Re: [PATCH 0/3] Nested TSC handling

[Bandan Das <bandan.das@stratus.com>]

On  0, Nadav Har'El <nyh@il.ibm.com> wrote:
> The following are patches I propose for fixing the bug discovered by Bandan
> Das and discussed in the "Nested VMX - L1 hangs on running L2" thread.
> 
> The first patch should fix the originally-reported bug, as explained in the
> aforementioned thread: A new x86_op read_l1_tsc() is called L1's TSC is
> needed, instead of assuming that calling kvm_read_msr() will do that
> (because this has to return L2's TSC when a nested guest is running).
> 
> The second and third patches fix relatively-unimportant corner cases in
> nested VMX and nested SVM TSC handling.
> 
> I'd appreciate it if the people who noticed this bug can verify that these
> patches indeed solve it for them.
I can verify that these patches work for me. Thanks Nadav and others for looking 
into this!

I have cc'ed Jan and Matt in case they would be interested in trying out 
these patches.


> Patch statistics:
> -----------------
> 
>  arch/x86/include/asm/kvm_host.h |    2 +
>  arch/x86/kvm/svm.c              |   13 ++++++--
>  arch/x86/kvm/vmx.c              |   48 +++++++++++++++++++++++-------
>  arch/x86/kvm/x86.c              |    8 ++---
>  4 files changed, 54 insertions(+), 17 deletions(-)
> 
> --
> Nadav Har'El
> IBM Haifa Research Lab