From: Marcelo Tosatti Date: Thu, 19 Dec 2013 17:28:51 +0000 (-0200) Subject: KVM: MMU: handle invalid root_hpa at __direct_map X-Git-Tag: v3.12.16~9 X-Git-Url: http://git.hungrycats.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=0334c2e03c00144d2ff68c97a1ff93a640883c57;p=linux KVM: MMU: handle invalid root_hpa at __direct_map commit 989c6b34f6a9480e397b170cc62237e89bf4fdb9 upstream. It is possible for __direct_map to be called on invalid root_hpa (-1), two examples: 1) try_async_pf -> can_do_async_pf -> vmx_interrupt_allowed -> nested_vmx_vmexit 2) vmx_handle_exit -> vmx_interrupt_allowed -> nested_vmx_vmexit Then to load_vmcs12_host_state and kvm_mmu_reset_context. Check for this possibility, let fault exception be regenerated. BZ: https://bugzilla.redhat.com/show_bug.cgi?id=924916 Signed-off-by: Marcelo Tosatti Signed-off-by: Paolo Bonzini Cc: Josh Boyer Signed-off-by: Jiri Slaby --- diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index dce0df8150df..0af8490eb9a9 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2664,6 +2664,9 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, int emulate = 0; gfn_t pseudo_gfn; + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return 0; + for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) { if (iterator.level == level) { mmu_set_spte(vcpu, iterator.sptep, ACC_ALL,