RE: [Xen-devel] RE: [PATCH 1/2] Vcpu hotplug: Move ACPI processor from \_PR to \_SB

["Jiang, Yunhong" <yunhong.jiang@xxxxxxxxx>]

>-----Original Message-----
>From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
>[mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Keir Fraser
>Sent: Friday, February 12, 2010 4:30 PM
>To: Jiang, Yunhong; Liu, Jinsong; xen-devel
>Subject: Re: [Xen-devel] RE: [PATCH 1/2] Vcpu hotplug: Move ACPI processor from
>\_PR to \_SB
>
>On 12/02/2010 02:57, "Jiang, Yunhong" <yunhong.jiang@xxxxxxxxx> wrote:
>
>> However, according to ACPI spec, the _PR is for ACPI1.0 compatible. We have 
>> no
>> idea which OS is ACPI 1.0 OS. As HeQing found ACPI 1.0 bugs in Win2K, so we
>> assume W2K is ACPI 1.0. We test shows W2K guest is ok with the _SB definition
>> in our testing. Maybe Win98/WinMe is ACPI 1.0, but we have no image for these
>> OS. But yes, that's a main issue for _SB method and we need more 
>> consideration
>> here.
>>
>> In fact, we have internal argue to choose _PR or _SB method before Jinsong's
>> initial patch sent out. Later _PR method is chosen because of the ACPI 1.0
>> compatible benifit, and kernel 30 version is ok. (IIRC, .32 kernel is not
>> released at that time).
>
>Well, that's tricky. 2.6.32 is supposed to be a long-term maintained kernel,
>so presumably there will be a 2.6.32.x along in the not-too-far future which
>fixes this Linux bug? I feel we're a bit close to the wire to make this
>change now.

We talked this issue with our colleague working on kernel side. As one key 
engineer is on vocation, I will get more information when he is back after 
Chinese New Year. 
But I'm not sure of Win2K8.

>
>I'd be a bit more comfortable if we had the cover of lots of other modern
>systems putting their processor objects under \_SB, but actually I've never
>seen one. Then again I haven't been looking at high-end systems supporting
>CPU hotplug and the like.

Yes. I only saw \_SB definition in system supporting CPU hotplug. In fact, in 
that system, the processor is defined under an container object in \_SB. As 
currently all system in our lab is shutdown for CNY, I can't find more system 
to check. And I suspect that we need care \_PR soluation, legacy OS support is 
an important usage model for virtualization.

One thing I noticed in my system is, there is a ACPI version option in my 
desktop system, and I remember I saw that option in other system also. So one 
possible solution is, place all processor definition under a seperated SSDT 
file. An option is provided so that build.c can select different SSDT based on 
user's input. But that make thing tricky still.

--jyh

>
> -- Keir
>
>
>
>_______________________________________________
>Xen-devel mailing list
>Xen-devel@xxxxxxxxxxxxxxxxxxx
>http://lists.xensource.com/xen-devel

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel