Re: 2.6.13-rc6-git10 test report [x86_64](WITHOUT NVIDIA MODULE)

[Sean Bruno <sean.bruno@dsl-only.net>]

On Fri, 2005-08-19 at 08:59 -0700, Peter Buckingham wrote:
> Andi Kleen wrote:
> > On Fri, Aug 19, 2005 at 08:52:15AM -0700, Peter Buckingham wrote:
> > 
> >>Andi Kleen wrote:
> >>
> >>>At least his original error message can only happen when  CONFIG_GART_IOMMU
> >>>is disabled.
> >>>
> >>>PCI-DMA:  More that 4GB of RAM and no IOMMU                                
> >>>PCI-DMA:  32bit PCI IO may malfunction.<6>PCI-DMA:  Disabling IOMMU        
> >>
> >>Yeah, I agree. In the later dmesgs, though, it seems to be enabled.
> > 
> > 
> > Those don't show any failure.
> 
> no they don't. basically it just says your bios hasn't configured enough 
> IOMMU space, so the kernel is going to do it anyway. it's really just a 
> warning or an fyi rather than an error. i may have chosen the word 
> poorly ;-)
> 
> in short Sean, this isn't a big deal. you only really need to change 
> this if you want to remove a warning from your dmesg output.
> 
> peter

Well, there doesn't appear to be any reference to a setting in my BIOS
for this size(IOMMU).  So I don't think that I can change it!  :(

The machine is working quite a bit better with pci=noacpi in leu of
disabling ACPI in the BIOS, but there are still those nasty errors in
reference to the ACPI tables being broken:
    ACPI-0362: *** Error: Looking up [\_SB_.PCI0.LNK0] in namespace,
AE_NOT_FOUND
search_node ffff8101428572c0 start_node ffff8101428572c0 return_node
0000000000000000

And this one about the 8254 timer:
..MP-BIOS bug: 8254 timer not connected to IO-APIC

And finally, I think that something else kind of wierd is happening with
the on-board sensors.  lm_sensors is having trouble detecting the fan
speeds and temperatures of the main board, but I will take that up with
their developers.

Sean