Hi,
I just noticed that Nested VMX got broken (at least in my tests) by commit
46199f33c29533e7ad2a7d2128dc30175d1d4157.
The specific change causing the problem was:
@@ -2220,7 +2216,6 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32
msr_index, u64 data)
break;
msr = find_msr_entry(vmx, msr_index);
if (msr) {
- vmx_load_host_state(vmx);
msr->data = data;
break;
}
And if anyone wants a quick workaround to making nested VMX work again,
returning this line fixes the problem.
I'm still trying to figure out why this line, which indeed seems unrelated
and unnecessary, is necessary for the correct functioning of nested VMX.
My (unsubstantiated) guess is that it isn't that it is actually necessary
in this point - it's just that it does something that should have been more
properly done in another place, but I've yet to figure out exactly what.
I'll send a patch when I have this figured out. If anybody else has any
guess, I'd love to hear.
Nadav.
--
Nadav Har'El | Tuesday, Mar 6 2012,
nyh@xxxxxxxxxxxxxxxxxxx |-----------------------------------------
Phone +972-523-790466, ICQ 13349191 |Sign in pool: "Welcome to our OOL. Notice
http://nadav.harel.org.il |there is no P, please keep it that way."
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
