[Xen-ia64-devel] FW: Xen and the Art of Linux/ia64 Virtualization

["Magenheimer, Dan (HP Labs Fort Collins)" <dan.magenheimer@xxxxxx>]

Hi all --

(In case you aren't a member of the linux-ia64 mailing list,
for some months I've been developing and posting a patch [1]
for Linux/ia64 that prepares a foundation such that it will be
easy to add Xen (or any VMM) transparent paravirtualization
in future releases.)

I submitted the latest "final" Linux/ia64 paravirtualization
patch the day after Linux 2.6.13 was released, in hope of
getting it in for 2.6.14.  Alas, it was not to be.

Though Arun is likely no longer following this list, I thought
it would still be a good forum for discussion about the
concerns Arun raised [2] about the adequacy of instruction
level virtualization on ia64.  I am aware of at least
two other ia64 VMMs (vBlades and vNUMA) in addition to Xen
that heavily utilize instruction level virtualization, and
Intel VT technology is also very similar.  However, there
may be other ia64 VMMs of which I am not aware.

Comments and opinions? 

Hopefully, the Xen/x86 team will continue to make progress
on getting their patches into Linux, which should resolve
Tony's other concern...

Thanks!
Dan

[1] http://www.gelato.unsw.edu.au/archives/linux-ia64/0508/15089.html
[2] http://www.gelato.unsw.edu.au/archives/linux-ia64/0508/15025.html
and: http://www.gelato.unsw.edu.au/archives/linux-ia64/0508/15028.html


> -----Original Message-----
> From: Luck, Tony [mailto:tony.luck@xxxxxxxxx] 
> Sent: Sunday, September 11, 2005 2:49 PM
> To: Magenheimer, Dan (HP Labs Fort Collins)
> Subject: RE: Xen and the Art of Linux/ia64 Virtualization
> 
> >Sorry to be a nag, but Keith's continued references to 2.6.14-rc1
> >being any minute now are making me nervous.
> >
> >Is there anything you need from me to make the virtualization
> >patch acceptable for 2.6.14?  Or should I just shut up and
> >be patient :-}
> 
> The alleged cut-off time for new stuff for 2.6.14 is in about
> eight hours now.
> 
> I'm not going to be merging the Xen patches this time around.  The
> primary open issue is one that Arun raised that I don't think you
> answered: Whether instruction level virtualization is adequate. I'm
> also waiting to see how the integration in other architectures goes.
> I don't see that there is a big rush to push this into the base.
> 
> -Tony
> 

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel