Re: dmidecode output, as requested (MacBook1,1)

[Len Brown <lenb@kernel.org>]

On Saturday 19 January 2008 09:55, Henk Koster wrote:
> On Friday 18 January 2008 22:48:00 you wrote:
> > 	Manufacturer: Apple Computer, Inc.
> > 	Product Name: MacBook1,1
> >
> > With 2.6.22, did you notice any functional difference with
> > "acpi_osi=!Linux"?
> 
> No difference
> 
> > (or since the default changed at 2.6.23, with 2.6.23 or later
> >  do you notice any difference with "acpi_osi=Linux"?)
> 
> No difference
> 
> > Please send me the output from acpidump.
> 
> Attached, running 2.6.23 now.

OSI(Linux) changes teh value of OSYS:

    OperationRegion (GNVS, SystemMemory, 0x7EED5C10, 0x0100)
    Field (GNVS, AnyAcc, Lock, Preserve)
    {
        OSYS,   16,

...

    Scope (\_SB)
    {
        Method (_INI, 0, NotSerialized)
        {
            If (CondRefOf (_OSI, Local0))
            {
                If (_OSI ("Darwin"))
                {
                    Store (0x2710, OSYS)
                }
                Else
                {
                    If (_OSI ("Linux"))
                    {
                        Store (0x03E8, OSYS)
                    }
                    Else
                    {
                        Store (0x07D1, OSYS)
                    }
                }
            }
            Else
            {
                Store (0x07D0, OSYS)
            }
OSYS 0x7D1 enables the UI bit in HPET _STA (Linux doen't care)
and is used by the OS=Darwin test method:

  Method (OSDW, 0, NotSerialized)
    {
        If (LEqual (OSYS, 0x2710))
        {
            Return (0x01)
        }
        Else
        {
            Return (0x00)
        }
    }

which treats Linux and Windows the same,
so OSI(Linux) is a NOP on the MacBook1,1

thanks,
-Len