From owner-freebsd-scsi Mon Apr 17 14:56:05 1995 Return-Path: freebsd-scsi-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA25164 for freebsd-scsi-outgoing; Mon, 17 Apr 1995 14:56:05 -0700 Received: from cs.sunysb.edu (sbcs.sunysb.edu [130.245.1.15]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id OAA25157 for ; Mon, 17 Apr 1995 14:56:03 -0700 Received: from syslab11.csdept (syslab11.cs.sunysb.edu [130.245.14.11]) by cs.sunysb.edu (8.6.11/8.6.9) with SMTP id RAA29886 for ; Mon, 17 Apr 1995 17:55:58 -0400 Date: Mon, 17 Apr 1995 17:55:58 -0400 From: Michael Vernick Message-Id: <199504172155.RAA29886@cs.sunysb.edu> Received: by syslab11.csdept (4.1/SMI-4.1) id AA04136; Mon, 17 Apr 95 17:52:46 EDT To: freebsd-scsi@FreeBSD.org Subject: SCSI command processing Sender: freebsd-scsi-owner@FreeBSD.org Precedence: bulk I have also done some testing of SCSI overhead processing. On average an 'Inquiry' command sent to the disk over an EISA bus takes 1.2 ms, that is going in and out of user space from start to finish. To retrieve 1 block from the disk cache also takes about the same amount of time, 1.22ms on average. I test this by first doing a read of a random block and then re-read it. The first read incurs the seek, rotation, and transfer overhead, the second read is satisfied by from the disk cache. So the 1.22 ms includes: going in and out of user space, 2 each of: EISA bus transfer, SCSI controller overhead, SCSI bus transfer and disk overhead. Thus, it would seem the disk overhead is quite small. mv From owner-freebsd-scsi Mon Apr 17 21:39:26 1995 Return-Path: freebsd-scsi-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id VAA13943 for freebsd-scsi-outgoing; Mon, 17 Apr 1995 21:39:26 -0700 Received: from mail.barrnet.net (mail.BARRNET.NET [131.119.246.7]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id VAA13789 ; Mon, 17 Apr 1995 21:35:23 -0700 Received: from trout.sri.MT.net (trout.sri.MT.net [204.182.243.12]) by mail.barrnet.net (8.6.10/MAIL-RELAY-LEN) with ESMTP id LAA04398; Fri, 14 Apr 1995 11:39:15 -0700 Received: (from nate@localhost) by trout.sri.MT.net (8.6.11/8.6.11) id MAA29401; Fri, 14 Apr 1995 12:40:20 -0600 Date: Fri, 14 Apr 1995 12:40:20 -0600 From: Nate Williams Message-Id: <199504141840.MAA29401@trout.sri.MT.net> In-Reply-To: Peter Dufault "Re: SCSI target" (Apr 14, 2:09pm) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: Peter Dufault , rgrimes@gndrsh.aac.dev.com (Rodney W. Grimes) Subject: Re: SCSI target Cc: freebsd-hackers@FreeBSD.org, freebsd-scsi@FreeBSD.org Sender: freebsd-scsi-owner@FreeBSD.org Precedence: bulk > > > However, the only host adapter that supports this is the Adaptec > > > 1542B. The 1542C will hang the bus big time if you put it into > > > target mode, so I've locked it out. > > > > Great! But one of my 1542B's bit the dust about a month ago so now > > I only have one of them :-( :-(. > > I only have one also. > > (...) > > If you loose that card and still need one to work with I can arrange > > to put an NCR 53C810 controller in your house on a permanent loan > > basis. > > I'm not planning on adding the support to the firmware - initially, > I could probably use the 1542B more than the 810 for test. If anyone > has a 1542B sitting in a closet (as I did until recently) feel free to > send it. If anyone has a controller which is better than a 1542B for a straight ISA box (No VLB frills), I'd be willing to trade it for my 1542B. I'm not real pleased with the performance out of the one I have, and would certainly be willing to trade up to get better disk performance. (I'm getting about 60% of what I could with my PD2100S) Nate From owner-freebsd-scsi Tue Apr 18 04:34:51 1995 Return-Path: freebsd-scsi-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA01030 for freebsd-scsi-outgoing; Tue, 18 Apr 1995 04:34:51 -0700 Received: from hda.com (hda.com [199.232.40.182]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id EAA01023 for ; Tue, 18 Apr 1995 04:34:47 -0700 Received: (dufault@localhost) by hda.com (8.6.9/8.3) id HAA02991; Tue, 18 Apr 1995 07:34:02 -0400 From: Peter Dufault Message-Id: <199504181134.HAA02991@hda.com> Subject: Re: SCSI command processing To: vernick@cs.sunysb.edu (Michael Vernick) Date: Tue, 18 Apr 1995 07:34:02 -0400 (EDT) Cc: freebsd-scsi@FreeBSD.org In-Reply-To: <199504172155.RAA29886@cs.sunysb.edu> from "Michael Vernick" at Apr 17, 95 05:55:58 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1371 Sender: freebsd-scsi-owner@FreeBSD.org Precedence: bulk Michael Vernick writes: > > I have also done some testing of SCSI overhead processing. On average > an 'Inquiry' command sent to the disk over an EISA bus takes 1.2 ms, > that is going in and out of user space from start to finish. To > retrieve 1 block from the disk cache also takes about the same amount > of time, 1.22ms on average. I test this by first doing a read of a > random block and then re-read it. The first read incurs the seek, > rotation, and transfer overhead, the second read is satisfied by from > the disk cache. So the 1.22 ms includes: going in and out of user > space, 2 each of: EISA bus transfer, SCSI controller overhead, SCSI > bus transfer and disk overhead. Thus, it would seem the disk overhead > is quite small. When I have some spare time I'm going to instrument the upper levels of the code. There is a fair amount of overhead in that software, but I suspect it is drowned by the overhead of the host adapter and the target firmware. The fact that Test Unit Ready takes forever makes me most suspicious of the host adapter firmware, and it would be nice to have a standard way of measuring the host adapter overhead across the different supported adapters. -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267