From owner-freebsd-hackers Sun Aug 23 19:16:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA00591 for freebsd-hackers-outgoing; Sun, 23 Aug 1998 19:16:51 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from alpha.delta.edu (alpha.delta.edu [161.133.129.3]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id TAA00556 for ; Sun, 23 Aug 1998 19:16:31 -0700 (PDT) (envelope-from dakott@alpha.delta.edu) Received: from pm343-25.dialip.mich.net by alpha.delta.edu; (5.65v3.0/1.1.8.2/06Jan97-0932AM) id AA13944; Sun, 23 Aug 1998 22:20:11 -0400 Received: from kott.my.domain (dakott@kott.my.domain [192.168.0.1]) by kott.my.domain (8.8.8/8.8.5) with SMTP id SAA26790 for ; Fri, 21 Aug 1998 18:58:57 -0400 (EDT) Date: Fri, 21 Aug 1998 18:58:57 -0400 (EDT) From: David Kott To: hackers@FreeBSD.ORG Subject: Adaptec 1542i Performance on -stable [was Re: SCSI Controller] In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > OK. If you can lay your hands on an Adaptec 1542, it should work fine. > It's the only ISA/VLB SCSI controller I have any experience with > (which doesn't mean it's the only good one). ISTR there are a few bugs > in our 1542 support, but I haven't paid any attention to it for some > time so I may be wrong. I've run -stable successfully on a few boxen > equipped with 1542s, but not recently. I am using a 1542 right now on this -stable machine. While I am not particularily overwhelmed by it's performance, I have had no problems that I could definately link to the adapter itself. Perhaps I have not configured it for optimal performance. I did have some strange kernel panics when I installed my hardrive in front of the CDROM that previously had occupied the bus alone. The drive would log timeout errors. Then, as now, my root partition/swap et. al was on this drive, the kernel would panic. Also, sporadically, the drive would load the kernel sucessfully for booting, but would just hang when the kernel attempted to mount the root partition. I configured the hardrive to supply the bus termination, instead of the CDROM, as well as recompiling my kernel sans the "TUNE_1542" option, and I haven't had a problem since. Here are some really coarse performance comparisons for you to ignore. :-) -d root@kott [/root]# dd if=/dev/rsd0a of=/dev/null count=1024 1024+0 records in 1024+0 records out 524288 bytes transferred in 3.619458 secs (144853 bytes/sec) root@kott [/root]# dd if=/dev/rwd2s1a of=/dev/null count=1024 1024+0 records in 1024+0 records out 524288 bytes transferred in 0.173658 secs (3019083 bytes/sec) chip2 rev 0 on pci0:7:1 wdc1 at 0x170-0x177 irq 15 flags 0xa0ffa0ff on isa wdc1: unit 0 (wd2): , 32-bit, multi-block-16 wd2: 6149MB (12594960 sectors), 13328 cyls, 15 heads, 63 S/T, 512 B/S aha0: Rev 41 (AHA-154x[AB]) V0.5, enabling residuals, target ops aha0: reading board settings, dma=7 int=11 id=7 (bus speed defaulted) aha0 at 0x330-0x333 irq 11 drq 7 on isa (aha0:0:0): "QUANTUM FIREBALL ST4.3S 0F0C" type 0 fixed SCSI 2 sd0(aha0:0:0): Direct-Access 4136MB (8471232 512 byte sectors) sd0(aha0:0:0): with 7068 cyls, 6 heads, and an average 199 sectors/track To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message