Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2019-02-05 Thread Jim Mattson
On Thu, Aug 17, 2017 at 9:37 AM Paolo Bonzini wrote: > > There is currently some confusion between nested and L1 GPAs. The > assignment to "direct" in kvm_mmu_page_fault tries to fix that, but > it is not enough. What this patch does is fence off the MMIO cache > completely when using shadow nes

Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-18 Thread Paolo Bonzini
On 18/08/2017 14:35, Radim Krčmář wrote: > >>> diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h >>> @@ -90,7 +90,11 @@ static inline u32 bit(int bitno) >>> static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu, >>> gva_t gva, gfn_t gfn, unsigned acc

Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-18 Thread Radim Krčmář
2017-08-18 09:59+0200, David Hildenbrand: > On 17.08.2017 18:36, Paolo Bonzini wrote: > > There is currently some confusion between nested and L1 GPAs. The > > assignment to "direct" in kvm_mmu_page_fault tries to fix that, but > > it is not enough. What this patch does is fence off the MMIO cach

Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-18 Thread David Hildenbrand
On 17.08.2017 18:36, Paolo Bonzini wrote: > There is currently some confusion between nested and L1 GPAs. The > assignment to "direct" in kvm_mmu_page_fault tries to fix that, but > it is not enough. What this patch does is fence off the MMIO cache > completely when using shadow nested page table

[PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-17 Thread Paolo Bonzini
There is currently some confusion between nested and L1 GPAs. The assignment to "direct" in kvm_mmu_page_fault tries to fix that, but it is not enough. What this patch does is fence off the MMIO cache completely when using shadow nested page tables, since we have neither a GVA nor an L1 GPA to pu

Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-17 Thread David Hildenbrand
On 11.08.2017 18:52, Paolo Bonzini wrote: > There is currently some confusion between nested and L1 GPAs. The > assignment to "direct" in kvm_mmu_page_fault tries to fix that, but > it is not enough. What this patch does is fence off the MMIO cache > completely when using shadow nested page table

Re: [PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-12 Thread Wanpeng Li
2017-08-12 0:52 GMT+08:00 Paolo Bonzini : > There is currently some confusion between nested and L1 GPAs. The > assignment to "direct" in kvm_mmu_page_fault tries to fix that, but > it is not enough. What this patch does is fence off the MMIO cache > completely when using shadow nested page table

[PATCH 3/3] KVM: x86: fix use of L1 MMIO areas in nested guests

2017-08-11 Thread Paolo Bonzini
There is currently some confusion between nested and L1 GPAs. The assignment to "direct" in kvm_mmu_page_fault tries to fix that, but it is not enough. What this patch does is fence off the MMIO cache completely when using shadow nested page tables, since we have neither a GVA nor an L1 GPA to pu