Re: [PATCH 0/28] nVMX: Nested VMX, v7

[Avi Kivity <avi@redhat.com>]

On 12/08/2010 06:59 PM, Nadav Har'El wrote:
> Hi,
>
> This is the seventh iteration of the nested VMX patch set. It fixes a bunch
> of bugs in the previous iteration, and in particular it now works correctly
> with EPT in the L0 hypervisor, so "ept=0" no longer needs to be specified.
>
> This new set of patches should apply to the current KVM trunk (I checked with
> 66fc6be8d2b04153b753182610f919faf9c705bc). In particular it uses the recently
> added is_guest_mode() function (common to both nested svm and vmx) instead of
> inventing our own flag.
>
> About nested VMX:
> -----------------
>
> The following 28 patches implement nested VMX support. This feature enables a
> guest to use the VMX APIs in order to run its own nested guests. In other
> words, it allows running hypervisors (that use VMX) under KVM.
> Multiple guest hypervisors can be run concurrently, and each of those can
> in turn host multiple guests.
>
> The theory behind this work, our implementation, and its performance
> characteristics were presented in OSDI 2010 (the USENIX Symposium on
> Operating Systems Design and Implementation). Our paper was titled
> "The Turtles Project: Design and Implementation of Nested Virtualization",
> and was awarded "Jay Lepreau Best Paper". The paper is available online, at:
>
> 	http://www.usenix.org/events/osdi10/tech/full_papers/Ben-Yehuda.pdf
>
> This patch set does not include all the features described in the paper.
> In particular, this patch set is missing nested EPT (shadow page tables are
> used in L1, while L0 can use shadow page tables or EPT). It is also missing
> some features required to run VMWare Server as a guest. These missing features
> will be sent as follow-on patchs.
>
> Running nested VMX:
> ------------------
>
> The current patches have a number of requirements, which will be relaxed in
> follow-on patches:
>
> 1. This version was only tested with KVM (64-bit) as a guest hypervisor, and
>     Linux as a nested guest.
>
> 2. SMP is supported in the code, but is unfortunately buggy in this version
>     and often leads to hangs. Use the "nosmp" option in the L0 (topmost)
>     kernel to avoid this bug (and to reduce your performance ;-))..

Any idea as to the cause?  There should be little interaction between 
host or guest smp and nvmx.

> 5. Nested VPID is not properly supported in this version. You must give the
>     "vpid=0" module options to kvm-intel to turn this feature off.

Do you mean host vpid here?  Likely you're not flushing the tlb when 
switching between guest and host.

-- 
error compiling committee.c: too many arguments to function