From owner-aic7xxx Wed Mar 17 5:27:46 1999 Delivered-To: aic7xxx@freebsd.org Received: from aloha.cc.columbia.edu (aloha.cc.columbia.edu [128.59.59.134]) by hub.freebsd.org (Postfix) with ESMTP id A0F3214ED1 for ; Wed, 17 Mar 1999 05:27:42 -0800 (PST) (envelope-from amlai@columbia.edu) Received: from localhost (amlai@localhost) by aloha.cc.columbia.edu (8.8.5/8.8.5) with SMTP id IAA06938; Wed, 17 Mar 1999 08:27:16 -0500 (EST) Date: Wed, 17 Mar 1999 08:27:15 -0500 (EST) From: Albert Max Lai Reply-To: Albert Max Lai To: "christian.klein" , nelson@seahunt.imat.com Cc: aic7xxx@FreeBSD.ORG Subject: Re: aic7895, IBM 9ZX, Quantum Viking, slow transfer rates In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-aic7xxx@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 16 Mar 1999, christian.klein wrote: As suggested, I have increased the file size to 512MB. I still see some strange results. Reading is indeed significantly faster on the 9ZX, but significantly slower on 2 of the 3 tests than the Viking. My CPU is dual PII 350. As far as I can tell, I do not have any IRQ sharing. And compared to Mr. Nelson's 9LZX, my drive is significantly slower (although the 9LZX is supposed to be slightly faster than the 9ZX though). Also, looking at the benchmark results from under NT4, my performance should be closer to Mr. Nelson's despite being slower than the LZX model. I've also checked the settings in the controller configuration as well as my bios and have not found anything to change. I have also updated my aic7xxx driver to the 5.1.11 version, but that did not change the results either. If anyone has any further suggestions, I would be greatful. (Also, if anyone happens to actually have a 9ZX, I'd be interested in seeing your results.) Thanks in advance! IBM 9ZX Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done... -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 512 3797 58.4 5934 10.0 3577 12.1 5933 82.5 10211 14.8 99.9 4.1 Quantum Viking Seeker 1...Seeker 3...Seeker 2...start 'em...done...done...done... -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 512 4291 65.8 7699 13.0 3017 10.4 5948 82.5 7514 10.8 66.1 2.3 From /proc/aic7xxx/0: Adaptec AIC7xxx driver version: 5.1.11/3.2.4 Compile Options: AIC7XXX_RESET_DELAY : 5 AIC7XXX_TAGGED_QUEUEING: Adapter Support Enabled Check below to see which devices use tagged queueing AIC7XXX_PAGE_ENABLE : Enabled (This is no longer an option) AIC7XXX_PROC_STATS : Disabled Adapter Configuration: SCSI Adapter: Adaptec AIC-7895 Ultra SCSI host adapter Ultra Wide Controller Channel A PCI MMAPed I/O Base: 0xfebfa000 Adapter SEEPROM Config: SEEPROM found and used. Adaptec SCSI BIOS: Enabled IRQ: 16 SCBs: Active 0, Max Active 2, Allocated 15, HW 32, Page 255 Interrupts: 203116 BIOS Control Word: 0x18a6 Adapter Control Word: 0x005f Extended Translation: Enabled Disconnect Enable Flags: 0xffff Ultra Enable Flags: 0x0003 Tag Queue Enable Flags: 0x0000 Ordered Queue Tag Flags: 0x0000 Default Tag Queue Depth: 8 Tagged Queue By Device array for aic7xxx host instance 0: {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} Actual queue depth per device for aic7xxx host instance 0: {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} Statistics: (scsi0:0:0:0) Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 Transinfo settings: current(12/8/1), goal(12/8/1), user(12/8/1) Total transfers 146247 (95376 reads and 50871 writes) (scsi0:0:1:0) Device using Narrow/Sync transfers at 20.0 MByte/sec, offset 15 Transinfo settings: current(12/15/0), goal(12/15/0), user(12/8/1) Total transfers 56718 (38687 reads and 18031 writes) (scsi0:0:5:0) Device using Narrow/Sync transfers at 4.4 MByte/sec, offset 15 Transinfo settings: current(51/15/0), goal(12/15/0), user(12/8/1) Total transfers 0 (0 reads and 0 writes) > 100 mb with bonnie is not a representative benchmark of _disk_ speed. use > 4 x your amount of physical memory. > > but, i have the same viking drive and a 4 gig narrow cheetah and both do > better than your benchmarks (p2-266/tyan/440FX). Have > you looked at your bios settings? is your 7895 sharing an irq with > something else? what does /proc/scsi/aic7xxx/* say?? How fast is your > processor? > > christian > > > > On Tue, 16 Mar 1999, Albert Max Lai wrote: > > > For some reason, my IBM DGVS09U (10020RPM 9ZX UW 9GB) is performing > > exactly the same as my Quantum Viking 4.5 NSE (7200RPM Narrow 4.5GB). I > > have a aic7895 on a Tyan S1836DLUAN, running the 2.2.3 kernel, with the > > 5.1.10 of the driver. They are both on bus 0 of the controller. Everything > > gets detected properly at boot time, and the output from hdparm seems to > > make sense. (Below are outputs from various ways of benchmarking the two > > drives.) Is the following what I should be seeing? (I don't think that I > > should though.) Thanks in advance. > > > > P.S. Doug, if you are reading this, thanks for a great driver! > > > > >From the 9ZX: > > [root@amlai2 misc]# /sbin/hdparm -tT /dev/sda > > > > /dev/sda: > > Timing buffer-cache reads: 64 MB in 0.66 seconds =96.97 MB/sec > > Timing buffered disk reads: 32 MB in 2.02 seconds =15.84 MB/sec > > > > >From the Viking: > > [root@amlai2 misc]# /sbin/hdparm -tT /dev/sdb > > > > /dev/sdb: > > Timing buffer-cache reads: 64 MB in 0.69 seconds =92.75 MB/sec > > Timing buffered disk reads: 32 MB in 3.12 seconds =10.26 MB/sec > > > > Timing buffered disk reads look right. > > > > > In actuality, when doing various bencharks, the IBM drive comes out > > to being approximately the same speed, if not sometimes slower: > > > > On the IBM 9ZX: > > File './Bonnie.5138', size: 104857600 > > Writing with putc()...done > > Rewriting...done > > Writing intelligently...done > > Reading with getc()...done > > Reading intelligently...done > > Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done... > > -------Sequential Output-------- ---Sequential Input----Random-- > > -Per Char- --Block--- -Rewrite-- -Per Char- --Block-----Seeks--- > > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > > 100 5934 91.9 9252 13.9 3755 12.1 4181 59.2 11708 18.2 204.6 5.8 > > > > On the Quantum Viking: > > File './Bonnie.5146', size: 104857600 > > Writing with putc()...done > > Rewriting...done > > Writing intelligently...done > > Reading with getc()...done > > Reading intelligently...done > > Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done... > > -------Sequential Output-------- ---Sequential Input----Random-- > > -Per Char- --Block--- -Rewrite-- -Per Char- --Block-----Seeks--- > > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > > 100 6219 95.5 10024 15.4 3065 10.3 5242 74.2 7845 11.1 133.93 3.8 > > > > The 9ZX test seems slower than it should be. > > > > Just for kicks, I did the following as a really idiotic benchmark: > > On the 9ZX: > > [root@amlai2 amlai]# time dd if=/dev/zero of=zeroes bs=1024k count=100 > > 100+0 records in > > 100+0 records out > > 0.01user 1.67system 0:11.71elapsed 14%CPU (0avgtext+0avgdata > > 0maxresident)k > > 0inputs+0outputs (88major+13minor)pagefaults 0swaps > > > > On the Viking: > > [root@amlai2 misc]# time dd if=/dev/zero of=zeroes bs=1024k count=100 > > 100+0 records in > > 100+0 records out > > 0.00user 1.87system 0:10.03elapsed 18%CPU (0avgtext+0avgdata > > 0maxresident)k > > 0inputs+0outputs (91major+13minor)pagefaults 0swaps > > > > > > -- > > Albert Lai 1018D East Campus > > Residential Computer Consultant 411 W. 116th St. > > Columbia University New York, NY 10027 > > http://www.columbia.edu/~aml61 (212)853-4854 > > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe aic7xxx" in the body of the message > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe aic7xxx" in the body of the message > -- Albert Lai 1018D East Campus Residential Computer Consultant 411 W. 116th St. Columbia University New York, NY 10027 http://www.columbia.edu/~aml61 (212)853-4854 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe aic7xxx" in the body of the message