From owner-freebsd-smp Sat Nov 6 13:18:45 1999 Delivered-To: freebsd-smp@freebsd.org Received: from guru.phone.net (guru.phone.net [216.240.39.120]) by hub.freebsd.org (Postfix) with SMTP id 426F214C40 for ; Sat, 6 Nov 1999 13:18:42 -0800 (PST) (envelope-from mwm@phone.net) Received: (qmail 1514 invoked by uid 100); 6 Nov 1999 21:18:42 -0000 From: Mike Meyer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14372.39730.659338.440569@guru.phone.net> Date: Sat, 6 Nov 1999 13:18:42 -0800 (PST) To: freebsd-smp@freebsd.org Subject: forwarded message from Mike Meyer X-Mailer: VM 6.72 under 21.1 (patch 3) "Acadia" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org [This is now *way* off-topic for smp - but where should it go?] sthaug@nethelp.no writes: ;->> > ;->With your typical ;->> > ;->1- or 2-drive system, IDE drives shouldn't be significantly slower than ;->> > ;->SCSI. ;->> > ;->> > My tests show different - see > > http://www.phone.net/home/mwm/disktest/disktest.html >. If there's ;->> > something I can do to improve the performance of the IDE drive, I'd ;->> > love to hear it. ;->> ;->> Did you explicitly configure your kernel to enable DMA? I couldn't ;->> find any mention of that on your page. No, I didn't. I just pulled the line from the GENERIC kernel and used that. The obvious question is why isn't DMA & multi-block read enabled by default in GENERIC? Since the flags indicate a *probe* for those capabilities, the kernel should still work for all systems, right? And most modern system should have controllers/drives that can use these things - and they make a *huge* difference - at least for me. ;->I'd say that the picture is not at all clearcut any more. My tests ;->indicate that SCSI drives still have an edge on EIDE drives for random ;->access, but it's not huge. The CPU usage is lower for SCSI in some ;->cases, but not all. Well, the claim was that for multi-drive systems SCSI still has an advantage. I can't test that one. Not sure what tests to use, either. ;->The DPTA-372730 is one darn fast disk! The raw data for the Maxtor - properly configured - indicate that it's not very far behind. You didn't say whether you had softupdates on. Adding the dmesg output for the drives would probably help. Speaking of which, here's the dmesg on the two drives in my system. Is there anything obvious that would improve the throughput on these drives before I start testing again? Nov 6 12:23:58 guru /kernel: wdc0: unit 0 (wd0): , DMA, 32-bit, multi-block-16 Nov 6 12:23:58 guru /kernel: da0: Fixed Direct Access SCSI-2 device Nov 6 12:23:58 guru /kernel: da0: 40.000MB/s transfers (20.000MHz, offset 15, 16bit), Tagged Queueing Enabled Nov 6 12:23:58 guru /kernel: da0: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C) Thanx,