Re: [PATCH] kvm: x86: nVMX: maintain internal copy of current VMCS

2016-07-15 Thread David Matlack
On Thu, Jul 14, 2016 at 1:33 AM, Paolo Bonzini wrote: > > > On 14/07/2016 02:16, David Matlack wrote: >> KVM maintains L1's current VMCS in guest memory, at the guest physical >> page identified by the argument to VMPTRLD. This makes hairy >> time-of-check to time-of-use bugs possible,as VCPUs can

Re: [PATCH] kvm: x86: nVMX: maintain internal copy of current VMCS

2016-07-14 Thread Paolo Bonzini
On 14/07/2016 02:16, David Matlack wrote: > KVM maintains L1's current VMCS in guest memory, at the guest physical > page identified by the argument to VMPTRLD. This makes hairy > time-of-check to time-of-use bugs possible,as VCPUs can be writing > the the VMCS page in memory while KVM is emulati

[PATCH] kvm: x86: nVMX: maintain internal copy of current VMCS

2016-07-13 Thread David Matlack
KVM maintains L1's current VMCS in guest memory, at the guest physical page identified by the argument to VMPTRLD. This makes hairy time-of-check to time-of-use bugs possible,as VCPUs can be writing the the VMCS page in memory while KVM is emulating VMLAUNCH and VMRESUME. The spec documents that w