Re: [Xen-devel] [PATCH 0 of 5] v2: Nested-p2m cleanups and locking changes

[Tim Deegan <Tim.Deegan@xxxxxxxxxx>]

At 11:46 +0100 on 27 Jun (1309175170), Tim Deegan wrote:
> This patch series tidies up a few bits ofthe nested p2m code.
> The main thing it does is reorganize the locking so that most of the
> changes to nested p2m tables happen only under the p2m lock, and the
> nestedp2m lock is only needed to reassign p2m tables to new cr3 values.

There are still a few things I'm not convinced about in the nested NPT
code:

 - The function that allocates new nested p2ms probably needs an
   overhaul, as I said in my last email. 
 - The flushing policy is a bit confusing: e.g., what exactly ought to 
   happen when the guest sets the tlb-control bits?  AFAICS the nested-p2ms 
   are already kept in sync with host-p2m changes, and we flush all 
   TLBs when we update nested-p2ms, so can we skip this extra flush?
 - Why is there a 10x increase in IPIs after this series?  I don't see
   what sequence of events sets the relevant cpumask bits to make this
   happen.

Cheers,

Tim.

-- 
Tim Deegan <Tim.Deegan@xxxxxxxxxx>
Principal Software Engineer, Xen Platform Team
Citrix Systems UK Ltd.  (Company #02937203, SL9 0BG)

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