Skip to contentSkip to conversation

Add bus lock VM exit support

PATCH RESEND Superseded Chenyi Qiang 2 participants 7 messages Oct 12, 2020 – Oct 26, 2020

Subsystems (3)
x86 KVM KVM x86

v4 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

Chenyi Qiang

author
#0 lore raw

v4 cover Add bus lock VM exit support

This patch series add the support for bus lock VM exit in KVM. It is a sub-feature of bus lock detection. When it is enabled by the VMM, the processor generates a "Bus Lock" VM exit following execution of an instruction if the processor detects that one or more bus locks were caused the instruction was being executed (due to either direct access by the instruction or stuffed accesses like through A/D updates).

This first patch applies Sean's refactor for vcpu_vmx.exit_reason available at https://patchwork.kernel.org/patch/11500659.
It is necessary as bus lock VM exit adds a new modifier bit(bit 26) in exit_reason field in VMCS.

The second patch is the enabling work for bus lock VM exit. Add the support to set the capability to enable bus lock vm exit. The current implementation just exit to user space when handling the bus lock detected in guest.

The concrete throttling policy in user space is still to be discussed. We can enforce ratelimit on bus lock in guest, inject some sleep time or maybe other ideas.

Document for Bus Lock Detection is now available at the latest "Intel Architecture Instruction Set Extensions Programming Reference".

Document Link:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html

---

Changelogs

v3->v4:
- rebase on top of v5.9
- some code cleanup.
- v3:https://lore.kernel.org/lkml/[email protected]/

v2->v3:
- use a bitmap to get/set the capability of bus lock detection. we support exit and off mode currently.
- put the handle of exiting to userspace in vmx.c, thus no need to define a shadow to track vmx->exit_reason.bus_lock_detected.
- remove the vcpu->stats.bus_locks since every bus lock exits to userspace.
- v2:https://lore.kernel.org/lkml/[email protected]/

v1->v2:
- resolve Vitaly's comment to introduce the KVM_EXIT_BUS_LOCK and a capability to enable it.
- add the support to exit to user space when handling bus locks.
- extend the vcpu->run->flags to indicate bus lock detected for other exit reasons when exiting to user space.
- v1:https://lore.kernel.org/lkml/[email protected]/

---

Chenyi Qiang (1):
KVM: VMX: Enable bus lock VM exit

Sean Christopherson (1):
KVM: VMX: Convert vcpu_vmx.exit_reason to a union

 arch/x86/include/asm/kvm_host.h    |   7 ++
 arch/x86/include/asm/vmx.h         |   1 +
 arch/x86/include/asm/vmxfeatures.h |   1 +
 arch/x86/include/uapi/asm/kvm.h    |   1 +
 arch/x86/include/uapi/asm/vmx.h    |   4 +-
 arch/x86/kvm/vmx/capabilities.h    |   6 ++
 arch/x86/kvm/vmx/nested.c          |  42 +++++++-----
 arch/x86/kvm/vmx/vmx.c             | 103 +++++++++++++++++++----------
 arch/x86/kvm/vmx/vmx.h             |  25 ++++++-
 arch/x86/kvm/x86.c                 |  29 +++++++-
 include/uapi/linux/kvm.h           |   5 ++
 11 files changed, 172 insertions(+), 52 deletions(-)
Signature · 1 line
2.17.1
author
#1 lore raw

PATCH RESEND v4 1/2 KVM: VMX: Convert vcpu_vmx.exit_reason to a union

From: Sean Christopherson <[email protected]>

Convert vcpu_vmx.exit_reason from a u32 to a union (of size u32). The full VM_EXIT_REASON field is comprised of a 16-bit basic exit reason in bits 15:0, and single-bit modifiers in bits 31:16.

Historically, KVM has only had to worry about handling the "failed VM-Entry" modifier, which could only be set in very specific flows and required dedicated handling. I.e. manually stripping the FAILED_VMENTRY bit was a somewhat viable approach. But even with only a single bit to worry about, KVM has had several bugs related to comparing a basic exit reason against the full exit reason store in vcpu_vmx.

Upcoming Intel features, e.g. SGX, will add new modifier bits that can be set on more or less any VM-Exit, as opposed to the significantly more restricted FAILED_VMENTRY, i.e. correctly handling everything in one-off flows isn't scalable. Tracking exit reason in a union forces code to explicitly choose between consuming the full exit reason and the basic exit, and is a convenient way to document and access the modifiers.

No functional change intended.

Cc Xiaoyao Li <[email protected]>
Signed-off-by Sean Christopherson <[email protected]>
---
 arch/x86/kvm/vmx/nested.c | 42 ++++++++++++++++---------
 arch/x86/kvm/vmx/vmx.c    | 66 ++++++++++++++++++++-------------------
 arch/x86/kvm/vmx/vmx.h    | 25 ++++++++++++++-
 3 files changed, 85 insertions(+), 48 deletions(-)
arch/x86/kvm/vmx/nested.c +27 -15
index 1bb6b31eb646..370c414e333c 100644
@@ -3304,7 +3304,11 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu,
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
 	enum vm_entry_failure_code entry_failure_code;
 	bool evaluate_pending_interrupts;
-	u32 exit_reason, failed_index;
+	union vmx_exit_reason exit_reason = {
+		.basic = EXIT_REASON_INVALID_STATE,
+		.failed_vmentry = 1,
+	};
+	u32 failed_index;
 
 	if (kvm_check_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu))
 		kvm_vcpu_flush_tlb_current(vcpu);

@@ -3354,7 +3358,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu,
 
 		if (nested_vmx_check_guest_state(vcpu, vmcs12,
 						 &entry_failure_code)) {
-			exit_reason = EXIT_REASON_INVALID_STATE;
+			exit_reason.basic = EXIT_REASON_INVALID_STATE;
 			vmcs12->exit_qualification = entry_failure_code;
 			goto vmentry_fail_vmexit;
 		}

@@ -3365,7 +3369,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu,
 		vcpu->arch.tsc_offset += vmcs12->tsc_offset;
 
 	if (prepare_vmcs02(vcpu, vmcs12, &entry_failure_code)) {
-		exit_reason = EXIT_REASON_INVALID_STATE;
+		exit_reason.basic = EXIT_REASON_INVALID_STATE;
 		vmcs12->exit_qualification = entry_failure_code;
 		goto vmentry_fail_vmexit_guest_mode;
 	}

@@ -3375,7 +3379,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu,
 						   vmcs12->vm_entry_msr_load_addr,
 						   vmcs12->vm_entry_msr_load_count);
 		if (failed_index) {
-			exit_reason = EXIT_REASON_MSR_LOAD_FAIL;
+			exit_reason.basic = EXIT_REASON_MSR_LOAD_FAIL;
 			vmcs12->exit_qualification = failed_index;
 			goto vmentry_fail_vmexit_guest_mode;
 		}

@@ -3443,7 +3447,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu,
 		return NVMX_VMENTRY_VMEXIT;
 
 	load_vmcs12_host_state(vcpu, vmcs12);
-	vmcs12->vm_exit_reason = exit_reason | VMX_EXIT_REASONS_FAILED_VMENTRY;
+	vmcs12->vm_exit_reason = exit_reason.full;
 	if (enable_shadow_vmcs || vmx->nested.hv_evmcs)
 		vmx->nested.need_vmcs12_to_shadow_sync = true;
 	return NVMX_VMENTRY_VMEXIT;

@@ -5496,7 +5500,12 @@ static int handle_vmfunc(struct kvm_vcpu *vcpu)
 	return kvm_skip_emulated_instruction(vcpu);
 
 fail:
-	nested_vmx_vmexit(vcpu, vmx->exit_reason,
+	/*
+	 * This is effectively a reflected VM-Exit, as opposed to a synthesized
+	 * nested VM-Exit.  Pass the original exit reason, i.e. don't hardcode
+	 * EXIT_REASON_VMFUNC as the exit reason.
+	 */
+	nested_vmx_vmexit(vcpu, vmx->exit_reason.full,
 			  vmx_get_intr_info(vcpu),
 			  vmx_get_exit_qual(vcpu));
 	return 1;

@@ -5564,7 +5573,8 @@ static bool nested_vmx_exit_handled_io(struct kvm_vcpu *vcpu,
  * MSR bitmap. This may be the case even when L0 doesn't use MSR bitmaps.
  */
 static bool nested_vmx_exit_handled_msr(struct kvm_vcpu *vcpu,
-	struct vmcs12 *vmcs12, u32 exit_reason)
+					struct vmcs12 *vmcs12,
+					union vmx_exit_reason exit_reason)
 {
 	u32 msr_index = kvm_rcx_read(vcpu);
 	gpa_t bitmap;

@@ -5578,7 +5588,7 @@ static bool nested_vmx_exit_handled_msr(struct kvm_vcpu *vcpu,
 	 * First we need to figure out which of the four to use:
 	 */
 	bitmap = vmcs12->msr_bitmap;
-	if (exit_reason == EXIT_REASON_MSR_WRITE)
+	if (exit_reason.basic == EXIT_REASON_MSR_WRITE)
 		bitmap += 2048;
 	if (msr_index >= 0xc0000000) {
 		msr_index -= 0xc0000000;

@@ -5715,11 +5725,12 @@ static bool nested_vmx_exit_handled_mtf(struct vmcs12 *vmcs12)
  * Return true if L0 wants to handle an exit from L2 regardless of whether or not
  * L1 wants the exit.  Only call this when in is_guest_mode (L2).
  */
-static bool nested_vmx_l0_wants_exit(struct kvm_vcpu *vcpu, u32 exit_reason)
+static bool nested_vmx_l0_wants_exit(struct kvm_vcpu *vcpu,
+				     union vmx_exit_reason exit_reason)
 {
 	u32 intr_info;
 
-	switch ((u16)exit_reason) {
+	switch ((u16)exit_reason.basic) {
 	case EXIT_REASON_EXCEPTION_NMI:
 		intr_info = vmx_get_intr_info(vcpu);
 		if (is_nmi(intr_info))

@@ -5775,12 +5786,13 @@ static bool nested_vmx_l0_wants_exit(struct kvm_vcpu *vcpu, u32 exit_reason)
  * Return 1 if L1 wants to intercept an exit from L2.  Only call this when in
  * is_guest_mode (L2).
  */
-static bool nested_vmx_l1_wants_exit(struct kvm_vcpu *vcpu, u32 exit_reason)
+static bool nested_vmx_l1_wants_exit(struct kvm_vcpu *vcpu,
+				     union vmx_exit_reason exit_reason)
 {
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
 	u32 intr_info;
 
-	switch ((u16)exit_reason) {
+	switch ((u16)exit_reason.basic) {
 	case EXIT_REASON_EXCEPTION_NMI:
 		intr_info = vmx_get_intr_info(vcpu);
 		if (is_nmi(intr_info))

@@ -5899,7 +5911,7 @@ static bool nested_vmx_l1_wants_exit(struct kvm_vcpu *vcpu, u32 exit_reason)
 bool nested_vmx_reflect_vmexit(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 exit_reason = vmx->exit_reason;
+	union vmx_exit_reason exit_reason = vmx->exit_reason;
 	unsigned long exit_qual;
 	u32 exit_intr_info;
 

@@ -5921,7 +5933,7 @@ bool nested_vmx_reflect_vmexit(struct kvm_vcpu *vcpu)
 	exit_intr_info = vmx_get_intr_info(vcpu);
 	exit_qual = vmx_get_exit_qual(vcpu);
 
-	trace_kvm_nested_vmexit(kvm_rip_read(vcpu), exit_reason, exit_qual,
+	trace_kvm_nested_vmexit(kvm_rip_read(vcpu), exit_reason.full, exit_qual,
 				vmx->idt_vectoring_info, exit_intr_info,
 				vmcs_read32(VM_EXIT_INTR_ERROR_CODE),
 				KVM_ISA_VMX);

@@ -5950,7 +5962,7 @@ bool nested_vmx_reflect_vmexit(struct kvm_vcpu *vcpu)
 	}
 
 reflect_vmexit:
-	nested_vmx_vmexit(vcpu, exit_reason, exit_intr_info, exit_qual);
+	nested_vmx_vmexit(vcpu, exit_reason.full, exit_intr_info, exit_qual);
 	return true;
 }
arch/x86/kvm/vmx/vmx.c +34 -32

149 lines not on this page · raw message

arch/x86/kvm/vmx/vmx.h +24 -1
index a0e47720f60c..ca250622b123 100644
@@ -92,6 +92,29 @@ struct pt_desc {
 	struct pt_ctx guest;
 };
 
+union vmx_exit_reason {
+	struct {
+		u32	basic			: 16;
+		u32	reserved16		: 1;
+		u32	reserved17		: 1;
+		u32	reserved18		: 1;
+		u32	reserved19		: 1;
+		u32	reserved20		: 1;
+		u32	reserved21		: 1;
+		u32	reserved22		: 1;
+		u32	reserved23		: 1;
+		u32	reserved24		: 1;
+		u32	reserved25		: 1;
+		u32	reserved26		: 1;
+		u32	enclave_mode		: 1;
+		u32	smi_pending_mtf		: 1;
+		u32	smi_from_vmx_root	: 1;
+		u32	reserved30		: 1;
+		u32	failed_vmentry		: 1;
+	};
+	u32 full;
+};
+
 /*
  * The nested_vmx structure is part of vcpu_vmx, and holds information we need
  * for correct emulation of VMX (i.e., nested VMX) on this vcpu.

@@ -266,7 +289,7 @@ struct vcpu_vmx {
 	int vpid;
 	bool emulation_required;
 
-	u32 exit_reason;
+	union vmx_exit_reason exit_reason;
 
 	/* Posted interrupt descriptor */
 	struct pi_desc pi_desc;
Signature · 1 line
2.17.1
#2 lore raw

Re: [RESEND v4 1/2] KVM: VMX: Convert vcpu_vmx.exit_reason to a union

Quoted 20 lines · Chenyi Qiang

From: Sean Christopherson <[email protected]>

Convert vcpu_vmx.exit_reason from a u32 to a union (of size u32). The full VM_EXIT_REASON field is comprised of a 16-bit basic exit reason in bits 15:0, and single-bit modifiers in bits 31:16.

Historically, KVM has only had to worry about handling the "failed VM-Entry" modifier, which could only be set in very specific flows and required dedicated handling. I.e. manually stripping the FAILED_VMENTRY bit was a somewhat viable approach. But even with only a single bit to worry about, KVM has had several bugs related to comparing a basic exit reason against the full exit reason store in vcpu_vmx.

Upcoming Intel features, e.g. SGX, will add new modifier bits that can be set on more or less any VM-Exit, as opposed to the significantly more restricted FAILED_VMENTRY, i.e. correctly handling everything in one-off flows isn't scalable. Tracking exit reason in a union forces code to explicitly choose between consuming the full exit reason and the basic exit, and is a convenient way to document and access the modifiers.

Quoted 4 lines

No functional change intended.

Cc Xiaoyao Li <[email protected]>
Signed-off-by Sean Christopherson <[email protected]>

This needs your SOB since you are involved in the handling of the patch.

author
#3 lore raw

Re: [RESEND v4 1/2] KVM: VMX: Convert vcpu_vmx.exit_reason to a union

Quoted 21 lines · Sean Christopherson
Quoted 20 lines · Chenyi Qiang

From: Sean Christopherson <[email protected]>

Convert vcpu_vmx.exit_reason from a u32 to a union (of size u32). The full VM_EXIT_REASON field is comprised of a 16-bit basic exit reason in bits 15:0, and single-bit modifiers in bits 31:16.

Historically, KVM has only had to worry about handling the "failed VM-Entry" modifier, which could only be set in very specific flows and required dedicated handling. I.e. manually stripping the FAILED_VMENTRY bit was a somewhat viable approach. But even with only a single bit to worry about, KVM has had several bugs related to comparing a basic exit reason against the full exit reason store in vcpu_vmx.

Upcoming Intel features, e.g. SGX, will add new modifier bits that can be set on more or less any VM-Exit, as opposed to the significantly more restricted FAILED_VMENTRY, i.e. correctly handling everything in one-off flows isn't scalable. Tracking exit reason in a union forces code to explicitly choose between consuming the full exit reason and the basic exit, and is a convenient way to document and access the modifiers.

Quoted 6 lines
Quoted 4 lines

No functional change intended.

Cc Xiaoyao Li <[email protected]>
Signed-off-by Sean Christopherson <[email protected]>

This needs your SOB since you are involved in the handling of the patch.

Sorry for the late response, my mailbox encountered some issues...

Will add the SOB here.

Quoted 1 line
author
#4 lore raw

PATCH RESEND v4 2/2 KVM: VMX: Enable bus lock VM exit

In reply to Chenyi Qiang

Virtual Machine can exploit bus locks to degrade the performance of system. Bus lock can be caused by split locked access to writeback(WB) memory or by using locks on uncacheable(UC) memory. The bus lock is typically >1000 cycles slower than an atomic operation within a cache line. It also disrupts performance on other cores (which must wait for the bus lock to be released before their memory operations can complete).

To address the threat, bus lock VM exit is introduced to notify the VMM when a bus lock was acquired, allowing it to enforce throttling or other policy based mitigations.

A VMM can enable VM exit due to bus locks by setting a new "Bus Lock Detection" VM-execution control(bit 30 of Secondary Processor-based VM execution controls). If delivery of this VM exit was preempted by a higher priority VM exit (e.g. EPT misconfiguration, EPT violation, APIC access VM exit, APIC write VM exit, exception bitmap exiting), bit 26 of exit reason in vmcs field is set to 1.

In current implementation, the KVM exposes this capability through KVM_CAP_X86_BUS_LOCK_EXIT. The user can get the supported mode bitmap (i.e. off and exit) and enable it explicitly (disabled by default). If bus locks in guest are detected by KVM, exit to user space even when current exit reason is handled by KVM internally. Set a new field KVM_RUN_BUS_LOCK in vcpu->run->flags to inform the user space that there is a bus lock in guest and it is preempted by a higher prioriy VM exit.

Document for Bus Lock VM exit is now available at the latest "Intel Architecture Instruction Set Extensions Programming Reference".

Document Link:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html

Co-developed-by Xiaoyao Li <[email protected]>
Signed-off-by Xiaoyao Li <[email protected]>
Signed-off-by Chenyi Qiang <[email protected]>
---
 arch/x86/include/asm/kvm_host.h    |  7 ++++++
 arch/x86/include/asm/vmx.h         |  1 +
 arch/x86/include/asm/vmxfeatures.h |  1 +
 arch/x86/include/uapi/asm/kvm.h    |  1 +
 arch/x86/include/uapi/asm/vmx.h    |  4 +++-
 arch/x86/kvm/vmx/capabilities.h    |  6 +++++
 arch/x86/kvm/vmx/vmx.c             | 37 ++++++++++++++++++++++++++++--
 arch/x86/kvm/vmx/vmx.h             |  2 +-
 arch/x86/kvm/x86.c                 | 29 ++++++++++++++++++++++-
 include/uapi/linux/kvm.h           |  5 ++++
 10 files changed, 88 insertions(+), 5 deletions(-)
arch/x86/include/asm/kvm_host.h +7 -0
index 5303dbc5c9bc..f5dd88553d19 100644
@@ -52,6 +52,9 @@
 #define KVM_DIRTY_LOG_MANUAL_CAPS   (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \
 					KVM_DIRTY_LOG_INITIALLY_SET)
 
+#define KVM_BUS_LOCK_DETECTION_VALID_MODE	(KVM_BUS_LOCK_DETECTION_OFF | \
+						 KVM_BUS_LOCK_DETECTION_EXIT)
+
 /* x86-specific vcpu->requests bit members */
 #define KVM_REQ_MIGRATE_TIMER		KVM_ARCH_REQ(0)
 #define KVM_REQ_REPORT_TPR_ACCESS	KVM_ARCH_REQ(1)

@@ -961,6 +964,8 @@ struct kvm_arch {
 	bool guest_can_read_msr_platform_info;
 	bool exception_payload_enabled;
 
+	bool bus_lock_detection_enabled;
+
 	struct kvm_pmu_event_filter *pmu_event_filter;
 	struct task_struct *nx_lpage_recovery_thread;
 };

@@ -1347,6 +1352,8 @@ extern u8   kvm_tsc_scaling_ratio_frac_bits;
 extern u64  kvm_max_tsc_scaling_ratio;
 /* 1ull << kvm_tsc_scaling_ratio_frac_bits */
 extern u64  kvm_default_tsc_scaling_ratio;
+/* bus lock detection supported? */
+extern bool kvm_has_bus_lock_exit;
 
 extern u64 kvm_mce_cap_supported;
arch/x86/include/asm/vmx.h +1 -0
index cd7de4b401fe..93a880bc31a7 100644
@@ -73,6 +73,7 @@
 #define SECONDARY_EXEC_PT_USE_GPA		VMCS_CONTROL_BIT(PT_USE_GPA)
 #define SECONDARY_EXEC_TSC_SCALING              VMCS_CONTROL_BIT(TSC_SCALING)
 #define SECONDARY_EXEC_ENABLE_USR_WAIT_PAUSE	VMCS_CONTROL_BIT(USR_WAIT_PAUSE)
+#define SECONDARY_EXEC_BUS_LOCK_DETECTION	VMCS_CONTROL_BIT(BUS_LOCK_DETECTION)
 
 #define PIN_BASED_EXT_INTR_MASK                 VMCS_CONTROL_BIT(INTR_EXITING)
 #define PIN_BASED_NMI_EXITING                   VMCS_CONTROL_BIT(NMI_EXITING)
arch/x86/include/asm/vmxfeatures.h +1 -0
index 9915990fd8cf..e80523346274 100644
@@ -83,5 +83,6 @@
 #define VMX_FEATURE_TSC_SCALING		( 2*32+ 25) /* Scale hardware TSC when read in guest */
 #define VMX_FEATURE_USR_WAIT_PAUSE	( 2*32+ 26) /* Enable TPAUSE, UMONITOR, UMWAIT in guest */
 #define VMX_FEATURE_ENCLV_EXITING	( 2*32+ 28) /* "" VM-Exit on ENCLV (leaf dependent) */
+#define VMX_FEATURE_BUS_LOCK_DETECTION	( 2*32+ 30) /* VM-Exit when bus lock caused */
 
 #endif /* _ASM_X86_VMXFEATURES_H */
arch/x86/include/uapi/asm/kvm.h +1 -0
index 0780f97c1850..a1471c05f7f9 100644
@@ -111,6 +111,7 @@ struct kvm_ioapic_state {
 #define KVM_NR_IRQCHIPS          3
 
 #define KVM_RUN_X86_SMM		 (1 << 0)
+#define KVM_RUN_BUS_LOCK         (1 << 1)
 
 /* for KVM_GET_REGS and KVM_SET_REGS */
 struct kvm_regs {
arch/x86/include/uapi/asm/vmx.h +3 -1
index b8ff9e8ac0d5..14c177c4afd5 100644
@@ -88,6 +88,7 @@
 #define EXIT_REASON_XRSTORS             64
 #define EXIT_REASON_UMWAIT              67
 #define EXIT_REASON_TPAUSE              68
+#define EXIT_REASON_BUS_LOCK            74
 
 #define VMX_EXIT_REASONS \
 	{ EXIT_REASON_EXCEPTION_NMI,         "EXCEPTION_NMI" }, \

@@ -148,7 +149,8 @@
 	{ EXIT_REASON_XSAVES,                "XSAVES" }, \
 	{ EXIT_REASON_XRSTORS,               "XRSTORS" }, \
 	{ EXIT_REASON_UMWAIT,                "UMWAIT" }, \
-	{ EXIT_REASON_TPAUSE,                "TPAUSE" }
+	{ EXIT_REASON_TPAUSE,                "TPAUSE" }, \
+	{ EXIT_REASON_BUS_LOCK,              "BUS_LOCK" }
 
 #define VMX_EXIT_REASON_FLAGS \
 	{ VMX_EXIT_REASONS_FAILED_VMENTRY,	"FAILED_VMENTRY" }
arch/x86/kvm/vmx/capabilities.h +6 -0
index 4bbd8b448d22..aa94535e6705 100644
@@ -262,6 +262,12 @@ static inline bool cpu_has_vmx_tsc_scaling(void)
 		SECONDARY_EXEC_TSC_SCALING;
 }
 
+static inline bool cpu_has_vmx_bus_lock_detection(void)
+{
+	return vmcs_config.cpu_based_2nd_exec_ctrl &
+	    SECONDARY_EXEC_BUS_LOCK_DETECTION;
+}
+
 static inline bool cpu_has_vmx_apicv(void)
 {
 	return cpu_has_vmx_apic_register_virt() &&
arch/x86/kvm/vmx/vmx.c +35 -2
index 01dedcedd2a9..22374cedac2b 100644
@@ -2476,7 +2476,8 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf,
 			SECONDARY_EXEC_ENABLE_USR_WAIT_PAUSE |
 			SECONDARY_EXEC_PT_USE_GPA |
 			SECONDARY_EXEC_PT_CONCEAL_VMX |
-			SECONDARY_EXEC_ENABLE_VMFUNC;
+			SECONDARY_EXEC_ENABLE_VMFUNC |
+			SECONDARY_EXEC_BUS_LOCK_DETECTION;
 		if (cpu_has_sgx())
 			opt2 |= SECONDARY_EXEC_ENCLS_EXITING;
 		if (adjust_vmx_controls(min2, opt2,

@@ -4259,6 +4260,9 @@ static void vmx_compute_secondary_exec_control(struct vcpu_vmx *vmx)
 		}
 	}
 
+	if (!vcpu->kvm->arch.bus_lock_detection_enabled)
+		exec_control &= ~SECONDARY_EXEC_BUS_LOCK_DETECTION;
+
 	vmx->secondary_exec_control = exec_control;
 }
 

@@ -5691,6 +5695,12 @@ static int handle_encls(struct kvm_vcpu *vcpu)
 	return 1;
 }
 
+static int handle_bus_lock(struct kvm_vcpu *vcpu)
+{
+	vcpu->run->exit_reason = KVM_EXIT_BUS_LOCK;
+	return 0;
+}
+
 /*
  * The exit handlers return 1 if the exit was handled fully and guest execution
  * may resume.  Otherwise they set the kvm_run parameter to indicate what needs

@@ -5747,6 +5757,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = {
 	[EXIT_REASON_VMFUNC]		      = handle_vmx_instruction,
 	[EXIT_REASON_PREEMPTION_TIMER]	      = handle_preemption_timer,
 	[EXIT_REASON_ENCLS]		      = handle_encls,
+	[EXIT_REASON_BUS_LOCK]                = handle_bus_lock,
 };
 
 static const int kvm_vmx_max_exit_handlers =

@@ -5985,7 +5996,7 @@ void dump_vmcs(void)
  * The guest has exited.  See if we can fix it or if we need userspace
  * assistance.
  */
-static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
+static int __vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	union vmx_exit_reason exit_reason = vmx->exit_reason;

@@ -6138,6 +6149,26 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 	return 0;
 }
 
+static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
+{
+	int ret = __vmx_handle_exit(vcpu, exit_fastpath);
+
+	/*
+	 * Even when current exit reason is handled by KVM internally, we
+	 * still need to exit to user space when bus lock detected to inform
+	 * that there is a bus lock in guest.
+	 */
+	if (to_vmx(vcpu)->exit_reason.bus_lock_detected) {
+		if (ret > 0)
+			vcpu->run->exit_reason = KVM_EXIT_BUS_LOCK;
+		else
+			vcpu->run->flags |= KVM_RUN_BUS_LOCK;
+		return 0;
+	}
+	vcpu->run->flags &= ~KVM_RUN_BUS_LOCK;
+	return ret;
+}
+
 /*
  * Software based L1D cache flush which is used when microcode providing
  * the cache control MSR is not loaded.

@@ -8104,6 +8135,8 @@ static __init int hardware_setup(void)
 		kvm_tsc_scaling_ratio_frac_bits = 48;
 	}
 
+	kvm_has_bus_lock_exit = cpu_has_vmx_bus_lock_detection();
+
 	set_bit(0, vmx_vpid_bitmap); /* 0 is reserved for host */
 
 	if (enable_ept)
arch/x86/kvm/vmx/vmx.h +1 -1
index ca250622b123..f78f9d5552b8 100644
@@ -105,7 +105,7 @@ union vmx_exit_reason {
 		u32	reserved23		: 1;
 		u32	reserved24		: 1;
 		u32	reserved25		: 1;
-		u32	reserved26		: 1;
+		u32	bus_lock_detected	: 1;
 		u32	enclave_mode		: 1;
 		u32	smi_pending_mtf		: 1;
 		u32	smi_from_vmx_root	: 1;
arch/x86/kvm/x86.c +28 -1
index ce856e0ece84..00eb23da3a73 100644
@@ -134,6 +134,8 @@ u64  __read_mostly kvm_max_tsc_scaling_ratio;
 EXPORT_SYMBOL_GPL(kvm_max_tsc_scaling_ratio);
 u64 __read_mostly kvm_default_tsc_scaling_ratio;
 EXPORT_SYMBOL_GPL(kvm_default_tsc_scaling_ratio);
+bool __read_mostly kvm_has_bus_lock_exit;
+EXPORT_SYMBOL_GPL(kvm_has_bus_lock_exit);
 
 /* tsc tolerance in parts per million - default to 1/2 of the NTP threshold */
 static u32 __read_mostly tsc_tolerance_ppm = 250;

@@ -3595,6 +3597,13 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_STEAL_TIME:
 		r = sched_info_on();
 		break;
+	case KVM_CAP_X86_BUS_LOCK_EXIT:
+		if (kvm_has_bus_lock_exit)
+			r = KVM_BUS_LOCK_DETECTION_OFF |
+			    KVM_BUS_LOCK_DETECTION_EXIT;
+		else
+			r = 0;
+		break;
 	default:
 		break;
 	}

@@ -5047,6 +5056,20 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 		kvm->arch.exception_payload_enabled = cap->args[0];
 		r = 0;
 		break;
+	case KVM_CAP_X86_BUS_LOCK_EXIT:
+		r = -EINVAL;
+		if (cap->args[0] & ~KVM_BUS_LOCK_DETECTION_VALID_MODE)
+			break;
+
+		if ((cap->args[0] & KVM_BUS_LOCK_DETECTION_OFF) &&
+		    (cap->args[0] & KVM_BUS_LOCK_DETECTION_EXIT))
+			break;
+
+		if (kvm_has_bus_lock_exit &&
+		    cap->args[0] & KVM_BUS_LOCK_DETECTION_EXIT)
+			kvm->arch.bus_lock_detection_enabled = true;
+		r = 0;
+		break;
 	default:
 		r = -EINVAL;
 		break;

@@ -7789,12 +7812,16 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu)
 	struct kvm_run *kvm_run = vcpu->run;
 
 	kvm_run->if_flag = (kvm_get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
-	kvm_run->flags = is_smm(vcpu) ? KVM_RUN_X86_SMM : 0;
 	kvm_run->cr8 = kvm_get_cr8(vcpu);
 	kvm_run->apic_base = kvm_get_apic_base(vcpu);
 	kvm_run->ready_for_interrupt_injection =
 		pic_in_kernel(vcpu->kvm) ||
 		kvm_vcpu_ready_for_interrupt_injection(vcpu);
+
+	if (is_smm(vcpu))
+		kvm_run->flags |= KVM_RUN_X86_SMM;
+	else
+		kvm_run->flags &= ~KVM_RUN_X86_SMM;
 }
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu)
include/uapi/linux/kvm.h +5 -0
index 7d8eced6f459..6cb25e909a82 100644
@@ -248,6 +248,7 @@ struct kvm_hyperv_exit {
 #define KVM_EXIT_IOAPIC_EOI       26
 #define KVM_EXIT_HYPERV           27
 #define KVM_EXIT_ARM_NISV         28
+#define KVM_EXIT_BUS_LOCK         29
 
 /* For KVM_EXIT_INTERNAL_ERROR */
 /* Emulate instruction failed. */

@@ -1037,6 +1038,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_SMALLER_MAXPHYADDR 185
 #define KVM_CAP_S390_DIAG318 186
 #define KVM_CAP_STEAL_TIME 187
+#define KVM_CAP_X86_BUS_LOCK_EXIT 188
 
 #ifdef KVM_CAP_IRQ_ROUTING
 

@@ -1691,4 +1693,7 @@ struct kvm_hyperv_eventfd {
 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE    (1 << 0)
 #define KVM_DIRTY_LOG_INITIALLY_SET            (1 << 1)
 
+#define KVM_BUS_LOCK_DETECTION_OFF             (1 << 0)
+#define KVM_BUS_LOCK_DETECTION_EXIT            (1 << 1)
+
 #endif /* __LINUX_KVM_H */
Signature · 1 line
2.17.1
#5 lore raw

Re: [RESEND v4 2/2] KVM: VMX: Enable bus lock VM exit

Quoted 8 lines · Chenyi Qiang
arch/x86/kvm/vmx/vmx.c +4 -0
@@ -6138,6 +6149,26 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 	return 0;
 }
 
+static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
+{
+	int ret = __vmx_handle_exit(vcpu, exit_fastpath);
+

arch/x86/kvm/vmx/vmx.c static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) · line 6165

Quoted 10 lines
arch/x86/kvm/vmx/vmx.c +10 -0
+	/*
+	 * Even when current exit reason is handled by KVM internally, we
+	 * still need to exit to user space when bus lock detected to inform
+	 * that there is a bus lock in guest.
+	 */
+	if (to_vmx(vcpu)->exit_reason.bus_lock_detected) {
+		if (ret > 0)
+			vcpu->run->exit_reason = KVM_EXIT_BUS_LOCK;
+		else
+			vcpu->run->flags |= KVM_RUN_BUS_LOCK;

This should always set flags.KVM_RUN_BUS_LOCK, e.g. so that userspace can always check flags.KVM_RUN_BUS_LOCK instead of having to check both the flag and the exit reason. As is, it's really bad because the flag is undefined, which could teach userspace to do the wrong thing.

arch/x86/kvm/vmx/vmx.c static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) · line 6168

Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +3 -0
+		return 0;
+	}
+	vcpu->run->flags &= ~KVM_RUN_BUS_LOCK;

Hmm, I feel like explicitly clearing flags is should be unnecessary. By that, I mean that's it's necessary in the current patch, bit I think we should figure out how to make that not be the case. With the current approach, every chunk of code that needs to set a flag also needs to clear it, which increases the odds of missing a case and ending up with a flag in an undefined state.

The easiest way I can think of is to add another prep patch that zeros run->flags at the beginning of kvm_arch_vcpu_ioctl_run(), and changes post_kvm_run_save() to do:

	if (is_smm(vcpu))
		kvm_run->flags |= KVM_RUN_X86_SMM;

Then this patch can omit clearing KVM_RUN_BUS_LOCK, and doesn't have to touch the SMM flag.

Quoted 6 lines
arch/x86/kvm/vmx/vmx.c +3 -0
+	return ret;
+}
+
 /*
  * Software based L1D cache flush which is used when microcode providing
  * the cache control MSR is not loaded.
author
#6 lore raw

Re: [RESEND v4 2/2] KVM: VMX: Enable bus lock VM exit

Quoted 16 lines · Sean Christopherson
Quoted 15 lines · Chenyi Qiang
arch/x86/kvm/vmx/vmx.c +11 -0
@@ -6138,6 +6149,26 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
  	return 0;
  }
 
+static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
+{
+	int ret = __vmx_handle_exit(vcpu, exit_fastpath);
+
+	/*
+	 * Even when current exit reason is handled by KVM internally, we
+	 * still need to exit to user space when bus lock detected to inform
+	 * that there is a bus lock in guest.
+	 */
+	if (to_vmx(vcpu)->exit_reason.bus_lock_detected) {
+		if (ret > 0)
Quoted 9 lines

arch/x86/kvm/vmx/vmx.c static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) · line 6165

Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +3 -0
+			vcpu->run->exit_reason = KVM_EXIT_BUS_LOCK;
+		else
+			vcpu->run->flags |= KVM_RUN_BUS_LOCK;

This should always set flags.KVM_RUN_BUS_LOCK, e.g. so that userspace can always check flags.KVM_RUN_BUS_LOCK instead of having to check both the flag and the exit reason. As is, it's really bad because the flag is undefined, which could teach userspace to do the wrong thing.

Got it.

Quoted 14 lines

arch/x86/kvm/vmx/vmx.c static int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) · line 6168

Quoted 3 lines
arch/x86/kvm/vmx/vmx.c +3 -0
+		return 0;
+	}
+	vcpu->run->flags &= ~KVM_RUN_BUS_LOCK;

Hmm, I feel like explicitly clearing flags is should be unnecessary. By that, I mean that's it's necessary in the current patch, bit I think we should figure out how to make that not be the case. With the current approach, every chunk of code that needs to set a flag also needs to clear it, which increases the odds of missing a case and ending up with a flag in an undefined state.

The easiest way I can think of is to add another prep patch that zeros run->flags at the beginning of kvm_arch_vcpu_ioctl_run(), and changes post_kvm_run_save() to do:

Quoted 6 lines
	if (is_smm(vcpu))
		kvm_run->flags |= KVM_RUN_X86_SMM;

Then this patch can omit clearing KVM_RUN_BUS_LOCK, and doesn't have to touch the SMM flag.

Make sense, I will add the prep patch.

Quoted 6 lines
Quoted 6 lines
arch/x86/kvm/vmx/vmx.c +3 -0
+	return ret;
+}
+
  /*
   * Software based L1D cache flush which is used when microcode providing
   * the cache control MSR is not loaded.