner-freebsd-scsi@FreeBSD.ORG Thu Jun 10 08:44:49 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1540216A4CE for ; Thu, 10 Jun 2004 08:44:49 +0000 (GMT) Received: from pan.salford.ac.uk (pan.salford.ac.uk [146.87.255.104]) by mx1.FreeBSD.org (Postfix) with SMTP id BDA1C43D1D for ; Thu, 10 Jun 2004 08:44:45 +0000 (GMT) (envelope-from M.S.Powell@salford.ac.uk) Received: (qmail 74571 invoked by uid 1236); 10 Jun 2004 08:44:05 -0000 Received: from M.S.Powell@salford.ac.uk by pan.salford.ac.uk by uid 401 with qmail-scanner-1.20 Clear:RC:1(146.87.255.76):. Processed in 0.022403 secs); 10 Jun 2004 08:44:05 -0000 Received: from [146.87.255.76] (HELO plato.salford.ac.uk) (146.87.255.76) by pan.salford.ac.uk (qpsmtpd/0.27-dev) with SMTP; Thu, 10 Jun 2004 09:44:05 +0100 Received: (qmail 82354 invoked by uid 1001); 10 Jun 2004 08:44:04 -0000 Received: from localhost (sendmail-bs@127.0.0.1) by localhost with SMTP; 10 Jun 2004 08:44:04 -0000 Date: Thu, 10 Jun 2004 09:44:04 +0100 (BST) From: "Mark Powell" To: freebsd-scsi@freebsd.org Message-ID: <20040610094010.Y76744@plato.salford.ac.uk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: RE: 4.10S amr v1.7.2.14 hangs at boot with Dell PERC3/DC (fwd) X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Jun 2004 08:44:49 -0000 Hi, I've been trying to even get a response from someone who admits that they maintain the amr driver. See below. Does anyone know who the right person to contact is? Cheers. NB Apologies to anyone irked by my copying basically a private exchange to this list. It's about scsi drivers not girlfriends, guys :) On Tue, 16 Mar 2004, Mark Powell wrote: > On Tue, 2 Mar 2004, Moore, Eric Dean wrote: > > > Mark - amr is maintained by Atul. > > Does Atul read his email? :) I never did receive a reponse to this problem report :( I notice that in 4.10S the same problem still exists. I'm having to build all kernels for Dell servers with amr cards with v1.7.2.13 to get them to boot. i.e. non-working versions are * $FreeBSD: src/sys/dev/amr/amr.c,v 1.7.2.14 2003/10/29 09:11:38 ps Exp $ * $FreeBSD: src/sys/dev/amr/amr_disk.c,v 1.5.2.6 2003/10/29 09:11:38 ps Exp $ * $FreeBSD: src/sys/dev/amr/amrvar.h,v 1.2.2.6 2003/10/29 09:11:38 ps Exp $ working versions are: * $FreeBSD: src/sys/dev/amr/amr.c,v 1.7.2.13 2003/01/15 13:41:18 emoore Exp $ * $FreeBSD: src/sys/dev/amr/amr_disk.c,v 1.5.2.5 2002/12/20 15:12:04 emoore Exp $ * $FreeBSD: src/sys/dev/amr/amrvar.h,v 1.2.2.5 2002/12/20 15:12:04 emoore Exp $ Changing these 3 files gets the server to boot. Surely this is quite a show-stopper to be ignored for 3 months? Many thanks. > Cheers. > > > > > > > On Tuesday, March 02, 2004 3:24 AM, Mark Powell wrote: > > > Hi, > > > I'm not sure if you're the author of the FreeBSD amr driver? Is the > > > below a known problem? > > > Cheers. > > > > > > ---------- Forwarded message ---------- > > > Date: Thu, 19 Feb 2004 09:48:55 -0800 > > > From: Mike Smith > > > To: Mark Powell > > > Cc: Atul Mukker > > > Subject: Re: amr v1.7.2.14 hangs at boot with Dell PERC3/DC > > > > > > > > > On Feb 19, 2004, at 5:24 AM, Mark Powell wrote: > > > > > > > Hi, > > > > I have a Poweredge 2550 with a Dell PERC3/DC card. If I use amr.c > > > > (and > > > > related files) from 30/10/03; i.e. v1.7.2.14; then the > > > kernel hangs at > > > > the > > > > point where it would bring the RAID controller up. If I > > > revert to the > > > > previous version of amr.c; i.e. v1.7.2.13; then all is fine. > > > > I tried searching to see if this was a known bug, but > > > couldn't find > > > > anything. > > > > I'm contacting you as a last resort, as I know you must be busy. > > > > > > Mark, > > > > > > The amr(4) driver is maintained by LSI these days; Atul, all yours. 8) > > > > > > = Mike > > > > > > > -- Mark Powell - UNIX System Administrator - The University of Salford Information Services Division, Clifford Whitworth Building, Salford University, Manchester, M5 4WT, UK. Tel: +44 161 295 4837 Fax: +44 161 295 5888 www.pgp.com for PGP key From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 09:46:43 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8776316A4CE for ; Fri, 11 Jun 2004 09:46:43 +0000 (GMT) Received: from usti.fccps.cz (usti.fccps.cz [194.108.74.250]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8339043D55 for ; Fri, 11 Jun 2004 09:46:42 +0000 (GMT) (envelope-from Frantisek.Rysanek@post.cz) Received: from localhost (localhost [127.0.0.1]) by usti.fccps.cz (Postfix) with ESMTP id A075377C87 for ; Fri, 11 Jun 2004 11:46:16 +0200 (CEST) Received: from usti.fccps.cz ([127.0.0.1]) by localhost (usti [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 22980-09 for ; Fri, 11 Jun 2004 11:46:15 +0200 (CEST) Received: from frr (unknown [10.0.0.238]) by usti.fccps.cz (Postfix) with ESMTP id 8E9C577C86 for ; Fri, 11 Jun 2004 11:46:14 +0200 (CEST) From: "Frantisek Rysanek" To: freebsd-scsi@freebsd.org Date: Fri, 11 Jun 2004 11:47:57 +0200 MIME-Version: 1.0 Message-ID: <40C99BED.12482.159E333F@localhost> Priority: normal X-mailer: Pegasus Mail for Windows (v4.12a) X-Virus-Scanned: by amavisd-new at fccps.cz Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Description: Mail message body X-Content-Filtered-By: Mailman/MimeDel 2.1.1 Subject: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 09:46:43 -0000 Dear SCSI experts and FreeBSD veterans, I'm looking for advice with an apparently simple SCSI issue: how do I set a maximum TCQ depth in FreeBSD? Scenario: - an U320 SCSI disk drive - Adaptec 39320 (aic7902) == U320 & PCI-X - FreeBSD 5.2.1 (and Linux 2.4.26, for comparison) - Bonnie (-s 1500, 4x in parallel, to trash the RAM buffers) - the system is a dual Xeon @ 2.6 GHz, 1 GB DDR266, ServerWorks GC-LE (PCI-X) I have encountered a SCSI disk device that tends to hang under load in FreeBSD - namely I have discovered that using Bonnie. The disk hangs when Bonnie says "rewriting" and in iostat you can see the TPS skyrocket. In Linux, with the default compile-time parameters to aic79xx.o, the disk works without a problem, no matter what Bonnie does to it. I figured out that it could be a problem with TCQ depth. The default in Linux/aic79xx is 32, so I cranked it up to 250, recompiled and ran Bonnie. Guess what - the disk hung at just the same point as in FreeBSD, only the TPS in linux managed to crawl yet somewhat higher. The hang invokes a flurry of error messages on the host system, both in Linux and FreeBSD - the precise wording is different, but essentially the drivers say that the SCSI bus is stuck and they keep trying to reset it, to no avail. The culprit device remains stuck until it's power-cycled. The datasheets for the disk device say that it can cope with 256 commands in the TCQ. Yet my experiments show that the highest safe value is about 224. 240 already makes the device crash. I tried to find the breakdown point by interval halving. An increase from 32 to 128 on the host side results in a significant performance improvement (about +50 %). A further increase to 192 or 224 is not much of an advantage. At 240, the disk chokes. To sum up, I'm fairly convinced that the disk can run just fine at a lower TCQ depth. The trouble is, that I can't find a way to set the TCQ depth in FreeBSD. Although the directory structure seems similar, the FreeBSD drivers for aic7xxx/aic79xx are significantly different from the Linux variety. The preprocessor macros and structs that handle TCQ settings in Linux are clearly missing in FreeBSD. I understand that FreeBSD is using a common SCSI and CAM layer, so perhaps I'd better be looking somewhere in there - only I can't find a hint of what to do. I tried looking into /usr/src/sys/conf/options /usr/src/sys/conf/NOTES /usr/src/sys/dev/aic7xxx/* etc. to no avail. I'm not too much of a coder, I get lost quickly in the CAM and SCSI layers. Any ideas are welcome. Thanks for your time reading this message :-) Frank Rysanek From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 12:04:49 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DB5BB16A4CE for ; Fri, 11 Jun 2004 12:04:49 +0000 (GMT) Received: from sockar.homeip.net (tourist.net1.nerim.net [62.212.109.212]) by mx1.FreeBSD.org (Postfix) with ESMTP id D165043D58 for ; Fri, 11 Jun 2004 12:04:48 +0000 (GMT) (envelope-from amon@sockar.homeip.net) Received: from sockar.homeip.net (localhost [127.0.0.1]) by sockar.homeip.net (8.12.9p2/8.12.9) with ESMTP id i5BC3YJL018056; Fri, 11 Jun 2004 14:03:34 +0200 (CEST) (envelope-from amon@sockar.homeip.net) Received: (from amon@localhost) by sockar.homeip.net (8.12.9p2/8.12.9/Submit) id i5BC3YuY018055; Fri, 11 Jun 2004 14:03:34 +0200 (CEST) (envelope-from amon) Date: Fri, 11 Jun 2004 14:03:34 +0200 From: Herve Boulouis To: Frantisek Rysanek Message-ID: <20040611140334.C70536@ra.aabs> References: <40C99BED.12482.159E333F@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.2.5.1i In-Reply-To: <40C99BED.12482.159E333F@localhost>; from Frantisek.Rysanek@post.cz on Fri, Jun 11, 2004 at 11:47:57AM +0200 cc: freebsd-scsi@freebsd.org Subject: Re: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 12:04:50 -0000 Le 11/06/2004 à 11:47, Frantisek Rysanek a écrit: > Dear SCSI experts and FreeBSD veterans, > > I'm looking for advice with an apparently simple SCSI issue: > how do I set a maximum TCQ depth in FreeBSD? I'm no kernel guru but maybe adding an entry to the xpt_quirk_table in sys/cam/cam_xpt.c for your disk would solve the problem ? -- Herve Boulouis From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 13:36:19 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8001B16A4CE for ; Fri, 11 Jun 2004 13:36:19 +0000 (GMT) Received: from mail.sandvine.com (sandvine.com [199.243.201.138]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9696D43D53 for ; Fri, 11 Jun 2004 13:36:18 +0000 (GMT) (envelope-from don@sandvine.com) Received: by mail.sandvine.com with Internet Mail Service (5.5.2657.72) id ; Fri, 11 Jun 2004 09:36:06 -0400 Message-ID: From: Don Bowman To: 'Frantisek Rysanek' , freebsd-scsi@freebsd.org Date: Fri, 11 Jun 2004 09:36:05 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain; charset="iso-8859-1" Subject: RE: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 13:36:19 -0000 From: Frantisek Rysanek [mailto:Frantisek.Rysanek@post.cz] > Dear SCSI experts and FreeBSD veterans, > > I'm looking for advice with an apparently simple SCSI issue: > how do I set a maximum TCQ depth in FreeBSD? > > Scenario: > - an U320 SCSI disk drive > - Adaptec 39320 (aic7902) == U320 & PCI-X > - FreeBSD 5.2.1 (and Linux 2.4.26, for comparison) > - Bonnie (-s 1500, 4x in parallel, to trash the RAM buffers) > - the system is a dual Xeon @ 2.6 GHz, 1 GB DDR266, > ServerWorks GC-LE (PCI-X) > ... drive timing out with high tag depth ... http://lists.freebsd.org/pipermail/freebsd-scsi/2003-August/000443.html shows a patch i did last year that achieves this. If you update the firmware on your drive you may find the problem is resolved as well. I cannot supply the firmware, but i can forward you a programming utility that may upgrade it for you. --don From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 15:55:00 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9399A16A4D0 for ; Fri, 11 Jun 2004 15:55:00 +0000 (GMT) Received: from mail.wolves.k12.mo.us (duey.wolves.k12.mo.us [207.160.214.9]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3ED8C43D54 for ; Fri, 11 Jun 2004 15:55:00 +0000 (GMT) (envelope-from cdillon@wolves.k12.mo.us) Received: from localhost (localhost [127.0.0.1]) by mail.wolves.k12.mo.us (Postfix) with ESMTP id 6AC7C20004; Fri, 11 Jun 2004 10:54:21 -0500 (CDT) Received: from mail.wolves.k12.mo.us ([127.0.0.1]) by localhost (mail.wolves.k12.mo.us [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 51910-02-21; Fri, 11 Jun 2004 10:54:20 -0500 (CDT) Received: by mail.wolves.k12.mo.us (Postfix, from userid 1001) id 1752B20003; Fri, 11 Jun 2004 10:54:20 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by mail.wolves.k12.mo.us (Postfix) with ESMTP id 15B6D1A956; Fri, 11 Jun 2004 10:54:20 -0500 (CDT) Date: Fri, 11 Jun 2004 10:54:20 -0500 (CDT) From: Chris Dillon To: Herve Boulouis In-Reply-To: <20040611140334.C70536@ra.aabs> Message-ID: <20040611104932.T52446@duey.wolves.k12.mo.us> References: <40C99BED.12482.159E333F@localhost> <20040611140334.C70536@ra.aabs> MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="0-245630583-1086969260=:52446" X-Virus-Scanned: by amavisd-new at wolves.k12.mo.us cc: Frantisek Rysanek cc: freebsd-scsi@freebsd.org Subject: Re: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 15:55:00 -0000 This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --0-245630583-1086969260=:52446 Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 8BIT On Fri, 11 Jun 2004, Herve Boulouis wrote: > Le 11/06/2004 à 11:47, Frantisek Rysanek a écrit: >> Dear SCSI experts and FreeBSD veterans, >> >> I'm looking for advice with an apparently simple SCSI issue: how do >> I set a maximum TCQ depth in FreeBSD? > > I'm no kernel guru but maybe adding an entry to the xpt_quirk_table > in sys/cam/cam_xpt.c for your disk would solve the problem ? The first thing you should do is upgrade the firmware on the drive and see if that solves the problem. If not, add an entry to the quirk table. You can also use camcontrol to limit the maximum number of transactions sent to the drive concurrently. To set the maximum queue depth on da0 to 32: camcontrol tags da0 -N 32 -- Chris Dillon - cdillon(at)wolves.k12.mo.us FreeBSD: The fastest, most open, and most stable OS on the planet - Available for IA32, IA64, AMD64, PC98, Alpha, and UltraSPARC architectures - PowerPC, ARM, MIPS, and S/390 under development - http://www.freebsd.org Q: Because it reverses the logical flow of conversation. A: Why is putting a reply at the top of the message frowned upon? --0-245630583-1086969260=:52446-- From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 18:19:02 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CF18616A4CE for ; Fri, 11 Jun 2004 18:19:02 +0000 (GMT) Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163]) by mx1.FreeBSD.org (Postfix) with ESMTP id 12A9243D45 for ; Fri, 11 Jun 2004 18:19:02 +0000 (GMT) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.12.11/8.12.11) with ESMTP id i5BIIspc022823; Fri, 11 Jun 2004 11:18:58 -0700 (PDT) (envelope-from truckman@FreeBSD.org) Message-Id: <200406111818.i5BIIspc022823@gw.catspoiler.org> Date: Fri, 11 Jun 2004 11:18:54 -0700 (PDT) From: Don Lewis To: Frantisek.Rysanek@post.cz In-Reply-To: <40C99BED.12482.159E333F@localhost> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii cc: freebsd-scsi@FreeBSD.org Subject: Re: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 18:19:03 -0000 On 11 Jun, Frantisek Rysanek wrote: > Dear SCSI experts and FreeBSD veterans, > > I'm looking for advice with an apparently simple SCSI issue: > how do I set a maximum TCQ depth in FreeBSD? > > Scenario: > - an U320 SCSI disk drive > - Adaptec 39320 (aic7902) == U320 & PCI-X > - FreeBSD 5.2.1 (and Linux 2.4.26, for comparison) > - Bonnie (-s 1500, 4x in parallel, to trash the RAM buffers) > - the system is a dual Xeon @ 2.6 GHz, 1 GB DDR266, > ServerWorks GC-LE (PCI-X) > > I have encountered a SCSI disk device that tends to hang > under load in FreeBSD - namely I have discovered that using > Bonnie. The disk hangs when Bonnie says "rewriting" and in > iostat you can see the TPS skyrocket. > > In Linux, with the default compile-time parameters to > aic79xx.o, the disk works without a problem, no matter what > Bonnie does to it. > > I figured out that it could be a problem with TCQ depth. The > default in Linux/aic79xx is 32, so I cranked it up to 250, > recompiled and ran Bonnie. Guess what - the disk hung at just > the same point as in FreeBSD, only the TPS in linux managed to > crawl yet somewhat higher. > > The hang invokes a flurry of error messages on the host > system, both in Linux and FreeBSD - the precise wording is > different, but essentially the drivers say that the SCSI bus > is stuck and they keep trying to reset it, to no avail. The > culprit device remains stuck until it's power-cycled. > > The datasheets for the disk device say that it can cope with > 256 commands in the TCQ. Yet my experiments show that > the highest safe value is about 224. 240 already makes the > device crash. > I tried to find the breakdown point by interval halving. > An increase from 32 to 128 on the host side results in > a significant performance improvement (about +50 %). > A further increase to 192 or 224 is not much of an advantage. > At 240, the disk chokes. > > To sum up, I'm fairly convinced that the disk can run just > fine at a lower TCQ depth. > > The trouble is, that I can't find a way to set the TCQ depth > in FreeBSD. Try the "camcontrol tags" command. As an alternative, you might want to use the "camcontrol modepage" command to turn off write caching, especially if you don't want the drive lie to soft updates about when data has been written to stable storage. The file system benchmarking that I"ve done with soft updates enabled on SCSI drives with TCQ has shown little difference in performance no matter whether write caching is enabled or not. The same drive controller buffer memory is used for the write cache and the TCQ cache in the drives that I've tested, so when write caching is enabled and the write cache fills up, the drive drops the number of outstanding tagged commands that it supports. I think the only real difference is when the drive sends back the acknowledgement that the data has been written. I have seen cases where a drive fails to service certain requests for an arbitrarily long period of time when it is saturated with a string of other (read?) requests that it can satisfy from cache, or at least that is what I think is going on. Reducing the TCQ depth and asking the vendor for a firmware fix were the two suggestions that I got. From owner-freebsd-scsi@FreeBSD.ORG Fri Jun 11 18:24:48 2004 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D17E116A4CE for ; Fri, 11 Jun 2004 18:24:48 +0000 (GMT) Received: from pooker.samsco.org (pooker.samsco.org [168.103.85.57]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4783343D39 for ; Fri, 11 Jun 2004 18:24:48 +0000 (GMT) (envelope-from scottl@freebsd.org) Received: from freebsd.org (junior-wifi.samsco.home [192.168.0.11]) (authenticated bits=0) by pooker.samsco.org (8.12.11/8.12.10) with ESMTP id i5BIQpe7012100; Fri, 11 Jun 2004 12:26:51 -0600 (MDT) (envelope-from scottl@freebsd.org) Message-ID: <40C9F8A9.2020407@freebsd.org> Date: Fri, 11 Jun 2004 12:23:37 -0600 From: Scott Long User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.6) Gecko/20040304 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Frantisek Rysanek References: <40C99BED.12482.159E333F@localhost> In-Reply-To: <40C99BED.12482.159E333F@localhost> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, hits=0.0 required=3.8 tests=none autolearn=no version=2.63 X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on pooker.samsco.org cc: freebsd-scsi@freebsd.org Subject: Re: how to limit SCSI TCQ depth in FreeBSD? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Jun 2004 18:24:48 -0000 Frantisek Rysanek wrote: > Dear SCSI experts and FreeBSD veterans, > > I'm looking for advice with an apparently simple SCSI issue: > how do I set a maximum TCQ depth in FreeBSD? > > Scenario: > - an U320 SCSI disk drive Knowing what brand, model, and fwrev of disks these are will help quite a bit. As others note, using camcontrol to limit the tags will work for now. Scott