Jakob Simon-Gaarde wrote:
> First of all I haven't just been talking, I have really spend days debugging 
> and profiling - so I think that should be respected in this context. I just 
> had to face the fact that I wasn't able to fix the problem.

Sorry if you viewed my answer disrespectful. It was not my intention. 
Any contribution is appreciated, and while your findings are currently a 
bit hard to explain, they are useful. But asking developers to react 
directly on even a small proportion of the mailing list mails would 
effectly shut down VirtualBox development. Which neither of us wants. So 
I've asked for a little patience, not more.

> The hack mentioned was not ment as a solution but as information - it was 
> something I used for manual profiling which turned out to have the curious 
> effect of speeding up performance.

So we both agree on the status of the hack. Have you tried the latest 
SVN code? As I said it may improve network performance in general (and 
NAT too). So feedback on that would be most welcome.

Klaus

_______________________________________________
vbox-dev mailing list
[email protected]
http://vbox.innotek.de/mailman/listinfo/vbox-dev

Reply via email to