Hello, I was just reading Stefan's API wiki page and something dawned on me. For the i82801xx, the LPC GPIO Base address is set to 0x500. According to Stefan's API wiki page the coreboot table is also at 0x500. The output from the kernel indicates a conflict at 0x500? Could this be causing a conflict???
Kernel: PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO PCI quirk: region 0500-053f claimed by ICH4 GPIO
Hi,
You are mixing up memory and IO ports memory. IO ports have their own address space accessible with inb/outb.
Rudolf
On Wed, 18 Jun 2008 18:12:48 +0200, Rudolf Marek r.marek@assembler.cz wrote:
Hi,
You are mixing up memory and IO ports memory. IO ports have their own address space accessible with inb/outb.
Ah, duh, your right. But what could be causing this then?
Kernel: PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO PCI quirk: region 0500-053f claimed by ICH4 GPIO
On Wed, Jun 18, 2008 at 12:33:32PM -0400, Joseph Smith wrote:
Ah, duh, your right. But what could be causing this then?
Kernel: PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO PCI quirk: region 0500-053f claimed by ICH4 GPIO
Can't say without more kernel message context.
//Peter
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Peter Stuge wrote:
On Wed, Jun 18, 2008 at 12:33:32PM -0400, Joseph Smith wrote:
Ah, duh, your right. But what could be causing this then?
Kernel: PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO PCI quirk: region 0500-053f claimed by ICH4 GPIO
Can't say without more kernel message context.
Well that the PNP subsytem needs to be aware on non-std bars which claim this IO. Imagine if PCI IO allocator in linux will for some non-assigned resource assign same region as above. Well this wont happen because PCI io usually starts at 1000 but in general it might be above 1000 too. I think this was the reason for this quirk. - The ACPI pnp might claim this region too and in fact ACPI should do that. Even more specially this quirk is here mostly when ACPI fails to do so and the IO is above 1000 ;)
You may ignore this as "mostly harmless". No really it is harmless for you.
Thanks, Rudolf
-----Original Message----- From: coreboot-bounces@coreboot.org [mailto:coreboot-bounces@coreboot.org] On Behalf Of Rudolf Marek Sent: Wednesday, June 18, 2008 6:42 PM To: Coreboot Subject: Re: [coreboot] coreboot table and i82801xx GPIO conflicts
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Peter Stuge wrote:
On Wed, Jun 18, 2008 at 12:33:32PM -0400, Joseph Smith wrote:
Ah, duh, your right. But what could be causing this then?
Kernel: PCI quirk: region 0400-047f claimed by ICH4 ACPI/GPIO/TCO PCI quirk: region 0500-053f claimed by ICH4 GPIO
Can't say without more kernel message context.
Well that the PNP subsytem needs to be aware on non-std bars which claim this IO. Imagine if PCI IO allocator in linux will for some non-assigned resource assign same region as above. Well this wont happen because PCI io usually starts at 1000 but in general it might be above 1000 too. I think this was the reason for this quirk. - The ACPI pnp might claim this region too and in fact ACPI should do that. Even more specially this quirk is here mostly when ACPI fails to do so and the IO is above 1000 ;)
You may ignore this as "mostly harmless". No really it is harmless for you.
Thanks Rudolf that makes sense. That would explain it because I also get this message from coreboot " Disabling local apic...done." (Although I have no idea why coreboot disables it?). And this from the kernel:
Local APIC disabled by BIOS -- you can enable it with "lapic"
You can check out the boot log here to see what I mean: http://www.coreboot.org/pipermail/coreboot/2008-March/032221.html
Thanks, Joseph Smith Set-Top-Linux www.settoplinux.org
-
Joseph Smith -
Peter Stuge -
Rudolf Marek