From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Williamson Subject: RE: lost thermal zones on 20040715 nc6000 Date: Tue, 24 Aug 2004 19:51:43 -0600 Sender: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Message-ID: <1093398703.2314.4.camel@localhost.localdomain> References: Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Errors-To: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: "Li, Shaohua" Cc: eric.valette-GANU6spQydw@public.gmane.org, Karol Kozimor , "Moore, Robert" , acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: linux-acpi@vger.kernel.org In case you didn't see, Linus posted a trivial patch that avoids the problem for me: http://marc.theaimsgroup.com/?l=linux-kernel&m=109339448319880&w=2 I still don't think we have any right to expose the hidden SMBus, it's just asking for trouble, but... Alex On Wed, 2004-08-25 at 09:27 +0800, Li, Shaohua wrote: > I basically agree ACPI should access the region exclusively. The patch > in Bug 3191 allows both SMBus driver and ACPI thermal zone driver access > it. But there is a risk, we have no way to synchronize the access. Only > one driver can access it one time. But somebody will argue SMBus driver > has more benefit. > > Thanks, > Shaohua > > >-----Original Message----- > >From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel- > >admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Alex Williamson > >Sent: Wednesday, August 25, 2004 5:24 AM > >To: Moore, Robert > >Cc: acpi-devel > >Subject: RE: [ACPI] lost thermal zones on 20040715 nc6000 > > > > > > I've nearly gotten to the bottom of this problem. It's not an > >interpreter bug, nor is it a BIOS bug. The address we were trying to > >read in AML is a SMBus controller address. This address sits on the > >hidden SMBus function in the Intel chipset. Unfortunately, the PCI > >quirks routine un-hides the devices. So far this is all the same as > the > >way it behaved prior to the 20040715 code drop. > > > > The new piece is the motherboard driver. Now this driver comes in > an > >claims the I/O port region that the SMBus controller is using. We've > >un-hidden the PCI function, so PCI wants that resource range. It can't > >get it so it moves the SMBus BAR to a different address. The AML is > now > >hosed cause it's hard-coded to use the address it programmed in (why > >shouldn't it, the device is hidden). Why are we un-hiding the SMBus > >function? The AML sets up an OpRegion for the address space it uses, > so > >should have exclusive access to that range. Suggestions? Thanks, > > > > Alex > > ------------------------------------------------------- SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33 Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift. http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285