From: Sean Christopherson Date: Fri, 11 Mar 2022 03:27:41 +0000 Subject: KVM: x86: avoid calling x86 emulator without a decoded instruction Git-commit: fee060cd52d69c114b62d1a2948ea9648b5131f9 Patch-mainline: v5.19-rc1 References: CVE-2022-1852 bsc#1199875 Whenever x86_decode_emulated_instruction() detects a breakpoint, it returns the value that kvm_vcpu_check_breakpoint() writes into its pass-by-reference second argument. Unfortunately this is completely bogus because the expected outcome of x86_decode_emulated_instruction is an EMULATION_* value. Then, if kvm_vcpu_check_breakpoint() does "*r = 0" (corresponding to a KVM_EXIT_DEBUG userspace exit), it is misunderstood as EMULATION_OK and x86_emulate_instruction() is called without having decoded the instruction. This causes various havoc from running with a stale emulation context. The fix is to move the call to kvm_vcpu_check_breakpoint() where it was before commit 4aa2691dcbd3 ("KVM: x86: Factor out x86 instruction emulation with decoding") introduced x86_decode_emulated_instruction(). The other caller of the function does not need breakpoint checks, because it is invoked as part of a vmexit and the processor has already checked those before executing the instruction that #GP'd. This fixes CVE-2022-1852. Reported-by: Qiuhao Li Reported-by: Gaoning Pan Reported-by: Yongkang Jia Fixes: 4aa2691dcbd3 ("KVM: x86: Factor out x86 instruction emulation with decoding") Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson Message-Id: <20220311032801.3467418-2-seanjc@google.com> [Rewrote commit message according to Qiuhao's report, since a patch already existed to fix the bug. - Paolo] Signed-off-by: Paolo Bonzini Acked-by: Joerg Roedel --- arch/x86/kvm/x86.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -7694,7 +7694,7 @@ int kvm_skip_emulated_instruction(struct } EXPORT_SYMBOL_GPL(kvm_skip_emulated_instruction); -static bool kvm_vcpu_check_breakpoint(struct kvm_vcpu *vcpu, int *r) +static bool kvm_vcpu_check_code_breakpoint(struct kvm_vcpu *vcpu, int *r) { if (unlikely(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) && (vcpu->arch.guest_debug_dr7 & DR7_BP_EN_MASK)) { @@ -7763,25 +7763,23 @@ static bool is_vmware_backdoor_opcode(st } /* - * Decode to be emulated instruction. Return EMULATION_OK if success. + * Decode an instruction for emulation. The caller is responsible for handling + * code breakpoints. Note, manually detecting code breakpoints is unnecessary + * (and wrong) when emulating on an intercepted fault-like exception[*], as + * code breakpoints have higher priority and thus have already been done by + * hardware. + * + * [*] Except #MC, which is higher priority, but KVM should never emulate in + * response to a machine check. */ int x86_decode_emulated_instruction(struct kvm_vcpu *vcpu, int emulation_type, void *insn, int insn_len) { - int r = EMULATION_OK; struct x86_emulate_ctxt *ctxt = vcpu->arch.emulate_ctxt; + int r; init_emulate_ctxt(vcpu); - /* - * We will reenter on the same instruction since we do not set - * complete_userspace_io. This does not handle watchpoints yet, - * those would be handled in the emulate_ops. - */ - if (!(emulation_type & EMULTYPE_SKIP) && - kvm_vcpu_check_breakpoint(vcpu, &r)) - return r; - r = x86_decode_insn(ctxt, insn, insn_len, emulation_type); trace_kvm_emulate_insn_start(vcpu); @@ -7814,6 +7812,15 @@ int x86_emulate_instruction(struct kvm_v if (!(emulation_type & EMULTYPE_NO_DECODE)) { kvm_clear_exception_queue(vcpu); + /* + * Return immediately if RIP hits a code breakpoint, such #DBs + * are fault-like and are higher priority than any faults on + * the code fetch itself. + */ + if (!(emulation_type & EMULTYPE_SKIP) && + kvm_vcpu_check_code_breakpoint(vcpu, &r)) + return r; + r = x86_decode_emulated_instruction(vcpu, emulation_type, insn, insn_len); if (r != EMULATION_OK) {