Re: [RFC] VMX CR3 cache

[Marcelo Tosatti <marcelo-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org>]

On Mon, Jan 28, 2008 at 06:17:34PM +0100, Ingo Molnar wrote:
> 
> * Marcelo Tosatti <marcelo-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org> wrote:
> 
> > lat_ctx numbers (output is "nr-procs overhead-in-us"):
> > 
> > cr3-cache:
> > "size=0k ovr=1.30
> > 2 6.63
> > "size=0k ovr=1.31
> > 4 7.43
> > "size=0k ovr=1.32
> > 8 11.02
> 
> when i did the testing then i was able to get zero VM exits in the 
> lat_ctx hotpath and get the same performance as on native. The above 
> numbers you got suggest that this is not the case. (your lat_ctx numbers 
> on native are probably around 1 usec, right?) 

Actually, native is around 5usec:

# bin/x86_64-linux-gnu/lat_ctx -N 10 2

"size=0k ovr=1.09
2 5.21

So it does get pretty close to native, but the problem is that its
difficult to see improvements on macrobenchmarks (even hackbench which
remotely resembles a real scenario is not improving).

> If that is the case, could 
> you check via profile=kvm (or whatever other method you use to profile 
> KVM) where the VM exits come from?
> 
> 	Ingo
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> kvm-devel mailing list
> kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/kvm-devel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/