Mark.Haywood wrote:

> Li, Aubrey wrote:
>> Li, Aubrey wrote:
>> 
>> 
>>>> Mark Haywood <> wrote:
>>>> 
>>>>> Mark Haywood wrote:
>>>>> 
>>>>>> Li, Aubrey wrote:
>>>>>> 
>>>>>>> Hi Andrei,
>>>>>>> 
>>>>>>> Andrei Dorofeev wrote:
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>>> Aubrey,
>>>>>>>> 
>>>>>>>> What version of the X8DTN BIOS are you using?  I have filed
>>>>>>>> this issue back in February on premier.intel.com and it
>>>>>>>> should've been resolved by now.   Do you see this with BIOS
>>>>>>>> version 3059W? 
>>>>>>>> 
>>>>>>>> 
>>>>>>> Right, right, the version is 3059W! And this is the newest
>>>>>>> version. From the BIOS changelog, they tested windows, redhat
>>>>>>> linux, I'm not sure if they care about what solaris reported, :(
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>>> Don't give up on using _PSD because of this.
>>>>>>>> 
>>>>>>>> 
>>>>>>> I believe before NHM, _PSD is seldom implemented. Using the
>>>>>>> vendor specific CPU topology should be an acceptable way to
>>>>>>> build domain info. I admit I don't have the knowledge about the
>>>>>>> other CPU vendor, like SPARC and AMD, what's the benefit of
>>>>>>> using _PSD? Othering than introducing panic, ;)
>>>>>>> 
>>>>>>> 
>>>>>> That's a very good question. What is the benefit of the _PSD? I
>>>>>> believe the _PSD was introduced to prevent Solaris and other
>>>>>> operating systems from having to do exactly what you are
>>>>>> proposing (i.e., introducing vendor specific details of
>>>>>> processor state domains into the operating system). The _PSD is
>>>>>> supposed to define a standard way for operating systems to
>>>>>> digest the domain data. Unless there is a really compelling
>>>>>> reason to ignore the _PSD, I would suggest that we continue to
>>>>>> use it. 
>>>>>> 
>>>>> How about a compromise. Since the _PSD often doesn't exist, we
>>>>> have to be able to determine the domains from the topology. Since
>>>>> we have to do this anyway, why don't we digest the _PSD and
>>>>> verify it afterwards using the topology. Why do this? So that we
>>>>> can report bad _PSDs to the BIOS developers so that the _PSDs
>>>>> become more reliable in the future. If a faulty _PSD is
>>>>> identified we can log a message to the console and use domains
>>>>> built from the topology. At some point it would be nice if we
>>>>> didn't have to continue determining the domains using topology.
>>>>> Sound reasonable? 
>>>>> 
>>>> That sounds a good suggestion. In early stage, we may need to
>>>> reveal bugs in BIOS implementation and push BIOS to behave
>>>> correctly. Eventually BIOS should be robust enough so we could
>>>> switch to BIOS. If we give up _PSD, it seems we are on the
>>>> opposite side of 
>>>> ACPI spect then.
>>>> 
>>>> 
>>> BIOS is not ACPI, hehe. If we use topology to check, why not
>>> use it at all?
>>> why we need to verify BIOS bug?  Anyway, the suggestion works to me.
>>> 
>>> Thanks,
>>> -Aubrey
>>> 
>> 
>> Wait a minute, I remember some platform has sub-domain in a package,
>> I wonder how do you want to verify if _PSD is right or not?
>> 
> 
> Are you suggesting that we can't verify the domains using the
> topology? If so, then I don't see how we could create the domains
> from the topology either. BTW, it is this type of complexity (and the
> possibility that the complexity will increase in the future) that
> makes me believe we should try to use the _PSD if possible.
> 

No, I'm not suggesting. 
I also wonder is there a platform that several packages belong to the same 
domain?
I just want to know when we verify the _PSD, can we do some vendor specific more
than cpu topology? I'm worried _PSD becomes wrong after topology checking.

Thanks,
-Aubrey

Reply via email to