Date: Sun, 5 Jul 2009 16:42:36 GMT From: Rene Schickbauer <cavac@magicbooks.org> To: freebsd-gnats-submit@FreeBSD.org Subject: i386/136356: SATA / ICH7 problems Message-ID: <200907051642.n65GgaN1084690@www.freebsd.org> Resent-Message-ID: <200907051650.n65Go2C4004476@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 136356 >Category: i386 >Synopsis: SATA / ICH7 problems >Confidential: no >Severity: critical >Priority: high >Responsible: freebsd-i386 >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Jul 05 16:50:02 UTC 2009 >Closed-Date: >Last-Modified: >Originator: Rene Schickbauer >Release: 8-current >Organization: >Environment: FreeBSD titan 8.0-CURRENT-200906 FreeBSD 8.0-CURRENT-200906 #1: Tue Jun 30 13:29:40 CEST 2009 root@titan:/usr/obj/usr/src/sys/GENERIC i386 >Description: Writing to a SATA disk on an ICH7M SATA 150 Controller brings the controller into an inaccessible/unstable state. In my EEE PC 901, i haved changed the second SSD disk from the Asus delivered one to a Patriot 32GB SSD. Installing FreeBSD on this disk and working with it seems fine. Accessing the build-in 4GB SSD by writing makes that controller inaccesible, sometimes even crashing the whole system. After rebooting, in some cases even the BIOS does not see the controller anymore - i have to remove AC cord and battery for a minute in Order to restore the system. The hardware itself is working ok, hours of stresstesting under windows do not seem to trigger that problem, while even a simple newfs under FreeBSD can crash the computer. >From what i have been able to see in the kernel (i'm not really good at kernel debugging...), there could be a problem that the ICH7 does not react properly to a reset command or the reset command itself is faulty. This seems to trigger on certain timeout conditions, when the slow SSD/Flashchips/Chipset sends a soft error because it is not quite able to keep up and need one or more commands resend. >How-To-Repeat: Get an EEE PC (or a Soekris Net5501 with a slow SATA disk - was able to reproduce the error there, too) and write to a SATA disk. >Fix: >Release-Note: >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200907051642.n65GgaN1084690>