Re: use of scratchpad registers

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Mon, Mar 25, 2013 at 10:07:22AM +0000, Jan Beulich wrote:
> Konrad,
> 
> in a response on thread "High CPU temp, suspend problem - xen
> 4.1.5-pre, linux 3.7.x", you say
> 
> "There are some BIOSes out there that I know that use the
>  scratchpad registers in IOH (so depending on the platform that
>  can be 0:0e.1 , Reg 0x84). If Xen or Linux touch it then the
>  P-states and C-states that the BIOS generates are buggy",
> 
> which makes me assume you have knowledge on why and where
> Xen or Linux would touch these registers. It is my understanding
> that these registers should be left alone by Xen, and be used by
> Linux only in ways consistent between native and running on top
> of Xen (i.e. through the same specific driver).
> 
> However, we've been having a report of there being a difference
> in behavior, which so far we weren't able to narrow down the
> origin of. Hence, if you know details here, I'd very much
> appreciate if you could share them.

Of course. Let me email the group that had let me know about this
scratchpad registers and see if they narrowed down _who_ was
poking at that. Once I get that information - or if they are
comfortable responding in a public forum, I will post it
here ASAP.


> 
> Thanks, Jan
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
>