From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: Errors with AHCI and 10K Raptor drive Date: Mon, 19 Mar 2007 14:17:14 -0700 Message-ID: <20070319141714.1632cee4@freekitty> References: <20070319075321.444c241e@freebird> <45FEA5D1.20101@pobox.com> <20070319131608.6d93fbf4@freekitty> <45FEF6B0.6020502@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: Received: from smtp.osdl.org ([65.172.181.24]:56716 "EHLO smtp.osdl.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964960AbXCSVRR (ORCPT ); Mon, 19 Mar 2007 17:17:17 -0400 In-Reply-To: <45FEF6B0.6020502@garzik.org> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Jeff Garzik Cc: linux-ide@vger.kernel.org On Mon, 19 Mar 2007 16:46:40 -0400 Jeff Garzik wrote: > Stephen Hemminger wrote: > > On Mon, 19 Mar 2007 11:01:37 -0400 > > Jeff Garzik wrote: > > > >> Stephen Hemminger wrote: > >>> This is a new drive, so I suspect some driver incompatibility between the > >>> driver (AHCI) and the hard drive (WD Raptor 10K 74G). > >> Driver incompatibilities between the driver (LLDD or libata-core) and a > >> hard drive are extremely rare. > >> > >> The most common problem on modern SATA drives is drive firmware bugs, at > >> the moment. > >> > > > > I moved the drive to another machine with ICH7 and it seems to work fine > > there. Strange? > > Does dmesg reflect any difference in AHCI version or PHY speed, on > either machine? Did you use the same cables in both tests? > > AHCI version ("0001.0100" == AHCI 1.1): > ... > ahci 0000:00:1f.2: AHCI 0001.0100 32 slots 4 ports 3 Gbps 0xf impl SATA mode > ... On the good machine: [ 30.728849] ata1: SATA max UDMA/133 cmd 0x000000000001e400 ctl 0x000000000001e082 bmdma 0x000000000001d880 irq 23 [ 30.890918] ata1.00: ATA-7: WDC WD740ADFD-00NLR1, 20.07P20, max UDMA/133 [ 30.890966] ata1.00: 145226112 sectors, multi 16: LBA48 NCQ (depth 0/32) [ 30.900957] ata1.00: configured for UDMA/133 Maybe since AHCI isn't enabled in the BIOS on the good machine, we aren't doing NCQ? > PHY configuration and speed: > ... > ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300) > ... > > SATA headers (ports) on the motherboard and SATA cables are known to be > weak links in the chain, commonly. I tried a port and cable swap on the bad machine. The good machine had a new cable, so it still could be the cable. -- Stephen Hemminger