Re: [Fwd: Re: [RFC] XI Shadow Page Table Mechanism]

[Anthony Liguori <aliguori@us.ibm.com>]

Robert Phillips wrote:
>
>     This isn't supported currently?  Since an HVM must go through 16
>     bit, 32
>     bit, and 64 bit mode to boot up, how can we start more than one
>     guest at
>     a time currently if this doesn't already work?
>
>
> Ed Smith's daily test results show there have been problems with SMP.  
> We haven't diagnosed these problems but, from reading the pre-XI 
> shadow code, it's not clear how it copes with multiple VCPUs in the 
> same domain running in different modes.

There's only a short period of time when this would be happening right?  
During boot up?

>     - how do you deal with large pages within the hypervisor?  do you
>     coalesce or just hope there is contiguous pages available? 
>
>
> We just hope that contiguous pages are available.  However we allocate 
> pages for the guest in large extents to maximize this likelihood.  In 
> practice this is very effective for guests created soon after boot 
> time.  There may be fragmentation problems later.

Any ideas about how to deal with this long term?  Large page support 
would also be useful for PV domains.  There was a lot of people at the 
last summit that were interested in this...

Thanks for the responses,

Anthony Liguori

>     - what is the performance benefit in saving the shadow pages for each
>     domain?  there's clearly a memory trade-off here so understanding the
>     performance gain seems important.
>
>
> The performance benefit appears to be substantial but we have not done 
> a thorough study yet.
>
>     - OOM can be dealt with in the existing code by just invalidating
>     existing mappings to free up pages.  what advantages do your approach
>     have to this?  (i realize we don't do this today but in theory, we
>     could).
>
>
> Ultimately XI deals with OOM  by tearing down cached shadow pages, 
> just as you say.  But it uses LRU to pick the victims.
>
>     Interesting stuff.  I'm eager to see the code.
>
>     Regards,
>
>     Anthony Liguori
>
>     > Thanks,
>     > -b
>     >
>     >
>     ------------------------------------------------------------------------
>     >
>     > _______________________________________________
>     > Xen-devel mailing list
>     > Xen-devel@lists.xensource.com <mailto:Xen-devel@lists.xensource.com>
>     > http://lists.xensource.com/xen-devel
>
>
>     --
>     ------------------------------------------------------------------------
>     Ben Thomas                                         Virtual Iron
>     Software
>     bthomas@virtualiron.com
>     <mailto:bthomas@virtualiron.com>                            Tower
>     1, Floor 2
>     978-849-1214                                       900 Chelmsford
>     Street
>                                                         Lowell, MA 01851
>
> ------------------------------------------------------------------------
> Robert S. Phillips                          Virtual Iron Software
> rsp.vi.xen@virtualiron.com 
> <mailto:rsp.vi.xen@virtualiron.com>             Tower 1, Floor 2
> 978-849-1220                                 900 Chelmsford Street
>                                                     Lowell, MA 01851