From owner-freebsd-stable@FreeBSD.ORG Mon May 21 22:45:24 2012 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 41F681065670 for ; Mon, 21 May 2012 22:45:24 +0000 (UTC) (envelope-from matheus@eternamente.info) Received: from phoenix.eternamente.info (phoenix.eternamente.info [109.169.62.232]) by mx1.freebsd.org (Postfix) with ESMTP id E08248FC1A for ; Mon, 21 May 2012 22:45:23 +0000 (UTC) Received: by phoenix.eternamente.info (Postfix, from userid 80) id 55B0F1CC48; Mon, 21 May 2012 22:39:03 -0300 (BRT) Received: from 187.114.205.80 (SquirrelMail authenticated user matheus) by eternamente.info with HTTP; Mon, 21 May 2012 22:39:03 -0300 Message-ID: Date: Mon, 21 May 2012 22:39:03 -0300 From: "Nenhum_de_Nos" To: freebsd-stable@freebsd.org User-Agent: SquirrelMail/1.4.21 MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal Subject: siiis + port multiplier + 9.0R X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 May 2012 22:45:24 -0000 hail, I have a problem using a port multiplier on 9.0R: pci SATA card: siis0@pci0:5:0:0: class=0x010400 card=0x71241095 chip=0x31241095 rev=0x01 hdr=0x00 vendor = 'Silicon Image, Inc.' device = 'SiI 3124 PCI-X Serial ATA Controller' class = mass storage subclass = RAID multiplier: pmp0 at siisch1 bus 0 scbus1 target 15 lun 0 pmp0: ATA-0 device pmp0: 300.000MB/s transfers (SATA 2.x, NONE, PIO 8192bytes) pmp0: 5 fan-out ports FreeBSD: FreeBSD lamneth 9.0-RELEASE FreeBSD 9.0-RELEASE #0: Tue Jan 3 07:46:30 UTC 2012 root@farrell.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC amd64 the error is: dmesg: May 21 22:20:45 lamneth kernel: siisch1: ... waiting for slots 40000000 May 21 22:20:45 lamneth kernel: siisch1: Timeout on slot 30 May 21 22:20:45 lamneth kernel: siisch1: siis_timeout is 00040000 ss 40812000 rs 40812000 es 00000000 sts 801e2000 serr 00000000 May 21 22:21:17 lamneth kernel: siisch1: Timeout on slot 26 May 21 22:21:17 lamneth kernel: siisch1: siis_timeout is 00040000 ss 64400000 rs 64400000 es 00000000 sts 801e2000 serr 00000000 May 21 22:21:17 lamneth kernel: siisch1: ... waiting for slots 60400000 May 21 22:21:17 lamneth kernel: siisch1: Timeout on slot 22 May 21 22:21:17 lamneth kernel: siisch1: siis_timeout is 00040000 ss 64400000 rs 64400000 es 00000000 sts 801e2000 serr 00000000 May 21 22:21:17 lamneth kernel: siisch1: ... waiting for slots 60000000 May 21 22:21:47 lamneth kernel: siisch1: Timeout on slot 29 May 21 22:21:47 lamneth kernel: siisch1: siis_timeout is 00040000 ss 64400000 rs 64400000 es 00000000 sts 801e2000 serr 00000000 May 21 22:21:47 lamneth kernel: siisch1: ... waiting for slots 40000000 May 21 22:21:47 lamneth kernel: siisch1: Timeout on slot 30 May 21 22:21:47 lamneth kernel: siisch1: siis_timeout is 00040000 ss 64400000 rs 64400000 es 00000000 sts 801e2000 serr 00000000 using the card without the multiplier, gets me around 7MB/s for ZFS+Geli. Using both, I get a great speed (almost this) and then I begin to see those messages above, and copy speed drops to hundreads of KB/s. I've used this multiplier with this card: ahci1@pci0:13:0:0: class=0x010601 card=0x10601b21 chip=0x06121b21 rev=0x01 hdr=0x00 vendor = 'ASMedia Technology Inc.' class = mass storage subclass = SATA on another box, no error. but the performance is too low using geli (soekris net6501-70). any reason for this, is a bug ? I'm looking forward on buying this: http://www.hwtools.net/Adapter/PM362.html or http://www.hwtools.net/adapter/pm2c.html, if the first would support multiplier, then all is fine. else, the latter would let me use the eSATA pcie card I use on soekris, and I think all would be fine also. if anyone would have any words of advice. My goal is to have a FreeBSD capable of ZFS, geli and lots of disks. My mainboard is intel d525mw and 4GB ram. so far, just using 4 disks performance is what I need. thanks, matheus -- We will call you Cygnus, The God of balance you shall be A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? http://en.wikipedia.org/wiki/Posting_style