From owner-cvs-all Wed Oct 14 15:37:46 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA09807 for cvs-all-outgoing; Wed, 14 Oct 1998 15:37:46 -0700 (PDT) (envelope-from owner-cvs-all@FreeBSD.ORG) Received: from metal.intt.org (metal.intt.org [206.109.108.186]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA09794; Wed, 14 Oct 1998 15:37:43 -0700 (PDT) (envelope-from smace@metal.intt.org) Received: (from smace@localhost) by metal.intt.org (8.9.1/8.9.1) id RAA21607; Wed, 14 Oct 1998 17:37:22 -0500 (CDT) From: Scott Mace Message-Id: <199810142237.RAA21607@metal.intt.org> Subject: Re: cvs commit: src/sys/cam cam_xpt.c In-Reply-To: <199810142226.QAA12957@panzer.plutotech.com> from "Kenneth D. Merry" at "Oct 14, 1998 4:26:35 pm" To: ken@plutotech.com (Kenneth D. Merry) Date: Wed, 14 Oct 1998 17:37:22 -0500 (CDT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, martin@tdc.on.ca X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk > Scott Mace wrote... > > > ken 1998/10/14 14:17:40 PDT > > > > > > Modified files: > > > sys/cam cam_xpt.c > > > Log: > > > Disable tagged queueing for the Seagate Elite 9GB drives. They tend to get > > > hung up when you send tags to them too quickly. (CAM is able to recover > > > from the problem, but this just avoids it altogether.) > > > > > > Reviewed by: gibbs > > > Reported by: Bret Ford > > > and: Martin Renters > > > > > > Revision Changes Path > > > 1.21 +10 -1 src/sys/cam/cam_xpt.c > > > > > > > This is odd, I have 2 elite 9's and they run fine > > with tagged queuing. What rev are the drives that fail? > > What are the symptoms exactly (maybe they don't run fine ?@?) > > Well, Bret Ford reported the following inquiry info: > > SEAGATE ST410800N 7110 > > Martin Renters hasn't yet sent me dmesg information. What firmware > revision do you have? > > The symptoms are that the drive just kinda locks up after heavy I/O. We > hit it with a BDR after a minute and it comes back. > > > In fact without tagged quueing they have very slow performance. > > I'm sure they do. > > If we can track it down to a specific firmware revision, we can narrow the > quirk entry a little more. > My scsi system looks like this: ahc0: rev 0x03 int a irq 11 on pci0.12.0 ahc0: aic7870 Single Channel A, SCSI Id=7, 16/255 SCBs Waiting 15 seconds for SCSI devices to settle sa0 at ahc0 bus 0 target 2 lun 0 sa0: Removable Sequential Access SCSI2 device sa0: 5.0MB/s transfers (5.0MHz, offset 15) sa1 at ahc0 bus 0 target 4 lun 0 sa1: Removable Sequential Access SCSI2 device sa1: 4.32MB/s transfers (4.32MHz, offset 11) da0 at ahc0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI2 device da0: 10.0MB/s transfers (10.0MHz, offset 15), Tagged Queueing Enabled da0: 2047MB (4194058 512 byte sectors: 255H 63S/T 261C) da1 at ahc0 bus 0 target 5 lun 0 da1: Fixed Direct Access SCSI2 device da1: 10.0MB/s transfers (10.0MHz, offset 15), Tagged Queueing Enabled da1: 8669MB (17755614 512 byte sectors: 255H 63S/T 1105C) da2 at ahc0 bus 0 target 6 lun 0 da2: Fixed Direct Access SCSI2 device da2: 10.0MB/s transfers (10.0MHz, offset 15), Tagged Queueing Enabled da2: 8669MB (17755614 512 byte sectors: 255H 63S/T 1105C) I have pounded the hell out of these drives in the past 2months with make worlds, cvs updates, hugh rm's, etc. Never had a single hang, hickup, or anyhting. I think the most current firmware is 0025, this is what the seagate repair center is shipping drives out with. Perhaps 7110 is an older firmware naming convention? The only thing I see is this: (da0:ahc0:0:0:0): tagged openings now 64 (da0:ahc0:0:0:0): tagged openings now 63 (da0:ahc0:0:0:0): tagged openings now 62 (da1:ahc0:0:5:0): tagged openings now 64 (da1:ahc0:0:5:0): tagged openings now 63 (da2:ahc0:0:6:0): tagged openings now 64 (da2:ahc0:0:6:0): tagged openings now 63 (da1:ahc0:0:5:0): tagged openings now 62 (da2:ahc0:0:6:0): tagged openings now 62 (da1:ahc0:0:5:0): tagged openings now 61 This is over time under heavy use, but never a lockup or hang or crash. Scott To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message