On Wednesday 30 July 2008 15:30:10 Nick Piggin wrote:
> On Wednesday 30 July 2008 14:39, Rusty Russell wrote:
> >     Before: 5.93 seconds
> >     After: 5.40 seconds
> >
> > (Counts of slow vs fastpath in this usage are 6092 and 2852462
> > respectively.)
> >
> > And more importantly for lguest, the code is simpler.
>
> Cool. That's actually a fairly significant straight-line performance
> improvement, given that the lguest processes are running one-at-a-time
> right? (and not contending on the locks or bouncing cachelines too much)

Yep.  I mean, the performance is abysmal either way (native for this is about 
10x faster), but it would be interesting to see kvm numbers as they have a 
more sophisticated shadow page table setup.  Not as trivial as lguest since 
they seem to use the mm sem to protect other things...

Cheers,
Rusty.
_______________________________________________
Lguest mailing list
[email protected]
https://ozlabs.org/mailman/listinfo/lguest

Reply via email to