From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jean Delvare Subject: Re: [lm-sensors] Could the k8temp driver be interfering with ACPI? Date: Wed, 21 Feb 2007 15:54:44 +0100 Message-ID: <20070221155444.99386b98.khali@linux-fr.org> References: <45D5EA88.7090300@redhat.com> <45D6DDCE.5050803@assembler.cz> <45D7461A.2040808@redhat.com> <45D8D6AE.3000204@assembler.cz> <45DB0EEA.2010205@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: Received: from smtp-103-wednesday.nerim.net ([62.4.16.103]:1263 "EHLO kraid.nerim.net" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751288AbXBUOzE (ORCPT ); Wed, 21 Feb 2007 09:55:04 -0500 In-Reply-To: <45DB0EEA.2010205@redhat.com> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Chuck Ebbert Cc: Rudolf Marek , linux-acpi@vger.kernel.org, linux-kernel , lm-sensors@lm-sensors.org Hi Chuck, On Tue, 20 Feb 2007 10:08:26 -0500, Chuck Ebbert wrote: > I blacklisted the k8temp driver (and the out-of-tree k8_edac driver > in Fedora) and the temps were still volatile, so that's not causing > it. Since then I've upgraded the system BIOS from F.06 to F.27 and > the problems _may_ have gone away. My own custom 2.6.19 kernel has > never been a problem, so I'm thinking it's one of these drivers > loaded by Fedora that I never even compile: > > i2c_core i2c-core doesn't touch the hardware by itself. > i2c_ec Presumably autoloaded by the ACPI subsystem, I guess your ACPI implementation includes an SMBus 2.0 controller. > i2c_piix4 i2c-piix4 will autoload if a supported PCI device is found on your system. Assuming this is the same physical bus as i2c_ec is exposing, it's no good to load both i2c-piix4 and i2c_ec at the same time. Unfortunately i2c_ec doesn't request the I/O resources it uses so this kind of conflict cannot be avoided currently. Can you try to load the i2c-dev driver, then run the following commands and report the results: $ i2cdetect -l For each bus listed: $ i2cdetect N (where N is the i2c bus number) > asus_acpi (on a Compaq???) > sbs This is a new battery driver used in conjunction with i2c_ec. I guess you have a smart battery in your laptop which is accessed through the SMBus. I found that this driver bypasses the i2c-core locking, which is really bad. I reported it one week ago: http://marc.theaimsgroup.com/?l=linux-acpi&m=117160531631100&w=2 (for some reason my original post wasn't archived) My patch wasn't applied, but the problems you describe could well be caused by this locking issue. So I suggest that you unload the sbs driver and see if things get better. If they do, you could try to apply my patch and load sbs again, and see if it fixes it. -- Jean Delvare From mboxrd@z Thu Jan 1 00:00:00 1970 From: khali@linux-fr.org (Jean Delvare) Date: Wed, 21 Feb 2007 14:54:44 +0000 Subject: [lm-sensors] Could the k8temp driver be interfering with ACPI? Message-Id: <20070221155444.99386b98.khali@linux-fr.org> List-Id: References: <45D5EA88.7090300@redhat.com> <45D6DDCE.5050803@assembler.cz> <45D7461A.2040808@redhat.com> <45D8D6AE.3000204@assembler.cz> <45DB0EEA.2010205@redhat.com> In-Reply-To: <45DB0EEA.2010205@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Chuck Ebbert Cc: Rudolf Marek , linux-acpi@vger.kernel.org, linux-kernel , lm-sensors@lm-sensors.org Hi Chuck, On Tue, 20 Feb 2007 10:08:26 -0500, Chuck Ebbert wrote: > I blacklisted the k8temp driver (and the out-of-tree k8_edac driver > in Fedora) and the temps were still volatile, so that's not causing > it. Since then I've upgraded the system BIOS from F.06 to F.27 and > the problems _may_ have gone away. My own custom 2.6.19 kernel has > never been a problem, so I'm thinking it's one of these drivers > loaded by Fedora that I never even compile: > > i2c_core i2c-core doesn't touch the hardware by itself. > i2c_ec Presumably autoloaded by the ACPI subsystem, I guess your ACPI implementation includes an SMBus 2.0 controller. > i2c_piix4 i2c-piix4 will autoload if a supported PCI device is found on your system. Assuming this is the same physical bus as i2c_ec is exposing, it's no good to load both i2c-piix4 and i2c_ec at the same time. Unfortunately i2c_ec doesn't request the I/O resources it uses so this kind of conflict cannot be avoided currently. Can you try to load the i2c-dev driver, then run the following commands and report the results: $ i2cdetect -l For each bus listed: $ i2cdetect N (where N is the i2c bus number) > asus_acpi (on a Compaq???) > sbs This is a new battery driver used in conjunction with i2c_ec. I guess you have a smart battery in your laptop which is accessed through the SMBus. I found that this driver bypasses the i2c-core locking, which is really bad. I reported it one week ago: http://marc.theaimsgroup.com/?l=linux-acpi&m7160531631100&w=2 (for some reason my original post wasn't archived) My patch wasn't applied, but the problems you describe could well be caused by this locking issue. So I suggest that you unload the sbs driver and see if things get better. If they do, you could try to apply my patch and load sbs again, and see if it fixes it. -- Jean Delvare