Re: KVM performance

[Rusty Russell <rusty-8n+1lVoiYb80n/F98K4Iww@public.gmane.org>]

On Sun, 2007-01-28 at 11:40 +0200, Avi Kivity wrote:
> Rusty Russell wrote:
>  I use virtbench (http://ozlabs.org/~rusty/virtbench) to try
> > to measure optimization results; it still needs more tests (and an
> > explicit kvm backend).   
> 
> A kvm backend would be appreciated.

Yes, and patches are most welcome 8)  Actually, I'll work on this week.

> >> - after modifying a pte, kvm doesn't preload the modified pte into 
> >> shadow, but instead lets the guest fault it in  
> > lguest doesn't either, but don't you still want the fault to update the
> > host accessed bit?
> 
> The story here is that the guest is handling a pagefault and writing the 
> new guest pte.  kvm traps the write (guest pagetables are write 
> protected), and has the option of updating the shadow pte to reflect the 
> guest pte.
> 
> The clever guest kernel will set the accessed bit (and the dirty bit on 
> writable ptes) to avoid an rmw cycle by the hardware pagetable walker.
> 
> [two instrumented runs later]
> 
> Both Linux and Windows seem to do this optimization.

Right.  This (trivial!) optimization wins lguest a good 10%:

Before:
    Time for one Copy-on-Write fault: 13622 nsec
    Time to exec client once: 1085481 nsec
    Time for one fork/exit/wait: 700796 nsec
After:
    Time for one Copy-on-Write fault: 12036 nsec
    Time to exec client once: 969899 nsec
    Time for one fork/exit/wait: 664601 nsec

Thanks!
Rusty.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV