Re: [Xen-users] Re: Hardware differences for 2 xen servers?

[Javier Guerra <javier@xxxxxxxxxxx>]

On Wednesday 21 June 2006 5:19 am, Matthew Palmer wrote:
> Seriously?  That sounds like a pretty bad failing of live-migration -- I'd
> *really* expect live migration to be robust against this sort of thing,

the issue is that most OSs (at least Linux does it) detects some features of 
the CPU at startup time, and decides then if it wants to use them or not.  
for example, it might find SSE2 and use it for memcpys; later get migrated to 
a Celeron... and crash on the next call to memcpy

migrating to 'bigger' systems shouldn't be a problem; unless you go from a 
32-bit system to PAE, or 64-bit...
-- 
Javier

Attachment: pgp1AJDcLK4tB.pgp
Description: PGP signature

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