Re: [PATCH v3] kvm-unit-tests : Basic architecture of VMX nested test case

[Gleb Natapov <gleb@redhat.com>]

On Tue, Jul 30, 2013 at 09:59:47AM +0200, Marius Vlad wrote:
> Hi Gleb,
> 
> On Tue, Jul 30, 2013 at 09:42:12AM +0200, Gleb Natapov wrote:
> > On Tue, Jul 30, 2013 at 09:31:17AM +0200, Marius Vlad wrote:
> > > Hi Arthur, 
> > > 
> > >    I'm trying to test your patch on a SandyBridge machine.
> > > 
> > >    Used 'nested=1' when loading kvm (from 3.9-1-amd64)
> > Should be kvm-intel, but you will get error otherwise, so I assume that
> > this is what you did.
> Indeed.
> > 
> > >    and qemu (emulator version 1.5.1 (Debian 1.5.0+dfsg-5)).
> > > 
> > >    Without nested=1 I get ``unhandled excecption 13'', so I presume
> > >    this is OK, with it, the test_vmx_capability() fails at
> > >    detecting vmx.
> > > 
> > >    I've used for qemu:
> > >    $ qemu-system-x86_64 -enable-kvm -device pc-testdev -device
> > >    isa-debug-exit,iobase=0xf4,iosize=0x4 -display none -serial stdio -device
> > >    pci-testdev -kernel x86/vmx.flat -cpu SandyBridge,+vmx,+x2apic -smp 1
> > > 
> > >    Are there any knobs which I should further tune?
> > > 
> > On the first glance you have everything that is needed. What is the output of
> > "cat /proc/cpuinfo" in the guest is?
I checked what test_vmx_capability() is actually doing and it uses
MSR_IA32_FEATURE_CONTROL MSR which is not supported in your version of
kernel. You need b3897a49e22fc173efa77527a447c714f753f681 commit, it is
not in any released version of kernel yet. You can just omit the test.

--
			Gleb.