linux-kernel - [PATCH 5.15 397/667] KVM: nVMX: Clear IDT vectoring on nested VM-Exit for double/triple fault

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20220607164946.651521052@linuxfoundation.org>
Date:   Tue,  7 Jun 2022 19:01:02 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Chenyi Qiang <chenyi.qiang@...el.com>,
        Sean Christopherson <seanjc@...gle.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Sasha Levin <sashal@...nel.org>
Subject: [PATCH 5.15 397/667] KVM: nVMX: Clear IDT vectoring on nested VM-Exit for double/triple fault

From: Sean Christopherson <seanjc@...gle.com>

[ Upstream commit 9bd1f0efa859b61950d109b32ff8d529cc33a3ad ]

Clear the IDT vectoring field in vmcs12 on next VM-Exit due to a double
or triple fault.  Per the SDM, a VM-Exit isn't considered to occur during
event delivery if the exit is due to an intercepted double fault or a
triple fault.  Opportunistically move the default clearing (no event
"pending") into the helper so that it's more obvious that KVM does indeed
handle this case.

Note, the double fault case is worded rather wierdly in the SDM:

  The original event results in a double-fault exception that causes the
  VM exit directly.

Temporarily ignoring injected events, double faults can _only_ occur if
an exception occurs while attempting to deliver a different exception,
i.e. there's _always_ an original event.  And for injected double fault,
while there's no original event, injected events are never subject to
interception.

Presumably the SDM is calling out that a the vectoring info will be valid
if a different exit occurs after a double fault, e.g. if a #PF occurs and
is intercepted while vectoring #DF, then the vectoring info will show the
double fault.  In other words, the clause can simply be read as:

  The VM exit is caused by a double-fault exception.

Fixes: 4704d0befb07 ("KVM: nVMX: Exiting from L2 to L1")
Cc: Chenyi Qiang <chenyi.qiang@...el.com>
Signed-off-by: Sean Christopherson <seanjc@...gle.com>
Message-Id: <20220407002315.78092-4-seanjc@...gle.com>
Signed-off-by: Paolo Bonzini <pbonzini@...hat.com>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 arch/x86/kvm/vmx/nested.c | 32 ++++++++++++++++++++++++++++----
 arch/x86/kvm/vmx/vmcs.h   |  5 +++++
 2 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
index e0b5a4a83241..f7bdb62d6cec 100644
--- a/arch/x86/kvm/vmx/nested.c
+++ b/arch/x86/kvm/vmx/nested.c
@@ -3678,12 +3678,34 @@ vmcs12_guest_cr4(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 }
 
 static void vmcs12_save_pending_event(struct kvm_vcpu *vcpu,
-				      struct vmcs12 *vmcs12)
+				      struct vmcs12 *vmcs12,
+				      u32 vm_exit_reason, u32 exit_intr_info)
 {
 	u32 idt_vectoring;
 	unsigned int nr;
 
-	if (vcpu->arch.exception.injected) {
+	/*
+	 * Per the SDM, VM-Exits due to double and triple faults are never
+	 * considered to occur during event delivery, even if the double/triple
+	 * fault is the result of an escalating vectoring issue.
+	 *
+	 * Note, the SDM qualifies the double fault behavior with "The original
+	 * event results in a double-fault exception".  It's unclear why the
+	 * qualification exists since exits due to double fault can occur only
+	 * while vectoring a different exception (injected events are never
+	 * subject to interception), i.e. there's _always_ an original event.
+	 *
+	 * The SDM also uses NMI as a confusing example for the "original event
+	 * causes the VM exit directly" clause.  NMI isn't special in any way,
+	 * the same rule applies to all events that cause an exit directly.
+	 * NMI is an odd choice for the example because NMIs can only occur on
+	 * instruction boundaries, i.e. they _can't_ occur during vectoring.
+	 */
+	if ((u16)vm_exit_reason == EXIT_REASON_TRIPLE_FAULT ||
+	    ((u16)vm_exit_reason == EXIT_REASON_EXCEPTION_NMI &&
+	     is_double_fault(exit_intr_info))) {
+		vmcs12->idt_vectoring_info_field = 0;
+	} else if (vcpu->arch.exception.injected) {
 		nr = vcpu->arch.exception.nr;
 		idt_vectoring = nr | VECTORING_INFO_VALID_MASK;
 
@@ -3716,6 +3738,8 @@ static void vmcs12_save_pending_event(struct kvm_vcpu *vcpu,
 			idt_vectoring |= INTR_TYPE_EXT_INTR;
 
 		vmcs12->idt_vectoring_info_field = idt_vectoring;
+	} else {
+		vmcs12->idt_vectoring_info_field = 0;
 	}
 }
 
@@ -4202,8 +4226,8 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12,
 		 * Transfer the event that L0 or L1 may wanted to inject into
 		 * L2 to IDT_VECTORING_INFO_FIELD.
 		 */
-		vmcs12->idt_vectoring_info_field = 0;
-		vmcs12_save_pending_event(vcpu, vmcs12);
+		vmcs12_save_pending_event(vcpu, vmcs12,
+					  vm_exit_reason, exit_intr_info);
 
 		vmcs12->vm_exit_intr_info = exit_intr_info;
 		vmcs12->vm_exit_instruction_len = vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
diff --git a/arch/x86/kvm/vmx/vmcs.h b/arch/x86/kvm/vmx/vmcs.h
index 6e5de2e2b0da..4de2a6e3b190 100644
--- a/arch/x86/kvm/vmx/vmcs.h
+++ b/arch/x86/kvm/vmx/vmcs.h
@@ -104,6 +104,11 @@ static inline bool is_breakpoint(u32 intr_info)
 	return is_exception_n(intr_info, BP_VECTOR);
 }
 
+static inline bool is_double_fault(u32 intr_info)
+{
+	return is_exception_n(intr_info, DF_VECTOR);
+}
+
 static inline bool is_page_fault(u32 intr_info)
 {
 	return is_exception_n(intr_info, PF_VECTOR);
-- 
2.35.1