Bug reports requested - acpi

[Roman Kurakin]

Nate Lawson wrote:

> Roman Kurakin wrote:
>
>> Nate Lawson wrote:
>>
>>> Roman Kurakin wrote:
>>>
>>>> Nate Lawson:
>>>>
>>>>> set hint.apic.0.disabled="1" at the loader prompt 
>>>>
>>>>
>>>> I'll try this tomorrow since I am going to sleep now. I hope it 
>>>> will reboot.
>>>> If not I'll try to get to the work to fix its state.
>>>
>>>
>>> I have not heard back from you about whether disabling the APIC (not 
>>> ACPI) alone fixes the problem.
>>
>>
>>
>> My answer was:
>>
>>    I was unaware of all that safe mode turns off, and than I last 
>> time check this I ovelooked
>>    that apic is also disabled. It seems that problem with APIC.
>>
>>    But could you tell me how changes in ACPICA affect APIC code?
>>
>> Sorry if it was not very clean. Yes, disabling apic helps.
>
>
> Please answer yes or no:  ACPI enabled but APIC disabled works 100%? 
> Safe mode disables both ACPI and APIC so it's not a good test of where 
> the problem is.  You should probably also test with ACPI disabled but 
> APIC enabled ("set hint.acpi.0.disabled=1").

default both enabled, so :
acpi.disabled - do not work
apic.disabled - do work

> The APIC code has two methods of enumerating PIC devices:  ACPI and MP 
> table.  With ACPI enabled, the APIC code gets its info from the MADT 
> table.  With ACPI disabled, it uses the MP table.  Many older (< 2001) 
> systems have problems with their MADT but an ok MP table.  I think 
> there are a few systems that have a correct MADT but flawed MP table 
> but that is rare.

>
> Thanks,