From owner-freebsd-scsi@FreeBSD.ORG Wed Aug 3 17:37:59 2011 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 26DEF1065670 for ; Wed, 3 Aug 2011 17:37:59 +0000 (UTC) (envelope-from ken@kdm.org) Received: from nargothrond.kdm.org (nargothrond.kdm.org [70.56.43.81]) by mx1.freebsd.org (Postfix) with ESMTP id CA0EE8FC08 for ; Wed, 3 Aug 2011 17:37:58 +0000 (UTC) Received: from nargothrond.kdm.org (localhost [127.0.0.1]) by nargothrond.kdm.org (8.14.2/8.14.2) with ESMTP id p73H2B08047847; Wed, 3 Aug 2011 11:02:11 -0600 (MDT) (envelope-from ken@nargothrond.kdm.org) Received: (from ken@localhost) by nargothrond.kdm.org (8.14.2/8.14.2/Submit) id p73H2AhV047846; Wed, 3 Aug 2011 11:02:10 -0600 (MDT) (envelope-from ken) Date: Wed, 3 Aug 2011 11:02:10 -0600 From: "Kenneth D. Merry" To: Gregory Wright Message-ID: <20110803170210.GA46958@nargothrond.kdm.org> References: <4E396DDD.9030602@antiope.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="gKMricLos+KVdGMg" Content-Disposition: inline In-Reply-To: <4E396DDD.9030602@antiope.com> User-Agent: Mutt/1.4.2i Cc: freebsd-scsi@freebsd.org Subject: Re: an mps probelm X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 03 Aug 2011 17:37:59 -0000 --gKMricLos+KVdGMg Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Wed, Aug 03, 2011 at 11:48:45 -0400, Gregory Wright wrote: > > Hi, > > I have an LSI SAS 9200-8e hba card. It's connected to a Dell > PowerVault 124T autoloader with an LTO4 drive. > > I installed 9.0-BETA1 yesterday and was able to successfully > detect the tape drive (mps driver), but the hba doesn't find the changer. > I used camcontrol to force rescanning all luns but that > didn't help either. Try rescanning the particular LUN for the changer device, like this: camcontrol rescan 6:4:1 (That is assuming the tape drive is at 6:4:0.) > Since I have just acquired this hardware I installed an OS > reported to work with this autoloader/hba combination, > ubuntu 11.04 server. On ubuntu, the changer is found: > > gwright@greenhouse-george:~$ lsscsi -g > [0:0:0:0] disk AMCC 9650SE-12M DISK 3.06 /dev/sdb /dev/sg1 > [0:0:0:1] disk AMCC 9650SE-12M DISK 3.06 /dev/sdc /dev/sg2 > [0:0:1:0] disk AMCC 9650SE-12M DISK 3.06 /dev/sdd /dev/sg3 > [1:0:0:0] tape IBM ULTRIUM-TD4 97F0 /dev/st0 /dev/sg5 > [1:0:0:1] mediumx DELL PV-124T 0080 /dev/sch0 /dev/sg6 > [2:0:0:0] disk ATA ST3160811AS 3.AA /dev/sda /dev/sg0 > [6:0:0:0] cd/dvd MITSUMI CD-ROM SR244W T01A /dev/sr0 /dev/sg4 > > (On ubuntu the driver is mpt2sas.) > > It seems as if the mps driver is not seeing the additional lun > hanging off the bus. Yeah, it looks like the path inquiry CCB in the driver is telling CAM not to scan LUNs. > This server is not in production yet so I can re-install 9.0-BETA1 > and run more tests, if that would help. Try the attached patch, and let me know whether it works. > FWIW, I set hw.mps.0.debug_level=1. In the dmesg I get > > greenhouse-george# dmesg | grep mps > mps0: port 0xa000-0xa0ff mem > 0xff4f0000-0xff4fffff,0xff480000-0xff4bffff irq 28 at device 0.0 on pci5 > mps0: Doorbell= 0x22000000 > mps0: Doorbell= 0x12000000 > mps0: IOCFacts : > mps0: Firmware: 02.15.63.00 > mps0: IOCCapabilities: > 185c > mps0: Doorbell= 0x10000000 > mps0: num_reqs 1024, num_replies 1055 > mps0: Doorbell= 0x10000000 > mps0: IOCInit status= 0x0 > mps0: writing postindex 1 > mps0: PortFacts : > mps0: writing postindex 2 > mps0: writing postindex 3 > mps0: mpssas_discovery_end: removing confighook > mps0: writing postindex 8 > mps0: SAS Device Page 0 : > mps0: writing postindex 9 > mps0: writing postindex 10 > mps0: SAS PHY Page 0 : > mps0: Triggering rescan of 6:4:-1 > mps0: writing postindex 11 > mps0: writing postindex 12 > mps0: writing postindex 13 > mps0: (6:4:0) IOCStatus= 0x45, ScsiStatus= 0x2, SCSIState= 0x1 > TransferCount= 0x0 > mps0: writing postindex 14 > mps0: (6:4:0) IOCStatus= 0x45, ScsiStatus= 0x0, SCSIState= 0x0 > TransferCount= 0x11 > mps0: writing postindex 15 > mps0: (6:4:0) IOCStatus= 0x45, ScsiStatus= 0x0, SCSIState= 0x0 > TransferCount= 0x56 > mps0: writing postindex 16 > mps0: (6:4:0) IOCStatus= 0x45, ScsiStatus= 0x0, SCSIState= 0x0 > TransferCount= 0xe > mps0: writing postindex 17 > mps0: (6:4:0) IOCStatus= 0x0, ScsiStatus= 0x2, SCSIState= 0x1 > TransferCount= 0x0 > mps0: writing postindex 18 > (probe4:mps0:0:4:0): TEST UNIT READY. CDB: 0 0 0 0 0 0 > (probe4:mps0:0:4:0): CAM status: SCSI Status Error > (probe4:mps0:0:4:0): SCSI status: Check Condition > (probe4:mps0:0:4:0): SCSI sense: NOT READY asc:3a,0 (Medium not present) > field rp > > > Please let me know what I can do to help debug this. Ken -- Kenneth Merry ken@FreeBSD.ORG --gKMricLos+KVdGMg Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="mps_sas.c.max_lun.20110803" ==== //depot/SpectraBSD/head/sys/dev/mps/mps_sas.c#29 - /usr/home/kenm/perforce4/SpectraBSD/head/sys/dev/mps/mps_sas.c ==== *** /tmp/tmp.96563.59 Wed Aug 3 10:56:49 2011 --- /usr/home/kenm/perforce4/SpectraBSD/head/sys/dev/mps/mps_sas.c Wed Aug 3 10:56:31 2011 *************** *** 921,927 **** cpi->hba_misc = PIM_NOBUSRESET; cpi->hba_eng_cnt = 0; cpi->max_target = sassc->sc->facts->MaxTargets - 1; ! cpi->max_lun = 0; cpi->initiator_id = 255; strncpy(cpi->sim_vid, "FreeBSD", SIM_IDLEN); strncpy(cpi->hba_vid, "LSILogic", HBA_IDLEN); --- 921,927 ---- cpi->hba_misc = PIM_NOBUSRESET; cpi->hba_eng_cnt = 0; cpi->max_target = sassc->sc->facts->MaxTargets - 1; ! cpi->max_lun = 8; cpi->initiator_id = 255; strncpy(cpi->sim_vid, "FreeBSD", SIM_IDLEN); strncpy(cpi->hba_vid, "LSILogic", HBA_IDLEN); --gKMricLos+KVdGMg--