From: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx> The KVM hypervisor may provide a guest with ability to defer remote TLB flush when the remote VCPU is not running. When this feature is used, the TLB flush will happen only when the remote VPCU is scheduled to run again. This will avoid unnecessary (and expensive) IPIs. Under certain circumstances, when a guest initiates such deferred action, the hypervisor may miss the request. It is also possible that the guest may mistakenly assume that it has already marked remote VCPU as needing a flush when in fact that request had already been processed by the hypervisor. In both cases this will result in an invalid translation being present in a vCPU, potentially allowing accesses to memory locations in that guest's address space that should not be accessible. Note that only intra-guest memory is vulnerable. The attached patches address both of these problems: 1. The first patch makes sure the hypervisor doesn't accidentally clear guest's remote flush request 2. The rest of the patches prevent the race between hypervisor acknowledging a remote flush request and guest issuing a new one. Boris Ostrovsky (5): x86/kvm: Be careful not to clear KVM_VCPU_FLUSH_TLB bit x86/kvm: Introduce kvm_(un)map_gfn() x86/kvm: Cache gfn to pfn translation x86/KVM: Make sure KVM_VCPU_FLUSH_TLB flag is not missed x86/KVM: Clean up host's steal time structure arch/x86/include/asm/kvm_host.h | 4 +- arch/x86/kvm/x86.c | 69 +++++++++++++++--------- include/linux/kvm_host.h | 5 ++ include/linux/kvm_types.h | 9 +++- virt/kvm/kvm_main.c | 113 ++++++++++++++++++++++++++++++++++------ 5 files changed, 154 insertions(+), 46 deletions(-) -- 1.8.3.1
