[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] x86/AMD: Add support for AMD's OSVW feature in guests

>>> On 19.01.12 at 16:22, Boris Ostrovsky <boris.ostrovsky@xxxxxxx> wrote:
> On 01/19/12 03:10, Jan Beulich wrote:
>> Btw., one more comment on the change to init_amd(): You will likely
>> need to distinguish BP and AP cases here - on the BP you want to
>> plainly write the values read from the MSRs to the global variables,
>> but on APs (with possibly different settings) you need to work
>> towards a setting of the global variables that apply to all of the
>> CPUs. This is not just for the (theoretical only?) hotplug case, but
>> particularly the one of a soft-offlined CPU that had its microcode
>> updated already in a way affecting the OSVW bits and is
>> subsequently being brought back online.
> Not sure I follow you. There is no difference in how BIOS/microcode sets 
> OSVW bits on AP or BP, that's based on silicon version (i.e. stepping) 
> and not where that silicon is plugged in.

But mixed stepping systems could have values in an AP that are
different from what the BP (and earlier APs) had.

> If a core is offlined that will presumably happen after it has already 
> gone through init_amd() and therefore global versions of the registers 
> already accounted for that processor's values. (The downside is that if 
> the offlined processors has more bugs than the rest then everyone is 
> considered as bad even though the bad processor is out of the action. 
> But I think the positives of updating global variables after a cpu is 
> offlined are not worth the complexity.)

Yes, optimizing for this case is very unlikely to be necessary.


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.