From owner-freebsd-scsi Sun Apr 28 04:07:23 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id EAA25344 for freebsd-scsi-outgoing; Sun, 28 Apr 1996 04:07:23 -0700 (PDT) Received: from ibp.ibp.fr (ibp.ibp.fr [132.227.60.30]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id EAA25306 for ; Sun, 28 Apr 1996 04:07:01 -0700 (PDT) Received: from blaise.ibp.fr (blaise.ibp.fr [132.227.60.1]) by ibp.ibp.fr (8.6.12/jtpda-5.0) with ESMTP id NAA05630 ; Sun, 28 Apr 1996 13:06:57 +0200 Received: from (uucp@localhost) by blaise.ibp.fr (8.6.12/jtpda-5.0) with UUCP id NAA05991 ; Sun, 28 Apr 1996 13:06:58 +0200 Received: (from roberto@localhost) by keltia.freenix.fr (8.7.5/keltia-uucp-2.7) id NAA04177; Sun, 28 Apr 1996 13:01:34 +0200 (MET DST) From: Ollivier Robert Message-Id: <199604281101.NAA04177@keltia.freenix.fr> Subject: Re: Adaptec 1542, Adaptec 1742, Bustek 747, Archive Viper, Help? To: peter@taronga.com (Peter da Silva) Date: Sun, 28 Apr 1996 13:01:33 +0200 (MET DST) Cc: scsi@freebsd.org In-Reply-To: <199604272249.RAA01804@bonkers.taronga.com> from Peter da Silva at "Apr 27, 96 05:49:58 pm" X-Operating-System: FreeBSD 2.2-CURRENT ctm#1916 X-Mailer: ELM [version 2.4ME+ PL15 (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 It seems that Peter da Silva said: > When I try to install the 1742 as a second controller, it interferes with > the 747. No matter what EISA config I set, if the 1742 is enabled the Bustek > BIOS tried to drive it as a Bustek. I don't understand EISA config that > well, so I could simply be thick. I'm surprised because I also have a Buslogic 747S and an Adaptec 1740A in my machine: eisa0: Probing for devices on the EISA bus bt0: at 0x330-0x333, 0x3c00-0x3cff irq 11 bt0: on eisa0 slot 3 bt0: Bt747 / 0-(32bit) bus bt0: reading board settings, busmastering, int=11 bt0: version 3.37, fast sync, parity, 32 mbxs, 32 ccbs [...] ahb0: at 0x5c00-0x5cff irq 12 on eisa0 slot 5 I run -CURRENT and the SCSI code has signifiantly changed since 2.0.5R... I don't remember whether I had the 1740A or not during 2.0.5R times but I've had any problem mixing the two. Is your 1740A in enhanced mode ? I can send you my EISA config. files if you want. -- Ollivier ROBERT -=- The daemon is FREE! -=- roberto@keltia.freenix.fr FreeBSD keltia.freenix.fr 2.2-CURRENT #13: Sun Apr 21 18:14:54 MET DST 1996 From owner-freebsd-scsi Sun Apr 28 10:17:40 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA11962 for freebsd-scsi-outgoing; Sun, 28 Apr 1996 10:17:40 -0700 (PDT) Received: from uuneo.neosoft.com (root@uuneo.neosoft.com [206.109.1.3]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id KAA11957 for ; Sun, 28 Apr 1996 10:17:38 -0700 (PDT) Received: (from taronga@localhost) by uuneo.neosoft.com (8.7.5/8.7.4) with UUCP id LAA05058 for scsi@freebsd.org; Sun, 28 Apr 1996 11:18:40 -0500 (CDT) Received: (from peter@localhost) by bonkers.taronga.com (8.6.11/8.6.9) id LAA17615 for scsi@freebsd.org; Sun, 28 Apr 1996 11:09:48 -0500 Date: Sun, 28 Apr 1996 11:09:48 -0500 From: peter@taronga.com (Peter da Silva) Message-Id: <199604281609.LAA17615@bonkers.taronga.com> To: scsi@freebsd.org Subject: Anaconda Update Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Damndest thing... Backups using DUMP have no problem writing to the whole tape. But when I tried AMANDA I got: /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error last message repeated 111 times From owner-freebsd-scsi Sun Apr 28 12:57:12 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA20158 for freebsd-scsi-outgoing; Sun, 28 Apr 1996 12:57:12 -0700 (PDT) Received: from cedb.dpcsys.com (cedb.DPCSYS.COM [165.90.143.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA20145 for ; Sun, 28 Apr 1996 12:57:09 -0700 (PDT) Received: from cedb (cedb.DPCSYS.COM [165.90.143.3]) by cedb.dpcsys.com (8.6.10/DPC-1.0) with SMTP id TAA13189; Sun, 28 Apr 1996 19:44:11 GMT Date: Sun, 28 Apr 1996 12:44:10 -0700 (PDT) From: Dan Busarow X-Sender: dan@cedb To: Peter da Silva cc: scsi@freebsd.org Subject: Re: Anaconda Update In-Reply-To: <199604281609.LAA17615@bonkers.taronga.com> 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 Sun, 28 Apr 1996, Peter da Silva wrote: > Backups using DUMP have no problem writing to the whole tape. But when > I tried AMANDA I got: I just picked up an Anaconda from CSC myself. Running 2.1.0R with an AH1542C. I'm getting medium errors and write errors fairly quickly (less than 100K), dump reports end of medium. Searching through the lists on www.freebsd.org I found several threads on this drive but no answers. Except for a cryptic "once I figured out active termination" that doesn't describe what needed to be done. Could anyone elaborate on what it takes to get this drive/controller working? I'm also planning on using it in conjuntion with Amanda. dmesg output attached. Thanks, Dan -- Dan Busarow DPC Systems Dana Point, California FreeBSD 2.1.0-RELEASE #0: Fri Apr 26 12:06:09 PDT 1996 root@saltcreek.beach.net:/usr/src/sys/compile/Beach-1 CPU: i486 DX2 (486-class CPU) Origin = "GenuineIntel" Id = 0x435 Stepping=5 Features=0x3 real memory = 8388608 (8192K bytes) avail memory = 6549504 (6396K bytes) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 not found at 0x280 ed1 at 0x300-0x31f irq 5 on isa ed1: address 00:40:05:16:78:82, type NE2000 (16 bit) cy0 irq 10 maddr 0xd4000 msize 8192 on isa sio0 not found at 0x3f8 sio1 not found at 0x2f8 lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface mse0: wrong signature ff mse0 not found at 0x23c fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 514MB (1054368 sectors), 1046 cyls, 16 heads, 63 S/T, 512 B/S wdc1 not found at 0x170 bt0 not found at 0x330 uha0 not found at 0x330 ahc1 not found ahb0 not found aha0 at 0x330-0x333 irq 11 drq 5 on isa aha0 waiting for scsi devices to settle (aha0:0:0): "CONNER CFP1060S 1.05GB 2035" type 0 fixed SCSI 2 sd0(aha0:0:0): Direct-Access 1013MB (2074880 512 byte sectors) (aha0:6:0): "ARCHIVE ANCDA 2750 28077 -002" type 1 removable SCSI 2 st0(aha0:6:0): Sequential-Access density code 0x0, drive empty aic0 not found at 0x340 nca0 not found at 0x1f88 nca1 not found at 0x350 sea0 not found wt0 not probed due to I/O address conflict with ed1 at 0x300 mcd0 not probed due to I/O address conflict with ed1 at 0x300 mcd1: timeout getting status mcd1 not found at 0x340 matcdc0 not found at 0x230 scd0 not found at 0x230 ie0 not found at 0x360 ep0 not probed due to I/O address conflict with ed1 at 0x300 ix0 not probed due to I/O address conflict with ed1 at 0x300 le0 not probed due to I/O address conflict with ed1 at 0x300 lnc0 not found at 0x280 lnc1 not probed due to I/O address conflict with ed1 at 0x300 ze0 not probed due to I/O address conflict with ed1 at 0x300 zp0 not probed due to I/O address conflict with ed1 at 0x300 npx0 on motherboard npx0: INT 16 interface From owner-freebsd-scsi Mon Apr 29 02:17:59 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id CAA04842 for freebsd-scsi-outgoing; Mon, 29 Apr 1996 02:17:59 -0700 (PDT) Received: from uuneo.neosoft.com (root@uuneo.neosoft.com [206.109.1.3]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id CAA04834 for ; Mon, 29 Apr 1996 02:17:52 -0700 (PDT) Received: (from taronga@localhost) by uuneo.neosoft.com (8.7.5/8.7.4) with UUCP id QAA27678 for scsi@freebsd.org; Sun, 28 Apr 1996 16:18:03 -0500 (CDT) Received: (from peter@localhost) by bonkers.taronga.com (8.6.11/8.6.9) id QAA02278 for scsi@freebsd.org; Sun, 28 Apr 1996 16:01:13 -0500 Date: Sun, 28 Apr 1996 16:01:13 -0500 From: peter@taronga.com (Peter da Silva) Message-Id: <199604282101.QAA02278@bonkers.taronga.com> To: scsi@freebsd.org Subject: Anaconda on Adaptec 154x ... Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk More errors from the Anaconda: Dumping with Amanda: /kernel: st0(aha0:5:0): MEDIUM ERROR info:2a8 asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR info:2a8 asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error Dumping with dump: /kernel: st0(aha0:5:0): MEDIUM ERROR info:2b0 asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR info:2e0 asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR info:2ac asc:c,0 Write error /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error The following came from trying to read and write DC600 and DC6150 tapes: /kernel: st0(aha0:5:0): NOT READY asc:4,1 /kernel: st0(aha0:5:0): Logical unit is in process of becoming ready /kernel: st0(aha0:5:0): NOT READY asc:4,1 /kernel: st0(aha0:5:0): Logical unit is in process of becoming ready /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0: oops not queued /kernel: st0: oops not queued /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0: oops not queued /kernel: st0: oops not queued /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0: oops not queued /kernel: st0: oops not queued /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:4,1 /kernel: st0(aha0:5:0): Logical unit is in process of becoming ready /kernel: st0(aha0:5:0): MEDIUM ERROR info:10000 asc:80,97 /kernel: st0(aha0:5:0): MEDIUM ERROR info:10000 asc:80,97 /kernel: st0(aha0:5:0): ILLEGAL REQUEST asc:52,0 Cartridge fault /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error /kernel: st0(aha0:5:0): NOT READY asc:3a,0 Medium not present /kernel: st0(aha0:5:0): MEDIUM ERROR asc:c,0 Write error From owner-freebsd-scsi Mon Apr 29 18:07:56 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id SAA27752 for freebsd-scsi-outgoing; Mon, 29 Apr 1996 18:07:56 -0700 (PDT) Received: from sumter.awod.com (awod.com [198.81.225.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id SAA27747 Mon, 29 Apr 1996 18:07:53 -0700 (PDT) Received: from Ken (tsunami.awod.com [198.81.225.31]) by sumter.awod.com (8.6.12/8.6.9) with SMTP id VAA23456; Mon, 29 Apr 1996 21:07:06 -0400 Message-Id: <1.5.4.32.19960430010743.009aa208@awod.com> X-Sender: klam@awod.com X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 29 Apr 1996 21:07:43 -0400 To: freebsd-current@FreeBSD.org From: Ken Lam Subject: scsi errors in CURRENT with AHA2940 Cc: scsi@FreeBSD.org Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk I've got a 2940 in a P5-90 with SiS chipset. The following is what I have from the logs, note that it doesn't crash, can't access the drives (input/output error, I believe). sd0(ahc0:0:0): timed out in dataout phase, SCSISIGI == 0x0 sd0(ahc0:0:0): BUS DEVICE RESET message queued. sd0(ahc0:0:0): timed out in dataout phase, SCSISIGI == 0x0 ahc0: Issued Channel A Bus Reset #1. 2 SCBs aborted sd1(ahc0:1:0): Power on, reset, or bus device reset occurred field replaceable unit: 1 I've been having problems with the system and haven't been able to identify the cause, I have checked termination, and swapped with known good cables. Suggestions? TIA, Ken --- Ken Lam "'Plug and Play' only worked with the original ATARI(tm)" From owner-freebsd-scsi Mon Apr 29 21:25:57 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA02934 for freebsd-scsi-outgoing; Mon, 29 Apr 1996 21:25:57 -0700 (PDT) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA02913 Mon, 29 Apr 1996 21:25:50 -0700 (PDT) Message-Id: <199604300425.VAA02913@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Ken Lam cc: freebsd-current@freebsd.org, scsi@freebsd.org Subject: Re: scsi errors in CURRENT with AHA2940 In-reply-to: Your message of "Mon, 29 Apr 1996 21:07:43 EDT." <1.5.4.32.19960430010743.009aa208@awod.com> Date: Mon, 29 Apr 1996 21:25:50 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk How recent are your sources? What kind of devices do you have on the chain? Are you using any AHC kernel config options? Under what kinds of loads are you seeing this problem? >I've got a 2940 in a P5-90 with SiS chipset. > >The following is what I have from the logs, note that it doesn't >crash, can't access the drives (input/output error, I believe). > >sd0(ahc0:0:0): timed out in dataout phase, SCSISIGI == 0x0 >sd0(ahc0:0:0): BUS DEVICE RESET message queued. >sd0(ahc0:0:0): timed out in dataout phase, SCSISIGI == 0x0 >ahc0: Issued Channel A Bus Reset #1. 2 SCBs aborted >sd1(ahc0:1:0): Power on, reset, or bus device reset occurred field replaceabl >e >unit: 1 > >I've been having problems with the system and haven't been able to >identify the cause, I have checked termination, and swapped with known >good cables. > >Suggestions? > >TIA, >Ken >--- >Ken Lam >"'Plug and Play' only worked with the original ATARI(tm)" > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Tue Apr 30 19:32:09 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA05019 for freebsd-scsi-outgoing; Tue, 30 Apr 1996 19:32:09 -0700 (PDT) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [204.214.4.2]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id TAA04945 for ; Tue, 30 Apr 1996 19:31:55 -0700 (PDT) Received: from max3-203.HiWAAY.net by fly.HiWAAY.net; (5.65v3.0/1.1.8.2/21Sep95-1003PM) id AA18259; Tue, 30 Apr 1996 21:31:44 -0500 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 30 Apr 1996 21:31:52 -0500 To: freebsd-scsi@freebsd.org From: dkelly@hiwaay.net (David Kelly) Subject: Buffered Writes? Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk How do I querry (and maybe enable) my hard drives for buffered writes? I have (2) supposedly identical ST3610N Seagate 500M drives and the one on SCSI ID #0 seems to run at half the thruput of the other on ID #1. I'm guessing one had internal caching enabled. The answer is mostly acedemic now as those drives are elsewhere, replaced with an ST32550N, but now I'd like to know more about how my Barracuda is configured. Studying scsi(8) suggests that is where the key is but at this time I know enough to know I shouldn't go playing with scsi(8) using random commands. System is a NexGen PCI-90 with Adaptec 2940. -- David Kelly N4HHE, n4hhe@amsat.org, dkelly@hiwaay.net ============================================================= To invent, you need a good imagination and a pile of junk. - Thomas Edison From owner-freebsd-scsi Wed May 1 01:52:03 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id BAA24191 for freebsd-scsi-outgoing; Wed, 1 May 1996 01:52:03 -0700 (PDT) Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id BAA24169 for ; Wed, 1 May 1996 01:51:58 -0700 (PDT) Received: from sax.sax.de by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id KAA00855; Wed, 1 May 1996 10:51:02 +0200 Received: by sax.sax.de (8.6.11/8.6.12-s1) with UUCP id KAA07357; Wed, 1 May 1996 10:51:02 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id KAA09890; Wed, 1 May 1996 10:50:21 +0200 (MET DST) From: J Wunsch Message-Id: <199605010850.KAA09890@uriah.heep.sax.de> Subject: Re: Buffered Writes? To: scsi@freebsd.org Date: Wed, 1 May 1996 10:50:20 +0200 (MET DST) Cc: dkelly@hiwaay.net (David Kelly) Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: from David Kelly at "Apr 30, 96 09:31:52 pm" 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 X-Mailer: ELM [version 2.4ME+ PL15 (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 As David Kelly wrote: > How do I querry (and maybe enable) my hard drives for buffered writes? I use ``scsi -f /dev/rsd0.ctl -m '' to browse through the available mode pages. -- 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 Wed May 1 05:30:18 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id FAA06419 for freebsd-scsi-outgoing; Wed, 1 May 1996 05:30:18 -0700 (PDT) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [204.214.4.2]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id FAA06414 for ; Wed, 1 May 1996 05:30:16 -0700 (PDT) Received: from max1-150.HiWAAY.net by fly.HiWAAY.net; (5.65v3.0/1.1.8.2/21Sep95-1003PM) id AA02175; Wed, 1 May 1996 07:30:00 -0500 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 1 May 1996 07:30:05 -0500 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch), scsi@freebsd.org From: dkelly@hiwaay.net (David Kelly) Subject: Re: Buffered Writes? Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 3:50 AM 5/1/96, J Wunsch wrote: >As David Kelly wrote: > >> How do I querry (and maybe enable) my hard drives for buffered writes? I > >use ``scsi -f /dev/rsd0.ctl -m '' to browse through >the available mode pages. Thanks. Guess I just "shoot for" available mode pages? Would there be any gaps in the numbering? First is zero? or one? -- David Kelly N4HHE, n4hhe@amsat.org, dkelly@hiwaay.net ============================================================= To invent, you need a good imagination and a pile of junk. - Thomas Edison From owner-freebsd-scsi Wed May 1 06:49:26 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA09874 for freebsd-scsi-outgoing; Wed, 1 May 1996 06:49:26 -0700 (PDT) Received: from iworks.InterWorks.org (deischen@iworks.interworks.org [128.255.18.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA09869 for ; Wed, 1 May 1996 06:49:25 -0700 (PDT) Received: by iworks.InterWorks.org (1.37.109.8/16.2) id AA01302; Wed, 1 May 1996 08:48:54 -0500 Message-Id: <9605011348.AA01302@iworks.InterWorks.org> Date: Wed, 1 May 1996 08:48:54 -0500 From: "Daniel M. Eischen" To: dkelly@hiwaay.net, scsi@freebsd.org Subject: Re: Buffered Writes? Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >>> How do I querry (and maybe enable) my hard drives for buffered writes? I >> >>use ``scsi -f /dev/rsd0.ctl -m '' to browse through >>the available mode pages. > >Thanks. Guess I just "shoot for" available mode pages? Would there be any >gaps in the numbering? First is zero? or one? Look in /usr/share/misc/scsi_modes :-) I believe what you want is mode page 8. Dan Eischen deischen@iworks.InterWorks.org From owner-freebsd-scsi Wed May 1 08:20:07 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA14606 for freebsd-scsi-outgoing; Wed, 1 May 1996 08:20:07 -0700 (PDT) Received: from www.interlog.com (www.interlog.com [198.53.145.8]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA14518 for ; Wed, 1 May 1996 08:19:44 -0700 (PDT) Received: from gold.interlog.com (root@gold.interlog.com [198.53.145.2]) by www.interlog.com (8.6.12/8.6.12) with ESMTP id LAA27874 for ; Wed, 1 May 1996 11:19:42 -0400 Received: from aks.interlog.com (aks.interlog.com [206.108.74.121]) by gold.interlog.com (8.6.10/8.6.10) with SMTP id LAA25522 for ; Wed, 1 May 1996 11:19:40 -0400 Received: by aks.interlog.com with Microsoft Mail id <01BB374F.6B769C20@aks.interlog.com>; Wed, 1 May 1996 11:14:52 -0400 Message-ID: <01BB374F.6B769C20@aks.interlog.com> From: Andrew Suessmuth To: "'freebsd-scsi@freebsd.org'" Subject: ASUS SP3G NCR problem Date: Wed, 1 May 1996 10:52:02 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Umm.... HELP! :-) I've just spent a futile night scouring the freebsd archives, looking = for answers to my SCSI boot problems, and couldn't find an answer (that = worked anyways. :-) ) I have a rev 1.8 PCI/I-SP3G ASUS motherboard with a rev 4 Saturn chip in = it and a rev 2 NCR 53c810 chip. I'm trying to boot using the 2.1.0 boot = floppy image. The devices on the SCSI chain are: Toshiba MK538FB hard = drive, Hitachi DK516C hard drive, ARCHIVE Python DAT, and a Sony = CDU-76S. All the PCI related BIOS settings are set to the BIOS = defaults. When FreeBSD reaches the point where it polls the NCR SCSI bus, it = starts scrolling a bunch error messages. I haven't recorded all the = messages it scrolls up the screen, but after correctly identifying the = device (it prints the right name on screen) it receives an M_REJECT and = a timeout for each device. One line in the middle of the scroll that = caught my eye might be helpful in IDing the problem: DEBUG: ioctl(3, TIOCCONS, NULL) =3D 0 (success) Another cute effect occurs when I exit the install program, causing a = reboot, my system reports a CMOS checksum error, and gets reset to the = default BIOS settings. I appologize in advance if this isn't the correct list to post this = question in (it seemed most fitting to me). Andrew Suessmuth aks@interlog.comTesting From owner-freebsd-scsi Wed May 1 10:51:02 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA04339 for freebsd-scsi-outgoing; Wed, 1 May 1996 10:51:02 -0700 (PDT) Received: from Sisyphos (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA04334 for ; Wed, 1 May 1996 10:50:56 -0700 (PDT) Received: by Sisyphos id AA12965 (5.67b/IDA-1.5 for scsi@freebsd.org); Wed, 1 May 1996 19:50:34 +0200 Message-Id: <199605011750.AA12965@Sisyphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Wed, 1 May 1996 19:50:34 +0200 In-Reply-To: Andrew Suessmuth "ASUS SP3G NCR problem" (May 1, 10:52) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Andrew Suessmuth Subject: Re: ASUS SP3G NCR problem Cc: scsi@freebsd.org Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On May 1, 10:52, Andrew Suessmuth wrote: } Subject: ASUS SP3G NCR problem } Umm.... HELP! :-) } } I've just spent a futile night scouring the freebsd archives, looking = } for answers to my SCSI boot problems, and couldn't find an answer (that = } worked anyways. :-) ) } } I have a rev 1.8 PCI/I-SP3G ASUS motherboard with a rev 4 Saturn chip in = } it and a rev 2 NCR 53c810 chip. Fine. The development system for the NCR driver, BTW. } I'm trying to boot using the 2.1.0 boot = } floppy image. The devices on the SCSI chain are: Toshiba MK538FB hard = } drive, Hitachi DK516C hard drive, ARCHIVE Python DAT, and a Sony = } CDU-76S. All the PCI related BIOS settings are set to the BIOS = } defaults. The Toshiba has caused problems before, the other devices look fine. I'm not sure whether the Sony CDROM can be used with synch. transfers, though. } When FreeBSD reaches the point where it polls the NCR SCSI bus, it = } starts scrolling a bunch error messages. I haven't recorded all the = } messages it scrolls up the screen, but after correctly identifying the = } device (it prints the right name on screen) it receives an M_REJECT and = } a timeout for each device. One line in the middle of the scroll that = There is a M_REJECT for **each** device ? Please check again. I'd expect a single device to cause all these error messages. There is the SCSI ID in the error messages: % (ncr0:2:0): "TOSHIBA MK537FB 6257" is a type 0 fixed SCSI % sd0(ncr0:2:0): Direct-Access % sd0(ncr0:2:0): asynchronous % sd0(ncr0:2:0): M_REJECT received (2:8) (This is from an earlier problem report, which also was caused by some problem with synch. transfer negotiation between the NCR driver and the Toshiba drive.) } caught my eye might be helpful in IDing the problem: } } DEBUG: ioctl(3, TIOCCONS, NULL) =3D 0 (success) This seems unrelated. } Another cute effect occurs when I exit the install program, causing a = } reboot, my system reports a CMOS checksum error, and gets reset to the = } default BIOS settings. This should not happen, and I never heard about this before ... } I appologize in advance if this isn't the correct list to post this = } question in (it seemed most fitting to me). It definitely is the correct list, at least as far as the NCR+Toshiba problem is concerned. Could you please try the boot floppy from the latest SNAP release ? If you can disconnect the Toshiba for the time of the installation, you may be able to boot the 2.1.0 floppy and install to the other disk drive. Once the system is up, you can rebuild the kernel, and you could use the Toshiba with asynch. transfers, at least. I could then send patches for you to try, and to get the synch. transfer negotiation going ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se From owner-freebsd-scsi Wed May 1 12:07:24 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA09404 for freebsd-scsi-outgoing; Wed, 1 May 1996 12:07:24 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA09391 for ; Wed, 1 May 1996 12:07:19 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id FAA06873; Thu, 2 May 1996 05:06:16 +1000 Date: Thu, 2 May 1996 05:06:16 +1000 From: Bruce Evans Message-Id: <199605011906.FAA06873@godzilla.zeta.org.au> To: aks@interlog.com, freebsd-scsi@FreeBSD.org Subject: Re: ASUS SP3G NCR problem Sender: owner-freebsd-scsi@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >I have a rev 1.8 PCI/I-SP3G ASUS motherboard with a rev 4 Saturn chip in = >it and a rev 2 NCR 53c810 chip. I'm trying to boot using the 2.1.0 boot = >floppy image. The devices on the SCSI chain are: Toshiba MK538FB hard = >drive, Hitachi DK516C hard drive, ARCHIVE Python DAT, and a Sony = >CDU-76S. All the PCI related BIOS settings are set to the BIOS = >defaults. Toshiba MK* drives don't work with ncr controller/driver unless command tags are disabled. Command tags are disabled by default in GENERIC in -current. For old versions you have to build a special kernel with tags disabled: option SCSI_NCR_DEFAULT_TAGS 0 or something like that. Bruce From owner-freebsd-scsi Wed May 1 16:12:01 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id QAA29832 for freebsd-scsi-outgoing; Wed, 1 May 1996 16:12:01 -0700 (PDT) Received: from sunrise.cs.berkeley.edu (root@sunrise.CS.Berkeley.EDU [128.32.38.121]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id QAA29805 for ; Wed, 1 May 1996 16:11:57 -0700 (PDT) Received: (from asami@localhost) by sunrise.cs.berkeley.edu (8.6.12/8.6.12) id QAA05975; Wed, 1 May 1996 16:12:48 -0700 Date: Wed, 1 May 1996 16:12:48 -0700 Message-Id: <199605012312.QAA05975@sunrise.cs.berkeley.edu> To: dkelly@hiwaay.net CC: freebsd-scsi@freebsd.org In-reply-to: (dkelly@hiwaay.net) Subject: Re: Buffered Writes? From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * How do I querry (and maybe enable) my hard drives for buffered writes? I * have (2) supposedly identical ST3610N Seagate 500M drives and the one on * SCSI ID #0 seems to run at half the thruput of the other on ID #1. I'm * guessing one had internal caching enabled. The answer is mostly acedemic * now as those drives are elsewhere, replaced with an ST32550N, but now I'd * like to know more about how my Barracuda is configured. Studying scsi(8) * suggests that is where the key is but at this time I know enough to know I * shouldn't go playing with scsi(8) using random commands. * * System is a NexGen PCI-90 with Adaptec 2940. Yes, the Barracuda comes without the write cache enabled. However, by turning on the tagged queuing (options AHC_TAGENABLE), we got the sequential write speed to get very close to the read speed: write read without tag 3.6 6.9 with tag 6.2 7.9 So we haven't actually looked into how to enable the write cache. Please let me know if you find something interesting. Satoshi From owner-freebsd-scsi Fri May 3 12:30:03 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA05568 for freebsd-scsi-outgoing; Fri, 3 May 1996 12:30:03 -0700 (PDT) Received: from s1.GANet.NET (s1.GANet.NET [199.18.201.2]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA05542 for ; Fri, 3 May 1996 12:30:00 -0700 (PDT) Received: (from ec0@localhost) by s1.GANet.NET (8.6.11/8.6.11) id PAA29593; Fri, 3 May 1996 15:28:31 -0400 Date: Fri, 3 May 1996 15:28:29 -0400 (EDT) From: Eric Chet To: Satoshi Asami cc: dkelly@HiWAAY.net, freebsd-scsi@FreeBSD.ORG Subject: Re: Buffered Writes? In-Reply-To: <199605012312.QAA05975@sunrise.cs.berkeley.edu> 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 Wed, 1 May 1996, Satoshi Asami wrote: > * How do I querry (and maybe enable) my hard drives for buffered writes? I > * have (2) supposedly identical ST3610N Seagate 500M drives and the one on > * SCSI ID #0 seems to run at half the thruput of the other on ID #1. I'm > * guessing one had internal caching enabled. The answer is mostly acedemic > * now as those drives are elsewhere, replaced with an ST32550N, but now I'd > * like to know more about how my Barracuda is configured. Studying scsi(8) > * suggests that is where the key is but at this time I know enough to know I > * shouldn't go playing with scsi(8) using random commands. > * > * System is a NexGen PCI-90 with Adaptec 2940. > > Yes, the Barracuda comes without the write cache enabled. However, by > turning on the tagged queuing (options AHC_TAGENABLE), we got the > sequential write speed to get very close to the read speed: > > write read > without tag 3.6 6.9 > with tag 6.2 7.9 > > So we haven't actually looked into how to enable the write cache. > > Please let me know if you find something interesting. > > Satoshi Hello Does anybody know how to or have a utility to turn on the write cache on a barracuda? I have a wide barracuda on a NCR 825 and would like to see if there is any performance difference. Peace, Eric From owner-freebsd-scsi Fri May 3 15:12:47 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id PAA15680 for freebsd-scsi-outgoing; Fri, 3 May 1996 15:12:47 -0700 (PDT) Received: from cs.sunysb.edu (sbcs.sunysb.edu [130.245.1.15]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id PAA15672 for ; Fri, 3 May 1996 15:12:42 -0700 (PDT) Received: from sbgrad9.csdept (sbgrad9.cs.sunysb.edu [130.245.2.29]) by cs.sunysb.edu (8.6.12/8.6.9) with SMTP id SAA18813 for ; Fri, 3 May 1996 18:15:45 -0400 Date: Fri, 3 May 1996 18:15:45 -0400 From: Michael Vernick Message-Id: <199605032215.SAA18813@cs.sunysb.edu> Received: by sbgrad9.csdept (4.1/SMI-4.1) id AA15695; Fri, 3 May 96 18:11:16 EDT To: scsi@freebsd.org Subject: Adaptec 3940 and Scatter/Gather Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The documentation specifies that the 3940 can perform scatter/gather. Does anyone know how to use this feature? Is it Adaptec specific? I didn't see anything about scatter/gather in the Scsi spec. mv From owner-freebsd-scsi Fri May 3 21:28:22 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA08028 for freebsd-scsi-outgoing; Fri, 3 May 1996 21:28:22 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA08020 for ; Fri, 3 May 1996 21:28:15 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id VAA00731; Fri, 3 May 1996 21:27:35 -0700 From: "Rodney W. Grimes" Message-Id: <199605040427.VAA00731@GndRsh.aac.dev.com> Subject: Re: Adaptec 3940 and Scatter/Gather To: vernick@CS.SunySB.EDU (Michael Vernick) Date: Fri, 3 May 1996 21:27:35 -0700 (PDT) Cc: scsi@freebsd.org In-Reply-To: <199605032215.SAA18813@cs.sunysb.edu> from Michael Vernick at "May 3, 96 06:15:45 pm" X-Mailer: ELM [version 2.4ME+ PL11 (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 documentation specifies that the 3940 can perform scatter/gather. As can the aha(1542, 1742, 2742, 2940, 2842), bt(542, 742, 747, 757, 946), etc etc... > Does anyone know how to use this feature? FreeBSD uses it on all the above cards, read the source code to understand how it is used. > Is it Adaptec specific? No, almost all bus master scsi controllers support scatter/gather. > I didn't see anything about scatter/gather in the Scsi spec. It has nothing to do with the scsi bus, it has to do with the fact that you can gather arbitrary blocks of host memory into one write command, or scatter one scsi read command to many memory blocks. This is very advantages when your I/O subsystem runs using physical addresses and the host OS is a virtual memory system. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-scsi Sat May 4 01:54:07 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id BAA13143 for freebsd-scsi-outgoing; Sat, 4 May 1996 01:54:07 -0700 (PDT) Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id BAA13133 for ; Sat, 4 May 1996 01:54:03 -0700 (PDT) Received: from sax.sax.de by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id KAA14260; Sat, 4 May 1996 10:53:55 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id KAA09549; Sat, 4 May 1996 10:53:49 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id KAA21491; Sat, 4 May 1996 10:47:24 +0200 (MET DST) From: J Wunsch Message-Id: <199605040847.KAA21491@uriah.heep.sax.de> Subject: Re: Adaptec 3940 and Scatter/Gather To: scsi@freebsd.org Date: Sat, 4 May 1996 10:47:23 +0200 (MET DST) Cc: vernick@CS.SunySB.EDU (Michael Vernick) Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199605032215.SAA18813@cs.sunysb.edu> from Michael Vernick at "May 3, 96 06:15:45 pm" 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 X-Mailer: ELM [version 2.4ME+ PL15 (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 As Michael Vernick wrote: > The documentation specifies that the 3940 can perform scatter/gather. > Does anyone know how to use this feature? Is it Adaptec specific? I > didn't see anything about scatter/gather in the Scsi spec. If you have it in a FreeBSD machine, it does scatter/gather. :-) Well, it cannot be in the SCSI specs, since it's nothing to care for on the SCSI interface, but it rather refers to the ability to split up a single SCSI transfer into multiple different memory locations on the host bus (as opposed to SCSI bus) interface. This comes really handy in a VM environment, since you can use one scatter/gather segment per page, so you don't need to care for the non-contiguousness of a logical buffer inside the physical memory. -- 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 Sat May 4 14:30:27 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id OAA28357 for freebsd-scsi-outgoing; Sat, 4 May 1996 14:30:27 -0700 (PDT) Received: from hda.com (ip76-max1-fitch.zipnet.net [199.232.245.76]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id OAA28347 for ; Sat, 4 May 1996 14:30:13 -0700 (PDT) Received: (from dufault@localhost) by hda.com (8.6.11/8.6.9) id RAA20783; Sat, 4 May 1996 17:33:19 -0400 From: Peter Dufault Message-Id: <199605042133.RAA20783@hda.com> Subject: Re: Buffered Writes? To: ec0@s1.GANet.NET (Eric Chet) Date: Sat, 4 May 1996 17:33:18 -0400 (EDT) Cc: asami@CS.Berkeley.EDU, dkelly@HiWAAY.net, freebsd-scsi@FreeBSD.ORG In-Reply-To: from "Eric Chet" at May 3, 96 03:28:29 pm Reply-to: hdalog@zipnet.net X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > Hello > Does anybody know how to or have a utility to turn on the > write cache on a barracuda? I have a wide barracuda on a NCR 825 and > would like to see if there is any performance difference. You would use the "-e" option of the mode page editor to edit mode page 8 and set the "WCE" (write cache enable) field to 1 to enable write back cacheing. Do "man 8 scsi". This will be similar to the async update of your file system meta data - you will introduce the problem of deferred errors, which presently will give you an error on the wrong disk transfer and treat the failed transfer as successful, and you'll have the possibility of not writing out the buffer on the disk in the case of power failure (unless your drive can write during spindown). -- Temporarily via "hdalog@zipnet.net"... Peter Dufault Real-Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267 From owner-freebsd-scsi Sat May 4 17:56:04 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id RAA22922 for freebsd-scsi-outgoing; Sat, 4 May 1996 17:56:04 -0700 (PDT) Received: from hauki.clinet.fi (root@hauki.clinet.fi [194.100.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id RAA22863 for ; Sat, 4 May 1996 17:56:00 -0700 (PDT) Received: from cantina.clinet.fi (root@cantina.clinet.fi [194.100.0.15]) by hauki.clinet.fi (8.7.5/8.6.4) with ESMTP id DAA18313; Sun, 5 May 1996 03:55:53 +0300 (EET DST) Received: (hsu@localhost) by cantina.clinet.fi (8.7.3/8.6.4) id DAA13322; Sun, 5 May 1996 03:55:53 +0300 (EET DST) Date: Sun, 5 May 1996 03:55:53 +0300 (EET DST) Message-Id: <199605050055.DAA13322@cantina.clinet.fi> From: Heikki Suonsivu To: hdalog@zipnet.net Cc: freebsd-scsi@freebsd.org In-reply-to: Peter Dufault's message of 5 May 1996 00:52:25 +0300 Subject: Re: Buffered Writes? Organization: Clinet Ltd, Espoo, Finland References: <199605042133.RAA20783@hda.com> Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk From: Peter Dufault > Does anybody know how to or have a utility to turn on the > write cache on a barracuda? I have a wide barracuda on a NCR 825 and You would use the "-e" option of the mode page editor to edit mode page 8 and set the "WCE" (write cache enable) field to 1 to enable write back cacheing. Do "man 8 scsi". This will be similar to the async update of your file system meta data - you will introduce the problem of deferred errors, which presently will give you an error on the wrong disk transfer and treat the failed transfer as successful, and you'll have the possibility of not writing out the buffer on the disk in the case of power failure (unless your drive can write during spindown). queue algorithm modifier usually is "write in issue order" so just setting WCE to 1 does not do cause any damage (and nor it usually improves the performance if the load is continous like news disk). But the more important issue is that I still haven't seen a disk which actually *does* have a write cache (pointers welcome :). Easy test is to write data to every second block on the disk. If there is a write cache, you should get more than block / ~70ms (one disk revolution) written, optimally you should get the disk platter speed / 2, but scsi handshake overhead usually makes this impossible. -- Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@clinet.fi mobile +358-40-5519679 work +358-0-4375360 fax -4555276 home -8031121 From owner-freebsd-scsi Sat May 4 23:27:37 1996 Return-Path: owner-freebsd-scsi Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id XAA22956 for freebsd-scsi-outgoing; Sat, 4 May 1996 23:27:37 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id XAA22903 for ; Sat, 4 May 1996 23:27:31 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id WAA01785; Sat, 4 May 1996 22:20:58 -0700 From: "Rodney W. Grimes" Message-Id: <199605050520.WAA01785@GndRsh.aac.dev.com> Subject: Re: Buffered Writes? To: hsu@clinet.fi (Heikki Suonsivu) Date: Sat, 4 May 1996 22:20:57 -0700 (PDT) Cc: hdalog@zipnet.net, freebsd-scsi@freebsd.org In-Reply-To: <199605050055.DAA13322@cantina.clinet.fi> from Heikki Suonsivu at "May 5, 96 03:55:53 am" X-Mailer: ELM [version 2.4ME+ PL11 (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 ... > of power failure (unless your drive can write during spindown). > > queue algorithm modifier usually is "write in issue order" so just setting > WCE to 1 does not do cause any damage (and nor it usually improves the > performance if the load is continous like news disk). But the more > important issue is that I still haven't seen a disk which actually *does* > have a write cache (pointers welcome :). Quantum drives are one of the few that both ship from the factory with WCE set to 1 (since there logic and spin down protection circutry actual work). Your following test should work and yeild patter speed/2 as long as you stay on track when using a Quantum Empire, Capella, Saturn, Atlas or Atlas II series drive (it may apply to other Quantum models as well). Most other vendors drives either don't have a true write cache, or they ship the drive with both WCE turned off and the memory (often called the buffer) dedicated to read operations only. The mode pages to get them to write cache are vendor specific and without the vendors manual for the drive your going to have a very hard time to get it turned on. > Easy test is to write data to > every second block on the disk. If there is a write cache, you should get > more than block / ~70ms (one disk revolution) written, optimally you should Pretty darn slow disk drive there, I think you meant 16mS (3600RPM) or 11.1mS (5400RPM) or perhaps 8mS (7200RPM). > get the disk platter speed / 2, but scsi handshake overhead usually makes > this impossible. SCSI handshake overhead is in the noise when looking at modern scsi disk drives and controllers. Depending on how fast you release the I/O's to the drive you can see a whole track written in 1 revolution of the disk when writting all the even sectors followed by all the odd sectors. To do this test correctly you must take into account the ZBR nature of the drive and the cache in the drive must be at least 1 track in size (they usually are closer to 1 cylinder in size total, and depending on how much is segmented for write purposes, etc.). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD