Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull kvm updates from Paolo Bonzini:
"S390:

- ultravisor communication device driver

- fix TEID on terminating storage key ops

RISC-V:

- Added Sv57x4 support for G-stage page table

- Added range based local HFENCE functions

- Added remote HFENCE functions based on VCPU requests

- Added ISA extension registers in ONE_REG interface

- Updated KVM RISC-V maintainers entry to cover selftests support

ARM:

- Add support for the ARMv8.6 WFxT extension

- Guard pages for the EL2 stacks

- Trap and emulate AArch32 ID registers to hide unsupported features

- Ability to select and save/restore the set of hypercalls exposed to
the guest

- Support for PSCI-initiated suspend in collaboration with userspace

- GICv3 register-based LPI invalidation support

- Move host PMU event merging into the vcpu data structure

- GICv3 ITS save/restore fixes

- The usual set of small-scale cleanups and fixes

x86:

- New ioctls to get/set TSC frequency for a whole VM

- Allow userspace to opt out of hypercall patching

- Only do MSR filtering for MSRs accessed by rdmsr/wrmsr

AMD SEV improvements:

- Add KVM_EXIT_SHUTDOWN metadata for SEV-ES

- V_TSC_AUX support

Nested virtualization improvements for AMD:

- Support for "nested nested" optimizations (nested vVMLOAD/VMSAVE,
nested vGIF)

- Allow AVIC to co-exist with a nested guest running

- Fixes for LBR virtualizations when a nested guest is running, and
nested LBR virtualization support

- PAUSE filtering for nested hypervisors

Guest support:

- Decoupling of vcpu_is_preempted from PV spinlocks"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (199 commits)
KVM: x86: Fix the intel_pt PMI handling wrongly considered from guest
KVM: selftests: x86: Sync the new name of the test case to .gitignore
Documentation: kvm: reorder ARM-specific section about KVM_SYSTEM_EVENT_SUSPEND
x86, kvm: use correct GFP flags for preemption disabled
KVM: LAPIC: Drop pending LAPIC timer injection when canceling the timer
x86/kvm: Alloc dummy async #PF token outside of raw spinlock
KVM: x86: avoid calling x86 emulator without a decoded instruction
KVM: SVM: Use kzalloc for sev ioctl interfaces to prevent kernel data leak
x86/fpu: KVM: Set the base guest FPU uABI size to sizeof(struct kvm_xsave)
s390/uv_uapi: depend on CONFIG_S390
KVM: selftests: x86: Fix test failure on arch lbr capable platforms
KVM: LAPIC: Trace LAPIC timer expiration on every vmentry
KVM: s390: selftest: Test suppression indication on key prot exception
KVM: s390: Don't indicate suppression on dirtying, failing memop
selftests: drivers/s390x: Add uvdevice tests
drivers/s390/char: Add Ultravisor io device
MAINTAINERS: Update KVM RISC-V entry to cover selftests support
RISC-V: KVM: Introduce ISA extension register
RISC-V: KVM: Cleanup stale TLB entries when host CPU changes
RISC-V: KVM: Add remote HFENCE functions based on VCPU requests
...

Documentation/arm64/cpu-feature-registers.rst

···
      +------------------------------+---------+---------+
      | RPRES                        | [7-4]   |    y    |
      +------------------------------+---------+---------+
+     | WFXT                         | [3-0]   |    y    |
+     +------------------------------+---------+---------+
 
 
 Appendix I: Example

Documentation/arm64/elf_hwcaps.rst

···
 
     Functionality implied by ID_AA64SMFR0_EL1.FA64 == 0b1.
 
+HWCAP2_WFXT
+
+    Functionality implied by ID_AA64ISAR2_EL1.WFXT == 0b0010.
+
 4. Unused AT_HWCAP bits
 -----------------------
 

Documentation/virt/kvm/api.rst

···
 	__u8 pad2[30];
   };
 
-If the KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag is returned from the
-KVM_CAP_XEN_HVM check, it may be set in the flags field of this ioctl.
-This requests KVM to generate the contents of the hypercall page
-automatically; hypercalls will be intercepted and passed to userspace
-through KVM_EXIT_XEN.  In this case, all of the blob size and address
-fields must be zero.
+If certain flags are returned from the KVM_CAP_XEN_HVM check, they may
+be set in the flags field of this ioctl:
+
+The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag requests KVM to generate
+the contents of the hypercall page automatically; hypercalls will be
+intercepted and passed to userspace through KVM_EXIT_XEN.  In this
+ase, all of the blob size and address fields must be zero.
+
+The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates to KVM that userspace
+will always use the KVM_XEN_HVM_EVTCHN_SEND ioctl to deliver event
+channel interrupts rather than manipulating the guest's shared_info
+structures directly. This, in turn, may allow KVM to enable features
+such as intercepting the SCHEDOP_poll hypercall to accelerate PV
+spinlock operation for the guest. Userspace may still use the ioctl
+to deliver events if it was advertised, even if userspace does not
+send this indication that it will always do so
 
 No other flags are currently valid in the struct kvm_xen_hvm_config.
 
···
                                  [s390]
    KVM_MP_STATE_LOAD             the vcpu is in a special load/startup state
                                  [s390]
+   KVM_MP_STATE_SUSPENDED        the vcpu is in a suspend state and is waiting
+                                 for a wakeup event [arm64]
    ==========================    ===============================================
 
 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
 these architectures.
 
-For arm64/riscv:
-^^^^^^^^^^^^^^^^
+For arm64:
+^^^^^^^^^^
+
+If a vCPU is in the KVM_MP_STATE_SUSPENDED state, KVM will emulate the
+architectural execution of a WFI instruction.
+
+If a wakeup event is recognized, KVM will exit to userspace with a
+KVM_SYSTEM_EVENT exit, where the event type is KVM_SYSTEM_EVENT_WAKEUP. If
+userspace wants to honor the wakeup, it must set the vCPU's MP state to
+KVM_MP_STATE_RUNNABLE. If it does not, KVM will continue to await a wakeup
+event in subsequent calls to KVM_RUN.
+
+.. warning::
+
+     If userspace intends to keep the vCPU in a SUSPENDED state, it is
+     strongly recommended that userspace take action to suppress the
+     wakeup event (such as masking an interrupt). Otherwise, subsequent
+     calls to KVM_RUN will immediately exit with a KVM_SYSTEM_EVENT_WAKEUP
+     event and inadvertently waste CPU cycles.
+
+     Additionally, if userspace takes action to suppress a wakeup event,
+     it is strongly recommended that it also restores the vCPU to its
+     original state when the vCPU is made RUNNABLE again. For example,
+     if userspace masked a pending interrupt to suppress the wakeup,
+     the interrupt should be unmasked before returning control to the
+     guest.
+
+For riscv:
+^^^^^^^^^^
 
 The only states that are valid are KVM_MP_STATE_STOPPED and
 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
···
 4.55 KVM_SET_TSC_KHZ
 --------------------
 
-:Capability: KVM_CAP_TSC_CONTROL
+:Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM_TSC_CONTROL
 :Architectures: x86
-:Type: vcpu ioctl
+:Type: vcpu ioctl / vm ioctl
 :Parameters: virtual tsc_khz
 :Returns: 0 on success, -1 on error
 
 Specifies the tsc frequency for the virtual machine. The unit of the
 frequency is KHz.
 
+If the KVM_CAP_VM_TSC_CONTROL capability is advertised, this can also
+be used as a vm ioctl to set the initial tsc frequency of subsequently
+created vCPUs.
 
 4.56 KVM_GET_TSC_KHZ
 --------------------
 
-:Capability: KVM_CAP_GET_TSC_KHZ
+:Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM_TSC_CONTROL
 :Architectures: x86
-:Type: vcpu ioctl
+:Type: vcpu ioctl / vm ioctl
 :Parameters: none
 :Returns: virtual tsc-khz on success, negative value on error
 
···
 
 After the vcpu's SVE configuration is finalized, further attempts to
 write this register will fail with EPERM.
+
+arm64 bitmap feature firmware pseudo-registers have the following bit pattern::
+
+  0x6030 0000 0016 <regno:16>
+
+The bitmap feature firmware registers exposes the hypercall services that
+are available for userspace to configure. The set bits corresponds to the
+services that are available for the guests to access. By default, KVM
+sets all the supported bits during VM initialization. The userspace can
+discover the available services via KVM_GET_ONE_REG, and write back the
+bitmap corresponding to the features that it wishes guests to see via
+KVM_SET_ONE_REG.
+
+Note: These registers are immutable once any of the vCPUs of the VM has
+run at least once. A KVM_SET_ONE_REG in such a scenario will return
+a -EBUSY to userspace.
+
+(See Documentation/virt/kvm/arm/hypercalls.rst for more details.)
 
 
 MIPS registers are mapped using the lower 32 bits.  The upper 16 of that is
···
 error number indicating the type of exception. This exception is also
 raised directly at the corresponding VCPU if the flag
 KVM_S390_MEMOP_F_INJECT_EXCEPTION is set.
+On protection exceptions, unless specified otherwise, the injected
+translation-exception identifier (TEID) indicates suppression.
 
 If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key
 protection is also in effect and may cause exceptions if accesses are
 prohibited given the access key designated by "key"; the valid range is 0..15.
 KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION
 is > 0.
+Since the accessed memory may span multiple pages and those pages might have
+different storage keys, it is possible that a protection exception occurs
+after memory has been modified. In this case, if the exception is injected,
+the TEID does not indicate suppression.
 
 Absolute read/write:
 ^^^^^^^^^^^^^^^^^^^^
···
 		struct {
 			__u64 gfn;
 		} shared_info;
-		__u64 pad[4];
+		struct {
+			__u32 send_port;
+			__u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */
+			__u32 flags;
+			union {
+				struct {
+					__u32 port;
+					__u32 vcpu;
+					__u32 priority;
+				} port;
+				struct {
+					__u32 port; /* Zero for eventfd */
+					__s32 fd;
+				} eventfd;
+				__u32 padding[4];
+			} deliver;
+		} evtchn;
+		__u32 xen_version;
+		__u64 pad[8];
 	} u;
   };
 
···
 
 KVM_XEN_ATTR_TYPE_UPCALL_VECTOR
   Sets the exception vector used to deliver Xen event channel upcalls.
+  This is the HVM-wide vector injected directly by the hypervisor
+  (not through the local APIC), typically configured by a guest via
+  HVM_PARAM_CALLBACK_IRQ.
+
+KVM_XEN_ATTR_TYPE_EVTCHN
+  This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
+  support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures
+  an outbound port number for interception of EVTCHNOP_send requests
+  from the guest. A given sending port number may be directed back
+  to a specified vCPU (by APIC ID) / port / priority on the guest,
+  or to trigger events on an eventfd. The vCPU and priority can be
+  changed by setting KVM_XEN_EVTCHN_UPDATE in a subsequent call,
+  but other fields cannot change for a given sending port. A port
+  mapping is removed by using KVM_XEN_EVTCHN_DEASSIGN in the flags
+  field.
+
+KVM_XEN_ATTR_TYPE_XEN_VERSION
+  This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
+  support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures
+  the 32-bit version code returned to the guest when it invokes the
+  XENVER_version call; typically (XEN_MAJOR << 16 | XEN_MINOR). PV
+  Xen guests will often use this to as a dummy hypercall to trigger
+  event channel delivery, so responding within the kernel without
+  exiting to userspace is beneficial.
 
 4.127 KVM_XEN_HVM_GET_ATTR
 --------------------------
···
 :Returns: 0 on success, < 0 on error
 
 Allows Xen VM attributes to be read. For the structure and types,
-see KVM_XEN_HVM_SET_ATTR above.
+see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_ATTR_TYPE_EVTCHN
+attribute cannot be read.
 
 4.128 KVM_XEN_VCPU_SET_ATTR
 ---------------------------
···
 			__u64 time_blocked;
 			__u64 time_offline;
 		} runstate;
+		__u32 vcpu_id;
+		struct {
+			__u32 port;
+			__u32 priority;
+			__u64 expires_ns;
+		} timer;
+		__u8 vector;
 	} u;
   };
 
···
   runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked
   or RUNSTATE_offline) to set the current accounted state as of the
   adjusted state_entry_time.
+
+KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID
+  This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
+  support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the Xen
+  vCPU ID of the given vCPU, to allow timer-related VCPU operations to
+  be intercepted by KVM.
+
+KVM_XEN_VCPU_ATTR_TYPE_TIMER
+  This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
+  support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the
+  event channel port/priority for the VIRQ_TIMER of the vCPU, as well
+  as allowing a pending timer to be saved/restored.
+
+KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR
+  This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
+  support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the
+  per-vCPU local APIC upcall vector, configured by a Xen guest with
+  the HVMOP_set_evtchn_upcall_vector hypercall. This is typically
+  used by Windows guests, and is distinct from the HVM-wide upcall
+  vector configured with HVM_PARAM_CALLBACK_IRQ.
+
 
 4.129 KVM_XEN_VCPU_GET_ATTR
 ---------------------------
···
 The offsets of the state save areas in struct kvm_xsave follow the contents
 of CPUID leaf 0xD on the host.
 
+4.135 KVM_XEN_HVM_EVTCHN_SEND
+-----------------------------
+
+:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND
+:Architectures: x86
+:Type: vm ioctl
+:Parameters: struct kvm_irq_routing_xen_evtchn
+:Returns: 0 on success, < 0 on error
+
+
+::
+
+   struct kvm_irq_routing_xen_evtchn {
+	__u32 port;
+	__u32 vcpu;
+	__u32 priority;
+   };
+
+This ioctl injects an event channel interrupt directly to the guest vCPU.
 
 5. The kvm_run structure
 ========================
···
   #define KVM_SYSTEM_EVENT_SHUTDOWN       1
   #define KVM_SYSTEM_EVENT_RESET          2
   #define KVM_SYSTEM_EVENT_CRASH          3
+  #define KVM_SYSTEM_EVENT_WAKEUP         4
+  #define KVM_SYSTEM_EVENT_SUSPEND        5
+  #define KVM_SYSTEM_EVENT_SEV_TERM       6
 			__u32 type;
                         __u32 ndata;
                         __u64 data[16];
···
    has requested a crash condition maintenance. Userspace can choose
    to ignore the request, or to gather VM memory core dump and/or
    reset/shutdown of the VM.
+ - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination.
+   The guest physical address of the guest's GHCB is stored in `data[0]`.
+ - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and
+   KVM has recognized a wakeup event. Userspace may honor this event by
+   marking the exiting vCPU as runnable, or deny it and call KVM_RUN again.
+ - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested a suspension of
+   the VM.
 
 If KVM_CAP_SYSTEM_EVENT_DATA is present, the 'data' field can contain
 architecture specific information for the system-level event.  Only
···
 Previous versions of Linux defined a `flags` member in this struct.  The
 field is now aliased to `data[0]`.  Userspace can assume that it is only
 written if ndata is greater than 0.
+
+For arm/arm64:
+--------------
+
+KVM_SYSTEM_EVENT_SUSPEND exits are enabled with the
+KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If a guest invokes the PSCI
+SYSTEM_SUSPEND function, KVM will exit to userspace with this event
+type.
+
+It is the sole responsibility of userspace to implement the PSCI
+SYSTEM_SUSPEND call according to ARM DEN0022D.b 5.19 "SYSTEM_SUSPEND".
+KVM does not change the vCPU's state before exiting to userspace, so
+the call parameters are left in-place in the vCPU registers.
+
+Userspace is _required_ to take action for such an exit. It must
+either:
+
+ - Honor the guest request to suspend the VM. Userspace can request
+   in-kernel emulation of suspension by setting the calling vCPU's
+   state to KVM_MP_STATE_SUSPENDED. Userspace must configure the vCPU's
+   state according to the parameters passed to the PSCI function when
+   the calling vCPU is resumed. See ARM DEN0022D.b 5.19.1 "Intended use"
+   for details on the function parameters.
+
+ - Deny the guest request to suspend the VM. See ARM DEN0022D.b 5.19.2
+   "Caller responsibilities" for possible return values.
 
 ::
 
···
                                     Additionally, when this quirk is disabled,
                                     KVM clears CPUID.01H:ECX[bit 3] if
                                     IA32_MISC_ENABLE[bit 18] is cleared.
+
+ KVM_X86_QUIRK_FIX_HYPERCALL_INSN   By default, KVM rewrites guest
+                                    VMMCALL/VMCALL instructions to match the
+                                    vendor's hypercall instruction for the
+                                    system. When this quirk is disabled, KVM
+                                    will no longer rewrite invalid guest
+                                    hypercall instructions. Executing the
+                                    incorrect hypercall instruction will
+                                    generate a #UD within the guest.
 =================================== ============================================
 
 8. Other capabilities.
···
   #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR	(1 << 0)
   #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL	(1 << 1)
   #define KVM_XEN_HVM_CONFIG_SHARED_INFO	(1 << 2)
-  #define KVM_XEN_HVM_CONFIG_RUNSTATE		(1 << 2)
-  #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL	(1 << 3)
+  #define KVM_XEN_HVM_CONFIG_RUNSTATE		(1 << 3)
+  #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL	(1 << 4)
+  #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND	(1 << 5)
 
 The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG
 ioctl is available, for the guest to set its hypercall page.
···
 The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries
 of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority
 field set to indicate 2 level event channel delivery.
+
+The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates that KVM supports
+injecting event channel events directly into the guest with the
+KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indicates support for the
+KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attributes and the
+KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_VECTOR vCPU attributes.
+related to event channel delivery, timers, and the XENVER_version
+interception.
 
 8.31 KVM_CAP_PPC_MULTITCE
 -------------------------
···
 At this time, KVM_PMU_CAP_DISABLE is the only capability.  Setting
 this capability will disable PMU virtualization for that VM.  Usermode
 should adjust CPUID leaf 0xA to reflect that the PMU is disabled.
+
+8.36 KVM_CAP_ARM_SYSTEM_SUSPEND
+-------------------------------
+
+:Capability: KVM_CAP_ARM_SYSTEM_SUSPEND
+:Architectures: arm64
+:Type: vm
+
+When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of
+type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request.
 
 9. Known KVM API problems
 =========================

Documentation/virt/kvm/arm/hypercalls.rst

···
+.. SPDX-License-Identifier: GPL-2.0
+
+=======================
+ARM Hypercall Interface
+=======================
+
+KVM handles the hypercall services as requested by the guests. New hypercall
+services are regularly made available by the ARM specification or by KVM (as
+vendor services) if they make sense from a virtualization point of view.
+
+This means that a guest booted on two different versions of KVM can observe
+two different "firmware" revisions. This could cause issues if a given guest
+is tied to a particular version of a hypercall service, or if a migration
+causes a different version to be exposed out of the blue to an unsuspecting
+guest.
+
+In order to remedy this situation, KVM exposes a set of "firmware
+pseudo-registers" that can be manipulated using the GET/SET_ONE_REG
+interface. These registers can be saved/restored by userspace, and set
+to a convenient value as required.
+
+The following registers are defined:
+
+* KVM_REG_ARM_PSCI_VERSION:
+
+  KVM implements the PSCI (Power State Coordination Interface)
+  specification in order to provide services such as CPU on/off, reset
+  and power-off to the guest.
+
+  - Only valid if the vcpu has the KVM_ARM_VCPU_PSCI_0_2 feature set
+    (and thus has already been initialized)
+  - Returns the current PSCI version on GET_ONE_REG (defaulting to the
+    highest PSCI version implemented by KVM and compatible with v0.2)
+  - Allows any PSCI version implemented by KVM and compatible with
+    v0.2 to be set with SET_ONE_REG
+  - Affects the whole VM (even if the register view is per-vcpu)
+
+* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+    Holds the state of the firmware support to mitigate CVE-2017-5715, as
+    offered by KVM to the guest via a HVC call. The workaround is described
+    under SMCCC_ARCH_WORKAROUND_1 in [1].
+
+  Accepted values are:
+
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL:
+      KVM does not offer
+      firmware support for the workaround. The mitigation status for the
+      guest is unknown.
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL:
+      The workaround HVC call is
+      available to the guest and required for the mitigation.
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED:
+      The workaround HVC call
+      is available to the guest, but it is not needed on this VCPU.
+
+* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+    Holds the state of the firmware support to mitigate CVE-2018-3639, as
+    offered by KVM to the guest via a HVC call. The workaround is described
+    under SMCCC_ARCH_WORKAROUND_2 in [1]_.
+
+  Accepted values are:
+
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL:
+      A workaround is not
+      available. KVM does not offer firmware support for the workaround.
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN:
+      The workaround state is
+      unknown. KVM does not offer firmware support for the workaround.
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL:
+      The workaround is available,
+      and can be disabled by a vCPU. If
+      KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED is set, it is active for
+      this vCPU.
+    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED:
+      The workaround is always active on this vCPU or it is not needed.
+
+
+Bitmap Feature Firmware Registers
+---------------------------------
+
+Contrary to the above registers, the following registers exposes the
+hypercall services in the form of a feature-bitmap to the userspace. This
+bitmap is translated to the services that are available to the guest.
+There is a register defined per service call owner and can be accessed via
+GET/SET_ONE_REG interface.
+
+By default, these registers are set with the upper limit of the features
+that are supported. This way userspace can discover all the usable
+hypercall services via GET_ONE_REG. The user-space can write-back the
+desired bitmap back via SET_ONE_REG. The features for the registers that
+are untouched, probably because userspace isn't aware of them, will be
+exposed as is to the guest.
+
+Note that KVM will not allow the userspace to configure the registers
+anymore once any of the vCPUs has run at least once. Instead, it will
+return a -EBUSY.
+
+The pseudo-firmware bitmap register are as follows:
+
+* KVM_REG_ARM_STD_BMAP:
+    Controls the bitmap of the ARM Standard Secure Service Calls.
+
+  The following bits are accepted:
+
+    Bit-0: KVM_REG_ARM_STD_BIT_TRNG_V1_0:
+      The bit represents the services offered under v1.0 of ARM True Random
+      Number Generator (TRNG) specification, ARM DEN0098.
+
+* KVM_REG_ARM_STD_HYP_BMAP:
+    Controls the bitmap of the ARM Standard Hypervisor Service Calls.
+
+  The following bits are accepted:
+
+    Bit-0: KVM_REG_ARM_STD_HYP_BIT_PV_TIME:
+      The bit represents the Paravirtualized Time service as represented by
+      ARM DEN0057A.
+
+* KVM_REG_ARM_VENDOR_HYP_BMAP:
+    Controls the bitmap of the Vendor specific Hypervisor Service Calls.
+
+  The following bits are accepted:
+
+    Bit-0: KVM_REG_ARM_VENDOR_HYP_BIT_FUNC_FEAT
+      The bit represents the ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID
+      and ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID function-ids.
+
+    Bit-1: KVM_REG_ARM_VENDOR_HYP_BIT_PTP:
+      The bit represents the Precision Time Protocol KVM service.
+
+Errors:
+
+    =======  =============================================================
+    -ENOENT   Unknown register accessed.
+    -EBUSY    Attempt a 'write' to the register after the VM has started.
+    -EINVAL   Invalid bitmap written to the register.
+    =======  =============================================================
+
+.. [1] https://developer.arm.com/-/media/developer/pdf/ARM_DEN_0070A_Firmware_interfaces_for_mitigating_CVE-2017-5715.pdf

Documentation/virt/kvm/arm/index.rst

···
    :maxdepth: 2
 
    hyp-abi
-   psci
+   hypercalls
    pvtime
    ptp_kvm

Documentation/virt/kvm/arm/psci.rst

···
-.. SPDX-License-Identifier: GPL-2.0
-
-=========================================
-Power State Coordination Interface (PSCI)
-=========================================
-
-KVM implements the PSCI (Power State Coordination Interface)
-specification in order to provide services such as CPU on/off, reset
-and power-off to the guest.
-
-The PSCI specification is regularly updated to provide new features,
-and KVM implements these updates if they make sense from a virtualization
-point of view.
-
-This means that a guest booted on two different versions of KVM can
-observe two different "firmware" revisions. This could cause issues if
-a given guest is tied to a particular PSCI revision (unlikely), or if
-a migration causes a different PSCI version to be exposed out of the
-blue to an unsuspecting guest.
-
-In order to remedy this situation, KVM exposes a set of "firmware
-pseudo-registers" that can be manipulated using the GET/SET_ONE_REG
-interface. These registers can be saved/restored by userspace, and set
-to a convenient value if required.
-
-The following register is defined:
-
-* KVM_REG_ARM_PSCI_VERSION:
-
-  - Only valid if the vcpu has the KVM_ARM_VCPU_PSCI_0_2 feature set
-    (and thus has already been initialized)
-  - Returns the current PSCI version on GET_ONE_REG (defaulting to the
-    highest PSCI version implemented by KVM and compatible with v0.2)
-  - Allows any PSCI version implemented by KVM and compatible with
-    v0.2 to be set with SET_ONE_REG
-  - Affects the whole VM (even if the register view is per-vcpu)
-
-* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
-    Holds the state of the firmware support to mitigate CVE-2017-5715, as
-    offered by KVM to the guest via a HVC call. The workaround is described
-    under SMCCC_ARCH_WORKAROUND_1 in [1].
-
-  Accepted values are:
-
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL:
-      KVM does not offer
-      firmware support for the workaround. The mitigation status for the
-      guest is unknown.
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL:
-      The workaround HVC call is
-      available to the guest and required for the mitigation.
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED:
-      The workaround HVC call
-      is available to the guest, but it is not needed on this VCPU.
-
-* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
-    Holds the state of the firmware support to mitigate CVE-2018-3639, as
-    offered by KVM to the guest via a HVC call. The workaround is described
-    under SMCCC_ARCH_WORKAROUND_2 in [1]_.
-
-  Accepted values are:
-
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL:
-      A workaround is not
-      available. KVM does not offer firmware support for the workaround.
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN:
-      The workaround state is
-      unknown. KVM does not offer firmware support for the workaround.
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL:
-      The workaround is available,
-      and can be disabled by a vCPU. If
-      KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED is set, it is active for
-      this vCPU.
-    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED:
-      The workaround is always active on this vCPU or it is not needed.
-
-.. [1] https://developer.arm.com/-/media/developer/pdf/ARM_DEN_0070A_Firmware_interfaces_for_mitigating_CVE-2017-5715.pdf

Documentation/virt/kvm/x86/mmu.rst

···
     Is 1 if the MMU instance cannot use A/D bits.  EPT did not have A/D
     bits before Haswell; shadow EPT page tables also cannot use A/D bits
     if the L1 hypervisor does not enable them.
+  role.passthrough:
+    The page is not backed by a guest page table, but its first entry
+    points to one.  This is set if NPT uses 5-level page tables (host
+    CR4.LA57=1) and is shadowing L1's 4-level NPT (L1 CR4.LA57=1).
   gfn:
     Either the guest page table containing the translations shadowed by this
     page, or the base page frame for linear translations.  See role.direct.

MAINTAINERS

···
 F:	arch/riscv/include/asm/kvm*
 F:	arch/riscv/include/uapi/asm/kvm*
 F:	arch/riscv/kvm/
+F:	tools/testing/selftests/kvm/*/riscv/
+F:	tools/testing/selftests/kvm/riscv/
 
 KERNEL VIRTUAL MACHINE for s390 (KVM/s390)
 M:	Christian Borntraeger <[e-mail]>
···
 F:	arch/s390/include/asm/gmap.h
 F:	arch/s390/include/asm/kvm*
 F:	arch/s390/include/uapi/asm/kvm*
+F:	arch/s390/include/uapi/asm/uvdevice.h
 F:	arch/s390/kernel/uv.c
 F:	arch/s390/kvm/
 F:	arch/s390/mm/gmap.c
+F:	drivers/s390/char/uvdevice.c
+F:	tools/testing/selftests/drivers/s390x/uvdevice/
 F:	tools/testing/selftests/kvm/*/s390x/
 F:	tools/testing/selftests/kvm/s390x/
 

arch/arm64/include/asm/barrier.h

···
 
 #define sev()		asm volatile("sev" : : : "memory")
 #define wfe()		asm volatile("wfe" : : : "memory")
+#define wfet(val)	asm volatile("msr s0_3_c1_c0_0, %0"	\
+				     : : "r" (val) : "memory")
 #define wfi()		asm volatile("wfi" : : : "memory")
+#define wfit(val)	asm volatile("msr s0_3_c1_c0_1, %0"	\
+				     : : "r" (val) : "memory")
 
 #define isb()		asm volatile("isb" : : : "memory")
 #define dmb(opt)	asm volatile("dmb " #opt : : : "memory")

arch/arm64/include/asm/cputype.h

···
 
 #define APPLE_CPU_PART_M1_ICESTORM	0x022
 #define APPLE_CPU_PART_M1_FIRESTORM	0x023
+#define APPLE_CPU_PART_M1_ICESTORM_PRO	0x024
+#define APPLE_CPU_PART_M1_FIRESTORM_PRO	0x025
+#define APPLE_CPU_PART_M1_ICESTORM_MAX	0x028
+#define APPLE_CPU_PART_M1_FIRESTORM_MAX	0x029
 
 #define MIDR_CORTEX_A53 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A53)
 #define MIDR_CORTEX_A57 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A57)
···
 #define MIDR_HISI_TSV110 MIDR_CPU_MODEL(ARM_CPU_IMP_HISI, HISI_CPU_PART_TSV110)
 #define MIDR_APPLE_M1_ICESTORM MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_ICESTORM)
 #define MIDR_APPLE_M1_FIRESTORM MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_FIRESTORM)
+#define MIDR_APPLE_M1_ICESTORM_PRO MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_ICESTORM_PRO)
+#define MIDR_APPLE_M1_FIRESTORM_PRO MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_FIRESTORM_PRO)
+#define MIDR_APPLE_M1_ICESTORM_MAX MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_ICESTORM_MAX)
+#define MIDR_APPLE_M1_FIRESTORM_MAX MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_FIRESTORM_MAX)
 
 /* Fujitsu Erratum 010001 affects A64FX 1.0 and 1.1, (v0r0 and v1r0) */
 #define MIDR_FUJITSU_ERRATUM_010001		MIDR_FUJITSU_A64FX

arch/arm64/include/asm/esr.h

···
 #define ESR_ELx_CV		(UL(1) << 24)
 #define ESR_ELx_COND_SHIFT	(20)
 #define ESR_ELx_COND_MASK	(UL(0xF) << ESR_ELx_COND_SHIFT)
-#define ESR_ELx_WFx_ISS_TI	(UL(1) << 0)
+#define ESR_ELx_WFx_ISS_RN	(UL(0x1F) << 5)
+#define ESR_ELx_WFx_ISS_RV	(UL(1) << 2)
+#define ESR_ELx_WFx_ISS_TI	(UL(3) << 0)
+#define ESR_ELx_WFx_ISS_WFxT	(UL(2) << 0)
 #define ESR_ELx_WFx_ISS_WFI	(UL(0) << 0)
 #define ESR_ELx_WFx_ISS_WFE	(UL(1) << 0)
 #define ESR_ELx_xVC_IMM_MASK	((UL(1) << 16) - 1)
···
 #define DISR_EL1_ESR_MASK	(ESR_ELx_AET | ESR_ELx_EA | ESR_ELx_FSC)
 
 /* ESR value templates for specific events */
-#define ESR_ELx_WFx_MASK	(ESR_ELx_EC_MASK | ESR_ELx_WFx_ISS_TI)
+#define ESR_ELx_WFx_MASK	(ESR_ELx_EC_MASK |			\
+				 (ESR_ELx_WFx_ISS_TI & ~ESR_ELx_WFx_ISS_WFxT))
 #define ESR_ELx_WFx_WFI_VAL	((ESR_ELx_EC_WFx << ESR_ELx_EC_SHIFT) |	\
 				 ESR_ELx_WFx_ISS_WFI)
 

arch/arm64/include/asm/hwcap.h

···
 #define KERNEL_HWCAP_SME_B16F32		__khwcap2_feature(SME_B16F32)
 #define KERNEL_HWCAP_SME_F32F32		__khwcap2_feature(SME_F32F32)
 #define KERNEL_HWCAP_SME_FA64		__khwcap2_feature(SME_FA64)
+#define KERNEL_HWCAP_WFXT		__khwcap2_feature(WFXT)
 
 /*
  * This yields a mask that user programs can use to figure out what

arch/arm64/include/asm/kvm_arm.h

···
  * FMO:		Override CPSR.F and enable signaling with VF
  * SWIO:	Turn set/way invalidates into set/way clean+invalidate
  * PTW:		Take a stage2 fault if a stage1 walk steps in device memory
+ * TID3:	Trap EL1 reads of group 3 ID registers
  */
 #define HCR_GUEST_FLAGS (HCR_TSC | HCR_TSW | HCR_TWE | HCR_TWI | HCR_VM | \
 			 HCR_BSU_IS | HCR_FB | HCR_TACR | \
 			 HCR_AMO | HCR_SWIO | HCR_TIDCP | HCR_RW | HCR_TLOR | \
-			 HCR_FMO | HCR_IMO | HCR_PTW )
+			 HCR_FMO | HCR_IMO | HCR_PTW | HCR_TID3 )
 #define HCR_VIRT_EXCP_MASK (HCR_VSE | HCR_VI | HCR_VF)
 #define HCR_HOST_NVHE_FLAGS (HCR_RW | HCR_API | HCR_APK | HCR_ATA)
 #define HCR_HOST_NVHE_PROTECTED_FLAGS (HCR_HOST_NVHE_FLAGS | HCR_TSC)

arch/arm64/include/asm/kvm_asm.h

···
 	unsigned long tcr_el2;
 	unsigned long tpidr_el2;
 	unsigned long stack_hyp_va;
+	unsigned long stack_pa;
 	phys_addr_t pgd_pa;
 	unsigned long hcr_el2;
 	unsigned long vttbr;

arch/arm64/include/asm/kvm_emulate.h

···
 
 	if (vcpu_el1_is_32bit(vcpu))
 		vcpu->arch.hcr_el2 &= ~HCR_RW;
-	else
-		/*
-		 * TID3: trap feature register accesses that we virtualise.
-		 * For now this is conditional, since no AArch32 feature regs
-		 * are currently virtualised.
-		 */
-		vcpu->arch.hcr_el2 |= HCR_TID3;
 
 	if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE) ||
 	    vcpu_el1_is_32bit(vcpu))

arch/arm64/include/asm/kvm_host.h

···
 #define KVM_REQ_RECORD_STEAL	KVM_ARCH_REQ(3)
 #define KVM_REQ_RELOAD_GICv4	KVM_ARCH_REQ(4)
 #define KVM_REQ_RELOAD_PMU	KVM_ARCH_REQ(5)
+#define KVM_REQ_SUSPEND		KVM_ARCH_REQ(6)
 
 #define KVM_DIRTY_LOG_MANUAL_CAPS   (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \
 				     KVM_DIRTY_LOG_INITIALLY_SET)
···
 struct kvm_arch_memory_slot {
 };
 
+/**
+ * struct kvm_smccc_features: Descriptor of the hypercall services exposed to the guests
+ *
+ * @std_bmap: Bitmap of standard secure service calls
+ * @std_hyp_bmap: Bitmap of standard hypervisor service calls
+ * @vendor_hyp_bmap: Bitmap of vendor specific hypervisor service calls
+ */
+struct kvm_smccc_features {
+	unsigned long std_bmap;
+	unsigned long std_hyp_bmap;
+	unsigned long vendor_hyp_bmap;
+};
+
 struct kvm_arch {
 	struct kvm_s2_mmu mmu;
 
 	/* VTCR_EL2 value for this VM */
 	u64    vtcr;
-
-	/* The maximum number of vCPUs depends on the used GIC model */
-	int max_vcpus;
 
 	/* Interrupt controller */
 	struct vgic_dist	vgic;
···
 	 */
 #define KVM_ARCH_FLAG_REG_WIDTH_CONFIGURED		3
 #define KVM_ARCH_FLAG_EL1_32BIT				4
+	/* PSCI SYSTEM_SUSPEND enabled for the guest */
+#define KVM_ARCH_FLAG_SYSTEM_SUSPEND_ENABLED		5
 
 	unsigned long flags;
 
···
 
 	u8 pfr0_csv2;
 	u8 pfr0_csv3;
+
+	/* Hypercall features firmware registers' descriptor */
+	struct kvm_smccc_features smccc_feat;
 };
 
 struct kvm_vcpu_fault_info {
···
 	struct kvm_vcpu *__hyp_running_vcpu;
 };
 
-struct kvm_pmu_events {
-	u32 events_host;
-	u32 events_guest;
-};
-
 struct kvm_host_data {
 	struct kvm_cpu_context host_ctxt;
-	struct kvm_pmu_events pmu_events;
 };
 
 struct kvm_host_psci_config {
···
 		u32	mdscr_el1;
 	} guest_debug_preserved;
 
-	/* vcpu power-off state */
-	bool power_off;
+	/* vcpu power state */
+	struct kvm_mp_state mp_state;
 
 	/* Don't run the guest (internal implementation need) */
 	bool pause;
···
 #define KVM_ARM64_FP_FOREIGN_FPSTATE	(1 << 14)
 #define KVM_ARM64_ON_UNSUPPORTED_CPU	(1 << 15) /* Physical CPU not in supported_cpus */
 #define KVM_ARM64_HOST_SME_ENABLED	(1 << 16) /* SME enabled for EL0 */
+#define KVM_ARM64_WFIT			(1 << 17) /* WFIT instruction trapped */
 
 #define KVM_GUESTDBG_VALID_MASK (KVM_GUESTDBG_ENABLE | \
 				 KVM_GUESTDBG_USE_SW_BP | \
···
 int kvm_handle_cp15_32(struct kvm_vcpu *vcpu);
 int kvm_handle_cp15_64(struct kvm_vcpu *vcpu);
 int kvm_handle_sys_reg(struct kvm_vcpu *vcpu);
+int kvm_handle_cp10_id(struct kvm_vcpu *vcpu);
 
 void kvm_reset_sys_regs(struct kvm_vcpu *vcpu);
 
-void kvm_sys_reg_table_init(void);
+int kvm_sys_reg_table_init(void);
 
 /* MMIO helpers */
 void kvm_mmio_write_buf(void *buf, unsigned int len, unsigned long data);
···
 #ifdef CONFIG_KVM
 void kvm_set_pmu_events(u32 set, struct perf_event_attr *attr);
 void kvm_clr_pmu_events(u32 clr);
-
-void kvm_vcpu_pmu_restore_guest(struct kvm_vcpu *vcpu);
-void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
 #else
 static inline void kvm_set_pmu_events(u32 set, struct perf_event_attr *attr) {}
 static inline void kvm_clr_pmu_events(u32 clr) {}
···
 #define kvm_has_mte(kvm)					\
 	(system_supports_mte() &&				\
 	 test_bit(KVM_ARCH_FLAG_MTE_ENABLED, &(kvm)->arch.flags))
-#define kvm_vcpu_has_pmu(vcpu)					\
-	(test_bit(KVM_ARM_VCPU_PMU_V3, (vcpu)->arch.features))
 
 int kvm_trng_call(struct kvm_vcpu *vcpu);
 #ifdef CONFIG_KVM
···
 #else
 static inline void kvm_hyp_reserve(void) { }
 #endif
+
+void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu);
+bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu);
 
 #endif /* __ARM64_KVM_HOST_H__ */

arch/arm64/include/asm/kvm_mmu.h

···
 int kvm_share_hyp(void *from, void *to);
 void kvm_unshare_hyp(void *from, void *to);
 int create_hyp_mappings(void *from, void *to, enum kvm_pgtable_prot prot);
+int __create_hyp_mappings(unsigned long start, unsigned long size,
+			  unsigned long phys, enum kvm_pgtable_prot prot);
+int hyp_alloc_private_va_range(size_t size, unsigned long *haddr);
 int create_hyp_io_mappings(phys_addr_t phys_addr, size_t size,
 			   void __iomem **kaddr,
 			   void __iomem **haddr);

arch/arm64/include/uapi/asm/hwcap.h

···
 #define HWCAP2_SME_B16F32	(1 << 28)
 #define HWCAP2_SME_F32F32	(1 << 29)
 #define HWCAP2_SME_FA64		(1 << 30)
+#define HWCAP2_WFXT		(1UL << 31)
 
 #endif /* _UAPI__ASM_HWCAP_H */

arch/arm64/include/uapi/asm/kvm.h

···
 #define KVM_ARM64_SVE_VLS_WORDS	\
 	((KVM_ARM64_SVE_VQ_MAX - KVM_ARM64_SVE_VQ_MIN) / 64 + 1)
 
+/* Bitmap feature firmware registers */
+#define KVM_REG_ARM_FW_FEAT_BMAP		(0x0016 << KVM_REG_ARM_COPROC_SHIFT)
+#define KVM_REG_ARM_FW_FEAT_BMAP_REG(r)		(KVM_REG_ARM64 | KVM_REG_SIZE_U64 | \
+						KVM_REG_ARM_FW_FEAT_BMAP |	\
+						((r) & 0xffff))
+
+#define KVM_REG_ARM_STD_BMAP			KVM_REG_ARM_FW_FEAT_BMAP_REG(0)
+
+enum {
+	KVM_REG_ARM_STD_BIT_TRNG_V1_0	= 0,
+#ifdef __KERNEL__
+	KVM_REG_ARM_STD_BMAP_BIT_COUNT,
+#endif
+};
+
+#define KVM_REG_ARM_STD_HYP_BMAP		KVM_REG_ARM_FW_FEAT_BMAP_REG(1)
+
+enum {
+	KVM_REG_ARM_STD_HYP_BIT_PV_TIME	= 0,
+#ifdef __KERNEL__
+	KVM_REG_ARM_STD_HYP_BMAP_BIT_COUNT,
+#endif
+};
+
+#define KVM_REG_ARM_VENDOR_HYP_BMAP		KVM_REG_ARM_FW_FEAT_BMAP_REG(2)
+
+enum {
+	KVM_REG_ARM_VENDOR_HYP_BIT_FUNC_FEAT	= 0,
+	KVM_REG_ARM_VENDOR_HYP_BIT_PTP		= 1,
+#ifdef __KERNEL__
+	KVM_REG_ARM_VENDOR_HYP_BMAP_BIT_COUNT,
+#endif
+};
+
 /* Device Control API: ARM VGIC */
 #define KVM_DEV_ARM_VGIC_GRP_ADDR	0
 #define KVM_DEV_ARM_VGIC_GRP_DIST_REGS	1

arch/arm64/kernel/cpufeature.c

···
 	ARM64_FTR_BITS(FTR_VISIBLE_IF_IS_ENABLED(CONFIG_ARM64_PTR_AUTH),
 		       FTR_STRICT, FTR_LOWER_SAFE, ID_AA64ISAR2_GPA3_SHIFT, 4, 0),
 	ARM64_FTR_BITS(FTR_VISIBLE, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64ISAR2_RPRES_SHIFT, 4, 0),
+	ARM64_FTR_BITS(FTR_VISIBLE, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64ISAR2_WFXT_SHIFT, 4, 0),
 	ARM64_FTR_END,
 };
 
···
 		.cpu_enable = fa64_kernel_enable,
 	},
 #endif /* CONFIG_ARM64_SME */
+	{
+		.desc = "WFx with timeout",
+		.capability = ARM64_HAS_WFXT,
+		.type = ARM64_CPUCAP_SYSTEM_FEATURE,
+		.sys_reg = SYS_ID_AA64ISAR2_EL1,
+		.sign = FTR_UNSIGNED,
+		.field_pos = ID_AA64ISAR2_WFXT_SHIFT,
+		.field_width = 4,
+		.matches = has_cpuid_feature,
+		.min_field_value = ID_AA64ISAR2_WFXT_SUPPORTED,
+	},
 	{},
 };
 
···
 	HWCAP_CAP(SYS_ID_AA64MMFR0_EL1, ID_AA64MMFR0_ECV_SHIFT, 4, FTR_UNSIGNED, 1, CAP_HWCAP, KERNEL_HWCAP_ECV),
 	HWCAP_CAP(SYS_ID_AA64MMFR1_EL1, ID_AA64MMFR1_AFP_SHIFT, 4, FTR_UNSIGNED, 1, CAP_HWCAP, KERNEL_HWCAP_AFP),
 	HWCAP_CAP(SYS_ID_AA64ISAR2_EL1, ID_AA64ISAR2_RPRES_SHIFT, 4, FTR_UNSIGNED, 1, CAP_HWCAP, KERNEL_HWCAP_RPRES),
+	HWCAP_CAP(SYS_ID_AA64ISAR2_EL1, ID_AA64ISAR2_WFXT_SHIFT, 4, FTR_UNSIGNED, ID_AA64ISAR2_WFXT_SUPPORTED, CAP_HWCAP, KERNEL_HWCAP_WFXT),
 #ifdef CONFIG_ARM64_SME
 	HWCAP_CAP(SYS_ID_AA64PFR1_EL1, ID_AA64PFR1_SME_SHIFT, 4, FTR_UNSIGNED, ID_AA64PFR1_SME, CAP_HWCAP, KERNEL_HWCAP_SME),
 	HWCAP_CAP(SYS_ID_AA64SMFR0_EL1, ID_AA64SMFR0_FA64_SHIFT, 1, FTR_UNSIGNED, ID_AA64SMFR0_FA64, CAP_HWCAP, KERNEL_HWCAP_SME_FA64),

arch/arm64/kernel/cpuinfo.c

···
 	[KERNEL_HWCAP_SME_B16F32]	= "smeb16f32",
 	[KERNEL_HWCAP_SME_F32F32]	= "smef32f32",
 	[KERNEL_HWCAP_SME_FA64]		= "smefa64",
+	[KERNEL_HWCAP_WFXT]		= "wfxt",
 };
 
 #ifdef CONFIG_COMPAT

arch/arm64/kvm/Makefile

···
 kvm-y += arm.o mmu.o mmio.o psci.o hypercalls.o pvtime.o \
 	 inject_fault.o va_layout.o handle_exit.o \
 	 guest.o debug.o reset.o sys_regs.o \
-	 vgic-sys-reg-v3.o fpsimd.o pmu.o pkvm.o \
+	 vgic-sys-reg-v3.o fpsimd.o pkvm.o \
 	 arch_timer.o trng.o vmid.o \
 	 vgic/vgic.o vgic/vgic-init.o \
 	 vgic/vgic-irqfd.o vgic/vgic-v2.o \
···
 	 vgic/vgic-mmio-v3.o vgic/vgic-kvm-device.o \
 	 vgic/vgic-its.o vgic/vgic-debug.o
 
-kvm-$(CONFIG_HW_PERF_EVENTS)  += pmu-emul.o
+kvm-$(CONFIG_HW_PERF_EVENTS)  += pmu-emul.o pmu.o
 
 always-y := hyp_constants.h hyp-constants.s
 

arch/arm64/kvm/arch_timer.c

···
 	return IRQ_HANDLED;
 }
 
-static u64 kvm_timer_compute_delta(struct arch_timer_context *timer_ctx)
+static u64 kvm_counter_compute_delta(struct arch_timer_context *timer_ctx,
+				     u64 val)
 {
-	u64 cval, now;
+	u64 now = kvm_phys_timer_read() - timer_get_offset(timer_ctx);
 
-	cval = timer_get_cval(timer_ctx);
-	now = kvm_phys_timer_read() - timer_get_offset(timer_ctx);
-
-	if (now < cval) {
+	if (now < val) {
 		u64 ns;
 
 		ns = cyclecounter_cyc2ns(timecounter->cc,
-					 cval - now,
+					 val - now,
 					 timecounter->mask,
 					 &timecounter->frac);
 		return ns;
···
 	return 0;
 }
 
+static u64 kvm_timer_compute_delta(struct arch_timer_context *timer_ctx)
+{
+	return kvm_counter_compute_delta(timer_ctx, timer_get_cval(timer_ctx));
+}
+
 static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx)
 {
 	WARN_ON(timer_ctx && timer_ctx->loaded);
 	return timer_ctx &&
 		((timer_get_ctl(timer_ctx) &
 		  (ARCH_TIMER_CTRL_IT_MASK | ARCH_TIMER_CTRL_ENABLE)) == ARCH_TIMER_CTRL_ENABLE);
+}
+
+static bool vcpu_has_wfit_active(struct kvm_vcpu *vcpu)
+{
+	return (cpus_have_final_cap(ARM64_HAS_WFXT) &&
+		(vcpu->arch.flags & KVM_ARM64_WFIT));
+}
+
+static u64 wfit_delay_ns(struct kvm_vcpu *vcpu)
+{
+	struct arch_timer_context *ctx = vcpu_vtimer(vcpu);
+	u64 val = vcpu_get_reg(vcpu, kvm_vcpu_sys_get_rt(vcpu));
+
+	return kvm_counter_compute_delta(ctx, val);
 }
 
 /*
···
 		if (kvm_timer_irq_can_fire(ctx))
 			min_delta = min(min_delta, kvm_timer_compute_delta(ctx));
 	}
+
+	if (vcpu_has_wfit_active(vcpu))
+		min_delta = min(min_delta, wfit_delay_ns(vcpu));
 
 	/* If none of timers can fire, then return 0 */
 	if (min_delta == ULLONG_MAX)
···
 	return cval <= now;
 }
 
-bool kvm_timer_is_pending(struct kvm_vcpu *vcpu)
+int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
-	struct timer_map map;
-
-	get_timer_map(vcpu, &map);
-
-	return kvm_timer_should_fire(map.direct_vtimer) ||
-	       kvm_timer_should_fire(map.direct_ptimer) ||
-	       kvm_timer_should_fire(map.emul_ptimer);
+	return vcpu_has_wfit_active(vcpu) && wfit_delay_ns(vcpu) == 0;
 }
 
 /*
···
 	 */
 	if (!kvm_timer_irq_can_fire(map.direct_vtimer) &&
 	    !kvm_timer_irq_can_fire(map.direct_ptimer) &&
-	    !kvm_timer_irq_can_fire(map.emul_ptimer))
+	    !kvm_timer_irq_can_fire(map.emul_ptimer) &&
+	    !vcpu_has_wfit_active(vcpu))
 		return;
 
 	/*

arch/arm64/kvm/arm.c

···
 		}
 		mutex_unlock(&kvm->lock);
 		break;
+	case KVM_CAP_ARM_SYSTEM_SUSPEND:
+		r = 0;
+		set_bit(KVM_ARCH_FLAG_SYSTEM_SUSPEND_ENABLED, &kvm->arch.flags);
+		break;
 	default:
 		r = -EINVAL;
 		break;
···
 	kvm_vgic_early_init(kvm);
 
 	/* The maximum number of VCPUs is limited by the host's GIC model */
-	kvm->arch.max_vcpus = kvm_arm_default_max_vcpus();
+	kvm->max_vcpus = kvm_arm_default_max_vcpus();
 
 	set_default_spectre(kvm);
+	kvm_arm_init_hypercalls(kvm);
 
 	return ret;
 out_free_stage2_pgd:
···
 	case KVM_CAP_SET_GUEST_DEBUG:
 	case KVM_CAP_VCPU_ATTRIBUTES:
 	case KVM_CAP_PTP_KVM:
+	case KVM_CAP_ARM_SYSTEM_SUSPEND:
 		r = 1;
 		break;
 	case KVM_CAP_SET_GUEST_DEBUG2:
···
 	case KVM_CAP_MAX_VCPUS:
 	case KVM_CAP_MAX_VCPU_ID:
 		if (kvm)
-			r = kvm->arch.max_vcpus;
+			r = kvm->max_vcpus;
 		else
 			r = kvm_arm_default_max_vcpus();
 		break;
···
 	if (irqchip_in_kernel(kvm) && vgic_initialized(kvm))
 		return -EBUSY;
 
-	if (id >= kvm->arch.max_vcpus)
+	if (id >= kvm->max_vcpus)
 		return -EINVAL;
 
 	return 0;
···
 	kvm_pmu_vcpu_destroy(vcpu);
 
 	kvm_arm_vcpu_destroy(vcpu);
-}
-
-int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
-{
-	return kvm_timer_is_pending(vcpu);
 }
 
 void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
···
 	vcpu->cpu = -1;
 }
 
-static void vcpu_power_off(struct kvm_vcpu *vcpu)
+void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu)
 {
-	vcpu->arch.power_off = true;
+	vcpu->arch.mp_state.mp_state = KVM_MP_STATE_STOPPED;
 	kvm_make_request(KVM_REQ_SLEEP, vcpu);
 	kvm_vcpu_kick(vcpu);
+}
+
+bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.mp_state.mp_state == KVM_MP_STATE_STOPPED;
+}
+
+static void kvm_arm_vcpu_suspend(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.mp_state.mp_state = KVM_MP_STATE_SUSPENDED;
+	kvm_make_request(KVM_REQ_SUSPEND, vcpu);
+	kvm_vcpu_kick(vcpu);
+}
+
+static bool kvm_arm_vcpu_suspended(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.mp_state.mp_state == KVM_MP_STATE_SUSPENDED;
 }
 
 int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state)
 {
-	if (vcpu->arch.power_off)
-		mp_state->mp_state = KVM_MP_STATE_STOPPED;
-	else
-		mp_state->mp_state = KVM_MP_STATE_RUNNABLE;
+	*mp_state = vcpu->arch.mp_state;
 
 	return 0;
 }
···
 
 	switch (mp_state->mp_state) {
 	case KVM_MP_STATE_RUNNABLE:
-		vcpu->arch.power_off = false;
+		vcpu->arch.mp_state = *mp_state;
 		break;
 	case KVM_MP_STATE_STOPPED:
-		vcpu_power_off(vcpu);
+		kvm_arm_vcpu_power_off(vcpu);
+		break;
+	case KVM_MP_STATE_SUSPENDED:
+		kvm_arm_vcpu_suspend(vcpu);
 		break;
 	default:
 		ret = -EINVAL;
···
 {
 	bool irq_lines = *vcpu_hcr(v) & (HCR_VI | HCR_VF);
 	return ((irq_lines || kvm_vgic_vcpu_pending_irq(v))
-		&& !v->arch.power_off && !v->arch.pause);
+		&& !kvm_arm_vcpu_stopped(v) && !v->arch.pause);
 }
 
 bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
···
 	}
 }
 
-static void vcpu_req_sleep(struct kvm_vcpu *vcpu)
+static void kvm_vcpu_sleep(struct kvm_vcpu *vcpu)
 {
 	struct rcuwait *wait = kvm_arch_vcpu_get_wait(vcpu);
 
 	rcuwait_wait_event(wait,
-			   (!vcpu->arch.power_off) &&(!vcpu->arch.pause),
+			   (!kvm_arm_vcpu_stopped(vcpu)) && (!vcpu->arch.pause),
 			   TASK_INTERRUPTIBLE);
 
-	if (vcpu->arch.power_off || vcpu->arch.pause) {
+	if (kvm_arm_vcpu_stopped(vcpu) || vcpu->arch.pause) {
 		/* Awaken to handle a signal, request we sleep again later. */
 		kvm_make_request(KVM_REQ_SLEEP, vcpu);
 	}
···
 	preempt_enable();
 
 	kvm_vcpu_halt(vcpu);
+	vcpu->arch.flags &= ~KVM_ARM64_WFIT;
 	kvm_clear_request(KVM_REQ_UNHALT, vcpu);
 
 	preempt_disable();
···
 	preempt_enable();
 }
 
-static void check_vcpu_requests(struct kvm_vcpu *vcpu)
+static int kvm_vcpu_suspend(struct kvm_vcpu *vcpu)
+{
+	if (!kvm_arm_vcpu_suspended(vcpu))
+		return 1;
+
+	kvm_vcpu_wfi(vcpu);
+
+	/*
+	 * The suspend state is sticky; we do not leave it until userspace
+	 * explicitly marks the vCPU as runnable. Request that we suspend again
+	 * later.
+	 */
+	kvm_make_request(KVM_REQ_SUSPEND, vcpu);
+
+	/*
+	 * Check to make sure the vCPU is actually runnable. If so, exit to
+	 * userspace informing it of the wakeup condition.
+	 */
+	if (kvm_arch_vcpu_runnable(vcpu)) {
+		memset(&vcpu->run->system_event, 0, sizeof(vcpu->run->system_event));
+		vcpu->run->system_event.type = KVM_SYSTEM_EVENT_WAKEUP;
+		vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
+		return 0;
+	}
+
+	/*
+	 * Otherwise, we were unblocked to process a different event, such as a
+	 * pending signal. Return 1 and allow kvm_arch_vcpu_ioctl_run() to
+	 * process the event.
+	 */
+	return 1;
+}
+
+/**
+ * check_vcpu_requests - check and handle pending vCPU requests
+ * @vcpu:	the VCPU pointer
+ *
+ * Return: 1 if we should enter the guest
+ *	   0 if we should exit to userspace
+ *	   < 0 if we should exit to userspace, where the return value indicates
+ *	   an error
+ */
+static int check_vcpu_requests(struct kvm_vcpu *vcpu)
 {
 	if (kvm_request_pending(vcpu)) {
 		if (kvm_check_request(KVM_REQ_SLEEP, vcpu))
-			vcpu_req_sleep(vcpu);
+			kvm_vcpu_sleep(vcpu);
 
 		if (kvm_check_request(KVM_REQ_VCPU_RESET, vcpu))
 			kvm_reset_vcpu(vcpu);
···
 		if (kvm_check_request(KVM_REQ_RELOAD_PMU, vcpu))
 			kvm_pmu_handle_pmcr(vcpu,
 					    __vcpu_sys_reg(vcpu, PMCR_EL0));
+
+		if (kvm_check_request(KVM_REQ_SUSPEND, vcpu))
+			return kvm_vcpu_suspend(vcpu);
 	}
+
+	return 1;
 }
 
 static bool vcpu_mode_is_bad_32bit(struct kvm_vcpu *vcpu)
···
 		if (!ret)
 			ret = 1;
 
-		check_vcpu_requests(vcpu);
+		if (ret > 0)
+			ret = check_vcpu_requests(vcpu);
 
 		/*
 		 * Preparing the interrupts to be injected also
···
 		local_irq_disable();
 
 		kvm_vgic_flush_hwstate(vcpu);
+
+		kvm_pmu_update_vcpu_events(vcpu);
 
 		/*
 		 * Ensure we set mode to IN_GUEST_MODE after we disable
···
 	 * Handle the "start in power-off" case.
 	 */
 	if (test_bit(KVM_ARM_VCPU_POWER_OFF, vcpu->arch.features))
-		vcpu_power_off(vcpu);
+		kvm_arm_vcpu_power_off(vcpu);
 	else
-		vcpu->arch.power_off = false;
+		vcpu->arch.mp_state.mp_state = KVM_MP_STATE_RUNNABLE;
 
 	return 0;
 }
···
 	tcr |= (idmap_t0sz & GENMASK(TCR_TxSZ_WIDTH - 1, 0)) << TCR_T0SZ_OFFSET;
 	params->tcr_el2 = tcr;
 
-	params->stack_hyp_va = kern_hyp_va(per_cpu(kvm_arm_hyp_stack_page, cpu) + PAGE_SIZE);
 	params->pgd_pa = kvm_mmu_get_httbr();
 	if (is_protected_kvm_enabled())
 		params->hcr_el2 = HCR_HOST_NVHE_PROTECTED_FLAGS;
···
 
 	kvm_register_perf_callbacks(NULL);
 
-	kvm_sys_reg_table_init();
-
 out:
 	if (err || !is_protected_kvm_enabled())
 		on_each_cpu(_kvm_arch_hardware_disable, NULL, 1);
···
 	 * Map the Hyp stack pages
 	 */
 	for_each_possible_cpu(cpu) {
+		struct kvm_nvhe_init_params *params = per_cpu_ptr_nvhe_sym(kvm_init_params, cpu);
 		char *stack_page = (char *)per_cpu(kvm_arm_hyp_stack_page, cpu);
-		err = create_hyp_mappings(stack_page, stack_page + PAGE_SIZE,
-					  PAGE_HYP);
+		unsigned long hyp_addr;
 
+		/*
+		 * Allocate a contiguous HYP private VA range for the stack
+		 * and guard page. The allocation is also aligned based on
+		 * the order of its size.
+		 */
+		err = hyp_alloc_private_va_range(PAGE_SIZE * 2, &hyp_addr);
+		if (err) {
+			kvm_err("Cannot allocate hyp stack guard page\n");
+			goto out_err;
+		}
+
+		/*
+		 * Since the stack grows downwards, map the stack to the page
+		 * at the higher address and leave the lower guard page
+		 * unbacked.
+		 *
+		 * Any valid stack address now has the PAGE_SHIFT bit as 1
+		 * and addresses corresponding to the guard page have the
+		 * PAGE_SHIFT bit as 0 - this is used for overflow detection.
+		 */
+		err = __create_hyp_mappings(hyp_addr + PAGE_SIZE, PAGE_SIZE,
+					    __pa(stack_page), PAGE_HYP);
 		if (err) {
 			kvm_err("Cannot map hyp stack\n");
 			goto out_err;
 		}
+
+		/*
+		 * Save the stack PA in nvhe_init_params. This will be needed
+		 * to recreate the stack mapping in protected nVHE mode.
+		 * __hyp_pa() won't do the right thing there, since the stack
+		 * has been mapped in the flexible private VA space.
+		 */
+		params->stack_pa = __pa(stack_page);
+
+		params->stack_hyp_va = hyp_addr + (2 * PAGE_SIZE);
 	}
 
 	for_each_possible_cpu(cpu) {
···
 	if (kvm_get_mode() == KVM_MODE_NONE) {
 		kvm_info("KVM disabled from command line\n");
 		return -ENODEV;
+	}
+
+	err = kvm_sys_reg_table_init();
+	if (err) {
+		kvm_info("Error initializing system register tables");
+		return err;
 	}
 
 	in_hyp_mode = is_kernel_in_hyp_mode();

arch/arm64/kvm/guest.c

···
 #include <linux/string.h>
 #include <linux/vmalloc.h>
 #include <linux/fs.h>
-#include <kvm/arm_psci.h>
+#include <kvm/arm_hypercalls.h>
 #include <asm/cputype.h>
 #include <linux/uaccess.h>
 #include <asm/fpsimd.h>
···
 
 	switch (reg->id & KVM_REG_ARM_COPROC_MASK) {
 	case KVM_REG_ARM_CORE:	return get_core_reg(vcpu, reg);
-	case KVM_REG_ARM_FW:	return kvm_arm_get_fw_reg(vcpu, reg);
+	case KVM_REG_ARM_FW:
+	case KVM_REG_ARM_FW_FEAT_BMAP:
+		return kvm_arm_get_fw_reg(vcpu, reg);
 	case KVM_REG_ARM64_SVE:	return get_sve_reg(vcpu, reg);
 	}
 
···
 
 	switch (reg->id & KVM_REG_ARM_COPROC_MASK) {
 	case KVM_REG_ARM_CORE:	return set_core_reg(vcpu, reg);
-	case KVM_REG_ARM_FW:	return kvm_arm_set_fw_reg(vcpu, reg);
+	case KVM_REG_ARM_FW:
+	case KVM_REG_ARM_FW_FEAT_BMAP:
+		return kvm_arm_set_fw_reg(vcpu, reg);
 	case KVM_REG_ARM64_SVE:	return set_sve_reg(vcpu, reg);
 	}
 

arch/arm64/kvm/handle_exit.c

···
  *
  * @vcpu:	the vcpu pointer
  *
- * WFE: Yield the CPU and come back to this vcpu when the scheduler
+ * WFE[T]: Yield the CPU and come back to this vcpu when the scheduler
  * decides to.
  * WFI: Simply call kvm_vcpu_halt(), which will halt execution of
  * world-switches and schedule other host processes until there is an
  * incoming IRQ or FIQ to the VM.
+ * WFIT: Same as WFI, with a timed wakeup implemented as a background timer
+ *
+ * WF{I,E}T can immediately return if the deadline has already expired.
  */
 static int kvm_handle_wfx(struct kvm_vcpu *vcpu)
 {
-	if (kvm_vcpu_get_esr(vcpu) & ESR_ELx_WFx_ISS_WFE) {
+	u64 esr = kvm_vcpu_get_esr(vcpu);
+
+	if (esr & ESR_ELx_WFx_ISS_WFE) {
 		trace_kvm_wfx_arm64(*vcpu_pc(vcpu), true);
 		vcpu->stat.wfe_exit_stat++;
-		kvm_vcpu_on_spin(vcpu, vcpu_mode_priv(vcpu));
 	} else {
 		trace_kvm_wfx_arm64(*vcpu_pc(vcpu), false);
 		vcpu->stat.wfi_exit_stat++;
-		kvm_vcpu_wfi(vcpu);
 	}
 
+	if (esr & ESR_ELx_WFx_ISS_WFxT) {
+		if (esr & ESR_ELx_WFx_ISS_RV) {
+			u64 val, now;
+
+			now = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_TIMER_CNT);
+			val = vcpu_get_reg(vcpu, kvm_vcpu_sys_get_rt(vcpu));
+
+			if (now >= val)
+				goto out;
+		} else {
+			/* Treat WFxT as WFx if RN is invalid */
+			esr &= ~ESR_ELx_WFx_ISS_WFxT;
+		}
+	}
+
+	if (esr & ESR_ELx_WFx_ISS_WFE) {
+		kvm_vcpu_on_spin(vcpu, vcpu_mode_priv(vcpu));
+	} else {
+		if (esr & ESR_ELx_WFx_ISS_WFxT)
+			vcpu->arch.flags |= KVM_ARM64_WFIT;
+
+		kvm_vcpu_wfi(vcpu);
+	}
+out:
 	kvm_incr_pc(vcpu);
 
 	return 1;
···
 	[ESR_ELx_EC_CP15_64]	= kvm_handle_cp15_64,
 	[ESR_ELx_EC_CP14_MR]	= kvm_handle_cp14_32,
 	[ESR_ELx_EC_CP14_LS]	= kvm_handle_cp14_load_store,
+	[ESR_ELx_EC_CP10_ID]	= kvm_handle_cp10_id,
 	[ESR_ELx_EC_CP14_64]	= kvm_handle_cp14_64,
 	[ESR_ELx_EC_HVC32]	= handle_hvc,
 	[ESR_ELx_EC_SMC32]	= handle_smc,
···
 	u64 elr_in_kimg = __phys_to_kimg(elr_phys);
 	u64 hyp_offset = elr_in_kimg - kaslr_offset() - elr_virt;
 	u64 mode = spsr & PSR_MODE_MASK;
+	u64 panic_addr = elr_virt + hyp_offset;
 
-	/*
-	 * The nVHE hyp symbols are not included by kallsyms to avoid issues
-	 * with aliasing. That means that the symbols cannot be printed with the
-	 * "%pS" format specifier, so fall back to the vmlinux address if
-	 * there's no better option.
-	 */
 	if (mode != PSR_MODE_EL2t && mode != PSR_MODE_EL2h) {
 		kvm_err("Invalid host exception to nVHE hyp!\n");
 	} else if (ESR_ELx_EC(esr) == ESR_ELx_EC_BRK64 &&
···
 		if (file)
 			kvm_err("nVHE hyp BUG at: %s:%u!\n", file, line);
 		else
-			kvm_err("nVHE hyp BUG at: %016llx!\n", elr_virt + hyp_offset);
+			kvm_err("nVHE hyp BUG at: [<%016llx>] %pB!\n", panic_addr,
+					(void *)panic_addr);
 	} else {
-		kvm_err("nVHE hyp panic at: %016llx!\n", elr_virt + hyp_offset);
+		kvm_err("nVHE hyp panic at: [<%016llx>] %pB!\n", panic_addr,
+				(void *)panic_addr);
 	}
 
 	/*

arch/arm64/kvm/hyp/include/nvhe/mm.h

···
 int pkvm_cpu_set_vector(enum arm64_hyp_spectre_vector slot);
 int pkvm_create_mappings(void *from, void *to, enum kvm_pgtable_prot prot);
 int pkvm_create_mappings_locked(void *from, void *to, enum kvm_pgtable_prot prot);
-unsigned long __pkvm_create_private_mapping(phys_addr_t phys, size_t size,
-					    enum kvm_pgtable_prot prot);
+int __pkvm_create_private_mapping(phys_addr_t phys, size_t size,
+				  enum kvm_pgtable_prot prot,
+				  unsigned long *haddr);
+int pkvm_alloc_private_va_range(size_t size, unsigned long *haddr);
 
 static inline void hyp_vmemmap_range(phys_addr_t phys, unsigned long size,
 				     unsigned long *start, unsigned long *end)

arch/arm64/kvm/hyp/nvhe/host.S

···
 	mov	lr, #(PSR_F_BIT | PSR_I_BIT | PSR_A_BIT | PSR_D_BIT |\
 		      PSR_MODE_EL1h)
 	msr	spsr_el2, lr
-	ldr	lr, =nvhe_hyp_panic_handler
+	adr_l	lr, nvhe_hyp_panic_handler
 	hyp_kimg_va lr, x6
 	msr	elr_el2, lr
 
···
 	add	sp, sp, #16
 	/*
 	 * Compute the idmap address of __kvm_handle_stub_hvc and
-	 * jump there. Since we use kimage_voffset, do not use the
-	 * HYP VA for __kvm_handle_stub_hvc, but the kernel VA instead
-	 * (by loading it from the constant pool).
+	 * jump there.
 	 *
 	 * Preserve x0-x4, which may contain stub parameters.
 	 */
-	ldr	x5, =__kvm_handle_stub_hvc
+	adr_l	x5, __kvm_handle_stub_hvc
 	hyp_pa	x5, x6
 	br	x5
 SYM_FUNC_END(__host_hvc)
···
 
 .macro invalid_host_el2_vect
 	.align 7
+
+	/*
+	 * Test whether the SP has overflowed, without corrupting a GPR.
+	 * nVHE hypervisor stacks are aligned so that the PAGE_SHIFT bit
+	 * of SP should always be 1.
+	 */
+	add	sp, sp, x0			// sp' = sp + x0
+	sub	x0, sp, x0			// x0' = sp' - x0 = (sp + x0) - x0 = sp
+	tbz	x0, #PAGE_SHIFT, .L__hyp_sp_overflow\@
+	sub	x0, sp, x0			// x0'' = sp' - x0' = (sp + x0) - sp = x0
+	sub	sp, sp, x0			// sp'' = sp' - x0 = (sp + x0) - x0 = sp
+
 	/* If a guest is loaded, panic out of it. */
 	stp	x0, x1, [sp, #-16]!
 	get_loaded_vcpu x0, x1
···
 	 * been partially clobbered by __host_enter.
 	 */
 	b	hyp_panic
+
+.L__hyp_sp_overflow\@:
+	/*
+	 * Reset SP to the top of the stack, to allow handling the hyp_panic.
+	 * This corrupts the stack but is ok, since we won't be attempting
+	 * any unwinding here.
+	 */
+	ldr_this_cpu	x0, kvm_init_params + NVHE_INIT_STACK_HYP_VA, x1
+	mov	sp, x0
+
+	b	hyp_panic_bad_stack
+	ASM_BUG()
 .endm
 
 .macro invalid_host_el1_vect

arch/arm64/kvm/hyp/nvhe/hyp-main.c

···
 	DECLARE_REG(size_t, size, host_ctxt, 2);
 	DECLARE_REG(enum kvm_pgtable_prot, prot, host_ctxt, 3);
 
-	cpu_reg(host_ctxt, 1) = __pkvm_create_private_mapping(phys, size, prot);
+	/*
+	 * __pkvm_create_private_mapping() populates a pointer with the
+	 * hypervisor start address of the allocation.
+	 *
+	 * However, handle___pkvm_create_private_mapping() hypercall crosses the
+	 * EL1/EL2 boundary so the pointer would not be valid in this context.
+	 *
+	 * Instead pass the allocation address as the return value (or return
+	 * ERR_PTR() on failure).
+	 */
+	unsigned long haddr;
+	int err = __pkvm_create_private_mapping(phys, size, prot, &haddr);
+
+	if (err)
+		haddr = (unsigned long)ERR_PTR(err);
+
+	cpu_reg(host_ctxt, 1) = haddr;
 }
 
 static void handle___pkvm_prot_finalize(struct kvm_cpu_context *host_ctxt)

arch/arm64/kvm/hyp/nvhe/mm.c

···
 	return err;
 }
 
-unsigned long __pkvm_create_private_mapping(phys_addr_t phys, size_t size,
-					    enum kvm_pgtable_prot prot)
+/**
+ * pkvm_alloc_private_va_range - Allocates a private VA range.
+ * @size:	The size of the VA range to reserve.
+ * @haddr:	The hypervisor virtual start address of the allocation.
+ *
+ * The private virtual address (VA) range is allocated above __io_map_base
+ * and aligned based on the order of @size.
+ *
+ * Return: 0 on success or negative error code on failure.
+ */
+int pkvm_alloc_private_va_range(size_t size, unsigned long *haddr)
+{
+	unsigned long base, addr;
+	int ret = 0;
+
+	hyp_spin_lock(&pkvm_pgd_lock);
+
+	/* Align the allocation based on the order of its size */
+	addr = ALIGN(__io_map_base, PAGE_SIZE << get_order(size));
+
+	/* The allocated size is always a multiple of PAGE_SIZE */
+	base = addr + PAGE_ALIGN(size);
+
+	/* Are we overflowing on the vmemmap ? */
+	if (!addr || base > __hyp_vmemmap)
+		ret = -ENOMEM;
+	else {
+		__io_map_base = base;
+		*haddr = addr;
+	}
+
+	hyp_spin_unlock(&pkvm_pgd_lock);
+
+	return ret;
+}
+
+int __pkvm_create_private_mapping(phys_addr_t phys, size_t size,
+				  enum kvm_pgtable_prot prot,
+				  unsigned long *haddr)
 {
 	unsigned long addr;
 	int err;
 
-	hyp_spin_lock(&pkvm_pgd_lock);
-
 	size = PAGE_ALIGN(size + offset_in_page(phys));
-	addr = __io_map_base;
-	__io_map_base += size;
+	err = pkvm_alloc_private_va_range(size, &addr);
+	if (err)
+		return err;
 
-	/* Are we overflowing on the vmemmap ? */
-	if (__io_map_base > __hyp_vmemmap) {
-		__io_map_base -= size;
-		addr = (unsigned long)ERR_PTR(-ENOMEM);
-		goto out;
-	}
+	err = __pkvm_create_mappings(addr, size, phys, prot);
+	if (err)
+		return err;
 
-	err = kvm_pgtable_hyp_map(&pkvm_pgtable, addr, size, phys, prot);
-	if (err) {
-		addr = (unsigned long)ERR_PTR(err);
-		goto out;
-	}
-
-	addr = addr + offset_in_page(phys);
-out:
-	hyp_spin_unlock(&pkvm_pgd_lock);
-
-	return addr;
+	*haddr = addr + offset_in_page(phys);
+	return err;
 }
 
 int pkvm_create_mappings_locked(void *from, void *to, enum kvm_pgtable_prot prot)
···
 int hyp_map_vectors(void)
 {
 	phys_addr_t phys;
-	void *bp_base;
+	unsigned long bp_base;
+	int ret;
 
 	if (!kvm_system_needs_idmapped_vectors()) {
 		__hyp_bp_vect_base = __bp_harden_hyp_vecs;
···
 	}
 
 	phys = __hyp_pa(__bp_harden_hyp_vecs);
-	bp_base = (void *)__pkvm_create_private_mapping(phys,
-							__BP_HARDEN_HYP_VECS_SZ,
-							PAGE_HYP_EXEC);
-	if (IS_ERR_OR_NULL(bp_base))
-		return PTR_ERR(bp_base);
+	ret = __pkvm_create_private_mapping(phys, __BP_HARDEN_HYP_VECS_SZ,
+					    PAGE_HYP_EXEC, &bp_base);
+	if (ret)
+		return ret;
 
-	__hyp_bp_vect_base = bp_base;
+	__hyp_bp_vect_base = (void *)bp_base;
 
 	return 0;
 }

arch/arm64/kvm/hyp/nvhe/setup.c

···
 		return ret;
 
 	for (i = 0; i < hyp_nr_cpus; i++) {
+		struct kvm_nvhe_init_params *params = per_cpu_ptr(&kvm_init_params, i);
+		unsigned long hyp_addr;
+
 		start = (void *)kern_hyp_va(per_cpu_base[i]);
 		end = start + PAGE_ALIGN(hyp_percpu_size);
 		ret = pkvm_create_mappings(start, end, PAGE_HYP);
 		if (ret)
 			return ret;
 
-		end = (void *)per_cpu_ptr(&kvm_init_params, i)->stack_hyp_va;
-		start = end - PAGE_SIZE;
-		ret = pkvm_create_mappings(start, end, PAGE_HYP);
+		/*
+		 * Allocate a contiguous HYP private VA range for the stack
+		 * and guard page. The allocation is also aligned based on
+		 * the order of its size.
+		 */
+		ret = pkvm_alloc_private_va_range(PAGE_SIZE * 2, &hyp_addr);
 		if (ret)
 			return ret;
+
+		/*
+		 * Since the stack grows downwards, map the stack to the page
+		 * at the higher address and leave the lower guard page
+		 * unbacked.
+		 *
+		 * Any valid stack address now has the PAGE_SHIFT bit as 1
+		 * and addresses corresponding to the guard page have the
+		 * PAGE_SHIFT bit as 0 - this is used for overflow detection.
+		 */
+		hyp_spin_lock(&pkvm_pgd_lock);
+		ret = kvm_pgtable_hyp_map(&pkvm_pgtable, hyp_addr + PAGE_SIZE,
+					PAGE_SIZE, params->stack_pa, PAGE_HYP);
+		hyp_spin_unlock(&pkvm_pgd_lock);
+		if (ret)
+			return ret;
+
+		/* Update stack_hyp_va to end of the stack's private VA range */
+		params->stack_hyp_va = hyp_addr + (2 * PAGE_SIZE);
 	}
 
 	/*

arch/arm64/kvm/hyp/nvhe/switch.c

···
 	}
 }
 
-/**
+/*
  * Disable host events, enable guest events
  */
-static bool __pmu_switch_to_guest(struct kvm_cpu_context *host_ctxt)
+#ifdef CONFIG_HW_PERF_EVENTS
+static bool __pmu_switch_to_guest(struct kvm_vcpu *vcpu)
 {
-	struct kvm_host_data *host;
-	struct kvm_pmu_events *pmu;
-
-	host = container_of(host_ctxt, struct kvm_host_data, host_ctxt);
-	pmu = &host->pmu_events;
+	struct kvm_pmu_events *pmu = &vcpu->arch.pmu.events;
 
 	if (pmu->events_host)
 		write_sysreg(pmu->events_host, pmcntenclr_el0);
···
 	return (pmu->events_host || pmu->events_guest);
 }
 
-/**
+/*
  * Disable guest events, enable host events
  */
-static void __pmu_switch_to_host(struct kvm_cpu_context *host_ctxt)
+static void __pmu_switch_to_host(struct kvm_vcpu *vcpu)
 {
-	struct kvm_host_data *host;
-	struct kvm_pmu_events *pmu;
-
-	host = container_of(host_ctxt, struct kvm_host_data, host_ctxt);
-	pmu = &host->pmu_events;
+	struct kvm_pmu_events *pmu = &vcpu->arch.pmu.events;
 
 	if (pmu->events_guest)
 		write_sysreg(pmu->events_guest, pmcntenclr_el0);
···
 	if (pmu->events_host)
 		write_sysreg(pmu->events_host, pmcntenset_el0);
 }
+#else
+#define __pmu_switch_to_guest(v)	({ false; })
+#define __pmu_switch_to_host(v)		do {} while (0)
+#endif
 
-/**
+/*
  * Handler for protected VM MSR, MRS or System instruction execution in AArch64.
  *
  * Returns true if the hypervisor has handled the exit, and control should go
···
 	 */
 	return (kvm_hyp_handle_sysreg(vcpu, exit_code) ||
 		kvm_handle_pvm_sysreg(vcpu, exit_code));
-}
-
-/**
- * Handler for protected floating-point and Advanced SIMD accesses.
- *
- * Returns true if the hypervisor has handled the exit, and control should go
- * back to the guest, or false if it hasn't.
- */
-static bool kvm_handle_pvm_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code)
-{
-	/* Linux guests assume support for floating-point and Advanced SIMD. */
-	BUILD_BUG_ON(!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_FP),
-				PVM_ID_AA64PFR0_ALLOW));
-	BUILD_BUG_ON(!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_ASIMD),
-				PVM_ID_AA64PFR0_ALLOW));
-
-	return kvm_hyp_handle_fpsimd(vcpu, exit_code);
 }
 
 static const exit_handler_fn hyp_exit_handlers[] = {
···
 	[0 ... ESR_ELx_EC_MAX]		= NULL,
 	[ESR_ELx_EC_SYS64]		= kvm_handle_pvm_sys64,
 	[ESR_ELx_EC_SVE]		= kvm_handle_pvm_restricted,
-	[ESR_ELx_EC_FP_ASIMD]		= kvm_handle_pvm_fpsimd,
+	[ESR_ELx_EC_FP_ASIMD]		= kvm_hyp_handle_fpsimd,
 	[ESR_ELx_EC_IABT_LOW]		= kvm_hyp_handle_iabt_low,
 	[ESR_ELx_EC_DABT_LOW]		= kvm_hyp_handle_dabt_low,
 	[ESR_ELx_EC_PAC]		= kvm_hyp_handle_ptrauth,
···
 	host_ctxt->__hyp_running_vcpu = vcpu;
 	guest_ctxt = &vcpu->arch.ctxt;
 
-	pmu_switch_needed = __pmu_switch_to_guest(host_ctxt);
+	pmu_switch_needed = __pmu_switch_to_guest(vcpu);
 
 	__sysreg_save_state_nvhe(host_ctxt);
 	/*
···
 	__debug_restore_host_buffers_nvhe(vcpu);
 
 	if (pmu_switch_needed)
-		__pmu_switch_to_host(host_ctxt);
+		__pmu_switch_to_host(vcpu);
 
 	/* Returning to host will clear PSR.I, remask PMR if needed */
 	if (system_uses_irq_prio_masking())
···
 	return exit_code;
 }
 
-void __noreturn hyp_panic(void)
+asmlinkage void __noreturn hyp_panic(void)
 {
 	u64 spsr = read_sysreg_el2(SYS_SPSR);
 	u64 elr = read_sysreg_el2(SYS_ELR);
···
 
 	__hyp_do_panic(host_ctxt, spsr, elr, par);
 	unreachable();
+}
+
+asmlinkage void __noreturn hyp_panic_bad_stack(void)
+{
+	hyp_panic();
 }
 
 asmlinkage void kvm_unexpected_el2_exception(void)

arch/arm64/kvm/hyp/nvhe/sys_regs.c

···
 	u64 set_mask = 0;
 	u64 allow_mask = PVM_ID_AA64PFR0_ALLOW;
 
-	if (!vcpu_has_sve(vcpu))
-		allow_mask &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_SVE);
-
 	set_mask |= get_restricted_features_unsigned(id_aa64pfr0_el1_sys_val,
 		PVM_ID_AA64PFR0_RESTRICT_UNSIGNED);
 

arch/arm64/kvm/hypercalls.c

···
 #include <kvm/arm_hypercalls.h>
 #include <kvm/arm_psci.h>
 
+#define KVM_ARM_SMCCC_STD_FEATURES				\
+	GENMASK(KVM_REG_ARM_STD_BMAP_BIT_COUNT - 1, 0)
+#define KVM_ARM_SMCCC_STD_HYP_FEATURES				\
+	GENMASK(KVM_REG_ARM_STD_HYP_BMAP_BIT_COUNT - 1, 0)
+#define KVM_ARM_SMCCC_VENDOR_HYP_FEATURES			\
+	GENMASK(KVM_REG_ARM_VENDOR_HYP_BMAP_BIT_COUNT - 1, 0)
+
 static void kvm_ptp_get_time(struct kvm_vcpu *vcpu, u64 *val)
 {
 	struct system_time_snapshot systime_snapshot;
···
 	val[3] = lower_32_bits(cycles);
 }
 
+static bool kvm_hvc_call_default_allowed(u32 func_id)
+{
+	switch (func_id) {
+	/*
+	 * List of function-ids that are not gated with the bitmapped
+	 * feature firmware registers, and are to be allowed for
+	 * servicing the call by default.
+	 */
+	case ARM_SMCCC_VERSION_FUNC_ID:
+	case ARM_SMCCC_ARCH_FEATURES_FUNC_ID:
+		return true;
+	default:
+		/* PSCI 0.2 and up is in the 0:0x1f range */
+		if (ARM_SMCCC_OWNER_NUM(func_id) == ARM_SMCCC_OWNER_STANDARD &&
+		    ARM_SMCCC_FUNC_NUM(func_id) <= 0x1f)
+			return true;
+
+		/*
+		 * KVM's PSCI 0.1 doesn't comply with SMCCC, and has
+		 * its own function-id base and range
+		 */
+		if (func_id >= KVM_PSCI_FN(0) && func_id <= KVM_PSCI_FN(3))
+			return true;
+
+		return false;
+	}
+}
+
+static bool kvm_hvc_call_allowed(struct kvm_vcpu *vcpu, u32 func_id)
+{
+	struct kvm_smccc_features *smccc_feat = &vcpu->kvm->arch.smccc_feat;
+
+	switch (func_id) {
+	case ARM_SMCCC_TRNG_VERSION:
+	case ARM_SMCCC_TRNG_FEATURES:
+	case ARM_SMCCC_TRNG_GET_UUID:
+	case ARM_SMCCC_TRNG_RND32:
+	case ARM_SMCCC_TRNG_RND64:
+		return test_bit(KVM_REG_ARM_STD_BIT_TRNG_V1_0,
+				&smccc_feat->std_bmap);
+	case ARM_SMCCC_HV_PV_TIME_FEATURES:
+	case ARM_SMCCC_HV_PV_TIME_ST:
+		return test_bit(KVM_REG_ARM_STD_HYP_BIT_PV_TIME,
+				&smccc_feat->std_hyp_bmap);
+	case ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID:
+	case ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID:
+		return test_bit(KVM_REG_ARM_VENDOR_HYP_BIT_FUNC_FEAT,
+				&smccc_feat->vendor_hyp_bmap);
+	case ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID:
+		return test_bit(KVM_REG_ARM_VENDOR_HYP_BIT_PTP,
+				&smccc_feat->vendor_hyp_bmap);
+	default:
+		return kvm_hvc_call_default_allowed(func_id);
+	}
+}
+
 int kvm_hvc_call_handler(struct kvm_vcpu *vcpu)
 {
+	struct kvm_smccc_features *smccc_feat = &vcpu->kvm->arch.smccc_feat;
 	u32 func_id = smccc_get_function(vcpu);
 	u64 val[4] = {SMCCC_RET_NOT_SUPPORTED};
 	u32 feature;
 	gpa_t gpa;
+
+	if (!kvm_hvc_call_allowed(vcpu, func_id))
+		goto out;
 
 	switch (func_id) {
 	case ARM_SMCCC_VERSION_FUNC_ID:
···
 			}
 			break;
 		case ARM_SMCCC_HV_PV_TIME_FEATURES:
-			val[0] = SMCCC_RET_SUCCESS;
+			if (test_bit(KVM_REG_ARM_STD_HYP_BIT_PV_TIME,
+				     &smccc_feat->std_hyp_bmap))
+				val[0] = SMCCC_RET_SUCCESS;
 			break;
 		}
 		break;
···
 		val[3] = ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_3;
 		break;
 	case ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID:
-		val[0] = BIT(ARM_SMCCC_KVM_FUNC_FEATURES);
-		val[0] |= BIT(ARM_SMCCC_KVM_FUNC_PTP);
+		val[0] = smccc_feat->vendor_hyp_bmap;
 		break;
 	case ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID:
 		kvm_ptp_get_time(vcpu, val);
···
 		return kvm_psci_call(vcpu);
 	}
 
+out:
 	smccc_set_retval(vcpu, val[0], val[1], val[2], val[3]);
 	return 1;
+}
+
+static const u64 kvm_arm_fw_reg_ids[] = {
+	KVM_REG_ARM_PSCI_VERSION,
+	KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1,
+	KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2,
+	KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3,
+	KVM_REG_ARM_STD_BMAP,
+	KVM_REG_ARM_STD_HYP_BMAP,
+	KVM_REG_ARM_VENDOR_HYP_BMAP,
+};
+
+void kvm_arm_init_hypercalls(struct kvm *kvm)
+{
+	struct kvm_smccc_features *smccc_feat = &kvm->arch.smccc_feat;
+
+	smccc_feat->std_bmap = KVM_ARM_SMCCC_STD_FEATURES;
+	smccc_feat->std_hyp_bmap = KVM_ARM_SMCCC_STD_HYP_FEATURES;
+	smccc_feat->vendor_hyp_bmap = KVM_ARM_SMCCC_VENDOR_HYP_FEATURES;
+}
+
+int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu)
+{
+	return ARRAY_SIZE(kvm_arm_fw_reg_ids);
+}
+
+int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(kvm_arm_fw_reg_ids); i++) {
+		if (put_user(kvm_arm_fw_reg_ids[i], uindices++))
+			return -EFAULT;
+	}
+
+	return 0;
+}
+
+#define KVM_REG_FEATURE_LEVEL_MASK	GENMASK(3, 0)
+
+/*
+ * Convert the workaround level into an easy-to-compare number, where higher
+ * values mean better protection.
+ */
+static int get_kernel_wa_level(u64 regid)
+{
+	switch (regid) {
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+		switch (arm64_get_spectre_v2_state()) {
+		case SPECTRE_VULNERABLE:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
+		case SPECTRE_MITIGATED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL;
+		case SPECTRE_UNAFFECTED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED;
+		}
+		return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+		switch (arm64_get_spectre_v4_state()) {
+		case SPECTRE_MITIGATED:
+			/*
+			 * As for the hypercall discovery, we pretend we
+			 * don't have any FW mitigation if SSBS is there at
+			 * all times.
+			 */
+			if (cpus_have_final_cap(ARM64_SSBS))
+				return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
+			fallthrough;
+		case SPECTRE_UNAFFECTED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED;
+		case SPECTRE_VULNERABLE:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
+		}
+		break;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
+		switch (arm64_get_spectre_bhb_state()) {
+		case SPECTRE_VULNERABLE:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_AVAIL;
+		case SPECTRE_MITIGATED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_AVAIL;
+		case SPECTRE_UNAFFECTED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_REQUIRED;
+		}
+		return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_AVAIL;
+	}
+
+	return -EINVAL;
+}
+
+int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
+{
+	struct kvm_smccc_features *smccc_feat = &vcpu->kvm->arch.smccc_feat;
+	void __user *uaddr = (void __user *)(long)reg->addr;
+	u64 val;
+
+	switch (reg->id) {
+	case KVM_REG_ARM_PSCI_VERSION:
+		val = kvm_psci_version(vcpu);
+		break;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
+		val = get_kernel_wa_level(reg->id) & KVM_REG_FEATURE_LEVEL_MASK;
+		break;
+	case KVM_REG_ARM_STD_BMAP:
+		val = READ_ONCE(smccc_feat->std_bmap);
+		break;
+	case KVM_REG_ARM_STD_HYP_BMAP:
+		val = READ_ONCE(smccc_feat->std_hyp_bmap);
+		break;
+	case KVM_REG_ARM_VENDOR_HYP_BMAP:
+		val = READ_ONCE(smccc_feat->vendor_hyp_bmap);
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	if (copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_arm_set_fw_reg_bmap(struct kvm_vcpu *vcpu, u64 reg_id, u64 val)
+{
+	int ret = 0;
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_smccc_features *smccc_feat = &kvm->arch.smccc_feat;
+	unsigned long *fw_reg_bmap, fw_reg_features;
+
+	switch (reg_id) {
+	case KVM_REG_ARM_STD_BMAP:
+		fw_reg_bmap = &smccc_feat->std_bmap;
+		fw_reg_features = KVM_ARM_SMCCC_STD_FEATURES;
+		break;
+	case KVM_REG_ARM_STD_HYP_BMAP:
+		fw_reg_bmap = &smccc_feat->std_hyp_bmap;
+		fw_reg_features = KVM_ARM_SMCCC_STD_HYP_FEATURES;
+		break;
+	case KVM_REG_ARM_VENDOR_HYP_BMAP:
+		fw_reg_bmap = &smccc_feat->vendor_hyp_bmap;
+		fw_reg_features = KVM_ARM_SMCCC_VENDOR_HYP_FEATURES;
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	/* Check for unsupported bit */
+	if (val & ~fw_reg_features)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+
+	if (test_bit(KVM_ARCH_FLAG_HAS_RAN_ONCE, &kvm->arch.flags) &&
+	    val != *fw_reg_bmap) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	WRITE_ONCE(*fw_reg_bmap, val);
+out:
+	mutex_unlock(&kvm->lock);
+	return ret;
+}
+
+int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
+{
+	void __user *uaddr = (void __user *)(long)reg->addr;
+	u64 val;
+	int wa_level;
+
+	if (copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	switch (reg->id) {
+	case KVM_REG_ARM_PSCI_VERSION:
+	{
+		bool wants_02;
+
+		wants_02 = test_bit(KVM_ARM_VCPU_PSCI_0_2, vcpu->arch.features);
+
+		switch (val) {
+		case KVM_ARM_PSCI_0_1:
+			if (wants_02)
+				return -EINVAL;
+			vcpu->kvm->arch.psci_version = val;
+			return 0;
+		case KVM_ARM_PSCI_0_2:
+		case KVM_ARM_PSCI_1_0:
+		case KVM_ARM_PSCI_1_1:
+			if (!wants_02)
+				return -EINVAL;
+			vcpu->kvm->arch.psci_version = val;
+			return 0;
+		}
+		break;
+	}
+
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
+		if (val & ~KVM_REG_FEATURE_LEVEL_MASK)
+			return -EINVAL;
+
+		if (get_kernel_wa_level(reg->id) < val)
+			return -EINVAL;
+
+		return 0;
+
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+		if (val & ~(KVM_REG_FEATURE_LEVEL_MASK |
+			    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED))
+			return -EINVAL;
+
+		/* The enabled bit must not be set unless the level is AVAIL. */
+		if ((val & KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED) &&
+		    (val & KVM_REG_FEATURE_LEVEL_MASK) != KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL)
+			return -EINVAL;
+
+		/*
+		 * Map all the possible incoming states to the only two we
+		 * really want to deal with.
+		 */
+		switch (val & KVM_REG_FEATURE_LEVEL_MASK) {
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL:
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN:
+			wa_level = KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
+			break;
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL:
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED:
+			wa_level = KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		/*
+		 * We can deal with NOT_AVAIL on NOT_REQUIRED, but not the
+		 * other way around.
+		 */
+		if (get_kernel_wa_level(reg->id) < wa_level)
+			return -EINVAL;
+
+		return 0;
+	case KVM_REG_ARM_STD_BMAP:
+	case KVM_REG_ARM_STD_HYP_BMAP:
+	case KVM_REG_ARM_VENDOR_HYP_BMAP:
+		return kvm_arm_set_fw_reg_bmap(vcpu, reg->id, val);
+	default:
+		return -ENOENT;
+	}
+
+	return -EINVAL;
 }

arch/arm64/kvm/mmu.c

···
 	return true;
 }
 
-static int __create_hyp_mappings(unsigned long start, unsigned long size,
-				 unsigned long phys, enum kvm_pgtable_prot prot)
+int __create_hyp_mappings(unsigned long start, unsigned long size,
+			  unsigned long phys, enum kvm_pgtable_prot prot)
 {
 	int err;
 
···
 	return 0;
 }
 
-static int __create_hyp_private_mapping(phys_addr_t phys_addr, size_t size,
-					unsigned long *haddr,
-					enum kvm_pgtable_prot prot)
+
+/**
+ * hyp_alloc_private_va_range - Allocates a private VA range.
+ * @size:	The size of the VA range to reserve.
+ * @haddr:	The hypervisor virtual start address of the allocation.
+ *
+ * The private virtual address (VA) range is allocated below io_map_base
+ * and aligned based on the order of @size.
+ *
+ * Return: 0 on success or negative error code on failure.
+ */
+int hyp_alloc_private_va_range(size_t size, unsigned long *haddr)
 {
 	unsigned long base;
 	int ret = 0;
-
-	if (!kvm_host_owns_hyp_mappings()) {
-		base = kvm_call_hyp_nvhe(__pkvm_create_private_mapping,
-					 phys_addr, size, prot);
-		if (IS_ERR_OR_NULL((void *)base))
-			return PTR_ERR((void *)base);
-		*haddr = base;
-
-		return 0;
-	}
 
 	mutex_lock(&kvm_hyp_pgd_mutex);
 
···
 	 *
 	 * The allocated size is always a multiple of PAGE_SIZE.
 	 */
-	size = PAGE_ALIGN(size + offset_in_page(phys_addr));
-	base = io_map_base - size;
+	base = io_map_base - PAGE_ALIGN(size);
+
+	/* Align the allocation based on the order of its size */
+	base = ALIGN_DOWN(base, PAGE_SIZE << get_order(size));
 
 	/*
 	 * Verify that BIT(VA_BITS - 1) hasn't been flipped by
···
 	if ((base ^ io_map_base) & BIT(VA_BITS - 1))
 		ret = -ENOMEM;
 	else
-		io_map_base = base;
+		*haddr = io_map_base = base;
 
 	mutex_unlock(&kvm_hyp_pgd_mutex);
 
-	if (ret)
-		goto out;
+	return ret;
+}
 
-	ret = __create_hyp_mappings(base, size, phys_addr, prot);
-	if (ret)
-		goto out;
+static int __create_hyp_private_mapping(phys_addr_t phys_addr, size_t size,
+					unsigned long *haddr,
+					enum kvm_pgtable_prot prot)
+{
+	unsigned long addr;
+	int ret = 0;
 
-	*haddr = base + offset_in_page(phys_addr);
-out:
+	if (!kvm_host_owns_hyp_mappings()) {
+		addr = kvm_call_hyp_nvhe(__pkvm_create_private_mapping,
+					 phys_addr, size, prot);
+		if (IS_ERR_VALUE(addr))
+			return addr;
+		*haddr = addr;
+
+		return 0;
+	}
+
+	size = PAGE_ALIGN(size + offset_in_page(phys_addr));
+	ret = hyp_alloc_private_va_range(size, &addr);
+	if (ret)
+		return ret;
+
+	ret = __create_hyp_mappings(addr, size, phys_addr, prot);
+	if (ret)
+		return ret;
+
+	*haddr = addr + offset_in_page(phys_addr);
 	return ret;
 }
 

arch/arm64/kvm/pmu-emul.c

···
 {
 	struct arm_pmu_entry *entry;
 
-	if (pmu->pmuver == 0 || pmu->pmuver == ID_AA64DFR0_PMUVER_IMP_DEF ||
-	    is_protected_kvm_enabled())
+	if (pmu->pmuver == 0 || pmu->pmuver == ID_AA64DFR0_PMUVER_IMP_DEF)
 		return;
 
 	mutex_lock(&arm_pmus_lock);

arch/arm64/kvm/pmu.c

···
  */
 #include <linux/kvm_host.h>
 #include <linux/perf_event.h>
-#include <asm/kvm_hyp.h>
+
+static DEFINE_PER_CPU(struct kvm_pmu_events, kvm_pmu_events);
 
 /*
  * Given the perf event attributes and system type, determine
···
 	return (attr->exclude_host != attr->exclude_guest);
 }
 
+struct kvm_pmu_events *kvm_get_pmu_events(void)
+{
+	return this_cpu_ptr(&kvm_pmu_events);
+}
+
 /*
  * Add events to track that we may want to switch at guest entry/exit
  * time.
  */
 void kvm_set_pmu_events(u32 set, struct perf_event_attr *attr)
 {
-	struct kvm_host_data *ctx = this_cpu_ptr_hyp_sym(kvm_host_data);
+	struct kvm_pmu_events *pmu = kvm_get_pmu_events();
 
-	if (!kvm_arm_support_pmu_v3() || !ctx || !kvm_pmu_switch_needed(attr))
+	if (!kvm_arm_support_pmu_v3() || !pmu || !kvm_pmu_switch_needed(attr))
 		return;
 
 	if (!attr->exclude_host)
-		ctx->pmu_events.events_host |= set;
+		pmu->events_host |= set;
 	if (!attr->exclude_guest)
-		ctx->pmu_events.events_guest |= set;
+		pmu->events_guest |= set;
 }
 
 /*
···
  */
 void kvm_clr_pmu_events(u32 clr)
 {
-	struct kvm_host_data *ctx = this_cpu_ptr_hyp_sym(kvm_host_data);
+	struct kvm_pmu_events *pmu = kvm_get_pmu_events();
 
-	if (!kvm_arm_support_pmu_v3() || !ctx)
+	if (!kvm_arm_support_pmu_v3() || !pmu)
 		return;
 
-	ctx->pmu_events.events_host &= ~clr;
-	ctx->pmu_events.events_guest &= ~clr;
+	pmu->events_host &= ~clr;
+	pmu->events_guest &= ~clr;
 }
 
 #define PMEVTYPER_READ_CASE(idx)				\
···
  */
 void kvm_vcpu_pmu_restore_guest(struct kvm_vcpu *vcpu)
 {
-	struct kvm_host_data *host;
+	struct kvm_pmu_events *pmu;
 	u32 events_guest, events_host;
 
 	if (!kvm_arm_support_pmu_v3() || !has_vhe())
 		return;
 
 	preempt_disable();
-	host = this_cpu_ptr_hyp_sym(kvm_host_data);
-	events_guest = host->pmu_events.events_guest;
-	events_host = host->pmu_events.events_host;
+	pmu = kvm_get_pmu_events();
+	events_guest = pmu->events_guest;
+	events_host = pmu->events_host;
 
 	kvm_vcpu_pmu_enable_el0(events_guest);
 	kvm_vcpu_pmu_disable_el0(events_host);
···
  */
 void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu)
 {
-	struct kvm_host_data *host;
+	struct kvm_pmu_events *pmu;
 	u32 events_guest, events_host;
 
 	if (!kvm_arm_support_pmu_v3() || !has_vhe())
 		return;
 
-	host = this_cpu_ptr_hyp_sym(kvm_host_data);
-	events_guest = host->pmu_events.events_guest;
-	events_host = host->pmu_events.events_host;
+	pmu = kvm_get_pmu_events();
+	events_guest = pmu->events_guest;
+	events_host = pmu->events_host;
 
 	kvm_vcpu_pmu_enable_el0(events_host);
 	kvm_vcpu_pmu_disable_el0(events_guest);

arch/arm64/kvm/psci.c

···
 	return PSCI_RET_SUCCESS;
 }
 
-static void kvm_psci_vcpu_off(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.power_off = true;
-	kvm_make_request(KVM_REQ_SLEEP, vcpu);
-	kvm_vcpu_kick(vcpu);
-}
-
 static inline bool kvm_psci_valid_affinity(struct kvm_vcpu *vcpu,
 					   unsigned long affinity)
 {
···
 	 */
 	if (!vcpu)
 		return PSCI_RET_INVALID_PARAMS;
-	if (!vcpu->arch.power_off) {
+	if (!kvm_arm_vcpu_stopped(vcpu)) {
 		if (kvm_psci_version(source_vcpu) != KVM_ARM_PSCI_0_1)
 			return PSCI_RET_ALREADY_ON;
 		else
···
 	kvm_make_request(KVM_REQ_VCPU_RESET, vcpu);
 
 	/*
-	 * Make sure the reset request is observed if the change to
-	 * power_off is observed.
+	 * Make sure the reset request is observed if the RUNNABLE mp_state is
+	 * observed.
 	 */
 	smp_wmb();
 
-	vcpu->arch.power_off = false;
+	vcpu->arch.mp_state.mp_state = KVM_MP_STATE_RUNNABLE;
 	kvm_vcpu_wake_up(vcpu);
 
 	return PSCI_RET_SUCCESS;
···
 		mpidr = kvm_vcpu_get_mpidr_aff(tmp);
 		if ((mpidr & target_affinity_mask) == target_affinity) {
 			matching_cpus++;
-			if (!tmp->arch.power_off)
+			if (!kvm_arm_vcpu_stopped(tmp))
 				return PSCI_0_2_AFFINITY_LEVEL_ON;
 		}
 	}
···
 	 * re-initialized.
 	 */
 	kvm_for_each_vcpu(i, tmp, vcpu->kvm)
-		tmp->arch.power_off = true;
+		tmp->arch.mp_state.mp_state = KVM_MP_STATE_STOPPED;
 	kvm_make_all_cpus_request(vcpu->kvm, KVM_REQ_SLEEP);
 
 	memset(&vcpu->run->system_event, 0, sizeof(vcpu->run->system_event));
···
 {
 	kvm_prepare_system_event(vcpu, KVM_SYSTEM_EVENT_RESET,
 				 KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2);
+}
+
+static void kvm_psci_system_suspend(struct kvm_vcpu *vcpu)
+{
+	struct kvm_run *run = vcpu->run;
+
+	memset(&run->system_event, 0, sizeof(vcpu->run->system_event));
+	run->system_event.type = KVM_SYSTEM_EVENT_SUSPEND;
+	run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
 }
 
 static void kvm_psci_narrow_to_32bit(struct kvm_vcpu *vcpu)
···
 		val = kvm_psci_vcpu_suspend(vcpu);
 		break;
 	case PSCI_0_2_FN_CPU_OFF:
-		kvm_psci_vcpu_off(vcpu);
+		kvm_arm_vcpu_power_off(vcpu);
 		val = PSCI_RET_SUCCESS;
 		break;
 	case PSCI_0_2_FN_CPU_ON:
···
 
 static int kvm_psci_1_x_call(struct kvm_vcpu *vcpu, u32 minor)
 {
+	unsigned long val = PSCI_RET_NOT_SUPPORTED;
 	u32 psci_fn = smccc_get_function(vcpu);
+	struct kvm *kvm = vcpu->kvm;
 	u32 arg;
-	unsigned long val;
 	int ret = 1;
 
 	switch(psci_fn) {
···
 		val = kvm_psci_check_allowed_function(vcpu, arg);
 		if (val)
 			break;
+
+		val = PSCI_RET_NOT_SUPPORTED;
 
 		switch(arg) {
 		case PSCI_0_2_FN_PSCI_VERSION:
···
 		case ARM_SMCCC_VERSION_FUNC_ID:
 			val = 0;
 			break;
+		case PSCI_1_0_FN_SYSTEM_SUSPEND:
+		case PSCI_1_0_FN64_SYSTEM_SUSPEND:
+			if (test_bit(KVM_ARCH_FLAG_SYSTEM_SUSPEND_ENABLED, &kvm->arch.flags))
+				val = 0;
+			break;
 		case PSCI_1_1_FN_SYSTEM_RESET2:
 		case PSCI_1_1_FN64_SYSTEM_RESET2:
-			if (minor >= 1) {
+			if (minor >= 1)
 				val = 0;
-				break;
-			}
-			fallthrough;
-		default:
-			val = PSCI_RET_NOT_SUPPORTED;
 			break;
+		}
+		break;
+	case PSCI_1_0_FN_SYSTEM_SUSPEND:
+		kvm_psci_narrow_to_32bit(vcpu);
+		fallthrough;
+	case PSCI_1_0_FN64_SYSTEM_SUSPEND:
+		/*
+		 * Return directly to userspace without changing the vCPU's
+		 * registers. Userspace depends on reading the SMCCC parameters
+		 * to implement SYSTEM_SUSPEND.
+		 */
+		if (test_bit(KVM_ARCH_FLAG_SYSTEM_SUSPEND_ENABLED, &kvm->arch.flags)) {
+			kvm_psci_system_suspend(vcpu);
+			return 0;
 		}
 		break;
 	case PSCI_1_1_FN_SYSTEM_RESET2:
···
 			val = PSCI_RET_INVALID_PARAMS;
 			break;
 		}
-		fallthrough;
+		break;
 	default:
 		return kvm_psci_0_2_call(vcpu);
 	}
···
 
 	switch (psci_fn) {
 	case KVM_PSCI_FN_CPU_OFF:
-		kvm_psci_vcpu_off(vcpu);
+		kvm_arm_vcpu_power_off(vcpu);
 		val = PSCI_RET_SUCCESS;
 		break;
 	case KVM_PSCI_FN_CPU_ON:
···
 	default:
 		return -EINVAL;
 	}
-}
-
-int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu)
-{
-	return 4;		/* PSCI version and three workaround registers */
-}
-
-int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
-{
-	if (put_user(KVM_REG_ARM_PSCI_VERSION, uindices++))
-		return -EFAULT;
-
-	if (put_user(KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1, uindices++))
-		return -EFAULT;
-
-	if (put_user(KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2, uindices++))
-		return -EFAULT;
-
-	if (put_user(KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3, uindices++))
-		return -EFAULT;
-
-	return 0;
-}
-
-#define KVM_REG_FEATURE_LEVEL_WIDTH	4
-#define KVM_REG_FEATURE_LEVEL_MASK	(BIT(KVM_REG_FEATURE_LEVEL_WIDTH) - 1)
-
-/*
- * Convert the workaround level into an easy-to-compare number, where higher
- * values mean better protection.
- */
-static int get_kernel_wa_level(u64 regid)
-{
-	switch (regid) {
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
-		switch (arm64_get_spectre_v2_state()) {
-		case SPECTRE_VULNERABLE:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
-		case SPECTRE_MITIGATED:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL;
-		case SPECTRE_UNAFFECTED:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED;
-		}
-		return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
-		switch (arm64_get_spectre_v4_state()) {
-		case SPECTRE_MITIGATED:
-			/*
-			 * As for the hypercall discovery, we pretend we
-			 * don't have any FW mitigation if SSBS is there at
-			 * all times.
-			 */
-			if (cpus_have_final_cap(ARM64_SSBS))
-				return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
-			fallthrough;
-		case SPECTRE_UNAFFECTED:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED;
-		case SPECTRE_VULNERABLE:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
-		}
-		break;
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
-		switch (arm64_get_spectre_bhb_state()) {
-		case SPECTRE_VULNERABLE:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_AVAIL;
-		case SPECTRE_MITIGATED:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_AVAIL;
-		case SPECTRE_UNAFFECTED:
-			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_REQUIRED;
-		}
-		return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3_NOT_AVAIL;
-	}
-
-	return -EINVAL;
-}
-
-int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
-{
-	void __user *uaddr = (void __user *)(long)reg->addr;
-	u64 val;
-
-	switch (reg->id) {
-	case KVM_REG_ARM_PSCI_VERSION:
-		val = kvm_psci_version(vcpu);
-		break;
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
-		val = get_kernel_wa_level(reg->id) & KVM_REG_FEATURE_LEVEL_MASK;
-		break;
-	default:
-		return -ENOENT;
-	}
-
-	if (copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)))
-		return -EFAULT;
-
-	return 0;
-}
-
-int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
-{
-	void __user *uaddr = (void __user *)(long)reg->addr;
-	u64 val;
-	int wa_level;
-
-	if (copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id)))
-		return -EFAULT;
-
-	switch (reg->id) {
-	case KVM_REG_ARM_PSCI_VERSION:
-	{
-		bool wants_02;
-
-		wants_02 = test_bit(KVM_ARM_VCPU_PSCI_0_2, vcpu->arch.features);
-
-		switch (val) {
-		case KVM_ARM_PSCI_0_1:
-			if (wants_02)
-				return -EINVAL;
-			vcpu->kvm->arch.psci_version = val;
-			return 0;
-		case KVM_ARM_PSCI_0_2:
-		case KVM_ARM_PSCI_1_0:
-		case KVM_ARM_PSCI_1_1:
-			if (!wants_02)
-				return -EINVAL;
-			vcpu->kvm->arch.psci_version = val;
-			return 0;
-		}
-		break;
-	}
-
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3:
-		if (val & ~KVM_REG_FEATURE_LEVEL_MASK)
-			return -EINVAL;
-
-		if (get_kernel_wa_level(reg->id) < val)
-			return -EINVAL;
-
-		return 0;
-
-	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
-		if (val & ~(KVM_REG_FEATURE_LEVEL_MASK |
-			    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED))
-			return -EINVAL;
-
-		/* The enabled bit must not be set unless the level is AVAIL. */
-		if ((val & KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED) &&
-		    (val & KVM_REG_FEATURE_LEVEL_MASK) != KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL)
-			return -EINVAL;
-
-		/*
-		 * Map all the possible incoming states to the only two we
-		 * really want to deal with.
-		 */
-		switch (val & KVM_REG_FEATURE_LEVEL_MASK) {
-		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL:
-		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN:
-			wa_level = KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
-			break;
-		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL:
-		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED:
-			wa_level = KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED;
-			break;
-		default:
-			return -EINVAL;
-		}
-
-		/*
-		 * We can deal with NOT_AVAIL on NOT_REQUIRED, but not the
-		 * other way around.
-		 */
-		if (get_kernel_wa_level(reg->id) < wa_level)
-			return -EINVAL;
-
-		return 0;
-	default:
-		return -ENOENT;
-	}
-
-	return -EINVAL;
 }

arch/arm64/kvm/sys_regs.c

···
 		if (!vcpu_has_ptrauth(vcpu))
 			val &= ~(ARM64_FEATURE_MASK(ID_AA64ISAR2_APA3) |
 				 ARM64_FEATURE_MASK(ID_AA64ISAR2_GPA3));
+		if (!cpus_have_final_cap(ARM64_HAS_WFXT))
+			val &= ~ARM64_FEATURE_MASK(ID_AA64ISAR2_WFXT);
 		break;
 	case SYS_ID_AA64DFR0_EL1:
 		/* Limit debug to ARMv8.0 */
···
 	{ Op1( 0), CRm( 2), .access = trap_raz_wi },
 };
 
+#define CP15_PMU_SYS_REG(_map, _Op1, _CRn, _CRm, _Op2)			\
+	AA32(_map),							\
+	Op1(_Op1), CRn(_CRn), CRm(_CRm), Op2(_Op2),			\
+	.visibility = pmu_visibility
+
 /* Macro to expand the PMEVCNTRn register */
 #define PMU_PMEVCNTR(n)							\
-	/* PMEVCNTRn */							\
-	{ Op1(0), CRn(0b1110),						\
-	  CRm((0b1000 | (((n) >> 3) & 0x3))), Op2(((n) & 0x7)),		\
-	  access_pmu_evcntr }
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 0b1110,				\
+	  (0b1000 | (((n) >> 3) & 0x3)), ((n) & 0x7)),			\
+	  .access = access_pmu_evcntr }
 
 /* Macro to expand the PMEVTYPERn register */
 #define PMU_PMEVTYPER(n)						\
-	/* PMEVTYPERn */						\
-	{ Op1(0), CRn(0b1110),						\
-	  CRm((0b1100 | (((n) >> 3) & 0x3))), Op2(((n) & 0x7)),		\
-	  access_pmu_evtyper }
-
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 0b1110,				\
+	  (0b1100 | (((n) >> 3) & 0x3)), ((n) & 0x7)),			\
+	  .access = access_pmu_evtyper }
 /*
  * Trapped cp15 registers. TTBR0/TTBR1 get a double encoding,
  * depending on the way they are accessed (as a 32bit or a 64bit
···
 	{ Op1( 0), CRn( 7), CRm(14), Op2( 2), access_dcsw },
 
 	/* PMU */
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 0), access_pmcr },
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 1), access_pmcnten },
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 2), access_pmcnten },
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 3), access_pmovs },
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 4), access_pmswinc },
-	{ Op1( 0), CRn( 9), CRm(12), Op2( 5), access_pmselr },
-	{ AA32(LO), Op1( 0), CRn( 9), CRm(12), Op2( 6), access_pmceid },
-	{ AA32(LO), Op1( 0), CRn( 9), CRm(12), Op2( 7), access_pmceid },
-	{ Op1( 0), CRn( 9), CRm(13), Op2( 0), access_pmu_evcntr },
-	{ Op1( 0), CRn( 9), CRm(13), Op2( 1), access_pmu_evtyper },
-	{ Op1( 0), CRn( 9), CRm(13), Op2( 2), access_pmu_evcntr },
-	{ Op1( 0), CRn( 9), CRm(14), Op2( 0), access_pmuserenr },
-	{ Op1( 0), CRn( 9), CRm(14), Op2( 1), access_pminten },
-	{ Op1( 0), CRn( 9), CRm(14), Op2( 2), access_pminten },
-	{ Op1( 0), CRn( 9), CRm(14), Op2( 3), access_pmovs },
-	{ AA32(HI), Op1( 0), CRn( 9), CRm(14), Op2( 4), access_pmceid },
-	{ AA32(HI), Op1( 0), CRn( 9), CRm(14), Op2( 5), access_pmceid },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 0), .access = access_pmcr },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 1), .access = access_pmcnten },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 2), .access = access_pmcnten },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 3), .access = access_pmovs },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 4), .access = access_pmswinc },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 12, 5), .access = access_pmselr },
+	{ CP15_PMU_SYS_REG(LO,     0, 9, 12, 6), .access = access_pmceid },
+	{ CP15_PMU_SYS_REG(LO,     0, 9, 12, 7), .access = access_pmceid },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 13, 0), .access = access_pmu_evcntr },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 13, 1), .access = access_pmu_evtyper },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 13, 2), .access = access_pmu_evcntr },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 14, 0), .access = access_pmuserenr },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 14, 1), .access = access_pminten },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 14, 2), .access = access_pminten },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 14, 3), .access = access_pmovs },
+	{ CP15_PMU_SYS_REG(HI,     0, 9, 14, 4), .access = access_pmceid },
+	{ CP15_PMU_SYS_REG(HI,     0, 9, 14, 5), .access = access_pmceid },
 	/* PMMIR */
-	{ Op1( 0), CRn( 9), CRm(14), Op2( 6), trap_raz_wi },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 9, 14, 6), .access = trap_raz_wi },
 
 	/* PRRR/MAIR0 */
 	{ AA32(LO), Op1( 0), CRn(10), CRm( 2), Op2( 0), access_vm_reg, NULL, MAIR_EL1 },
···
 	PMU_PMEVTYPER(29),
 	PMU_PMEVTYPER(30),
 	/* PMCCFILTR */
-	{ Op1(0), CRn(14), CRm(15), Op2(7), access_pmu_evtyper },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 14, 15, 7), .access = access_pmu_evtyper },
 
 	{ Op1(1), CRn( 0), CRm( 0), Op2(0), access_ccsidr },
 	{ Op1(1), CRn( 0), CRm( 0), Op2(1), access_clidr },
···
 
 static const struct sys_reg_desc cp15_64_regs[] = {
 	{ Op1( 0), CRn( 0), CRm( 2), Op2( 0), access_vm_reg, NULL, TTBR0_EL1 },
-	{ Op1( 0), CRn( 0), CRm( 9), Op2( 0), access_pmu_evcntr },
+	{ CP15_PMU_SYS_REG(DIRECT, 0, 0, 9, 0), .access = access_pmu_evcntr },
 	{ Op1( 0), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_SGI1R */
 	{ Op1( 1), CRn( 0), CRm( 2), Op2( 0), access_vm_reg, NULL, TTBR1_EL1 },
 	{ Op1( 1), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_ASGI1R */
···
 	{ SYS_DESC(SYS_AARCH32_CNTP_CVAL),    access_arch_timer },
 };
 
-static int check_sysreg_table(const struct sys_reg_desc *table, unsigned int n,
-			      bool is_32)
+static bool check_sysreg_table(const struct sys_reg_desc *table, unsigned int n,
+			       bool is_32)
 {
 	unsigned int i;
 
 	for (i = 0; i < n; i++) {
 		if (!is_32 && table[i].reg && !table[i].reset) {
-			kvm_err("sys_reg table %p entry %d has lacks reset\n",
-				table, i);
-			return 1;
+			kvm_err("sys_reg table %pS entry %d lacks reset\n", &table[i], i);
+			return false;
 		}
 
 		if (i && cmp_sys_reg(&table[i-1], &table[i]) >= 0) {
-			kvm_err("sys_reg table %p out of order (%d)\n", table, i - 1);
-			return 1;
+			kvm_err("sys_reg table %pS entry %d out of order\n", &table[i - 1], i - 1);
+			return false;
 		}
 	}
 
-	return 0;
+	return true;
 }
 
 int kvm_handle_cp14_load_store(struct kvm_vcpu *vcpu)
···
  * @table: array of trap descriptors
  * @num: size of the trap descriptor array
  *
- * Return 0 if the access has been handled, and -1 if not.
+ * Return true if the access has been handled, false if not.
  */
-static int emulate_cp(struct kvm_vcpu *vcpu,
-		      struct sys_reg_params *params,
-		      const struct sys_reg_desc *table,
-		      size_t num)
+static bool emulate_cp(struct kvm_vcpu *vcpu,
+		       struct sys_reg_params *params,
+		       const struct sys_reg_desc *table,
+		       size_t num)
 {
 	const struct sys_reg_desc *r;
 
 	if (!table)
-		return -1;	/* Not handled */
+		return false;	/* Not handled */
 
 	r = find_reg(params, table, num);
 
 	if (r) {
 		perform_access(vcpu, params, r);
-		return 0;
+		return true;
 	}
 
 	/* Not handled */
-	return -1;
+	return false;
 }
 
 static void unhandled_cp_access(struct kvm_vcpu *vcpu,
···
 	 * potential register operation in the case of a read and return
 	 * with success.
 	 */
-	if (!emulate_cp(vcpu, &params, global, nr_global)) {
+	if (emulate_cp(vcpu, &params, global, nr_global)) {
 		/* Split up the value between registers for the read side */
 		if (!params.is_write) {
 			vcpu_set_reg(vcpu, Rt, lower_32_bits(params.regval));
···
 	return 1;
 }
 
+static bool emulate_sys_reg(struct kvm_vcpu *vcpu, struct sys_reg_params *params);
+
+/*
+ * The CP10 ID registers are architecturally mapped to AArch64 feature
+ * registers. Abuse that fact so we can rely on the AArch64 handler for accesses
+ * from AArch32.
+ */
+static bool kvm_esr_cp10_id_to_sys64(u64 esr, struct sys_reg_params *params)
+{
+	u8 reg_id = (esr >> 10) & 0xf;
+	bool valid;
+
+	params->is_write = ((esr & 1) == 0);
+	params->Op0 = 3;
+	params->Op1 = 0;
+	params->CRn = 0;
+	params->CRm = 3;
+
+	/* CP10 ID registers are read-only */
+	valid = !params->is_write;
+
+	switch (reg_id) {
+	/* MVFR0 */
+	case 0b0111:
+		params->Op2 = 0;
+		break;
+	/* MVFR1 */
+	case 0b0110:
+		params->Op2 = 1;
+		break;
+	/* MVFR2 */
+	case 0b0101:
+		params->Op2 = 2;
+		break;
+	default:
+		valid = false;
+	}
+
+	if (valid)
+		return true;
+
+	kvm_pr_unimpl("Unhandled cp10 register %s: %u\n",
+		      params->is_write ? "write" : "read", reg_id);
+	return false;
+}
+
+/**
+ * kvm_handle_cp10_id() - Handles a VMRS trap on guest access to a 'Media and
+ *			  VFP Register' from AArch32.
+ * @vcpu: The vCPU pointer
+ *
+ * MVFR{0-2} are architecturally mapped to the AArch64 MVFR{0-2}_EL1 registers.
+ * Work out the correct AArch64 system register encoding and reroute to the
+ * AArch64 system register emulation.
+ */
+int kvm_handle_cp10_id(struct kvm_vcpu *vcpu)
+{
+	int Rt = kvm_vcpu_sys_get_rt(vcpu);
+	u64 esr = kvm_vcpu_get_esr(vcpu);
+	struct sys_reg_params params;
+
+	/* UNDEF on any unhandled register access */
+	if (!kvm_esr_cp10_id_to_sys64(esr, &params)) {
+		kvm_inject_undefined(vcpu);
+		return 1;
+	}
+
+	if (emulate_sys_reg(vcpu, &params))
+		vcpu_set_reg(vcpu, Rt, params.regval);
+
+	return 1;
+}
+
+/**
+ * kvm_emulate_cp15_id_reg() - Handles an MRC trap on a guest CP15 access where
+ *			       CRn=0, which corresponds to the AArch32 feature
+ *			       registers.
+ * @vcpu: the vCPU pointer
+ * @params: the system register access parameters.
+ *
+ * Our cp15 system register tables do not enumerate the AArch32 feature
+ * registers. Conveniently, our AArch64 table does, and the AArch32 system
+ * register encoding can be trivially remapped into the AArch64 for the feature
+ * registers: Append op0=3, leaving op1, CRn, CRm, and op2 the same.
+ *
+ * According to DDI0487G.b G7.3.1, paragraph "Behavior of VMSAv8-32 32-bit
+ * System registers with (coproc=0b1111, CRn==c0)", read accesses from this
+ * range are either UNKNOWN or RES0. Rerouting remains architectural as we
+ * treat undefined registers in this range as RAZ.
+ */
+static int kvm_emulate_cp15_id_reg(struct kvm_vcpu *vcpu,
+				   struct sys_reg_params *params)
+{
+	int Rt = kvm_vcpu_sys_get_rt(vcpu);
+
+	/* Treat impossible writes to RO registers as UNDEFINED */
+	if (params->is_write) {
+		unhandled_cp_access(vcpu, params);
+		return 1;
+	}
+
+	params->Op0 = 3;
+
+	/*
+	 * All registers where CRm > 3 are known to be UNKNOWN/RAZ from AArch32.
+	 * Avoid conflicting with future expansion of AArch64 feature registers
+	 * and simply treat them as RAZ here.
+	 */
+	if (params->CRm > 3)
+		params->regval = 0;
+	else if (!emulate_sys_reg(vcpu, params))
+		return 1;
+
+	vcpu_set_reg(vcpu, Rt, params->regval);
+	return 1;
+}
+
 /**
  * kvm_handle_cp_32 -- handles a mrc/mcr trap on a guest CP14/CP15 access
  * @vcpu: The VCPU pointer
  * @run:  The kvm_run struct
  */
 static int kvm_handle_cp_32(struct kvm_vcpu *vcpu,
+			    struct sys_reg_params *params,
 			    const struct sys_reg_desc *global,
 			    size_t nr_global)
 {
-	struct sys_reg_params params;
-	u64 esr = kvm_vcpu_get_esr(vcpu);
 	int Rt  = kvm_vcpu_sys_get_rt(vcpu);
 
-	params.CRm = (esr >> 1) & 0xf;
-	params.regval = vcpu_get_reg(vcpu, Rt);
-	params.is_write = ((esr & 1) == 0);
-	params.CRn = (esr >> 10) & 0xf;
-	params.Op0 = 0;
-	params.Op1 = (esr >> 14) & 0x7;
-	params.Op2 = (esr >> 17) & 0x7;
+	params->regval = vcpu_get_reg(vcpu, Rt);
 
-	if (!emulate_cp(vcpu, &params, global, nr_global)) {
-		if (!params.is_write)
-			vcpu_set_reg(vcpu, Rt, params.regval);
+	if (emulate_cp(vcpu, params, global, nr_global)) {
+		if (!params->is_write)
+			vcpu_set_reg(vcpu, Rt, params->regval);
 		return 1;
 	}
 
-	unhandled_cp_access(vcpu, &params);
+	unhandled_cp_access(vcpu, params);
 	return 1;
 }
 
···
 
 int kvm_handle_cp15_32(struct kvm_vcpu *vcpu)
 {
-	return kvm_handle_cp_32(vcpu, cp15_regs, ARRAY_SIZE(cp15_regs));
+	struct sys_reg_params params;
+
+	params = esr_cp1x_32_to_params(kvm_vcpu_get_esr(vcpu));
+
+	/*
+	 * Certain AArch32 ID registers are handled by rerouting to the AArch64
+	 * system register table. Registers in the ID range where CRm=0 are
+	 * excluded from this scheme as they do not trivially map into AArch64
+	 * system register encodings.
+	 */
+	if (params.Op1 == 0 && params.CRn == 0 && params.CRm)
+		return kvm_emulate_cp15_id_reg(vcpu, &params);
+
+	return kvm_handle_cp_32(vcpu, &params, cp15_regs, ARRAY_SIZE(cp15_regs));
 }
 
 int kvm_handle_cp14_64(struct kvm_vcpu *vcpu)
···
 
 int kvm_handle_cp14_32(struct kvm_vcpu *vcpu)
 {
-	return kvm_handle_cp_32(vcpu, cp14_regs, ARRAY_SIZE(cp14_regs));
+	struct sys_reg_params params;
+
+	params = esr_cp1x_32_to_params(kvm_vcpu_get_esr(vcpu));
+
+	return kvm_handle_cp_32(vcpu, &params, cp14_regs, ARRAY_SIZE(cp14_regs));
 }
 
 static bool is_imp_def_sys_reg(struct sys_reg_params *params)
···
 	return params->Op0 == 3 && (params->CRn & 0b1011) == 0b1011;
 }
 
-static int emulate_sys_reg(struct kvm_vcpu *vcpu,
+/**
+ * emulate_sys_reg - Emulate a guest access to an AArch64 system register
+ * @vcpu: The VCPU pointer
+ * @params: Decoded system register parameters
+ *
+ * Return: true if the system register access was successful, false otherwise.
+ */
+static bool emulate_sys_reg(struct kvm_vcpu *vcpu,
 			   struct sys_reg_params *params)
 {
 	const struct sys_reg_desc *r;
···
 
 	if (likely(r)) {
 		perform_access(vcpu, params, r);
-	} else if (is_imp_def_sys_reg(params)) {
+		return true;
+	}
+
+	if (is_imp_def_sys_reg(params)) {
 		kvm_inject_undefined(vcpu);
 	} else {
 		print_sys_reg_msg(params,
···
 				  *vcpu_pc(vcpu), *vcpu_cpsr(vcpu));
 		kvm_inject_undefined(vcpu);
 	}
-	return 1;
+	return false;
 }
 
 /**
···
 	struct sys_reg_params params;
 	unsigned long esr = kvm_vcpu_get_esr(vcpu);
 	int Rt = kvm_vcpu_sys_get_rt(vcpu);
-	int ret;
 
 	trace_kvm_handle_sys_reg(esr);
 
 	params = esr_sys64_to_params(esr);
 	params.regval = vcpu_get_reg(vcpu, Rt);
 
-	ret = emulate_sys_reg(vcpu, &params);
+	if (!emulate_sys_reg(vcpu, &params))
+		return 1;
 
 	if (!params.is_write)
 		vcpu_set_reg(vcpu, Rt, params.regval);
-	return ret;
+	return 1;
 }
 
 /******************************************************************************
···
 	return write_demux_regids(uindices);
 }
 
-void kvm_sys_reg_table_init(void)
+int kvm_sys_reg_table_init(void)
 {
+	bool valid = true;
 	unsigned int i;
 	struct sys_reg_desc clidr;
 
 	/* Make sure tables are unique and in order. */
-	BUG_ON(check_sysreg_table(sys_reg_descs, ARRAY_SIZE(sys_reg_descs), false));
-	BUG_ON(check_sysreg_table(cp14_regs, ARRAY_SIZE(cp14_regs), true));
-	BUG_ON(check_sysreg_table(cp14_64_regs, ARRAY_SIZE(cp14_64_regs), true));
-	BUG_ON(check_sysreg_table(cp15_regs, ARRAY_SIZE(cp15_regs), true));
-	BUG_ON(check_sysreg_table(cp15_64_regs, ARRAY_SIZE(cp15_64_regs), true));
-	BUG_ON(check_sysreg_table(invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs), false));
+	valid &= check_sysreg_table(sys_reg_descs, ARRAY_SIZE(sys_reg_descs), false);
+	valid &= check_sysreg_table(cp14_regs, ARRAY_SIZE(cp14_regs), true);
+	valid &= check_sysreg_table(cp14_64_regs, ARRAY_SIZE(cp14_64_regs), true);
+	valid &= check_sysreg_table(cp15_regs, ARRAY_SIZE(cp15_regs), true);
+	valid &= check_sysreg_table(cp15_64_regs, ARRAY_SIZE(cp15_64_regs), true);
+	valid &= check_sysreg_table(invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs), false);
+
+	if (!valid)
+		return -EINVAL;
 
 	/* We abuse the reset function to overwrite the table itself. */
 	for (i = 0; i < ARRAY_SIZE(invariant_sys_regs); i++)
···
 			break;
 	/* Clear all higher bits. */
 	cache_levels &= (1 << (i*3))-1;
+
+	return 0;
 }

arch/arm64/kvm/sys_regs.h

···
 				  .Op2 = ((esr) >> 17) & 0x7,                  \
 				  .is_write = !((esr) & 1) })
 
+#define esr_cp1x_32_to_params(esr)						\
+	((struct sys_reg_params){ .Op1 = ((esr) >> 14) & 0x7,			\
+				  .CRn = ((esr) >> 10) & 0xf,			\
+				  .CRm = ((esr) >> 1) & 0xf,			\
+				  .Op2 = ((esr) >> 17) & 0x7,			\
+				  .is_write = !((esr) & 1) })
+
 struct sys_reg_desc {
 	/* Sysreg string for debug */
 	const char *name;
 
 	enum {
-		AA32_ZEROHIGH,
+		AA32_DIRECT,
 		AA32_LO,
 		AA32_HI,
 	} aarch32_map;

arch/arm64/kvm/vgic/vgic-init.c

···
 	ret = 0;
 
 	if (type == KVM_DEV_TYPE_ARM_VGIC_V2)
-		kvm->arch.max_vcpus = VGIC_V2_MAX_CPUS;
+		kvm->max_vcpus = VGIC_V2_MAX_CPUS;
 	else
-		kvm->arch.max_vcpus = VGIC_V3_MAX_CPUS;
+		kvm->max_vcpus = VGIC_V3_MAX_CPUS;
 
-	if (atomic_read(&kvm->online_vcpus) > kvm->arch.max_vcpus) {
+	if (atomic_read(&kvm->online_vcpus) > kvm->max_vcpus) {
 		ret = -E2BIG;
 		goto out_unlock;
 	}
···
 
 	vgic_debug_init(kvm);
 
-	dist->implementation_rev = 2;
+	/*
+	 * If userspace didn't set the GIC implementation revision,
+	 * default to the latest and greatest. You know want it.
+	 */
+	if (!dist->implementation_rev)
+		dist->implementation_rev = KVM_VGIC_IMP_REV_LATEST;
 	dist->initialized = true;
 
 out:

arch/arm64/kvm/vgic/vgic-its.c

···
 	if (!vcpu)
 		return E_ITS_INT_UNMAPPED_INTERRUPT;
 
-	if (!vcpu->arch.vgic_cpu.lpis_enabled)
+	if (!vgic_lpis_enabled(vcpu))
 		return -EBUSY;
 
 	vgic_its_cache_translation(kvm, its, devid, eventid, ite->irq);
···
 	return update_affinity(ite->irq, vcpu);
 }
 
+static bool __is_visible_gfn_locked(struct vgic_its *its, gpa_t gpa)
+{
+	gfn_t gfn = gpa >> PAGE_SHIFT;
+	int idx;
+	bool ret;
+
+	idx = srcu_read_lock(&its->dev->kvm->srcu);
+	ret = kvm_is_visible_gfn(its->dev->kvm, gfn);
+	srcu_read_unlock(&its->dev->kvm->srcu, idx);
+	return ret;
+}
+
 /*
  * Check whether an ID can be stored into the corresponding guest table.
  * For a direct table this is pretty easy, but gets a bit nasty for
···
 	u64 indirect_ptr, type = GITS_BASER_TYPE(baser);
 	phys_addr_t base = GITS_BASER_ADDR_48_to_52(baser);
 	int esz = GITS_BASER_ENTRY_SIZE(baser);
-	int index, idx;
-	gfn_t gfn;
-	bool ret;
+	int index;
 
 	switch (type) {
 	case GITS_BASER_TYPE_DEVICE:
···
 			return false;
 
 		addr = base + id * esz;
-		gfn = addr >> PAGE_SHIFT;
 
 		if (eaddr)
 			*eaddr = addr;
 
-		goto out;
+		return __is_visible_gfn_locked(its, addr);
 	}
 
 	/* calculate and check the index into the 1st level */
···
 	/* Find the address of the actual entry */
 	index = id % (SZ_64K / esz);
 	indirect_ptr += index * esz;
-	gfn = indirect_ptr >> PAGE_SHIFT;
 
 	if (eaddr)
 		*eaddr = indirect_ptr;
 
-out:
-	idx = srcu_read_lock(&its->dev->kvm->srcu);
-	ret = kvm_is_visible_gfn(its->dev->kvm, gfn);
-	srcu_read_unlock(&its->dev->kvm->srcu, idx);
-	return ret;
+	return __is_visible_gfn_locked(its, indirect_ptr);
 }
 
+/*
+ * Check whether an event ID can be stored in the corresponding Interrupt
+ * Translation Table, which starts at device->itt_addr.
+ */
+static bool vgic_its_check_event_id(struct vgic_its *its, struct its_device *device,
+		u32 event_id)
+{
+	const struct vgic_its_abi *abi = vgic_its_get_abi(its);
+	int ite_esz = abi->ite_esz;
+	gpa_t gpa;
+
+	/* max table size is: BIT_ULL(device->num_eventid_bits) * ite_esz */
+	if (event_id >= BIT_ULL(device->num_eventid_bits))
+		return false;
+
+	gpa = device->itt_addr + event_id * ite_esz;
+	return __is_visible_gfn_locked(its, gpa);
+}
+
+/*
+ * Add a new collection into the ITS collection table.
+ * Returns 0 on success, and a negative error value for generic errors.
+ */
 static int vgic_its_alloc_collection(struct vgic_its *its,
 				     struct its_collection **colp,
 				     u32 coll_id)
 {
 	struct its_collection *collection;
-
-	if (!vgic_its_check_id(its, its->baser_coll_table, coll_id, NULL))
-		return E_ITS_MAPC_COLLECTION_OOR;
 
 	collection = kzalloc(sizeof(*collection), GFP_KERNEL_ACCOUNT);
 	if (!collection)
···
 	if (!device)
 		return E_ITS_MAPTI_UNMAPPED_DEVICE;
 
-	if (event_id >= BIT_ULL(device->num_eventid_bits))
+	if (!vgic_its_check_event_id(its, device, event_id))
 		return E_ITS_MAPTI_ID_OOR;
 
 	if (its_cmd_get_command(its_cmd) == GITS_CMD_MAPTI)
···
 
 	collection = find_collection(its, coll_id);
 	if (!collection) {
-		int ret = vgic_its_alloc_collection(its, &collection, coll_id);
+		int ret;
+
+		if (!vgic_its_check_id(its, its->baser_coll_table, coll_id, NULL))
+			return E_ITS_MAPC_COLLECTION_OOR;
+
+		ret = vgic_its_alloc_collection(its, &collection, coll_id);
 		if (ret)
 			return ret;
 		new_coll = collection;
···
 		if (!collection) {
 			int ret;
 
+			if (!vgic_its_check_id(its, its->baser_coll_table,
+						coll_id, NULL))
+				return E_ITS_MAPC_COLLECTION_OOR;
+
 			ret = vgic_its_alloc_collection(its, &collection,
 							coll_id);
 			if (ret)
···
 	return 0;
 }
 
+int vgic_its_inv_lpi(struct kvm *kvm, struct vgic_irq *irq)
+{
+	return update_lpi_config(kvm, irq, NULL, true);
+}
+
 /*
  * The INV command syncs the configuration bits from the memory table.
  * Must be called with the its_lock mutex held.
···
 	if (!ite)
 		return E_ITS_INV_UNMAPPED_INTERRUPT;
 
-	return update_lpi_config(kvm, ite->irq, NULL, true);
+	return vgic_its_inv_lpi(kvm, ite->irq);
+}
+
+/**
+ * vgic_its_invall - invalidate all LPIs targetting a given vcpu
+ * @vcpu: the vcpu for which the RD is targetted by an invalidation
+ *
+ * Contrary to the INVALL command, this targets a RD instead of a
+ * collection, and we don't need to hold the its_lock, since no ITS is
+ * involved here.
+ */
+int vgic_its_invall(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+	int irq_count, i = 0;
+	u32 *intids;
+
+	irq_count = vgic_copy_lpi_list(kvm, vcpu, &intids);
+	if (irq_count < 0)
+		return irq_count;
+
+	for (i = 0; i < irq_count; i++) {
+		struct vgic_irq *irq = vgic_get_irq(kvm, NULL, intids[i]);
+		if (!irq)
+			continue;
+		update_lpi_config(kvm, irq, vcpu, false);
+		vgic_put_irq(kvm, irq);
+	}
+
+	kfree(intids);
+
+	if (vcpu->arch.vgic_cpu.vgic_v3.its_vpe.its_vm)
+		its_invall_vpe(&vcpu->arch.vgic_cpu.vgic_v3.its_vpe);
+
+	return 0;
 }
 
 /*
···
 	u32 coll_id = its_cmd_get_collection(its_cmd);
 	struct its_collection *collection;
 	struct kvm_vcpu *vcpu;
-	struct vgic_irq *irq;
-	u32 *intids;
-	int irq_count, i;
 
 	collection = find_collection(its, coll_id);
 	if (!its_is_collection_mapped(collection))
 		return E_ITS_INVALL_UNMAPPED_COLLECTION;
 
 	vcpu = kvm_get_vcpu(kvm, collection->target_addr);
-
-	irq_count = vgic_copy_lpi_list(kvm, vcpu, &intids);
-	if (irq_count < 0)
-		return irq_count;
-
-	for (i = 0; i < irq_count; i++) {
-		irq = vgic_get_irq(kvm, NULL, intids[i]);
-		if (!irq)
-			continue;
-		update_lpi_config(kvm, irq, vcpu, false);
-		vgic_put_irq(kvm, irq);
-	}
-
-	kfree(intids);
-
-	if (vcpu->arch.vgic_cpu.vgic_v3.its_vpe.its_vm)
-		its_invall_vpe(&vcpu->arch.vgic_cpu.vgic_v3.its_vpe);
+	vgic_its_invall(vcpu);
 
 	return 0;
 }
···
 	if (!collection)
 		return -EINVAL;
 
+	if (!vgic_its_check_event_id(its, dev, event_id))
+		return -EINVAL;
+
 	ite = vgic_its_alloc_ite(dev, collection, event_id);
 	if (IS_ERR(ite))
 		return PTR_ERR(ite);
···
 		vcpu = kvm_get_vcpu(kvm, collection->target_addr);
 
 	irq = vgic_add_lpi(kvm, lpi_id, vcpu);
-	if (IS_ERR(irq))
+	if (IS_ERR(irq)) {
+		its_free_ite(kvm, ite);
 		return PTR_ERR(irq);
+	}
 	ite->irq = irq;
 
 	return offset;
···
 				void *ptr, void *opaque)
 {
 	struct its_device *dev;
+	u64 baser = its->baser_device_table;
 	gpa_t itt_addr;
 	u8 num_eventid_bits;
 	u64 entry = *(u64 *)ptr;
···
 
 	/* dte entry is valid */
 	offset = (entry & KVM_ITS_DTE_NEXT_MASK) >> KVM_ITS_DTE_NEXT_SHIFT;
+
+	if (!vgic_its_check_id(its, baser, id, NULL))
+		return -EINVAL;
 
 	dev = vgic_its_alloc_device(its, id, itt_addr, num_eventid_bits);
 	if (IS_ERR(dev))
···
 	if (ret > 0)
 		ret = 0;
 
+	if (ret < 0)
+		vgic_its_free_device_list(its->dev->kvm, its);
+
 	return ret;
 }
 
···
 	return kvm_write_guest_lock(its->dev->kvm, gpa, &val, esz);
 }
 
+/*
+ * Restore a collection entry into the ITS collection table.
+ * Return +1 on success, 0 if the entry was invalid (which should be
+ * interpreted as end-of-table), and a negative error value for generic errors.
+ */
 static int vgic_its_restore_cte(struct vgic_its *its, gpa_t gpa, int esz)
 {
 	struct its_collection *collection;
···
 	collection = find_collection(its, coll_id);
 	if (collection)
 		return -EEXIST;
+
+	if (!vgic_its_check_id(its, its->baser_coll_table, coll_id, NULL))
+		return -EINVAL;
+
 	ret = vgic_its_alloc_collection(its, &collection, coll_id);
 	if (ret)
 		return ret;
···
 	if (ret > 0)
 		return 0;
 
+	if (ret < 0)
+		vgic_its_free_collection_list(its->dev->kvm, its);
+
 	return ret;
 }
 
···
 	if (ret)
 		return ret;
 
-	return vgic_its_restore_device_tables(its);
+	ret = vgic_its_restore_device_tables(its);
+	if (ret)
+		vgic_its_free_collection_list(its->dev->kvm, its);
+	return ret;
 }
 
 static int vgic_its_commit_v0(struct vgic_its *its)

arch/arm64/kvm/vgic/vgic-mmio-v2.c

···
 					   gpa_t addr, unsigned int len,
 					   unsigned long val)
 {
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	u32 reg;
+
 	switch (addr & 0x0c) {
 	case GIC_DIST_IIDR:
-		if (val != vgic_mmio_read_v2_misc(vcpu, addr, len))
+		reg = vgic_mmio_read_v2_misc(vcpu, addr, len);
+		if ((reg ^ val) & ~GICD_IIDR_REVISION_MASK)
 			return -EINVAL;
 
 		/*
···
 		 * migration from old kernels to new kernels with legacy
 		 * userspace.
 		 */
-		vcpu->kvm->arch.vgic.v2_groups_user_writable = true;
-		return 0;
+		reg = FIELD_GET(GICD_IIDR_REVISION_MASK, reg);
+		switch (reg) {
+		case KVM_VGIC_IMP_REV_2:
+		case KVM_VGIC_IMP_REV_3:
+			vcpu->kvm->arch.vgic.v2_groups_user_writable = true;
+			dist->implementation_rev = reg;
+			return 0;
+		default:
+			return -EINVAL;
+		}
 	}
 
 	vgic_mmio_write_v2_misc(vcpu, addr, len, val);

arch/arm64/kvm/vgic/vgic-mmio-v3.c

···
 					   unsigned long val)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	u32 reg;
 
 	switch (addr & 0x0c) {
 	case GICD_TYPER2:
-	case GICD_IIDR:
 		if (val != vgic_mmio_read_v3_misc(vcpu, addr, len))
 			return -EINVAL;
 		return 0;
+	case GICD_IIDR:
+		reg = vgic_mmio_read_v3_misc(vcpu, addr, len);
+		if ((reg ^ val) & ~GICD_IIDR_REVISION_MASK)
+			return -EINVAL;
+
+		reg = FIELD_GET(GICD_IIDR_REVISION_MASK, reg);
+		switch (reg) {
+		case KVM_VGIC_IMP_REV_2:
+		case KVM_VGIC_IMP_REV_3:
+			dist->implementation_rev = reg;
+			return 0;
+		default:
+			return -EINVAL;
+		}
 	case GICD_CTLR:
 		/* Not a GICv4.1? No HW SGIs */
 		if (!kvm_vgic_global_state.has_gicv4_1)
···
 	vgic_put_irq(vcpu->kvm, irq);
 }
 
+bool vgic_lpis_enabled(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	return atomic_read(&vgic_cpu->ctlr) == GICR_CTLR_ENABLE_LPIS;
+}
+
 static unsigned long vgic_mmio_read_v3r_ctlr(struct kvm_vcpu *vcpu,
 					     gpa_t addr, unsigned int len)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	unsigned long val;
 
-	return vgic_cpu->lpis_enabled ? GICR_CTLR_ENABLE_LPIS : 0;
+	val = atomic_read(&vgic_cpu->ctlr);
+	if (vgic_get_implementation_rev(vcpu) >= KVM_VGIC_IMP_REV_3)
+		val |= GICR_CTLR_IR | GICR_CTLR_CES;
+
+	return val;
 }
-
 
 static void vgic_mmio_write_v3r_ctlr(struct kvm_vcpu *vcpu,
 				     gpa_t addr, unsigned int len,
 				     unsigned long val)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
-	bool was_enabled = vgic_cpu->lpis_enabled;
+	u32 ctlr;
 
 	if (!vgic_has_its(vcpu->kvm))
 		return;
 
-	vgic_cpu->lpis_enabled = val & GICR_CTLR_ENABLE_LPIS;
+	if (!(val & GICR_CTLR_ENABLE_LPIS)) {
+		/*
+		 * Don't disable if RWP is set, as there already an
+		 * ongoing disable. Funky guest...
+		 */
+		ctlr = atomic_cmpxchg_acquire(&vgic_cpu->ctlr,
+					      GICR_CTLR_ENABLE_LPIS,
+					      GICR_CTLR_RWP);
+		if (ctlr != GICR_CTLR_ENABLE_LPIS)
+			return;
 
-	if (was_enabled && !vgic_cpu->lpis_enabled) {
 		vgic_flush_pending_lpis(vcpu);
 		vgic_its_invalidate_cache(vcpu->kvm);
-	}
+		atomic_set_release(&vgic_cpu->ctlr, 0);
+	} else {
+		ctlr = atomic_cmpxchg_acquire(&vgic_cpu->ctlr, 0,
+					      GICR_CTLR_ENABLE_LPIS);
+		if (ctlr != 0)
+			return;
 
-	if (!was_enabled && vgic_cpu->lpis_enabled)
 		vgic_enable_lpis(vcpu);
+	}
 }
 
 static bool vgic_mmio_vcpu_rdist_is_last(struct kvm_vcpu *vcpu)
···
 				     unsigned long val)
 {
 	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
-	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
 	u64 old_propbaser, propbaser;
 
 	/* Storing a value with LPIs already enabled is undefined */
-	if (vgic_cpu->lpis_enabled)
+	if (vgic_lpis_enabled(vcpu))
 		return;
 
 	do {
···
 	u64 old_pendbaser, pendbaser;
 
 	/* Storing a value with LPIs already enabled is undefined */
-	if (vgic_cpu->lpis_enabled)
+	if (vgic_lpis_enabled(vcpu))
 		return;
 
 	do {
···
 		pendbaser = vgic_sanitise_pendbaser(pendbaser);
 	} while (cmpxchg64(&vgic_cpu->pendbaser, old_pendbaser,
 			   pendbaser) != old_pendbaser);
+}
+
+static unsigned long vgic_mmio_read_sync(struct kvm_vcpu *vcpu,
+					 gpa_t addr, unsigned int len)
+{
+	return !!atomic_read(&vcpu->arch.vgic_cpu.syncr_busy);
+}
+
+static void vgic_set_rdist_busy(struct kvm_vcpu *vcpu, bool busy)
+{
+	if (busy) {
+		atomic_inc(&vcpu->arch.vgic_cpu.syncr_busy);
+		smp_mb__after_atomic();
+	} else {
+		smp_mb__before_atomic();
+		atomic_dec(&vcpu->arch.vgic_cpu.syncr_busy);
+	}
+}
+
+static void vgic_mmio_write_invlpi(struct kvm_vcpu *vcpu,
+				   gpa_t addr, unsigned int len,
+				   unsigned long val)
+{
+	struct vgic_irq *irq;
+
+	/*
+	 * If the guest wrote only to the upper 32bit part of the
+	 * register, drop the write on the floor, as it is only for
+	 * vPEs (which we don't support for obvious reasons).
+	 *
+	 * Also discard the access if LPIs are not enabled.
+	 */
+	if ((addr & 4) || !vgic_lpis_enabled(vcpu))
+		return;
+
+	vgic_set_rdist_busy(vcpu, true);
+
+	irq = vgic_get_irq(vcpu->kvm, NULL, lower_32_bits(val));
+	if (irq) {
+		vgic_its_inv_lpi(vcpu->kvm, irq);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	vgic_set_rdist_busy(vcpu, false);
+}
+
+static void vgic_mmio_write_invall(struct kvm_vcpu *vcpu,
+				   gpa_t addr, unsigned int len,
+				   unsigned long val)
+{
+	/* See vgic_mmio_write_invlpi() for the early return rationale */
+	if ((addr & 4) || !vgic_lpis_enabled(vcpu))
+		return;
+
+	vgic_set_rdist_busy(vcpu, true);
+	vgic_its_invall(vcpu);
+	vgic_set_rdist_busy(vcpu, false);
 }
 
 /*
···
 	REGISTER_DESC_WITH_LENGTH(GICR_PENDBASER,
 		vgic_mmio_read_pendbase, vgic_mmio_write_pendbase, 8,
 		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_INVLPIR,
+		vgic_mmio_read_raz, vgic_mmio_write_invlpi, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_INVALLR,
+		vgic_mmio_read_raz, vgic_mmio_write_invall, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_SYNCR,
+		vgic_mmio_read_sync, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
 	REGISTER_DESC_WITH_LENGTH(GICR_IDREGS,
 		vgic_mmio_read_v3_idregs, vgic_mmio_write_wi, 48,
 		VGIC_ACCESS_32bit),

arch/arm64/kvm/vgic/vgic-v3.c

···
 static const struct midr_range broken_seis[] = {
 	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_ICESTORM),
 	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_FIRESTORM),
+	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_ICESTORM_PRO),
+	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_FIRESTORM_PRO),
+	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_ICESTORM_MAX),
+	MIDR_ALL_VERSIONS(MIDR_APPLE_M1_FIRESTORM_MAX),
 	{},
 };
 

arch/arm64/kvm/vgic/vgic.h

···
 #define DEBUG_SPINLOCK_BUG_ON(p)
 #endif
 
+static inline u32 vgic_get_implementation_rev(struct kvm_vcpu *vcpu)
+{
+	return vcpu->kvm->arch.vgic.implementation_rev;
+}
+
 /* Requires the irq_lock to be held by the caller. */
 static inline bool irq_is_pending(struct vgic_irq *irq)
 {
···
 		(base < d->vgic_dist_base + KVM_VGIC_V3_DIST_SIZE);
 }
 
+bool vgic_lpis_enabled(struct kvm_vcpu *vcpu);
 int vgic_copy_lpi_list(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 **intid_ptr);
 int vgic_its_resolve_lpi(struct kvm *kvm, struct vgic_its *its,
 			 u32 devid, u32 eventid, struct vgic_irq **irq);
···
 void vgic_lpi_translation_cache_init(struct kvm *kvm);
 void vgic_lpi_translation_cache_destroy(struct kvm *kvm);
 void vgic_its_invalidate_cache(struct kvm *kvm);
+
+/* GICv4.1 MMIO interface */
+int vgic_its_inv_lpi(struct kvm *kvm, struct vgic_irq *irq);
+int vgic_its_invall(struct kvm_vcpu *vcpu);
 
 bool vgic_supports_direct_msis(struct kvm *kvm);
 int vgic_v4_init(struct kvm *kvm);

arch/arm64/lib/delay.c

···
 {
 	cycles_t start = get_cycles();
 
-	if (arch_timer_evtstrm_available()) {
+	if (cpus_have_const_cap(ARM64_HAS_WFXT)) {
+		u64 end = start + cycles;
+
+		/*
+		 * Start with WFIT. If an interrupt makes us resume
+		 * early, use a WFET loop to complete the delay.
+		 */
+		wfit(end);
+		while ((get_cycles() - start) < cycles)
+			wfet(end);
+	} else 	if (arch_timer_evtstrm_available()) {
 		const cycles_t timer_evt_period =
 			USECS_TO_CYCLES(ARCH_TIMER_EVT_STREAM_PERIOD_US);
 

arch/arm64/tools/cpucaps

···
 HAS_SYSREG_GIC_CPUIF
 HAS_TLB_RANGE
 HAS_VIRT_HOST_EXTN
+HAS_WFXT
 HW_DBM
 KVM_PROTECTED_MODE
 MISMATCHED_CACHE_TYPE

arch/riscv/include/asm/csr.h

···
 #define HGATP_MODE_SV32X4	_AC(1, UL)
 #define HGATP_MODE_SV39X4	_AC(8, UL)
 #define HGATP_MODE_SV48X4	_AC(9, UL)
+#define HGATP_MODE_SV57X4	_AC(10, UL)
 
 #define HGATP32_MODE_SHIFT	31
 #define HGATP32_VMID_SHIFT	22

arch/riscv/include/asm/kvm_host.h

···
 #include <linux/types.h>
 #include <linux/kvm.h>
 #include <linux/kvm_types.h>
+#include <linux/spinlock.h>
 #include <asm/csr.h>
 #include <asm/kvm_vcpu_fp.h>
 #include <asm/kvm_vcpu_timer.h>
 
-#define KVM_MAX_VCPUS			\
-	((HGATP_VMID_MASK >> HGATP_VMID_SHIFT) + 1)
+#define KVM_MAX_VCPUS			1024
 
 #define KVM_HALT_POLL_NS_DEFAULT	500000
 
···
 	KVM_ARCH_REQ_FLAGS(0, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
 #define KVM_REQ_VCPU_RESET		KVM_ARCH_REQ(1)
 #define KVM_REQ_UPDATE_HGATP		KVM_ARCH_REQ(2)
+#define KVM_REQ_FENCE_I			\
+	KVM_ARCH_REQ_FLAGS(3, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
+#define KVM_REQ_HFENCE_GVMA_VMID_ALL	KVM_REQ_TLB_FLUSH
+#define KVM_REQ_HFENCE_VVMA_ALL		\
+	KVM_ARCH_REQ_FLAGS(4, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
+#define KVM_REQ_HFENCE			\
+	KVM_ARCH_REQ_FLAGS(5, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
+
+enum kvm_riscv_hfence_type {
+	KVM_RISCV_HFENCE_UNKNOWN = 0,
+	KVM_RISCV_HFENCE_GVMA_VMID_GPA,
+	KVM_RISCV_HFENCE_VVMA_ASID_GVA,
+	KVM_RISCV_HFENCE_VVMA_ASID_ALL,
+	KVM_RISCV_HFENCE_VVMA_GVA,
+};
+
+struct kvm_riscv_hfence {
+	enum kvm_riscv_hfence_type type;
+	unsigned long asid;
+	unsigned long order;
+	gpa_t addr;
+	gpa_t size;
+};
+
+#define KVM_RISCV_VCPU_MAX_HFENCE	64
 
 struct kvm_vm_stat {
 	struct kvm_vm_stat_generic generic;
···
 };
 
 struct kvm_arch {
-	/* stage2 vmid */
+	/* G-stage vmid */
 	struct kvm_vmid vmid;
 
-	/* stage2 page table */
+	/* G-stage page table */
 	pgd_t *pgd;
 	phys_addr_t pgd_phys;
 
···
 	/* VCPU ran at least once */
 	bool ran_atleast_once;
 
+	/* Last Host CPU on which Guest VCPU exited */
+	int last_exit_cpu;
+
 	/* ISA feature bits (similar to MISA) */
 	unsigned long isa;
 
···
 	/* VCPU Timer */
 	struct kvm_vcpu_timer timer;
 
+	/* HFENCE request queue */
+	spinlock_t hfence_lock;
+	unsigned long hfence_head;
+	unsigned long hfence_tail;
+	struct kvm_riscv_hfence hfence_queue[KVM_RISCV_VCPU_MAX_HFENCE];
+
 	/* MMIO instruction details */
 	struct kvm_mmio_decode mmio_decode;
 
···
 
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 
-void __kvm_riscv_hfence_gvma_vmid_gpa(unsigned long gpa_divby_4,
-				      unsigned long vmid);
-void __kvm_riscv_hfence_gvma_vmid(unsigned long vmid);
-void __kvm_riscv_hfence_gvma_gpa(unsigned long gpa_divby_4);
-void __kvm_riscv_hfence_gvma_all(void);
+#define KVM_RISCV_GSTAGE_TLB_MIN_ORDER		12
 
-int kvm_riscv_stage2_map(struct kvm_vcpu *vcpu,
+void kvm_riscv_local_hfence_gvma_vmid_gpa(unsigned long vmid,
+					  gpa_t gpa, gpa_t gpsz,
+					  unsigned long order);
+void kvm_riscv_local_hfence_gvma_vmid_all(unsigned long vmid);
+void kvm_riscv_local_hfence_gvma_gpa(gpa_t gpa, gpa_t gpsz,
+				     unsigned long order);
+void kvm_riscv_local_hfence_gvma_all(void);
+void kvm_riscv_local_hfence_vvma_asid_gva(unsigned long vmid,
+					  unsigned long asid,
+					  unsigned long gva,
+					  unsigned long gvsz,
+					  unsigned long order);
+void kvm_riscv_local_hfence_vvma_asid_all(unsigned long vmid,
+					  unsigned long asid);
+void kvm_riscv_local_hfence_vvma_gva(unsigned long vmid,
+				     unsigned long gva, unsigned long gvsz,
+				     unsigned long order);
+void kvm_riscv_local_hfence_vvma_all(unsigned long vmid);
+
+void kvm_riscv_local_tlb_sanitize(struct kvm_vcpu *vcpu);
+
+void kvm_riscv_fence_i_process(struct kvm_vcpu *vcpu);
+void kvm_riscv_hfence_gvma_vmid_all_process(struct kvm_vcpu *vcpu);
+void kvm_riscv_hfence_vvma_all_process(struct kvm_vcpu *vcpu);
+void kvm_riscv_hfence_process(struct kvm_vcpu *vcpu);
+
+void kvm_riscv_fence_i(struct kvm *kvm,
+		       unsigned long hbase, unsigned long hmask);
+void kvm_riscv_hfence_gvma_vmid_gpa(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    gpa_t gpa, gpa_t gpsz,
+				    unsigned long order);
+void kvm_riscv_hfence_gvma_vmid_all(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask);
+void kvm_riscv_hfence_vvma_asid_gva(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    unsigned long gva, unsigned long gvsz,
+				    unsigned long order, unsigned long asid);
+void kvm_riscv_hfence_vvma_asid_all(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    unsigned long asid);
+void kvm_riscv_hfence_vvma_gva(struct kvm *kvm,
+			       unsigned long hbase, unsigned long hmask,
+			       unsigned long gva, unsigned long gvsz,
+			       unsigned long order);
+void kvm_riscv_hfence_vvma_all(struct kvm *kvm,
+			       unsigned long hbase, unsigned long hmask);
+
+int kvm_riscv_gstage_map(struct kvm_vcpu *vcpu,
 			 struct kvm_memory_slot *memslot,
 			 gpa_t gpa, unsigned long hva, bool is_write);
-int kvm_riscv_stage2_alloc_pgd(struct kvm *kvm);
-void kvm_riscv_stage2_free_pgd(struct kvm *kvm);
-void kvm_riscv_stage2_update_hgatp(struct kvm_vcpu *vcpu);
-void kvm_riscv_stage2_mode_detect(void);
-unsigned long kvm_riscv_stage2_mode(void);
-int kvm_riscv_stage2_gpa_bits(void);
+int kvm_riscv_gstage_alloc_pgd(struct kvm *kvm);
+void kvm_riscv_gstage_free_pgd(struct kvm *kvm);
+void kvm_riscv_gstage_update_hgatp(struct kvm_vcpu *vcpu);
+void kvm_riscv_gstage_mode_detect(void);
+unsigned long kvm_riscv_gstage_mode(void);
+int kvm_riscv_gstage_gpa_bits(void);
 
-void kvm_riscv_stage2_vmid_detect(void);
-unsigned long kvm_riscv_stage2_vmid_bits(void);
-int kvm_riscv_stage2_vmid_init(struct kvm *kvm);
-bool kvm_riscv_stage2_vmid_ver_changed(struct kvm_vmid *vmid);
-void kvm_riscv_stage2_vmid_update(struct kvm_vcpu *vcpu);
+void kvm_riscv_gstage_vmid_detect(void);
+unsigned long kvm_riscv_gstage_vmid_bits(void);
+int kvm_riscv_gstage_vmid_init(struct kvm *kvm);
+bool kvm_riscv_gstage_vmid_ver_changed(struct kvm_vmid *vmid);
+void kvm_riscv_gstage_vmid_update(struct kvm_vcpu *vcpu);
 
 void __kvm_riscv_unpriv_trap(void);
 

arch/riscv/include/uapi/asm/kvm.h

···
 	__u64 state;
 };
 
+/*
+ * ISA extension IDs specific to KVM. This is not the same as the host ISA
+ * extension IDs as that is internal to the host and should not be exposed
+ * to the guest. This should always be contiguous to keep the mapping simple
+ * in KVM implementation.
+ */
+enum KVM_RISCV_ISA_EXT_ID {
+	KVM_RISCV_ISA_EXT_A = 0,
+	KVM_RISCV_ISA_EXT_C,
+	KVM_RISCV_ISA_EXT_D,
+	KVM_RISCV_ISA_EXT_F,
+	KVM_RISCV_ISA_EXT_H,
+	KVM_RISCV_ISA_EXT_I,
+	KVM_RISCV_ISA_EXT_M,
+	KVM_RISCV_ISA_EXT_MAX,
+};
+
 /* Possible states for kvm_riscv_timer */
 #define KVM_RISCV_TIMER_STATE_OFF	0
 #define KVM_RISCV_TIMER_STATE_ON	1
···
 #define KVM_REG_RISCV_FP_D		(0x06 << KVM_REG_RISCV_TYPE_SHIFT)
 #define KVM_REG_RISCV_FP_D_REG(name)	\
 		(offsetof(struct __riscv_d_ext_state, name) / sizeof(__u64))
+
+/* ISA Extension registers are mapped as type 7 */
+#define KVM_REG_RISCV_ISA_EXT		(0x07 << KVM_REG_RISCV_TYPE_SHIFT)
 
 #endif
 

arch/riscv/kvm/main.c

···
 		return -ENODEV;
 	}
 
-	kvm_riscv_stage2_mode_detect();
+	kvm_riscv_gstage_mode_detect();
 
-	kvm_riscv_stage2_vmid_detect();
+	kvm_riscv_gstage_vmid_detect();
 
 	kvm_info("hypervisor extension available\n");
 
-	switch (kvm_riscv_stage2_mode()) {
+	switch (kvm_riscv_gstage_mode()) {
 	case HGATP_MODE_SV32X4:
 		str = "Sv32x4";
 		break;
···
 	case HGATP_MODE_SV48X4:
 		str = "Sv48x4";
 		break;
+	case HGATP_MODE_SV57X4:
+		str = "Sv57x4";
+		break;
 	default:
 		return -ENODEV;
 	}
 	kvm_info("using %s G-stage page table format\n", str);
 
-	kvm_info("VMID %ld bits available\n", kvm_riscv_stage2_vmid_bits());
+	kvm_info("VMID %ld bits available\n", kvm_riscv_gstage_vmid_bits());
 
 	return 0;
 }

arch/riscv/kvm/mmu.c

···
 #include <asm/csr.h>
 #include <asm/page.h>
 #include <asm/pgtable.h>
-#include <asm/sbi.h>
 
 #ifdef CONFIG_64BIT
-static unsigned long stage2_mode = (HGATP_MODE_SV39X4 << HGATP_MODE_SHIFT);
-static unsigned long stage2_pgd_levels = 3;
-#define stage2_index_bits	9
+static unsigned long gstage_mode = (HGATP_MODE_SV39X4 << HGATP_MODE_SHIFT);
+static unsigned long gstage_pgd_levels = 3;
+#define gstage_index_bits	9
 #else
-static unsigned long stage2_mode = (HGATP_MODE_SV32X4 << HGATP_MODE_SHIFT);
-static unsigned long stage2_pgd_levels = 2;
-#define stage2_index_bits	10
+static unsigned long gstage_mode = (HGATP_MODE_SV32X4 << HGATP_MODE_SHIFT);
+static unsigned long gstage_pgd_levels = 2;
+#define gstage_index_bits	10
 #endif
 
-#define stage2_pgd_xbits	2
-#define stage2_pgd_size	(1UL << (HGATP_PAGE_SHIFT + stage2_pgd_xbits))
-#define stage2_gpa_bits	(HGATP_PAGE_SHIFT + \
-			 (stage2_pgd_levels * stage2_index_bits) + \
-			 stage2_pgd_xbits)
-#define stage2_gpa_size	((gpa_t)(1ULL << stage2_gpa_bits))
+#define gstage_pgd_xbits	2
+#define gstage_pgd_size	(1UL << (HGATP_PAGE_SHIFT + gstage_pgd_xbits))
+#define gstage_gpa_bits	(HGATP_PAGE_SHIFT + \
+			 (gstage_pgd_levels * gstage_index_bits) + \
+			 gstage_pgd_xbits)
+#define gstage_gpa_size	((gpa_t)(1ULL << gstage_gpa_bits))
 
-#define stage2_pte_leaf(__ptep)	\
+#define gstage_pte_leaf(__ptep)	\
 	(pte_val(*(__ptep)) & (_PAGE_READ | _PAGE_WRITE | _PAGE_EXEC))
 
-static inline unsigned long stage2_pte_index(gpa_t addr, u32 level)
+static inline unsigned long gstage_pte_index(gpa_t addr, u32 level)
 {
 	unsigned long mask;
-	unsigned long shift = HGATP_PAGE_SHIFT + (stage2_index_bits * level);
+	unsigned long shift = HGATP_PAGE_SHIFT + (gstage_index_bits * level);
 
-	if (level == (stage2_pgd_levels - 1))
-		mask = (PTRS_PER_PTE * (1UL << stage2_pgd_xbits)) - 1;
+	if (level == (gstage_pgd_levels - 1))
+		mask = (PTRS_PER_PTE * (1UL << gstage_pgd_xbits)) - 1;
 	else
 		mask = PTRS_PER_PTE - 1;
 
 	return (addr >> shift) & mask;
 }
 
-static inline unsigned long stage2_pte_page_vaddr(pte_t pte)
+static inline unsigned long gstage_pte_page_vaddr(pte_t pte)
 {
 	return (unsigned long)pfn_to_virt(pte_val(pte) >> _PAGE_PFN_SHIFT);
 }
 
-static int stage2_page_size_to_level(unsigned long page_size, u32 *out_level)
+static int gstage_page_size_to_level(unsigned long page_size, u32 *out_level)
 {
 	u32 i;
 	unsigned long psz = 1UL << 12;
 
-	for (i = 0; i < stage2_pgd_levels; i++) {
-		if (page_size == (psz << (i * stage2_index_bits))) {
+	for (i = 0; i < gstage_pgd_levels; i++) {
+		if (page_size == (psz << (i * gstage_index_bits))) {
 			*out_level = i;
 			return 0;
 		}
···
 	return -EINVAL;
 }
 
-static int stage2_level_to_page_size(u32 level, unsigned long *out_pgsize)
+static int gstage_level_to_page_order(u32 level, unsigned long *out_pgorder)
 {
-	if (stage2_pgd_levels < level)
+	if (gstage_pgd_levels < level)
 		return -EINVAL;
 
-	*out_pgsize = 1UL << (12 + (level * stage2_index_bits));
-
+	*out_pgorder = 12 + (level * gstage_index_bits);
 	return 0;
 }
 
-static bool stage2_get_leaf_entry(struct kvm *kvm, gpa_t addr,
+static int gstage_level_to_page_size(u32 level, unsigned long *out_pgsize)
+{
+	int rc;
+	unsigned long page_order = PAGE_SHIFT;
+
+	rc = gstage_level_to_page_order(level, &page_order);
+	if (rc)
+		return rc;
+
+	*out_pgsize = BIT(page_order);
+	return 0;
+}
+
+static bool gstage_get_leaf_entry(struct kvm *kvm, gpa_t addr,
 				  pte_t **ptepp, u32 *ptep_level)
 {
 	pte_t *ptep;
-	u32 current_level = stage2_pgd_levels - 1;
+	u32 current_level = gstage_pgd_levels - 1;
 
 	*ptep_level = current_level;
 	ptep = (pte_t *)kvm->arch.pgd;
-	ptep = &ptep[stage2_pte_index(addr, current_level)];
+	ptep = &ptep[gstage_pte_index(addr, current_level)];
 	while (ptep && pte_val(*ptep)) {
-		if (stage2_pte_leaf(ptep)) {
+		if (gstage_pte_leaf(ptep)) {
 			*ptep_level = current_level;
 			*ptepp = ptep;
 			return true;
···
 		if (current_level) {
 			current_level--;
 			*ptep_level = current_level;
-			ptep = (pte_t *)stage2_pte_page_vaddr(*ptep);
-			ptep = &ptep[stage2_pte_index(addr, current_level)];
+			ptep = (pte_t *)gstage_pte_page_vaddr(*ptep);
+			ptep = &ptep[gstage_pte_index(addr, current_level)];
 		} else {
 			ptep = NULL;
 		}
···
 	return false;
 }
 
-static void stage2_remote_tlb_flush(struct kvm *kvm, u32 level, gpa_t addr)
+static void gstage_remote_tlb_flush(struct kvm *kvm, u32 level, gpa_t addr)
 {
-	unsigned long size = PAGE_SIZE;
-	struct kvm_vmid *vmid = &kvm->arch.vmid;
+	unsigned long order = PAGE_SHIFT;
 
-	if (stage2_level_to_page_size(level, &size))
+	if (gstage_level_to_page_order(level, &order))
 		return;
-	addr &= ~(size - 1);
+	addr &= ~(BIT(order) - 1);
 
-	/*
-	 * TODO: Instead of cpu_online_mask, we should only target CPUs
-	 * where the Guest/VM is running.
-	 */
-	preempt_disable();
-	sbi_remote_hfence_gvma_vmid(cpu_online_mask, addr, size,
-				    READ_ONCE(vmid->vmid));
-	preempt_enable();
+	kvm_riscv_hfence_gvma_vmid_gpa(kvm, -1UL, 0, addr, BIT(order), order);
 }
 
-static int stage2_set_pte(struct kvm *kvm, u32 level,
+static int gstage_set_pte(struct kvm *kvm, u32 level,
 			   struct kvm_mmu_memory_cache *pcache,
 			   gpa_t addr, const pte_t *new_pte)
 {
-	u32 current_level = stage2_pgd_levels - 1;
+	u32 current_level = gstage_pgd_levels - 1;
 	pte_t *next_ptep = (pte_t *)kvm->arch.pgd;
-	pte_t *ptep = &next_ptep[stage2_pte_index(addr, current_level)];
+	pte_t *ptep = &next_ptep[gstage_pte_index(addr, current_level)];
 
 	if (current_level < level)
 		return -EINVAL;
 
 	while (current_level != level) {
-		if (stage2_pte_leaf(ptep))
+		if (gstage_pte_leaf(ptep))
 			return -EEXIST;
 
 		if (!pte_val(*ptep)) {
···
 			*ptep = pfn_pte(PFN_DOWN(__pa(next_ptep)),
 					__pgprot(_PAGE_TABLE));
 		} else {
-			if (stage2_pte_leaf(ptep))
+			if (gstage_pte_leaf(ptep))
 				return -EEXIST;
-			next_ptep = (pte_t *)stage2_pte_page_vaddr(*ptep);
+			next_ptep = (pte_t *)gstage_pte_page_vaddr(*ptep);
 		}
 
 		current_level--;
-		ptep = &next_ptep[stage2_pte_index(addr, current_level)];
+		ptep = &next_ptep[gstage_pte_index(addr, current_level)];
 	}
 
 	*ptep = *new_pte;
-	if (stage2_pte_leaf(ptep))
-		stage2_remote_tlb_flush(kvm, current_level, addr);
+	if (gstage_pte_leaf(ptep))
+		gstage_remote_tlb_flush(kvm, current_level, addr);
 
 	return 0;
 }
 
-static int stage2_map_page(struct kvm *kvm,
+static int gstage_map_page(struct kvm *kvm,
 			   struct kvm_mmu_memory_cache *pcache,
 			   gpa_t gpa, phys_addr_t hpa,
 			   unsigned long page_size,
···
 	pte_t new_pte;
 	pgprot_t prot;
 
-	ret = stage2_page_size_to_level(page_size, &level);
+	ret = gstage_page_size_to_level(page_size, &level);
 	if (ret)
 		return ret;
 
···
 	 *    PTE so that software can update these bits.
 	 *
 	 * We support both options mentioned above. To achieve this, we
-	 * always set 'A' and 'D' PTE bits at time of creating stage2
+	 * always set 'A' and 'D' PTE bits at time of creating G-stage
 	 * mapping. To support KVM dirty page logging with both options
-	 * mentioned above, we will write-protect stage2 PTEs to track
+	 * mentioned above, we will write-protect G-stage PTEs to track
 	 * dirty pages.
 	 */
 
···
 	new_pte = pfn_pte(PFN_DOWN(hpa), prot);
 	new_pte = pte_mkdirty(new_pte);
 
-	return stage2_set_pte(kvm, level, pcache, gpa, &new_pte);
+	return gstage_set_pte(kvm, level, pcache, gpa, &new_pte);
 }
 
-enum stage2_op {
-	STAGE2_OP_NOP = 0,	/* Nothing */
-	STAGE2_OP_CLEAR,	/* Clear/Unmap */
-	STAGE2_OP_WP,		/* Write-protect */
+enum gstage_op {
+	GSTAGE_OP_NOP = 0,	/* Nothing */
+	GSTAGE_OP_CLEAR,	/* Clear/Unmap */
+	GSTAGE_OP_WP,		/* Write-protect */
 };
 
-static void stage2_op_pte(struct kvm *kvm, gpa_t addr,
-			  pte_t *ptep, u32 ptep_level, enum stage2_op op)
+static void gstage_op_pte(struct kvm *kvm, gpa_t addr,
+			  pte_t *ptep, u32 ptep_level, enum gstage_op op)
 {
 	int i, ret;
 	pte_t *next_ptep;
 	u32 next_ptep_level;
 	unsigned long next_page_size, page_size;
 
-	ret = stage2_level_to_page_size(ptep_level, &page_size);
+	ret = gstage_level_to_page_size(ptep_level, &page_size);
 	if (ret)
 		return;
 
···
 	if (!pte_val(*ptep))
 		return;
 
-	if (ptep_level && !stage2_pte_leaf(ptep)) {
-		next_ptep = (pte_t *)stage2_pte_page_vaddr(*ptep);
+	if (ptep_level && !gstage_pte_leaf(ptep)) {
+		next_ptep = (pte_t *)gstage_pte_page_vaddr(*ptep);
 		next_ptep_level = ptep_level - 1;
-		ret = stage2_level_to_page_size(next_ptep_level,
+		ret = gstage_level_to_page_size(next_ptep_level,
 						&next_page_size);
 		if (ret)
 			return;
 
-		if (op == STAGE2_OP_CLEAR)
+		if (op == GSTAGE_OP_CLEAR)
 			set_pte(ptep, __pte(0));
 		for (i = 0; i < PTRS_PER_PTE; i++)
-			stage2_op_pte(kvm, addr + i * next_page_size,
+			gstage_op_pte(kvm, addr + i * next_page_size,
 					&next_ptep[i], next_ptep_level, op);
-		if (op == STAGE2_OP_CLEAR)
+		if (op == GSTAGE_OP_CLEAR)
 			put_page(virt_to_page(next_ptep));
 	} else {
-		if (op == STAGE2_OP_CLEAR)
+		if (op == GSTAGE_OP_CLEAR)
 			set_pte(ptep, __pte(0));
-		else if (op == STAGE2_OP_WP)
+		else if (op == GSTAGE_OP_WP)
 			set_pte(ptep, __pte(pte_val(*ptep) & ~_PAGE_WRITE));
-		stage2_remote_tlb_flush(kvm, ptep_level, addr);
+		gstage_remote_tlb_flush(kvm, ptep_level, addr);
 	}
 }
 
-static void stage2_unmap_range(struct kvm *kvm, gpa_t start,
+static void gstage_unmap_range(struct kvm *kvm, gpa_t start,
 			       gpa_t size, bool may_block)
 {
 	int ret;
···
 	gpa_t addr = start, end = start + size;
 
 	while (addr < end) {
-		found_leaf = stage2_get_leaf_entry(kvm, addr,
+		found_leaf = gstage_get_leaf_entry(kvm, addr,
 						   &ptep, &ptep_level);
-		ret = stage2_level_to_page_size(ptep_level, &page_size);
+		ret = gstage_level_to_page_size(ptep_level, &page_size);
 		if (ret)
 			break;
 
···
 			goto next;
 
 		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
-			stage2_op_pte(kvm, addr, ptep,
-				      ptep_level, STAGE2_OP_CLEAR);
+			gstage_op_pte(kvm, addr, ptep,
+				      ptep_level, GSTAGE_OP_CLEAR);
 
 next:
 		addr += page_size;
···
 	}
 }
 
-static void stage2_wp_range(struct kvm *kvm, gpa_t start, gpa_t end)
+static void gstage_wp_range(struct kvm *kvm, gpa_t start, gpa_t end)
 {
 	int ret;
 	pte_t *ptep;
···
 	unsigned long page_size;
 
 	while (addr < end) {
-		found_leaf = stage2_get_leaf_entry(kvm, addr,
+		found_leaf = gstage_get_leaf_entry(kvm, addr,
 						   &ptep, &ptep_level);
-		ret = stage2_level_to_page_size(ptep_level, &page_size);
+		ret = gstage_level_to_page_size(ptep_level, &page_size);
 		if (ret)
 			break;
 
···
 			goto next;
 
 		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
-			stage2_op_pte(kvm, addr, ptep,
-				      ptep_level, STAGE2_OP_WP);
+			gstage_op_pte(kvm, addr, ptep,
+				      ptep_level, GSTAGE_OP_WP);
 
 next:
 		addr += page_size;
 	}
 }
 
-static void stage2_wp_memory_region(struct kvm *kvm, int slot)
+static void gstage_wp_memory_region(struct kvm *kvm, int slot)
 {
 	struct kvm_memslots *slots = kvm_memslots(kvm);
 	struct kvm_memory_slot *memslot = id_to_memslot(slots, slot);
···
 	phys_addr_t end = (memslot->base_gfn + memslot->npages) << PAGE_SHIFT;
 
 	spin_lock(&kvm->mmu_lock);
-	stage2_wp_range(kvm, start, end);
+	gstage_wp_range(kvm, start, end);
 	spin_unlock(&kvm->mmu_lock);
 	kvm_flush_remote_tlbs(kvm);
 }
 
-static int stage2_ioremap(struct kvm *kvm, gpa_t gpa, phys_addr_t hpa,
+static int gstage_ioremap(struct kvm *kvm, gpa_t gpa, phys_addr_t hpa,
 			  unsigned long size, bool writable)
 {
 	pte_t pte;
···
 		if (!writable)
 			pte = pte_wrprotect(pte);
 
-		ret = kvm_mmu_topup_memory_cache(&pcache, stage2_pgd_levels);
+		ret = kvm_mmu_topup_memory_cache(&pcache, gstage_pgd_levels);
 		if (ret)
 			goto out;
 
 		spin_lock(&kvm->mmu_lock);
-		ret = stage2_set_pte(kvm, 0, &pcache, addr, &pte);
+		ret = gstage_set_pte(kvm, 0, &pcache, addr, &pte);
 		spin_unlock(&kvm->mmu_lock);
 		if (ret)
 			goto out;
···
 	phys_addr_t start = (base_gfn +  __ffs(mask)) << PAGE_SHIFT;
 	phys_addr_t end = (base_gfn + __fls(mask) + 1) << PAGE_SHIFT;
 
-	stage2_wp_range(kvm, start, end);
+	gstage_wp_range(kvm, start, end);
 }
 
 void kvm_arch_sync_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot)
···
 
 void kvm_arch_flush_shadow_all(struct kvm *kvm)
 {
-	kvm_riscv_stage2_free_pgd(kvm);
+	kvm_riscv_gstage_free_pgd(kvm);
 }
 
 void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
···
 	phys_addr_t size = slot->npages << PAGE_SHIFT;
 
 	spin_lock(&kvm->mmu_lock);
-	stage2_unmap_range(kvm, gpa, size, false);
+	gstage_unmap_range(kvm, gpa, size, false);
 	spin_unlock(&kvm->mmu_lock);
 }
 
···
 	 * the memory slot is write protected.
 	 */
 	if (change != KVM_MR_DELETE && new->flags & KVM_MEM_LOG_DIRTY_PAGES)
-		stage2_wp_memory_region(kvm, new->id);
+		gstage_wp_memory_region(kvm, new->id);
 }
 
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
···
 	 * space addressable by the KVM guest GPA space.
 	 */
 	if ((new->base_gfn + new->npages) >=
-	    (stage2_gpa_size >> PAGE_SHIFT))
+	    (gstage_gpa_size >> PAGE_SHIFT))
 		return -EFAULT;
 
 	hva = new->userspace_addr;
···
 				goto out;
 			}
 
-			ret = stage2_ioremap(kvm, gpa, pa,
+			ret = gstage_ioremap(kvm, gpa, pa,
 					     vm_end - vm_start, writable);
 			if (ret)
 				break;
···
 
 	spin_lock(&kvm->mmu_lock);
 	if (ret)
-		stage2_unmap_range(kvm, base_gpa, size, false);
+		gstage_unmap_range(kvm, base_gpa, size, false);
 	spin_unlock(&kvm->mmu_lock);
 
 out:
···
 	if (!kvm->arch.pgd)
 		return false;
 
-	stage2_unmap_range(kvm, range->start << PAGE_SHIFT,
+	gstage_unmap_range(kvm, range->start << PAGE_SHIFT,
 			   (range->end - range->start) << PAGE_SHIFT,
 			   range->may_block);
 	return false;
···
 
 	WARN_ON(range->end - range->start != 1);
 
-	ret = stage2_map_page(kvm, NULL, range->start << PAGE_SHIFT,
+	ret = gstage_map_page(kvm, NULL, range->start << PAGE_SHIFT,
 			      __pfn_to_phys(pfn), PAGE_SIZE, true, true);
 	if (ret) {
-		kvm_debug("Failed to map stage2 page (error %d)\n", ret);
+		kvm_debug("Failed to map G-stage page (error %d)\n", ret);
 		return true;
 	}
 
···
 
 	WARN_ON(size != PAGE_SIZE && size != PMD_SIZE && size != PGDIR_SIZE);
 
-	if (!stage2_get_leaf_entry(kvm, range->start << PAGE_SHIFT,
+	if (!gstage_get_leaf_entry(kvm, range->start << PAGE_SHIFT,
 				   &ptep, &ptep_level))
 		return false;
 
···
 
 	WARN_ON(size != PAGE_SIZE && size != PMD_SIZE && size != PGDIR_SIZE);
 
-	if (!stage2_get_leaf_entry(kvm, range->start << PAGE_SHIFT,
+	if (!gstage_get_leaf_entry(kvm, range->start << PAGE_SHIFT,
 				   &ptep, &ptep_level))
 		return false;
 
 	return pte_young(*ptep);
 }
 
-int kvm_riscv_stage2_map(struct kvm_vcpu *vcpu,
+int kvm_riscv_gstage_map(struct kvm_vcpu *vcpu,
 			 struct kvm_memory_slot *memslot,
 			 gpa_t gpa, unsigned long hva, bool is_write)
 {
···
 	}
 
 	/* We need minimum second+third level pages */
-	ret = kvm_mmu_topup_memory_cache(pcache, stage2_pgd_levels);
+	ret = kvm_mmu_topup_memory_cache(pcache, gstage_pgd_levels);
 	if (ret) {
-		kvm_err("Failed to topup stage2 cache\n");
+		kvm_err("Failed to topup G-stage cache\n");
 		return ret;
 	}
 
···
 	if (writeable) {
 		kvm_set_pfn_dirty(hfn);
 		mark_page_dirty(kvm, gfn);
-		ret = stage2_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
+		ret = gstage_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
 				      vma_pagesize, false, true);
 	} else {
-		ret = stage2_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
+		ret = gstage_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
 				      vma_pagesize, true, true);
 	}
 
 	if (ret)
-		kvm_err("Failed to map in stage2\n");
+		kvm_err("Failed to map in G-stage\n");
 
 out_unlock:
 	spin_unlock(&kvm->mmu_lock);
···
 	return ret;
 }
 
-int kvm_riscv_stage2_alloc_pgd(struct kvm *kvm)
+int kvm_riscv_gstage_alloc_pgd(struct kvm *kvm)
 {
 	struct page *pgd_page;
 
···
 	}
 
 	pgd_page = alloc_pages(GFP_KERNEL | __GFP_ZERO,
-				get_order(stage2_pgd_size));
+				get_order(gstage_pgd_size));
 	if (!pgd_page)
 		return -ENOMEM;
 	kvm->arch.pgd = page_to_virt(pgd_page);
···
 	return 0;
 }
 
-void kvm_riscv_stage2_free_pgd(struct kvm *kvm)
+void kvm_riscv_gstage_free_pgd(struct kvm *kvm)
 {
 	void *pgd = NULL;
 
 	spin_lock(&kvm->mmu_lock);
 	if (kvm->arch.pgd) {
-		stage2_unmap_range(kvm, 0UL, stage2_gpa_size, false);
+		gstage_unmap_range(kvm, 0UL, gstage_gpa_size, false);
 		pgd = READ_ONCE(kvm->arch.pgd);
 		kvm->arch.pgd = NULL;
 		kvm->arch.pgd_phys = 0;
···
 	spin_unlock(&kvm->mmu_lock);
 
 	if (pgd)
-		free_pages((unsigned long)pgd, get_order(stage2_pgd_size));
+		free_pages((unsigned long)pgd, get_order(gstage_pgd_size));
 }
 
-void kvm_riscv_stage2_update_hgatp(struct kvm_vcpu *vcpu)
+void kvm_riscv_gstage_update_hgatp(struct kvm_vcpu *vcpu)
 {
-	unsigned long hgatp = stage2_mode;
+	unsigned long hgatp = gstage_mode;
 	struct kvm_arch *k = &vcpu->kvm->arch;
 
 	hgatp |= (READ_ONCE(k->vmid.vmid) << HGATP_VMID_SHIFT) &
···
 
 	csr_write(CSR_HGATP, hgatp);
 
-	if (!kvm_riscv_stage2_vmid_bits())
-		__kvm_riscv_hfence_gvma_all();
+	if (!kvm_riscv_gstage_vmid_bits())
+		kvm_riscv_local_hfence_gvma_all();
 }
 
-void kvm_riscv_stage2_mode_detect(void)
+void kvm_riscv_gstage_mode_detect(void)
 {
 #ifdef CONFIG_64BIT
-	/* Try Sv48x4 stage2 mode */
+	/* Try Sv57x4 G-stage mode */
+	csr_write(CSR_HGATP, HGATP_MODE_SV57X4 << HGATP_MODE_SHIFT);
+	if ((csr_read(CSR_HGATP) >> HGATP_MODE_SHIFT) == HGATP_MODE_SV57X4) {
+		gstage_mode = (HGATP_MODE_SV57X4 << HGATP_MODE_SHIFT);
+		gstage_pgd_levels = 5;
+		goto skip_sv48x4_test;
+	}
+
+	/* Try Sv48x4 G-stage mode */
 	csr_write(CSR_HGATP, HGATP_MODE_SV48X4 << HGATP_MODE_SHIFT);
 	if ((csr_read(CSR_HGATP) >> HGATP_MODE_SHIFT) == HGATP_MODE_SV48X4) {
-		stage2_mode = (HGATP_MODE_SV48X4 << HGATP_MODE_SHIFT);
-		stage2_pgd_levels = 4;
+		gstage_mode = (HGATP_MODE_SV48X4 << HGATP_MODE_SHIFT);
+		gstage_pgd_levels = 4;
 	}
-	csr_write(CSR_HGATP, 0);
+skip_sv48x4_test:
 
-	__kvm_riscv_hfence_gvma_all();
+	csr_write(CSR_HGATP, 0);
+	kvm_riscv_local_hfence_gvma_all();
 #endif
 }
 
-unsigned long kvm_riscv_stage2_mode(void)
+unsigned long kvm_riscv_gstage_mode(void)
 {
-	return stage2_mode >> HGATP_MODE_SHIFT;
+	return gstage_mode >> HGATP_MODE_SHIFT;
 }
 
-int kvm_riscv_stage2_gpa_bits(void)
+int kvm_riscv_gstage_gpa_bits(void)
 {
-	return stage2_gpa_bits;
+	return gstage_gpa_bits;
 }

arch/riscv/kvm/tlb.S

···
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Copyright (C) 2019 Western Digital Corporation or its affiliates.
- *
- * Authors:
- *     Anup Patel <[e-mail]>
- */
-
-#include <linux/linkage.h>
-#include <asm/asm.h>
-
-	.text
-	.altmacro
-	.option norelax
-
-	/*
-	 * Instruction encoding of hfence.gvma is:
-	 * HFENCE.GVMA rs1, rs2
-	 * HFENCE.GVMA zero, rs2
-	 * HFENCE.GVMA rs1
-	 * HFENCE.GVMA
-	 *
-	 * rs1!=zero and rs2!=zero ==> HFENCE.GVMA rs1, rs2
-	 * rs1==zero and rs2!=zero ==> HFENCE.GVMA zero, rs2
-	 * rs1!=zero and rs2==zero ==> HFENCE.GVMA rs1
-	 * rs1==zero and rs2==zero ==> HFENCE.GVMA
-	 *
-	 * Instruction encoding of HFENCE.GVMA is:
-	 * 0110001 rs2(5) rs1(5) 000 00000 1110011
-	 */
-
-ENTRY(__kvm_riscv_hfence_gvma_vmid_gpa)
-	/*
-	 * rs1 = a0 (GPA >> 2)
-	 * rs2 = a1 (VMID)
-	 * HFENCE.GVMA a0, a1
-	 * 0110001 01011 01010 000 00000 1110011
-	 */
-	.word 0x62b50073
-	ret
-ENDPROC(__kvm_riscv_hfence_gvma_vmid_gpa)
-
-ENTRY(__kvm_riscv_hfence_gvma_vmid)
-	/*
-	 * rs1 = zero
-	 * rs2 = a0 (VMID)
-	 * HFENCE.GVMA zero, a0
-	 * 0110001 01010 00000 000 00000 1110011
-	 */
-	.word 0x62a00073
-	ret
-ENDPROC(__kvm_riscv_hfence_gvma_vmid)
-
-ENTRY(__kvm_riscv_hfence_gvma_gpa)
-	/*
-	 * rs1 = a0 (GPA >> 2)
-	 * rs2 = zero
-	 * HFENCE.GVMA a0
-	 * 0110001 00000 01010 000 00000 1110011
-	 */
-	.word 0x62050073
-	ret
-ENDPROC(__kvm_riscv_hfence_gvma_gpa)
-
-ENTRY(__kvm_riscv_hfence_gvma_all)
-	/*
-	 * rs1 = zero
-	 * rs2 = zero
-	 * HFENCE.GVMA
-	 * 0110001 00000 00000 000 00000 1110011
-	 */
-	.word 0x62000073
-	ret
-ENDPROC(__kvm_riscv_hfence_gvma_all)

arch/riscv/kvm/tlb.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2022 Ventana Micro Systems Inc.
+ */
+
+#include <linux/bitmap.h>
+#include <linux/cpumask.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/smp.h>
+#include <linux/kvm_host.h>
+#include <asm/cacheflush.h>
+#include <asm/csr.h>
+
+/*
+ * Instruction encoding of hfence.gvma is:
+ * HFENCE.GVMA rs1, rs2
+ * HFENCE.GVMA zero, rs2
+ * HFENCE.GVMA rs1
+ * HFENCE.GVMA
+ *
+ * rs1!=zero and rs2!=zero ==> HFENCE.GVMA rs1, rs2
+ * rs1==zero and rs2!=zero ==> HFENCE.GVMA zero, rs2
+ * rs1!=zero and rs2==zero ==> HFENCE.GVMA rs1
+ * rs1==zero and rs2==zero ==> HFENCE.GVMA
+ *
+ * Instruction encoding of HFENCE.GVMA is:
+ * 0110001 rs2(5) rs1(5) 000 00000 1110011
+ */
+
+void kvm_riscv_local_hfence_gvma_vmid_gpa(unsigned long vmid,
+					  gpa_t gpa, gpa_t gpsz,
+					  unsigned long order)
+{
+	gpa_t pos;
+
+	if (PTRS_PER_PTE < (gpsz >> order)) {
+		kvm_riscv_local_hfence_gvma_vmid_all(vmid);
+		return;
+	}
+
+	for (pos = gpa; pos < (gpa + gpsz); pos += BIT(order)) {
+		/*
+		 * rs1 = a0 (GPA >> 2)
+		 * rs2 = a1 (VMID)
+		 * HFENCE.GVMA a0, a1
+		 * 0110001 01011 01010 000 00000 1110011
+		 */
+		asm volatile ("srli a0, %0, 2\n"
+			      "add a1, %1, zero\n"
+			      ".word 0x62b50073\n"
+			      :: "r" (pos), "r" (vmid)
+			      : "a0", "a1", "memory");
+	}
+}
+
+void kvm_riscv_local_hfence_gvma_vmid_all(unsigned long vmid)
+{
+	/*
+	 * rs1 = zero
+	 * rs2 = a0 (VMID)
+	 * HFENCE.GVMA zero, a0
+	 * 0110001 01010 00000 000 00000 1110011
+	 */
+	asm volatile ("add a0, %0, zero\n"
+		      ".word 0x62a00073\n"
+		      :: "r" (vmid) : "a0", "memory");
+}
+
+void kvm_riscv_local_hfence_gvma_gpa(gpa_t gpa, gpa_t gpsz,
+				     unsigned long order)
+{
+	gpa_t pos;
+
+	if (PTRS_PER_PTE < (gpsz >> order)) {
+		kvm_riscv_local_hfence_gvma_all();
+		return;
+	}
+
+	for (pos = gpa; pos < (gpa + gpsz); pos += BIT(order)) {
+		/*
+		 * rs1 = a0 (GPA >> 2)
+		 * rs2 = zero
+		 * HFENCE.GVMA a0
+		 * 0110001 00000 01010 000 00000 1110011
+		 */
+		asm volatile ("srli a0, %0, 2\n"
+			      ".word 0x62050073\n"
+			      :: "r" (pos) : "a0", "memory");
+	}
+}
+
+void kvm_riscv_local_hfence_gvma_all(void)
+{
+	/*
+	 * rs1 = zero
+	 * rs2 = zero
+	 * HFENCE.GVMA
+	 * 0110001 00000 00000 000 00000 1110011
+	 */
+	asm volatile (".word 0x62000073" ::: "memory");
+}
+
+/*
+ * Instruction encoding of hfence.gvma is:
+ * HFENCE.VVMA rs1, rs2
+ * HFENCE.VVMA zero, rs2
+ * HFENCE.VVMA rs1
+ * HFENCE.VVMA
+ *
+ * rs1!=zero and rs2!=zero ==> HFENCE.VVMA rs1, rs2
+ * rs1==zero and rs2!=zero ==> HFENCE.VVMA zero, rs2
+ * rs1!=zero and rs2==zero ==> HFENCE.VVMA rs1
+ * rs1==zero and rs2==zero ==> HFENCE.VVMA
+ *
+ * Instruction encoding of HFENCE.VVMA is:
+ * 0010001 rs2(5) rs1(5) 000 00000 1110011
+ */
+
+void kvm_riscv_local_hfence_vvma_asid_gva(unsigned long vmid,
+					  unsigned long asid,
+					  unsigned long gva,
+					  unsigned long gvsz,
+					  unsigned long order)
+{
+	unsigned long pos, hgatp;
+
+	if (PTRS_PER_PTE < (gvsz >> order)) {
+		kvm_riscv_local_hfence_vvma_asid_all(vmid, asid);
+		return;
+	}
+
+	hgatp = csr_swap(CSR_HGATP, vmid << HGATP_VMID_SHIFT);
+
+	for (pos = gva; pos < (gva + gvsz); pos += BIT(order)) {
+		/*
+		 * rs1 = a0 (GVA)
+		 * rs2 = a1 (ASID)
+		 * HFENCE.VVMA a0, a1
+		 * 0010001 01011 01010 000 00000 1110011
+		 */
+		asm volatile ("add a0, %0, zero\n"
+			      "add a1, %1, zero\n"
+			      ".word 0x22b50073\n"
+			      :: "r" (pos), "r" (asid)
+			      : "a0", "a1", "memory");
+	}
+
+	csr_write(CSR_HGATP, hgatp);
+}
+
+void kvm_riscv_local_hfence_vvma_asid_all(unsigned long vmid,
+					  unsigned long asid)
+{
+	unsigned long hgatp;
+
+	hgatp = csr_swap(CSR_HGATP, vmid << HGATP_VMID_SHIFT);
+
+	/*
+	 * rs1 = zero
+	 * rs2 = a0 (ASID)
+	 * HFENCE.VVMA zero, a0
+	 * 0010001 01010 00000 000 00000 1110011
+	 */
+	asm volatile ("add a0, %0, zero\n"
+		      ".word 0x22a00073\n"
+		      :: "r" (asid) : "a0", "memory");
+
+	csr_write(CSR_HGATP, hgatp);
+}
+
+void kvm_riscv_local_hfence_vvma_gva(unsigned long vmid,
+				     unsigned long gva, unsigned long gvsz,
+				     unsigned long order)
+{
+	unsigned long pos, hgatp;
+
+	if (PTRS_PER_PTE < (gvsz >> order)) {
+		kvm_riscv_local_hfence_vvma_all(vmid);
+		return;
+	}
+
+	hgatp = csr_swap(CSR_HGATP, vmid << HGATP_VMID_SHIFT);
+
+	for (pos = gva; pos < (gva + gvsz); pos += BIT(order)) {
+		/*
+		 * rs1 = a0 (GVA)
+		 * rs2 = zero
+		 * HFENCE.VVMA a0
+		 * 0010001 00000 01010 000 00000 1110011
+		 */
+		asm volatile ("add a0, %0, zero\n"
+			      ".word 0x22050073\n"
+			      :: "r" (pos) : "a0", "memory");
+	}
+
+	csr_write(CSR_HGATP, hgatp);
+}
+
+void kvm_riscv_local_hfence_vvma_all(unsigned long vmid)
+{
+	unsigned long hgatp;
+
+	hgatp = csr_swap(CSR_HGATP, vmid << HGATP_VMID_SHIFT);
+
+	/*
+	 * rs1 = zero
+	 * rs2 = zero
+	 * HFENCE.VVMA
+	 * 0010001 00000 00000 000 00000 1110011
+	 */
+	asm volatile (".word 0x22000073" ::: "memory");
+
+	csr_write(CSR_HGATP, hgatp);
+}
+
+void kvm_riscv_local_tlb_sanitize(struct kvm_vcpu *vcpu)
+{
+	unsigned long vmid;
+
+	if (!kvm_riscv_gstage_vmid_bits() ||
+	    vcpu->arch.last_exit_cpu == vcpu->cpu)
+		return;
+
+	/*
+	 * On RISC-V platforms with hardware VMID support, we share same
+	 * VMID for all VCPUs of a particular Guest/VM. This means we might
+	 * have stale G-stage TLB entries on the current Host CPU due to
+	 * some other VCPU of the same Guest which ran previously on the
+	 * current Host CPU.
+	 *
+	 * To cleanup stale TLB entries, we simply flush all G-stage TLB
+	 * entries by VMID whenever underlying Host CPU changes for a VCPU.
+	 */
+
+	vmid = READ_ONCE(vcpu->kvm->arch.vmid.vmid);
+	kvm_riscv_local_hfence_gvma_vmid_all(vmid);
+}
+
+void kvm_riscv_fence_i_process(struct kvm_vcpu *vcpu)
+{
+	local_flush_icache_all();
+}
+
+void kvm_riscv_hfence_gvma_vmid_all_process(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vmid *vmid;
+
+	vmid = &vcpu->kvm->arch.vmid;
+	kvm_riscv_local_hfence_gvma_vmid_all(READ_ONCE(vmid->vmid));
+}
+
+void kvm_riscv_hfence_vvma_all_process(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vmid *vmid;
+
+	vmid = &vcpu->kvm->arch.vmid;
+	kvm_riscv_local_hfence_vvma_all(READ_ONCE(vmid->vmid));
+}
+
+static bool vcpu_hfence_dequeue(struct kvm_vcpu *vcpu,
+				struct kvm_riscv_hfence *out_data)
+{
+	bool ret = false;
+	struct kvm_vcpu_arch *varch = &vcpu->arch;
+
+	spin_lock(&varch->hfence_lock);
+
+	if (varch->hfence_queue[varch->hfence_head].type) {
+		memcpy(out_data, &varch->hfence_queue[varch->hfence_head],
+		       sizeof(*out_data));
+		varch->hfence_queue[varch->hfence_head].type = 0;
+
+		varch->hfence_head++;
+		if (varch->hfence_head == KVM_RISCV_VCPU_MAX_HFENCE)
+			varch->hfence_head = 0;
+
+		ret = true;
+	}
+
+	spin_unlock(&varch->hfence_lock);
+
+	return ret;
+}
+
+static bool vcpu_hfence_enqueue(struct kvm_vcpu *vcpu,
+				const struct kvm_riscv_hfence *data)
+{
+	bool ret = false;
+	struct kvm_vcpu_arch *varch = &vcpu->arch;
+
+	spin_lock(&varch->hfence_lock);
+
+	if (!varch->hfence_queue[varch->hfence_tail].type) {
+		memcpy(&varch->hfence_queue[varch->hfence_tail],
+		       data, sizeof(*data));
+
+		varch->hfence_tail++;
+		if (varch->hfence_tail == KVM_RISCV_VCPU_MAX_HFENCE)
+			varch->hfence_tail = 0;
+
+		ret = true;
+	}
+
+	spin_unlock(&varch->hfence_lock);
+
+	return ret;
+}
+
+void kvm_riscv_hfence_process(struct kvm_vcpu *vcpu)
+{
+	struct kvm_riscv_hfence d = { 0 };
+	struct kvm_vmid *v = &vcpu->kvm->arch.vmid;
+
+	while (vcpu_hfence_dequeue(vcpu, &d)) {
+		switch (d.type) {
+		case KVM_RISCV_HFENCE_UNKNOWN:
+			break;
+		case KVM_RISCV_HFENCE_GVMA_VMID_GPA:
+			kvm_riscv_local_hfence_gvma_vmid_gpa(
+						READ_ONCE(v->vmid),
+						d.addr, d.size, d.order);
+			break;
+		case KVM_RISCV_HFENCE_VVMA_ASID_GVA:
+			kvm_riscv_local_hfence_vvma_asid_gva(
+						READ_ONCE(v->vmid), d.asid,
+						d.addr, d.size, d.order);
+			break;
+		case KVM_RISCV_HFENCE_VVMA_ASID_ALL:
+			kvm_riscv_local_hfence_vvma_asid_all(
+						READ_ONCE(v->vmid), d.asid);
+			break;
+		case KVM_RISCV_HFENCE_VVMA_GVA:
+			kvm_riscv_local_hfence_vvma_gva(
+						READ_ONCE(v->vmid),
+						d.addr, d.size, d.order);
+			break;
+		default:
+			break;
+		}
+	}
+}
+
+static void make_xfence_request(struct kvm *kvm,
+				unsigned long hbase, unsigned long hmask,
+				unsigned int req, unsigned int fallback_req,
+				const struct kvm_riscv_hfence *data)
+{
+	unsigned long i;
+	struct kvm_vcpu *vcpu;
+	unsigned int actual_req = req;
+	DECLARE_BITMAP(vcpu_mask, KVM_MAX_VCPUS);
+
+	bitmap_clear(vcpu_mask, 0, KVM_MAX_VCPUS);
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		if (hbase != -1UL) {
+			if (vcpu->vcpu_id < hbase)
+				continue;
+			if (!(hmask & (1UL << (vcpu->vcpu_id - hbase))))
+				continue;
+		}
+
+		bitmap_set(vcpu_mask, i, 1);
+
+		if (!data || !data->type)
+			continue;
+
+		/*
+		 * Enqueue hfence data to VCPU hfence queue. If we don't
+		 * have space in the VCPU hfence queue then fallback to
+		 * a more conservative hfence request.
+		 */
+		if (!vcpu_hfence_enqueue(vcpu, data))
+			actual_req = fallback_req;
+	}
+
+	kvm_make_vcpus_request_mask(kvm, actual_req, vcpu_mask);
+}
+
+void kvm_riscv_fence_i(struct kvm *kvm,
+		       unsigned long hbase, unsigned long hmask)
+{
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_FENCE_I,
+			    KVM_REQ_FENCE_I, NULL);
+}
+
+void kvm_riscv_hfence_gvma_vmid_gpa(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    gpa_t gpa, gpa_t gpsz,
+				    unsigned long order)
+{
+	struct kvm_riscv_hfence data;
+
+	data.type = KVM_RISCV_HFENCE_GVMA_VMID_GPA;
+	data.asid = 0;
+	data.addr = gpa;
+	data.size = gpsz;
+	data.order = order;
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE,
+			    KVM_REQ_HFENCE_GVMA_VMID_ALL, &data);
+}
+
+void kvm_riscv_hfence_gvma_vmid_all(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask)
+{
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE_GVMA_VMID_ALL,
+			    KVM_REQ_HFENCE_GVMA_VMID_ALL, NULL);
+}
+
+void kvm_riscv_hfence_vvma_asid_gva(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    unsigned long gva, unsigned long gvsz,
+				    unsigned long order, unsigned long asid)
+{
+	struct kvm_riscv_hfence data;
+
+	data.type = KVM_RISCV_HFENCE_VVMA_ASID_GVA;
+	data.asid = asid;
+	data.addr = gva;
+	data.size = gvsz;
+	data.order = order;
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE,
+			    KVM_REQ_HFENCE_VVMA_ALL, &data);
+}
+
+void kvm_riscv_hfence_vvma_asid_all(struct kvm *kvm,
+				    unsigned long hbase, unsigned long hmask,
+				    unsigned long asid)
+{
+	struct kvm_riscv_hfence data;
+
+	data.type = KVM_RISCV_HFENCE_VVMA_ASID_ALL;
+	data.asid = asid;
+	data.addr = data.size = data.order = 0;
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE,
+			    KVM_REQ_HFENCE_VVMA_ALL, &data);
+}
+
+void kvm_riscv_hfence_vvma_gva(struct kvm *kvm,
+			       unsigned long hbase, unsigned long hmask,
+			       unsigned long gva, unsigned long gvsz,
+			       unsigned long order)
+{
+	struct kvm_riscv_hfence data;
+
+	data.type = KVM_RISCV_HFENCE_VVMA_GVA;
+	data.asid = 0;
+	data.addr = gva;
+	data.size = gvsz;
+	data.order = order;
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE,
+			    KVM_REQ_HFENCE_VVMA_ALL, &data);
+}
+
+void kvm_riscv_hfence_vvma_all(struct kvm *kvm,
+			       unsigned long hbase, unsigned long hmask)
+{
+	make_xfence_request(kvm, hbase, hmask, KVM_REQ_HFENCE_VVMA_ALL,
+			    KVM_REQ_HFENCE_VVMA_ALL, NULL);
+}

arch/riscv/kvm/vcpu.c

···
 	if (loaded)
 		kvm_arch_vcpu_put(vcpu);
 
+	vcpu->arch.last_exit_cpu = -1;
+
 	memcpy(csr, reset_csr, sizeof(*csr));
 
 	memcpy(cntx, reset_cntx, sizeof(*cntx));
···
 
 	WRITE_ONCE(vcpu->arch.irqs_pending, 0);
 	WRITE_ONCE(vcpu->arch.irqs_pending_mask, 0);
+
+	vcpu->arch.hfence_head = 0;
+	vcpu->arch.hfence_tail = 0;
+	memset(vcpu->arch.hfence_queue, 0, sizeof(vcpu->arch.hfence_queue));
 
 	/* Reset the guest CSRs for hotplug usecase */
 	if (loaded)
···
 
 	/* Setup ISA features available to VCPU */
 	vcpu->arch.isa = riscv_isa_extension_base(NULL) & KVM_RISCV_ISA_ALLOWED;
+
+	/* Setup VCPU hfence queue */
+	spin_lock_init(&vcpu->arch.hfence_lock);
 
 	/* Setup reset state of shadow SSTATUS and HSTATUS CSRs */
 	cntx = &vcpu->arch.guest_reset_context;
···
 	/* Cleanup VCPU timer */
 	kvm_riscv_vcpu_timer_deinit(vcpu);
 
-	/* Free unused pages pre-allocated for Stage2 page table mappings */
+	/* Free unused pages pre-allocated for G-stage page table mappings */
 	kvm_mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);
 }
 
···
 	return 0;
 }
 
+/* Mapping between KVM ISA Extension ID & Host ISA extension ID */
+static unsigned long kvm_isa_ext_arr[] = {
+	RISCV_ISA_EXT_a,
+	RISCV_ISA_EXT_c,
+	RISCV_ISA_EXT_d,
+	RISCV_ISA_EXT_f,
+	RISCV_ISA_EXT_h,
+	RISCV_ISA_EXT_i,
+	RISCV_ISA_EXT_m,
+};
+
+static int kvm_riscv_vcpu_get_reg_isa_ext(struct kvm_vcpu *vcpu,
+					  const struct kvm_one_reg *reg)
+{
+	unsigned long __user *uaddr =
+			(unsigned long __user *)(unsigned long)reg->addr;
+	unsigned long reg_num = reg->id & ~(KVM_REG_ARCH_MASK |
+					    KVM_REG_SIZE_MASK |
+					    KVM_REG_RISCV_ISA_EXT);
+	unsigned long reg_val = 0;
+	unsigned long host_isa_ext;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+
+	if (reg_num >= KVM_RISCV_ISA_EXT_MAX || reg_num >= ARRAY_SIZE(kvm_isa_ext_arr))
+		return -EINVAL;
+
+	host_isa_ext = kvm_isa_ext_arr[reg_num];
+	if (__riscv_isa_extension_available(&vcpu->arch.isa, host_isa_ext))
+		reg_val = 1; /* Mark the given extension as available */
+
+	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_set_reg_isa_ext(struct kvm_vcpu *vcpu,
+					  const struct kvm_one_reg *reg)
+{
+	unsigned long __user *uaddr =
+			(unsigned long __user *)(unsigned long)reg->addr;
+	unsigned long reg_num = reg->id & ~(KVM_REG_ARCH_MASK |
+					    KVM_REG_SIZE_MASK |
+					    KVM_REG_RISCV_ISA_EXT);
+	unsigned long reg_val;
+	unsigned long host_isa_ext;
+	unsigned long host_isa_ext_mask;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+
+	if (reg_num >= KVM_RISCV_ISA_EXT_MAX || reg_num >= ARRAY_SIZE(kvm_isa_ext_arr))
+		return -EINVAL;
+
+	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	host_isa_ext = kvm_isa_ext_arr[reg_num];
+	if (!__riscv_isa_extension_available(NULL, host_isa_ext))
+		return	-EOPNOTSUPP;
+
+	if (host_isa_ext >= RISCV_ISA_EXT_BASE &&
+	    host_isa_ext < RISCV_ISA_EXT_MAX) {
+		/*
+		 * Multi-letter ISA extension. Currently there is no provision
+		 * to enable/disable the multi-letter ISA extensions for guests.
+		 * Return success if the request is to enable any ISA extension
+		 * that is available in the hardware.
+		 * Return -EOPNOTSUPP otherwise.
+		 */
+		if (!reg_val)
+			return -EOPNOTSUPP;
+		else
+			return 0;
+	}
+
+	/* Single letter base ISA extension */
+	if (!vcpu->arch.ran_atleast_once) {
+		host_isa_ext_mask = BIT_MASK(host_isa_ext);
+		if (!reg_val && (host_isa_ext_mask & KVM_RISCV_ISA_DISABLE_ALLOWED))
+			vcpu->arch.isa &= ~host_isa_ext_mask;
+		else
+			vcpu->arch.isa |= host_isa_ext_mask;
+		vcpu->arch.isa &= riscv_isa_extension_base(NULL);
+		vcpu->arch.isa &= KVM_RISCV_ISA_ALLOWED;
+		kvm_riscv_vcpu_fp_reset(vcpu);
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
 static int kvm_riscv_vcpu_set_reg(struct kvm_vcpu *vcpu,
 				  const struct kvm_one_reg *reg)
 {
···
 	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_FP_D)
 		return kvm_riscv_vcpu_set_reg_fp(vcpu, reg,
 						 KVM_REG_RISCV_FP_D);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_ISA_EXT)
+		return kvm_riscv_vcpu_set_reg_isa_ext(vcpu, reg);
 
 	return -EINVAL;
 }
···
 	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_FP_D)
 		return kvm_riscv_vcpu_get_reg_fp(vcpu, reg,
 						 KVM_REG_RISCV_FP_D);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_ISA_EXT)
+		return kvm_riscv_vcpu_get_reg_isa_ext(vcpu, reg);
 
 	return -EINVAL;
 }
···
 	csr_write(CSR_HVIP, csr->hvip);
 	csr_write(CSR_VSATP, csr->vsatp);
 
-	kvm_riscv_stage2_update_hgatp(vcpu);
+	kvm_riscv_gstage_update_hgatp(vcpu);
 
 	kvm_riscv_vcpu_timer_restore(vcpu);
 
···
 			kvm_riscv_reset_vcpu(vcpu);
 
 		if (kvm_check_request(KVM_REQ_UPDATE_HGATP, vcpu))
-			kvm_riscv_stage2_update_hgatp(vcpu);
+			kvm_riscv_gstage_update_hgatp(vcpu);
 
-		if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu))
-			__kvm_riscv_hfence_gvma_all();
+		if (kvm_check_request(KVM_REQ_FENCE_I, vcpu))
+			kvm_riscv_fence_i_process(vcpu);
+
+		/*
+		 * The generic KVM_REQ_TLB_FLUSH is same as
+		 * KVM_REQ_HFENCE_GVMA_VMID_ALL
+		 */
+		if (kvm_check_request(KVM_REQ_HFENCE_GVMA_VMID_ALL, vcpu))
+			kvm_riscv_hfence_gvma_vmid_all_process(vcpu);
+
+		if (kvm_check_request(KVM_REQ_HFENCE_VVMA_ALL, vcpu))
+			kvm_riscv_hfence_vvma_all_process(vcpu);
+
+		if (kvm_check_request(KVM_REQ_HFENCE, vcpu))
+			kvm_riscv_hfence_process(vcpu);
 	}
 }
 
···
 {
 	guest_state_enter_irqoff();
 	__kvm_riscv_switch_to(&vcpu->arch);
+	vcpu->arch.last_exit_cpu = vcpu->cpu;
 	guest_state_exit_irqoff();
 }
 
···
 		/* Check conditions before entering the guest */
 		cond_resched();
 
-		kvm_riscv_stage2_vmid_update(vcpu);
+		kvm_riscv_gstage_vmid_update(vcpu);
 
 		kvm_riscv_check_vcpu_requests(vcpu);
 
···
 		kvm_riscv_update_hvip(vcpu);
 
 		if (ret <= 0 ||
-		    kvm_riscv_stage2_vmid_ver_changed(&vcpu->kvm->arch.vmid) ||
+		    kvm_riscv_gstage_vmid_ver_changed(&vcpu->kvm->arch.vmid) ||
 		    kvm_request_pending(vcpu)) {
 			vcpu->mode = OUTSIDE_GUEST_MODE;
 			local_irq_enable();
···
 			kvm_vcpu_srcu_read_lock(vcpu);
 			continue;
 		}
+
+		/*
+		 * Cleanup stale TLB enteries
+		 *
+		 * Note: This should be done after G-stage VMID has been
+		 * updated using kvm_riscv_gstage_vmid_ver_changed()
+		 */
+		kvm_riscv_local_tlb_sanitize(vcpu);
 
 		guest_timing_enter_irqoff();
 

arch/riscv/kvm/vcpu_exit.c

···
 	return 0;
 }
 
-static int stage2_page_fault(struct kvm_vcpu *vcpu, struct kvm_run *run,
+static int gstage_page_fault(struct kvm_vcpu *vcpu, struct kvm_run *run,
 			     struct kvm_cpu_trap *trap)
 {
 	struct kvm_memory_slot *memslot;
···
 		};
 	}
 
-	ret = kvm_riscv_stage2_map(vcpu, memslot, fault_addr, hva,
+	ret = kvm_riscv_gstage_map(vcpu, memslot, fault_addr, hva,
 		(trap->scause == EXC_STORE_GUEST_PAGE_FAULT) ? true : false);
 	if (ret < 0)
 		return ret;
···
 	case EXC_LOAD_GUEST_PAGE_FAULT:
 	case EXC_STORE_GUEST_PAGE_FAULT:
 		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)
-			ret = stage2_page_fault(vcpu, run, trap);
+			ret = gstage_page_fault(vcpu, run, trap);
 		break;
 	case EXC_SUPERVISOR_SYSCALL:
 		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)

arch/riscv/kvm/vcpu_sbi_replace.c

···
 				      struct kvm_cpu_trap *utrap, bool *exit)
 {
 	int ret = 0;
-	unsigned long i;
-	struct cpumask cm;
-	struct kvm_vcpu *tmp;
 	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
 	unsigned long hmask = cp->a0;
 	unsigned long hbase = cp->a1;
 	unsigned long funcid = cp->a6;
 
-	cpumask_clear(&cm);
-	kvm_for_each_vcpu(i, tmp, vcpu->kvm) {
-		if (hbase != -1UL) {
-			if (tmp->vcpu_id < hbase)
-				continue;
-			if (!(hmask & (1UL << (tmp->vcpu_id - hbase))))
-				continue;
-		}
-		if (tmp->cpu < 0)
-			continue;
-		cpumask_set_cpu(tmp->cpu, &cm);
-	}
-
 	switch (funcid) {
 	case SBI_EXT_RFENCE_REMOTE_FENCE_I:
-		ret = sbi_remote_fence_i(&cm);
+		kvm_riscv_fence_i(vcpu->kvm, hbase, hmask);
 		break;
 	case SBI_EXT_RFENCE_REMOTE_SFENCE_VMA:
-		ret = sbi_remote_hfence_vvma(&cm, cp->a2, cp->a3);
+		if (cp->a2 == 0 && cp->a3 == 0)
+			kvm_riscv_hfence_vvma_all(vcpu->kvm, hbase, hmask);
+		else
+			kvm_riscv_hfence_vvma_gva(vcpu->kvm, hbase, hmask,
+						  cp->a2, cp->a3, PAGE_SHIFT);
 		break;
 	case SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID:
-		ret = sbi_remote_hfence_vvma_asid(&cm, cp->a2,
-						  cp->a3, cp->a4);
+		if (cp->a2 == 0 && cp->a3 == 0)
+			kvm_riscv_hfence_vvma_asid_all(vcpu->kvm,
+						       hbase, hmask, cp->a4);
+		else
+			kvm_riscv_hfence_vvma_asid_gva(vcpu->kvm,
+						       hbase, hmask,
+						       cp->a2, cp->a3,
+						       PAGE_SHIFT, cp->a4);
 		break;
 	case SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA:
 	case SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA_VMID:
 	case SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA:
 	case SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID:
-	/* TODO: implement for nested hypervisor case */
+		/*
+		 * Until nested virtualization is implemented, the
+		 * SBI HFENCE calls should be treated as NOPs
+		 */
+		break;
 	default:
 		ret = -EOPNOTSUPP;
 	}

arch/riscv/kvm/vcpu_sbi_v01.c

···
 	int i, ret = 0;
 	u64 next_cycle;
 	struct kvm_vcpu *rvcpu;
-	struct cpumask cm;
 	struct kvm *kvm = vcpu->kvm;
 	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
 
···
 		if (utrap->scause)
 			break;
 
-		cpumask_clear(&cm);
-		for_each_set_bit(i, &hmask, BITS_PER_LONG) {
-			rvcpu = kvm_get_vcpu_by_id(vcpu->kvm, i);
-			if (rvcpu->cpu < 0)
-				continue;
-			cpumask_set_cpu(rvcpu->cpu, &cm);
-		}
 		if (cp->a7 == SBI_EXT_0_1_REMOTE_FENCE_I)
-			ret = sbi_remote_fence_i(&cm);
-		else if (cp->a7 == SBI_EXT_0_1_REMOTE_SFENCE_VMA)
-			ret = sbi_remote_hfence_vvma(&cm, cp->a1, cp->a2);
-		else
-			ret = sbi_remote_hfence_vvma_asid(&cm, cp->a1, cp->a2, cp->a3);
+			kvm_riscv_fence_i(vcpu->kvm, 0, hmask);
+		else if (cp->a7 == SBI_EXT_0_1_REMOTE_SFENCE_VMA) {
+			if (cp->a1 == 0 && cp->a2 == 0)
+				kvm_riscv_hfence_vvma_all(vcpu->kvm,
+							  0, hmask);
+			else
+				kvm_riscv_hfence_vvma_gva(vcpu->kvm,
+							  0, hmask,
+							  cp->a1, cp->a2,
+							  PAGE_SHIFT);
+		} else {
+			if (cp->a1 == 0 && cp->a2 == 0)
+				kvm_riscv_hfence_vvma_asid_all(vcpu->kvm,
+							       0, hmask,
+							       cp->a3);
+			else
+				kvm_riscv_hfence_vvma_asid_gva(vcpu->kvm,
+							       0, hmask,
+							       cp->a1, cp->a2,
+							       PAGE_SHIFT,
+							       cp->a3);
+		}
 		break;
 	default:
 		ret = -EINVAL;

arch/riscv/kvm/vm.c

···
 {
 	int r;
 
-	r = kvm_riscv_stage2_alloc_pgd(kvm);
+	r = kvm_riscv_gstage_alloc_pgd(kvm);
 	if (r)
 		return r;
 
-	r = kvm_riscv_stage2_vmid_init(kvm);
+	r = kvm_riscv_gstage_vmid_init(kvm);
 	if (r) {
-		kvm_riscv_stage2_free_pgd(kvm);
+		kvm_riscv_gstage_free_pgd(kvm);
 		return r;
 	}
 
···
 		r = KVM_USER_MEM_SLOTS;
 		break;
 	case KVM_CAP_VM_GPA_BITS:
-		r = kvm_riscv_stage2_gpa_bits();
+		r = kvm_riscv_gstage_gpa_bits();
 		break;
 	default:
 		r = 0;

arch/riscv/kvm/vmid.c

···
 #include <linux/errno.h>
 #include <linux/err.h>
 #include <linux/module.h>
+#include <linux/smp.h>
 #include <linux/kvm_host.h>
 #include <asm/csr.h>
-#include <asm/sbi.h>
 
 static unsigned long vmid_version = 1;
 static unsigned long vmid_next;
 static unsigned long vmid_bits;
 static DEFINE_SPINLOCK(vmid_lock);
 
-void kvm_riscv_stage2_vmid_detect(void)
+void kvm_riscv_gstage_vmid_detect(void)
 {
 	unsigned long old;
 
···
 	csr_write(CSR_HGATP, old);
 
 	/* We polluted local TLB so flush all guest TLB */
-	__kvm_riscv_hfence_gvma_all();
+	kvm_riscv_local_hfence_gvma_all();
 
 	/* We don't use VMID bits if they are not sufficient */
 	if ((1UL << vmid_bits) < num_possible_cpus())
 		vmid_bits = 0;
 }
 
-unsigned long kvm_riscv_stage2_vmid_bits(void)
+unsigned long kvm_riscv_gstage_vmid_bits(void)
 {
 	return vmid_bits;
 }
 
-int kvm_riscv_stage2_vmid_init(struct kvm *kvm)
+int kvm_riscv_gstage_vmid_init(struct kvm *kvm)
 {
 	/* Mark the initial VMID and VMID version invalid */
 	kvm->arch.vmid.vmid_version = 0;
···
 	return 0;
 }
 
-bool kvm_riscv_stage2_vmid_ver_changed(struct kvm_vmid *vmid)
+bool kvm_riscv_gstage_vmid_ver_changed(struct kvm_vmid *vmid)
 {
 	if (!vmid_bits)
 		return false;
···
 			READ_ONCE(vmid_version));
 }
 
-void kvm_riscv_stage2_vmid_update(struct kvm_vcpu *vcpu)
+static void __local_hfence_gvma_all(void *info)
+{
+	kvm_riscv_local_hfence_gvma_all();
+}
+
+void kvm_riscv_gstage_vmid_update(struct kvm_vcpu *vcpu)
 {
 	unsigned long i;
 	struct kvm_vcpu *v;
 	struct kvm_vmid *vmid = &vcpu->kvm->arch.vmid;
 
-	if (!kvm_riscv_stage2_vmid_ver_changed(vmid))
+	if (!kvm_riscv_gstage_vmid_ver_changed(vmid))
 		return;
 
 	spin_lock(&vmid_lock);
···
 	 * We need to re-check the vmid_version here to ensure that if
 	 * another vcpu already allocated a valid vmid for this vm.
 	 */
-	if (!kvm_riscv_stage2_vmid_ver_changed(vmid)) {
+	if (!kvm_riscv_gstage_vmid_ver_changed(vmid)) {
 		spin_unlock(&vmid_lock);
 		return;
 	}
···
 		 * instances is invalid and we have force VMID re-assignement
 		 * for all Guest instances. The Guest instances that were not
 		 * running will automatically pick-up new VMIDs because will
-		 * call kvm_riscv_stage2_vmid_update() whenever they enter
+		 * call kvm_riscv_gstage_vmid_update() whenever they enter
 		 * in-kernel run loop. For Guest instances that are already
 		 * running, we force VM exits on all host CPUs using IPI and
 		 * flush all Guest TLBs.
 		 */
-		sbi_remote_hfence_gvma(cpu_online_mask, 0, 0);
+		on_each_cpu_mask(cpu_online_mask, __local_hfence_gvma_all,
+				 NULL, 1);
 	}
 
 	vmid->vmid = vmid_next;
···
 
 	spin_unlock(&vmid_lock);
 
-	/* Request stage2 page table update for all VCPUs */
+	/* Request G-stage page table update for all VCPUs */
 	kvm_for_each_vcpu(i, v, vcpu->kvm)
 		kvm_make_request(KVM_REQ_UPDATE_HGATP, v);
 }

arch/s390/include/asm/uv.h

···
 /*
  * Ultravisor Interfaces
  *
- * Copyright IBM Corp. 2019
+ * Copyright IBM Corp. 2019, 2022
  *
  * Author(s):
  *	Vasily Gorbik <[e-mail]>
···
 #define UVC_CMD_UNPIN_PAGE_SHARED	0x0342
 #define UVC_CMD_SET_SHARED_ACCESS	0x1000
 #define UVC_CMD_REMOVE_SHARED_ACCESS	0x1001
+#define UVC_CMD_RETR_ATTEST		0x1020
 
 /* Bits in installed uv calls */
 enum uv_cmds_inst {
···
 	BIT_UVC_CMD_UNSHARE_ALL = 20,
 	BIT_UVC_CMD_PIN_PAGE_SHARED = 21,
 	BIT_UVC_CMD_UNPIN_PAGE_SHARED = 22,
+	BIT_UVC_CMD_RETR_ATTEST = 28,
 };
 
 enum uv_feat_ind {
···
 	u64 reserved08[3];
 	u64 paddr;
 	u64 reserved28;
+} __packed __aligned(8);
+
+/* Retrieve Attestation Measurement */
+struct uv_cb_attest {
+	struct uv_cb_header header;	/* 0x0000 */
+	u64 reserved08[2];		/* 0x0008 */
+	u64 arcb_addr;			/* 0x0018 */
+	u64 cont_token;			/* 0x0020 */
+	u8  reserved28[6];		/* 0x0028 */
+	u16 user_data_len;		/* 0x002e */
+	u8  user_data[256];		/* 0x0030 */
+	u32 reserved130[3];		/* 0x0130 */
+	u32 meas_len;			/* 0x013c */
+	u64 meas_addr;			/* 0x0140 */
+	u8  config_uid[16];		/* 0x0148 */
+	u32 reserved158;		/* 0x0158 */
+	u32 add_data_len;		/* 0x015c */
+	u64 add_data_addr;		/* 0x0160 */
+	u64 reserved168[4];		/* 0x0168 */
 } __packed __aligned(8);
 
 static inline int __uv_call(unsigned long r1, unsigned long r2)

arch/s390/include/uapi/asm/uvdevice.h

···
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ *  Copyright IBM Corp. 2022
+ *  Author(s): Steffen Eiden <[e-mail]>
+ */
+#ifndef __S390_ASM_UVDEVICE_H
+#define __S390_ASM_UVDEVICE_H
+
+#include <linux/types.h>
+
+struct uvio_ioctl_cb {
+	__u32 flags;
+	__u16 uv_rc;			/* UV header rc value */
+	__u16 uv_rrc;			/* UV header rrc value */
+	__u64 argument_addr;		/* Userspace address of uvio argument */
+	__u32 argument_len;
+	__u8  reserved14[0x40 - 0x14];	/* must be zero */
+};
+
+#define UVIO_ATT_USER_DATA_LEN		0x100
+#define UVIO_ATT_UID_LEN		0x10
+struct uvio_attest {
+	__u64 arcb_addr;				/* 0x0000 */
+	__u64 meas_addr;				/* 0x0008 */
+	__u64 add_data_addr;				/* 0x0010 */
+	__u8  user_data[UVIO_ATT_USER_DATA_LEN];	/* 0x0018 */
+	__u8  config_uid[UVIO_ATT_UID_LEN];		/* 0x0118 */
+	__u32 arcb_len;					/* 0x0128 */
+	__u32 meas_len;					/* 0x012c */
+	__u32 add_data_len;				/* 0x0130 */
+	__u16 user_data_len;				/* 0x0134 */
+	__u16 reserved136;				/* 0x0136 */
+};
+
+/*
+ * The following max values define an upper length for the IOCTL in/out buffers.
+ * However, they do not represent the maximum the Ultravisor allows which is
+ * often way smaller. By allowing larger buffer sizes we hopefully do not need
+ * to update the code with every machine update. It is therefore possible for
+ * userspace to request more memory than actually used by kernel/UV.
+ */
+#define UVIO_ATT_ARCB_MAX_LEN		0x100000
+#define UVIO_ATT_MEASUREMENT_MAX_LEN	0x8000
+#define UVIO_ATT_ADDITIONAL_MAX_LEN	0x8000
+
+#define UVIO_DEVICE_NAME "uv"
+#define UVIO_TYPE_UVC 'u'
+
+#define UVIO_IOCTL_ATT _IOWR(UVIO_TYPE_UVC, 0x01, struct uvio_ioctl_cb)
+
+#endif /* __S390_ASM_UVDEVICE_H */

arch/s390/kvm/gaccess.c

···
 	PROT_TYPE_IEP  = 4,
 };
 
-static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva,
-		     u8 ar, enum gacc_mode mode, enum prot_type prot)
+static int trans_exc_ending(struct kvm_vcpu *vcpu, int code, unsigned long gva, u8 ar,
+			    enum gacc_mode mode, enum prot_type prot, bool terminate)
 {
 	struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm;
 	struct trans_exc_code_bits *tec;
···
 		case PROT_TYPE_DAT:
 			tec->b61 = 1;
 			break;
+		}
+		if (terminate) {
+			tec->b56 = 0;
+			tec->b60 = 0;
+			tec->b61 = 0;
 		}
 		fallthrough;
 	case PGM_ASCE_TYPE:
···
 		break;
 	}
 	return code;
+}
+
+static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva, u8 ar,
+		     enum gacc_mode mode, enum prot_type prot)
+{
+	return trans_exc_ending(vcpu, code, gva, ar, mode, prot, false);
 }
 
 static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce,
···
 		data += fragment_len;
 		ga = kvm_s390_logical_to_effective(vcpu, ga + fragment_len);
 	}
-	if (rc > 0)
-		rc = trans_exc(vcpu, rc, ga, ar, mode, prot);
+	if (rc > 0) {
+		bool terminate = (mode == GACC_STORE) && (idx > 0);
+
+		rc = trans_exc_ending(vcpu, rc, ga, ar, mode, prot, terminate);
+	}
 out_unlock:
 	if (need_ipte_lock)
 		ipte_unlock(vcpu);

arch/x86/include/asm/cpufeatures.h

···
 #define X86_FEATURE_SEV			(19*32+ 1) /* AMD Secure Encrypted Virtualization */
 #define X86_FEATURE_VM_PAGE_FLUSH	(19*32+ 2) /* "" VM Page Flush MSR is supported */
 #define X86_FEATURE_SEV_ES		(19*32+ 3) /* AMD Secure Encrypted Virtualization - Encrypted State */
+#define X86_FEATURE_V_TSC_AUX		(19*32+ 9) /* "" Virtual TSC_AUX */
 #define X86_FEATURE_SME_COHERENT	(19*32+10) /* "" AMD hardware-enforced cache coherency */
 
 /*

arch/x86/include/asm/kvm-x86-ops.h

···
 KVM_X86_OP(msr_filter_changed)
 KVM_X86_OP(complete_emulated_msr)
 KVM_X86_OP(vcpu_deliver_sipi_vector)
+KVM_X86_OP_OPTIONAL_RET0(vcpu_get_apicv_inhibit_reasons);
 
 #undef KVM_X86_OP
 #undef KVM_X86_OP_OPTIONAL

arch/x86/include/asm/kvm-x86-pmu-ops.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#if !defined(KVM_X86_PMU_OP) || !defined(KVM_X86_PMU_OP_OPTIONAL)
+BUILD_BUG_ON(1)
+#endif
+
+/*
+ * KVM_X86_PMU_OP() and KVM_X86_PMU_OP_OPTIONAL() are used to help generate
+ * both DECLARE/DEFINE_STATIC_CALL() invocations and
+ * "static_call_update()" calls.
+ *
+ * KVM_X86_PMU_OP_OPTIONAL() can be used for those functions that can have
+ * a NULL definition, for example if "static_call_cond()" will be used
+ * at the call sites.
+ */
+KVM_X86_PMU_OP(pmc_perf_hw_id)
+KVM_X86_PMU_OP(pmc_is_enabled)
+KVM_X86_PMU_OP(pmc_idx_to_pmc)
+KVM_X86_PMU_OP(rdpmc_ecx_to_pmc)
+KVM_X86_PMU_OP(msr_idx_to_pmc)
+KVM_X86_PMU_OP(is_valid_rdpmc_ecx)
+KVM_X86_PMU_OP(is_valid_msr)
+KVM_X86_PMU_OP(get_msr)
+KVM_X86_PMU_OP(set_msr)
+KVM_X86_PMU_OP(refresh)
+KVM_X86_PMU_OP(init)
+KVM_X86_PMU_OP(reset)
+KVM_X86_PMU_OP_OPTIONAL(deliver_pmi)
+KVM_X86_PMU_OP_OPTIONAL(cleanup)
+
+#undef KVM_X86_PMU_OP
+#undef KVM_X86_PMU_OP_OPTIONAL

arch/x86/include/asm/kvm_host.h

···
 /*
  * kvm_mmu_page_role tracks the properties of a shadow page (where shadow page
  * also includes TDP pages) to determine whether or not a page can be used in
- * the given MMU context.  This is a subset of the overall kvm_mmu_role to
+ * the given MMU context.  This is a subset of the overall kvm_cpu_role to
  * minimize the size of kvm_memory_slot.arch.gfn_track, i.e. allows allocating
  * 2 bytes per gfn instead of 4 bytes per gfn.
  *
- * Indirect upper-level shadow pages are tracked for write-protection via
+ * Upper-level shadow pages having gptes are tracked for write-protection via
  * gfn_track.  As above, gfn_track is a 16 bit counter, so KVM must not create
  * more than 2^16-1 upper-level shadow pages at a single gfn, otherwise
  * gfn_track will overflow and explosions will ensure.
···
 		unsigned smap_andnot_wp:1;
 		unsigned ad_disabled:1;
 		unsigned guest_mode:1;
-		unsigned :6;
+		unsigned passthrough:1;
+		unsigned :5;
 
 		/*
 		 * This is left at the top of the word so that
···
 	struct {
 		unsigned int valid:1;
 		unsigned int execonly:1;
-		unsigned int cr0_pg:1;
-		unsigned int cr4_pae:1;
 		unsigned int cr4_pse:1;
 		unsigned int cr4_pke:1;
 		unsigned int cr4_smap:1;
···
 	};
 };
 
-union kvm_mmu_role {
+union kvm_cpu_role {
 	u64 as_u64;
 	struct {
 		union kvm_mmu_page_role base;
···
 			 struct kvm_mmu_page *sp);
 	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva, hpa_t root_hpa);
 	struct kvm_mmu_root_info root;
-	union kvm_mmu_role mmu_role;
-	u8 root_level;
-	u8 shadow_root_level;
-	u8 ept_ad;
-	bool direct_map;
-	struct kvm_mmu_root_info prev_roots[KVM_MMU_NUM_PREV_ROOTS];
-
-	/*
-	 * Bitmap; bit set = permission fault
-	 * Byte index: page fault error code [4:1]
-	 * Bit index: pte permissions in ACC_* format
-	 */
-	u8 permissions[16];
+	union kvm_cpu_role cpu_role;
+	union kvm_mmu_page_role root_role;
 
 	/*
 	* The pkru_mask indicates if protection key checks are needed.  It
···
 	* Each domain has 2 bits which are ANDed with AD and WD from PKRU.
 	*/
 	u32 pkru_mask;
+
+	struct kvm_mmu_root_info prev_roots[KVM_MMU_NUM_PREV_ROOTS];
+
+	/*
+	 * Bitmap; bit set = permission fault
+	 * Byte index: page fault error code [4:1]
+	 * Bit index: pte permissions in ACC_* format
+	 */
+	u8 permissions[16];
 
 	u64 *pae_root;
 	u64 *pml4_root;
···
 struct kvm_vcpu_xen {
 	u64 hypercall_rip;
 	u32 current_runstate;
-	bool vcpu_info_set;
-	bool vcpu_time_info_set;
-	bool runstate_set;
-	struct gfn_to_hva_cache vcpu_info_cache;
-	struct gfn_to_hva_cache vcpu_time_info_cache;
-	struct gfn_to_hva_cache runstate_cache;
+	u8 upcall_vector;
+	struct gfn_to_pfn_cache vcpu_info_cache;
+	struct gfn_to_pfn_cache vcpu_time_info_cache;
+	struct gfn_to_pfn_cache runstate_cache;
 	u64 last_steal;
 	u64 runstate_entry_time;
 	u64 runstate_times[4];
 	unsigned long evtchn_pending_sel;
+	u32 vcpu_id; /* The Xen / ACPI vCPU ID */
+	u32 timer_virq;
+	u64 timer_expires; /* In guest epoch */
+	atomic_t timer_pending;
+	struct hrtimer timer;
+	int poll_evtchn;
+	struct timer_list poll_timer;
 };
 
 struct kvm_vcpu_arch {
···
 	gpa_t time;
 	struct pvclock_vcpu_time_info hv_clock;
 	unsigned int hw_tsc_khz;
-	struct gfn_to_hva_cache pv_time;
-	bool pv_time_enabled;
+	struct gfn_to_pfn_cache pv_time;
 	/* set guest stopped flag in pvclock flags field */
 	bool pvclock_set_guest_stopped_request;
 
···
 
 /* Xen emulation context */
 struct kvm_xen {
+	u32 xen_version;
 	bool long_mode;
 	u8 upcall_vector;
 	struct gfn_to_pfn_cache shinfo_cache;
+	struct idr evtchn_ports;
+	unsigned long poll_mask[BITS_TO_LONGS(KVM_MAX_VCPUS)];
 };
 
 enum kvm_irqchip_mode {
···
 	u64 cur_tsc_offset;
 	u64 cur_tsc_generation;
 	int nr_vcpus_matched_tsc;
+
+	u32 default_tsc_khz;
 
 	seqcount_raw_spinlock_t pvclock_sc;
 	bool use_master_clock;
···
 
 struct kvm_vcpu_stat {
 	struct kvm_vcpu_stat_generic generic;
+	u64 pf_taken;
 	u64 pf_fixed;
+	u64 pf_emulate;
+	u64 pf_spurious;
+	u64 pf_fast;
+	u64 pf_mmio_spte_created;
 	u64 pf_guest;
 	u64 tlb_flush;
 	u64 invlpg;
···
 	int cpu_dirty_log_size;
 	void (*update_cpu_dirty_logging)(struct kvm_vcpu *vcpu);
 
-	/* pmu operations of sub-arch */
-	const struct kvm_pmu_ops *pmu_ops;
 	const struct kvm_x86_nested_ops *nested_ops;
 
 	void (*vcpu_blocking)(struct kvm_vcpu *vcpu);
···
 	int (*complete_emulated_msr)(struct kvm_vcpu *vcpu, int err);
 
 	void (*vcpu_deliver_sipi_vector)(struct kvm_vcpu *vcpu, u8 vector);
+
+	/*
+	 * Returns vCPU specific APICv inhibit reasons
+	 */
+	unsigned long (*vcpu_get_apicv_inhibit_reasons)(struct kvm_vcpu *vcpu);
 };
 
 struct kvm_x86_nested_ops {
 	void (*leave_nested)(struct kvm_vcpu *vcpu);
 	int (*check_events)(struct kvm_vcpu *vcpu);
+	bool (*handle_page_fault_workaround)(struct kvm_vcpu *vcpu,
+					     struct x86_exception *fault);
 	bool (*hv_timer_pending)(struct kvm_vcpu *vcpu);
 	void (*triple_fault)(struct kvm_vcpu *vcpu);
 	int (*get_state)(struct kvm_vcpu *vcpu,
···
 	unsigned int (*handle_intel_pt_intr)(void);
 
 	struct kvm_x86_ops *runtime_ops;
+	struct kvm_pmu_ops *pmu_ops;
 };
 
 struct kvm_arch_async_pf {
···
 #define KVM_X86_OP_OPTIONAL KVM_X86_OP
 #define KVM_X86_OP_OPTIONAL_RET0 KVM_X86_OP
 #include <asm/kvm-x86-ops.h>
-
-static inline void kvm_ops_static_call_update(void)
-{
-#define __KVM_X86_OP(func) \
-	static_call_update(kvm_x86_##func, kvm_x86_ops.func);
-#define KVM_X86_OP(func) \
-	WARN_ON(!kvm_x86_ops.func); __KVM_X86_OP(func)
-#define KVM_X86_OP_OPTIONAL __KVM_X86_OP
-#define KVM_X86_OP_OPTIONAL_RET0(func) \
-	static_call_update(kvm_x86_##func, (void *)kvm_x86_ops.func ? : \
-					   (void *)__static_call_return0);
-#include <asm/kvm-x86-ops.h>
-#undef __KVM_X86_OP
-}
 
 #define __KVM_HAVE_ARCH_VM_ALLOC
 static inline struct kvm *kvm_arch_alloc_vm(void)
···
 				struct x86_exception *exception);
 
 bool kvm_apicv_activated(struct kvm *kvm);
+bool kvm_vcpu_apicv_activated(struct kvm_vcpu *vcpu);
 void kvm_vcpu_update_apicv(struct kvm_vcpu *vcpu);
 void __kvm_set_or_clear_apicv_inhibit(struct kvm *kvm,
 				      enum kvm_apicv_inhibit reason, bool set);
···
 	 KVM_X86_QUIRK_CD_NW_CLEARED |		\
 	 KVM_X86_QUIRK_LAPIC_MMIO_HOLE |	\
 	 KVM_X86_QUIRK_OUT_7E_INC_RIP |		\
-	 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT)
+	 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT |	\
+	 KVM_X86_QUIRK_FIX_HYPERCALL_INSN)
 
 #endif /* _ASM_X86_KVM_HOST_H */

arch/x86/include/asm/uaccess.h

···
 
 #endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 
+#ifdef CONFIG_CC_HAS_ASM_GOTO_TIED_OUTPUT
+#define __try_cmpxchg_user_asm(itype, ltype, _ptr, _pold, _new, label)	({ \
+	bool success;							\
+	__typeof__(_ptr) _old = (__typeof__(_ptr))(_pold);		\
+	__typeof__(*(_ptr)) __old = *_old;				\
+	__typeof__(*(_ptr)) __new = (_new);				\
+	asm_volatile_goto("\n"						\
+		     "1: " LOCK_PREFIX "cmpxchg"itype" %[new], %[ptr]\n"\
+		     _ASM_EXTABLE_UA(1b, %l[label])			\
+		     : CC_OUT(z) (success),				\
+		       [ptr] "+m" (*_ptr),				\
+		       [old] "+a" (__old)				\
+		     : [new] ltype (__new)				\
+		     : "memory"						\
+		     : label);						\
+	if (unlikely(!success))						\
+		*_old = __old;						\
+	likely(success);					})
+
+#ifdef CONFIG_X86_32
+#define __try_cmpxchg64_user_asm(_ptr, _pold, _new, label)	({	\
+	bool success;							\
+	__typeof__(_ptr) _old = (__typeof__(_ptr))(_pold);		\
+	__typeof__(*(_ptr)) __old = *_old;				\
+	__typeof__(*(_ptr)) __new = (_new);				\
+	asm_volatile_goto("\n"						\
+		     "1: " LOCK_PREFIX "cmpxchg8b %[ptr]\n"		\
+		     _ASM_EXTABLE_UA(1b, %l[label])			\
+		     : CC_OUT(z) (success),				\
+		       "+A" (__old),					\
+		       [ptr] "+m" (*_ptr)				\
+		     : "b" ((u32)__new),				\
+		       "c" ((u32)((u64)__new >> 32))			\
+		     : "memory"						\
+		     : label);						\
+	if (unlikely(!success))						\
+		*_old = __old;						\
+	likely(success);					})
+#endif // CONFIG_X86_32
+#else  // !CONFIG_CC_HAS_ASM_GOTO_TIED_OUTPUT
+#define __try_cmpxchg_user_asm(itype, ltype, _ptr, _pold, _new, label)	({ \
+	int __err = 0;							\
+	bool success;							\
+	__typeof__(_ptr) _old = (__typeof__(_ptr))(_pold);		\
+	__typeof__(*(_ptr)) __old = *_old;				\
+	__typeof__(*(_ptr)) __new = (_new);				\
+	asm volatile("\n"						\
+		     "1: " LOCK_PREFIX "cmpxchg"itype" %[new], %[ptr]\n"\
+		     CC_SET(z)						\
+		     "2:\n"						\
+		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG,	\
+					   %[errout])			\
+		     : CC_OUT(z) (success),				\
+		       [errout] "+r" (__err),				\
+		       [ptr] "+m" (*_ptr),				\
+		       [old] "+a" (__old)				\
+		     : [new] ltype (__new)				\
+		     : "memory", "cc");					\
+	if (unlikely(__err))						\
+		goto label;						\
+	if (unlikely(!success))						\
+		*_old = __old;						\
+	likely(success);					})
+
+#ifdef CONFIG_X86_32
+/*
+ * Unlike the normal CMPXCHG, hardcode ECX for both success/fail and error.
+ * There are only six GPRs available and four (EAX, EBX, ECX, and EDX) are
+ * hardcoded by CMPXCHG8B, leaving only ESI and EDI.  If the compiler uses
+ * both ESI and EDI for the memory operand, compilation will fail if the error
+ * is an input+output as there will be no register available for input.
+ */
+#define __try_cmpxchg64_user_asm(_ptr, _pold, _new, label)	({	\
+	int __result;							\
+	__typeof__(_ptr) _old = (__typeof__(_ptr))(_pold);		\
+	__typeof__(*(_ptr)) __old = *_old;				\
+	__typeof__(*(_ptr)) __new = (_new);				\
+	asm volatile("\n"						\
+		     "1: " LOCK_PREFIX "cmpxchg8b %[ptr]\n"		\
+		     "mov $0, %%ecx\n\t"				\
+		     "setz %%cl\n"					\
+		     "2:\n"						\
+		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %%ecx) \
+		     : [result]"=c" (__result),				\
+		       "+A" (__old),					\
+		       [ptr] "+m" (*_ptr)				\
+		     : "b" ((u32)__new),				\
+		       "c" ((u32)((u64)__new >> 32))			\
+		     : "memory", "cc");					\
+	if (unlikely(__result < 0))					\
+		goto label;						\
+	if (unlikely(!__result))					\
+		*_old = __old;						\
+	likely(__result);					})
+#endif // CONFIG_X86_32
+#endif // CONFIG_CC_HAS_ASM_GOTO_TIED_OUTPUT
+
 /* FIXME: this hack is definitely wrong -AK */
 struct __large_struct { unsigned long buf[100]; };
 #define __m(x) (*(struct __large_struct __user *)(x))
···
 	if (unlikely(__gu_err)) goto err_label;					\
 } while (0)
 #endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
+
+extern void __try_cmpxchg_user_wrong_size(void);
+
+#ifndef CONFIG_X86_32
+#define __try_cmpxchg64_user_asm(_ptr, _oldp, _nval, _label)		\
+	__try_cmpxchg_user_asm("q", "r", (_ptr), (_oldp), (_nval), _label)
+#endif
+
+/*
+ * Force the pointer to u<size> to match the size expected by the asm helper.
+ * clang/LLVM compiles all cases and only discards the unused paths after
+ * processing errors, which breaks i386 if the pointer is an 8-byte value.
+ */
+#define unsafe_try_cmpxchg_user(_ptr, _oldp, _nval, _label) ({			\
+	bool __ret;								\
+	__chk_user_ptr(_ptr);							\
+	switch (sizeof(*(_ptr))) {						\
+	case 1:	__ret = __try_cmpxchg_user_asm("b", "q",			\
+					       (__force u8 *)(_ptr), (_oldp),	\
+					       (_nval), _label);		\
+		break;								\
+	case 2:	__ret = __try_cmpxchg_user_asm("w", "r",			\
+					       (__force u16 *)(_ptr), (_oldp),	\
+					       (_nval), _label);		\
+		break;								\
+	case 4:	__ret = __try_cmpxchg_user_asm("l", "r",			\
+					       (__force u32 *)(_ptr), (_oldp),	\
+					       (_nval), _label);		\
+		break;								\
+	case 8:	__ret = __try_cmpxchg64_user_asm((__force u64 *)(_ptr), (_oldp),\
+						 (_nval), _label);		\
+		break;								\
+	default: __try_cmpxchg_user_wrong_size();				\
+	}									\
+	__ret;						})
+
+/* "Returns" 0 on success, 1 on failure, -EFAULT if the access faults. */
+#define __try_cmpxchg_user(_ptr, _oldp, _nval, _label)	({		\
+	int __ret = -EFAULT;						\
+	__uaccess_begin_nospec();					\
+	__ret = !unsafe_try_cmpxchg_user(_ptr, _oldp, _nval, _label);	\
+_label:									\
+	__uaccess_end();						\
+	__ret;								\
+							})
 
 /*
  * We want the unsafe accessors to always be inlined and use

arch/x86/include/asm/vmx.h

···
 #define EPT_VIOLATION_ACC_READ_BIT	0
 #define EPT_VIOLATION_ACC_WRITE_BIT	1
 #define EPT_VIOLATION_ACC_INSTR_BIT	2
-#define EPT_VIOLATION_READABLE_BIT	3
-#define EPT_VIOLATION_WRITABLE_BIT	4
-#define EPT_VIOLATION_EXECUTABLE_BIT	5
+#define EPT_VIOLATION_RWX_SHIFT		3
+#define EPT_VIOLATION_GVA_IS_VALID_BIT	7
 #define EPT_VIOLATION_GVA_TRANSLATED_BIT 8
 #define EPT_VIOLATION_ACC_READ		(1 << EPT_VIOLATION_ACC_READ_BIT)
 #define EPT_VIOLATION_ACC_WRITE		(1 << EPT_VIOLATION_ACC_WRITE_BIT)
 #define EPT_VIOLATION_ACC_INSTR		(1 << EPT_VIOLATION_ACC_INSTR_BIT)
-#define EPT_VIOLATION_READABLE		(1 << EPT_VIOLATION_READABLE_BIT)
-#define EPT_VIOLATION_WRITABLE		(1 << EPT_VIOLATION_WRITABLE_BIT)
-#define EPT_VIOLATION_EXECUTABLE	(1 << EPT_VIOLATION_EXECUTABLE_BIT)
+#define EPT_VIOLATION_RWX_MASK		(VMX_EPT_RWX_MASK << EPT_VIOLATION_RWX_SHIFT)
+#define EPT_VIOLATION_GVA_IS_VALID	(1 << EPT_VIOLATION_GVA_IS_VALID_BIT)
 #define EPT_VIOLATION_GVA_TRANSLATED	(1 << EPT_VIOLATION_GVA_TRANSLATED_BIT)
 
 /*

arch/x86/include/uapi/asm/kvm.h

···
 	struct kvm_vcpu_events events;
 };
 
-#define KVM_X86_QUIRK_LINT0_REENABLED	   (1 << 0)
-#define KVM_X86_QUIRK_CD_NW_CLEARED	   (1 << 1)
-#define KVM_X86_QUIRK_LAPIC_MMIO_HOLE	   (1 << 2)
-#define KVM_X86_QUIRK_OUT_7E_INC_RIP	   (1 << 3)
-#define KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT (1 << 4)
+#define KVM_X86_QUIRK_LINT0_REENABLED		(1 << 0)
+#define KVM_X86_QUIRK_CD_NW_CLEARED		(1 << 1)
+#define KVM_X86_QUIRK_LAPIC_MMIO_HOLE		(1 << 2)
+#define KVM_X86_QUIRK_OUT_7E_INC_RIP		(1 << 3)
+#define KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT	(1 << 4)
+#define KVM_X86_QUIRK_FIX_HYPERCALL_INSN	(1 << 5)
 
 #define KVM_STATE_NESTED_FORMAT_VMX	0
 #define KVM_STATE_NESTED_FORMAT_SVM	1

arch/x86/kernel/asm-offsets_64.c

···
 
 #include <asm/ia32.h>
 
-#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS)
+#if defined(CONFIG_KVM_GUEST)
 #include <asm/kvm_para.h>
 #endif
 
···
 	BLANK();
 #endif
 
-#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS)
+#if defined(CONFIG_KVM_GUEST)
 	OFFSET(KVM_STEAL_TIME_preempted, kvm_steal_time, preempted);
 	BLANK();
 #endif

arch/x86/kernel/fpu/core.c

···
 #include <asm/traps.h>
 #include <asm/irq_regs.h>
 
+#include <uapi/asm/kvm.h>
+
 #include <linux/hardirq.h>
 #include <linux/pkeys.h>
 #include <linux/vmalloc.h>
···
 	gfpu->fpstate		= fpstate;
 	gfpu->xfeatures		= fpu_user_cfg.default_features;
 	gfpu->perm		= fpu_user_cfg.default_features;
-	gfpu->uabi_size		= fpu_user_cfg.default_size;
+
+	/*
+	 * KVM sets the FP+SSE bits in the XSAVE header when copying FPU state
+	 * to userspace, even when XSAVE is unsupported, so that restoring FPU
+	 * state on a different CPU that does support XSAVE can cleanly load
+	 * the incoming state using its natural XSAVE.  In other words, KVM's
+	 * uABI size may be larger than this host's default size.  Conversely,
+	 * the default size should never be larger than KVM's base uABI size;
+	 * all features that can expand the uABI size must be opt-in.
+	 */
+	gfpu->uabi_size		= sizeof(struct kvm_xsave);
+	if (WARN_ON_ONCE(fpu_user_cfg.default_size > gfpu->uabi_size))
+		gfpu->uabi_size = fpu_user_cfg.default_size;
+
 	fpu_init_guest_permissions(gfpu);
 
 	return true;

arch/x86/kernel/kvm.c

···
 {
 	u32 key = hash_32(token, KVM_TASK_SLEEP_HASHBITS);
 	struct kvm_task_sleep_head *b = &async_pf_sleepers[key];
-	struct kvm_task_sleep_node *n;
+	struct kvm_task_sleep_node *n, *dummy = NULL;
 
 	if (token == ~0) {
 		apf_task_wake_all();
···
 	n = _find_apf_task(b, token);
 	if (!n) {
 		/*
-		 * async PF was not yet handled.
-		 * Add dummy entry for the token.
+		 * Async #PF not yet handled, add a dummy entry for the token.
+		 * Allocating the token must be down outside of the raw lock
+		 * as the allocator is preemptible on PREEMPT_RT kernels.
 		 */
-		n = kzalloc(sizeof(*n), GFP_ATOMIC);
-		if (!n) {
-			/*
-			 * Allocation failed! Busy wait while other cpu
-			 * handles async PF.
-			 */
+		if (!dummy) {
 			raw_spin_unlock(&b->lock);
-			cpu_relax();
+			dummy = kzalloc(sizeof(*dummy), GFP_ATOMIC);
+
+			/*
+			 * Continue looping on allocation failure, eventually
+			 * the async #PF will be handled and allocating a new
+			 * node will be unnecessary.
+			 */
+			if (!dummy)
+				cpu_relax();
+
+			/*
+			 * Recheck for async #PF completion before enqueueing
+			 * the dummy token to avoid duplicate list entries.
+			 */
 			goto again;
 		}
-		n->token = token;
-		n->cpu = smp_processor_id();
-		init_swait_queue_head(&n->wq);
-		hlist_add_head(&n->link, &b->list);
+		dummy->token = token;
+		dummy->cpu = smp_processor_id();
+		init_swait_queue_head(&dummy->wq);
+		hlist_add_head(&dummy->link, &b->list);
+		dummy = NULL;
 	} else {
 		apf_task_wake_one(n);
 	}
 	raw_spin_unlock(&b->lock);
-	return;
+
+	/* A dummy token might be allocated and ultimately not used.  */
+	if (dummy)
+		kfree(dummy);
 }
 EXPORT_SYMBOL_GPL(kvm_async_pf_task_wake);
 
···
 }
 #endif
 
+#if defined(CONFIG_X86_32) || !defined(CONFIG_SMP)
+bool __kvm_vcpu_is_preempted(long cpu);
+
+__visible bool __kvm_vcpu_is_preempted(long cpu)
+{
+	struct kvm_steal_time *src = &per_cpu(steal_time, cpu);
+
+	return !!(src->preempted & KVM_VCPU_PREEMPTED);
+}
+PV_CALLEE_SAVE_REGS_THUNK(__kvm_vcpu_is_preempted);
+
+#else
+
+#include <asm/asm-offsets.h>
+
+extern bool __raw_callee_save___kvm_vcpu_is_preempted(long);
+
+/*
+ * Hand-optimize version for x86-64 to avoid 8 64-bit register saving and
+ * restoring to/from the stack.
+ */
+asm(
+".pushsection .text;"
+".global __raw_callee_save___kvm_vcpu_is_preempted;"
+".type __raw_callee_save___kvm_vcpu_is_preempted, @function;"
+"__raw_callee_save___kvm_vcpu_is_preempted:"
+ASM_ENDBR
+"movq	__per_cpu_offset(,%rdi,8), %rax;"
+"cmpb	$0, " __stringify(KVM_STEAL_TIME_preempted) "+steal_time(%rax);"
+"setne	%al;"
+ASM_RET
+".size __raw_callee_save___kvm_vcpu_is_preempted, .-__raw_callee_save___kvm_vcpu_is_preempted;"
+".popsection");
+
+#endif
+
 static void __init kvm_guest_init(void)
 {
 	int i;
···
 	if (kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
 		has_steal_clock = 1;
 		static_call_update(pv_steal_clock, kvm_steal_clock);
+
+		pv_ops.lock.vcpu_is_preempted =
+			PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
 	}
 
 	if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
···
 	}
 }
 
-#ifdef CONFIG_X86_32
-__visible bool __kvm_vcpu_is_preempted(long cpu)
-{
-	struct kvm_steal_time *src = &per_cpu(steal_time, cpu);
-
-	return !!(src->preempted & KVM_VCPU_PREEMPTED);
-}
-PV_CALLEE_SAVE_REGS_THUNK(__kvm_vcpu_is_preempted);
-
-#else
-
-#include <asm/asm-offsets.h>
-
-extern bool __raw_callee_save___kvm_vcpu_is_preempted(long);
-
-/*
- * Hand-optimize version for x86-64 to avoid 8 64-bit register saving and
- * restoring to/from the stack.
- */
-asm(
-".pushsection .text;"
-".global __raw_callee_save___kvm_vcpu_is_preempted;"
-".type __raw_callee_save___kvm_vcpu_is_preempted, @function;"
-"__raw_callee_save___kvm_vcpu_is_preempted:"
-ASM_ENDBR
-"movq	__per_cpu_offset(,%rdi,8), %rax;"
-"cmpb	$0, " __stringify(KVM_STEAL_TIME_preempted) "+steal_time(%rax);"
-"setne	%al;"
-ASM_RET
-".size __raw_callee_save___kvm_vcpu_is_preempted, .-__raw_callee_save___kvm_vcpu_is_preempted;"
-".popsection");
-
-#endif
-
 /*
  * Setup pv_lock_ops to exploit KVM_FEATURE_PV_UNHALT if present.
  */
···
 	pv_ops.lock.wait = kvm_wait;
 	pv_ops.lock.kick = kvm_kick_cpu;
 
-	if (kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
-		pv_ops.lock.vcpu_is_preempted =
-			PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
-	}
 	/*
 	 * When PV spinlock is enabled which is preferred over
 	 * virt_spin_lock(), virt_spin_lock_key's value is meaningless.

arch/x86/kernel/kvmclock.c

···
 
 static int __init kvm_setup_vsyscall_timeinfo(void)
 {
-	if (!kvm_para_available() || !kvmclock)
+	if (!kvm_para_available() || !kvmclock || nopv)
 		return 0;
 
 	kvmclock_init_mem();

arch/x86/kvm/i8259.c

···
 				 */
 				irq2 = 7;
 			intno = s->pics[1].irq_base + irq2;
-			irq = irq2 + 8;
 		} else
 			intno = s->pics[0].irq_base + irq;
 	} else {

arch/x86/kvm/irq.c

···
  */
 int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
-	if (lapic_in_kernel(vcpu))
-		return apic_has_pending_timer(vcpu);
+	int r = 0;
 
-	return 0;
+	if (lapic_in_kernel(vcpu))
+		r = apic_has_pending_timer(vcpu);
+	if (kvm_xen_timer_enabled(vcpu))
+		r += kvm_xen_has_pending_timer(vcpu);
+
+	return r;
 }
 EXPORT_SYMBOL(kvm_cpu_has_pending_timer);
 
···
 {
 	if (lapic_in_kernel(vcpu))
 		kvm_inject_apic_timer_irqs(vcpu);
+	if (kvm_xen_timer_enabled(vcpu))
+		kvm_xen_inject_timer_irqs(vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs);
 

arch/x86/kvm/irq_comm.c

···
 		if (!level)
 			return -1;
 
-		return kvm_xen_set_evtchn_fast(e, kvm);
+		return kvm_xen_set_evtchn_fast(&e->xen_evtchn, kvm);
 #endif
 	default:
 		break;

arch/x86/kvm/lapic.c

···
 	if (apic->lapic_timer.hv_timer_in_use)
 		cancel_hv_timer(apic);
 	preempt_enable();
+	atomic_set(&apic->lapic_timer.pending, 0);
 }
 
 static void apic_update_lvtt(struct kvm_lapic *apic)
···
 	tsc_deadline = apic->lapic_timer.expired_tscdeadline;
 	apic->lapic_timer.expired_tscdeadline = 0;
 	guest_tsc = kvm_read_l1_tsc(vcpu, rdtsc());
-	apic->lapic_timer.advance_expire_delta = guest_tsc - tsc_deadline;
+	trace_kvm_wait_lapic_expire(vcpu->vcpu_id, guest_tsc - tsc_deadline);
 
 	if (lapic_timer_advance_dynamic) {
-		adjust_lapic_timer_advance(vcpu, apic->lapic_timer.advance_expire_delta);
+		adjust_lapic_timer_advance(vcpu, guest_tsc - tsc_deadline);
 		/*
 		 * If the timer fired early, reread the TSC to account for the
 		 * overhead of the above adjustment to avoid waiting longer

arch/x86/kvm/lapic.h

···
 	u64 tscdeadline;
 	u64 expired_tscdeadline;
 	u32 timer_advance_ns;
-	s64 advance_expire_delta;
 	atomic_t pending;			/* accumulated triggered timers */
 	bool hv_timer_in_use;
 };

arch/x86/kvm/mmu.h

···
 	return (1ULL << (max_gpa_bits - PAGE_SHIFT)) - 1;
 }
 
+static inline u8 kvm_get_shadow_phys_bits(void)
+{
+	/*
+	 * boot_cpu_data.x86_phys_bits is reduced when MKTME or SME are detected
+	 * in CPU detection code, but the processor treats those reduced bits as
+	 * 'keyID' thus they are not reserved bits. Therefore KVM needs to look at
+	 * the physical address bits reported by CPUID.
+	 */
+	if (likely(boot_cpu_data.extended_cpuid_level >= 0x80000008))
+		return cpuid_eax(0x80000008) & 0xff;
+
+	/*
+	 * Quite weird to have VMX or SVM but not MAXPHYADDR; probably a VM with
+	 * custom CPUID.  Proceed with whatever the kernel found since these features
+	 * aren't virtualizable (SME/SEV also require CPUIDs higher than 0x80000008).
+	 */
+	return boot_cpu_data.x86_phys_bits;
+}
+
 void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 mmio_mask, u64 access_mask);
+void kvm_mmu_set_me_spte_mask(u64 me_value, u64 me_mask);
 void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only);
 
 void kvm_init_mmu(struct kvm_vcpu *vcpu);
···
 		return;
 
 	static_call(kvm_x86_load_mmu_pgd)(vcpu, root_hpa,
-					  vcpu->arch.mmu->shadow_root_level);
-}
-
-struct kvm_page_fault {
-	/* arguments to kvm_mmu_do_page_fault.  */
-	const gpa_t addr;
-	const u32 error_code;
-	const bool prefetch;
-
-	/* Derived from error_code.  */
-	const bool exec;
-	const bool write;
-	const bool present;
-	const bool rsvd;
-	const bool user;
-
-	/* Derived from mmu and global state.  */
-	const bool is_tdp;
-	const bool nx_huge_page_workaround_enabled;
-
-	/*
-	 * Whether a >4KB mapping can be created or is forbidden due to NX
-	 * hugepages.
-	 */
-	bool huge_page_disallowed;
-
-	/*
-	 * Maximum page size that can be created for this fault; input to
-	 * FNAME(fetch), __direct_map and kvm_tdp_mmu_map.
-	 */
-	u8 max_level;
-
-	/*
-	 * Page size that can be created based on the max_level and the
-	 * page size used by the host mapping.
-	 */
-	u8 req_level;
-
-	/*
-	 * Page size that will be created based on the req_level and
-	 * huge_page_disallowed.
-	 */
-	u8 goal_level;
-
-	/* Shifted addr, or result of guest page table walk if addr is a gva.  */
-	gfn_t gfn;
-
-	/* The memslot containing gfn. May be NULL. */
-	struct kvm_memory_slot *slot;
-
-	/* Outputs of kvm_faultin_pfn.  */
-	kvm_pfn_t pfn;
-	hva_t hva;
-	bool map_writable;
-};
-
-int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
-
-extern int nx_huge_pages;
-static inline bool is_nx_huge_page_enabled(void)
-{
-	return READ_ONCE(nx_huge_pages);
-}
-
-static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
-					u32 err, bool prefetch)
-{
-	struct kvm_page_fault fault = {
-		.addr = cr2_or_gpa,
-		.error_code = err,
-		.exec = err & PFERR_FETCH_MASK,
-		.write = err & PFERR_WRITE_MASK,
-		.present = err & PFERR_PRESENT_MASK,
-		.rsvd = err & PFERR_RSVD_MASK,
-		.user = err & PFERR_USER_MASK,
-		.prefetch = prefetch,
-		.is_tdp = likely(vcpu->arch.mmu->page_fault == kvm_tdp_page_fault),
-		.nx_huge_page_workaround_enabled = is_nx_huge_page_enabled(),
-
-		.max_level = KVM_MAX_HUGEPAGE_LEVEL,
-		.req_level = PG_LEVEL_4K,
-		.goal_level = PG_LEVEL_4K,
-	};
-#ifdef CONFIG_RETPOLINE
-	if (fault.is_tdp)
-		return kvm_tdp_page_fault(vcpu, &fault);
-#endif
-	return vcpu->arch.mmu->page_fault(vcpu, &fault);
+					  vcpu->arch.mmu->root_role.level);
 }
 
 /*

arch/x86/kvm/mmu/mmu.c

···
 
 /*
  * Yes, lot's of underscores.  They're a hint that you probably shouldn't be
- * reading from the role_regs.  Once the mmu_role is constructed, it becomes
+ * reading from the role_regs.  Once the root_role is constructed, it becomes
  * the single source of truth for the MMU's state.
  */
 #define BUILD_MMU_ROLE_REGS_ACCESSOR(reg, name, flag)			\
-static inline bool __maybe_unused ____is_##reg##_##name(struct kvm_mmu_role_regs *regs)\
+static inline bool __maybe_unused					\
+____is_##reg##_##name(const struct kvm_mmu_role_regs *regs)		\
 {									\
 	return !!(regs->reg & flag);					\
 }
···
 #define BUILD_MMU_ROLE_ACCESSOR(base_or_ext, reg, name)		\
 static inline bool __maybe_unused is_##reg##_##name(struct kvm_mmu *mmu)	\
 {								\
-	return !!(mmu->mmu_role. base_or_ext . reg##_##name);	\
+	return !!(mmu->cpu_role. base_or_ext . reg##_##name);	\
 }
-BUILD_MMU_ROLE_ACCESSOR(ext,  cr0, pg);
 BUILD_MMU_ROLE_ACCESSOR(base, cr0, wp);
 BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, pse);
-BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, pae);
 BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, smep);
 BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, smap);
 BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, pke);
 BUILD_MMU_ROLE_ACCESSOR(ext,  cr4, la57);
 BUILD_MMU_ROLE_ACCESSOR(base, efer, nx);
+BUILD_MMU_ROLE_ACCESSOR(ext,  efer, lma);
+
+static inline bool is_cr0_pg(struct kvm_mmu *mmu)
+{
+        return mmu->cpu_role.base.level > 0;
+}
+
+static inline bool is_cr4_pae(struct kvm_mmu *mmu)
+{
+        return !mmu->cpu_role.base.has_4_byte_gpte;
+}
 
 static struct kvm_mmu_role_regs vcpu_to_role_regs(struct kvm_vcpu *vcpu)
 {
···
 	};
 
 	return regs;
-}
-
-static int role_regs_to_root_level(struct kvm_mmu_role_regs *regs)
-{
-	if (!____is_cr0_pg(regs))
-		return 0;
-	else if (____is_efer_lma(regs))
-		return ____is_cr4_la57(regs) ? PT64_ROOT_5LEVEL :
-					       PT64_ROOT_4LEVEL;
-	else if (____is_cr4_pae(regs))
-		return PT32E_ROOT_LEVEL;
-	else
-		return PT32_ROOT_LEVEL;
 }
 
 static inline bool kvm_available_flush_tlb_with_range(void)
···
 
 static gfn_t kvm_mmu_page_get_gfn(struct kvm_mmu_page *sp, int index)
 {
+	if (sp->role.passthrough)
+		return sp->gfn;
+
 	if (!sp->role.direct)
 		return sp->gfns[index];
 
···
 
 static void kvm_mmu_page_set_gfn(struct kvm_mmu_page *sp, int index, gfn_t gfn)
 {
+	if (sp->role.passthrough) {
+		WARN_ON_ONCE(gfn != sp->gfn);
+		return;
+	}
+
 	if (!sp->role.direct) {
 		sp->gfns[index] = gfn;
 		return;
···
 
 static void slot_rmap_walk_next(struct slot_rmap_walk_iterator *iterator)
 {
-	if (++iterator->rmap <= iterator->end_rmap) {
+	while (++iterator->rmap <= iterator->end_rmap) {
 		iterator->gfn += (1UL << KVM_HPAGE_GFN_SHIFT(iterator->level));
-		return;
+
+		if (iterator->rmap->val)
+			return;
 	}
 
 	if (++iterator->level > iterator->end_level) {
···
 static void kvm_mmu_commit_zap_page(struct kvm *kvm,
 				    struct list_head *invalid_list);
 
+static bool sp_has_gptes(struct kvm_mmu_page *sp)
+{
+	if (sp->role.direct)
+		return false;
+
+	if (sp->role.passthrough)
+		return false;
+
+	return true;
+}
+
 #define for_each_valid_sp(_kvm, _sp, _list)				\
 	hlist_for_each_entry(_sp, _list, hash_link)			\
 		if (is_obsolete_sp((_kvm), (_sp))) {			\
 		} else
 
-#define for_each_gfn_indirect_valid_sp(_kvm, _sp, _gfn)			\
+#define for_each_gfn_valid_sp_with_gptes(_kvm, _sp, _gfn)		\
 	for_each_valid_sp(_kvm, _sp,					\
 	  &(_kvm)->arch.mmu_page_hash[kvm_page_table_hashfn(_gfn)])	\
-		if ((_sp)->gfn != (_gfn) || (_sp)->role.direct) {} else
+		if ((_sp)->gfn != (_gfn) || !sp_has_gptes(_sp)) {} else
 
-static bool kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
+static int kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 			 struct list_head *invalid_list)
 {
 	int ret = vcpu->arch.mmu->sync_page(vcpu, sp);
 
-	if (ret < 0) {
+	if (ret < 0)
 		kvm_mmu_prepare_zap_page(vcpu->kvm, sp, invalid_list);
-		return false;
-	}
-
-	return !!ret;
+	return ret;
 }
 
 static bool kvm_mmu_remote_flush_or_zap(struct kvm *kvm,
···
 
 		for_each_sp(pages, sp, parents, i) {
 			kvm_unlink_unsync_page(vcpu->kvm, sp);
-			flush |= kvm_sync_page(vcpu, sp, &invalid_list);
+			flush |= kvm_sync_page(vcpu, sp, &invalid_list) > 0;
 			mmu_pages_clear_parents(&parents);
 		}
 		if (need_resched() || rwlock_needbreak(&vcpu->kvm->mmu_lock)) {
···
 					     int direct,
 					     unsigned int access)
 {
-	bool direct_mmu = vcpu->arch.mmu->direct_map;
+	bool direct_mmu = vcpu->arch.mmu->root_role.direct;
 	union kvm_mmu_page_role role;
 	struct hlist_head *sp_list;
 	unsigned quadrant;
 	struct kvm_mmu_page *sp;
+	int ret;
 	int collisions = 0;
 	LIST_HEAD(invalid_list);
 
-	role = vcpu->arch.mmu->mmu_role.base;
+	role = vcpu->arch.mmu->root_role;
 	role.level = level;
 	role.direct = direct;
 	role.access = access;
···
 		quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1;
 		role.quadrant = quadrant;
 	}
+	if (level <= vcpu->arch.mmu->cpu_role.base.level)
+		role.passthrough = 0;
 
 	sp_list = &vcpu->kvm->arch.mmu_page_hash[kvm_page_table_hashfn(gfn)];
 	for_each_valid_sp(vcpu->kvm, sp, sp_list) {
···
 			 * If the sync fails, the page is zapped.  If so, break
 			 * in order to rebuild it.
 			 */
-			if (!kvm_sync_page(vcpu, sp, &invalid_list))
+			ret = kvm_sync_page(vcpu, sp, &invalid_list);
+			if (ret < 0)
 				break;
 
 			WARN_ON(!list_empty(&invalid_list));
-			kvm_flush_remote_tlbs(vcpu->kvm);
+			if (ret > 0)
+				kvm_flush_remote_tlbs(vcpu->kvm);
 		}
 
 		__clear_sp_write_flooding_count(sp);
···
 	sp->gfn = gfn;
 	sp->role = role;
 	hlist_add_head(&sp->hash_link, sp_list);
-	if (!direct) {
+	if (sp_has_gptes(sp)) {
 		account_shadowed(vcpu->kvm, sp);
 		if (level == PG_LEVEL_4K && kvm_vcpu_write_protect_gfn(vcpu, gfn))
 			kvm_flush_remote_tlbs_with_address(vcpu->kvm, gfn, 1);
···
 {
 	iterator->addr = addr;
 	iterator->shadow_addr = root;
-	iterator->level = vcpu->arch.mmu->shadow_root_level;
+	iterator->level = vcpu->arch.mmu->root_role.level;
 
 	if (iterator->level >= PT64_ROOT_4LEVEL &&
-	    vcpu->arch.mmu->root_level < PT64_ROOT_4LEVEL &&
-	    !vcpu->arch.mmu->direct_map)
+	    vcpu->arch.mmu->cpu_role.base.level < PT64_ROOT_4LEVEL &&
+	    !vcpu->arch.mmu->root_role.direct)
 		iterator->level = PT32E_ROOT_LEVEL;
 
 	if (iterator->level == PT32E_ROOT_LEVEL) {
···
 	/* Zapping children means active_mmu_pages has become unstable. */
 	list_unstable = *nr_zapped;
 
-	if (!sp->role.invalid && !sp->role.direct)
+	if (!sp->role.invalid && sp_has_gptes(sp))
 		unaccount_shadowed(kvm, sp);
 
 	if (sp->unsync)
···
 	pgprintk("%s: looking for gfn %llx\n", __func__, gfn);
 	r = 0;
 	write_lock(&kvm->mmu_lock);
-	for_each_gfn_indirect_valid_sp(kvm, sp, gfn) {
+	for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) {
 		pgprintk("%s: gfn %llx role %x\n", __func__, gfn,
 			 sp->role.word);
 		r = 1;
···
 	gpa_t gpa;
 	int r;
 
-	if (vcpu->arch.mmu->direct_map)
+	if (vcpu->arch.mmu->root_role.direct)
 		return 0;
 
 	gpa = kvm_mmu_gva_to_gpa_read(vcpu, gva, NULL);
···
 	 * that case, KVM must complete emulation of the guest TLB flush before
 	 * allowing shadow pages to become unsync (writable by the guest).
 	 */
-	for_each_gfn_indirect_valid_sp(kvm, sp, gfn) {
+	for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) {
 		if (!can_unsync)
 			return -EPERM;
 
···
 		 *sptep, write_fault, gfn);
 
 	if (unlikely(is_noslot_pfn(pfn))) {
+		vcpu->stat.pf_mmio_spte_created++;
 		mark_mmio_spte(vcpu, sptep, gfn, pte_access);
 		return RET_PF_EMULATE;
 	}
···
 		return ret;
 
 	direct_pte_prefetch(vcpu, it.sptep);
-	++vcpu->stat.pf_fixed;
 	return ret;
 }
 
···
 	return -EFAULT;
 }
 
-static bool handle_abnormal_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
-				unsigned int access, int *ret_val)
+static int handle_abnormal_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
+			       unsigned int access)
 {
 	/* The pfn is invalid, report the error! */
-	if (unlikely(is_error_pfn(fault->pfn))) {
-		*ret_val = kvm_handle_bad_page(vcpu, fault->gfn, fault->pfn);
-		return true;
-	}
+	if (unlikely(is_error_pfn(fault->pfn)))
+		return kvm_handle_bad_page(vcpu, fault->gfn, fault->pfn);
 
 	if (unlikely(!fault->slot)) {
 		gva_t gva = fault->is_tdp ? 0 : fault->addr;
···
 		 * and only if L1's MAXPHYADDR is inaccurate with respect to
 		 * the hardware's).
 		 */
-		if (unlikely(!shadow_mmio_value) ||
-		    unlikely(fault->gfn > kvm_mmu_max_gfn())) {
-			*ret_val = RET_PF_EMULATE;
-			return true;
-		}
+		if (unlikely(!enable_mmio_caching) ||
+		    unlikely(fault->gfn > kvm_mmu_max_gfn()))
+			return RET_PF_EMULATE;
 	}
 
-	return false;
+	return RET_PF_CONTINUE;
 }
 
 static bool page_fault_can_be_fast(struct kvm_page_fault *fault)
 {
 	/*
-	 * Do not fix the mmio spte with invalid generation number which
-	 * need to be updated by slow page fault path.
+	 * Page faults with reserved bits set, i.e. faults on MMIO SPTEs, only
+	 * reach the common page fault handler if the SPTE has an invalid MMIO
+	 * generation number.  Refreshing the MMIO generation needs to go down
+	 * the slow path.  Note, EPT Misconfigs do NOT set the PRESENT flag!
 	 */
 	if (fault->rsvd)
 		return false;
 
-	/* See if the page fault is due to an NX violation */
-	if (unlikely(fault->exec && fault->present))
-		return false;
-
 	/*
 	 * #PF can be fast if:
-	 * 1. The shadow page table entry is not present, which could mean that
-	 *    the fault is potentially caused by access tracking (if enabled).
-	 * 2. The shadow page table entry is present and the fault
-	 *    is caused by write-protect, that means we just need change the W
-	 *    bit of the spte which can be done out of mmu-lock.
 	 *
-	 * However, if access tracking is disabled we know that a non-present
-	 * page must be a genuine page fault where we have to create a new SPTE.
-	 * So, if access tracking is disabled, we return true only for write
-	 * accesses to a present page.
+	 * 1. The shadow page table entry is not present and A/D bits are
+	 *    disabled _by KVM_, which could mean that the fault is potentially
+	 *    caused by access tracking (if enabled).  If A/D bits are enabled
+	 *    by KVM, but disabled by L1 for L2, KVM is forced to disable A/D
+	 *    bits for L2 and employ access tracking, but the fast page fault
+	 *    mechanism only supports direct MMUs.
+	 * 2. The shadow page table entry is present, the access is a write,
+	 *    and no reserved bits are set (MMIO SPTEs cannot be "fixed"), i.e.
+	 *    the fault was caused by a write-protection violation.  If the
+	 *    SPTE is MMU-writable (determined later), the fault can be fixed
+	 *    by setting the Writable bit, which can be done out of mmu_lock.
 	 */
+	if (!fault->present)
+		return !kvm_ad_enabled();
 
-	return shadow_acc_track_mask != 0 || (fault->write && fault->present);
+	/*
+	 * Note, instruction fetches and writes are mutually exclusive, ignore
+	 * the "exec" flag.
+	 */
+	return fault->write;
 }
 
 /*
···
 
 		new_spte = spte;
 
-		if (is_access_track_spte(spte))
+		/*
+		 * KVM only supports fixing page faults outside of MMU lock for
+		 * direct MMUs, nested MMUs are always indirect, and KVM always
+		 * uses A/D bits for non-nested MMUs.  Thus, if A/D bits are
+		 * enabled, the SPTE can't be an access-tracked SPTE.
+		 */
+		if (unlikely(!kvm_ad_enabled()) && is_access_track_spte(spte))
 			new_spte = restore_acc_track_spte(new_spte);
 
 		/*
-		 * Currently, to simplify the code, write-protection can
-		 * be removed in the fast path only if the SPTE was
-		 * write-protected for dirty-logging or access tracking.
+		 * To keep things simple, only SPTEs that are MMU-writable can
+		 * be made fully writable outside of mmu_lock, e.g. only SPTEs
+		 * that were write-protected for dirty-logging or access
+		 * tracking are handled here.  Don't bother checking if the
+		 * SPTE is writable to prioritize running with A/D bits enabled.
+		 * The is_access_allowed() check above handles the common case
+		 * of the fault being spurious, and the SPTE is known to be
+		 * shadow-present, i.e. except for access tracking restoration
+		 * making the new SPTE writable, the check is wasteful.
 		 */
 		if (fault->write && is_mmu_writable_spte(spte)) {
 			new_spte |= PT_WRITABLE_MASK;
···
 
 	trace_fast_page_fault(vcpu, fault, sptep, spte, ret);
 	walk_shadow_page_lockless_end(vcpu);
+
+	if (ret != RET_PF_INVALID)
+		vcpu->stat.pf_fast++;
 
 	return ret;
 }
···
 	 * This should not be called while L2 is active, L2 can't invalidate
 	 * _only_ its own roots, e.g. INVVPID unconditionally exits.
 	 */
-	WARN_ON_ONCE(mmu->mmu_role.base.guest_mode);
+	WARN_ON_ONCE(mmu->root_role.guest_mode);
 
 	for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++) {
 		root_hpa = mmu->prev_roots[i].hpa;
···
 static int mmu_alloc_direct_roots(struct kvm_vcpu *vcpu)
 {
 	struct kvm_mmu *mmu = vcpu->arch.mmu;
-	u8 shadow_root_level = mmu->shadow_root_level;
+	u8 shadow_root_level = mmu->root_role.level;
 	hpa_t root;
 	unsigned i;
 	int r;
···
 	 * On SVM, reading PDPTRs might access guest memory, which might fault
 	 * and thus might sleep.  Grab the PDPTRs before acquiring mmu_lock.
 	 */
-	if (mmu->root_level == PT32E_ROOT_LEVEL) {
+	if (mmu->cpu_role.base.level == PT32E_ROOT_LEVEL) {
 		for (i = 0; i < 4; ++i) {
 			pdptrs[i] = mmu->get_pdptr(vcpu, i);
 			if (!(pdptrs[i] & PT_PRESENT_MASK))
···
 	 * Do we shadow a long mode page table? If so we need to
 	 * write-protect the guests page table root.
 	 */
-	if (mmu->root_level >= PT64_ROOT_4LEVEL) {
+	if (mmu->cpu_role.base.level >= PT64_ROOT_4LEVEL) {
 		root = mmu_alloc_root(vcpu, root_gfn, 0,
-				      mmu->shadow_root_level, false);
+				      mmu->root_role.level, false);
 		mmu->root.hpa = root;
 		goto set_root_pgd;
 	}
···
 	 * or a PAE 3-level page table. In either case we need to be aware that
 	 * the shadow page table may be a PAE or a long mode page table.
 	 */
-	pm_mask = PT_PRESENT_MASK | shadow_me_mask;
-	if (mmu->shadow_root_level >= PT64_ROOT_4LEVEL) {
+	pm_mask = PT_PRESENT_MASK | shadow_me_value;
+	if (mmu->root_role.level >= PT64_ROOT_4LEVEL) {
 		pm_mask |= PT_ACCESSED_MASK | PT_WRITABLE_MASK | PT_USER_MASK;
 
 		if (WARN_ON_ONCE(!mmu->pml4_root)) {
···
 		}
 		mmu->pml4_root[0] = __pa(mmu->pae_root) | pm_mask;
 
-		if (mmu->shadow_root_level == PT64_ROOT_5LEVEL) {
+		if (mmu->root_role.level == PT64_ROOT_5LEVEL) {
 			if (WARN_ON_ONCE(!mmu->pml5_root)) {
 				r = -EIO;
 				goto out_unlock;
···
 	for (i = 0; i < 4; ++i) {
 		WARN_ON_ONCE(IS_VALID_PAE_ROOT(mmu->pae_root[i]));
 
-		if (mmu->root_level == PT32E_ROOT_LEVEL) {
+		if (mmu->cpu_role.base.level == PT32E_ROOT_LEVEL) {
 			if (!(pdptrs[i] & PT_PRESENT_MASK)) {
 				mmu->pae_root[i] = INVALID_PAE_ROOT;
 				continue;
···
 		mmu->pae_root[i] = root | pm_mask;
 	}
 
-	if (mmu->shadow_root_level == PT64_ROOT_5LEVEL)
+	if (mmu->root_role.level == PT64_ROOT_5LEVEL)
 		mmu->root.hpa = __pa(mmu->pml5_root);
-	else if (mmu->shadow_root_level == PT64_ROOT_4LEVEL)
+	else if (mmu->root_role.level == PT64_ROOT_4LEVEL)
 		mmu->root.hpa = __pa(mmu->pml4_root);
 	else
 		mmu->root.hpa = __pa(mmu->pae_root);
···
 static int mmu_alloc_special_roots(struct kvm_vcpu *vcpu)
 {
 	struct kvm_mmu *mmu = vcpu->arch.mmu;
-	bool need_pml5 = mmu->shadow_root_level > PT64_ROOT_4LEVEL;
+	bool need_pml5 = mmu->root_role.level > PT64_ROOT_4LEVEL;
 	u64 *pml5_root = NULL;
 	u64 *pml4_root = NULL;
 	u64 *pae_root;
···
 	 * equivalent level in the guest's NPT to shadow.  Allocate the tables
 	 * on demand, as running a 32-bit L1 VMM on 64-bit KVM is very rare.
 	 */
-	if (mmu->direct_map || mmu->root_level >= PT64_ROOT_4LEVEL ||
-	    mmu->shadow_root_level < PT64_ROOT_4LEVEL)
+	if (mmu->root_role.direct ||
+	    mmu->cpu_role.base.level >= PT64_ROOT_4LEVEL ||
+	    mmu->root_role.level < PT64_ROOT_4LEVEL)
 		return 0;
 
 	/*
···
 	int i;
 	struct kvm_mmu_page *sp;
 
-	if (vcpu->arch.mmu->direct_map)
+	if (vcpu->arch.mmu->root_role.direct)
 		return;
 
 	if (!VALID_PAGE(vcpu->arch.mmu->root.hpa))
···
 
 	vcpu_clear_mmio_info(vcpu, MMIO_GVA_ANY);
 
-	if (vcpu->arch.mmu->root_level >= PT64_ROOT_4LEVEL) {
+	if (vcpu->arch.mmu->cpu_role.base.level >= PT64_ROOT_4LEVEL) {
 		hpa_t root = vcpu->arch.mmu->root.hpa;
 		sp = to_shadow_page(root);
 
···
 
 	arch.token = alloc_apf_token(vcpu);
 	arch.gfn = gfn;
-	arch.direct_map = vcpu->arch.mmu->direct_map;
+	arch.direct_map = vcpu->arch.mmu->root_role.direct;
 	arch.cr3 = vcpu->arch.mmu->get_guest_pgd(vcpu);
 
 	return kvm_setup_async_pf(vcpu, cr2_or_gpa,
 				  kvm_vcpu_gfn_to_hva(vcpu, gfn), &arch);
 }
 
-static bool kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, int *r)
+void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
+{
+	int r;
+
+	if ((vcpu->arch.mmu->root_role.direct != work->arch.direct_map) ||
+	      work->wakeup_all)
+		return;
+
+	r = kvm_mmu_reload(vcpu);
+	if (unlikely(r))
+		return;
+
+	if (!vcpu->arch.mmu->root_role.direct &&
+	      work->arch.cr3 != vcpu->arch.mmu->get_guest_pgd(vcpu))
+		return;
+
+	kvm_mmu_do_page_fault(vcpu, work->cr2_or_gpa, 0, true);
+}
+
+static int kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
 	struct kvm_memory_slot *slot = fault->slot;
 	bool async;
···
 	 * be zapped before KVM inserts a new MMIO SPTE for the gfn.
 	 */
 	if (slot && (slot->flags & KVM_MEMSLOT_INVALID))
-		goto out_retry;
+		return RET_PF_RETRY;
 
 	if (!kvm_is_visible_memslot(slot)) {
 		/* Don't expose private memslots to L2. */
···
 			fault->slot = NULL;
 			fault->pfn = KVM_PFN_NOSLOT;
 			fault->map_writable = false;
-			return false;
+			return RET_PF_CONTINUE;
 		}
 		/*
 		 * If the APIC access page exists but is disabled, go directly
···
 		 * when the AVIC is re-enabled.
 		 */
 		if (slot && slot->id == APIC_ACCESS_PAGE_PRIVATE_MEMSLOT &&
-		    !kvm_apicv_activated(vcpu->kvm)) {
-			*r = RET_PF_EMULATE;
-			return true;
-		}
+		    !kvm_apicv_activated(vcpu->kvm))
+			return RET_PF_EMULATE;
 	}
 
 	async = false;
···
 					  fault->write, &fault->map_writable,
 					  &fault->hva);
 	if (!async)
-		return false; /* *pfn has correct page already */
+		return RET_PF_CONTINUE; /* *pfn has correct page already */
 
 	if (!fault->prefetch && kvm_can_do_async_pf(vcpu)) {
 		trace_kvm_try_async_get_page(fault->addr, fault->gfn);
 		if (kvm_find_async_pf_gfn(vcpu, fault->gfn)) {
 			trace_kvm_async_pf_doublefault(fault->addr, fault->gfn);
 			kvm_make_request(KVM_REQ_APF_HALT, vcpu);
-			goto out_retry;
-		} else if (kvm_arch_setup_async_pf(vcpu, fault->addr, fault->gfn))
-			goto out_retry;
+			return RET_PF_RETRY;
+		} else if (kvm_arch_setup_async_pf(vcpu, fault->addr, fault->gfn)) {
+			return RET_PF_RETRY;
+		}
 	}
 
 	fault->pfn = __gfn_to_pfn_memslot(slot, fault->gfn, false, NULL,
 					  fault->write, &fault->map_writable,
 					  &fault->hva);
-	return false;
-
-out_retry:
-	*r = RET_PF_RETRY;
-	return true;
+	return RET_PF_CONTINUE;
 }
 
 /*
···
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (kvm_faultin_pfn(vcpu, fault, &r))
+	r = kvm_faultin_pfn(vcpu, fault);
+	if (r != RET_PF_CONTINUE)
 		return r;
 
-	if (handle_abnormal_pfn(vcpu, fault, ACC_ALL, &r))
+	r = handle_abnormal_pfn(vcpu, fault, ACC_ALL);
+	if (r != RET_PF_CONTINUE)
 		return r;
 
 	r = RET_PF_RETRY;
···
 	context->gva_to_gpa = nonpaging_gva_to_gpa;
 	context->sync_page = nonpaging_sync_page;
 	context->invlpg = NULL;
-	context->direct_map = true;
 }
 
 static inline bool is_root_usable(struct kvm_mmu_root_info *root, gpa_t pgd,
···
 void kvm_mmu_new_pgd(struct kvm_vcpu *vcpu, gpa_t new_pgd)
 {
 	struct kvm_mmu *mmu = vcpu->arch.mmu;
-	union kvm_mmu_page_role new_role = mmu->mmu_role.base;
+	union kvm_mmu_page_role new_role = mmu->root_role;
 
 	if (!fast_pgd_switch(vcpu->kvm, mmu, new_pgd, new_role)) {
 		/* kvm_mmu_ensure_valid_pgd will set up a new root.  */
···
 			     guest_cpuid_has(vcpu, X86_FEATURE_GBPAGES);
 }
 
-static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu,
-				  struct kvm_mmu *context)
+static void reset_guest_rsvds_bits_mask(struct kvm_vcpu *vcpu,
+					struct kvm_mmu *context)
 {
 	__reset_rsvds_bits_mask(&context->guest_rsvd_check,
 				vcpu->arch.reserved_gpa_bits,
-				context->root_level, is_efer_nx(context),
+				context->cpu_role.base.level, is_efer_nx(context),
 				guest_can_use_gbpages(vcpu),
 				is_cr4_pse(context),
 				guest_cpuid_is_amd_or_hygon(vcpu));
···
 static void reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu,
 					struct kvm_mmu *context)
 {
-	/*
-	 * KVM uses NX when TDP is disabled to handle a variety of scenarios,
-	 * notably for huge SPTEs if iTLB multi-hit mitigation is enabled and
-	 * to generate correct permissions for CR0.WP=0/CR4.SMEP=1/EFER.NX=0.
-	 * The iTLB multi-hit workaround can be toggled at any time, so assume
-	 * NX can be used by any non-nested shadow MMU to avoid having to reset
-	 * MMU contexts.  Note, KVM forces EFER.NX=1 when TDP is disabled.
-	 */
-	bool uses_nx = is_efer_nx(context) || !tdp_enabled;
-
 	/* @amd adds a check on bit of SPTEs, which KVM shouldn't use anyways. */
 	bool is_amd = true;
 	/* KVM doesn't use 2-level page tables for the shadow MMU. */
···
 	struct rsvd_bits_validate *shadow_zero_check;
 	int i;
 
-	WARN_ON_ONCE(context->shadow_root_level < PT32E_ROOT_LEVEL);
+	WARN_ON_ONCE(context->root_role.level < PT32E_ROOT_LEVEL);
 
 	shadow_zero_check = &context->shadow_zero_check;
 	__reset_rsvds_bits_mask(shadow_zero_check, reserved_hpa_bits(),
-				context->shadow_root_level, uses_nx,
+				context->root_role.level,
+				context->root_role.efer_nx,
 				guest_can_use_gbpages(vcpu), is_pse, is_amd);
 
 	if (!shadow_me_mask)
 		return;
 
-	for (i = context->shadow_root_level; --i >= 0;) {
-		shadow_zero_check->rsvd_bits_mask[0][i] &= ~shadow_me_mask;
-		shadow_zero_check->rsvd_bits_mask[1][i] &= ~shadow_me_mask;
+	for (i = context->root_role.level; --i >= 0;) {
+		/*
+		 * So far shadow_me_value is a constant during KVM's life
+		 * time.  Bits in shadow_me_value are allowed to be set.
+		 * Bits in shadow_me_mask but not in shadow_me_value are
+		 * not allowed to be set.
+		 */
+		shadow_zero_check->rsvd_bits_mask[0][i] |= shadow_me_mask;
+		shadow_zero_check->rsvd_bits_mask[1][i] |= shadow_me_mask;
+		shadow_zero_check->rsvd_bits_mask[0][i] &= ~shadow_me_value;
+		shadow_zero_check->rsvd_bits_mask[1][i] &= ~shadow_me_value;
 	}
 
 }
···
 
 	if (boot_cpu_is_amd())
 		__reset_rsvds_bits_mask(shadow_zero_check, reserved_hpa_bits(),
-					context->shadow_root_level, false,
+					context->root_role.level, false,
 					boot_cpu_has(X86_FEATURE_GBPAGES),
 					false, true);
 	else
···
 	if (!shadow_me_mask)
 		return;
 
-	for (i = context->shadow_root_level; --i >= 0;) {
+	for (i = context->root_role.level; --i >= 0;) {
 		shadow_zero_check->rsvd_bits_mask[0][i] &= ~shadow_me_mask;
 		shadow_zero_check->rsvd_bits_mask[1][i] &= ~shadow_me_mask;
 	}
···
 	if (!is_cr0_pg(mmu))
 		return;
 
-	reset_rsvds_bits_mask(vcpu, mmu);
+	reset_guest_rsvds_bits_mask(vcpu, mmu);
 	update_permission_bitmask(mmu, false);
 	update_pkru_bitmask(mmu);
 }
···
 	context->gva_to_gpa = paging64_gva_to_gpa;
 	context->sync_page = paging64_sync_page;
 	context->invlpg = paging64_invlpg;
-	context->direct_map = false;
 }
 
 static void paging32_init_context(struct kvm_mmu *context)
···
 	context->gva_to_gpa = paging32_gva_to_gpa;
 	context->sync_page = paging32_sync_page;
 	context->invlpg = paging32_invlpg;
-	context->direct_map = false;
 }
 
-static union kvm_mmu_extended_role kvm_calc_mmu_role_ext(struct kvm_vcpu *vcpu,
-							 struct kvm_mmu_role_regs *regs)
+static union kvm_cpu_role
+kvm_calc_cpu_role(struct kvm_vcpu *vcpu, const struct kvm_mmu_role_regs *regs)
 {
-	union kvm_mmu_extended_role ext = {0};
-
-	if (____is_cr0_pg(regs)) {
-		ext.cr0_pg = 1;
-		ext.cr4_pae = ____is_cr4_pae(regs);
-		ext.cr4_smep = ____is_cr4_smep(regs);
-		ext.cr4_smap = ____is_cr4_smap(regs);
-		ext.cr4_pse = ____is_cr4_pse(regs);
-
-		/* PKEY and LA57 are active iff long mode is active. */
-		ext.cr4_pke = ____is_efer_lma(regs) && ____is_cr4_pke(regs);
-		ext.cr4_la57 = ____is_efer_lma(regs) && ____is_cr4_la57(regs);
-		ext.efer_lma = ____is_efer_lma(regs);
-	}
-
-	ext.valid = 1;
-
-	return ext;
-}
-
-static union kvm_mmu_role kvm_calc_mmu_role_common(struct kvm_vcpu *vcpu,
-						   struct kvm_mmu_role_regs *regs,
-						   bool base_only)
-{
-	union kvm_mmu_role role = {0};
+	union kvm_cpu_role role = {0};
 
 	role.base.access = ACC_ALL;
-	if (____is_cr0_pg(regs)) {
-		role.base.efer_nx = ____is_efer_nx(regs);
-		role.base.cr0_wp = ____is_cr0_wp(regs);
-	}
 	role.base.smm = is_smm(vcpu);
 	role.base.guest_mode = is_guest_mode(vcpu);
+	role.ext.valid = 1;
 
-	if (base_only)
+	if (!____is_cr0_pg(regs)) {
+		role.base.direct = 1;
 		return role;
+	}
 
-	role.ext = kvm_calc_mmu_role_ext(vcpu, regs);
+	role.base.efer_nx = ____is_efer_nx(regs);
+	role.base.cr0_wp = ____is_cr0_wp(regs);
+	role.base.smep_andnot_wp = ____is_cr4_smep(regs) && !____is_cr0_wp(regs);
+	role.base.smap_andnot_wp = ____is_cr4_smap(regs) && !____is_cr0_wp(regs);
+	role.base.has_4_byte_gpte = !____is_cr4_pae(regs);
 
+	if (____is_efer_lma(regs))
+		role.base.level = ____is_cr4_la57(regs) ? PT64_ROOT_5LEVEL
+							: PT64_ROOT_4LEVEL;
+	else if (____is_cr4_pae(regs))
+		role.base.level = PT32E_ROOT_LEVEL;
+	else
+		role.base.level = PT32_ROOT_LEVEL;
+
+	role.ext.cr4_smep = ____is_cr4_smep(regs);
+	role.ext.cr4_smap = ____is_cr4_smap(regs);
+	role.ext.cr4_pse = ____is_cr4_pse(regs);
+
+	/* PKEY and LA57 are active iff long mode is active. */
+	role.ext.cr4_pke = ____is_efer_lma(regs) && ____is_cr4_pke(regs);
+	role.ext.cr4_la57 = ____is_efer_lma(regs) && ____is_cr4_la57(regs);
+	role.ext.efer_lma = ____is_efer_lma(regs);
 	return role;
 }
 
···
 	return max_tdp_level;
 }
 
-static union kvm_mmu_role
+static union kvm_mmu_page_role
 kvm_calc_tdp_mmu_root_page_role(struct kvm_vcpu *vcpu,
-				struct kvm_mmu_role_regs *regs, bool base_only)
+				union kvm_cpu_role cpu_role)
 {
-	union kvm_mmu_role role = kvm_calc_mmu_role_common(vcpu, regs, base_only);
+	union kvm_mmu_page_role role = {0};
 
-	role.base.ad_disabled = (shadow_accessed_mask == 0);
-	role.base.level = kvm_mmu_get_tdp_level(vcpu);
-	role.base.direct = true;
-	role.base.has_4_byte_gpte = false;
+	role.access = ACC_ALL;
+	role.cr0_wp = true;
+	role.efer_nx = true;
+	role.smm = cpu_role.base.smm;
+	role.guest_mode = cpu_role.base.guest_mode;
+	role.ad_disabled = !kvm_ad_enabled();
+	role.level = kvm_mmu_get_tdp_level(vcpu);
+	role.direct = true;
+	role.has_4_byte_gpte = false;
 
 	return role;
 }
 
-static void init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
+static void init_kvm_tdp_mmu(struct kvm_vcpu *vcpu,
+			     union kvm_cpu_role cpu_role)
 {
 	struct kvm_mmu *context = &vcpu->arch.root_mmu;
-	struct kvm_mmu_role_regs regs = vcpu_to_role_regs(vcpu);
-	union kvm_mmu_role new_role =
-		kvm_calc_tdp_mmu_root_page_role(vcpu, &regs, false);
+	union kvm_mmu_page_role root_role = kvm_calc_tdp_mmu_root_page_role(vcpu, cpu_role);
 
-	if (new_role.as_u64 == context->mmu_role.as_u64)
+	if (cpu_role.as_u64 == context->cpu_role.as_u64 &&
+	    root_role.word == context->root_role.word)
 		return;
 
-	context->mmu_role.as_u64 = new_role.as_u64;
+	context->cpu_role.as_u64 = cpu_role.as_u64;
+	context->root_role.word = root_role.word;
 	context->page_fault = kvm_tdp_page_fault;
 	context->sync_page = nonpaging_sync_page;
 	context->invlpg = NULL;
-	context->shadow_root_level = kvm_mmu_get_tdp_level(vcpu);
-	context->direct_map = true;
 	context->get_guest_pgd = get_cr3;
 	context->get_pdptr = kvm_pdptr_read;
 	context->inject_page_fault = kvm_inject_page_fault;
-	context->root_level = role_regs_to_root_level(&regs);
 
 	if (!is_cr0_pg(context))
 		context->gva_to_gpa = nonpaging_gva_to_gpa;
···
 	reset_tdp_shadow_zero_bits_mask(context);
 }
 
-static union kvm_mmu_role
-kvm_calc_shadow_root_page_role_common(struct kvm_vcpu *vcpu,
-				      struct kvm_mmu_role_regs *regs, bool base_only)
-{
-	union kvm_mmu_role role = kvm_calc_mmu_role_common(vcpu, regs, base_only);
-
-	role.base.smep_andnot_wp = role.ext.cr4_smep && !____is_cr0_wp(regs);
-	role.base.smap_andnot_wp = role.ext.cr4_smap && !____is_cr0_wp(regs);
-	role.base.has_4_byte_gpte = ____is_cr0_pg(regs) && !____is_cr4_pae(regs);
-
-	return role;
-}
-
-static union kvm_mmu_role
-kvm_calc_shadow_mmu_root_page_role(struct kvm_vcpu *vcpu,
-				   struct kvm_mmu_role_regs *regs, bool base_only)
-{
-	union kvm_mmu_role role =
-		kvm_calc_shadow_root_page_role_common(vcpu, regs, base_only);
-
-	role.base.direct = !____is_cr0_pg(regs);
-
-	if (!____is_efer_lma(regs))
-		role.base.level = PT32E_ROOT_LEVEL;
-	else if (____is_cr4_la57(regs))
-		role.base.level = PT64_ROOT_5LEVEL;
-	else
-		role.base.level = PT64_ROOT_4LEVEL;
-
-	return role;
-}
-
 static void shadow_mmu_init_context(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
-				    struct kvm_mmu_role_regs *regs,
-				    union kvm_mmu_role new_role)
+				    union kvm_cpu_role cpu_role,
+				    union kvm_mmu_page_role root_role)
 {
-	if (new_role.as_u64 == context->mmu_role.as_u64)
+	if (cpu_role.as_u64 == context->cpu_role.as_u64 &&
+	    root_role.word == context->root_role.word)
 		return;
 
-	context->mmu_role.as_u64 = new_role.as_u64;
+	context->cpu_role.as_u64 = cpu_role.as_u64;
+	context->root_role.word = root_role.word;
 
 	if (!is_cr0_pg(context))
 		nonpaging_init_context(context);
···
 		paging64_init_context(context);
 	else
 		paging32_init_context(context);
-	context->root_level = role_regs_to_root_level(regs);
 
 	reset_guest_paging_metadata(vcpu, context);
-	context->shadow_root_level = new_role.base.level;
-
 	reset_shadow_zero_bits_mask(vcpu, context);
 }
 
 static void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu,
-				struct kvm_mmu_role_regs *regs)
+				union kvm_cpu_role cpu_role)
 {
 	struct kvm_mmu *context = &vcpu->arch.root_mmu;
-	union kvm_mmu_role new_role =
-		kvm_calc_shadow_mmu_root_page_role(vcpu, regs, false);
+	union kvm_mmu_page_role root_role;
 
-	shadow_mmu_init_context(vcpu, context, regs, new_role);
-}
+	root_role = cpu_role.base;
 
-static union kvm_mmu_role
-kvm_calc_shadow_npt_root_page_role(struct kvm_vcpu *vcpu,
-				   struct kvm_mmu_role_regs *regs)
-{
-	union kvm_mmu_role role =
-		kvm_calc_shadow_root_page_role_common(vcpu, regs, false);
+	/* KVM uses PAE paging whenever the guest isn't using 64-bit paging. */
+	root_role.level = max_t(u32, root_role.level, PT32E_ROOT_LEVEL);
 
-	role.base.direct = false;
-	role.base.level = kvm_mmu_get_tdp_level(vcpu);
+	/*
+	 * KVM forces EFER.NX=1 when TDP is disabled, reflect it in the MMU role.
+	 * KVM uses NX when TDP is disabled to handle a variety of scenarios,
+	 * notably for huge SPTEs if iTLB multi-hit mitigation is enabled and
+	 * to generate correct permissions for CR0.WP=0/CR4.SMEP=1/EFER.NX=0.
+	 * The iTLB multi-hit workaround can be toggled at any time, so assume
+	 * NX can be used by any non-nested shadow MMU to avoid having to reset
+	 * MMU contexts.
+	 */
+	root_role.efer_nx = true;
 
-	return role;
+	shadow_mmu_init_context(vcpu, context, cpu_role, root_role);
 }
 
 void kvm_init_shadow_npt_mmu(struct kvm_vcpu *vcpu, unsigned long cr0,
···
 		.cr4 = cr4 & ~X86_CR4_PKE,
 		.efer = efer,
 	};
-	union kvm_mmu_role new_role;
+	union kvm_cpu_role cpu_role = kvm_calc_cpu_role(vcpu, &regs);
+	union kvm_mmu_page_role root_role;
 
-	new_role = kvm_calc_shadow_npt_root_page_role(vcpu, &regs);
+	/* NPT requires CR0.PG=1. */
+	WARN_ON_ONCE(cpu_role.base.direct);
 
-	shadow_mmu_init_context(vcpu, context, &regs, new_role);
+	root_role = cpu_role.base;
+	root_role.level = kvm_mmu_get_tdp_level(vcpu);
+	if (root_role.level == PT64_ROOT_5LEVEL &&
+	    cpu_role.base.level == PT64_ROOT_4LEVEL)
+		root_role.passthrough = 1;
+
+	shadow_mmu_init_context(vcpu, context, cpu_role, root_role);
 	kvm_mmu_new_pgd(vcpu, nested_cr3);
 }
 EXPORT_SYMBOL_GPL(kvm_init_shadow_npt_mmu);
 
-static union kvm_mmu_role
+static union kvm_cpu_role
 kvm_calc_shadow_ept_root_page_role(struct kvm_vcpu *vcpu, bool accessed_dirty,
 				   bool execonly, u8 level)
 {
-	union kvm_mmu_role role = {0};
+	union kvm_cpu_role role = {0};
 
-	/* SMM flag is inherited from root_mmu */
-	role.base.smm = vcpu->arch.root_mmu.mmu_role.base.smm;
-
+	/*
+	 * KVM does not support SMM transfer monitors, and consequently does not
+	 * support the "entry to SMM" control either.  role.base.smm is always 0.
+	 */
+	WARN_ON_ONCE(is_smm(vcpu));
 	role.base.level = level;
 	role.base.has_4_byte_gpte = false;
 	role.base.direct = false;
···
 	role.base.guest_mode = true;
 	role.base.access = ACC_ALL;
 
-	/* EPT, and thus nested EPT, does not consume CR0, CR4, nor EFER. */
 	role.ext.word = 0;
 	role.ext.execonly = execonly;
 	role.ext.valid = 1;
···
 {
 	struct kvm_mmu *context = &vcpu->arch.guest_mmu;
 	u8 level = vmx_eptp_page_walk_level(new_eptp);
-	union kvm_mmu_role new_role =
+	union kvm_cpu_role new_mode =
 		kvm_calc_shadow_ept_root_page_role(vcpu, accessed_dirty,
 						   execonly, level);
 
-	if (new_role.as_u64 != context->mmu_role.as_u64) {
-		context->mmu_role.as_u64 = new_role.as_u64;
+	if (new_mode.as_u64 != context->cpu_role.as_u64) {
+		/* EPT, and thus nested EPT, does not consume CR0, CR4, nor EFER. */
+		context->cpu_role.as_u64 = new_mode.as_u64;
+		context->root_role.word = new_mode.base.word;
 
-		context->shadow_root_level = level;
-
-		context->ept_ad = accessed_dirty;
 		context->page_fault = ept_page_fault;
 		context->gva_to_gpa = ept_gva_to_gpa;
 		context->sync_page = ept_sync_page;
 		context->invlpg = ept_invlpg;
-		context->root_level = level;
-		context->direct_map = false;
+
 		update_permission_bitmask(context, true);
 		context->pkru_mask = 0;
 		reset_rsvds_bits_mask_ept(vcpu, context, execonly, huge_page_level);
···
 }
 EXPORT_SYMBOL_GPL(kvm_init_shadow_ept_mmu);
 
-static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
+static void init_kvm_softmmu(struct kvm_vcpu *vcpu,
+			     union kvm_cpu_role cpu_role)
 {
 	struct kvm_mmu *context = &vcpu->arch.root_mmu;
-	struct kvm_mmu_role_regs regs = vcpu_to_role_regs(vcpu);
 
-	kvm_init_shadow_mmu(vcpu, &regs);
+	kvm_init_shadow_mmu(vcpu, cpu_role);
 
 	context->get_guest_pgd     = get_cr3;
 	context->get_pdptr         = kvm_pdptr_read;
 	context->inject_page_fault = kvm_inject_page_fault;
 }
 
-static union kvm_mmu_role
-kvm_calc_nested_mmu_role(struct kvm_vcpu *vcpu, struct kvm_mmu_role_regs *regs)
+static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu,
+				union kvm_cpu_role new_mode)
 {
-	union kvm_mmu_role role;
-
-	role = kvm_calc_shadow_root_page_role_common(vcpu, regs, false);
-
-	/*
-	 * Nested MMUs are used only for walking L2's gva->gpa, they never have
-	 * shadow pages of their own and so "direct" has no meaning.   Set it
-	 * to "true" to try to detect bogus usage of the nested MMU.
-	 */
-	role.base.direct = true;
-	role.base.level = role_regs_to_root_level(regs);
-	return role;
-}
-
-static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
-{
-	struct kvm_mmu_role_regs regs = vcpu_to_role_regs(vcpu);
-	union kvm_mmu_role new_role = kvm_calc_nested_mmu_role(vcpu, &regs);
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
-	if (new_role.as_u64 == g_context->mmu_role.as_u64)
+	if (new_mode.as_u64 == g_context->cpu_role.as_u64)
 		return;
 
-	g_context->mmu_role.as_u64 = new_role.as_u64;
+	g_context->cpu_role.as_u64   = new_mode.as_u64;
 	g_context->get_guest_pgd     = get_cr3;
 	g_context->get_pdptr         = kvm_pdptr_read;
 	g_context->inject_page_fault = kvm_inject_page_fault;
-	g_context->root_level        = new_role.base.level;
 
 	/*
 	 * L2 page tables are never shadowed, so there is no need to sync
···
 
 void kvm_init_mmu(struct kvm_vcpu *vcpu)
 {
+	struct kvm_mmu_role_regs regs = vcpu_to_role_regs(vcpu);
+	union kvm_cpu_role cpu_role = kvm_calc_cpu_role(vcpu, &regs);
+
 	if (mmu_is_nested(vcpu))
-		init_kvm_nested_mmu(vcpu);
+		init_kvm_nested_mmu(vcpu, cpu_role);
 	else if (tdp_enabled)
-		init_kvm_tdp_mmu(vcpu);
+		init_kvm_tdp_mmu(vcpu, cpu_role);
 	else
-		init_kvm_softmmu(vcpu);
+		init_kvm_softmmu(vcpu, cpu_role);
 }
 EXPORT_SYMBOL_GPL(kvm_init_mmu);
 
···
 	 * problem is swept under the rug; KVM's CPUID API is horrific and
 	 * it's all but impossible to solve it without introducing a new API.
 	 */
-	vcpu->arch.root_mmu.mmu_role.ext.valid = 0;
-	vcpu->arch.guest_mmu.mmu_role.ext.valid = 0;
-	vcpu->arch.nested_mmu.mmu_role.ext.valid = 0;
+	vcpu->arch.root_mmu.root_role.word = 0;
+	vcpu->arch.guest_mmu.root_role.word = 0;
+	vcpu->arch.nested_mmu.root_role.word = 0;
+	vcpu->arch.root_mmu.cpu_role.ext.valid = 0;
+	vcpu->arch.guest_mmu.cpu_role.ext.valid = 0;
+	vcpu->arch.nested_mmu.cpu_role.ext.valid = 0;
 	kvm_mmu_reset_context(vcpu);
 
 	/*
···
 {
 	int r;
 
-	r = mmu_topup_memory_caches(vcpu, !vcpu->arch.mmu->direct_map);
+	r = mmu_topup_memory_caches(vcpu, !vcpu->arch.mmu->root_role.direct);
 	if (r)
 		goto out;
 	r = mmu_alloc_special_roots(vcpu);
 	if (r)
 		goto out;
-	if (vcpu->arch.mmu->direct_map)
+	if (vcpu->arch.mmu->root_role.direct)
 		r = mmu_alloc_direct_roots(vcpu);
 	else
 		r = mmu_alloc_shadow_roots(vcpu);
···
 
 	++vcpu->kvm->stat.mmu_pte_write;
 
-	for_each_gfn_indirect_valid_sp(vcpu->kvm, sp, gfn) {
+	for_each_gfn_valid_sp_with_gptes(vcpu->kvm, sp, gfn) {
 		if (detect_write_misaligned(sp, gpa, bytes) ||
 		      detect_write_flooding(sp)) {
 			kvm_mmu_prepare_zap_page(vcpu->kvm, sp, &invalid_list);
···
 	write_unlock(&vcpu->kvm->mmu_lock);
 }
 
-int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u64 error_code,
+int noinline kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u64 error_code,
 		       void *insn, int insn_len)
 {
 	int r, emulation_type = EMULTYPE_PF;
-	bool direct = vcpu->arch.mmu->direct_map;
+	bool direct = vcpu->arch.mmu->root_role.direct;
 
 	if (WARN_ON(!VALID_PAGE(vcpu->arch.mmu->root.hpa)))
 		return RET_PF_RETRY;
···
 	 * paging in both guests. If true, we simply unprotect the page
 	 * and resume the guest.
 	 */
-	if (vcpu->arch.mmu->direct_map &&
+	if (vcpu->arch.mmu->root_role.direct &&
 	    (error_code & PFERR_NESTED_GUEST_PAGE) == PFERR_NESTED_GUEST_PAGE) {
 		kvm_mmu_unprotect_page(vcpu->kvm, gpa_to_gfn(cr2_or_gpa));
 		return 1;
···
 	if (!tdp_enabled)
 		set_memory_decrypted((unsigned long)mmu->pae_root, 1);
 	else
-		WARN_ON_ONCE(shadow_me_mask);
+		WARN_ON_ONCE(shadow_me_value);
 
 	for (i = 0; i < 4; ++i)
 		mmu->pae_root[i] = INVALID_PAE_ROOT;
···
 	 */
 	BUILD_BUG_ON(sizeof(union kvm_mmu_page_role) != sizeof(u32));
 	BUILD_BUG_ON(sizeof(union kvm_mmu_extended_role) != sizeof(u32));
-	BUILD_BUG_ON(sizeof(union kvm_mmu_role) != sizeof(u64));
+	BUILD_BUG_ON(sizeof(union kvm_cpu_role) != sizeof(u64));
 
 	kvm_mmu_reset_all_pte_masks();
 

arch/x86/kvm/mmu/mmu_internal.h

···
 					u64 start_gfn, u64 pages);
 unsigned int pte_list_count(struct kvm_rmap_head *rmap_head);
 
+extern int nx_huge_pages;
+static inline bool is_nx_huge_page_enabled(void)
+{
+	return READ_ONCE(nx_huge_pages);
+}
+
+struct kvm_page_fault {
+	/* arguments to kvm_mmu_do_page_fault.  */
+	const gpa_t addr;
+	const u32 error_code;
+	const bool prefetch;
+
+	/* Derived from error_code.  */
+	const bool exec;
+	const bool write;
+	const bool present;
+	const bool rsvd;
+	const bool user;
+
+	/* Derived from mmu and global state.  */
+	const bool is_tdp;
+	const bool nx_huge_page_workaround_enabled;
+
+	/*
+	 * Whether a >4KB mapping can be created or is forbidden due to NX
+	 * hugepages.
+	 */
+	bool huge_page_disallowed;
+
+	/*
+	 * Maximum page size that can be created for this fault; input to
+	 * FNAME(fetch), __direct_map and kvm_tdp_mmu_map.
+	 */
+	u8 max_level;
+
+	/*
+	 * Page size that can be created based on the max_level and the
+	 * page size used by the host mapping.
+	 */
+	u8 req_level;
+
+	/*
+	 * Page size that will be created based on the req_level and
+	 * huge_page_disallowed.
+	 */
+	u8 goal_level;
+
+	/* Shifted addr, or result of guest page table walk if addr is a gva.  */
+	gfn_t gfn;
+
+	/* The memslot containing gfn. May be NULL. */
+	struct kvm_memory_slot *slot;
+
+	/* Outputs of kvm_faultin_pfn.  */
+	kvm_pfn_t pfn;
+	hva_t hva;
+	bool map_writable;
+};
+
+int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
+
 /*
- * Return values of handle_mmio_page_fault, mmu.page_fault, and fast_page_fault().
+ * Return values of handle_mmio_page_fault(), mmu.page_fault(), fast_page_fault(),
+ * and of course kvm_mmu_do_page_fault().
  *
+ * RET_PF_CONTINUE: So far, so good, keep handling the page fault.
  * RET_PF_RETRY: let CPU fault again on the address.
  * RET_PF_EMULATE: mmio page fault, emulate the instruction directly.
  * RET_PF_INVALID: the spte is invalid, let the real page fault path update it.
···
  *
  * Any names added to this enum should be exported to userspace for use in
  * tracepoints via TRACE_DEFINE_ENUM() in mmutrace.h
+ *
+ * Note, all values must be greater than or equal to zero so as not to encroach
+ * on -errno return values.  Somewhat arbitrarily use '0' for CONTINUE, which
+ * will allow for efficient machine code when checking for CONTINUE, e.g.
+ * "TEST %rax, %rax, JNZ", as all "stop!" values are non-zero.
  */
 enum {
-	RET_PF_RETRY = 0,
+	RET_PF_CONTINUE = 0,
+	RET_PF_RETRY,
 	RET_PF_EMULATE,
 	RET_PF_INVALID,
 	RET_PF_FIXED,
 	RET_PF_SPURIOUS,
 };
+
+static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
+					u32 err, bool prefetch)
+{
+	struct kvm_page_fault fault = {
+		.addr = cr2_or_gpa,
+		.error_code = err,
+		.exec = err & PFERR_FETCH_MASK,
+		.write = err & PFERR_WRITE_MASK,
+		.present = err & PFERR_PRESENT_MASK,
+		.rsvd = err & PFERR_RSVD_MASK,
+		.user = err & PFERR_USER_MASK,
+		.prefetch = prefetch,
+		.is_tdp = likely(vcpu->arch.mmu->page_fault == kvm_tdp_page_fault),
+		.nx_huge_page_workaround_enabled = is_nx_huge_page_enabled(),
+
+		.max_level = KVM_MAX_HUGEPAGE_LEVEL,
+		.req_level = PG_LEVEL_4K,
+		.goal_level = PG_LEVEL_4K,
+	};
+	int r;
+
+	/*
+	 * Async #PF "faults", a.k.a. prefetch faults, are not faults from the
+	 * guest perspective and have already been counted at the time of the
+	 * original fault.
+	 */
+	if (!prefetch)
+		vcpu->stat.pf_taken++;
+
+	if (IS_ENABLED(CONFIG_RETPOLINE) && fault.is_tdp)
+		r = kvm_tdp_page_fault(vcpu, &fault);
+	else
+		r = vcpu->arch.mmu->page_fault(vcpu, &fault);
+
+	/*
+	 * Similar to above, prefetch faults aren't truly spurious, and the
+	 * async #PF path doesn't do emulation.  Do count faults that are fixed
+	 * by the async #PF handler though, otherwise they'll never be counted.
+	 */
+	if (r == RET_PF_FIXED)
+		vcpu->stat.pf_fixed++;
+	else if (prefetch)
+		;
+	else if (r == RET_PF_EMULATE)
+		vcpu->stat.pf_emulate++;
+	else if (r == RET_PF_SPURIOUS)
+		vcpu->stat.pf_spurious++;
+	return r;
+}
 
 int kvm_mmu_max_mapping_level(struct kvm *kvm,
 			      const struct kvm_memory_slot *slot, gfn_t gfn,

arch/x86/kvm/mmu/mmutrace.h

···
 	{ PFERR_RSVD_MASK, "RSVD" },	\
 	{ PFERR_FETCH_MASK, "F" }
 
+TRACE_DEFINE_ENUM(RET_PF_CONTINUE);
 TRACE_DEFINE_ENUM(RET_PF_RETRY);
 TRACE_DEFINE_ENUM(RET_PF_EMULATE);
 TRACE_DEFINE_ENUM(RET_PF_INVALID);

arch/x86/kvm/mmu/paging_tmpl.h

···
 	#define PT_LEVEL_BITS PT64_LEVEL_BITS
 	#define PT_GUEST_DIRTY_SHIFT 9
 	#define PT_GUEST_ACCESSED_SHIFT 8
-	#define PT_HAVE_ACCESSED_DIRTY(mmu) ((mmu)->ept_ad)
+	#define PT_HAVE_ACCESSED_DIRTY(mmu) (!(mmu)->cpu_role.base.ad_disabled)
 	#ifdef CONFIG_X86_64
 	#define CMPXCHG "cmpxchgq"
 	#endif
···
 	       FNAME(is_bad_mt_xwr)(&mmu->guest_rsvd_check, gpte);
 }
 
-static int FNAME(cmpxchg_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
-			       pt_element_t __user *ptep_user, unsigned index,
-			       pt_element_t orig_pte, pt_element_t new_pte)
-{
-	signed char r;
-
-	if (!user_access_begin(ptep_user, sizeof(pt_element_t)))
-		return -EFAULT;
-
-#ifdef CMPXCHG
-	asm volatile("1:" LOCK_PREFIX CMPXCHG " %[new], %[ptr]\n"
-		     "setnz %b[r]\n"
-		     "2:"
-		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %k[r])
-		     : [ptr] "+m" (*ptep_user),
-		       [old] "+a" (orig_pte),
-		       [r] "=q" (r)
-		     : [new] "r" (new_pte)
-		     : "memory");
-#else
-	asm volatile("1:" LOCK_PREFIX "cmpxchg8b %[ptr]\n"
-		     "setnz %b[r]\n"
-		     "2:"
-		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %k[r])
-		     : [ptr] "+m" (*ptep_user),
-		       [old] "+A" (orig_pte),
-		       [r] "=q" (r)
-		     : [new_lo] "b" ((u32)new_pte),
-		       [new_hi] "c" ((u32)(new_pte >> 32))
-		     : "memory");
-#endif
-
-	user_access_end();
-	return r;
-}
-
 static bool FNAME(prefetch_invalid_gpte)(struct kvm_vcpu *vcpu,
 				  struct kvm_mmu_page *sp, u64 *spte,
 				  u64 gpte)
···
 	if (!FNAME(is_present_gpte)(gpte))
 		goto no_present;
 
-	/* if accessed bit is not supported prefetch non accessed gpte */
+	/* Prefetch only accessed entries (unless A/D bits are disabled). */
 	if (PT_HAVE_ACCESSED_DIRTY(vcpu->arch.mmu) &&
 	    !(gpte & PT_GUEST_ACCESSED_MASK))
 		goto no_present;
···
 		if (unlikely(!walker->pte_writable[level - 1]))
 			continue;
 
-		ret = FNAME(cmpxchg_gpte)(vcpu, mmu, ptep_user, index, orig_pte, pte);
+		ret = __try_cmpxchg_user(ptep_user, &orig_pte, pte, fault);
 		if (ret)
 			return ret;
 
···
 	 * is not reserved and does not indicate a large page at this level,
 	 * so clear PT_PAGE_SIZE_MASK in gpte if that is the case.
 	 */
-	gpte &= level - (PT32_ROOT_LEVEL + mmu->mmu_role.ext.cr4_pse);
+	gpte &= level - (PT32_ROOT_LEVEL + mmu->cpu_role.ext.cr4_pse);
 #endif
 	/*
 	 * PG_LEVEL_4K always terminates.  The RHS has bit 7 set
···
 
 	trace_kvm_mmu_pagetable_walk(addr, access);
 retry_walk:
-	walker->level = mmu->root_level;
+	walker->level = mmu->cpu_role.base.level;
 	pte           = mmu->get_guest_pgd(vcpu);
 	have_ad       = PT_HAVE_ACCESSED_DIRTY(mmu);
 
···
 	 * The other bits are set to 0.
 	 */
 	if (!(errcode & PFERR_RSVD_MASK)) {
-		vcpu->arch.exit_qualification &= 0x180;
+		vcpu->arch.exit_qualification &= (EPT_VIOLATION_GVA_IS_VALID |
+						  EPT_VIOLATION_GVA_TRANSLATED);
 		if (write_fault)
 			vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_WRITE;
 		if (user_fault)
 			vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_READ;
 		if (fetch_fault)
 			vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_INSTR;
-		vcpu->arch.exit_qualification |= (pte_access & 0x7) << 3;
+
+		/*
+		 * Note, pte_access holds the raw RWX bits from the EPTE, not
+		 * ACC_*_MASK flags!
+		 */
+		vcpu->arch.exit_qualification |= (pte_access & VMX_EPT_RWX_MASK) <<
+						 EPT_VIOLATION_RWX_SHIFT;
 	}
 #endif
 	walker->fault.address = addr;
···
 	WARN_ON_ONCE(gw->gfn != base_gfn);
 	direct_access = gw->pte_access;
 
-	top_level = vcpu->arch.mmu->root_level;
+	top_level = vcpu->arch.mmu->cpu_role.base.level;
 	if (top_level == PT32E_ROOT_LEVEL)
 		top_level = PT32_ROOT_LEVEL;
 	/*
···
 		return ret;
 
 	FNAME(pte_prefetch)(vcpu, gw, it.sptep);
-	++vcpu->stat.pf_fixed;
 	return ret;
 
 out_gpte_changed:
···
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (kvm_faultin_pfn(vcpu, fault, &r))
+	r = kvm_faultin_pfn(vcpu, fault);
+	if (r != RET_PF_CONTINUE)
 		return r;
 
-	if (handle_abnormal_pfn(vcpu, fault, walker.pte_access, &r))
+	r = handle_abnormal_pfn(vcpu, fault, walker.pte_access);
+	if (r != RET_PF_CONTINUE)
 		return r;
 
 	/*
···
  */
 static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
-	union kvm_mmu_page_role mmu_role = vcpu->arch.mmu->mmu_role.base;
+	union kvm_mmu_page_role root_role = vcpu->arch.mmu->root_role;
 	int i;
 	bool host_writable;
 	gpa_t first_pte_gpa;
···
 		.level = 0xf,
 		.access = 0x7,
 		.quadrant = 0x3,
+		.passthrough = 0x1,
 	};
 
 	/*
···
 	 * reserved bits checks will be wrong, etc...
 	 */
 	if (WARN_ON_ONCE(sp->role.direct ||
-			 (sp->role.word ^ mmu_role.word) & ~sync_role_ign.word))
+			 (sp->role.word ^ root_role.word) & ~sync_role_ign.word))
 		return -1;
 
 	first_pte_gpa = FNAME(get_level1_sp_gpa)(sp);

arch/x86/kvm/mmu/spte.c

···
 #include <asm/memtype.h>
 #include <asm/vmx.h>
 
-static bool __read_mostly enable_mmio_caching = true;
+bool __read_mostly enable_mmio_caching = true;
 module_param_named(mmio_caching, enable_mmio_caching, bool, 0444);
 
 u64 __read_mostly shadow_host_writable_mask;
···
 u64 __read_mostly shadow_mmio_mask;
 u64 __read_mostly shadow_mmio_access_mask;
 u64 __read_mostly shadow_present_mask;
+u64 __read_mostly shadow_me_value;
 u64 __read_mostly shadow_me_mask;
 u64 __read_mostly shadow_acc_track_mask;
 
···
 	else
 		pte_access &= ~ACC_WRITE_MASK;
 
-	if (!kvm_is_mmio_pfn(pfn))
-		spte |= shadow_me_mask;
+	if (shadow_me_value && !kvm_is_mmio_pfn(pfn))
+		spte |= shadow_me_value;
 
 	spte |= (u64)pfn << PAGE_SHIFT;
 
···
 	u64 spte = SPTE_MMU_PRESENT_MASK;
 
 	spte |= __pa(child_pt) | shadow_present_mask | PT_WRITABLE_MASK |
-		shadow_user_mask | shadow_x_mask | shadow_me_mask;
+		shadow_user_mask | shadow_x_mask | shadow_me_value;
 
 	if (ad_disabled)
 		spte |= SPTE_TDP_AD_DISABLED_MASK;
···
 	new_spte = mark_spte_for_access_track(new_spte);
 
 	return new_spte;
-}
-
-static u8 kvm_get_shadow_phys_bits(void)
-{
-	/*
-	 * boot_cpu_data.x86_phys_bits is reduced when MKTME or SME are detected
-	 * in CPU detection code, but the processor treats those reduced bits as
-	 * 'keyID' thus they are not reserved bits. Therefore KVM needs to look at
-	 * the physical address bits reported by CPUID.
-	 */
-	if (likely(boot_cpu_data.extended_cpuid_level >= 0x80000008))
-		return cpuid_eax(0x80000008) & 0xff;
-
-	/*
-	 * Quite weird to have VMX or SVM but not MAXPHYADDR; probably a VM with
-	 * custom CPUID.  Proceed with whatever the kernel found since these features
-	 * aren't virtualizable (SME/SEV also require CPUIDs higher than 0x80000008).
-	 */
-	return boot_cpu_data.x86_phys_bits;
 }
 
 u64 mark_spte_for_access_track(u64 spte)
···
 	    WARN_ON(mmio_value && (REMOVED_SPTE & mmio_mask) == mmio_value))
 		mmio_value = 0;
 
+	if (!mmio_value)
+		enable_mmio_caching = false;
+
 	shadow_mmio_value = mmio_value;
 	shadow_mmio_mask  = mmio_mask;
 	shadow_mmio_access_mask = access_mask;
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask);
+
+void kvm_mmu_set_me_spte_mask(u64 me_value, u64 me_mask)
+{
+	/* shadow_me_value must be a subset of shadow_me_mask */
+	if (WARN_ON(me_value & ~me_mask))
+		me_value = me_mask = 0;
+
+	shadow_me_value = me_value;
+	shadow_me_mask = me_mask;
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_set_me_spte_mask);
 
 void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only)
 {
···
 	shadow_x_mask		= VMX_EPT_EXECUTABLE_MASK;
 	shadow_present_mask	= has_exec_only ? 0ull : VMX_EPT_READABLE_MASK;
 	shadow_acc_track_mask	= VMX_EPT_RWX_MASK;
-	shadow_me_mask		= 0ull;
-
 	shadow_host_writable_mask = EPT_SPTE_HOST_WRITABLE;
 	shadow_mmu_writable_mask  = EPT_SPTE_MMU_WRITABLE;
 
···
 	shadow_x_mask		= 0;
 	shadow_present_mask	= PT_PRESENT_MASK;
 	shadow_acc_track_mask	= 0;
-	shadow_me_mask		= sme_me_mask;
+	shadow_me_mask		= 0;
+	shadow_me_value		= 0;
 
 	shadow_host_writable_mask = DEFAULT_SPTE_HOST_WRITABLE;
 	shadow_mmu_writable_mask  = DEFAULT_SPTE_MMU_WRITABLE;

arch/x86/kvm/mmu/spte.h

···
 
 #include "mmu_internal.h"
 
+extern bool __read_mostly enable_mmio_caching;
+
 /*
  * A MMU present SPTE is backed by actual memory and may or may not be present
  * in hardware.  E.g. MMIO SPTEs are not considered present.  Use bit 11, as it
···
 extern u64 __read_mostly shadow_mmio_mask;
 extern u64 __read_mostly shadow_mmio_access_mask;
 extern u64 __read_mostly shadow_present_mask;
+extern u64 __read_mostly shadow_me_value;
 extern u64 __read_mostly shadow_me_mask;
 
 /*
···
 static inline bool is_mmio_spte(u64 spte)
 {
 	return (spte & shadow_mmio_mask) == shadow_mmio_value &&
-	       likely(shadow_mmio_value);
+	       likely(enable_mmio_caching);
 }
 
 static inline bool is_shadow_present_pte(u64 pte)
 {
 	return !!(pte & SPTE_MMU_PRESENT_MASK);
+}
+
+/*
+ * Returns true if A/D bits are supported in hardware and are enabled by KVM.
+ * When enabled, KVM uses A/D bits for all non-nested MMUs.  Because L1 can
+ * disable A/D bits in EPTP12, SP and SPTE variants are needed to handle the
+ * scenario where KVM is using A/D bits for L1, but not L2.
+ */
+static inline bool kvm_ad_enabled(void)
+{
+	return !!shadow_accessed_mask;
 }
 
 static inline bool sp_ad_disabled(struct kvm_mmu_page *sp)

arch/x86/kvm/mmu/tdp_mmu.c

···
 
 hpa_t kvm_tdp_mmu_get_vcpu_root_hpa(struct kvm_vcpu *vcpu)
 {
-	union kvm_mmu_page_role role = vcpu->arch.mmu->mmu_role.base;
+	union kvm_mmu_page_role role = vcpu->arch.mmu->root_role;
 	struct kvm *kvm = vcpu->kvm;
 	struct kvm_mmu_page *root;
 
···
 
 	/* If a MMIO SPTE is installed, the MMIO will need to be emulated. */
 	if (unlikely(is_mmio_spte(new_spte))) {
+		vcpu->stat.pf_mmio_spte_created++;
 		trace_mark_mmio_spte(rcu_dereference(iter->sptep), iter->gfn,
 				     new_spte);
 		ret = RET_PF_EMULATE;
···
 		trace_kvm_mmu_set_spte(iter->level, iter->gfn,
 				       rcu_dereference(iter->sptep));
 	}
-
-	/*
-	 * Increase pf_fixed in both RET_PF_EMULATE and RET_PF_FIXED to be
-	 * consistent with legacy MMU behavior.
-	 */
-	if (ret != RET_PF_SPURIOUS)
-		vcpu->stat.pf_fixed++;
 
 	return ret;
 }
···
 			   struct kvm_mmu_page *sp, bool account_nx,
 			   bool shared)
 {
-	u64 spte = make_nonleaf_spte(sp->spt, !shadow_accessed_mask);
+	u64 spte = make_nonleaf_spte(sp->spt, !kvm_ad_enabled());
 	int ret = 0;
 
 	if (shared) {
···
 	gfn_t gfn = addr >> PAGE_SHIFT;
 	int leaf = -1;
 
-	*root_level = vcpu->arch.mmu->shadow_root_level;
+	*root_level = vcpu->arch.mmu->root_role.level;
 
 	tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) {
 		leaf = iter.level;

arch/x86/kvm/pmu.c

···
  *        * AMD:   [0 .. AMD64_NUM_COUNTERS-1] <=> gp counters
  */
 
+static struct kvm_pmu_ops kvm_pmu_ops __read_mostly;
+
+#define KVM_X86_PMU_OP(func)					     \
+	DEFINE_STATIC_CALL_NULL(kvm_x86_pmu_##func,			     \
+				*(((struct kvm_pmu_ops *)0)->func));
+#define KVM_X86_PMU_OP_OPTIONAL KVM_X86_PMU_OP
+#include <asm/kvm-x86-pmu-ops.h>
+
+void kvm_pmu_ops_update(const struct kvm_pmu_ops *pmu_ops)
+{
+	memcpy(&kvm_pmu_ops, pmu_ops, sizeof(kvm_pmu_ops));
+
+#define __KVM_X86_PMU_OP(func) \
+	static_call_update(kvm_x86_pmu_##func, kvm_pmu_ops.func);
+#define KVM_X86_PMU_OP(func) \
+	WARN_ON(!kvm_pmu_ops.func); __KVM_X86_PMU_OP(func)
+#define KVM_X86_PMU_OP_OPTIONAL __KVM_X86_PMU_OP
+#include <asm/kvm-x86-pmu-ops.h>
+#undef __KVM_X86_PMU_OP
+}
+
+static inline bool pmc_is_enabled(struct kvm_pmc *pmc)
+{
+	return static_call(kvm_x86_pmu_pmc_is_enabled)(pmc);
+}
+
 static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
 {
 	struct kvm_pmu *pmu = container_of(irq_work, struct kvm_pmu, irq_work);
···
 			  ARCH_PERFMON_EVENTSEL_CMASK |
 			  HSW_IN_TX |
 			  HSW_IN_TX_CHECKPOINTED))) {
-		config = kvm_x86_ops.pmu_ops->pmc_perf_hw_id(pmc);
+		config = static_call(kvm_x86_pmu_pmc_perf_hw_id)(pmc);
 		if (config != PERF_COUNT_HW_MAX)
 			type = PERF_TYPE_HARDWARE;
 	}
···
 
 	pmc->current_config = (u64)ctrl;
 	pmc_reprogram_counter(pmc, PERF_TYPE_HARDWARE,
-			      kvm_x86_ops.pmu_ops->pmc_perf_hw_id(pmc),
+			      static_call(kvm_x86_pmu_pmc_perf_hw_id)(pmc),
 			      !(en_field & 0x2), /* exclude user */
 			      !(en_field & 0x1), /* exclude kernel */
 			      pmi);
···
 
 void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx)
 {
-	struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, pmc_idx);
+	struct kvm_pmc *pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, pmc_idx);
 
 	if (!pmc)
 		return;
···
 	int bit;
 
 	for_each_set_bit(bit, pmu->reprogram_pmi, X86_PMC_IDX_MAX) {
-		struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, bit);
+		struct kvm_pmc *pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, bit);
 
 		if (unlikely(!pmc || !pmc->perf_event)) {
 			clear_bit(bit, pmu->reprogram_pmi);
···
 /* check if idx is a valid index to access PMU */
 bool kvm_pmu_is_valid_rdpmc_ecx(struct kvm_vcpu *vcpu, unsigned int idx)
 {
-	return kvm_x86_ops.pmu_ops->is_valid_rdpmc_ecx(vcpu, idx);
+	return static_call(kvm_x86_pmu_is_valid_rdpmc_ecx)(vcpu, idx);
 }
 
 bool is_vmware_backdoor_pmc(u32 pmc_idx)
···
 	if (is_vmware_backdoor_pmc(idx))
 		return kvm_pmu_rdpmc_vmware(vcpu, idx, data);
 
-	pmc = kvm_x86_ops.pmu_ops->rdpmc_ecx_to_pmc(vcpu, idx, &mask);
+	pmc = static_call(kvm_x86_pmu_rdpmc_ecx_to_pmc)(vcpu, idx, &mask);
 	if (!pmc)
 		return 1;
 
···
 void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu)
 {
 	if (lapic_in_kernel(vcpu)) {
-		if (kvm_x86_ops.pmu_ops->deliver_pmi)
-			kvm_x86_ops.pmu_ops->deliver_pmi(vcpu);
+		static_call_cond(kvm_x86_pmu_deliver_pmi)(vcpu);
 		kvm_apic_local_deliver(vcpu->arch.apic, APIC_LVTPC);
 	}
 }
 
 bool kvm_pmu_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 {
-	return kvm_x86_ops.pmu_ops->msr_idx_to_pmc(vcpu, msr) ||
-		kvm_x86_ops.pmu_ops->is_valid_msr(vcpu, msr);
+	return static_call(kvm_x86_pmu_msr_idx_to_pmc)(vcpu, msr) ||
+		static_call(kvm_x86_pmu_is_valid_msr)(vcpu, msr);
 }
 
 static void kvm_pmu_mark_pmc_in_use(struct kvm_vcpu *vcpu, u32 msr)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
-	struct kvm_pmc *pmc = kvm_x86_ops.pmu_ops->msr_idx_to_pmc(vcpu, msr);
+	struct kvm_pmc *pmc = static_call(kvm_x86_pmu_msr_idx_to_pmc)(vcpu, msr);
 
 	if (pmc)
 		__set_bit(pmc->idx, pmu->pmc_in_use);
···
 
 int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 {
-	return kvm_x86_ops.pmu_ops->get_msr(vcpu, msr_info);
+	return static_call(kvm_x86_pmu_get_msr)(vcpu, msr_info);
 }
 
 int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 {
 	kvm_pmu_mark_pmc_in_use(vcpu, msr_info->index);
-	return kvm_x86_ops.pmu_ops->set_msr(vcpu, msr_info);
+	return static_call(kvm_x86_pmu_set_msr)(vcpu, msr_info);
 }
 
 /* refresh PMU settings. This function generally is called when underlying
···
  */
 void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 {
-	kvm_x86_ops.pmu_ops->refresh(vcpu);
+	static_call(kvm_x86_pmu_refresh)(vcpu);
 }
 
 void kvm_pmu_reset(struct kvm_vcpu *vcpu)
···
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 
 	irq_work_sync(&pmu->irq_work);
-	kvm_x86_ops.pmu_ops->reset(vcpu);
+	static_call(kvm_x86_pmu_reset)(vcpu);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)
···
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 
 	memset(pmu, 0, sizeof(*pmu));
-	kvm_x86_ops.pmu_ops->init(vcpu);
+	static_call(kvm_x86_pmu_init)(vcpu);
 	init_irq_work(&pmu->irq_work, kvm_pmi_trigger_fn);
 	pmu->event_count = 0;
 	pmu->need_cleanup = false;
···
 		      pmu->pmc_in_use, X86_PMC_IDX_MAX);
 
 	for_each_set_bit(i, bitmask, X86_PMC_IDX_MAX) {
-		pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, i);
+		pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, i);
 
 		if (pmc && pmc->perf_event && !pmc_speculative_in_use(pmc))
 			pmc_stop_counter(pmc);
 	}
 
-	if (kvm_x86_ops.pmu_ops->cleanup)
-		kvm_x86_ops.pmu_ops->cleanup(vcpu);
+	static_call_cond(kvm_x86_pmu_cleanup)(vcpu);
 
 	bitmap_zero(pmu->pmc_in_use, X86_PMC_IDX_MAX);
 }
···
 	unsigned int config;
 
 	pmc->eventsel &= (ARCH_PERFMON_EVENTSEL_EVENT | ARCH_PERFMON_EVENTSEL_UMASK);
-	config = kvm_x86_ops.pmu_ops->pmc_perf_hw_id(pmc);
+	config = static_call(kvm_x86_pmu_pmc_perf_hw_id)(pmc);
 	pmc->eventsel = old_eventsel;
 	return config == perf_hw_id;
 }
···
 	int i;
 
 	for_each_set_bit(i, pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX) {
-		pmc = kvm_x86_ops.pmu_ops->pmc_idx_to_pmc(pmu, i);
+		pmc = static_call(kvm_x86_pmu_pmc_idx_to_pmc)(pmu, i);
 
 		if (!pmc || !pmc_is_enabled(pmc) || !pmc_speculative_in_use(pmc))
 			continue;

arch/x86/kvm/pmu.h

···
 	void (*cleanup)(struct kvm_vcpu *vcpu);
 };
 
+void kvm_pmu_ops_update(const struct kvm_pmu_ops *pmu_ops);
+
 static inline u64 pmc_bitmask(struct kvm_pmc *pmc)
 {
 	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
···
 static inline bool pmc_is_fixed(struct kvm_pmc *pmc)
 {
 	return pmc->type == KVM_PMC_FIXED;
-}
-
-static inline bool pmc_is_enabled(struct kvm_pmc *pmc)
-{
-	return kvm_x86_ops.pmu_ops->pmc_is_enabled(pmc);
 }
 
 static inline bool kvm_valid_perf_global_ctrl(struct kvm_pmu *pmu,

arch/x86/kvm/svm/avic.c

···
 	return err;
 }
 
-void avic_init_vmcb(struct vcpu_svm *svm)
+void avic_init_vmcb(struct vcpu_svm *svm, struct vmcb *vmcb)
 {
-	struct vmcb *vmcb = svm->vmcb;
 	struct kvm_svm *kvm_svm = to_kvm_svm(svm->vcpu.kvm);
 	phys_addr_t bpa = __sme_set(page_to_phys(svm->avic_backing_page));
 	phys_addr_t lpa = __sme_set(page_to_phys(kvm_svm->avic_logical_id_table_page));
···
 	put_cpu();
 }
 
-static void avic_kick_target_vcpus(struct kvm *kvm, struct kvm_lapic *source,
-				   u32 icrl, u32 icrh)
+/*
+ * A fast-path version of avic_kick_target_vcpus(), which attempts to match
+ * destination APIC ID to vCPU without looping through all vCPUs.
+ */
+static int avic_kick_target_vcpus_fast(struct kvm *kvm, struct kvm_lapic *source,
+				       u32 icrl, u32 icrh, u32 index)
 {
+	u32 dest, apic_id;
 	struct kvm_vcpu *vcpu;
+	int dest_mode = icrl & APIC_DEST_MASK;
+	int shorthand = icrl & APIC_SHORT_MASK;
+	struct kvm_svm *kvm_svm = to_kvm_svm(kvm);
+	u32 *avic_logical_id_table = page_address(kvm_svm->avic_logical_id_table_page);
+
+	if (shorthand != APIC_DEST_NOSHORT)
+		return -EINVAL;
+
+	/*
+	 * The AVIC incomplete IPI #vmexit info provides index into
+	 * the physical APIC ID table, which can be used to derive
+	 * guest physical APIC ID.
+	 */
+	if (dest_mode == APIC_DEST_PHYSICAL) {
+		apic_id = index;
+	} else {
+		if (!apic_x2apic_mode(source)) {
+			/* For xAPIC logical mode, the index is for logical APIC table. */
+			apic_id = avic_logical_id_table[index] & 0x1ff;
+		} else {
+			return -EINVAL;
+		}
+	}
+
+	/*
+	 * Assuming vcpu ID is the same as physical apic ID,
+	 * and use it to retrieve the target vCPU.
+	 */
+	vcpu = kvm_get_vcpu_by_id(kvm, apic_id);
+	if (!vcpu)
+		return -EINVAL;
+
+	if (apic_x2apic_mode(vcpu->arch.apic))
+		dest = icrh;
+	else
+		dest = GET_APIC_DEST_FIELD(icrh);
+
+	/*
+	 * Try matching the destination APIC ID with the vCPU.
+	 */
+	if (kvm_apic_match_dest(vcpu, source, shorthand, dest, dest_mode)) {
+		vcpu->arch.apic->irr_pending = true;
+		svm_complete_interrupt_delivery(vcpu,
+						icrl & APIC_MODE_MASK,
+						icrl & APIC_INT_LEVELTRIG,
+						icrl & APIC_VECTOR_MASK);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static void avic_kick_target_vcpus(struct kvm *kvm, struct kvm_lapic *source,
+				   u32 icrl, u32 icrh, u32 index)
+{
 	unsigned long i;
+	struct kvm_vcpu *vcpu;
+
+	if (!avic_kick_target_vcpus_fast(kvm, source, icrl, icrh, index))
+		return;
+
+	trace_kvm_avic_kick_vcpu_slowpath(icrh, icrl, index);
 
 	/*
 	 * Wake any target vCPUs that are blocking, i.e. waiting for a wake
···
 	u32 icrh = svm->vmcb->control.exit_info_1 >> 32;
 	u32 icrl = svm->vmcb->control.exit_info_1;
 	u32 id = svm->vmcb->control.exit_info_2 >> 32;
-	u32 index = svm->vmcb->control.exit_info_2 & 0xFF;
+	u32 index = svm->vmcb->control.exit_info_2 & 0x1FF;
 	struct kvm_lapic *apic = vcpu->arch.apic;
 
 	trace_kvm_avic_incomplete_ipi(vcpu->vcpu_id, icrh, icrl, id, index);
···
 		 * set the appropriate IRR bits on the valid target
 		 * vcpus. So, we just need to kick the appropriate vcpu.
 		 */
-		avic_kick_target_vcpus(vcpu->kvm, apic, icrl, icrh);
+		avic_kick_target_vcpus(vcpu->kvm, apic, icrl, icrh, index);
 		break;
 	case AVIC_IPI_FAILURE_INVALID_TARGET:
 		break;
···
 	}
 
 	return 1;
+}
+
+unsigned long avic_vcpu_get_apicv_inhibit_reasons(struct kvm_vcpu *vcpu)
+{
+	if (is_guest_mode(vcpu))
+		return APICV_INHIBIT_REASON_NESTED;
+	return 0;
 }
 
 static u32 *avic_get_logical_id_entry(struct kvm_vcpu *vcpu, u32 ldr, bool flat)

arch/x86/kvm/svm/nested.c

···
 				       struct x86_exception *fault)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb *vmcb = svm->vmcb;
 
-	if (svm->vmcb->control.exit_code != SVM_EXIT_NPF) {
+	if (vmcb->control.exit_code != SVM_EXIT_NPF) {
 		/*
 		 * TODO: track the cause of the nested page fault, and
 		 * correctly fill in the high bits of exit_info_1.
 		 */
-		svm->vmcb->control.exit_code = SVM_EXIT_NPF;
-		svm->vmcb->control.exit_code_hi = 0;
-		svm->vmcb->control.exit_info_1 = (1ULL << 32);
-		svm->vmcb->control.exit_info_2 = fault->address;
+		vmcb->control.exit_code = SVM_EXIT_NPF;
+		vmcb->control.exit_code_hi = 0;
+		vmcb->control.exit_info_1 = (1ULL << 32);
+		vmcb->control.exit_info_2 = fault->address;
 	}
 
-	svm->vmcb->control.exit_info_1 &= ~0xffffffffULL;
-	svm->vmcb->control.exit_info_1 |= fault->error_code;
+	vmcb->control.exit_info_1 &= ~0xffffffffULL;
+	vmcb->control.exit_info_1 |= fault->error_code;
 
 	nested_svm_vmexit(svm);
 }
 
-static void svm_inject_page_fault_nested(struct kvm_vcpu *vcpu, struct x86_exception *fault)
+static bool nested_svm_handle_page_fault_workaround(struct kvm_vcpu *vcpu,
+						    struct x86_exception *fault)
 {
-       struct vcpu_svm *svm = to_svm(vcpu);
-       WARN_ON(!is_guest_mode(vcpu));
+	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb *vmcb = svm->vmcb;
+
+ 	WARN_ON(!is_guest_mode(vcpu));
 
 	if (vmcb12_is_intercept(&svm->nested.ctl,
 				INTERCEPT_EXCEPTION_OFFSET + PF_VECTOR) &&
-	    !svm->nested.nested_run_pending) {
-               svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + PF_VECTOR;
-               svm->vmcb->control.exit_code_hi = 0;
-               svm->vmcb->control.exit_info_1 = fault->error_code;
-               svm->vmcb->control.exit_info_2 = fault->address;
-               nested_svm_vmexit(svm);
-       } else {
-               kvm_inject_page_fault(vcpu, fault);
-       }
+	    !WARN_ON_ONCE(svm->nested.nested_run_pending)) {
+	     	vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + PF_VECTOR;
+		vmcb->control.exit_code_hi = 0;
+		vmcb->control.exit_info_1 = fault->error_code;
+		vmcb->control.exit_info_2 = fault->address;
+		nested_svm_vmexit(svm);
+		return true;
+	}
+
+	return false;
 }
 
 static u64 nested_svm_get_tdp_pdptr(struct kvm_vcpu *vcpu, int index)
···
 	vcpu->arch.walk_mmu = &vcpu->arch.root_mmu;
 }
 
+static bool nested_vmcb_needs_vls_intercept(struct vcpu_svm *svm)
+{
+	if (!svm->v_vmload_vmsave_enabled)
+		return true;
+
+	if (!nested_npt_enabled(svm))
+		return true;
+
+	if (!(svm->nested.ctl.virt_ext & VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK))
+		return true;
+
+	return false;
+}
+
 void recalc_intercepts(struct vcpu_svm *svm)
 {
 	struct vmcb_control_area *c, *h;
···
 	if (!intercept_smi)
 		vmcb_clr_intercept(c, INTERCEPT_SMI);
 
-	vmcb_set_intercept(c, INTERCEPT_VMLOAD);
-	vmcb_set_intercept(c, INTERCEPT_VMSAVE);
+	if (nested_vmcb_needs_vls_intercept(svm)) {
+		/*
+		 * If the virtual VMLOAD/VMSAVE is not enabled for the L2,
+		 * we must intercept these instructions to correctly
+		 * emulate them in case L1 doesn't intercept them.
+		 */
+		vmcb_set_intercept(c, INTERCEPT_VMLOAD);
+		vmcb_set_intercept(c, INTERCEPT_VMSAVE);
+	} else {
+		WARN_ON(!(c->virt_ext & VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK));
+	}
 }
 
 /*
···
 		 */
 		mask &= ~V_IRQ_MASK;
 	}
+
+	if (nested_vgif_enabled(svm))
+		mask |= V_GIF_MASK;
+
 	svm->nested.ctl.int_ctl        &= ~mask;
 	svm->nested.ctl.int_ctl        |= svm->vmcb->control.int_ctl & mask;
 }
···
 	}
 
 	vmcb12->control.exit_int_info = exit_int_info;
-}
-
-static inline bool nested_npt_enabled(struct vcpu_svm *svm)
-{
-	return svm->nested.ctl.nested_ctl & SVM_NESTED_CTL_NP_ENABLE;
 }
 
 static void nested_svm_transition_tlb_flush(struct kvm_vcpu *vcpu)
···
 static void nested_vmcb02_prepare_save(struct vcpu_svm *svm, struct vmcb *vmcb12)
 {
 	bool new_vmcb12 = false;
+	struct vmcb *vmcb01 = svm->vmcb01.ptr;
+	struct vmcb *vmcb02 = svm->nested.vmcb02.ptr;
 
 	nested_vmcb02_compute_g_pat(svm);
 
···
 	}
 
 	if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_SEG))) {
-		svm->vmcb->save.es = vmcb12->save.es;
-		svm->vmcb->save.cs = vmcb12->save.cs;
-		svm->vmcb->save.ss = vmcb12->save.ss;
-		svm->vmcb->save.ds = vmcb12->save.ds;
-		svm->vmcb->save.cpl = vmcb12->save.cpl;
-		vmcb_mark_dirty(svm->vmcb, VMCB_SEG);
+		vmcb02->save.es = vmcb12->save.es;
+		vmcb02->save.cs = vmcb12->save.cs;
+		vmcb02->save.ss = vmcb12->save.ss;
+		vmcb02->save.ds = vmcb12->save.ds;
+		vmcb02->save.cpl = vmcb12->save.cpl;
+		vmcb_mark_dirty(vmcb02, VMCB_SEG);
 	}
 
 	if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_DT))) {
-		svm->vmcb->save.gdtr = vmcb12->save.gdtr;
-		svm->vmcb->save.idtr = vmcb12->save.idtr;
-		vmcb_mark_dirty(svm->vmcb, VMCB_DT);
+		vmcb02->save.gdtr = vmcb12->save.gdtr;
+		vmcb02->save.idtr = vmcb12->save.idtr;
+		vmcb_mark_dirty(vmcb02, VMCB_DT);
 	}
 
 	kvm_set_rflags(&svm->vcpu, vmcb12->save.rflags | X86_EFLAGS_FIXED);
···
 	kvm_rip_write(&svm->vcpu, vmcb12->save.rip);
 
 	/* In case we don't even reach vcpu_run, the fields are not updated */
-	svm->vmcb->save.rax = vmcb12->save.rax;
-	svm->vmcb->save.rsp = vmcb12->save.rsp;
-	svm->vmcb->save.rip = vmcb12->save.rip;
+	vmcb02->save.rax = vmcb12->save.rax;
+	vmcb02->save.rsp = vmcb12->save.rsp;
+	vmcb02->save.rip = vmcb12->save.rip;
 
 	/* These bits will be set properly on the first execution when new_vmc12 is true */
 	if (unlikely(new_vmcb12 || vmcb_is_dirty(vmcb12, VMCB_DR))) {
-		svm->vmcb->save.dr7 = svm->nested.save.dr7 | DR7_FIXED_1;
+		vmcb02->save.dr7 = svm->nested.save.dr7 | DR7_FIXED_1;
 		svm->vcpu.arch.dr6  = svm->nested.save.dr6 | DR6_ACTIVE_LOW;
-		vmcb_mark_dirty(svm->vmcb, VMCB_DR);
+		vmcb_mark_dirty(vmcb02, VMCB_DR);
+	}
+
+	if (unlikely(svm->lbrv_enabled && (svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK))) {
+		/*
+		 * Reserved bits of DEBUGCTL are ignored.  Be consistent with
+		 * svm_set_msr's definition of reserved bits.
+		 */
+		svm_copy_lbrs(vmcb02, vmcb12);
+		vmcb02->save.dbgctl &= ~DEBUGCTL_RESERVED_BITS;
+		svm_update_lbrv(&svm->vcpu);
+
+	} else if (unlikely(vmcb01->control.virt_ext & LBR_CTL_ENABLE_MASK)) {
+		svm_copy_lbrs(vmcb02, vmcb01);
 	}
 }
 
 static void nested_vmcb02_prepare_control(struct vcpu_svm *svm)
 {
-	const u32 int_ctl_vmcb01_bits =
-		V_INTR_MASKING_MASK | V_GIF_MASK | V_GIF_ENABLE_MASK;
-
-	const u32 int_ctl_vmcb12_bits = V_TPR_MASK | V_IRQ_INJECTION_BITS_MASK;
+	u32 int_ctl_vmcb01_bits = V_INTR_MASKING_MASK;
+	u32 int_ctl_vmcb12_bits = V_TPR_MASK | V_IRQ_INJECTION_BITS_MASK;
 
 	struct kvm_vcpu *vcpu = &svm->vcpu;
+	struct vmcb *vmcb01 = svm->vmcb01.ptr;
+	struct vmcb *vmcb02 = svm->nested.vmcb02.ptr;
 
 	/*
 	 * Filled at exit: exit_code, exit_code_hi, exit_info_1, exit_info_2,
 	 * exit_int_info, exit_int_info_err, next_rip, insn_len, insn_bytes.
 	 */
 
-	/*
-	 * Also covers avic_vapic_bar, avic_backing_page, avic_logical_id,
-	 * avic_physical_id.
-	 */
-	WARN_ON(kvm_apicv_activated(svm->vcpu.kvm));
+	if (svm->vgif_enabled && (svm->nested.ctl.int_ctl & V_GIF_ENABLE_MASK))
+		int_ctl_vmcb12_bits |= (V_GIF_MASK | V_GIF_ENABLE_MASK);
+	else
+		int_ctl_vmcb01_bits |= (V_GIF_MASK | V_GIF_ENABLE_MASK);
 
 	/* Copied from vmcb01.  msrpm_base can be overwritten later.  */
-	svm->vmcb->control.nested_ctl = svm->vmcb01.ptr->control.nested_ctl;
-	svm->vmcb->control.iopm_base_pa = svm->vmcb01.ptr->control.iopm_base_pa;
-	svm->vmcb->control.msrpm_base_pa = svm->vmcb01.ptr->control.msrpm_base_pa;
+	vmcb02->control.nested_ctl = vmcb01->control.nested_ctl;
+	vmcb02->control.iopm_base_pa = vmcb01->control.iopm_base_pa;
+	vmcb02->control.msrpm_base_pa = vmcb01->control.msrpm_base_pa;
 
 	/* Done at vmrun: asid.  */
 
 	/* Also overwritten later if necessary.  */
-	svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
+	vmcb02->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
 
 	/* nested_cr3.  */
 	if (nested_npt_enabled(svm))
···
 			svm->nested.ctl.tsc_offset,
 			svm->tsc_ratio_msr);
 
-	svm->vmcb->control.tsc_offset = vcpu->arch.tsc_offset;
+	vmcb02->control.tsc_offset = vcpu->arch.tsc_offset;
 
 	if (svm->tsc_ratio_msr != kvm_default_tsc_scaling_ratio) {
 		WARN_ON(!svm->tsc_scaling_enabled);
 		nested_svm_update_tsc_ratio_msr(vcpu);
 	}
 
-	svm->vmcb->control.int_ctl             =
+	vmcb02->control.int_ctl             =
 		(svm->nested.ctl.int_ctl & int_ctl_vmcb12_bits) |
-		(svm->vmcb01.ptr->control.int_ctl & int_ctl_vmcb01_bits);
+		(vmcb01->control.int_ctl & int_ctl_vmcb01_bits);
 
-	svm->vmcb->control.int_vector          = svm->nested.ctl.int_vector;
-	svm->vmcb->control.int_state           = svm->nested.ctl.int_state;
-	svm->vmcb->control.event_inj           = svm->nested.ctl.event_inj;
-	svm->vmcb->control.event_inj_err       = svm->nested.ctl.event_inj_err;
+	vmcb02->control.int_vector          = svm->nested.ctl.int_vector;
+	vmcb02->control.int_state           = svm->nested.ctl.int_state;
+	vmcb02->control.event_inj           = svm->nested.ctl.event_inj;
+	vmcb02->control.event_inj_err       = svm->nested.ctl.event_inj_err;
+
+	vmcb02->control.virt_ext            = vmcb01->control.virt_ext &
+					      LBR_CTL_ENABLE_MASK;
+	if (svm->lbrv_enabled)
+		vmcb02->control.virt_ext  |=
+			(svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK);
+
+	if (!nested_vmcb_needs_vls_intercept(svm))
+		vmcb02->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
+
+	if (kvm_pause_in_guest(svm->vcpu.kvm)) {
+		/* use guest values since host doesn't use them */
+		vmcb02->control.pause_filter_count =
+				svm->pause_filter_enabled ?
+				svm->nested.ctl.pause_filter_count : 0;
+
+		vmcb02->control.pause_filter_thresh =
+				svm->pause_threshold_enabled ?
+				svm->nested.ctl.pause_filter_thresh : 0;
+
+	} else if (!vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_PAUSE)) {
+		/* use host values when guest doesn't use them */
+		vmcb02->control.pause_filter_count = vmcb01->control.pause_filter_count;
+		vmcb02->control.pause_filter_thresh = vmcb01->control.pause_filter_thresh;
+	} else {
+		/*
+		 * Intercept every PAUSE otherwise and
+		 * ignore both host and guest values
+		 */
+		vmcb02->control.pause_filter_count = 0;
+		vmcb02->control.pause_filter_thresh = 0;
+	}
 
 	nested_svm_transition_tlb_flush(vcpu);
 
···
 	if (ret)
 		return ret;
 
-	if (!npt_enabled)
-		vcpu->arch.mmu->inject_page_fault = svm_inject_page_fault_nested;
-
 	if (!from_vmrun)
 		kvm_make_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu);
 
 	svm_set_gif(svm, true);
+
+	if (kvm_vcpu_apicv_active(vcpu))
+		kvm_make_request(KVM_REQ_APICV_UPDATE, vcpu);
 
 	return 0;
 }
···
 	struct vmcb *vmcb12;
 	struct kvm_host_map map;
 	u64 vmcb12_gpa;
+	struct vmcb *vmcb01 = svm->vmcb01.ptr;
 
 	if (!svm->nested.hsave_msr) {
 		kvm_inject_gp(vcpu, 0);
···
 	 * Since vmcb01 is not in use, we can use it to store some of the L1
 	 * state.
 	 */
-	svm->vmcb01.ptr->save.efer   = vcpu->arch.efer;
-	svm->vmcb01.ptr->save.cr0    = kvm_read_cr0(vcpu);
-	svm->vmcb01.ptr->save.cr4    = vcpu->arch.cr4;
-	svm->vmcb01.ptr->save.rflags = kvm_get_rflags(vcpu);
-	svm->vmcb01.ptr->save.rip    = kvm_rip_read(vcpu);
+	vmcb01->save.efer   = vcpu->arch.efer;
+	vmcb01->save.cr0    = kvm_read_cr0(vcpu);
+	vmcb01->save.cr4    = vcpu->arch.cr4;
+	vmcb01->save.rflags = kvm_get_rflags(vcpu);
+	vmcb01->save.rip    = kvm_rip_read(vcpu);
 
 	if (!npt_enabled)
-		svm->vmcb01.ptr->save.cr3 = kvm_read_cr3(vcpu);
+		vmcb01->save.cr3 = kvm_read_cr3(vcpu);
 
 	svm->nested.nested_run_pending = 1;
 
···
 int nested_svm_vmexit(struct vcpu_svm *svm)
 {
 	struct kvm_vcpu *vcpu = &svm->vcpu;
+	struct vmcb *vmcb01 = svm->vmcb01.ptr;
+	struct vmcb *vmcb02 = svm->nested.vmcb02.ptr;
 	struct vmcb *vmcb12;
-	struct vmcb *vmcb = svm->vmcb;
 	struct kvm_host_map map;
 	int rc;
-
-	/* Triple faults in L2 should never escape. */
-	WARN_ON_ONCE(kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu));
 
 	rc = kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.vmcb12_gpa), &map);
 	if (rc) {
···
 
 	/* Give the current vmcb to the guest */
 
-	vmcb12->save.es     = vmcb->save.es;
-	vmcb12->save.cs     = vmcb->save.cs;
-	vmcb12->save.ss     = vmcb->save.ss;
-	vmcb12->save.ds     = vmcb->save.ds;
-	vmcb12->save.gdtr   = vmcb->save.gdtr;
-	vmcb12->save.idtr   = vmcb->save.idtr;
+	vmcb12->save.es     = vmcb02->save.es;
+	vmcb12->save.cs     = vmcb02->save.cs;
+	vmcb12->save.ss     = vmcb02->save.ss;
+	vmcb12->save.ds     = vmcb02->save.ds;
+	vmcb12->save.gdtr   = vmcb02->save.gdtr;
+	vmcb12->save.idtr   = vmcb02->save.idtr;
 	vmcb12->save.efer   = svm->vcpu.arch.efer;
 	vmcb12->save.cr0    = kvm_read_cr0(vcpu);
 	vmcb12->save.cr3    = kvm_read_cr3(vcpu);
-	vmcb12->save.cr2    = vmcb->save.cr2;
+	vmcb12->save.cr2    = vmcb02->save.cr2;
 	vmcb12->save.cr4    = svm->vcpu.arch.cr4;
 	vmcb12->save.rflags = kvm_get_rflags(vcpu);
 	vmcb12->save.rip    = kvm_rip_read(vcpu);
 	vmcb12->save.rsp    = kvm_rsp_read(vcpu);
 	vmcb12->save.rax    = kvm_rax_read(vcpu);
-	vmcb12->save.dr7    = vmcb->save.dr7;
+	vmcb12->save.dr7    = vmcb02->save.dr7;
 	vmcb12->save.dr6    = svm->vcpu.arch.dr6;
-	vmcb12->save.cpl    = vmcb->save.cpl;
+	vmcb12->save.cpl    = vmcb02->save.cpl;
 
-	vmcb12->control.int_state         = vmcb->control.int_state;
-	vmcb12->control.exit_code         = vmcb->control.exit_code;
-	vmcb12->control.exit_code_hi      = vmcb->control.exit_code_hi;
-	vmcb12->control.exit_info_1       = vmcb->control.exit_info_1;
-	vmcb12->control.exit_info_2       = vmcb->control.exit_info_2;
+	vmcb12->control.int_state         = vmcb02->control.int_state;
+	vmcb12->control.exit_code         = vmcb02->control.exit_code;
+	vmcb12->control.exit_code_hi      = vmcb02->control.exit_code_hi;
+	vmcb12->control.exit_info_1       = vmcb02->control.exit_info_1;
+	vmcb12->control.exit_info_2       = vmcb02->control.exit_info_2;
 
 	if (vmcb12->control.exit_code != SVM_EXIT_ERR)
 		nested_save_pending_event_to_vmcb12(svm, vmcb12);
 
 	if (svm->nrips_enabled)
-		vmcb12->control.next_rip  = vmcb->control.next_rip;
+		vmcb12->control.next_rip  = vmcb02->control.next_rip;
 
 	vmcb12->control.int_ctl           = svm->nested.ctl.int_ctl;
 	vmcb12->control.tlb_ctl           = svm->nested.ctl.tlb_ctl;
 	vmcb12->control.event_inj         = svm->nested.ctl.event_inj;
 	vmcb12->control.event_inj_err     = svm->nested.ctl.event_inj_err;
 
+	if (!kvm_pause_in_guest(vcpu->kvm) && vmcb02->control.pause_filter_count)
+		vmcb01->control.pause_filter_count = vmcb02->control.pause_filter_count;
+
 	nested_svm_copy_common_state(svm->nested.vmcb02.ptr, svm->vmcb01.ptr);
 
 	svm_switch_vmcb(svm, &svm->vmcb01);
+
+	if (unlikely(svm->lbrv_enabled && (svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK))) {
+		svm_copy_lbrs(vmcb12, vmcb02);
+		svm_update_lbrv(vcpu);
+	} else if (unlikely(vmcb01->control.virt_ext & LBR_CTL_ENABLE_MASK)) {
+		svm_copy_lbrs(vmcb01, vmcb02);
+		svm_update_lbrv(vcpu);
+	}
 
 	/*
 	 * On vmexit the  GIF is set to false and
 	 * no event can be injected in L1.
 	 */
 	svm_set_gif(svm, false);
-	svm->vmcb->control.exit_int_info = 0;
+	vmcb01->control.exit_int_info = 0;
 
 	svm->vcpu.arch.tsc_offset = svm->vcpu.arch.l1_tsc_offset;
-	if (svm->vmcb->control.tsc_offset != svm->vcpu.arch.tsc_offset) {
-		svm->vmcb->control.tsc_offset = svm->vcpu.arch.tsc_offset;
-		vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
+	if (vmcb01->control.tsc_offset != svm->vcpu.arch.tsc_offset) {
+		vmcb01->control.tsc_offset = svm->vcpu.arch.tsc_offset;
+		vmcb_mark_dirty(vmcb01, VMCB_INTERCEPTS);
 	}
 
 	if (svm->tsc_ratio_msr != kvm_default_tsc_scaling_ratio) {
···
 	/*
 	 * Restore processor state that had been saved in vmcb01
 	 */
-	kvm_set_rflags(vcpu, svm->vmcb->save.rflags);
-	svm_set_efer(vcpu, svm->vmcb->save.efer);
-	svm_set_cr0(vcpu, svm->vmcb->save.cr0 | X86_CR0_PE);
-	svm_set_cr4(vcpu, svm->vmcb->save.cr4);
-	kvm_rax_write(vcpu, svm->vmcb->save.rax);
-	kvm_rsp_write(vcpu, svm->vmcb->save.rsp);
-	kvm_rip_write(vcpu, svm->vmcb->save.rip);
+	kvm_set_rflags(vcpu, vmcb01->save.rflags);
+	svm_set_efer(vcpu, vmcb01->save.efer);
+	svm_set_cr0(vcpu, vmcb01->save.cr0 | X86_CR0_PE);
+	svm_set_cr4(vcpu, vmcb01->save.cr4);
+	kvm_rax_write(vcpu, vmcb01->save.rax);
+	kvm_rsp_write(vcpu, vmcb01->save.rsp);
+	kvm_rip_write(vcpu, vmcb01->save.rip);
 
 	svm->vcpu.arch.dr7 = DR7_FIXED_1;
 	kvm_update_dr7(&svm->vcpu);
···
 
 	nested_svm_uninit_mmu_context(vcpu);
 
-	rc = nested_svm_load_cr3(vcpu, svm->vmcb->save.cr3, false, true);
+	rc = nested_svm_load_cr3(vcpu, vmcb01->save.cr3, false, true);
 	if (rc)
 		return 1;
 
···
 	 * right now so that it an be accounted for before we execute
 	 * L1's next instruction.
 	 */
-	if (unlikely(svm->vmcb->save.rflags & X86_EFLAGS_TF))
+	if (unlikely(vmcb01->save.rflags & X86_EFLAGS_TF))
 		kvm_queue_exception(&(svm->vcpu), DB_VECTOR);
+
+	/*
+	 * Un-inhibit the AVIC right away, so that other vCPUs can start
+	 * to benefit from it right away.
+	 */
+	if (kvm_apicv_activated(vcpu->kvm))
+		kvm_vcpu_update_apicv(vcpu);
 
 	return 0;
 }
···
 static void nested_svm_inject_exception_vmexit(struct vcpu_svm *svm)
 {
 	unsigned int nr = svm->vcpu.arch.exception.nr;
+	struct vmcb *vmcb = svm->vmcb;
 
-	svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr;
-	svm->vmcb->control.exit_code_hi = 0;
+	vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr;
+	vmcb->control.exit_code_hi = 0;
 
 	if (svm->vcpu.arch.exception.has_error_code)
-		svm->vmcb->control.exit_info_1 = svm->vcpu.arch.exception.error_code;
+		vmcb->control.exit_info_1 = svm->vcpu.arch.exception.error_code;
 
 	/*
 	 * EXITINFO2 is undefined for all exception intercepts other
···
 	 */
 	if (nr == PF_VECTOR) {
 		if (svm->vcpu.arch.exception.nested_apf)
-			svm->vmcb->control.exit_info_2 = svm->vcpu.arch.apf.nested_apf_token;
+			vmcb->control.exit_info_2 = svm->vcpu.arch.apf.nested_apf_token;
 		else if (svm->vcpu.arch.exception.has_payload)
-			svm->vmcb->control.exit_info_2 = svm->vcpu.arch.exception.payload;
+			vmcb->control.exit_info_2 = svm->vcpu.arch.exception.payload;
 		else
-			svm->vmcb->control.exit_info_2 = svm->vcpu.arch.cr2;
+			vmcb->control.exit_info_2 = svm->vcpu.arch.cr2;
 	} else if (nr == DB_VECTOR) {
 		/* See inject_pending_event.  */
 		kvm_deliver_exception_payload(&svm->vcpu);
···
 struct kvm_x86_nested_ops svm_nested_ops = {
 	.leave_nested = svm_leave_nested,
 	.check_events = svm_check_nested_events,
+	.handle_page_fault_workaround = nested_svm_handle_page_fault_workaround,
 	.triple_fault = nested_svm_triple_fault,
 	.get_nested_state_pages = svm_get_nested_state_pages,
 	.get_state = svm_get_nested_state,

arch/x86/kvm/svm/pmu.c

···
 	}
 }
 
-struct kvm_pmu_ops amd_pmu_ops = {
+struct kvm_pmu_ops amd_pmu_ops __initdata = {
 	.pmc_perf_hw_id = amd_pmc_perf_hw_id,
 	.pmc_is_enabled = amd_pmc_is_enabled,
 	.pmc_idx_to_pmc = amd_pmc_idx_to_pmc,

arch/x86/kvm/svm/sev.c

···
 		if (params.len > SEV_FW_BLOB_MAX_SIZE)
 			return -EINVAL;
 
-		blob = kmalloc(params.len, GFP_KERNEL_ACCOUNT);
+		blob = kzalloc(params.len, GFP_KERNEL_ACCOUNT);
 		if (!blob)
 			return -ENOMEM;
 
···
 	if (!IS_ALIGNED(dst_paddr, 16) ||
 	    !IS_ALIGNED(paddr,     16) ||
 	    !IS_ALIGNED(size,      16)) {
-		tpage = (void *)alloc_page(GFP_KERNEL);
+		tpage = (void *)alloc_page(GFP_KERNEL | __GFP_ZERO);
 		if (!tpage)
 			return -ENOMEM;
 
···
 		if (params.len > SEV_FW_BLOB_MAX_SIZE)
 			return -EINVAL;
 
-		blob = kmalloc(params.len, GFP_KERNEL_ACCOUNT);
+		blob = kzalloc(params.len, GFP_KERNEL_ACCOUNT);
 		if (!blob)
 			return -ENOMEM;
 
···
 		return -EINVAL;
 
 	/* allocate the memory to hold the session data blob */
-	session_data = kmalloc(params.session_len, GFP_KERNEL_ACCOUNT);
+	session_data = kzalloc(params.session_len, GFP_KERNEL_ACCOUNT);
 	if (!session_data)
 		return -ENOMEM;
 
···
 
 	/* allocate memory for header and transport buffer */
 	ret = -ENOMEM;
-	hdr = kmalloc(params.hdr_len, GFP_KERNEL_ACCOUNT);
+	hdr = kzalloc(params.hdr_len, GFP_KERNEL_ACCOUNT);
 	if (!hdr)
 		goto e_unpin;
 
-	trans_data = kmalloc(params.trans_len, GFP_KERNEL_ACCOUNT);
+	trans_data = kzalloc(params.trans_len, GFP_KERNEL_ACCOUNT);
 	if (!trans_data)
 		goto e_free_hdr;
 
···
 		pr_info("SEV-ES guest requested termination: %#llx:%#llx\n",
 			reason_set, reason_code);
 
-		ret = -EINVAL;
-		break;
+		vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
+		vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SEV_TERM;
+		vcpu->run->system_event.ndata = 1;
+		vcpu->run->system_event.data[0] = control->ghcb_gpa;
+
+		return 0;
 	}
 	default:
 		/* Error, keep GHCB MSR value as-is */
···
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 1, 1);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1);
+
+	if (boot_cpu_has(X86_FEATURE_V_TSC_AUX) &&
+	    (guest_cpuid_has(&svm->vcpu, X86_FEATURE_RDTSCP) ||
+	     guest_cpuid_has(&svm->vcpu, X86_FEATURE_RDPID))) {
+		set_msr_interception(vcpu, svm->msrpm, MSR_TSC_AUX, 1, 1);
+		if (guest_cpuid_has(&svm->vcpu, X86_FEATURE_RDTSCP))
+			svm_clr_intercept(svm, INTERCEPT_RDTSCP);
+	}
 }
 
 void sev_es_vcpu_reset(struct vcpu_svm *svm)

arch/x86/kvm/svm/svm.c

···
 #define SEG_TYPE_LDT 2
 #define SEG_TYPE_BUSY_TSS16 3
 
-#define DEBUGCTL_RESERVED_BITS (~(0x3fULL))
-
 static bool erratum_383_found __read_mostly;
 
 u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly;
···
 	{ .index = MSR_EFER,				.always = false },
 	{ .index = MSR_IA32_CR_PAT,			.always = false },
 	{ .index = MSR_AMD64_SEV_ES_GHCB,		.always = true  },
+	{ .index = MSR_TSC_AUX,				.always = false },
 	{ .index = MSR_INVALID,				.always = false },
 };
 
···
 module_param(vls, int, 0444);
 
 /* enable/disable Virtual GIF */
-static int vgif = true;
+int vgif = true;
 module_param(vgif, int, 0444);
 
 /* enable/disable LBR virtualization */
···
  */
 static bool avic;
 module_param(avic, bool, 0444);
+
+static bool force_avic;
+module_param_unsafe(force_avic, bool, 0444);
 
 bool __read_mostly dump_invalid_vmcb;
 module_param(dump_invalid_vmcb, bool, 0644);
···
 	}
 }
 
+void svm_copy_lbrs(struct vmcb *to_vmcb, struct vmcb *from_vmcb)
+{
+	to_vmcb->save.dbgctl		= from_vmcb->save.dbgctl;
+	to_vmcb->save.br_from		= from_vmcb->save.br_from;
+	to_vmcb->save.br_to		= from_vmcb->save.br_to;
+	to_vmcb->save.last_excp_from	= from_vmcb->save.last_excp_from;
+	to_vmcb->save.last_excp_to	= from_vmcb->save.last_excp_to;
+
+	vmcb_mark_dirty(to_vmcb, VMCB_LBR);
+}
+
 static void svm_enable_lbrv(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
···
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 1, 1);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1);
+
+	/* Move the LBR msrs to the vmcb02 so that the guest can see them. */
+	if (is_guest_mode(vcpu))
+		svm_copy_lbrs(svm->vmcb, svm->vmcb01.ptr);
 }
 
 static void svm_disable_lbrv(struct kvm_vcpu *vcpu)
···
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 0, 0);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 0, 0);
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 0, 0);
+
+	/*
+	 * Move the LBR msrs back to the vmcb01 to avoid copying them
+	 * on nested guest entries.
+	 */
+	if (is_guest_mode(vcpu))
+		svm_copy_lbrs(svm->vmcb01.ptr, svm->vmcb);
+}
+
+static int svm_get_lbr_msr(struct vcpu_svm *svm, u32 index)
+{
+	/*
+	 * If the LBR virtualization is disabled, the LBR msrs are always
+	 * kept in the vmcb01 to avoid copying them on nested guest entries.
+	 *
+	 * If nested, and the LBR virtualization is enabled/disabled, the msrs
+	 * are moved between the vmcb01 and vmcb02 as needed.
+	 */
+	struct vmcb *vmcb =
+		(svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK) ?
+			svm->vmcb : svm->vmcb01.ptr;
+
+	switch (index) {
+	case MSR_IA32_DEBUGCTLMSR:
+		return vmcb->save.dbgctl;
+	case MSR_IA32_LASTBRANCHFROMIP:
+		return vmcb->save.br_from;
+	case MSR_IA32_LASTBRANCHTOIP:
+		return vmcb->save.br_to;
+	case MSR_IA32_LASTINTFROMIP:
+		return vmcb->save.last_excp_from;
+	case MSR_IA32_LASTINTTOIP:
+		return vmcb->save.last_excp_to;
+	default:
+		KVM_BUG(false, svm->vcpu.kvm,
+			"%s: Unknown MSR 0x%x", __func__, index);
+		return 0;
+	}
+}
+
+void svm_update_lbrv(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	bool enable_lbrv = svm_get_lbr_msr(svm, MSR_IA32_DEBUGCTLMSR) &
+					   DEBUGCTLMSR_LBR;
+
+	bool current_enable_lbrv = !!(svm->vmcb->control.virt_ext &
+				      LBR_CTL_ENABLE_MASK);
+
+	if (unlikely(is_guest_mode(vcpu) && svm->lbrv_enabled))
+		if (unlikely(svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK))
+			enable_lbrv = true;
+
+	if (enable_lbrv == current_enable_lbrv)
+		return;
+
+	if (enable_lbrv)
+		svm_enable_lbrv(vcpu);
+	else
+		svm_disable_lbrv(vcpu);
 }
 
 void disable_nmi_singlestep(struct vcpu_svm *svm)
···
 	struct vmcb_control_area *control = &svm->vmcb->control;
 	int old = control->pause_filter_count;
 
+	if (kvm_pause_in_guest(vcpu->kvm) || !old)
+		return;
+
 	control->pause_filter_count = __grow_ple_window(old,
 							pause_filter_count,
 							pause_filter_count_grow,
···
 	struct vcpu_svm *svm = to_svm(vcpu);
 	struct vmcb_control_area *control = &svm->vmcb->control;
 	int old = control->pause_filter_count;
+
+	if (kvm_pause_in_guest(vcpu->kvm) || !old)
+		return;
 
 	control->pause_filter_count =
 				__shrink_ple_window(old,
···
 
 		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 0, 0);
 		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 0, 0);
+
+		svm->v_vmload_vmsave_enabled = false;
 	} else {
 		/*
 		 * If hardware supports Virtual VMLOAD VMSAVE then enable it
···
 static void init_vmcb(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
-	struct vmcb_control_area *control = &svm->vmcb->control;
-	struct vmcb_save_area *save = &svm->vmcb->save;
+	struct vmcb *vmcb = svm->vmcb01.ptr;
+	struct vmcb_control_area *control = &vmcb->control;
+	struct vmcb_save_area *save = &vmcb->save;
 
 	svm_set_intercept(svm, INTERCEPT_CR0_READ);
 	svm_set_intercept(svm, INTERCEPT_CR3_READ);
···
 		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SPEC_CTRL, 1, 1);
 
 	if (kvm_vcpu_apicv_active(vcpu))
-		avic_init_vmcb(svm);
+		avic_init_vmcb(svm, vmcb);
 
 	if (vgif) {
 		svm_clr_intercept(svm, INTERCEPT_STGI);
···
 		}
 	}
 
-	svm_hv_init_vmcb(svm->vmcb);
+	svm_hv_init_vmcb(vmcb);
 	init_vmcb_after_set_cpuid(vcpu);
 
-	vmcb_mark_all_dirty(svm->vmcb);
+	vmcb_mark_all_dirty(vmcb);
 
 	enable_gif(svm);
 }
···
 	/*
 	 * The following fields are ignored when AVIC is enabled
 	 */
-	WARN_ON(kvm_apicv_activated(svm->vcpu.kvm));
+	WARN_ON(kvm_vcpu_apicv_activated(&svm->vcpu));
 
 	svm_set_intercept(svm, INTERCEPT_VINTR);
 
···
 		 * Likewise, clear the VINTR intercept, we will set it
 		 * again while processing KVM_REQ_EVENT if needed.
 		 */
-		if (vgif_enabled(svm))
+		if (vgif)
 			svm_clr_intercept(svm, INTERCEPT_STGI);
 		if (svm_is_intercept(svm, INTERCEPT_VINTR))
 			svm_clear_vintr(svm);
···
 		 * in use, we still rely on the VINTR intercept (rather than
 		 * STGI) to detect an open interrupt window.
 		*/
-		if (!vgif_enabled(svm))
+		if (!vgif)
 			svm_clear_vintr(svm);
 	}
 }
···
 	case MSR_TSC_AUX:
 		msr_info->data = svm->tsc_aux;
 		break;
-	/*
-	 * Nobody will change the following 5 values in the VMCB so we can
-	 * safely return them on rdmsr. They will always be 0 until LBRV is
-	 * implemented.
-	 */
 	case MSR_IA32_DEBUGCTLMSR:
-		msr_info->data = svm->vmcb->save.dbgctl;
-		break;
 	case MSR_IA32_LASTBRANCHFROMIP:
-		msr_info->data = svm->vmcb->save.br_from;
-		break;
 	case MSR_IA32_LASTBRANCHTOIP:
-		msr_info->data = svm->vmcb->save.br_to;
-		break;
 	case MSR_IA32_LASTINTFROMIP:
-		msr_info->data = svm->vmcb->save.last_excp_from;
-		break;
 	case MSR_IA32_LASTINTTOIP:
-		msr_info->data = svm->vmcb->save.last_excp_to;
+		msr_info->data = svm_get_lbr_msr(svm, msr_info->index);
 		break;
 	case MSR_VM_HSAVE_PA:
 		msr_info->data = svm->nested.hsave_msr;
···
 		if (data & DEBUGCTL_RESERVED_BITS)
 			return 1;
 
-		svm->vmcb->save.dbgctl = data;
-		vmcb_mark_dirty(svm->vmcb, VMCB_LBR);
-		if (data & (1ULL<<0))
-			svm_enable_lbrv(vcpu);
+		if (svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK)
+			svm->vmcb->save.dbgctl = data;
 		else
-			svm_disable_lbrv(vcpu);
+			svm->vmcb01.ptr->save.dbgctl = data;
+
+		svm_update_lbrv(vcpu);
+
 		break;
 	case MSR_VM_HSAVE_PA:
 		/*
···
 	svm_clear_vintr(to_svm(vcpu));
 
 	/*
-	 * For AVIC, the only reason to end up here is ExtINTs.
+	 * If not running nested, for AVIC, the only reason to end up here is ExtINTs.
 	 * In this case AVIC was temporarily disabled for
 	 * requesting the IRQ window and we have to re-enable it.
+	 *
+	 * If running nested, still remove the VM wide AVIC inhibit to
+	 * support case in which the interrupt window was requested when the
+	 * vCPU was not running nested.
+
+	 * All vCPUs which run still run nested, will remain to have their
+	 * AVIC still inhibited due to per-cpu AVIC inhibition.
 	 */
 	kvm_clear_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN);
 
···
 static int pause_interception(struct kvm_vcpu *vcpu)
 {
 	bool in_kernel;
-
 	/*
 	 * CPL is not made available for an SEV-ES guest, therefore
 	 * vcpu->arch.preempted_in_kernel can never be true.  Just
···
 	 */
 	in_kernel = !sev_es_guest(vcpu->kvm) && svm_get_cpl(vcpu) == 0;
 
-	if (!kvm_pause_in_guest(vcpu->kvm))
-		grow_ple_window(vcpu);
+	grow_ple_window(vcpu);
 
 	kvm_vcpu_on_spin(vcpu, in_kernel);
 	return kvm_skip_emulated_instruction(vcpu);
···
 	 * enabled, the STGI interception will not occur. Enable the irq
 	 * window under the assumption that the hardware will set the GIF.
 	 */
-	if (vgif_enabled(svm) || gif_set(svm)) {
+	if (vgif || gif_set(svm)) {
 		/*
 		 * IRQ window is not needed when AVIC is enabled,
 		 * unless we have pending ExtINT since it cannot be injected
-		 * via AVIC. In such case, we need to temporarily disable AVIC,
+		 * via AVIC. In such case, KVM needs to temporarily disable AVIC,
 		 * and fallback to injecting IRQ via V_IRQ.
+		 *
+		 * If running nested, AVIC is already locally inhibited
+		 * on this vCPU, therefore there is no need to request
+		 * the VM wide AVIC inhibition.
 		 */
-		kvm_set_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN);
+		if (!is_guest_mode(vcpu))
+			kvm_set_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN);
+
 		svm_set_vintr(svm);
 	}
 }
···
 		return; /* IRET will cause a vm exit */
 
 	if (!gif_set(svm)) {
-		if (vgif_enabled(svm))
+		if (vgif)
 			svm_set_intercept(svm, INTERCEPT_STGI);
 		return; /* STGI will cause a vm exit */
 	}
···
 		hv_track_root_tdp(vcpu, root_hpa);
 
 		cr3 = vcpu->arch.cr3;
-	} else if (vcpu->arch.mmu->shadow_root_level >= PT64_ROOT_4LEVEL) {
+	} else if (vcpu->arch.mmu->root_role.level >= PT64_ROOT_4LEVEL) {
 		cr3 = __sme_set(root_hpa) | kvm_get_active_pcid(vcpu);
 	} else {
 		/* PCID in the guest should be impossible with a 32-bit MMU. */
···
 			     guest_cpuid_has(vcpu, X86_FEATURE_NRIPS);
 
 	svm->tsc_scaling_enabled = tsc_scaling && guest_cpuid_has(vcpu, X86_FEATURE_TSCRATEMSR);
+	svm->lbrv_enabled = lbrv && guest_cpuid_has(vcpu, X86_FEATURE_LBRV);
+
+	svm->v_vmload_vmsave_enabled = vls && guest_cpuid_has(vcpu, X86_FEATURE_V_VMSAVE_VMLOAD);
+
+	svm->pause_filter_enabled = kvm_cpu_cap_has(X86_FEATURE_PAUSEFILTER) &&
+			guest_cpuid_has(vcpu, X86_FEATURE_PAUSEFILTER);
+
+	svm->pause_threshold_enabled = kvm_cpu_cap_has(X86_FEATURE_PFTHRESHOLD) &&
+			guest_cpuid_has(vcpu, X86_FEATURE_PFTHRESHOLD);
+
+	svm->vgif_enabled = vgif && guest_cpuid_has(vcpu, X86_FEATURE_VGIF);
 
 	svm_recalc_instruction_intercepts(vcpu, svm);
 
···
 		 */
 		if (guest_cpuid_has(vcpu, X86_FEATURE_X2APIC))
 			kvm_set_apicv_inhibit(kvm, APICV_INHIBIT_REASON_X2APIC);
-
-		/*
-		 * Currently, AVIC does not work with nested virtualization.
-		 * So, we disable AVIC when cpuid for SVM is set in the L1 guest.
-		 */
-		if (nested && guest_cpuid_has(vcpu, X86_FEATURE_SVM))
-			kvm_set_apicv_inhibit(kvm, APICV_INHIBIT_REASON_NESTED);
 	}
 	init_vmcb_after_set_cpuid(vcpu);
 }
···
 	svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP];
 	svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP];
 
-	ret = nested_svm_vmexit(svm);
+	ret = nested_svm_simple_vmexit(svm, SVM_EXIT_SW);
 	if (ret)
 		return ret;
 
···
 	struct vcpu_svm *svm = to_svm(vcpu);
 
 	if (!gif_set(svm)) {
-		if (vgif_enabled(svm))
+		if (vgif)
 			svm_set_intercept(svm, INTERCEPT_STGI);
 		/* STGI will cause a vm exit */
 	} else {
···
 
 	.sched_in = svm_sched_in,
 
-	.pmu_ops = &amd_pmu_ops,
 	.nested_ops = &svm_nested_ops,
 
 	.deliver_interrupt = svm_deliver_interrupt,
···
 	.complete_emulated_msr = svm_complete_emulated_msr,
 
 	.vcpu_deliver_sipi_vector = svm_vcpu_deliver_sipi_vector,
+	.vcpu_get_apicv_inhibit_reasons = avic_vcpu_get_apicv_inhibit_reasons,
 };
 
 /*
···
 
 		if (tsc_scaling)
 			kvm_cpu_cap_set(X86_FEATURE_TSCRATEMSR);
+
+		if (vls)
+			kvm_cpu_cap_set(X86_FEATURE_V_VMSAVE_VMLOAD);
+		if (lbrv)
+			kvm_cpu_cap_set(X86_FEATURE_LBRV);
+
+		if (boot_cpu_has(X86_FEATURE_PAUSEFILTER))
+			kvm_cpu_cap_set(X86_FEATURE_PAUSEFILTER);
+
+		if (boot_cpu_has(X86_FEATURE_PFTHRESHOLD))
+			kvm_cpu_cap_set(X86_FEATURE_PFTHRESHOLD);
+
+		if (vgif)
+			kvm_cpu_cap_set(X86_FEATURE_VGIF);
 
 		/* Nested VM can receive #VMEXIT instead of triggering #GP */
 		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
···
 			  get_npt_level(), PG_LEVEL_1G);
 	pr_info("kvm: Nested Paging %sabled\n", npt_enabled ? "en" : "dis");
 
+	/* Setup shadow_me_value and shadow_me_mask */
+	kvm_mmu_set_me_spte_mask(sme_me_mask, sme_me_mask);
+
 	/* Note, SEV setup consumes npt_enabled. */
 	sev_hardware_setup();
 
···
 			nrips = false;
 	}
 
-	enable_apicv = avic = avic && npt_enabled && boot_cpu_has(X86_FEATURE_AVIC);
+	enable_apicv = avic = avic && npt_enabled && (boot_cpu_has(X86_FEATURE_AVIC) || force_avic);
 
 	if (enable_apicv) {
-		pr_info("AVIC enabled\n");
+		if (!boot_cpu_has(X86_FEATURE_AVIC)) {
+			pr_warn("AVIC is not supported in CPUID but force enabled");
+			pr_warn("Your system might crash and burn");
+		} else
+			pr_info("AVIC enabled\n");
 
 		amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier);
 	} else {
 		svm_x86_ops.vcpu_blocking = NULL;
 		svm_x86_ops.vcpu_unblocking = NULL;
+		svm_x86_ops.vcpu_get_apicv_inhibit_reasons = NULL;
 	}
 
 	if (vls) {
···
 	.check_processor_compatibility = svm_check_processor_compat,
 
 	.runtime_ops = &svm_x86_ops,
+	.pmu_ops = &amd_pmu_ops,
 };
 
 static int __init svm_init(void)

arch/x86/kvm/svm/svm.h

···
 #define	IOPM_SIZE PAGE_SIZE * 3
 #define	MSRPM_SIZE PAGE_SIZE * 2
 
-#define MAX_DIRECT_ACCESS_MSRS	20
+#define MAX_DIRECT_ACCESS_MSRS	21
 #define MSRPM_OFFSETS	16
 extern u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly;
 extern bool npt_enabled;
+extern int vgif;
 extern bool intercept_smi;
 
 /*
···
 	unsigned int3_injected;
 	unsigned long int3_rip;
 
-	/* cached guest cpuid flags for faster access */
+	/* optional nested SVM features that are enabled for this guest  */
 	bool nrips_enabled                : 1;
 	bool tsc_scaling_enabled          : 1;
+	bool v_vmload_vmsave_enabled      : 1;
+	bool lbrv_enabled                 : 1;
+	bool pause_filter_enabled         : 1;
+	bool pause_threshold_enabled      : 1;
+	bool vgif_enabled                 : 1;
 
 	u32 ldr_reg;
 	u32 dfr_reg;
···
 	return vmcb_is_intercept(&svm->vmcb->control, bit);
 }
 
-static inline bool vgif_enabled(struct vcpu_svm *svm)
+static inline bool nested_vgif_enabled(struct vcpu_svm *svm)
 {
-	return !!(svm->vmcb->control.int_ctl & V_GIF_ENABLE_MASK);
+	return svm->vgif_enabled && (svm->nested.ctl.int_ctl & V_GIF_ENABLE_MASK);
+}
+
+static inline struct vmcb *get_vgif_vmcb(struct vcpu_svm *svm)
+{
+	if (!vgif)
+		return NULL;
+
+	if (is_guest_mode(&svm->vcpu) && !nested_vgif_enabled(svm))
+		return svm->nested.vmcb02.ptr;
+	else
+		return svm->vmcb01.ptr;
 }
 
 static inline void enable_gif(struct vcpu_svm *svm)
 {
-	if (vgif_enabled(svm))
-		svm->vmcb->control.int_ctl |= V_GIF_MASK;
+	struct vmcb *vmcb = get_vgif_vmcb(svm);
+
+	if (vmcb)
+		vmcb->control.int_ctl |= V_GIF_MASK;
 	else
 		svm->vcpu.arch.hflags |= HF_GIF_MASK;
 }
 
 static inline void disable_gif(struct vcpu_svm *svm)
 {
-	if (vgif_enabled(svm))
-		svm->vmcb->control.int_ctl &= ~V_GIF_MASK;
+	struct vmcb *vmcb = get_vgif_vmcb(svm);
+
+	if (vmcb)
+		vmcb->control.int_ctl &= ~V_GIF_MASK;
 	else
 		svm->vcpu.arch.hflags &= ~HF_GIF_MASK;
 }
 
 static inline bool gif_set(struct vcpu_svm *svm)
 {
-	if (vgif_enabled(svm))
-		return !!(svm->vmcb->control.int_ctl & V_GIF_MASK);
+	struct vmcb *vmcb = get_vgif_vmcb(svm);
+
+	if (vmcb)
+		return !!(vmcb->control.int_ctl & V_GIF_MASK);
 	else
 		return !!(svm->vcpu.arch.hflags & HF_GIF_MASK);
 }
 
+static inline bool nested_npt_enabled(struct vcpu_svm *svm)
+{
+	return svm->nested.ctl.nested_ctl & SVM_NESTED_CTL_NP_ENABLE;
+}
+
 /* svm.c */
 #define MSR_INVALID				0xffffffffU
+
+#define DEBUGCTL_RESERVED_BITS (~(0x3fULL))
 
 extern bool dump_invalid_vmcb;
 
···
 u32 *svm_vcpu_alloc_msrpm(void);
 void svm_vcpu_init_msrpm(struct kvm_vcpu *vcpu, u32 *msrpm);
 void svm_vcpu_free_msrpm(u32 *msrpm);
+void svm_copy_lbrs(struct vmcb *to_vmcb, struct vmcb *from_vmcb);
+void svm_update_lbrv(struct kvm_vcpu *vcpu);
 
 int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer);
 void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
···
 int avic_ga_log_notifier(u32 ga_tag);
 void avic_vm_destroy(struct kvm *kvm);
 int avic_vm_init(struct kvm *kvm);
-void avic_init_vmcb(struct vcpu_svm *svm);
+void avic_init_vmcb(struct vcpu_svm *svm, struct vmcb *vmcb);
 int avic_incomplete_ipi_interception(struct kvm_vcpu *vcpu);
 int avic_unaccelerated_access_interception(struct kvm_vcpu *vcpu);
 int avic_init_vcpu(struct vcpu_svm *svm);
···
 void avic_vcpu_blocking(struct kvm_vcpu *vcpu);
 void avic_vcpu_unblocking(struct kvm_vcpu *vcpu);
 void avic_ring_doorbell(struct kvm_vcpu *vcpu);
+unsigned long avic_vcpu_get_apicv_inhibit_reasons(struct kvm_vcpu *vcpu);
 
 /* sev.c */
 

arch/x86/kvm/trace.h

···
 		  __entry->vmid, __entry->vcpuid)
 );
 
+TRACE_EVENT(kvm_avic_kick_vcpu_slowpath,
+	    TP_PROTO(u32 icrh, u32 icrl, u32 index),
+	    TP_ARGS(icrh, icrl, index),
+
+	TP_STRUCT__entry(
+		__field(u32, icrh)
+		__field(u32, icrl)
+		__field(u32, index)
+	),
+
+	TP_fast_assign(
+		__entry->icrh = icrh;
+		__entry->icrl = icrl;
+		__entry->index = index;
+	),
+
+	TP_printk("icrh:icrl=%#08x:%08x, index=%u",
+		  __entry->icrh, __entry->icrl, __entry->index)
+);
+
 TRACE_EVENT(kvm_hv_timer_state,
 		TP_PROTO(unsigned int vcpu_id, unsigned int hv_timer_in_use),
 		TP_ARGS(vcpu_id, hv_timer_in_use),

arch/x86/kvm/vmx/nested.c

···
 	return 0;
 }
 
-
-static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu,
-		struct x86_exception *fault)
+static bool nested_vmx_handle_page_fault_workaround(struct kvm_vcpu *vcpu,
+						    struct x86_exception *fault)
 {
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
 
 	WARN_ON(!is_guest_mode(vcpu));
 
 	if (nested_vmx_is_page_fault_vmexit(vmcs12, fault->error_code) &&
-		!to_vmx(vcpu)->nested.nested_run_pending) {
+	    !WARN_ON_ONCE(to_vmx(vcpu)->nested.nested_run_pending)) {
 		vmcs12->vm_exit_intr_error_code = fault->error_code;
 		nested_vmx_vmexit(vcpu, EXIT_REASON_EXCEPTION_NMI,
 				  PF_VECTOR | INTR_TYPE_HARD_EXCEPTION |
 				  INTR_INFO_DELIVER_CODE_MASK | INTR_INFO_VALID_MASK,
 				  fault->address);
-	} else {
-		kvm_inject_page_fault(vcpu, fault);
+		return true;
 	}
+	return false;
 }
 
 static int nested_vmx_check_io_bitmap_controls(struct kvm_vcpu *vcpu,
···
 		vmcs_write64(GUEST_PDPTR3, vmcs12->guest_pdptr3);
 	}
 
-	if (!enable_ept)
-		vcpu->arch.walk_mmu->inject_page_fault = vmx_inject_page_fault_nested;
-
 	if ((vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL) &&
 	    WARN_ON_ONCE(kvm_set_msr(vcpu, MSR_CORE_PERF_GLOBAL_CTRL,
 				     vmcs12->guest_ia32_perf_global_ctrl))) {
···
 }
 
 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;
 
···
 			idt_vectoring |= INTR_TYPE_EXT_INTR;
 
 		vmcs12->idt_vectoring_info_field = idt_vectoring;
+	} else {
+		vmcs12->idt_vectoring_info_field = 0;
 	}
 }
 
···
 	if (to_vmx(vcpu)->exit_reason.enclave_mode)
 		vmcs12->vm_exit_reason |= VMX_EXIT_REASONS_SGX_ENCLAVE_MODE;
 	vmcs12->exit_qualification = exit_qualification;
-	vmcs12->vm_exit_intr_info = exit_intr_info;
 
-	vmcs12->idt_vectoring_info_field = 0;
-	vmcs12->vm_exit_instruction_len = vmcs_read32(VM_EXIT_INSTRUCTION_LEN);
-	vmcs12->vmx_instruction_info = vmcs_read32(VMX_INSTRUCTION_INFO);
-
+	/*
+	 * On VM-Exit due to a failed VM-Entry, the VMCS isn't marked launched
+	 * and only EXIT_REASON and EXIT_QUALIFICATION are updated, all other
+	 * exit info fields are unmodified.
+	 */
 	if (!(vmcs12->vm_exit_reason & VMX_EXIT_REASONS_FAILED_VMENTRY)) {
 		vmcs12->launch_state = 1;
 
···
 		 * Transfer the event that L0 or L1 may wanted to inject into
 		 * L2 to IDT_VECTORING_INFO_FIELD.
 		 */
-		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);
+		vmcs12->vmx_instruction_info = vmcs_read32(VMX_INSTRUCTION_INFO);
 
 		/*
 		 * According to spec, there's no need to store the guest's
···
 
 	/* trying to cancel vmlaunch/vmresume is a bug */
 	WARN_ON_ONCE(vmx->nested.nested_run_pending);
-
-	/* Similarly, triple faults in L2 should never escape. */
-	WARN_ON_ONCE(kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu));
 
 	if (kvm_check_request(KVM_REQ_GET_NESTED_STATE_PAGES, vcpu)) {
 		/*
···
 struct kvm_x86_nested_ops vmx_nested_ops = {
 	.leave_nested = vmx_leave_nested,
 	.check_events = vmx_check_nested_events,
+	.handle_page_fault_workaround = nested_vmx_handle_page_fault_workaround,
 	.hv_timer_pending = nested_vmx_preemption_timer_pending,
 	.triple_fault = nested_vmx_triple_fault,
 	.get_state = vmx_get_nested_state,

arch/x86/kvm/vmx/pmu_intel.c

···
 		intel_pmu_release_guest_lbr_event(vcpu);
 }
 
-struct kvm_pmu_ops intel_pmu_ops = {
+struct kvm_pmu_ops intel_pmu_ops __initdata = {
 	.pmc_perf_hw_id = intel_pmc_perf_hw_id,
 	.pmc_is_enabled = intel_pmc_is_enabled,
 	.pmc_idx_to_pmc = intel_pmc_idx_to_pmc,

arch/x86/kvm/vmx/posted_intr.c

···
 void pi_wakeup_handler(void)
 {
 	int cpu = smp_processor_id();
+	struct list_head *wakeup_list = &per_cpu(wakeup_vcpus_on_cpu, cpu);
+	raw_spinlock_t *spinlock = &per_cpu(wakeup_vcpus_on_cpu_lock, cpu);
 	struct vcpu_vmx *vmx;
 
-	raw_spin_lock(&per_cpu(wakeup_vcpus_on_cpu_lock, cpu));
-	list_for_each_entry(vmx, &per_cpu(wakeup_vcpus_on_cpu, cpu),
-			    pi_wakeup_list) {
+	raw_spin_lock(spinlock);
+	list_for_each_entry(vmx, wakeup_list, pi_wakeup_list) {
 
 		if (pi_test_on(&vmx->pi_desc))
 			kvm_vcpu_wake_up(&vmx->vcpu);
 	}
-	raw_spin_unlock(&per_cpu(wakeup_vcpus_on_cpu_lock, cpu));
+	raw_spin_unlock(spinlock);
 }
 
 void __init pi_init_cpu(int cpu)
···
 			continue;
 		}
 
-		vcpu_info.pi_desc_addr = __pa(&to_vmx(vcpu)->pi_desc);
+		vcpu_info.pi_desc_addr = __pa(vcpu_to_pi_desc(vcpu));
 		vcpu_info.vector = irq.vector;
 
 		trace_kvm_pi_irte_update(host_irq, vcpu->vcpu_id, e->gsi,

arch/x86/kvm/vmx/vmcs.h

···
 	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);

arch/x86/kvm/vmx/vmx.c

···
 				&_cpu_based_exec_control) < 0)
 		return -EIO;
 #ifdef CONFIG_X86_64
-	if ((_cpu_based_exec_control & CPU_BASED_TPR_SHADOW))
+	if (_cpu_based_exec_control & CPU_BASED_TPR_SHADOW)
 		_cpu_based_exec_control &= ~CPU_BASED_CR8_LOAD_EXITING &
 					   ~CPU_BASED_CR8_STORE_EXITING;
 #endif
···
 
 	if (enable_ept)
 		ept_sync_context(construct_eptp(vcpu, root_hpa,
-						mmu->shadow_root_level));
+						mmu->root_role.level));
 	else
 		vpid_sync_context(vmx_get_current_vpid(vcpu));
 }
···
 	if (cpu_has_secondary_exec_ctrls())
 		secondary_exec_controls_set(vmx, vmx_secondary_exec_control(vmx));
 
-	if (kvm_vcpu_apicv_active(&vmx->vcpu)) {
+	if (enable_apicv && lapic_in_kernel(&vmx->vcpu)) {
 		vmcs_write64(EOI_EXIT_BITMAP0, 0);
 		vmcs_write64(EOI_EXIT_BITMAP1, 0);
 		vmcs_write64(EOI_EXIT_BITMAP2, 0);
···
 	error_code |= (exit_qualification & EPT_VIOLATION_ACC_INSTR)
 		      ? PFERR_FETCH_MASK : 0;
 	/* ept page table entry is present? */
-	error_code |= (exit_qualification &
-		       (EPT_VIOLATION_READABLE | EPT_VIOLATION_WRITABLE |
-			EPT_VIOLATION_EXECUTABLE))
+	error_code |= (exit_qualification & EPT_VIOLATION_RWX_MASK)
 		      ? PFERR_PRESENT_MASK : 0;
 
 	error_code |= (exit_qualification & EPT_VIOLATION_GVA_TRANSLATED) != 0 ?
···
 	.cpu_dirty_log_size = PML_ENTITY_NUM,
 	.update_cpu_dirty_logging = vmx_update_cpu_dirty_logging,
 
-	.pmu_ops = &intel_pmu_ops,
 	.nested_ops = &vmx_nested_ops,
 
 	.pi_update_irte = vmx_pi_update_irte,
···
 	struct kvm_vcpu *vcpu = kvm_get_running_vcpu();
 
 	/* '0' on failure so that the !PT case can use a RET0 static call. */
-	if (!kvm_arch_pmi_in_guest(vcpu))
+	if (!vcpu || !kvm_handling_nmi_from_guest(vcpu))
 		return 0;
 
 	kvm_make_request(KVM_REQ_PMI, vcpu);
···
 
 	for (i = 0; i < ARRAY_SIZE(vmx_uret_msrs_list); ++i)
 		kvm_add_user_return_msr(vmx_uret_msrs_list[i]);
+}
+
+static void __init vmx_setup_me_spte_mask(void)
+{
+	u64 me_mask = 0;
+
+	/*
+	 * kvm_get_shadow_phys_bits() returns shadow_phys_bits.  Use
+	 * the former to avoid exposing shadow_phys_bits.
+	 *
+	 * On pre-MKTME system, boot_cpu_data.x86_phys_bits equals to
+	 * shadow_phys_bits.  On MKTME and/or TDX capable systems,
+	 * boot_cpu_data.x86_phys_bits holds the actual physical address
+	 * w/o the KeyID bits, and shadow_phys_bits equals to MAXPHYADDR
+	 * reported by CPUID.  Those bits between are KeyID bits.
+	 */
+	if (boot_cpu_data.x86_phys_bits != kvm_get_shadow_phys_bits())
+		me_mask = rsvd_bits(boot_cpu_data.x86_phys_bits,
+			kvm_get_shadow_phys_bits() - 1);
+	/*
+	 * Unlike SME, host kernel doesn't support setting up any
+	 * MKTME KeyID on Intel platforms.  No memory encryption
+	 * bits should be included into the SPTE.
+	 */
+	kvm_mmu_set_me_spte_mask(0, me_mask);
 }
 
 static struct kvm_x86_init_ops vmx_init_ops __initdata;
···
 		kvm_mmu_set_ept_masks(enable_ept_ad_bits,
 				      cpu_has_vmx_ept_execute_only());
 
+	/*
+	 * Setup shadow_me_value/shadow_me_mask to include MKTME KeyID
+	 * bits to shadow_zero_check.
+	 */
+	vmx_setup_me_spte_mask();
+
 	kvm_configure_mmu(enable_ept, 0, vmx_get_max_tdp_level(),
 			  ept_caps_to_lpage_level(vmx_capability.ept));
 
···
 	.handle_intel_pt_intr = NULL,
 
 	.runtime_ops = &vmx_x86_ops,
+	.pmu_ops = &intel_pmu_ops,
 };
 
 static void vmx_cleanup_l1d_flush(void)

arch/x86/kvm/x86.c

···
 
 const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = {
 	KVM_GENERIC_VCPU_STATS(),
+	STATS_DESC_COUNTER(VCPU, pf_taken),
 	STATS_DESC_COUNTER(VCPU, pf_fixed),
+	STATS_DESC_COUNTER(VCPU, pf_emulate),
+	STATS_DESC_COUNTER(VCPU, pf_spurious),
+	STATS_DESC_COUNTER(VCPU, pf_fast),
+	STATS_DESC_COUNTER(VCPU, pf_mmio_spte_created),
 	STATS_DESC_COUNTER(VCPU, pf_guest),
 	STATS_DESC_COUNTER(VCPU, tlb_flush),
 	STATS_DESC_COUNTER(VCPU, invlpg),
···
 }
 EXPORT_SYMBOL_GPL(kvm_inject_page_fault);
 
+/* Returns true if the page fault was immediately morphed into a VM-Exit. */
 bool kvm_inject_emulated_page_fault(struct kvm_vcpu *vcpu,
 				    struct x86_exception *fault)
 {
···
 		kvm_mmu_invalidate_gva(vcpu, fault_mmu, fault->address,
 				       fault_mmu->root.hpa);
 
+	/*
+	 * A workaround for KVM's bad exception handling.  If KVM injected an
+	 * exception into L2, and L2 encountered a #PF while vectoring the
+	 * injected exception, manually check to see if L1 wants to intercept
+	 * #PF, otherwise queuing the #PF will lead to #DF or a lost exception.
+	 * In all other cases, defer the check to nested_ops->check_events(),
+	 * which will correctly handle priority (this does not).  Note, other
+	 * exceptions, e.g. #GP, are theoretically affected, #PF is simply the
+	 * most problematic, e.g. when L0 and L1 are both intercepting #PF for
+	 * shadow paging.
+	 *
+	 * TODO: Rewrite exception handling to track injected and pending
+	 *       (VM-Exit) exceptions separately.
+	 */
+	if (unlikely(vcpu->arch.exception.injected && is_guest_mode(vcpu)) &&
+	    kvm_x86_ops.nested_ops->handle_page_fault_workaround(vcpu, fault))
+		return true;
+
 	fault_mmu->inject_page_fault(vcpu, fault);
-	return fault->nested_page_fault;
+	return false;
 }
 EXPORT_SYMBOL_GPL(kvm_inject_emulated_page_fault);
 
···
 			wrmsrl(MSR_IA32_XSS, vcpu->arch.ia32_xss);
 	}
 
+#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
 	if (static_cpu_has(X86_FEATURE_PKU) &&
-	    (kvm_read_cr4_bits(vcpu, X86_CR4_PKE) ||
-	     (vcpu->arch.xcr0 & XFEATURE_MASK_PKRU)) &&
-	    vcpu->arch.pkru != vcpu->arch.host_pkru)
+	    vcpu->arch.pkru != vcpu->arch.host_pkru &&
+	    ((vcpu->arch.xcr0 & XFEATURE_MASK_PKRU) ||
+	     kvm_read_cr4_bits(vcpu, X86_CR4_PKE)))
 		write_pkru(vcpu->arch.pkru);
+#endif /* CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS */
 }
 EXPORT_SYMBOL_GPL(kvm_load_guest_xsave_state);
 
···
 	if (vcpu->arch.guest_state_protected)
 		return;
 
+#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
 	if (static_cpu_has(X86_FEATURE_PKU) &&
-	    (kvm_read_cr4_bits(vcpu, X86_CR4_PKE) ||
-	     (vcpu->arch.xcr0 & XFEATURE_MASK_PKRU))) {
+	    ((vcpu->arch.xcr0 & XFEATURE_MASK_PKRU) ||
+	     kvm_read_cr4_bits(vcpu, X86_CR4_PKE))) {
 		vcpu->arch.pkru = rdpkru();
 		if (vcpu->arch.pkru != vcpu->arch.host_pkru)
 			write_pkru(vcpu->arch.host_pkru);
 	}
+#endif /* CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS */
 
 	if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) {
 
···
 	kvm_make_request(KVM_REQ_GLOBAL_CLOCK_UPDATE, vcpu);
 
 	/* we verify if the enable bit is set... */
-	vcpu->arch.pv_time_enabled = false;
-	if (!(system_time & 1))
-		return;
-
-	if (!kvm_gfn_to_hva_cache_init(vcpu->kvm,
-				       &vcpu->arch.pv_time, system_time & ~1ULL,
-				       sizeof(struct pvclock_vcpu_time_info)))
-		vcpu->arch.pv_time_enabled = true;
+	if (system_time & 1) {
+		kvm_gfn_to_pfn_cache_init(vcpu->kvm, &vcpu->arch.pv_time, vcpu,
+					  KVM_HOST_USES_PFN, system_time & ~1ULL,
+					  sizeof(struct pvclock_vcpu_time_info));
+	} else {
+		kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.pv_time);
+	}
 
 	return;
 }
···
 	return data.clock;
 }
 
-static void kvm_setup_pvclock_page(struct kvm_vcpu *v,
-				   struct gfn_to_hva_cache *cache,
-				   unsigned int offset)
+static void kvm_setup_guest_pvclock(struct kvm_vcpu *v,
+				    struct gfn_to_pfn_cache *gpc,
+				    unsigned int offset)
 {
 	struct kvm_vcpu_arch *vcpu = &v->arch;
-	struct pvclock_vcpu_time_info guest_hv_clock;
+	struct pvclock_vcpu_time_info *guest_hv_clock;
+	unsigned long flags;
 
-	if (unlikely(kvm_read_guest_offset_cached(v->kvm, cache,
-		&guest_hv_clock, offset, sizeof(guest_hv_clock))))
-		return;
+	read_lock_irqsave(&gpc->lock, flags);
+	while (!kvm_gfn_to_pfn_cache_check(v->kvm, gpc, gpc->gpa,
+					   offset + sizeof(*guest_hv_clock))) {
+		read_unlock_irqrestore(&gpc->lock, flags);
 
-	/* This VCPU is paused, but it's legal for a guest to read another
+		if (kvm_gfn_to_pfn_cache_refresh(v->kvm, gpc, gpc->gpa,
+						 offset + sizeof(*guest_hv_clock)))
+			return;
+
+		read_lock_irqsave(&gpc->lock, flags);
+	}
+
+	guest_hv_clock = (void *)(gpc->khva + offset);
+
+	/*
+	 * This VCPU is paused, but it's legal for a guest to read another
 	 * VCPU's kvmclock, so we really have to follow the specification where
 	 * it says that version is odd if data is being modified, and even after
 	 * it is consistent.
-	 *
-	 * Version field updates must be kept separate.  This is because
-	 * kvm_write_guest_cached might use a "rep movs" instruction, and
-	 * writes within a string instruction are weakly ordered.  So there
-	 * are three writes overall.
-	 *
-	 * As a small optimization, only write the version field in the first
-	 * and third write.  The vcpu->pv_time cache is still valid, because the
-	 * version field is the first in the struct.
 	 */
-	BUILD_BUG_ON(offsetof(struct pvclock_vcpu_time_info, version) != 0);
 
-	if (guest_hv_clock.version & 1)
-		++guest_hv_clock.version;  /* first time write, random junk */
-
-	vcpu->hv_clock.version = guest_hv_clock.version + 1;
-	kvm_write_guest_offset_cached(v->kvm, cache,
-				      &vcpu->hv_clock, offset,
-				      sizeof(vcpu->hv_clock.version));
-
+	guest_hv_clock->version = vcpu->hv_clock.version = (guest_hv_clock->version + 1) | 1;
 	smp_wmb();
 
 	/* retain PVCLOCK_GUEST_STOPPED if set in guest copy */
-	vcpu->hv_clock.flags |= (guest_hv_clock.flags & PVCLOCK_GUEST_STOPPED);
+	vcpu->hv_clock.flags |= (guest_hv_clock->flags & PVCLOCK_GUEST_STOPPED);
 
 	if (vcpu->pvclock_set_guest_stopped_request) {
 		vcpu->hv_clock.flags |= PVCLOCK_GUEST_STOPPED;
 		vcpu->pvclock_set_guest_stopped_request = false;
 	}
 
-	trace_kvm_pvclock_update(v->vcpu_id, &vcpu->hv_clock);
-
-	kvm_write_guest_offset_cached(v->kvm, cache,
-				      &vcpu->hv_clock, offset,
-				      sizeof(vcpu->hv_clock));
-
+	memcpy(guest_hv_clock, &vcpu->hv_clock, sizeof(*guest_hv_clock));
 	smp_wmb();
 
-	vcpu->hv_clock.version++;
-	kvm_write_guest_offset_cached(v->kvm, cache,
-				     &vcpu->hv_clock, offset,
-				     sizeof(vcpu->hv_clock.version));
+	guest_hv_clock->version = ++vcpu->hv_clock.version;
+
+	mark_page_dirty_in_slot(v->kvm, gpc->memslot, gpc->gpa >> PAGE_SHIFT);
+	read_unlock_irqrestore(&gpc->lock, flags);
+
+	trace_kvm_pvclock_update(v->vcpu_id, &vcpu->hv_clock);
 }
 
 static int kvm_guest_time_update(struct kvm_vcpu *v)
···
 
 	vcpu->hv_clock.flags = pvclock_flags;
 
-	if (vcpu->pv_time_enabled)
-		kvm_setup_pvclock_page(v, &vcpu->pv_time, 0);
-	if (vcpu->xen.vcpu_info_set)
-		kvm_setup_pvclock_page(v, &vcpu->xen.vcpu_info_cache,
-				       offsetof(struct compat_vcpu_info, time));
-	if (vcpu->xen.vcpu_time_info_set)
-		kvm_setup_pvclock_page(v, &vcpu->xen.vcpu_time_info_cache, 0);
+	if (vcpu->pv_time.active)
+		kvm_setup_guest_pvclock(v, &vcpu->pv_time, 0);
+	if (vcpu->xen.vcpu_info_cache.active)
+		kvm_setup_guest_pvclock(v, &vcpu->xen.vcpu_info_cache,
+					offsetof(struct compat_vcpu_info, time));
+	if (vcpu->xen.vcpu_time_info_cache.active)
+		kvm_setup_guest_pvclock(v, &vcpu->xen.vcpu_time_info_cache, 0);
 	kvm_hv_setup_tsc_page(v->kvm, &vcpu->hv_clock);
 	return 0;
 }
···
 
 static void kvmclock_reset(struct kvm_vcpu *vcpu)
 {
-	vcpu->arch.pv_time_enabled = false;
+	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.pv_time);
 	vcpu->arch.time = 0;
 }
 
···
 		r = KVM_XEN_HVM_CONFIG_HYPERCALL_MSR |
 		    KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL |
 		    KVM_XEN_HVM_CONFIG_SHARED_INFO |
-		    KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL;
+		    KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL |
+		    KVM_XEN_HVM_CONFIG_EVTCHN_SEND;
 		if (sched_info_on())
 			r |= KVM_XEN_HVM_CONFIG_RUNSTATE;
 		break;
···
 		r = boot_cpu_has(X86_FEATURE_XSAVE);
 		break;
 	case KVM_CAP_TSC_CONTROL:
+	case KVM_CAP_VM_TSC_CONTROL:
 		r = kvm_has_tsc_control;
 		break;
 	case KVM_CAP_X2APIC_API:
···
  */
 static int kvm_set_guest_paused(struct kvm_vcpu *vcpu)
 {
-	if (!vcpu->arch.pv_time_enabled)
+	if (!vcpu->arch.pv_time.active)
 		return -EINVAL;
 	vcpu->arch.pvclock_set_guest_stopped_request = true;
 	kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
···
 
 	mutex_lock(&kvm->lock);
 	kvm_for_each_vcpu(i, vcpu, kvm) {
-		if (!vcpu->arch.pv_time_enabled)
+		if (!vcpu->arch.pv_time.active)
 			continue;
 
 		ret = kvm_set_guest_paused(vcpu);
···
 		r = kvm_xen_hvm_set_attr(kvm, &xha);
 		break;
 	}
+	case KVM_XEN_HVM_EVTCHN_SEND: {
+		struct kvm_irq_routing_xen_evtchn uxe;
+
+		r = -EFAULT;
+		if (copy_from_user(&uxe, argp, sizeof(uxe)))
+			goto out;
+		r = kvm_xen_hvm_evtchn_send(kvm, &uxe);
+		break;
+	}
 #endif
 	case KVM_SET_CLOCK:
 		r = kvm_vm_ioctl_set_clock(kvm, argp);
···
 	case KVM_GET_CLOCK:
 		r = kvm_vm_ioctl_get_clock(kvm, argp);
 		break;
+	case KVM_SET_TSC_KHZ: {
+		u32 user_tsc_khz;
+
+		r = -EINVAL;
+		user_tsc_khz = (u32)arg;
+
+		if (kvm_has_tsc_control &&
+		    user_tsc_khz >= kvm_max_guest_tsc_khz)
+			goto out;
+
+		if (user_tsc_khz == 0)
+			user_tsc_khz = tsc_khz;
+
+		WRITE_ONCE(kvm->arch.default_tsc_khz, user_tsc_khz);
+		r = 0;
+
+		goto out;
+	}
+	case KVM_GET_TSC_KHZ: {
+		r = READ_ONCE(kvm->arch.default_tsc_khz);
+		goto out;
+	}
 	case KVM_MEMORY_ENCRYPT_OP: {
 		r = -ENOTTY;
 		if (!kvm_x86_ops.mem_enc_ioctl)
···
 				   exception, &write_emultor);
 }
 
-#define CMPXCHG_TYPE(t, ptr, old, new) \
-	(cmpxchg((t *)(ptr), *(t *)(old), *(t *)(new)) == *(t *)(old))
-
-#ifdef CONFIG_X86_64
-#  define CMPXCHG64(ptr, old, new) CMPXCHG_TYPE(u64, ptr, old, new)
-#else
-#  define CMPXCHG64(ptr, old, new) \
-	(cmpxchg64((u64 *)(ptr), *(u64 *)(old), *(u64 *)(new)) == *(u64 *)(old))
-#endif
+#define emulator_try_cmpxchg_user(t, ptr, old, new) \
+	(__try_cmpxchg_user((t __user *)(ptr), (t *)(old), *(t *)(new), efault ## t))
 
 static int emulator_cmpxchg_emulated(struct x86_emulate_ctxt *ctxt,
 				     unsigned long addr,
···
 				     unsigned int bytes,
 				     struct x86_exception *exception)
 {
-	struct kvm_host_map map;
 	struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt);
 	u64 page_line_mask;
+	unsigned long hva;
 	gpa_t gpa;
-	char *kaddr;
-	bool exchanged;
+	int r;
 
 	/* guests cmpxchg8b have to be emulated atomically */
 	if (bytes > 8 || (bytes & (bytes - 1)))
···
 	if (((gpa + bytes - 1) & page_line_mask) != (gpa & page_line_mask))
 		goto emul_write;
 
-	if (kvm_vcpu_map(vcpu, gpa_to_gfn(gpa), &map))
+	hva = kvm_vcpu_gfn_to_hva(vcpu, gpa_to_gfn(gpa));
+	if (kvm_is_error_hva(hva))
 		goto emul_write;
 
-	kaddr = map.hva + offset_in_page(gpa);
+	hva += offset_in_page(gpa);
 
 	switch (bytes) {
 	case 1:
-		exchanged = CMPXCHG_TYPE(u8, kaddr, old, new);
+		r = emulator_try_cmpxchg_user(u8, hva, old, new);
 		break;
 	case 2:
-		exchanged = CMPXCHG_TYPE(u16, kaddr, old, new);
+		r = emulator_try_cmpxchg_user(u16, hva, old, new);
 		break;
 	case 4:
-		exchanged = CMPXCHG_TYPE(u32, kaddr, old, new);
+		r = emulator_try_cmpxchg_user(u32, hva, old, new);
 		break;
 	case 8:
-		exchanged = CMPXCHG64(kaddr, old, new);
+		r = emulator_try_cmpxchg_user(u64, hva, old, new);
 		break;
 	default:
 		BUG();
 	}
 
-	kvm_vcpu_unmap(vcpu, &map, true);
-
-	if (!exchanged)
+	if (r < 0)
+		return X86EMUL_UNHANDLEABLE;
+	if (r)
 		return X86EMUL_CMPXCHG_FAILED;
 
 	kvm_page_track_write(vcpu, gpa, new, bytes);
···
 	    WARN_ON_ONCE(!(emulation_type & EMULTYPE_PF)))
 		return false;
 
-	if (!vcpu->arch.mmu->direct_map) {
+	if (!vcpu->arch.mmu->root_role.direct) {
 		/*
 		 * Write permission should be allowed since only
 		 * write access need to be emulated.
···
 	kvm_release_pfn_clean(pfn);
 
 	/* The instructions are well-emulated on direct mmu. */
-	if (vcpu->arch.mmu->direct_map) {
+	if (vcpu->arch.mmu->root_role.direct) {
 		unsigned int indirect_shadow_pages;
 
 		write_lock(&vcpu->kvm->mmu_lock);
···
 	vcpu->arch.last_retry_eip = ctxt->eip;
 	vcpu->arch.last_retry_addr = cr2_or_gpa;
 
-	if (!vcpu->arch.mmu->direct_map)
+	if (!vcpu->arch.mmu->root_role.direct)
 		gpa = kvm_mmu_gva_to_gpa_write(vcpu, cr2_or_gpa, NULL);
 
 	kvm_mmu_unprotect_page(vcpu->kvm, gpa_to_gfn(gpa));
···
 }
 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)) {
···
 }
 
 /*
- * 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);
 
···
 
 	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);
···
 		ctxt->exception.address = cr2_or_gpa;
 
 		/* With shadow page tables, cr2 contains a GVA or nGPA. */
-		if (vcpu->arch.mmu->direct_map) {
+		if (vcpu->arch.mmu->root_role.direct) {
 			ctxt->gpa_available = true;
 			ctxt->gpa_val = cr2_or_gpa;
 		}
···
 
 static void kvm_timer_init(void)
 {
-	max_tsc_khz = tsc_khz;
-
 	if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) {
-#ifdef CONFIG_CPU_FREQ
-		struct cpufreq_policy *policy;
-		int cpu;
+		max_tsc_khz = tsc_khz;
 
-		cpu = get_cpu();
-		policy = cpufreq_cpu_get(cpu);
-		if (policy) {
-			if (policy->cpuinfo.max_freq)
-				max_tsc_khz = policy->cpuinfo.max_freq;
-			cpufreq_cpu_put(policy);
+		if (IS_ENABLED(CONFIG_CPU_FREQ)) {
+			struct cpufreq_policy *policy;
+			int cpu;
+
+			cpu = get_cpu();
+			policy = cpufreq_cpu_get(cpu);
+			if (policy) {
+				if (policy->cpuinfo.max_freq)
+					max_tsc_khz = policy->cpuinfo.max_freq;
+				cpufreq_cpu_put(policy);
+			}
+			put_cpu();
 		}
-		put_cpu();
-#endif
 		cpufreq_register_notifier(&kvmclock_cpufreq_notifier_block,
 					  CPUFREQ_TRANSITION_NOTIFIER);
 	}
···
 }
 EXPORT_SYMBOL_GPL(kvm_apicv_activated);
 
+bool kvm_vcpu_apicv_activated(struct kvm_vcpu *vcpu)
+{
+	ulong vm_reasons = READ_ONCE(vcpu->kvm->arch.apicv_inhibit_reasons);
+	ulong vcpu_reasons = static_call(kvm_x86_vcpu_get_apicv_inhibit_reasons)(vcpu);
+
+	return (vm_reasons | vcpu_reasons) == 0;
+}
+EXPORT_SYMBOL_GPL(kvm_vcpu_apicv_activated);
 
 static void set_or_clear_apicv_inhibit(unsigned long *inhibits,
 				       enum kvm_apicv_inhibit reason, bool set)
···
 	struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt);
 	char instruction[3];
 	unsigned long rip = kvm_rip_read(vcpu);
+
+	/*
+	 * If the quirk is disabled, synthesize a #UD and let the guest pick up
+	 * the pieces.
+	 */
+	if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_FIX_HYPERCALL_INSN)) {
+		ctxt->exception.error_code_valid = false;
+		ctxt->exception.vector = UD_VECTOR;
+		ctxt->have_exception = true;
+		return X86EMUL_PROPAGATE_FAULT;
+	}
 
 	static_call(kvm_x86_patch_hypercall)(vcpu, instruction);
 
···
 
 	down_read(&vcpu->kvm->arch.apicv_update_lock);
 
-	activate = kvm_apicv_activated(vcpu->kvm);
+	activate = kvm_vcpu_apicv_activated(vcpu);
+
 	if (vcpu->arch.apicv_active == activate)
 		goto out;
 
···
 		 * per-VM state, and responsing vCPUs must wait for the update
 		 * to complete before servicing KVM_REQ_APICV_UPDATE.
 		 */
-		WARN_ON_ONCE(kvm_apicv_activated(vcpu->kvm) != kvm_vcpu_apicv_active(vcpu));
+		WARN_ON_ONCE(kvm_vcpu_apicv_activated(vcpu) != kvm_vcpu_apicv_active(vcpu));
 
 		exit_fastpath = static_call(kvm_x86_vcpu_run)(vcpu);
 		if (likely(exit_fastpath != EXIT_FASTPATH_REENTER_GUEST))
···
 	 * acceptable for all known use cases.
 	 */
 	guest_timing_exit_irqoff();
-
-	if (lapic_in_kernel(vcpu)) {
-		s64 delta = vcpu->arch.apic->lapic_timer.advance_expire_delta;
-		if (delta != S64_MIN) {
-			trace_kvm_wait_lapic_expire(vcpu->vcpu_id, delta);
-			vcpu->arch.apic->lapic_timer.advance_expire_delta = S64_MIN;
-		}
-	}
 
 	local_irq_enable();
 	preempt_enable();
···
 			break;
 
 		kvm_clear_request(KVM_REQ_UNBLOCK, vcpu);
+		if (kvm_xen_has_pending_events(vcpu))
+			kvm_xen_inject_pending_events(vcpu);
+
 		if (kvm_cpu_has_pending_timer(vcpu))
 			kvm_inject_pending_timer_irqs(vcpu);
 
···
 
 	vcpu->arch.arch_capabilities = kvm_get_arch_capabilities();
 	vcpu->arch.msr_platform_info = MSR_PLATFORM_INFO_CPUID_FAULT;
+	kvm_xen_init_vcpu(vcpu);
 	kvm_vcpu_mtrr_init(vcpu);
 	vcpu_load(vcpu);
-	kvm_set_tsc_khz(vcpu, max_tsc_khz);
+	kvm_set_tsc_khz(vcpu, vcpu->kvm->arch.default_tsc_khz);
 	kvm_vcpu_reset(vcpu, false);
 	kvm_init_mmu(vcpu);
 	vcpu_put(vcpu);
···
 	free_cpumask_var(vcpu->arch.wbinvd_dirty_mask);
 	fpu_free_guest_fpstate(&vcpu->arch.guest_fpu);
 
+	kvm_xen_destroy_vcpu(vcpu);
 	kvm_hv_vcpu_uninit(vcpu);
 	kvm_pmu_destroy(vcpu);
 	kfree(vcpu->arch.mce_banks);
···
 	drop_user_return_notifiers();
 }
 
+static inline void kvm_ops_update(struct kvm_x86_init_ops *ops)
+{
+	memcpy(&kvm_x86_ops, ops->runtime_ops, sizeof(kvm_x86_ops));
+
+#define __KVM_X86_OP(func) \
+	static_call_update(kvm_x86_##func, kvm_x86_ops.func);
+#define KVM_X86_OP(func) \
+	WARN_ON(!kvm_x86_ops.func); __KVM_X86_OP(func)
+#define KVM_X86_OP_OPTIONAL __KVM_X86_OP
+#define KVM_X86_OP_OPTIONAL_RET0(func) \
+	static_call_update(kvm_x86_##func, (void *)kvm_x86_ops.func ? : \
+					   (void *)__static_call_return0);
+#include <asm/kvm-x86-ops.h>
+#undef __KVM_X86_OP
+
+	kvm_pmu_ops_update(ops->pmu_ops);
+}
+
 int kvm_arch_hardware_setup(void *opaque)
 {
 	struct kvm_x86_init_ops *ops = opaque;
···
 	if (r != 0)
 		return r;
 
-	memcpy(&kvm_x86_ops, ops->runtime_ops, sizeof(kvm_x86_ops));
-	kvm_ops_static_call_update();
+	kvm_ops_update(ops);
 
 	kvm_register_perf_callbacks(ops->handle_intel_pt_intr);
 
···
 	pvclock_update_vm_gtod_copy(kvm);
 	raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags);
 
+	kvm->arch.default_tsc_khz = max_tsc_khz ? : tsc_khz;
 	kvm->arch.guest_can_read_msr_platform_info = true;
 	kvm->arch.enable_pmu = enable_pmu;
 
···
 	vcpu_put(vcpu);
 }
 
-static void kvm_free_vcpus(struct kvm *kvm)
+static void kvm_unload_vcpu_mmus(struct kvm *kvm)
 {
 	unsigned long i;
 	struct kvm_vcpu *vcpu;
 
-	/*
-	 * Unpin any mmu pages first.
-	 */
 	kvm_for_each_vcpu(i, vcpu, kvm) {
 		kvm_clear_async_pf_completion_queue(vcpu);
 		kvm_unload_vcpu_mmu(vcpu);
 	}
-
-	kvm_destroy_vcpus(kvm);
 }
 
 void kvm_arch_sync_events(struct kvm *kvm)
···
 		__x86_set_memory_region(kvm, TSS_PRIVATE_MEMSLOT, 0, 0);
 		mutex_unlock(&kvm->slots_lock);
 	}
+	kvm_unload_vcpu_mmus(kvm);
 	static_call_cond(kvm_x86_vm_destroy)(kvm);
 	kvm_free_msr_filter(srcu_dereference_check(kvm->arch.msr_filter, &kvm->srcu, 1));
 	kvm_pic_destroy(kvm);
 	kvm_ioapic_destroy(kvm);
-	kvm_free_vcpus(kvm);
+	kvm_destroy_vcpus(kvm);
 	kvfree(rcu_dereference_check(kvm->arch.apic_map, 1));
 	kfree(srcu_dereference_check(kvm->arch.pmu_event_filter, &kvm->srcu, 1));
 	kvm_mmu_uninit_vm(kvm);
···
 	    kvm_x86_ops.nested_ops->hv_timer_pending(vcpu))
 		return true;
 
+	if (kvm_xen_has_pending_events(vcpu))
+		return true;
+
+	if (kvm_test_request(KVM_REQ_TRIPLE_FAULT, vcpu))
+		return true;
+
 	return false;
 }
 
···
 	kvm_make_request(KVM_REQ_EVENT, vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_set_rflags);
-
-void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
-{
-	int r;
-
-	if ((vcpu->arch.mmu->direct_map != work->arch.direct_map) ||
-	      work->wakeup_all)
-		return;
-
-	r = kvm_mmu_reload(vcpu);
-	if (unlikely(r))
-		return;
-
-	if (!vcpu->arch.mmu->direct_map &&
-	      work->arch.cr3 != vcpu->arch.mmu->get_guest_pgd(vcpu))
-		return;
-
-	kvm_mmu_do_page_fault(vcpu, work->cr2_or_gpa, 0, true);
-}
 
 static inline u32 kvm_async_pf_hash_fn(gfn_t gfn)
 {
···
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_unaccelerated_access);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_incomplete_ipi);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_ga_log);
+EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_kick_vcpu_slowpath);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_apicv_accept_irq);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_enter);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_exit);

arch/x86/kvm/xen.c

···
 #include "x86.h"
 #include "xen.h"
 #include "hyperv.h"
+#include "lapic.h"
 
+#include <linux/eventfd.h>
 #include <linux/kvm_host.h>
 #include <linux/sched/stat.h>
 
 #include <trace/events/kvm.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/vcpu.h>
+#include <xen/interface/version.h>
 #include <xen/interface/event_channel.h>
+#include <xen/interface/sched.h>
 
 #include "trace.h"
+
+static int kvm_xen_set_evtchn(struct kvm_xen_evtchn *xe, struct kvm *kvm);
+static int kvm_xen_setattr_evtchn(struct kvm *kvm, struct kvm_xen_hvm_attr *data);
+static bool kvm_xen_hcall_evtchn_send(struct kvm_vcpu *vcpu, u64 param, u64 *r);
 
 DEFINE_STATIC_KEY_DEFERRED_FALSE(kvm_xen_enabled, HZ);
 
···
 	return ret;
 }
 
+void kvm_xen_inject_timer_irqs(struct kvm_vcpu *vcpu)
+{
+	if (atomic_read(&vcpu->arch.xen.timer_pending) > 0) {
+		struct kvm_xen_evtchn e;
+
+		e.vcpu_id = vcpu->vcpu_id;
+		e.vcpu_idx = vcpu->vcpu_idx;
+		e.port = vcpu->arch.xen.timer_virq;
+		e.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL;
+
+		kvm_xen_set_evtchn(&e, vcpu->kvm);
+
+		vcpu->arch.xen.timer_expires = 0;
+		atomic_set(&vcpu->arch.xen.timer_pending, 0);
+	}
+}
+
+static enum hrtimer_restart xen_timer_callback(struct hrtimer *timer)
+{
+	struct kvm_vcpu *vcpu = container_of(timer, struct kvm_vcpu,
+					     arch.xen.timer);
+	if (atomic_read(&vcpu->arch.xen.timer_pending))
+		return HRTIMER_NORESTART;
+
+	atomic_inc(&vcpu->arch.xen.timer_pending);
+	kvm_make_request(KVM_REQ_UNBLOCK, vcpu);
+	kvm_vcpu_kick(vcpu);
+
+	return HRTIMER_NORESTART;
+}
+
+static void kvm_xen_start_timer(struct kvm_vcpu *vcpu, u64 guest_abs, s64 delta_ns)
+{
+	atomic_set(&vcpu->arch.xen.timer_pending, 0);
+	vcpu->arch.xen.timer_expires = guest_abs;
+
+	if (delta_ns <= 0) {
+		xen_timer_callback(&vcpu->arch.xen.timer);
+	} else {
+		ktime_t ktime_now = ktime_get();
+		hrtimer_start(&vcpu->arch.xen.timer,
+			      ktime_add_ns(ktime_now, delta_ns),
+			      HRTIMER_MODE_ABS_HARD);
+	}
+}
+
+static void kvm_xen_stop_timer(struct kvm_vcpu *vcpu)
+{
+	hrtimer_cancel(&vcpu->arch.xen.timer);
+	vcpu->arch.xen.timer_expires = 0;
+	atomic_set(&vcpu->arch.xen.timer_pending, 0);
+}
+
+static void kvm_xen_init_timer(struct kvm_vcpu *vcpu)
+{
+	hrtimer_init(&vcpu->arch.xen.timer, CLOCK_MONOTONIC,
+		     HRTIMER_MODE_ABS_HARD);
+	vcpu->arch.xen.timer.function = xen_timer_callback;
+}
+
 static void kvm_xen_update_runstate(struct kvm_vcpu *v, int state)
 {
 	struct kvm_vcpu_xen *vx = &v->arch.xen;
···
 void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
 {
 	struct kvm_vcpu_xen *vx = &v->arch.xen;
-	struct gfn_to_hva_cache *ghc = &vx->runstate_cache;
-	struct kvm_memslots *slots = kvm_memslots(v->kvm);
-	bool atomic = (state == RUNSTATE_runnable);
-	uint64_t state_entry_time;
-	int __user *user_state;
-	uint64_t __user *user_times;
+	struct gfn_to_pfn_cache *gpc = &vx->runstate_cache;
+	uint64_t *user_times;
+	unsigned long flags;
+	size_t user_len;
+	int *user_state;
 
 	kvm_xen_update_runstate(v, state);
 
-	if (!vx->runstate_set)
+	if (!vx->runstate_cache.active)
 		return;
 
-	if (unlikely(slots->generation != ghc->generation || kvm_is_error_hva(ghc->hva)) &&
-	    kvm_gfn_to_hva_cache_init(v->kvm, ghc, ghc->gpa, ghc->len))
-		return;
+	if (IS_ENABLED(CONFIG_64BIT) && v->kvm->arch.xen.long_mode)
+		user_len = sizeof(struct vcpu_runstate_info);
+	else
+		user_len = sizeof(struct compat_vcpu_runstate_info);
 
-	/* We made sure it fits in a single page */
-	BUG_ON(!ghc->memslot);
+	read_lock_irqsave(&gpc->lock, flags);
+	while (!kvm_gfn_to_pfn_cache_check(v->kvm, gpc, gpc->gpa,
+					   user_len)) {
+		read_unlock_irqrestore(&gpc->lock, flags);
 
-	if (atomic)
-		pagefault_disable();
+		/* When invoked from kvm_sched_out() we cannot sleep */
+		if (state == RUNSTATE_runnable)
+			return;
+
+		if (kvm_gfn_to_pfn_cache_refresh(v->kvm, gpc, gpc->gpa, user_len))
+			return;
+
+		read_lock_irqsave(&gpc->lock, flags);
+	}
 
 	/*
 	 * The only difference between 32-bit and 64-bit versions of the
···
 	 */
 	BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state) != 0);
 	BUILD_BUG_ON(offsetof(struct compat_vcpu_runstate_info, state) != 0);
-	user_state = (int __user *)ghc->hva;
-
 	BUILD_BUG_ON(sizeof(struct compat_vcpu_runstate_info) != 0x2c);
-
-	user_times = (uint64_t __user *)(ghc->hva +
-					 offsetof(struct compat_vcpu_runstate_info,
-						  state_entry_time));
 #ifdef CONFIG_X86_64
 	BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state_entry_time) !=
 		     offsetof(struct compat_vcpu_runstate_info, state_entry_time) + 4);
 	BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, time) !=
 		     offsetof(struct compat_vcpu_runstate_info, time) + 4);
-
-	if (v->kvm->arch.xen.long_mode)
-		user_times = (uint64_t __user *)(ghc->hva +
-						 offsetof(struct vcpu_runstate_info,
-							  state_entry_time));
 #endif
+
+	user_state = gpc->khva;
+
+	if (IS_ENABLED(CONFIG_64BIT) && v->kvm->arch.xen.long_mode)
+		user_times = gpc->khva + offsetof(struct vcpu_runstate_info,
+						  state_entry_time);
+	else
+		user_times = gpc->khva + offsetof(struct compat_vcpu_runstate_info,
+						  state_entry_time);
+
 	/*
 	 * First write the updated state_entry_time at the appropriate
 	 * location determined by 'offset'.
 	 */
-	state_entry_time = vx->runstate_entry_time;
-	state_entry_time |= XEN_RUNSTATE_UPDATE;
-
 	BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, state_entry_time) !=
-		     sizeof(state_entry_time));
+		     sizeof(user_times[0]));
 	BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state_entry_time) !=
-		     sizeof(state_entry_time));
+		     sizeof(user_times[0]));
 
-	if (__put_user(state_entry_time, user_times))
-		goto out;
+	user_times[0] = vx->runstate_entry_time | XEN_RUNSTATE_UPDATE;
 	smp_wmb();
 
 	/*
···
 	BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state) !=
 		     sizeof(vx->current_runstate));
 
-	if (__put_user(vx->current_runstate, user_state))
-		goto out;
+	*user_state = vx->current_runstate;
 
 	/*
 	 * Write the actual runstate times immediately after the
···
 	BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, time) !=
 		     sizeof(vx->runstate_times));
 
-	if (__copy_to_user(user_times + 1, vx->runstate_times, sizeof(vx->runstate_times)))
-		goto out;
+	memcpy(user_times + 1, vx->runstate_times, sizeof(vx->runstate_times));
 	smp_wmb();
 
 	/*
 	 * Finally, clear the XEN_RUNSTATE_UPDATE bit in the guest's
 	 * runstate_entry_time field.
 	 */
-	state_entry_time &= ~XEN_RUNSTATE_UPDATE;
-	__put_user(state_entry_time, user_times);
+	user_times[0] &= ~XEN_RUNSTATE_UPDATE;
 	smp_wmb();
 
- out:
-	mark_page_dirty_in_slot(v->kvm, ghc->memslot, ghc->gpa >> PAGE_SHIFT);
+	read_unlock_irqrestore(&gpc->lock, flags);
 
-	if (atomic)
-		pagefault_enable();
+	mark_page_dirty_in_slot(v->kvm, gpc->memslot, gpc->gpa >> PAGE_SHIFT);
+}
+
+static void kvm_xen_inject_vcpu_vector(struct kvm_vcpu *v)
+{
+	struct kvm_lapic_irq irq = { };
+	int r;
+
+	irq.dest_id = v->vcpu_id;
+	irq.vector = v->arch.xen.upcall_vector;
+	irq.dest_mode = APIC_DEST_PHYSICAL;
+	irq.shorthand = APIC_DEST_NOSHORT;
+	irq.delivery_mode = APIC_DM_FIXED;
+	irq.level = 1;
+
+	/* The fast version will always work for physical unicast */
+	WARN_ON_ONCE(!kvm_irq_delivery_to_apic_fast(v->kvm, NULL, &irq, &r, NULL));
+}
+
+/*
+ * On event channel delivery, the vcpu_info may not have been accessible.
+ * In that case, there are bits in vcpu->arch.xen.evtchn_pending_sel which
+ * need to be marked into the vcpu_info (and evtchn_upcall_pending set).
+ * Do so now that we can sleep in the context of the vCPU to bring the
+ * page in, and refresh the pfn cache for it.
+ */
+void kvm_xen_inject_pending_events(struct kvm_vcpu *v)
+{
+	unsigned long evtchn_pending_sel = READ_ONCE(v->arch.xen.evtchn_pending_sel);
+	struct gfn_to_pfn_cache *gpc = &v->arch.xen.vcpu_info_cache;
+	unsigned long flags;
+
+	if (!evtchn_pending_sel)
+		return;
+
+	/*
+	 * Yes, this is an open-coded loop. But that's just what put_user()
+	 * does anyway. Page it in and retry the instruction. We're just a
+	 * little more honest about it.
+	 */
+	read_lock_irqsave(&gpc->lock, flags);
+	while (!kvm_gfn_to_pfn_cache_check(v->kvm, gpc, gpc->gpa,
+					   sizeof(struct vcpu_info))) {
+		read_unlock_irqrestore(&gpc->lock, flags);
+
+		if (kvm_gfn_to_pfn_cache_refresh(v->kvm, gpc, gpc->gpa,
+						 sizeof(struct vcpu_info)))
+			return;
+
+		read_lock_irqsave(&gpc->lock, flags);
+	}
+
+	/* Now gpc->khva is a valid kernel address for the vcpu_info */
+	if (IS_ENABLED(CONFIG_64BIT) && v->kvm->arch.xen.long_mode) {
+		struct vcpu_info *vi = gpc->khva;
+
+		asm volatile(LOCK_PREFIX "orq %0, %1\n"
+			     "notq %0\n"
+			     LOCK_PREFIX "andq %0, %2\n"
+			     : "=r" (evtchn_pending_sel),
+			       "+m" (vi->evtchn_pending_sel),
+			       "+m" (v->arch.xen.evtchn_pending_sel)
+			     : "0" (evtchn_pending_sel));
+		WRITE_ONCE(vi->evtchn_upcall_pending, 1);
+	} else {
+		u32 evtchn_pending_sel32 = evtchn_pending_sel;
+		struct compat_vcpu_info *vi = gpc->khva;
+
+		asm volatile(LOCK_PREFIX "orl %0, %1\n"
+			     "notl %0\n"
+			     LOCK_PREFIX "andl %0, %2\n"
+			     : "=r" (evtchn_pending_sel32),
+			       "+m" (vi->evtchn_pending_sel),
+			       "+m" (v->arch.xen.evtchn_pending_sel)
+			     : "0" (evtchn_pending_sel32));
+		WRITE_ONCE(vi->evtchn_upcall_pending, 1);
+	}
+	read_unlock_irqrestore(&gpc->lock, flags);
+
+	/* For the per-vCPU lapic vector, deliver it as MSI. */
+	if (v->arch.xen.upcall_vector)
+		kvm_xen_inject_vcpu_vector(v);
+
+	mark_page_dirty_in_slot(v->kvm, gpc->memslot, gpc->gpa >> PAGE_SHIFT);
 }
 
 int __kvm_xen_has_interrupt(struct kvm_vcpu *v)
 {
-	unsigned long evtchn_pending_sel = READ_ONCE(v->arch.xen.evtchn_pending_sel);
-	bool atomic = in_atomic() || !task_is_running(current);
-	int err;
+	struct gfn_to_pfn_cache *gpc = &v->arch.xen.vcpu_info_cache;
+	unsigned long flags;
 	u8 rc = 0;
 
 	/*
 	 * If the global upcall vector (HVMIRQ_callback_vector) is set and
 	 * the vCPU's evtchn_upcall_pending flag is set, the IRQ is pending.
 	 */
-	struct gfn_to_hva_cache *ghc = &v->arch.xen.vcpu_info_cache;
-	struct kvm_memslots *slots = kvm_memslots(v->kvm);
-	bool ghc_valid = slots->generation == ghc->generation &&
-		!kvm_is_error_hva(ghc->hva) && ghc->memslot;
-
-	unsigned int offset = offsetof(struct vcpu_info, evtchn_upcall_pending);
 
 	/* No need for compat handling here */
 	BUILD_BUG_ON(offsetof(struct vcpu_info, evtchn_upcall_pending) !=
···
 	BUILD_BUG_ON(sizeof(rc) !=
 		     sizeof_field(struct compat_vcpu_info, evtchn_upcall_pending));
 
-	/*
-	 * For efficiency, this mirrors the checks for using the valid
-	 * cache in kvm_read_guest_offset_cached(), but just uses
-	 * __get_user() instead. And falls back to the slow path.
-	 */
-	if (!evtchn_pending_sel && ghc_valid) {
-		/* Fast path */
-		pagefault_disable();
-		err = __get_user(rc, (u8 __user *)ghc->hva + offset);
-		pagefault_enable();
-		if (!err)
-			return rc;
-	}
+	read_lock_irqsave(&gpc->lock, flags);
+	while (!kvm_gfn_to_pfn_cache_check(v->kvm, gpc, gpc->gpa,
+					   sizeof(struct vcpu_info))) {
+		read_unlock_irqrestore(&gpc->lock, flags);
 
-	/* Slow path */
+		/*
+		 * This function gets called from kvm_vcpu_block() after setting the
+		 * task to TASK_INTERRUPTIBLE, to see if it needs to wake immediately
+		 * from a HLT. So we really mustn't sleep. If the page ended up absent
+		 * at that point, just return 1 in order to trigger an immediate wake,
+		 * and we'll end up getting called again from a context where we *can*
+		 * fault in the page and wait for it.
+		 */
+		if (in_atomic() || !task_is_running(current))
+			return 1;
 
-	/*
-	 * This function gets called from kvm_vcpu_block() after setting the
-	 * task to TASK_INTERRUPTIBLE, to see if it needs to wake immediately
-	 * from a HLT. So we really mustn't sleep. If the page ended up absent
-	 * at that point, just return 1 in order to trigger an immediate wake,
-	 * and we'll end up getting called again from a context where we *can*
-	 * fault in the page and wait for it.
-	 */
-	if (atomic)
-		return 1;
-
-	if (!ghc_valid) {
-		err = kvm_gfn_to_hva_cache_init(v->kvm, ghc, ghc->gpa, ghc->len);
-		if (err || !ghc->memslot) {
+		if (kvm_gfn_to_pfn_cache_refresh(v->kvm, gpc, gpc->gpa,
+						 sizeof(struct vcpu_info))) {
 			/*
 			 * If this failed, userspace has screwed up the
 			 * vcpu_info mapping. No interrupts for you.
 			 */
 			return 0;
 		}
+		read_lock_irqsave(&gpc->lock, flags);
 	}
 
-	/*
-	 * Now we have a valid (protected by srcu) userspace HVA in
-	 * ghc->hva which points to the struct vcpu_info. If there
-	 * are any bits in the in-kernel evtchn_pending_sel then
-	 * we need to write those to the guest vcpu_info and set
-	 * its evtchn_upcall_pending flag. If there aren't any bits
-	 * to add, we only want to *check* evtchn_upcall_pending.
-	 */
-	if (evtchn_pending_sel) {
-		bool long_mode = v->kvm->arch.xen.long_mode;
-
-		if (!user_access_begin((void __user *)ghc->hva, sizeof(struct vcpu_info)))
-			return 0;
-
-		if (IS_ENABLED(CONFIG_64BIT) && long_mode) {
-			struct vcpu_info __user *vi = (void __user *)ghc->hva;
-
-			/* Attempt to set the evtchn_pending_sel bits in the
-			 * guest, and if that succeeds then clear the same
-			 * bits in the in-kernel version. */
-			asm volatile("1:\t" LOCK_PREFIX "orq %0, %1\n"
-				     "\tnotq %0\n"
-				     "\t" LOCK_PREFIX "andq %0, %2\n"
-				     "2:\n"
-				     _ASM_EXTABLE_UA(1b, 2b)
-				     : "=r" (evtchn_pending_sel),
-				       "+m" (vi->evtchn_pending_sel),
-				       "+m" (v->arch.xen.evtchn_pending_sel)
-				     : "0" (evtchn_pending_sel));
-		} else {
-			struct compat_vcpu_info __user *vi = (void __user *)ghc->hva;
-			u32 evtchn_pending_sel32 = evtchn_pending_sel;
-
-			/* Attempt to set the evtchn_pending_sel bits in the
-			 * guest, and if that succeeds then clear the same
-			 * bits in the in-kernel version. */
-			asm volatile("1:\t" LOCK_PREFIX "orl %0, %1\n"
-				     "\tnotl %0\n"
-				     "\t" LOCK_PREFIX "andl %0, %2\n"
-				     "2:\n"
-				     _ASM_EXTABLE_UA(1b, 2b)
-				     : "=r" (evtchn_pending_sel32),
-				       "+m" (vi->evtchn_pending_sel),
-				       "+m" (v->arch.xen.evtchn_pending_sel)
-				     : "0" (evtchn_pending_sel32));
-		}
-		rc = 1;
-		unsafe_put_user(rc, (u8 __user *)ghc->hva + offset, err);
-
-	err:
-		user_access_end();
-
-		mark_page_dirty_in_slot(v->kvm, ghc->memslot, ghc->gpa >> PAGE_SHIFT);
-	} else {
-		__get_user(rc, (u8 __user *)ghc->hva + offset);
-	}
-
+	rc = ((struct vcpu_info *)gpc->khva)->evtchn_upcall_pending;
+	read_unlock_irqrestore(&gpc->lock, flags);
 	return rc;
 }
 
···
 {
 	int r = -ENOENT;
 
-	mutex_lock(&kvm->lock);
 
 	switch (data->type) {
 	case KVM_XEN_ATTR_TYPE_LONG_MODE:
 		if (!IS_ENABLED(CONFIG_64BIT) && data->u.long_mode) {
 			r = -EINVAL;
 		} else {
+			mutex_lock(&kvm->lock);
 			kvm->arch.xen.long_mode = !!data->u.long_mode;
+			mutex_unlock(&kvm->lock);
 			r = 0;
 		}
 		break;
 
 	case KVM_XEN_ATTR_TYPE_SHARED_INFO:
+		mutex_lock(&kvm->lock);
 		r = kvm_xen_shared_info_init(kvm, data->u.shared_info.gfn);
+		mutex_unlock(&kvm->lock);
 		break;
 
 	case KVM_XEN_ATTR_TYPE_UPCALL_VECTOR:
 		if (data->u.vector && data->u.vector < 0x10)
 			r = -EINVAL;
 		else {
+			mutex_lock(&kvm->lock);
 			kvm->arch.xen.upcall_vector = data->u.vector;
+			mutex_unlock(&kvm->lock);
 			r = 0;
 		}
+		break;
+
+	case KVM_XEN_ATTR_TYPE_EVTCHN:
+		r = kvm_xen_setattr_evtchn(kvm, data);
+		break;
+
+	case KVM_XEN_ATTR_TYPE_XEN_VERSION:
+		mutex_lock(&kvm->lock);
+		kvm->arch.xen.xen_version = data->u.xen_version;
+		mutex_unlock(&kvm->lock);
+		r = 0;
 		break;
 
 	default:
 		break;
 	}
 
-	mutex_unlock(&kvm->lock);
 	return r;
 }
 
···
 		r = 0;
 		break;
 
+	case KVM_XEN_ATTR_TYPE_XEN_VERSION:
+		data->u.xen_version = kvm->arch.xen.xen_version;
+		r = 0;
+		break;
+
 	default:
 		break;
 	}
···
 			     offsetof(struct compat_vcpu_info, time));
 
 		if (data->u.gpa == GPA_INVALID) {
-			vcpu->arch.xen.vcpu_info_set = false;
+			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.xen.vcpu_info_cache);
 			r = 0;
 			break;
 		}
 
-		/* It must fit within a single page */
-		if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct vcpu_info) > PAGE_SIZE) {
-			r = -EINVAL;
-			break;
-		}
-
-		r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
+		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
 					      &vcpu->arch.xen.vcpu_info_cache,
-					      data->u.gpa,
+					      NULL, KVM_HOST_USES_PFN, data->u.gpa,
 					      sizeof(struct vcpu_info));
-		if (!r) {
-			vcpu->arch.xen.vcpu_info_set = true;
+		if (!r)
 			kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
-		}
+
 		break;
 
 	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO:
 		if (data->u.gpa == GPA_INVALID) {
-			vcpu->arch.xen.vcpu_time_info_set = false;
+			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+						     &vcpu->arch.xen.vcpu_time_info_cache);
 			r = 0;
 			break;
 		}
 
-		/* It must fit within a single page */
-		if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct pvclock_vcpu_time_info) > PAGE_SIZE) {
-			r = -EINVAL;
-			break;
-		}
-
-		r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
+		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
 					      &vcpu->arch.xen.vcpu_time_info_cache,
-					      data->u.gpa,
+					      NULL, KVM_HOST_USES_PFN, data->u.gpa,
 					      sizeof(struct pvclock_vcpu_time_info));
-		if (!r) {
-			vcpu->arch.xen.vcpu_time_info_set = true;
+		if (!r)
 			kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
-		}
 		break;
 
 	case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR:
···
 			break;
 		}
 		if (data->u.gpa == GPA_INVALID) {
-			vcpu->arch.xen.runstate_set = false;
+			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+						     &vcpu->arch.xen.runstate_cache);
 			r = 0;
 			break;
 		}
 
-		/* It must fit within a single page */
-		if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct vcpu_runstate_info) > PAGE_SIZE) {
-			r = -EINVAL;
-			break;
-		}
-
-		r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
+		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
 					      &vcpu->arch.xen.runstate_cache,
-					      data->u.gpa,
+					      NULL, KVM_HOST_USES_PFN, data->u.gpa,
 					      sizeof(struct vcpu_runstate_info));
-		if (!r) {
-			vcpu->arch.xen.runstate_set = true;
-		}
 		break;
 
 	case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT:
···
 		r = 0;
 		break;
 
+	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID:
+		if (data->u.vcpu_id >= KVM_MAX_VCPUS)
+			r = -EINVAL;
+		else {
+			vcpu->arch.xen.vcpu_id = data->u.vcpu_id;
+			r = 0;
+		}
+		break;
+
+	case KVM_XEN_VCPU_ATTR_TYPE_TIMER:
+		if (data->u.timer.port) {
+			if (data->u.timer.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL) {
+				r = -EINVAL;
+				break;
+			}
+			vcpu->arch.xen.timer_virq = data->u.timer.port;
+			kvm_xen_init_timer(vcpu);
+
+			/* Restart the timer if it's set */
+			if (data->u.timer.expires_ns)
+				kvm_xen_start_timer(vcpu, data->u.timer.expires_ns,
+						    data->u.timer.expires_ns -
+						    get_kvmclock_ns(vcpu->kvm));
+		} else if (kvm_xen_timer_enabled(vcpu)) {
+			kvm_xen_stop_timer(vcpu);
+			vcpu->arch.xen.timer_virq = 0;
+		}
+
+		r = 0;
+		break;
+
+	case KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR:
+		if (data->u.vector && data->u.vector < 0x10)
+			r = -EINVAL;
+		else {
+			vcpu->arch.xen.upcall_vector = data->u.vector;
+			r = 0;
+		}
+		break;
+
 	default:
 		break;
 	}
···
 
 	switch (data->type) {
 	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO:
-		if (vcpu->arch.xen.vcpu_info_set)
+		if (vcpu->arch.xen.vcpu_info_cache.active)
 			data->u.gpa = vcpu->arch.xen.vcpu_info_cache.gpa;
 		else
 			data->u.gpa = GPA_INVALID;
···
 		break;
 
 	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO:
-		if (vcpu->arch.xen.vcpu_time_info_set)
+		if (vcpu->arch.xen.vcpu_time_info_cache.active)
 			data->u.gpa = vcpu->arch.xen.vcpu_time_info_cache.gpa;
 		else
 			data->u.gpa = GPA_INVALID;
···
 			r = -EOPNOTSUPP;
 			break;
 		}
-		if (vcpu->arch.xen.runstate_set) {
+		if (vcpu->arch.xen.runstate_cache.active) {
 			data->u.gpa = vcpu->arch.xen.runstate_cache.gpa;
 			r = 0;
 		}
···
 
 	case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST:
 		r = -EINVAL;
+		break;
+
+	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID:
+		data->u.vcpu_id = vcpu->arch.xen.vcpu_id;
+		r = 0;
+		break;
+
+	case KVM_XEN_VCPU_ATTR_TYPE_TIMER:
+		data->u.timer.port = vcpu->arch.xen.timer_virq;
+		data->u.timer.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL;
+		data->u.timer.expires_ns = vcpu->arch.xen.timer_expires;
+		r = 0;
+		break;
+
+	case KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR:
+		data->u.vector = vcpu->arch.xen.upcall_vector;
+		r = 0;
 		break;
 
 	default:
···
 
 int kvm_xen_hvm_config(struct kvm *kvm, struct kvm_xen_hvm_config *xhc)
 {
-	if (xhc->flags & ~KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL)
+	/* Only some feature flags need to be *enabled* by userspace */
+	u32 permitted_flags = KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL |
+		KVM_XEN_HVM_CONFIG_EVTCHN_SEND;
+
+	if (xhc->flags & ~permitted_flags)
 		return -EINVAL;
 
 	/*
···
 	return 0;
 }
 
-void kvm_xen_init_vm(struct kvm *kvm)
-{
-}
-
-void kvm_xen_destroy_vm(struct kvm *kvm)
-{
-	kvm_gfn_to_pfn_cache_destroy(kvm, &kvm->arch.xen.shinfo_cache);
-
-	if (kvm->arch.xen_hvm_config.msr)
-		static_branch_slow_dec_deferred(&kvm_xen_enabled);
-}
-
 static int kvm_xen_hypercall_set_result(struct kvm_vcpu *vcpu, u64 result)
 {
 	kvm_rax_write(vcpu, result);
···
 	return kvm_xen_hypercall_set_result(vcpu, run->xen.u.hcall.result);
 }
 
+static bool wait_pending_event(struct kvm_vcpu *vcpu, int nr_ports,
+			       evtchn_port_t *ports)
+{
+	struct kvm *kvm = vcpu->kvm;
+	struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache;
+	unsigned long *pending_bits;
+	unsigned long flags;
+	bool ret = true;
+	int idx, i;
+
+	read_lock_irqsave(&gpc->lock, flags);
+	idx = srcu_read_lock(&kvm->srcu);
+	if (!kvm_gfn_to_pfn_cache_check(kvm, gpc, gpc->gpa, PAGE_SIZE))
+		goto out_rcu;
+
+	ret = false;
+	if (IS_ENABLED(CONFIG_64BIT) && kvm->arch.xen.long_mode) {
+		struct shared_info *shinfo = gpc->khva;
+		pending_bits = (unsigned long *)&shinfo->evtchn_pending;
+	} else {
+		struct compat_shared_info *shinfo = gpc->khva;
+		pending_bits = (unsigned long *)&shinfo->evtchn_pending;
+	}
+
+	for (i = 0; i < nr_ports; i++) {
+		if (test_bit(ports[i], pending_bits)) {
+			ret = true;
+			break;
+		}
+	}
+
+ out_rcu:
+	srcu_read_unlock(&kvm->srcu, idx);
+	read_unlock_irqrestore(&gpc->lock, flags);
+
+	return ret;
+}
+
+static bool kvm_xen_schedop_poll(struct kvm_vcpu *vcpu, bool longmode,
+				 u64 param, u64 *r)
+{
+	int idx, i;
+	struct sched_poll sched_poll;
+	evtchn_port_t port, *ports;
+	gpa_t gpa;
+
+	if (!longmode || !lapic_in_kernel(vcpu) ||
+	    !(vcpu->kvm->arch.xen_hvm_config.flags & KVM_XEN_HVM_CONFIG_EVTCHN_SEND))
+		return false;
+
+	idx = srcu_read_lock(&vcpu->kvm->srcu);
+	gpa = kvm_mmu_gva_to_gpa_system(vcpu, param, NULL);
+	srcu_read_unlock(&vcpu->kvm->srcu, idx);
+
+	if (!gpa || kvm_vcpu_read_guest(vcpu, gpa, &sched_poll,
+					sizeof(sched_poll))) {
+		*r = -EFAULT;
+		return true;
+	}
+
+	if (unlikely(sched_poll.nr_ports > 1)) {
+		/* Xen (unofficially) limits number of pollers to 128 */
+		if (sched_poll.nr_ports > 128) {
+			*r = -EINVAL;
+			return true;
+		}
+
+		ports = kmalloc_array(sched_poll.nr_ports,
+				      sizeof(*ports), GFP_KERNEL);
+		if (!ports) {
+			*r = -ENOMEM;
+			return true;
+		}
+	} else
+		ports = &port;
+
+	for (i = 0; i < sched_poll.nr_ports; i++) {
+		idx = srcu_read_lock(&vcpu->kvm->srcu);
+		gpa = kvm_mmu_gva_to_gpa_system(vcpu,
+						(gva_t)(sched_poll.ports + i),
+						NULL);
+		srcu_read_unlock(&vcpu->kvm->srcu, idx);
+
+		if (!gpa || kvm_vcpu_read_guest(vcpu, gpa,
+						&ports[i], sizeof(port))) {
+			*r = -EFAULT;
+			goto out;
+		}
+	}
+
+	if (sched_poll.nr_ports == 1)
+		vcpu->arch.xen.poll_evtchn = port;
+	else
+		vcpu->arch.xen.poll_evtchn = -1;
+
+	set_bit(kvm_vcpu_get_idx(vcpu), vcpu->kvm->arch.xen.poll_mask);
+
+	if (!wait_pending_event(vcpu, sched_poll.nr_ports, ports)) {
+		vcpu->arch.mp_state = KVM_MP_STATE_HALTED;
+
+		if (sched_poll.timeout)
+			mod_timer(&vcpu->arch.xen.poll_timer,
+				  jiffies + nsecs_to_jiffies(sched_poll.timeout));
+
+		kvm_vcpu_halt(vcpu);
+
+		if (sched_poll.timeout)
+			del_timer(&vcpu->arch.xen.poll_timer);
+
+		vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
+		kvm_clear_request(KVM_REQ_UNHALT, vcpu);
+	}
+
+	vcpu->arch.xen.poll_evtchn = 0;
+	*r = 0;
+out:
+	/* Really, this is only needed in case of timeout */
+	clear_bit(kvm_vcpu_get_idx(vcpu), vcpu->kvm->arch.xen.poll_mask);
+
+	if (unlikely(sched_poll.nr_ports > 1))
+		kfree(ports);
+	return true;
+}
+
+static void cancel_evtchn_poll(struct timer_list *t)
+{
+	struct kvm_vcpu *vcpu = from_timer(vcpu, t, arch.xen.poll_timer);
+
+	kvm_make_request(KVM_REQ_UNBLOCK, vcpu);
+	kvm_vcpu_kick(vcpu);
+}
+
+static bool kvm_xen_hcall_sched_op(struct kvm_vcpu *vcpu, bool longmode,
+				   int cmd, u64 param, u64 *r)
+{
+	switch (cmd) {
+	case SCHEDOP_poll:
+		if (kvm_xen_schedop_poll(vcpu, longmode, param, r))
+			return true;
+		fallthrough;
+	case SCHEDOP_yield:
+		kvm_vcpu_on_spin(vcpu, true);
+		*r = 0;
+		return true;
+	default:
+		break;
+	}
+
+	return false;
+}
+
+struct compat_vcpu_set_singleshot_timer {
+    uint64_t timeout_abs_ns;
+    uint32_t flags;
+} __attribute__((packed));
+
+static bool kvm_xen_hcall_vcpu_op(struct kvm_vcpu *vcpu, bool longmode, int cmd,
+				  int vcpu_id, u64 param, u64 *r)
+{
+	struct vcpu_set_singleshot_timer oneshot;
+	s64 delta;
+	gpa_t gpa;
+	int idx;
+
+	if (!kvm_xen_timer_enabled(vcpu))
+		return false;
+
+	switch (cmd) {
+	case VCPUOP_set_singleshot_timer:
+		if (vcpu->arch.xen.vcpu_id != vcpu_id) {
+			*r = -EINVAL;
+			return true;
+		}
+		idx = srcu_read_lock(&vcpu->kvm->srcu);
+		gpa = kvm_mmu_gva_to_gpa_system(vcpu, param, NULL);
+		srcu_read_unlock(&vcpu->kvm->srcu, idx);
+
+		/*
+		 * The only difference for 32-bit compat is the 4 bytes of
+		 * padding after the interesting part of the structure. So
+		 * for a faithful emulation of Xen we have to *try* to copy
+		 * the padding and return -EFAULT if we can't. Otherwise we
+		 * might as well just have copied the 12-byte 32-bit struct.
+		 */
+		BUILD_BUG_ON(offsetof(struct compat_vcpu_set_singleshot_timer, timeout_abs_ns) !=
+			     offsetof(struct vcpu_set_singleshot_timer, timeout_abs_ns));
+		BUILD_BUG_ON(sizeof_field(struct compat_vcpu_set_singleshot_timer, timeout_abs_ns) !=
+			     sizeof_field(struct vcpu_set_singleshot_timer, timeout_abs_ns));
+		BUILD_BUG_ON(offsetof(struct compat_vcpu_set_singleshot_timer, flags) !=
+			     offsetof(struct vcpu_set_singleshot_timer, flags));
+		BUILD_BUG_ON(sizeof_field(struct compat_vcpu_set_singleshot_timer, flags) !=
+			     sizeof_field(struct vcpu_set_singleshot_timer, flags));
+
+		if (!gpa ||
+		    kvm_vcpu_read_guest(vcpu, gpa, &oneshot, longmode ? sizeof(oneshot) :
+					sizeof(struct compat_vcpu_set_singleshot_timer))) {
+			*r = -EFAULT;
+			return true;
+		}
+
+		delta = oneshot.timeout_abs_ns - get_kvmclock_ns(vcpu->kvm);
+		if ((oneshot.flags & VCPU_SSHOTTMR_future) && delta < 0) {
+			*r = -ETIME;
+			return true;
+		}
+
+		kvm_xen_start_timer(vcpu, oneshot.timeout_abs_ns, delta);
+		*r = 0;
+		return true;
+
+	case VCPUOP_stop_singleshot_timer:
+		if (vcpu->arch.xen.vcpu_id != vcpu_id) {
+			*r = -EINVAL;
+			return true;
+		}
+		kvm_xen_stop_timer(vcpu);
+		*r = 0;
+		return true;
+	}
+
+	return false;
+}
+
+static bool kvm_xen_hcall_set_timer_op(struct kvm_vcpu *vcpu, uint64_t timeout,
+				       u64 *r)
+{
+	if (!kvm_xen_timer_enabled(vcpu))
+		return false;
+
+	if (timeout) {
+		uint64_t guest_now = get_kvmclock_ns(vcpu->kvm);
+		int64_t delta = timeout - guest_now;
+
+		/* Xen has a 'Linux workaround' in do_set_timer_op() which
+		 * checks for negative absolute timeout values (caused by
+		 * integer overflow), and for values about 13 days in the
+		 * future (2^50ns) which would be caused by jiffies
+		 * overflow. For those cases, it sets the timeout 100ms in
+		 * the future (not *too* soon, since if a guest really did
+		 * set a long timeout on purpose we don't want to keep
+		 * churning CPU time by waking it up).
+		 */
+		if (unlikely((int64_t)timeout < 0 ||
+			     (delta > 0 && (uint32_t) (delta >> 50) != 0))) {
+			delta = 100 * NSEC_PER_MSEC;
+			timeout = guest_now + delta;
+		}
+
+		kvm_xen_start_timer(vcpu, timeout, delta);
+	} else {
+		kvm_xen_stop_timer(vcpu);
+	}
+
+	*r = 0;
+	return true;
+}
+
 int kvm_xen_hypercall(struct kvm_vcpu *vcpu)
 {
 	bool longmode;
-	u64 input, params[6];
+	u64 input, params[6], r = -ENOSYS;
+	bool handled = false;
 
 	input = (u64)kvm_register_read(vcpu, VCPU_REGS_RAX);
 
···
 	trace_kvm_xen_hypercall(input, params[0], params[1], params[2],
 				params[3], params[4], params[5]);
 
+	switch (input) {
+	case __HYPERVISOR_xen_version:
+		if (params[0] == XENVER_version && vcpu->kvm->arch.xen.xen_version) {
+			r = vcpu->kvm->arch.xen.xen_version;
+			handled = true;
+		}
+		break;
+	case __HYPERVISOR_event_channel_op:
+		if (params[0] == EVTCHNOP_send)
+			handled = kvm_xen_hcall_evtchn_send(vcpu, params[1], &r);
+		break;
+	case __HYPERVISOR_sched_op:
+		handled = kvm_xen_hcall_sched_op(vcpu, longmode, params[0],
+						 params[1], &r);
+		break;
+	case __HYPERVISOR_vcpu_op:
+		handled = kvm_xen_hcall_vcpu_op(vcpu, longmode, params[0], params[1],
+						params[2], &r);
+		break;
+	case __HYPERVISOR_set_timer_op: {
+		u64 timeout = params[0];
+		/* In 32-bit mode, the 64-bit timeout is in two 32-bit params. */
+		if (!longmode)
+			timeout |= params[1] << 32;
+		handled = kvm_xen_hcall_set_timer_op(vcpu, timeout, &r);
+		break;
+	}
+	default:
+		break;
+	}
+
+	if (handled)
+		return kvm_xen_hypercall_set_result(vcpu, r);
+
 	vcpu->run->exit_reason = KVM_EXIT_XEN;
 	vcpu->run->xen.type = KVM_EXIT_XEN_HCALL;
 	vcpu->run->xen.u.hcall.longmode = longmode;
···
 		return COMPAT_EVTCHN_2L_NR_CHANNELS;
 }
 
+static void kvm_xen_check_poller(struct kvm_vcpu *vcpu, int port)
+{
+	int poll_evtchn = vcpu->arch.xen.poll_evtchn;
+
+	if ((poll_evtchn == port || poll_evtchn == -1) &&
+	    test_and_clear_bit(kvm_vcpu_get_idx(vcpu), vcpu->kvm->arch.xen.poll_mask)) {
+		kvm_make_request(KVM_REQ_UNBLOCK, vcpu);
+		kvm_vcpu_kick(vcpu);
+	}
+}
+
 /*
- * This follows the kvm_set_irq() API, so it returns:
+ * The return value from this function is propagated to kvm_set_irq() API,
+ * so it returns:
  *  < 0   Interrupt was ignored (masked or not delivered for other reasons)
  *  = 0   Interrupt was coalesced (previous irq is still pending)
  *  > 0   Number of CPUs interrupt was delivered to
+ *
+ * It is also called directly from kvm_arch_set_irq_inatomic(), where the
+ * only check on its return value is a comparison with -EWOULDBLOCK'.
  */
-int kvm_xen_set_evtchn_fast(struct kvm_kernel_irq_routing_entry *e,
-			    struct kvm *kvm)
+int kvm_xen_set_evtchn_fast(struct kvm_xen_evtchn *xe, struct kvm *kvm)
 {
 	struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache;
 	struct kvm_vcpu *vcpu;
···
 	unsigned long flags;
 	int port_word_bit;
 	bool kick_vcpu = false;
-	int idx;
-	int rc;
+	int vcpu_idx, idx, rc;
 
-	vcpu = kvm_get_vcpu_by_id(kvm, e->xen_evtchn.vcpu);
-	if (!vcpu)
-		return -1;
+	vcpu_idx = READ_ONCE(xe->vcpu_idx);
+	if (vcpu_idx >= 0)
+		vcpu = kvm_get_vcpu(kvm, vcpu_idx);
+	else {
+		vcpu = kvm_get_vcpu_by_id(kvm, xe->vcpu_id);
+		if (!vcpu)
+			return -EINVAL;
+		WRITE_ONCE(xe->vcpu_idx, kvm_vcpu_get_idx(vcpu));
+	}
 
-	if (!vcpu->arch.xen.vcpu_info_set)
-		return -1;
+	if (!vcpu->arch.xen.vcpu_info_cache.active)
+		return -EINVAL;
 
-	if (e->xen_evtchn.port >= max_evtchn_port(kvm))
-		return -1;
+	if (xe->port >= max_evtchn_port(kvm))
+		return -EINVAL;
 
 	rc = -EWOULDBLOCK;
-	read_lock_irqsave(&gpc->lock, flags);
 
 	idx = srcu_read_lock(&kvm->srcu);
+
+	read_lock_irqsave(&gpc->lock, flags);
 	if (!kvm_gfn_to_pfn_cache_check(kvm, gpc, gpc->gpa, PAGE_SIZE))
 		goto out_rcu;
 
···
 		struct shared_info *shinfo = gpc->khva;
 		pending_bits = (unsigned long *)&shinfo->evtchn_pending;
 		mask_bits = (unsigned long *)&shinfo->evtchn_mask;
-		port_word_bit = e->xen_evtchn.port / 64;
+		port_word_bit = xe->port / 64;
 	} else {
 		struct compat_shared_info *shinfo = gpc->khva;
 		pending_bits = (unsigned long *)&shinfo->evtchn_pending;
 		mask_bits = (unsigned long *)&shinfo->evtchn_mask;
-		port_word_bit = e->xen_evtchn.port / 32;
+		port_word_bit = xe->port / 32;
 	}
 
 	/*
···
 	 * already set, then we kick the vCPU in question to write to the
 	 * *real* evtchn_pending_sel in its own guest vcpu_info struct.
 	 */
-	if (test_and_set_bit(e->xen_evtchn.port, pending_bits)) {
+	if (test_and_set_bit(xe->port, pending_bits)) {
 		rc = 0; /* It was already raised */
-	} else if (test_bit(e->xen_evtchn.port, mask_bits)) {
-		rc = -1; /* Masked */
+	} else if (test_bit(xe->port, mask_bits)) {
+		rc = -ENOTCONN; /* Masked */
+		kvm_xen_check_poller(vcpu, xe->port);
 	} else {
-		rc = 1; /* Delivered. But was the vCPU waking already? */
-		if (!test_and_set_bit(port_word_bit, &vcpu->arch.xen.evtchn_pending_sel))
-			kick_vcpu = true;
+		rc = 1; /* Delivered to the bitmap in shared_info. */
+		/* Now switch to the vCPU's vcpu_info to set the index and pending_sel */
+		read_unlock_irqrestore(&gpc->lock, flags);
+		gpc = &vcpu->arch.xen.vcpu_info_cache;
+
+		read_lock_irqsave(&gpc->lock, flags);
+		if (!kvm_gfn_to_pfn_cache_check(kvm, gpc, gpc->gpa, sizeof(struct vcpu_info))) {
+			/*
+			 * Could not access the vcpu_info. Set the bit in-kernel
+			 * and prod the vCPU to deliver it for itself.
+			 */
+			if (!test_and_set_bit(port_word_bit, &vcpu->arch.xen.evtchn_pending_sel))
+				kick_vcpu = true;
+			goto out_rcu;
+		}
+
+		if (IS_ENABLED(CONFIG_64BIT) && kvm->arch.xen.long_mode) {
+			struct vcpu_info *vcpu_info = gpc->khva;
+			if (!test_and_set_bit(port_word_bit, &vcpu_info->evtchn_pending_sel)) {
+				WRITE_ONCE(vcpu_info->evtchn_upcall_pending, 1);
+				kick_vcpu = true;
+			}
+		} else {
+			struct compat_vcpu_info *vcpu_info = gpc->khva;
+			if (!test_and_set_bit(port_word_bit,
+					      (unsigned long *)&vcpu_info->evtchn_pending_sel)) {
+				WRITE_ONCE(vcpu_info->evtchn_upcall_pending, 1);
+				kick_vcpu = true;
+			}
+		}
+
+		/* For the per-vCPU lapic vector, deliver it as MSI. */
+		if (kick_vcpu && vcpu->arch.xen.upcall_vector) {
+			kvm_xen_inject_vcpu_vector(vcpu);
+			kick_vcpu = false;
+		}
 	}
 
  out_rcu:
-	srcu_read_unlock(&kvm->srcu, idx);
 	read_unlock_irqrestore(&gpc->lock, flags);
+	srcu_read_unlock(&kvm->srcu, idx);
 
 	if (kick_vcpu) {
-		kvm_make_request(KVM_REQ_EVENT, vcpu);
+		kvm_make_request(KVM_REQ_UNBLOCK, vcpu);
 		kvm_vcpu_kick(vcpu);
 	}
 
 	return rc;
 }
 
-/* This is the version called from kvm_set_irq() as the .set function */
-static int evtchn_set_fn(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm,
-			 int irq_source_id, int level, bool line_status)
+static int kvm_xen_set_evtchn(struct kvm_xen_evtchn *xe, struct kvm *kvm)
 {
 	bool mm_borrowed = false;
 	int rc;
 
-	if (!level)
-		return -1;
-
-	rc = kvm_xen_set_evtchn_fast(e, kvm);
+	rc = kvm_xen_set_evtchn_fast(xe, kvm);
 	if (rc != -EWOULDBLOCK)
 		return rc;
 
···
 		struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache;
 		int idx;
 
-		rc = kvm_xen_set_evtchn_fast(e, kvm);
+		rc = kvm_xen_set_evtchn_fast(xe, kvm);
 		if (rc != -EWOULDBLOCK)
 			break;
 
···
 	return rc;
 }
 
+/* This is the version called from kvm_set_irq() as the .set function */
+static int evtchn_set_fn(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm,
+			 int irq_source_id, int level, bool line_status)
+{
+	if (!level)
+		return -EINVAL;
+
+	return kvm_xen_set_evtchn(&e->xen_evtchn, kvm);
+}
+
+/*
+ * Set up an event channel interrupt from the KVM IRQ routing table.
+ * Used for e.g. PIRQ from passed through physical devices.
+ */
 int kvm_xen_setup_evtchn(struct kvm *kvm,
 			 struct kvm_kernel_irq_routing_entry *e,
 			 const struct kvm_irq_routing_entry *ue)
 
 {
+	struct kvm_vcpu *vcpu;
+
 	if (ue->u.xen_evtchn.port >= max_evtchn_port(kvm))
 		return -EINVAL;
 
···
 	if (ue->u.xen_evtchn.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL)
 		return -EINVAL;
 
+	/*
+	 * Xen gives us interesting mappings from vCPU index to APIC ID,
+	 * which means kvm_get_vcpu_by_id() has to iterate over all vCPUs
+	 * to find it. Do that once at setup time, instead of every time.
+	 * But beware that on live update / live migration, the routing
+	 * table might be reinstated before the vCPU threads have finished
+	 * recreating their vCPUs.
+	 */
+	vcpu = kvm_get_vcpu_by_id(kvm, ue->u.xen_evtchn.vcpu);
+	if (vcpu)
+		e->xen_evtchn.vcpu_idx = kvm_vcpu_get_idx(vcpu);
+	else
+		e->xen_evtchn.vcpu_idx = -1;
+
 	e->xen_evtchn.port = ue->u.xen_evtchn.port;
-	e->xen_evtchn.vcpu = ue->u.xen_evtchn.vcpu;
+	e->xen_evtchn.vcpu_id = ue->u.xen_evtchn.vcpu;
 	e->xen_evtchn.priority = ue->u.xen_evtchn.priority;
 	e->set = evtchn_set_fn;
 
 	return 0;
+}
+
+/*
+ * Explicit event sending from userspace with KVM_XEN_HVM_EVTCHN_SEND ioctl.
+ */
+int kvm_xen_hvm_evtchn_send(struct kvm *kvm, struct kvm_irq_routing_xen_evtchn *uxe)
+{
+	struct kvm_xen_evtchn e;
+	int ret;
+
+	if (!uxe->port || uxe->port >= max_evtchn_port(kvm))
+		return -EINVAL;
+
+	/* We only support 2 level event channels for now */
+	if (uxe->priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL)
+		return -EINVAL;
+
+	e.port = uxe->port;
+	e.vcpu_id = uxe->vcpu;
+	e.vcpu_idx = -1;
+	e.priority = uxe->priority;
+
+	ret = kvm_xen_set_evtchn(&e, kvm);
+
+	/*
+	 * None of that 'return 1 if it actually got delivered' nonsense.
+	 * We don't care if it was masked (-ENOTCONN) either.
+	 */
+	if (ret > 0 || ret == -ENOTCONN)
+		ret = 0;
+
+	return ret;
+}
+
+/*
+ * Support for *outbound* event channel events via the EVTCHNOP_send hypercall.
+ */
+struct evtchnfd {
+	u32 send_port;
+	u32 type;
+	union {
+		struct kvm_xen_evtchn port;
+		struct {
+			u32 port; /* zero */
+			struct eventfd_ctx *ctx;
+		} eventfd;
+	} deliver;
+};
+
+/*
+ * Update target vCPU or priority for a registered sending channel.
+ */
+static int kvm_xen_eventfd_update(struct kvm *kvm,
+				  struct kvm_xen_hvm_attr *data)
+{
+	u32 port = data->u.evtchn.send_port;
+	struct evtchnfd *evtchnfd;
+
+	if (!port || port >= max_evtchn_port(kvm))
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+	evtchnfd = idr_find(&kvm->arch.xen.evtchn_ports, port);
+	mutex_unlock(&kvm->lock);
+
+	if (!evtchnfd)
+		return -ENOENT;
+
+	/* For an UPDATE, nothing may change except the priority/vcpu */
+	if (evtchnfd->type != data->u.evtchn.type)
+		return -EINVAL;
+
+	/*
+	 * Port cannot change, and if it's zero that was an eventfd
+	 * which can't be changed either.
+	 */
+	if (!evtchnfd->deliver.port.port ||
+	    evtchnfd->deliver.port.port != data->u.evtchn.deliver.port.port)
+		return -EINVAL;
+
+	/* We only support 2 level event channels for now */
+	if (data->u.evtchn.deliver.port.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+	evtchnfd->deliver.port.priority = data->u.evtchn.deliver.port.priority;
+	if (evtchnfd->deliver.port.vcpu_id != data->u.evtchn.deliver.port.vcpu) {
+		evtchnfd->deliver.port.vcpu_id = data->u.evtchn.deliver.port.vcpu;
+		evtchnfd->deliver.port.vcpu_idx = -1;
+	}
+	mutex_unlock(&kvm->lock);
+	return 0;
+}
+
+/*
+ * Configure the target (eventfd or local port delivery) for sending on
+ * a given event channel.
+ */
+static int kvm_xen_eventfd_assign(struct kvm *kvm,
+				  struct kvm_xen_hvm_attr *data)
+{
+	u32 port = data->u.evtchn.send_port;
+	struct eventfd_ctx *eventfd = NULL;
+	struct evtchnfd *evtchnfd = NULL;
+	int ret = -EINVAL;
+
+	if (!port || port >= max_evtchn_port(kvm))
+		return -EINVAL;
+
+	evtchnfd = kzalloc(sizeof(struct evtchnfd), GFP_KERNEL);
+	if (!evtchnfd)
+		return -ENOMEM;
+
+	switch(data->u.evtchn.type) {
+	case EVTCHNSTAT_ipi:
+		/* IPI  must map back to the same port# */
+		if (data->u.evtchn.deliver.port.port != data->u.evtchn.send_port)
+			goto out; /* -EINVAL */
+		break;
+
+	case EVTCHNSTAT_interdomain:
+		if (data->u.evtchn.deliver.port.port) {
+			if (data->u.evtchn.deliver.port.port >= max_evtchn_port(kvm))
+				goto out; /* -EINVAL */
+		} else {
+			eventfd = eventfd_ctx_fdget(data->u.evtchn.deliver.eventfd.fd);
+			if (IS_ERR(eventfd)) {
+				ret = PTR_ERR(eventfd);
+				goto out;
+			}
+		}
+		break;
+
+	case EVTCHNSTAT_virq:
+	case EVTCHNSTAT_closed:
+	case EVTCHNSTAT_unbound:
+	case EVTCHNSTAT_pirq:
+	default: /* Unknown event channel type */
+		goto out; /* -EINVAL */
+	}
+
+	evtchnfd->send_port = data->u.evtchn.send_port;
+	evtchnfd->type = data->u.evtchn.type;
+	if (eventfd) {
+		evtchnfd->deliver.eventfd.ctx = eventfd;
+	} else {
+		/* We only support 2 level event channels for now */
+		if (data->u.evtchn.deliver.port.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL)
+			goto out; /* -EINVAL; */
+
+		evtchnfd->deliver.port.port = data->u.evtchn.deliver.port.port;
+		evtchnfd->deliver.port.vcpu_id = data->u.evtchn.deliver.port.vcpu;
+		evtchnfd->deliver.port.vcpu_idx = -1;
+		evtchnfd->deliver.port.priority = data->u.evtchn.deliver.port.priority;
+	}
+
+	mutex_lock(&kvm->lock);
+	ret = idr_alloc(&kvm->arch.xen.evtchn_ports, evtchnfd, port, port + 1,
+			GFP_KERNEL);
+	mutex_unlock(&kvm->lock);
+	if (ret >= 0)
+		return 0;
+
+	if (ret == -ENOSPC)
+		ret = -EEXIST;
+out:
+	if (eventfd)
+		eventfd_ctx_put(eventfd);
+	kfree(evtchnfd);
+	return ret;
+}
+
+static int kvm_xen_eventfd_deassign(struct kvm *kvm, u32 port)
+{
+	struct evtchnfd *evtchnfd;
+
+	mutex_lock(&kvm->lock);
+	evtchnfd = idr_remove(&kvm->arch.xen.evtchn_ports, port);
+	mutex_unlock(&kvm->lock);
+
+	if (!evtchnfd)
+		return -ENOENT;
+
+	if (kvm)
+		synchronize_srcu(&kvm->srcu);
+	if (!evtchnfd->deliver.port.port)
+		eventfd_ctx_put(evtchnfd->deliver.eventfd.ctx);
+	kfree(evtchnfd);
+	return 0;
+}
+
+static int kvm_xen_eventfd_reset(struct kvm *kvm)
+{
+	struct evtchnfd *evtchnfd;
+	int i;
+
+	mutex_lock(&kvm->lock);
+	idr_for_each_entry(&kvm->arch.xen.evtchn_ports, evtchnfd, i) {
+		idr_remove(&kvm->arch.xen.evtchn_ports, evtchnfd->send_port);
+		synchronize_srcu(&kvm->srcu);
+		if (!evtchnfd->deliver.port.port)
+			eventfd_ctx_put(evtchnfd->deliver.eventfd.ctx);
+		kfree(evtchnfd);
+	}
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+}
+
+static int kvm_xen_setattr_evtchn(struct kvm *kvm, struct kvm_xen_hvm_attr *data)
+{
+	u32 port = data->u.evtchn.send_port;
+
+	if (data->u.evtchn.flags == KVM_XEN_EVTCHN_RESET)
+		return kvm_xen_eventfd_reset(kvm);
+
+	if (!port || port >= max_evtchn_port(kvm))
+		return -EINVAL;
+
+	if (data->u.evtchn.flags == KVM_XEN_EVTCHN_DEASSIGN)
+		return kvm_xen_eventfd_deassign(kvm, port);
+	if (data->u.evtchn.flags == KVM_XEN_EVTCHN_UPDATE)
+		return kvm_xen_eventfd_update(kvm, data);
+	if (data->u.evtchn.flags)
+		return -EINVAL;
+
+	return kvm_xen_eventfd_assign(kvm, data);
+}
+
+static bool kvm_xen_hcall_evtchn_send(struct kvm_vcpu *vcpu, u64 param, u64 *r)
+{
+	struct evtchnfd *evtchnfd;
+	struct evtchn_send send;
+	gpa_t gpa;
+	int idx;
+
+	idx = srcu_read_lock(&vcpu->kvm->srcu);
+	gpa = kvm_mmu_gva_to_gpa_system(vcpu, param, NULL);
+	srcu_read_unlock(&vcpu->kvm->srcu, idx);
+
+	if (!gpa || kvm_vcpu_read_guest(vcpu, gpa, &send, sizeof(send))) {
+		*r = -EFAULT;
+		return true;
+	}
+
+	/* The evtchn_ports idr is protected by vcpu->kvm->srcu */
+	evtchnfd = idr_find(&vcpu->kvm->arch.xen.evtchn_ports, send.port);
+	if (!evtchnfd)
+		return false;
+
+	if (evtchnfd->deliver.port.port) {
+		int ret = kvm_xen_set_evtchn(&evtchnfd->deliver.port, vcpu->kvm);
+		if (ret < 0 && ret != -ENOTCONN)
+			return false;
+	} else {
+		eventfd_signal(evtchnfd->deliver.eventfd.ctx, 1);
+	}
+
+	*r = 0;
+	return true;
+}
+
+void kvm_xen_init_vcpu(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.xen.vcpu_id = vcpu->vcpu_idx;
+	vcpu->arch.xen.poll_evtchn = 0;
+	timer_setup(&vcpu->arch.xen.poll_timer, cancel_evtchn_poll, 0);
+}
+
+void kvm_xen_destroy_vcpu(struct kvm_vcpu *vcpu)
+{
+	if (kvm_xen_timer_enabled(vcpu))
+		kvm_xen_stop_timer(vcpu);
+
+	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+				     &vcpu->arch.xen.runstate_cache);
+	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+				     &vcpu->arch.xen.vcpu_info_cache);
+	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+				     &vcpu->arch.xen.vcpu_time_info_cache);
+	del_timer_sync(&vcpu->arch.xen.poll_timer);
+}
+
+void kvm_xen_init_vm(struct kvm *kvm)
+{
+	idr_init(&kvm->arch.xen.evtchn_ports);
+}
+
+void kvm_xen_destroy_vm(struct kvm *kvm)
+{
+	struct evtchnfd *evtchnfd;
+	int i;
+
+	kvm_gfn_to_pfn_cache_destroy(kvm, &kvm->arch.xen.shinfo_cache);
+
+	idr_for_each_entry(&kvm->arch.xen.evtchn_ports, evtchnfd, i) {
+		if (!evtchnfd->deliver.port.port)
+			eventfd_ctx_put(evtchnfd->deliver.eventfd.ctx);
+		kfree(evtchnfd);
+	}
+	idr_destroy(&kvm->arch.xen.evtchn_ports);
+
+	if (kvm->arch.xen_hvm_config.msr)
+		static_branch_slow_dec_deferred(&kvm_xen_enabled);
 }

arch/x86/kvm/xen.h

···
 extern struct static_key_false_deferred kvm_xen_enabled;
 
 int __kvm_xen_has_interrupt(struct kvm_vcpu *vcpu);
+void kvm_xen_inject_pending_events(struct kvm_vcpu *vcpu);
 int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data);
 int kvm_xen_vcpu_get_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data);
 int kvm_xen_hvm_set_attr(struct kvm *kvm, struct kvm_xen_hvm_attr *data);
 int kvm_xen_hvm_get_attr(struct kvm *kvm, struct kvm_xen_hvm_attr *data);
+int kvm_xen_hvm_evtchn_send(struct kvm *kvm, struct kvm_irq_routing_xen_evtchn *evt);
 int kvm_xen_write_hypercall_page(struct kvm_vcpu *vcpu, u64 data);
 int kvm_xen_hvm_config(struct kvm *kvm, struct kvm_xen_hvm_config *xhc);
 void kvm_xen_init_vm(struct kvm *kvm);
 void kvm_xen_destroy_vm(struct kvm *kvm);
-
-int kvm_xen_set_evtchn_fast(struct kvm_kernel_irq_routing_entry *e,
+void kvm_xen_init_vcpu(struct kvm_vcpu *vcpu);
+void kvm_xen_destroy_vcpu(struct kvm_vcpu *vcpu);
+int kvm_xen_set_evtchn_fast(struct kvm_xen_evtchn *xe,
 			    struct kvm *kvm);
 int kvm_xen_setup_evtchn(struct kvm *kvm,
 			 struct kvm_kernel_irq_routing_entry *e,
···
 static inline int kvm_xen_has_interrupt(struct kvm_vcpu *vcpu)
 {
 	if (static_branch_unlikely(&kvm_xen_enabled.key) &&
-	    vcpu->arch.xen.vcpu_info_set && vcpu->kvm->arch.xen.upcall_vector)
+	    vcpu->arch.xen.vcpu_info_cache.active &&
+	    vcpu->kvm->arch.xen.upcall_vector)
 		return __kvm_xen_has_interrupt(vcpu);
 
 	return 0;
 }
+
+static inline bool kvm_xen_has_pending_events(struct kvm_vcpu *vcpu)
+{
+	return static_branch_unlikely(&kvm_xen_enabled.key) &&
+		vcpu->arch.xen.evtchn_pending_sel;
+}
+
+static inline bool kvm_xen_timer_enabled(struct kvm_vcpu *vcpu)
+{
+	return !!vcpu->arch.xen.timer_virq;
+}
+
+static inline int kvm_xen_has_pending_timer(struct kvm_vcpu *vcpu)
+{
+	if (kvm_xen_hypercall_enabled(vcpu->kvm) && kvm_xen_timer_enabled(vcpu))
+		return atomic_read(&vcpu->arch.xen.timer_pending);
+
+	return 0;
+}
+
+void kvm_xen_inject_timer_irqs(struct kvm_vcpu *vcpu);
 #else
 static inline int kvm_xen_write_hypercall_page(struct kvm_vcpu *vcpu, u64 data)
 {
···
 }
 
 static inline void kvm_xen_destroy_vm(struct kvm *kvm)
+{
+}
+
+static inline void kvm_xen_init_vcpu(struct kvm_vcpu *vcpu)
+{
+}
+
+static inline void kvm_xen_destroy_vcpu(struct kvm_vcpu *vcpu)
 {
 }
 
···
 static inline int kvm_xen_has_interrupt(struct kvm_vcpu *vcpu)
 {
 	return 0;
+}
+
+static inline void kvm_xen_inject_pending_events(struct kvm_vcpu *vcpu)
+{
+}
+
+static inline bool kvm_xen_has_pending_events(struct kvm_vcpu *vcpu)
+{
+	return false;
+}
+
+static inline int kvm_xen_has_pending_timer(struct kvm_vcpu *vcpu)
+{
+	return 0;
+}
+
+static inline void kvm_xen_inject_timer_irqs(struct kvm_vcpu *vcpu)
+{
+}
+
+static inline bool kvm_xen_timer_enabled(struct kvm_vcpu *vcpu)
+{
+	return false;
 }
 #endif
 

drivers/s390/char/Kconfig

···
 	  This option enables the Open-for-Business interface to the s390
 	  Service Element.
 
+config S390_UV_UAPI
+	def_tristate m
+	prompt "Ultravisor userspace API"
+	depends on S390
+	help
+	  Selecting exposes parts of the UV interface to userspace
+	  by providing a misc character device at /dev/uv.
+	  Using IOCTLs one can interact with the UV.
+	  The device is only available if the Ultravisor
+	  Facility (158) is present.
+
 config S390_TAPE
 	def_tristate m
 	prompt "S/390 tape device support"

drivers/s390/char/Makefile

···
 obj-$(CONFIG_MONWRITER) += monwriter.o
 obj-$(CONFIG_S390_VMUR) += vmur.o
 obj-$(CONFIG_CRASH_DUMP) += sclp_sdias.o zcore.o
+obj-$(CONFIG_S390_UV_UAPI) += uvdevice.o
 
 hmcdrv-objs := hmcdrv_mod.o hmcdrv_dev.o hmcdrv_ftp.o hmcdrv_cache.o diag_ftp.o sclp_ftp.o
 obj-$(CONFIG_HMC_DRV) += hmcdrv.o

drivers/s390/char/uvdevice.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Copyright IBM Corp. 2022
+ *  Author(s): Steffen Eiden <[e-mail]>
+ *
+ *  This file provides a Linux misc device to give userspace access to some
+ *  Ultravisor (UV) functions. The device only accepts IOCTLs and will only
+ *  be present if the Ultravisor facility (158) is present.
+ *
+ *  When userspace sends a valid IOCTL uvdevice will copy the input data to
+ *  kernel space, do some basic validity checks to avoid kernel/system
+ *  corruption. Any other check that the Ultravisor does will not be done by
+ *  the uvdevice to keep changes minimal when adding new functionalities
+ *  to existing UV-calls.
+ *  After the checks uvdevice builds a corresponding
+ *  Ultravisor Call Control Block, and sends the request to the Ultravisor.
+ *  Then, it copies the response, including the return codes, back to userspace.
+ *  It is the responsibility of the userspace to check for any error issued
+ *  by UV and to interpret the UV response. The uvdevice acts as a communication
+ *  channel for userspace to the Ultravisor.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+
+#include <asm/uvdevice.h>
+#include <asm/uv.h>
+
+static int uvio_build_uvcb_attest(struct uv_cb_attest *uvcb_attest, u8 *arcb,
+				  u8 *meas, u8 *add_data, struct uvio_attest *uvio_attest)
+{
+	void __user *user_buf_arcb = (void __user *)uvio_attest->arcb_addr;
+
+	if (copy_from_user(arcb, user_buf_arcb, uvio_attest->arcb_len))
+		return -EFAULT;
+
+	uvcb_attest->header.len = sizeof(*uvcb_attest);
+	uvcb_attest->header.cmd = UVC_CMD_RETR_ATTEST;
+	uvcb_attest->arcb_addr = (u64)arcb;
+	uvcb_attest->cont_token = 0;
+	uvcb_attest->user_data_len = uvio_attest->user_data_len;
+	memcpy(uvcb_attest->user_data, uvio_attest->user_data, sizeof(uvcb_attest->user_data));
+	uvcb_attest->meas_len = uvio_attest->meas_len;
+	uvcb_attest->meas_addr = (u64)meas;
+	uvcb_attest->add_data_len = uvio_attest->add_data_len;
+	uvcb_attest->add_data_addr = (u64)add_data;
+
+	return 0;
+}
+
+static int uvio_copy_attest_result_to_user(struct uv_cb_attest *uvcb_attest,
+					   struct uvio_ioctl_cb *uv_ioctl,
+					   u8 *measurement, u8 *add_data,
+					   struct uvio_attest *uvio_attest)
+{
+	struct uvio_attest __user *user_uvio_attest = (void __user *)uv_ioctl->argument_addr;
+	void __user *user_buf_add = (void __user *)uvio_attest->add_data_addr;
+	void __user *user_buf_meas = (void __user *)uvio_attest->meas_addr;
+	void __user *user_buf_uid = &user_uvio_attest->config_uid;
+
+	if (copy_to_user(user_buf_meas, measurement, uvio_attest->meas_len))
+		return -EFAULT;
+	if (add_data && copy_to_user(user_buf_add, add_data, uvio_attest->add_data_len))
+		return -EFAULT;
+	if (copy_to_user(user_buf_uid, uvcb_attest->config_uid, sizeof(uvcb_attest->config_uid)))
+		return -EFAULT;
+	return 0;
+}
+
+static int get_uvio_attest(struct uvio_ioctl_cb *uv_ioctl, struct uvio_attest *uvio_attest)
+{
+	u8 __user *user_arg_buf = (u8 __user *)uv_ioctl->argument_addr;
+
+	if (copy_from_user(uvio_attest, user_arg_buf, sizeof(*uvio_attest)))
+		return -EFAULT;
+
+	if (uvio_attest->arcb_len > UVIO_ATT_ARCB_MAX_LEN)
+		return -EINVAL;
+	if (uvio_attest->arcb_len == 0)
+		return -EINVAL;
+	if (uvio_attest->meas_len > UVIO_ATT_MEASUREMENT_MAX_LEN)
+		return -EINVAL;
+	if (uvio_attest->meas_len == 0)
+		return -EINVAL;
+	if (uvio_attest->add_data_len > UVIO_ATT_ADDITIONAL_MAX_LEN)
+		return -EINVAL;
+	if (uvio_attest->reserved136)
+		return -EINVAL;
+	return 0;
+}
+
+/**
+ * uvio_attestation() - Perform a Retrieve Attestation Measurement UVC.
+ *
+ * @uv_ioctl: ioctl control block
+ *
+ * uvio_attestation() does a Retrieve Attestation Measurement Ultravisor Call.
+ * It verifies that the given userspace addresses are valid and request sizes
+ * are sane. Every other check is made by the Ultravisor (UV) and won't result
+ * in a negative return value. It copies the input to kernelspace, builds the
+ * request, sends the UV-call, and copies the result to userspace.
+ *
+ * The Attestation Request has two input and two outputs.
+ * ARCB and User Data are inputs for the UV generated by userspace.
+ * Measurement and Additional Data are outputs for userspace generated by UV.
+ *
+ * The Attestation Request Control Block (ARCB) is a cryptographically verified
+ * and secured request to UV and User Data is some plaintext data which is
+ * going to be included in the Attestation Measurement calculation.
+ *
+ * Measurement is a cryptographic measurement of the callers properties,
+ * optional data configured by the ARCB and the user data. If specified by the
+ * ARCB, UV will add some Additional Data to the measurement calculation.
+ * This Additional Data is then returned as well.
+ *
+ * If the Retrieve Attestation Measurement UV facility is not present,
+ * UV will return invalid command rc. This won't be fenced in the driver
+ * and does not result in a negative return value.
+ *
+ * Context: might sleep
+ *
+ * Return: 0 on success or a negative error code on error.
+ */
+static int uvio_attestation(struct uvio_ioctl_cb *uv_ioctl)
+{
+	struct uv_cb_attest *uvcb_attest = NULL;
+	struct uvio_attest *uvio_attest = NULL;
+	u8 *measurement = NULL;
+	u8 *add_data = NULL;
+	u8 *arcb = NULL;
+	int ret;
+
+	ret = -EINVAL;
+	if (uv_ioctl->argument_len != sizeof(*uvio_attest))
+		goto out;
+
+	ret = -ENOMEM;
+	uvio_attest = kzalloc(sizeof(*uvio_attest), GFP_KERNEL);
+	if (!uvio_attest)
+		goto out;
+
+	ret = get_uvio_attest(uv_ioctl, uvio_attest);
+	if (ret)
+		goto out;
+
+	ret = -ENOMEM;
+	arcb = kvzalloc(uvio_attest->arcb_len, GFP_KERNEL);
+	measurement = kvzalloc(uvio_attest->meas_len, GFP_KERNEL);
+	if (!arcb || !measurement)
+		goto out;
+
+	if (uvio_attest->add_data_len) {
+		add_data = kvzalloc(uvio_attest->add_data_len, GFP_KERNEL);
+		if (!add_data)
+			goto out;
+	}
+
+	uvcb_attest = kzalloc(sizeof(*uvcb_attest), GFP_KERNEL);
+	if (!uvcb_attest)
+		goto out;
+
+	ret = uvio_build_uvcb_attest(uvcb_attest, arcb,  measurement, add_data, uvio_attest);
+	if (ret)
+		goto out;
+
+	uv_call_sched(0, (u64)uvcb_attest);
+
+	uv_ioctl->uv_rc = uvcb_attest->header.rc;
+	uv_ioctl->uv_rrc = uvcb_attest->header.rrc;
+
+	ret = uvio_copy_attest_result_to_user(uvcb_attest, uv_ioctl, measurement, add_data,
+					      uvio_attest);
+out:
+	kvfree(arcb);
+	kvfree(measurement);
+	kvfree(add_data);
+	kfree(uvio_attest);
+	kfree(uvcb_attest);
+	return ret;
+}
+
+static int uvio_copy_and_check_ioctl(struct uvio_ioctl_cb *ioctl, void __user *argp)
+{
+	if (copy_from_user(ioctl, argp, sizeof(*ioctl)))
+		return -EFAULT;
+	if (ioctl->flags != 0)
+		return -EINVAL;
+	if (memchr_inv(ioctl->reserved14, 0, sizeof(ioctl->reserved14)))
+		return -EINVAL;
+
+	return 0;
+}
+
+/*
+ * IOCTL entry point for the Ultravisor device.
+ */
+static long uvio_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	struct uvio_ioctl_cb uv_ioctl = { };
+	long ret;
+
+	switch (cmd) {
+	case UVIO_IOCTL_ATT:
+		ret = uvio_copy_and_check_ioctl(&uv_ioctl, argp);
+		if (ret)
+			return ret;
+		ret = uvio_attestation(&uv_ioctl);
+		break;
+	default:
+		ret = -ENOIOCTLCMD;
+		break;
+	}
+	if (ret)
+		return ret;
+
+	if (copy_to_user(argp, &uv_ioctl, sizeof(uv_ioctl)))
+		ret = -EFAULT;
+
+	return ret;
+}
+
+static const struct file_operations uvio_dev_fops = {
+	.owner = THIS_MODULE,
+	.unlocked_ioctl = uvio_ioctl,
+	.llseek = no_llseek,
+};
+
+static struct miscdevice uvio_dev_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = UVIO_DEVICE_NAME,
+	.fops = &uvio_dev_fops,
+};
+
+static void __exit uvio_dev_exit(void)
+{
+	misc_deregister(&uvio_dev_miscdev);
+}
+
+static int __init uvio_dev_init(void)
+{
+	if (!test_facility(158))
+		return -ENXIO;
+	return misc_register(&uvio_dev_miscdev);
+}
+
+module_init(uvio_dev_init);
+module_exit(uvio_dev_exit);
+
+MODULE_AUTHOR("IBM Corporation");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Ultravisor UAPI driver");

include/kvm/arm_arch_timer.h

···
 int kvm_arm_timer_get_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr);
 int kvm_arm_timer_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr);
 
-bool kvm_timer_is_pending(struct kvm_vcpu *vcpu);
-
 u64 kvm_phys_timer_read(void);
 
 void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu);

include/kvm/arm_hypercalls.h

···
 	vcpu_set_reg(vcpu, 3, a3);
 }
 
+struct kvm_one_reg;
+
+void kvm_arm_init_hypercalls(struct kvm *kvm);
+int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu);
+int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices);
+int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
+int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
+
 #endif

include/kvm/arm_pmu.h

···
 	struct perf_event *perf_event;
 };
 
+struct kvm_pmu_events {
+	u32 events_host;
+	u32 events_guest;
+};
+
 struct kvm_pmu {
-	int irq_num;
+	struct irq_work overflow_work;
+	struct kvm_pmu_events events;
 	struct kvm_pmc pmc[ARMV8_PMU_MAX_COUNTERS];
 	DECLARE_BITMAP(chained, ARMV8_PMU_MAX_COUNTER_PAIRS);
+	int irq_num;
 	bool created;
 	bool irq_level;
-	struct irq_work overflow_work;
 };
 
 struct arm_pmu_entry {
···
 int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu,
 			    struct kvm_device_attr *attr);
 int kvm_arm_pmu_v3_enable(struct kvm_vcpu *vcpu);
+
+struct kvm_pmu_events *kvm_get_pmu_events(void);
+void kvm_vcpu_pmu_restore_guest(struct kvm_vcpu *vcpu);
+void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu);
+
+#define kvm_vcpu_has_pmu(vcpu)					\
+	(test_bit(KVM_ARM_VCPU_PMU_V3, (vcpu)->arch.features))
+
+/*
+ * Updates the vcpu's view of the pmu events for this cpu.
+ * Must be called before every vcpu run after disabling interrupts, to ensure
+ * that an interrupt cannot fire and update the structure.
+ */
+#define kvm_pmu_update_vcpu_events(vcpu)				\
+	do {								\
+		if (!has_vhe() && kvm_vcpu_has_pmu(vcpu))		\
+			vcpu->arch.pmu.events = *kvm_get_pmu_events();	\
+	} while (0)
+
 #else
 struct kvm_pmu {
 };
···
 {
 	return 0;
 }
+
+#define kvm_vcpu_has_pmu(vcpu)		({ false; })
+static inline void kvm_pmu_update_vcpu_events(struct kvm_vcpu *vcpu) {}
+static inline void kvm_vcpu_pmu_restore_guest(struct kvm_vcpu *vcpu) {}
+static inline void kvm_vcpu_pmu_restore_host(struct kvm_vcpu *vcpu) {}
 
 #endif
 

include/kvm/arm_psci.h

···
 
 int kvm_psci_call(struct kvm_vcpu *vcpu);
 
-struct kvm_one_reg;
-
-int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu);
-int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices);
-int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
-int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
-
 #endif /* __KVM_ARM_PSCI_H__ */

include/kvm/arm_vgic.h

···
 
 	/* Implementation revision as reported in the GICD_IIDR */
 	u32			implementation_rev;
+#define KVM_VGIC_IMP_REV_2	2 /* GICv2 restorable groups */
+#define KVM_VGIC_IMP_REV_3	3 /* GICv3 GICR_CTLR.{IW,CES,RWP} */
+#define KVM_VGIC_IMP_REV_LATEST	KVM_VGIC_IMP_REV_3
 
 	/* Userspace can write to GICv2 IGROUPR */
 	bool			v2_groups_user_writable;
···
 	struct vgic_io_device	rd_iodev;
 	struct vgic_redist_region *rdreg;
 	u32 rdreg_index;
+	atomic_t syncr_busy;
 
 	/* Contains the attributes and gpa of the LPI pending tables. */
 	u64 pendbaser;
-
-	bool lpis_enabled;
+	/* GICR_CTLR.{ENABLE_LPIS,RWP} */
+	atomic_t ctlr;
 
 	/* Cache guest priority bits */
 	u32 num_pri_bits;

include/linux/kvm_host.h

···
 
 struct kvm_xen_evtchn {
 	u32 port;
-	u32 vcpu;
+	u32 vcpu_id;
+	int vcpu_idx;
 	u32 priority;
 };
 
···
 	 * and is accessed atomically.
 	 */
 	atomic_t online_vcpus;
+	int max_vcpus;
 	int created_vcpus;
 	int last_boosted_vcpu;
 	struct list_head vm_list;

include/uapi/linux/kvm.h

···
 #define KVM_SYSTEM_EVENT_SHUTDOWN       1
 #define KVM_SYSTEM_EVENT_RESET          2
 #define KVM_SYSTEM_EVENT_CRASH          3
+#define KVM_SYSTEM_EVENT_WAKEUP         4
+#define KVM_SYSTEM_EVENT_SUSPEND        5
+#define KVM_SYSTEM_EVENT_SEV_TERM       6
 			__u32 type;
 			__u32 ndata;
 			union {
···
 #define KVM_MP_STATE_OPERATING         7
 #define KVM_MP_STATE_LOAD              8
 #define KVM_MP_STATE_AP_RESET_HOLD     9
+#define KVM_MP_STATE_SUSPENDED         10
 
 struct kvm_mp_state {
 	__u32 mp_state;
···
 #define KVM_CAP_S390_MEM_OP_EXTENSION 211
 #define KVM_CAP_PMU_CAPABILITY 212
 #define KVM_CAP_DISABLE_QUIRKS2 213
-/* #define KVM_CAP_VM_TSC_CONTROL 214 */
+#define KVM_CAP_VM_TSC_CONTROL 214
 #define KVM_CAP_SYSTEM_EVENT_DATA 215
+#define KVM_CAP_ARM_SYSTEM_SUSPEND 216
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
···
 #define KVM_XEN_HVM_CONFIG_SHARED_INFO		(1 << 2)
 #define KVM_XEN_HVM_CONFIG_RUNSTATE		(1 << 3)
 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL	(1 << 4)
+#define KVM_XEN_HVM_CONFIG_EVTCHN_SEND		(1 << 5)
 
 struct kvm_xen_hvm_config {
 	__u32 flags;
···
 #define KVM_SET_PIT2              _IOW(KVMIO,  0xa0, struct kvm_pit_state2)
 /* Available with KVM_CAP_PPC_GET_PVINFO */
 #define KVM_PPC_GET_PVINFO	  _IOW(KVMIO,  0xa1, struct kvm_ppc_pvinfo)
-/* Available with KVM_CAP_TSC_CONTROL */
+/* Available with KVM_CAP_TSC_CONTROL for a vCPU, or with
+*  KVM_CAP_VM_TSC_CONTROL to set defaults for a VM */
 #define KVM_SET_TSC_KHZ           _IO(KVMIO,  0xa2)
 #define KVM_GET_TSC_KHZ           _IO(KVMIO,  0xa3)
 /* Available with KVM_CAP_PCI_2_3 */
···
 		struct {
 			__u64 gfn;
 		} shared_info;
+		struct {
+			__u32 send_port;
+			__u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */
+			__u32 flags;
+#define KVM_XEN_EVTCHN_DEASSIGN		(1 << 0)
+#define KVM_XEN_EVTCHN_UPDATE		(1 << 1)
+#define KVM_XEN_EVTCHN_RESET		(1 << 2)
+			/*
+			 * Events sent by the guest are either looped back to
+			 * the guest itself (potentially on a different port#)
+			 * or signalled via an eventfd.
+			 */
+			union {
+				struct {
+					__u32 port;
+					__u32 vcpu;
+					__u32 priority;
+				} port;
+				struct {
+					__u32 port; /* Zero for eventfd */
+					__s32 fd;
+				} eventfd;
+				__u32 padding[4];
+			} deliver;
+		} evtchn;
+		__u32 xen_version;
 		__u64 pad[8];
 	} u;
 };
···
 #define KVM_XEN_ATTR_TYPE_LONG_MODE		0x0
 #define KVM_XEN_ATTR_TYPE_SHARED_INFO		0x1
 #define KVM_XEN_ATTR_TYPE_UPCALL_VECTOR		0x2
+/* Available with KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND */
+#define KVM_XEN_ATTR_TYPE_EVTCHN		0x3
+#define KVM_XEN_ATTR_TYPE_XEN_VERSION		0x4
 
 /* Per-vCPU Xen attributes */
 #define KVM_XEN_VCPU_GET_ATTR	_IOWR(KVMIO, 0xca, struct kvm_xen_vcpu_attr)
 #define KVM_XEN_VCPU_SET_ATTR	_IOW(KVMIO,  0xcb, struct kvm_xen_vcpu_attr)
+
+/* Available with KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND */
+#define KVM_XEN_HVM_EVTCHN_SEND	_IOW(KVMIO,  0xd0, struct kvm_irq_routing_xen_evtchn)
 
 #define KVM_GET_SREGS2             _IOR(KVMIO,  0xcc, struct kvm_sregs2)
 #define KVM_SET_SREGS2             _IOW(KVMIO,  0xcd, struct kvm_sregs2)
···
 			__u64 time_blocked;
 			__u64 time_offline;
 		} runstate;
+		__u32 vcpu_id;
+		struct {
+			__u32 port;
+			__u32 priority;
+			__u64 expires_ns;
+		} timer;
+		__u8 vector;
 	} u;
 };
 
···
 #define KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT	0x3
 #define KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA	0x4
 #define KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST	0x5
+/* Available with KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND */
+#define KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID		0x6
+#define KVM_XEN_VCPU_ATTR_TYPE_TIMER		0x7
+#define KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR	0x8
 
 /* Secure Encrypted Virtualization command */
 enum sev_cmd_id {

init/Kconfig

···
 	depends on CC_HAS_ASM_GOTO
 	def_bool $(success,echo 'int foo(int x) { asm goto ("": "=r"(x) ::: bar); return x; bar: return 0; }' | $(CC) -x c - -c -o /dev/null)
 
+config CC_HAS_ASM_GOTO_TIED_OUTPUT
+	depends on CC_HAS_ASM_GOTO_OUTPUT
+	# Detect buggy gcc and clang, fixed in gcc-11 clang-14.
+	def_bool $(success,echo 'int foo(int *x) { asm goto (".long (%l[bar]) - .\n": "+m"(*x) ::: bar); return *x; bar: return 0; }' | $CC -x c - -c -o /dev/null)
+
 config TOOLS_SUPPORT_RELR
 	def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh)
 

scripts/kallsyms.c

···
 		".L",			/* local labels, .LBB,.Ltmpxxx,.L__unnamed_xx,.LASANPC, etc. */
 		"__crc_",		/* modversions */
 		"__efistub_",		/* arm64 EFI stub namespace */
-		"__kvm_nvhe_",		/* arm64 non-VHE KVM namespace */
+		"__kvm_nvhe_$",		/* arm64 local symbols in non-VHE KVM namespace */
+		"__kvm_nvhe_.L",	/* arm64 local symbols in non-VHE KVM namespace */
 		"__AArch64ADRPThunk_",	/* arm64 lld */
 		"__ARMV5PILongThunk_",	/* arm lld */
 		"__ARMV7PILongThunk_",

tools/include/linux/arm-smccc.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2015, Linaro Limited
+ */
+#ifndef __LINUX_ARM_SMCCC_H
+#define __LINUX_ARM_SMCCC_H
+
+#include <linux/const.h>
+
+/*
+ * This file provides common defines for ARM SMC Calling Convention as
+ * specified in
+ * https://developer.arm.com/docs/den0028/latest
+ *
+ * This code is up-to-date with version DEN 0028 C
+ */
+
+#define ARM_SMCCC_STD_CALL	        _AC(0,U)
+#define ARM_SMCCC_FAST_CALL	        _AC(1,U)
+#define ARM_SMCCC_TYPE_SHIFT		31
+
+#define ARM_SMCCC_SMC_32		0
+#define ARM_SMCCC_SMC_64		1
+#define ARM_SMCCC_CALL_CONV_SHIFT	30
+
+#define ARM_SMCCC_OWNER_MASK		0x3F
+#define ARM_SMCCC_OWNER_SHIFT		24
+
+#define ARM_SMCCC_FUNC_MASK		0xFFFF
+
+#define ARM_SMCCC_IS_FAST_CALL(smc_val)	\
+	((smc_val) & (ARM_SMCCC_FAST_CALL << ARM_SMCCC_TYPE_SHIFT))
+#define ARM_SMCCC_IS_64(smc_val) \
+	((smc_val) & (ARM_SMCCC_SMC_64 << ARM_SMCCC_CALL_CONV_SHIFT))
+#define ARM_SMCCC_FUNC_NUM(smc_val)	((smc_val) & ARM_SMCCC_FUNC_MASK)
+#define ARM_SMCCC_OWNER_NUM(smc_val) \
+	(((smc_val) >> ARM_SMCCC_OWNER_SHIFT) & ARM_SMCCC_OWNER_MASK)
+
+#define ARM_SMCCC_CALL_VAL(type, calling_convention, owner, func_num) \
+	(((type) << ARM_SMCCC_TYPE_SHIFT) | \
+	((calling_convention) << ARM_SMCCC_CALL_CONV_SHIFT) | \
+	(((owner) & ARM_SMCCC_OWNER_MASK) << ARM_SMCCC_OWNER_SHIFT) | \
+	((func_num) & ARM_SMCCC_FUNC_MASK))
+
+#define ARM_SMCCC_OWNER_ARCH		0
+#define ARM_SMCCC_OWNER_CPU		1
+#define ARM_SMCCC_OWNER_SIP		2
+#define ARM_SMCCC_OWNER_OEM		3
+#define ARM_SMCCC_OWNER_STANDARD	4
+#define ARM_SMCCC_OWNER_STANDARD_HYP	5
+#define ARM_SMCCC_OWNER_VENDOR_HYP	6
+#define ARM_SMCCC_OWNER_TRUSTED_APP	48
+#define ARM_SMCCC_OWNER_TRUSTED_APP_END	49
+#define ARM_SMCCC_OWNER_TRUSTED_OS	50
+#define ARM_SMCCC_OWNER_TRUSTED_OS_END	63
+
+#define ARM_SMCCC_FUNC_QUERY_CALL_UID  0xff01
+
+#define ARM_SMCCC_QUIRK_NONE		0
+#define ARM_SMCCC_QUIRK_QCOM_A6		1 /* Save/restore register a6 */
+
+#define ARM_SMCCC_VERSION_1_0		0x10000
+#define ARM_SMCCC_VERSION_1_1		0x10001
+#define ARM_SMCCC_VERSION_1_2		0x10002
+#define ARM_SMCCC_VERSION_1_3		0x10003
+
+#define ARM_SMCCC_1_3_SVE_HINT		0x10000
+
+#define ARM_SMCCC_VERSION_FUNC_ID					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 0)
+
+#define ARM_SMCCC_ARCH_FEATURES_FUNC_ID					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 1)
+
+#define ARM_SMCCC_ARCH_SOC_ID						\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 2)
+
+#define ARM_SMCCC_ARCH_WORKAROUND_1					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 0x8000)
+
+#define ARM_SMCCC_ARCH_WORKAROUND_2					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 0x7fff)
+
+#define ARM_SMCCC_ARCH_WORKAROUND_3					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   0, 0x3fff)
+
+#define ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID				\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   ARM_SMCCC_OWNER_VENDOR_HYP,			\
+			   ARM_SMCCC_FUNC_QUERY_CALL_UID)
+
+/* KVM UID value: 28b46fb6-2ec5-11e9-a9ca-4b564d003a74 */
+#define ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_0	0xb66fb428U
+#define ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_1	0xe911c52eU
+#define ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_2	0x564bcaa9U
+#define ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_3	0x743a004dU
+
+/* KVM "vendor specific" services */
+#define ARM_SMCCC_KVM_FUNC_FEATURES		0
+#define ARM_SMCCC_KVM_FUNC_PTP			1
+#define ARM_SMCCC_KVM_FUNC_FEATURES_2		127
+#define ARM_SMCCC_KVM_NUM_FUNCS			128
+
+#define ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID			\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   ARM_SMCCC_OWNER_VENDOR_HYP,			\
+			   ARM_SMCCC_KVM_FUNC_FEATURES)
+
+#define SMCCC_ARCH_WORKAROUND_RET_UNAFFECTED	1
+
+/*
+ * ptp_kvm is a feature used for time sync between vm and host.
+ * ptp_kvm module in guest kernel will get service from host using
+ * this hypercall ID.
+ */
+#define ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID				\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
+			   ARM_SMCCC_SMC_32,				\
+			   ARM_SMCCC_OWNER_VENDOR_HYP,			\
+			   ARM_SMCCC_KVM_FUNC_PTP)
+
+/* ptp_kvm counter type ID */
+#define KVM_PTP_VIRT_COUNTER			0
+#define KVM_PTP_PHYS_COUNTER			1
+
+/* Paravirtualised time calls (defined by ARM DEN0057A) */
+#define ARM_SMCCC_HV_PV_TIME_FEATURES				\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_64,			\
+			   ARM_SMCCC_OWNER_STANDARD_HYP,	\
+			   0x20)
+
+#define ARM_SMCCC_HV_PV_TIME_ST					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_64,			\
+			   ARM_SMCCC_OWNER_STANDARD_HYP,	\
+			   0x21)
+
+/* TRNG entropy source calls (defined by ARM DEN0098) */
+#define ARM_SMCCC_TRNG_VERSION					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_32,			\
+			   ARM_SMCCC_OWNER_STANDARD,		\
+			   0x50)
+
+#define ARM_SMCCC_TRNG_FEATURES					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_32,			\
+			   ARM_SMCCC_OWNER_STANDARD,		\
+			   0x51)
+
+#define ARM_SMCCC_TRNG_GET_UUID					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_32,			\
+			   ARM_SMCCC_OWNER_STANDARD,		\
+			   0x52)
+
+#define ARM_SMCCC_TRNG_RND32					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_32,			\
+			   ARM_SMCCC_OWNER_STANDARD,		\
+			   0x53)
+
+#define ARM_SMCCC_TRNG_RND64					\
+	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,			\
+			   ARM_SMCCC_SMC_64,			\
+			   ARM_SMCCC_OWNER_STANDARD,		\
+			   0x53)
+
+/*
+ * Return codes defined in ARM DEN 0070A
+ * ARM DEN 0070A is now merged/consolidated into ARM DEN 0028 C
+ */
+#define SMCCC_RET_SUCCESS			0
+#define SMCCC_RET_NOT_SUPPORTED			-1
+#define SMCCC_RET_NOT_REQUIRED			-2
+#define SMCCC_RET_INVALID_PARAMETER		-3
+
+#endif /*__LINUX_ARM_SMCCC_H*/

tools/testing/selftests/Makefile

···
 TARGETS += cpu-hotplug
 TARGETS += damon
 TARGETS += drivers/dma-buf
+TARGETS += drivers/s390x/uvdevice
 TARGETS += efivarfs
 TARGETS += exec
 TARGETS += filesystems

tools/testing/selftests/drivers/.gitignore

···
 # SPDX-License-Identifier: GPL-2.0-only
 /dma-buf/udmabuf
+/s390x/uvdevice/test_uvdevice

tools/testing/selftests/drivers/s390x/uvdevice/Makefile

···
+include ../../../../../build/Build.include
+
+UNAME_M := $(shell uname -m)
+
+ifneq ($(UNAME_M),s390x)
+nothing:
+.PHONY: all clean run_tests install
+.SILENT:
+else
+
+TEST_GEN_PROGS := test_uvdevice
+
+top_srcdir ?= ../../../../../..
+KSFT_KHDR_INSTALL := 1
+khdr_dir = $(top_srcdir)/usr/include
+LINUX_TOOL_ARCH_INCLUDE = $(top_srcdir)/tools/arch/$(ARCH)/include
+
+CFLAGS += -Wall -Werror -static -I$(khdr_dir) -I$(LINUX_TOOL_ARCH_INCLUDE)
+
+include ../../../lib.mk
+
+endif

tools/testing/selftests/drivers/s390x/uvdevice/config

···
+CONFIG_S390_UV_UAPI=y

tools/testing/selftests/drivers/s390x/uvdevice/test_uvdevice.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  selftest for the Ultravisor UAPI device
+ *
+ *  Copyright IBM Corp. 2022
+ *  Author(s): Steffen Eiden <[e-mail]>
+ */
+
+#include <stdint.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+
+#include <asm/uvdevice.h>
+
+#include "../../../kselftest_harness.h"
+
+#define UV_PATH  "/dev/uv"
+#define BUFFER_SIZE 0x200
+FIXTURE(uvio_fixture) {
+	int uv_fd;
+	struct uvio_ioctl_cb uvio_ioctl;
+	uint8_t buffer[BUFFER_SIZE];
+	__u64 fault_page;
+};
+
+FIXTURE_VARIANT(uvio_fixture) {
+	unsigned long ioctl_cmd;
+	uint32_t arg_size;
+};
+
+FIXTURE_VARIANT_ADD(uvio_fixture, att) {
+	.ioctl_cmd = UVIO_IOCTL_ATT,
+	.arg_size = sizeof(struct uvio_attest),
+};
+
+FIXTURE_SETUP(uvio_fixture)
+{
+	self->uv_fd = open(UV_PATH, O_ACCMODE);
+
+	self->uvio_ioctl.argument_addr = (__u64)self->buffer;
+	self->uvio_ioctl.argument_len = variant->arg_size;
+	self->fault_page =
+		(__u64)mmap(NULL, (size_t)getpagesize(), PROT_NONE, MAP_ANONYMOUS, -1, 0);
+}
+
+FIXTURE_TEARDOWN(uvio_fixture)
+{
+	if (self->uv_fd)
+		close(self->uv_fd);
+	munmap((void *)self->fault_page, (size_t)getpagesize());
+}
+
+TEST_F(uvio_fixture, fault_ioctl_arg)
+{
+	int rc, errno_cache;
+
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, NULL);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, self->fault_page);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+}
+
+TEST_F(uvio_fixture, fault_uvio_arg)
+{
+	int rc, errno_cache;
+
+	self->uvio_ioctl.argument_addr = 0;
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+
+	self->uvio_ioctl.argument_addr = self->fault_page;
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+}
+
+/*
+ * Test to verify that IOCTLs with invalid values in the ioctl_control block
+ * are rejected.
+ */
+TEST_F(uvio_fixture, inval_ioctl_cb)
+{
+	int rc, errno_cache;
+
+	self->uvio_ioctl.argument_len = 0;
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+
+	self->uvio_ioctl.argument_len = (uint32_t)-1;
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+	self->uvio_ioctl.argument_len = variant->arg_size;
+
+	self->uvio_ioctl.flags = (uint32_t)-1;
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+	self->uvio_ioctl.flags = 0;
+
+	memset(self->uvio_ioctl.reserved14, 0xff, sizeof(self->uvio_ioctl.reserved14));
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+
+	memset(&self->uvio_ioctl, 0x11, sizeof(self->uvio_ioctl));
+	rc = ioctl(self->uv_fd, variant->ioctl_cmd, &self->uvio_ioctl);
+	ASSERT_EQ(rc, -1);
+}
+
+TEST_F(uvio_fixture, inval_ioctl_cmd)
+{
+	int rc, errno_cache;
+	uint8_t nr = _IOC_NR(variant->ioctl_cmd);
+	unsigned long cmds[] = {
+		_IOWR('a', nr, struct uvio_ioctl_cb),
+		_IOWR(UVIO_TYPE_UVC, nr, int),
+		_IO(UVIO_TYPE_UVC, nr),
+		_IOR(UVIO_TYPE_UVC, nr, struct uvio_ioctl_cb),
+		_IOW(UVIO_TYPE_UVC, nr, struct uvio_ioctl_cb),
+	};
+
+	for (size_t i = 0; i < ARRAY_SIZE(cmds); i++) {
+		rc = ioctl(self->uv_fd, cmds[i], &self->uvio_ioctl);
+		errno_cache = errno;
+		ASSERT_EQ(rc, -1);
+		ASSERT_EQ(errno_cache, ENOTTY);
+	}
+}
+
+struct test_attest_buffer {
+	uint8_t arcb[0x180];
+	uint8_t meas[64];
+	uint8_t add[32];
+};
+
+FIXTURE(attest_fixture) {
+	int uv_fd;
+	struct uvio_ioctl_cb uvio_ioctl;
+	struct uvio_attest uvio_attest;
+	struct test_attest_buffer attest_buffer;
+	__u64 fault_page;
+};
+
+FIXTURE_SETUP(attest_fixture)
+{
+	self->uv_fd = open(UV_PATH, O_ACCMODE);
+
+	self->uvio_ioctl.argument_addr = (__u64)&self->uvio_attest;
+	self->uvio_ioctl.argument_len = sizeof(self->uvio_attest);
+
+	self->uvio_attest.arcb_addr = (__u64)&self->attest_buffer.arcb;
+	self->uvio_attest.arcb_len = sizeof(self->attest_buffer.arcb);
+
+	self->uvio_attest.meas_addr = (__u64)&self->attest_buffer.meas;
+	self->uvio_attest.meas_len = sizeof(self->attest_buffer.meas);
+
+	self->uvio_attest.add_data_addr = (__u64)&self->attest_buffer.add;
+	self->uvio_attest.add_data_len = sizeof(self->attest_buffer.add);
+	self->fault_page =
+		(__u64)mmap(NULL, (size_t)getpagesize(), PROT_NONE, MAP_ANONYMOUS, -1, 0);
+}
+
+FIXTURE_TEARDOWN(attest_fixture)
+{
+	if (self->uv_fd)
+		close(self->uv_fd);
+	munmap((void *)self->fault_page, (size_t)getpagesize());
+}
+
+static void att_inval_sizes_test(uint32_t *size, uint32_t max_size, bool test_zero,
+				 struct __test_metadata *_metadata,
+				 FIXTURE_DATA(attest_fixture) *self)
+{
+	int rc, errno_cache;
+	uint32_t tmp = *size;
+
+	if (test_zero) {
+		*size = 0;
+		rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+		errno_cache = errno;
+		ASSERT_EQ(rc, -1);
+		ASSERT_EQ(errno_cache, EINVAL);
+	}
+	*size = max_size + 1;
+	rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+	*size = tmp;
+}
+
+/*
+ * Test to verify that attestation IOCTLs with invalid values in the UVIO
+ * attestation control block are rejected.
+ */
+TEST_F(attest_fixture, att_inval_request)
+{
+	int rc, errno_cache;
+
+	att_inval_sizes_test(&self->uvio_attest.add_data_len, UVIO_ATT_ADDITIONAL_MAX_LEN,
+			     false, _metadata, self);
+	att_inval_sizes_test(&self->uvio_attest.meas_len, UVIO_ATT_MEASUREMENT_MAX_LEN,
+			     true, _metadata, self);
+	att_inval_sizes_test(&self->uvio_attest.arcb_len, UVIO_ATT_ARCB_MAX_LEN,
+			     true, _metadata, self);
+
+	self->uvio_attest.reserved136 = (uint16_t)-1;
+	rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EINVAL);
+
+	memset(&self->uvio_attest, 0x11, sizeof(self->uvio_attest));
+	rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+	ASSERT_EQ(rc, -1);
+}
+
+static void att_inval_addr_test(__u64 *addr, struct __test_metadata *_metadata,
+				FIXTURE_DATA(attest_fixture) *self)
+{
+	int rc, errno_cache;
+	__u64 tmp = *addr;
+
+	*addr = 0;
+	rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+	*addr = self->fault_page;
+	rc = ioctl(self->uv_fd, UVIO_IOCTL_ATT, &self->uvio_ioctl);
+	errno_cache = errno;
+	ASSERT_EQ(rc, -1);
+	ASSERT_EQ(errno_cache, EFAULT);
+	*addr = tmp;
+}
+
+TEST_F(attest_fixture, att_inval_addr)
+{
+	att_inval_addr_test(&self->uvio_attest.arcb_addr, _metadata, self);
+	att_inval_addr_test(&self->uvio_attest.add_data_addr, _metadata, self);
+	att_inval_addr_test(&self->uvio_attest.meas_addr, _metadata, self);
+}
+
+static void __attribute__((constructor)) __constructor_order_last(void)
+{
+	if (!__constructor_order)
+		__constructor_order = _CONSTRUCTOR_ORDER_BACKWARD;
+}
+
+int main(int argc, char **argv)
+{
+	int fd = open(UV_PATH, O_ACCMODE);
+
+	if (fd < 0)
+		ksft_exit_skip("No uv-device or cannot access " UV_PATH  "\n"
+			       "Enable CONFIG_S390_UV_UAPI and check the access rights on "
+			       UV_PATH ".\n");
+	close(fd);
+	return test_harness_run(argc, argv);
+}

tools/testing/selftests/kvm/.gitignore

···
 /aarch64/arch_timer
 /aarch64/debug-exceptions
 /aarch64/get-reg-list
-/aarch64/psci_cpu_on_test
+/aarch64/hypercalls
+/aarch64/psci_test
 /aarch64/vcpu_width_config
 /aarch64/vgic_init
 /aarch64/vgic_irq
···
 /x86_64/debug_regs
 /x86_64/evmcs_test
 /x86_64/emulator_error_test
+/x86_64/fix_hypercall_test
 /x86_64/get_msr_index_features
 /x86_64/kvm_clock_test
 /x86_64/kvm_pv_test
···
 /x86_64/xen_shinfo_test
 /x86_64/xen_vmcall_test
 /x86_64/xss_msr_test
-/x86_64/vmx_pmu_msrs_test
+/x86_64/vmx_pmu_caps_test
 /access_tracking_perf_test
 /demand_paging_test
 /dirty_log_test

tools/testing/selftests/kvm/Makefile

···
 TEST_GEN_PROGS_x86_64 += x86_64/get_msr_index_features
 TEST_GEN_PROGS_x86_64 += x86_64/evmcs_test
 TEST_GEN_PROGS_x86_64 += x86_64/emulator_error_test
+TEST_GEN_PROGS_x86_64 += x86_64/fix_hypercall_test
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_clock
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_cpuid
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_features
···
 TEST_GEN_PROGS_x86_64 += x86_64/vmx_preemption_timer_test
 TEST_GEN_PROGS_x86_64 += x86_64/svm_vmcall_test
 TEST_GEN_PROGS_x86_64 += x86_64/svm_int_ctl_test
+TEST_GEN_PROGS_x86_64 += x86_64/tsc_scaling_sync
 TEST_GEN_PROGS_x86_64 += x86_64/sync_regs_test
 TEST_GEN_PROGS_x86_64 += x86_64/userspace_io_test
 TEST_GEN_PROGS_x86_64 += x86_64/userspace_msr_exit_test
···
 TEST_GEN_PROGS_x86_64 += x86_64/xss_msr_test
 TEST_GEN_PROGS_x86_64 += x86_64/debug_regs
 TEST_GEN_PROGS_x86_64 += x86_64/tsc_msrs_test
-TEST_GEN_PROGS_x86_64 += x86_64/vmx_pmu_msrs_test
+TEST_GEN_PROGS_x86_64 += x86_64/vmx_pmu_caps_test
 TEST_GEN_PROGS_x86_64 += x86_64/xen_shinfo_test
 TEST_GEN_PROGS_x86_64 += x86_64/xen_vmcall_test
 TEST_GEN_PROGS_x86_64 += x86_64/sev_migrate_tests
···
 TEST_GEN_PROGS_aarch64 += aarch64/arch_timer
 TEST_GEN_PROGS_aarch64 += aarch64/debug-exceptions
 TEST_GEN_PROGS_aarch64 += aarch64/get-reg-list
-TEST_GEN_PROGS_aarch64 += aarch64/psci_cpu_on_test
+TEST_GEN_PROGS_aarch64 += aarch64/hypercalls
+TEST_GEN_PROGS_aarch64 += aarch64/psci_test
 TEST_GEN_PROGS_aarch64 += aarch64/vcpu_width_config
 TEST_GEN_PROGS_aarch64 += aarch64/vgic_init
 TEST_GEN_PROGS_aarch64 += aarch64/vgic_irq

tools/testing/selftests/kvm/aarch64/get-reg-list.c

···
 			    "%s: Unexpected bits set in FW reg id: 0x%llx", config_name(c), id);
 		printf("\tKVM_REG_ARM_FW_REG(%lld),\n", id & 0xffff);
 		break;
+	case KVM_REG_ARM_FW_FEAT_BMAP:
+		TEST_ASSERT(id == KVM_REG_ARM_FW_FEAT_BMAP_REG(id & 0xffff),
+			    "%s: Unexpected bits set in the bitmap feature FW reg id: 0x%llx", config_name(c), id);
+		printf("\tKVM_REG_ARM_FW_FEAT_BMAP_REG(%lld),\n", id & 0xffff);
+		break;
 	case KVM_REG_ARM64_SVE:
 		if (has_cap(c, KVM_CAP_ARM_SVE))
 			printf("\t%s,\n", sve_id_to_str(c, id));
···
 	KVM_REG_ARM_FW_REG(1),		/* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1 */
 	KVM_REG_ARM_FW_REG(2),		/* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2 */
 	KVM_REG_ARM_FW_REG(3),		/* KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3 */
+	KVM_REG_ARM_FW_FEAT_BMAP_REG(0),	/* KVM_REG_ARM_STD_BMAP */
+	KVM_REG_ARM_FW_FEAT_BMAP_REG(1),	/* KVM_REG_ARM_STD_HYP_BMAP */
+	KVM_REG_ARM_FW_FEAT_BMAP_REG(2),	/* KVM_REG_ARM_VENDOR_HYP_BMAP */
 	ARM64_SYS_REG(3, 3, 14, 3, 1),	/* CNTV_CTL_EL0 */
 	ARM64_SYS_REG(3, 3, 14, 3, 2),	/* CNTV_CVAL_EL0 */
 	ARM64_SYS_REG(3, 3, 14, 0, 2),

tools/testing/selftests/kvm/aarch64/hypercalls.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+
+/* hypercalls: Check the ARM64's psuedo-firmware bitmap register interface.
+ *
+ * The test validates the basic hypercall functionalities that are exposed
+ * via the psuedo-firmware bitmap register. This includes the registers'
+ * read/write behavior before and after the VM has started, and if the
+ * hypercalls are properly masked or unmasked to the guest when disabled or
+ * enabled from the KVM userspace, respectively.
+ */
+
+#include <errno.h>
+#include <linux/arm-smccc.h>
+#include <asm/kvm.h>
+#include <kvm_util.h>
+
+#include "processor.h"
+
+#define FW_REG_ULIMIT_VAL(max_feat_bit) (GENMASK(max_feat_bit, 0))
+
+/* Last valid bits of the bitmapped firmware registers */
+#define KVM_REG_ARM_STD_BMAP_BIT_MAX		0
+#define KVM_REG_ARM_STD_HYP_BMAP_BIT_MAX	0
+#define KVM_REG_ARM_VENDOR_HYP_BMAP_BIT_MAX	1
+
+struct kvm_fw_reg_info {
+	uint64_t reg;		/* Register definition */
+	uint64_t max_feat_bit;	/* Bit that represents the upper limit of the feature-map */
+};
+
+#define FW_REG_INFO(r)			\
+	{					\
+		.reg = r,			\
+		.max_feat_bit = r##_BIT_MAX,	\
+	}
+
+static const struct kvm_fw_reg_info fw_reg_info[] = {
+	FW_REG_INFO(KVM_REG_ARM_STD_BMAP),
+	FW_REG_INFO(KVM_REG_ARM_STD_HYP_BMAP),
+	FW_REG_INFO(KVM_REG_ARM_VENDOR_HYP_BMAP),
+};
+
+enum test_stage {
+	TEST_STAGE_REG_IFACE,
+	TEST_STAGE_HVC_IFACE_FEAT_DISABLED,
+	TEST_STAGE_HVC_IFACE_FEAT_ENABLED,
+	TEST_STAGE_HVC_IFACE_FALSE_INFO,
+	TEST_STAGE_END,
+};
+
+static int stage = TEST_STAGE_REG_IFACE;
+
+struct test_hvc_info {
+	uint32_t func_id;
+	uint64_t arg1;
+};
+
+#define TEST_HVC_INFO(f, a1)	\
+	{			\
+		.func_id = f,	\
+		.arg1 = a1,	\
+	}
+
+static const struct test_hvc_info hvc_info[] = {
+	/* KVM_REG_ARM_STD_BMAP */
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_VERSION, 0),
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_FEATURES, ARM_SMCCC_TRNG_RND64),
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_GET_UUID, 0),
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_RND32, 0),
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_RND64, 0),
+
+	/* KVM_REG_ARM_STD_HYP_BMAP */
+	TEST_HVC_INFO(ARM_SMCCC_ARCH_FEATURES_FUNC_ID, ARM_SMCCC_HV_PV_TIME_FEATURES),
+	TEST_HVC_INFO(ARM_SMCCC_HV_PV_TIME_FEATURES, ARM_SMCCC_HV_PV_TIME_ST),
+	TEST_HVC_INFO(ARM_SMCCC_HV_PV_TIME_ST, 0),
+
+	/* KVM_REG_ARM_VENDOR_HYP_BMAP */
+	TEST_HVC_INFO(ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID,
+			ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID),
+	TEST_HVC_INFO(ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID, 0),
+	TEST_HVC_INFO(ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID, KVM_PTP_VIRT_COUNTER),
+};
+
+/* Feed false hypercall info to test the KVM behavior */
+static const struct test_hvc_info false_hvc_info[] = {
+	/* Feature support check against a different family of hypercalls */
+	TEST_HVC_INFO(ARM_SMCCC_TRNG_FEATURES, ARM_SMCCC_VENDOR_HYP_KVM_PTP_FUNC_ID),
+	TEST_HVC_INFO(ARM_SMCCC_ARCH_FEATURES_FUNC_ID, ARM_SMCCC_TRNG_RND64),
+	TEST_HVC_INFO(ARM_SMCCC_HV_PV_TIME_FEATURES, ARM_SMCCC_TRNG_RND64),
+};
+
+static void guest_test_hvc(const struct test_hvc_info *hc_info)
+{
+	unsigned int i;
+	struct arm_smccc_res res;
+	unsigned int hvc_info_arr_sz;
+
+	hvc_info_arr_sz =
+	hc_info == hvc_info ? ARRAY_SIZE(hvc_info) : ARRAY_SIZE(false_hvc_info);
+
+	for (i = 0; i < hvc_info_arr_sz; i++, hc_info++) {
+		memset(&res, 0, sizeof(res));
+		smccc_hvc(hc_info->func_id, hc_info->arg1, 0, 0, 0, 0, 0, 0, &res);
+
+		switch (stage) {
+		case TEST_STAGE_HVC_IFACE_FEAT_DISABLED:
+		case TEST_STAGE_HVC_IFACE_FALSE_INFO:
+			GUEST_ASSERT_3(res.a0 == SMCCC_RET_NOT_SUPPORTED,
+					res.a0, hc_info->func_id, hc_info->arg1);
+			break;
+		case TEST_STAGE_HVC_IFACE_FEAT_ENABLED:
+			GUEST_ASSERT_3(res.a0 != SMCCC_RET_NOT_SUPPORTED,
+					res.a0, hc_info->func_id, hc_info->arg1);
+			break;
+		default:
+			GUEST_ASSERT_1(0, stage);
+		}
+	}
+}
+
+static void guest_code(void)
+{
+	while (stage != TEST_STAGE_END) {
+		switch (stage) {
+		case TEST_STAGE_REG_IFACE:
+			break;
+		case TEST_STAGE_HVC_IFACE_FEAT_DISABLED:
+		case TEST_STAGE_HVC_IFACE_FEAT_ENABLED:
+			guest_test_hvc(hvc_info);
+			break;
+		case TEST_STAGE_HVC_IFACE_FALSE_INFO:
+			guest_test_hvc(false_hvc_info);
+			break;
+		default:
+			GUEST_ASSERT_1(0, stage);
+		}
+
+		GUEST_SYNC(stage);
+	}
+
+	GUEST_DONE();
+}
+
+static int set_fw_reg(struct kvm_vm *vm, uint64_t id, uint64_t val)
+{
+	struct kvm_one_reg reg = {
+		.id = id,
+		.addr = (uint64_t)&val,
+	};
+
+	return _vcpu_ioctl(vm, 0, KVM_SET_ONE_REG, &reg);
+}
+
+static void get_fw_reg(struct kvm_vm *vm, uint64_t id, uint64_t *addr)
+{
+	struct kvm_one_reg reg = {
+		.id = id,
+		.addr = (uint64_t)addr,
+	};
+
+	vcpu_ioctl(vm, 0, KVM_GET_ONE_REG, &reg);
+}
+
+struct st_time {
+	uint32_t rev;
+	uint32_t attr;
+	uint64_t st_time;
+};
+
+#define STEAL_TIME_SIZE		((sizeof(struct st_time) + 63) & ~63)
+#define ST_GPA_BASE		(1 << 30)
+
+static void steal_time_init(struct kvm_vm *vm)
+{
+	uint64_t st_ipa = (ulong)ST_GPA_BASE;
+	unsigned int gpages;
+	struct kvm_device_attr dev = {
+		.group = KVM_ARM_VCPU_PVTIME_CTRL,
+		.attr = KVM_ARM_VCPU_PVTIME_IPA,
+		.addr = (uint64_t)&st_ipa,
+	};
+
+	gpages = vm_calc_num_guest_pages(VM_MODE_DEFAULT, STEAL_TIME_SIZE);
+	vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, ST_GPA_BASE, 1, gpages, 0);
+
+	vcpu_ioctl(vm, 0, KVM_SET_DEVICE_ATTR, &dev);
+}
+
+static void test_fw_regs_before_vm_start(struct kvm_vm *vm)
+{
+	uint64_t val;
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < ARRAY_SIZE(fw_reg_info); i++) {
+		const struct kvm_fw_reg_info *reg_info = &fw_reg_info[i];
+
+		/* First 'read' should be an upper limit of the features supported */
+		get_fw_reg(vm, reg_info->reg, &val);
+		TEST_ASSERT(val == FW_REG_ULIMIT_VAL(reg_info->max_feat_bit),
+			"Expected all the features to be set for reg: 0x%lx; expected: 0x%lx; read: 0x%lx\n",
+			reg_info->reg, FW_REG_ULIMIT_VAL(reg_info->max_feat_bit), val);
+
+		/* Test a 'write' by disabling all the features of the register map */
+		ret = set_fw_reg(vm, reg_info->reg, 0);
+		TEST_ASSERT(ret == 0,
+			"Failed to clear all the features of reg: 0x%lx; ret: %d\n",
+			reg_info->reg, errno);
+
+		get_fw_reg(vm, reg_info->reg, &val);
+		TEST_ASSERT(val == 0,
+			"Expected all the features to be cleared for reg: 0x%lx\n", reg_info->reg);
+
+		/*
+		 * Test enabling a feature that's not supported.
+		 * Avoid this check if all the bits are occupied.
+		 */
+		if (reg_info->max_feat_bit < 63) {
+			ret = set_fw_reg(vm, reg_info->reg, BIT(reg_info->max_feat_bit + 1));
+			TEST_ASSERT(ret != 0 && errno == EINVAL,
+			"Unexpected behavior or return value (%d) while setting an unsupported feature for reg: 0x%lx\n",
+			errno, reg_info->reg);
+		}
+	}
+}
+
+static void test_fw_regs_after_vm_start(struct kvm_vm *vm)
+{
+	uint64_t val;
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < ARRAY_SIZE(fw_reg_info); i++) {
+		const struct kvm_fw_reg_info *reg_info = &fw_reg_info[i];
+
+		/*
+		 * Before starting the VM, the test clears all the bits.
+		 * Check if that's still the case.
+		 */
+		get_fw_reg(vm, reg_info->reg, &val);
+		TEST_ASSERT(val == 0,
+			"Expected all the features to be cleared for reg: 0x%lx\n",
+			reg_info->reg);
+
+		/*
+		 * Since the VM has run at least once, KVM shouldn't allow modification of
+		 * the registers and should return EBUSY. Set the registers and check for
+		 * the expected errno.
+		 */
+		ret = set_fw_reg(vm, reg_info->reg, FW_REG_ULIMIT_VAL(reg_info->max_feat_bit));
+		TEST_ASSERT(ret != 0 && errno == EBUSY,
+		"Unexpected behavior or return value (%d) while setting a feature while VM is running for reg: 0x%lx\n",
+		errno, reg_info->reg);
+	}
+}
+
+static struct kvm_vm *test_vm_create(void)
+{
+	struct kvm_vm *vm;
+
+	vm = vm_create_default(0, 0, guest_code);
+
+	ucall_init(vm, NULL);
+	steal_time_init(vm);
+
+	return vm;
+}
+
+static struct kvm_vm *test_guest_stage(struct kvm_vm *vm)
+{
+	struct kvm_vm *ret_vm = vm;
+
+	pr_debug("Stage: %d\n", stage);
+
+	switch (stage) {
+	case TEST_STAGE_REG_IFACE:
+		test_fw_regs_after_vm_start(vm);
+		break;
+	case TEST_STAGE_HVC_IFACE_FEAT_DISABLED:
+		/* Start a new VM so that all the features are now enabled by default */
+		kvm_vm_free(vm);
+		ret_vm = test_vm_create();
+		break;
+	case TEST_STAGE_HVC_IFACE_FEAT_ENABLED:
+	case TEST_STAGE_HVC_IFACE_FALSE_INFO:
+		break;
+	default:
+		TEST_FAIL("Unknown test stage: %d\n", stage);
+	}
+
+	stage++;
+	sync_global_to_guest(vm, stage);
+
+	return ret_vm;
+}
+
+static void test_run(void)
+{
+	struct kvm_vm *vm;
+	struct ucall uc;
+	bool guest_done = false;
+
+	vm = test_vm_create();
+
+	test_fw_regs_before_vm_start(vm);
+
+	while (!guest_done) {
+		vcpu_run(vm, 0);
+
+		switch (get_ucall(vm, 0, &uc)) {
+		case UCALL_SYNC:
+			vm = test_guest_stage(vm);
+			break;
+		case UCALL_DONE:
+			guest_done = true;
+			break;
+		case UCALL_ABORT:
+			TEST_FAIL("%s at %s:%ld\n\tvalues: 0x%lx, 0x%lx; 0x%lx, stage: %u",
+			(const char *)uc.args[0], __FILE__, uc.args[1],
+			uc.args[2], uc.args[3], uc.args[4], stage);
+			break;
+		default:
+			TEST_FAIL("Unexpected guest exit\n");
+		}
+	}
+
+	kvm_vm_free(vm);
+}
+
+int main(void)
+{
+	setbuf(stdout, NULL);
+
+	test_run();
+	return 0;
+}

tools/testing/selftests/kvm/aarch64/psci_cpu_on_test.c

···
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * psci_cpu_on_test - Test that the observable state of a vCPU targeted by the
- * CPU_ON PSCI call matches what the caller requested.
- *
- * Copyright (c) 2021 Google LLC.
- *
- * This is a regression test for a race between KVM servicing the PSCI call and
- * userspace reading the vCPUs registers.
- */
-
-#define _GNU_SOURCE
-
-#include <linux/psci.h>
-
-#include "kvm_util.h"
-#include "processor.h"
-#include "test_util.h"
-
-#define VCPU_ID_SOURCE 0
-#define VCPU_ID_TARGET 1
-
-#define CPU_ON_ENTRY_ADDR 0xfeedf00dul
-#define CPU_ON_CONTEXT_ID 0xdeadc0deul
-
-static uint64_t psci_cpu_on(uint64_t target_cpu, uint64_t entry_addr,
-			    uint64_t context_id)
-{
-	register uint64_t x0 asm("x0") = PSCI_0_2_FN64_CPU_ON;
-	register uint64_t x1 asm("x1") = target_cpu;
-	register uint64_t x2 asm("x2") = entry_addr;
-	register uint64_t x3 asm("x3") = context_id;
-
-	asm("hvc #0"
-	    : "=r"(x0)
-	    : "r"(x0), "r"(x1), "r"(x2), "r"(x3)
-	    : "memory");
-
-	return x0;
-}
-
-static uint64_t psci_affinity_info(uint64_t target_affinity,
-				   uint64_t lowest_affinity_level)
-{
-	register uint64_t x0 asm("x0") = PSCI_0_2_FN64_AFFINITY_INFO;
-	register uint64_t x1 asm("x1") = target_affinity;
-	register uint64_t x2 asm("x2") = lowest_affinity_level;
-
-	asm("hvc #0"
-	    : "=r"(x0)
-	    : "r"(x0), "r"(x1), "r"(x2)
-	    : "memory");
-
-	return x0;
-}
-
-static void guest_main(uint64_t target_cpu)
-{
-	GUEST_ASSERT(!psci_cpu_on(target_cpu, CPU_ON_ENTRY_ADDR, CPU_ON_CONTEXT_ID));
-	uint64_t target_state;
-
-	do {
-		target_state = psci_affinity_info(target_cpu, 0);
-
-		GUEST_ASSERT((target_state == PSCI_0_2_AFFINITY_LEVEL_ON) ||
-			     (target_state == PSCI_0_2_AFFINITY_LEVEL_OFF));
-	} while (target_state != PSCI_0_2_AFFINITY_LEVEL_ON);
-
-	GUEST_DONE();
-}
-
-int main(void)
-{
-	uint64_t target_mpidr, obs_pc, obs_x0;
-	struct kvm_vcpu_init init;
-	struct kvm_vm *vm;
-	struct ucall uc;
-
-	vm = vm_create(VM_MODE_DEFAULT, DEFAULT_GUEST_PHY_PAGES, O_RDWR);
-	kvm_vm_elf_load(vm, program_invocation_name);
-	ucall_init(vm, NULL);
-
-	vm_ioctl(vm, KVM_ARM_PREFERRED_TARGET, &init);
-	init.features[0] |= (1 << KVM_ARM_VCPU_PSCI_0_2);
-
-	aarch64_vcpu_add_default(vm, VCPU_ID_SOURCE, &init, guest_main);
-
-	/*
-	 * make sure the target is already off when executing the test.
-	 */
-	init.features[0] |= (1 << KVM_ARM_VCPU_POWER_OFF);
-	aarch64_vcpu_add_default(vm, VCPU_ID_TARGET, &init, guest_main);
-
-	get_reg(vm, VCPU_ID_TARGET, KVM_ARM64_SYS_REG(SYS_MPIDR_EL1), &target_mpidr);
-	vcpu_args_set(vm, VCPU_ID_SOURCE, 1, target_mpidr & MPIDR_HWID_BITMASK);
-	vcpu_run(vm, VCPU_ID_SOURCE);
-
-	switch (get_ucall(vm, VCPU_ID_SOURCE, &uc)) {
-	case UCALL_DONE:
-		break;
-	case UCALL_ABORT:
-		TEST_FAIL("%s at %s:%ld", (const char *)uc.args[0], __FILE__,
-			  uc.args[1]);
-		break;
-	default:
-		TEST_FAIL("Unhandled ucall: %lu", uc.cmd);
-	}
-
-	get_reg(vm, VCPU_ID_TARGET, ARM64_CORE_REG(regs.pc), &obs_pc);
-	get_reg(vm, VCPU_ID_TARGET, ARM64_CORE_REG(regs.regs[0]), &obs_x0);
-
-	TEST_ASSERT(obs_pc == CPU_ON_ENTRY_ADDR,
-		    "unexpected target cpu pc: %lx (expected: %lx)",
-		    obs_pc, CPU_ON_ENTRY_ADDR);
-	TEST_ASSERT(obs_x0 == CPU_ON_CONTEXT_ID,
-		    "unexpected target context id: %lx (expected: %lx)",
-		    obs_x0, CPU_ON_CONTEXT_ID);
-
-	kvm_vm_free(vm);
-	return 0;
-}

tools/testing/selftests/kvm/aarch64/psci_test.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * psci_cpu_on_test - Test that the observable state of a vCPU targeted by the
+ * CPU_ON PSCI call matches what the caller requested.
+ *
+ * Copyright (c) 2021 Google LLC.
+ *
+ * This is a regression test for a race between KVM servicing the PSCI call and
+ * userspace reading the vCPUs registers.
+ */
+
+#define _GNU_SOURCE
+
+#include <linux/psci.h>
+
+#include "kvm_util.h"
+#include "processor.h"
+#include "test_util.h"
+
+#define VCPU_ID_SOURCE 0
+#define VCPU_ID_TARGET 1
+
+#define CPU_ON_ENTRY_ADDR 0xfeedf00dul
+#define CPU_ON_CONTEXT_ID 0xdeadc0deul
+
+static uint64_t psci_cpu_on(uint64_t target_cpu, uint64_t entry_addr,
+			    uint64_t context_id)
+{
+	struct arm_smccc_res res;
+
+	smccc_hvc(PSCI_0_2_FN64_CPU_ON, target_cpu, entry_addr, context_id,
+		  0, 0, 0, 0, &res);
+
+	return res.a0;
+}
+
+static uint64_t psci_affinity_info(uint64_t target_affinity,
+				   uint64_t lowest_affinity_level)
+{
+	struct arm_smccc_res res;
+
+	smccc_hvc(PSCI_0_2_FN64_AFFINITY_INFO, target_affinity, lowest_affinity_level,
+		  0, 0, 0, 0, 0, &res);
+
+	return res.a0;
+}
+
+static uint64_t psci_system_suspend(uint64_t entry_addr, uint64_t context_id)
+{
+	struct arm_smccc_res res;
+
+	smccc_hvc(PSCI_1_0_FN64_SYSTEM_SUSPEND, entry_addr, context_id,
+		  0, 0, 0, 0, 0, &res);
+
+	return res.a0;
+}
+
+static uint64_t psci_features(uint32_t func_id)
+{
+	struct arm_smccc_res res;
+
+	smccc_hvc(PSCI_1_0_FN_PSCI_FEATURES, func_id, 0, 0, 0, 0, 0, 0, &res);
+
+	return res.a0;
+}
+
+static void vcpu_power_off(struct kvm_vm *vm, uint32_t vcpuid)
+{
+	struct kvm_mp_state mp_state = {
+		.mp_state = KVM_MP_STATE_STOPPED,
+	};
+
+	vcpu_set_mp_state(vm, vcpuid, &mp_state);
+}
+
+static struct kvm_vm *setup_vm(void *guest_code)
+{
+	struct kvm_vcpu_init init;
+	struct kvm_vm *vm;
+
+	vm = vm_create(VM_MODE_DEFAULT, DEFAULT_GUEST_PHY_PAGES, O_RDWR);
+	kvm_vm_elf_load(vm, program_invocation_name);
+	ucall_init(vm, NULL);
+
+	vm_ioctl(vm, KVM_ARM_PREFERRED_TARGET, &init);
+	init.features[0] |= (1 << KVM_ARM_VCPU_PSCI_0_2);
+
+	aarch64_vcpu_add_default(vm, VCPU_ID_SOURCE, &init, guest_code);
+	aarch64_vcpu_add_default(vm, VCPU_ID_TARGET, &init, guest_code);
+
+	return vm;
+}
+
+static void enter_guest(struct kvm_vm *vm, uint32_t vcpuid)
+{
+	struct ucall uc;
+
+	vcpu_run(vm, vcpuid);
+	if (get_ucall(vm, vcpuid, &uc) == UCALL_ABORT)
+		TEST_FAIL("%s at %s:%ld", (const char *)uc.args[0], __FILE__,
+			  uc.args[1]);
+}
+
+static void assert_vcpu_reset(struct kvm_vm *vm, uint32_t vcpuid)
+{
+	uint64_t obs_pc, obs_x0;
+
+	get_reg(vm, vcpuid, ARM64_CORE_REG(regs.pc), &obs_pc);
+	get_reg(vm, vcpuid, ARM64_CORE_REG(regs.regs[0]), &obs_x0);
+
+	TEST_ASSERT(obs_pc == CPU_ON_ENTRY_ADDR,
+		    "unexpected target cpu pc: %lx (expected: %lx)",
+		    obs_pc, CPU_ON_ENTRY_ADDR);
+	TEST_ASSERT(obs_x0 == CPU_ON_CONTEXT_ID,
+		    "unexpected target context id: %lx (expected: %lx)",
+		    obs_x0, CPU_ON_CONTEXT_ID);
+}
+
+static void guest_test_cpu_on(uint64_t target_cpu)
+{
+	uint64_t target_state;
+
+	GUEST_ASSERT(!psci_cpu_on(target_cpu, CPU_ON_ENTRY_ADDR, CPU_ON_CONTEXT_ID));
+
+	do {
+		target_state = psci_affinity_info(target_cpu, 0);
+
+		GUEST_ASSERT((target_state == PSCI_0_2_AFFINITY_LEVEL_ON) ||
+			     (target_state == PSCI_0_2_AFFINITY_LEVEL_OFF));
+	} while (target_state != PSCI_0_2_AFFINITY_LEVEL_ON);
+
+	GUEST_DONE();
+}
+
+static void host_test_cpu_on(void)
+{
+	uint64_t target_mpidr;
+	struct kvm_vm *vm;
+	struct ucall uc;
+
+	vm = setup_vm(guest_test_cpu_on);
+
+	/*
+	 * make sure the target is already off when executing the test.
+	 */
+	vcpu_power_off(vm, VCPU_ID_TARGET);
+
+	get_reg(vm, VCPU_ID_TARGET, KVM_ARM64_SYS_REG(SYS_MPIDR_EL1), &target_mpidr);
+	vcpu_args_set(vm, VCPU_ID_SOURCE, 1, target_mpidr & MPIDR_HWID_BITMASK);
+	enter_guest(vm, VCPU_ID_SOURCE);
+
+	if (get_ucall(vm, VCPU_ID_SOURCE, &uc) != UCALL_DONE)
+		TEST_FAIL("Unhandled ucall: %lu", uc.cmd);
+
+	assert_vcpu_reset(vm, VCPU_ID_TARGET);
+	kvm_vm_free(vm);
+}
+
+static void enable_system_suspend(struct kvm_vm *vm)
+{
+	struct kvm_enable_cap cap = {
+		.cap = KVM_CAP_ARM_SYSTEM_SUSPEND,
+	};
+
+	vm_enable_cap(vm, &cap);
+}
+
+static void guest_test_system_suspend(void)
+{
+	uint64_t ret;
+
+	/* assert that SYSTEM_SUSPEND is discoverable */
+	GUEST_ASSERT(!psci_features(PSCI_1_0_FN_SYSTEM_SUSPEND));
+	GUEST_ASSERT(!psci_features(PSCI_1_0_FN64_SYSTEM_SUSPEND));
+
+	ret = psci_system_suspend(CPU_ON_ENTRY_ADDR, CPU_ON_CONTEXT_ID);
+	GUEST_SYNC(ret);
+}
+
+static void host_test_system_suspend(void)
+{
+	struct kvm_run *run;
+	struct kvm_vm *vm;
+
+	vm = setup_vm(guest_test_system_suspend);
+	enable_system_suspend(vm);
+
+	vcpu_power_off(vm, VCPU_ID_TARGET);
+	run = vcpu_state(vm, VCPU_ID_SOURCE);
+
+	enter_guest(vm, VCPU_ID_SOURCE);
+
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_SYSTEM_EVENT,
+		    "Unhandled exit reason: %u (%s)",
+		    run->exit_reason, exit_reason_str(run->exit_reason));
+	TEST_ASSERT(run->system_event.type == KVM_SYSTEM_EVENT_SUSPEND,
+		    "Unhandled system event: %u (expected: %u)",
+		    run->system_event.type, KVM_SYSTEM_EVENT_SUSPEND);
+
+	kvm_vm_free(vm);
+}
+
+int main(void)
+{
+	if (!kvm_check_cap(KVM_CAP_ARM_SYSTEM_SUSPEND)) {
+		print_skip("KVM_CAP_ARM_SYSTEM_SUSPEND not supported");
+		exit(KSFT_SKIP);
+	}
+
+	host_test_cpu_on();
+	host_test_system_suspend();
+	return 0;
+}

tools/testing/selftests/kvm/include/aarch64/processor.h

···
 	asm volatile("msr daifset, #3" : : : "memory");
 }
 
+/**
+ * struct arm_smccc_res - Result from SMC/HVC call
+ * @a0-a3 result values from registers 0 to 3
+ */
+struct arm_smccc_res {
+	unsigned long a0;
+	unsigned long a1;
+	unsigned long a2;
+	unsigned long a3;
+};
+
+/**
+ * smccc_hvc - Invoke a SMCCC function using the hvc conduit
+ * @function_id: the SMCCC function to be called
+ * @arg0-arg6: SMCCC function arguments, corresponding to registers x1-x7
+ * @res: pointer to write the return values from registers x0-x3
+ *
+ */
+void smccc_hvc(uint32_t function_id, uint64_t arg0, uint64_t arg1,
+	       uint64_t arg2, uint64_t arg3, uint64_t arg4, uint64_t arg5,
+	       uint64_t arg6, struct arm_smccc_res *res);
+
 #endif /* SELFTEST_KVM_PROCESSOR_H */

tools/testing/selftests/kvm/include/riscv/processor.h

···
 #define SATP_ASID_SHIFT				44
 #define SATP_ASID_MASK				_AC(0xFFFF, UL)
 
-#define SBI_EXT_EXPERIMENTAL_START	0x08000000
-#define SBI_EXT_EXPERIMENTAL_END	0x08FFFFFF
+#define SBI_EXT_EXPERIMENTAL_START		0x08000000
+#define SBI_EXT_EXPERIMENTAL_END		0x08FFFFFF
 
-#define KVM_RISCV_SELFTESTS_SBI_EXT	SBI_EXT_EXPERIMENTAL_END
+#define KVM_RISCV_SELFTESTS_SBI_EXT		SBI_EXT_EXPERIMENTAL_END
+#define KVM_RISCV_SELFTESTS_SBI_UCALL		0
+#define KVM_RISCV_SELFTESTS_SBI_UNEXP		1
 
 struct sbiret {
 	long error;

tools/testing/selftests/kvm/lib/aarch64/processor.c

···
 {
        guest_modes_append_default();
 }
+
+void smccc_hvc(uint32_t function_id, uint64_t arg0, uint64_t arg1,
+	       uint64_t arg2, uint64_t arg3, uint64_t arg4, uint64_t arg5,
+	       uint64_t arg6, struct arm_smccc_res *res)
+{
+	asm volatile("mov   w0, %w[function_id]\n"
+		     "mov   x1, %[arg0]\n"
+		     "mov   x2, %[arg1]\n"
+		     "mov   x3, %[arg2]\n"
+		     "mov   x4, %[arg3]\n"
+		     "mov   x5, %[arg4]\n"
+		     "mov   x6, %[arg5]\n"
+		     "mov   x7, %[arg6]\n"
+		     "hvc   #0\n"
+		     "mov   %[res0], x0\n"
+		     "mov   %[res1], x1\n"
+		     "mov   %[res2], x2\n"
+		     "mov   %[res3], x3\n"
+		     : [res0] "=r"(res->a0), [res1] "=r"(res->a1),
+		       [res2] "=r"(res->a2), [res3] "=r"(res->a3)
+		     : [function_id] "r"(function_id), [arg0] "r"(arg0),
+		       [arg1] "r"(arg1), [arg2] "r"(arg2), [arg3] "r"(arg3),
+		       [arg4] "r"(arg4), [arg5] "r"(arg5), [arg6] "r"(arg6)
+		     : "x0", "x1", "x2", "x3", "x4", "x5", "x6", "x7");
+}

tools/testing/selftests/kvm/lib/riscv/processor.c

···
 		core.regs.t3, core.regs.t4, core.regs.t5, core.regs.t6);
 }
 
-static void __aligned(16) guest_hang(void)
+static void __aligned(16) guest_unexp_trap(void)
 {
-	while (1)
-		;
+	sbi_ecall(KVM_RISCV_SELFTESTS_SBI_EXT,
+		  KVM_RISCV_SELFTESTS_SBI_UNEXP,
+		  0, 0, 0, 0, 0, 0);
 }
 
 void vm_vcpu_add_default(struct kvm_vm *vm, uint32_t vcpuid, void *guest_code)
···
 
 	/* Setup default exception vector of guest */
 	set_reg(vm, vcpuid, RISCV_CSR_REG(stvec),
-		(unsigned long)guest_hang);
+		(unsigned long)guest_unexp_trap);
 }
 
 void vcpu_args_set(struct kvm_vm *vm, uint32_t vcpuid, unsigned int num, ...)
···
 		case 7:
 			id = RISCV_CORE_REG(regs.a7);
 			break;
-		};
+		}
 		set_reg(vm, vcpuid, id, va_arg(ap, uint64_t));
 	}
 

tools/testing/selftests/kvm/lib/riscv/ucall.c

···
 		uc.args[i] = va_arg(va, uint64_t);
 	va_end(va);
 
-	sbi_ecall(KVM_RISCV_SELFTESTS_SBI_EXT, 0, (vm_vaddr_t)&uc,
-		  0, 0, 0, 0, 0);
+	sbi_ecall(KVM_RISCV_SELFTESTS_SBI_EXT,
+		  KVM_RISCV_SELFTESTS_SBI_UCALL,
+		  (vm_vaddr_t)&uc, 0, 0, 0, 0, 0);
 }
 
 uint64_t get_ucall(struct kvm_vm *vm, uint32_t vcpu_id, struct ucall *uc)
···
 		memset(uc, 0, sizeof(*uc));
 
 	if (run->exit_reason == KVM_EXIT_RISCV_SBI &&
-	    run->riscv_sbi.extension_id == KVM_RISCV_SELFTESTS_SBI_EXT &&
-	    run->riscv_sbi.function_id == 0) {
-		memcpy(&ucall, addr_gva2hva(vm, run->riscv_sbi.args[0]),
-			sizeof(ucall));
+	    run->riscv_sbi.extension_id == KVM_RISCV_SELFTESTS_SBI_EXT) {
+		switch (run->riscv_sbi.function_id) {
+		case KVM_RISCV_SELFTESTS_SBI_UCALL:
+			memcpy(&ucall, addr_gva2hva(vm,
+			       run->riscv_sbi.args[0]), sizeof(ucall));
 
-		vcpu_run_complete_io(vm, vcpu_id);
-		if (uc)
-			memcpy(uc, &ucall, sizeof(ucall));
+			vcpu_run_complete_io(vm, vcpu_id);
+			if (uc)
+				memcpy(uc, &ucall, sizeof(ucall));
+
+			break;
+		case KVM_RISCV_SELFTESTS_SBI_UNEXP:
+			vcpu_dump(stderr, vm, vcpu_id, 2);
+			TEST_ASSERT(0, "Unexpected trap taken by guest");
+			break;
+		default:
+			break;
+		}
 	}
 
 	return ucall.cmd;

tools/testing/selftests/kvm/s390x/memop.c

···
 #include <string.h>
 #include <sys/ioctl.h>
 
+#include <linux/bits.h>
+
 #include "test_util.h"
 #include "kvm_util.h"
 
···
 #define SIDA_OFFSET(o) ._sida_offset = 1, .sida_offset = (o)
 #define AR(a) ._ar = 1, .ar = (a)
 #define KEY(a) .f_key = 1, .key = (a)
+#define INJECT .f_inject = 1
 
 #define CHECK_N_DO(f, ...) ({ f(__VA_ARGS__, CHECK_ONLY); f(__VA_ARGS__); })
 
···
 	TEST_ASSERT(rv == 4, "Should result in protection exception");		\
 })
 
+static void guest_error_key(void)
+{
+	GUEST_SYNC(STAGE_INITED);
+	set_storage_key_range(mem1, PAGE_SIZE, 0x18);
+	set_storage_key_range(mem1 + PAGE_SIZE, sizeof(mem1) - PAGE_SIZE, 0x98);
+	GUEST_SYNC(STAGE_SKEYS_SET);
+	GUEST_SYNC(STAGE_IDLED);
+}
+
 static void test_errors_key(void)
 {
-	struct test_default t = test_default_init(guest_copy_key_fetch_prot);
+	struct test_default t = test_default_init(guest_error_key);
 
 	HOST_SYNC(t.vcpu, STAGE_INITED);
 	HOST_SYNC(t.vcpu, STAGE_SKEYS_SET);
···
 	CHECK_N_DO(ERR_PROT_MOP, t.vcpu, LOGICAL, READ, mem2, t.size, GADDR_V(mem2), KEY(2));
 	CHECK_N_DO(ERR_PROT_MOP, t.vm, ABSOLUTE, WRITE, mem1, t.size, GADDR_V(mem1), KEY(2));
 	CHECK_N_DO(ERR_PROT_MOP, t.vm, ABSOLUTE, READ, mem2, t.size, GADDR_V(mem2), KEY(2));
+
+	kvm_vm_free(t.kvm_vm);
+}
+
+static void test_termination(void)
+{
+	struct test_default t = test_default_init(guest_error_key);
+	uint64_t prefix;
+	uint64_t teid;
+	uint64_t teid_mask = BIT(63 - 56) | BIT(63 - 60) | BIT(63 - 61);
+	uint64_t psw[2];
+
+	HOST_SYNC(t.vcpu, STAGE_INITED);
+	HOST_SYNC(t.vcpu, STAGE_SKEYS_SET);
+
+	/* vcpu, mismatching keys after first page */
+	ERR_PROT_MOP(t.vcpu, LOGICAL, WRITE, mem1, t.size, GADDR_V(mem1), KEY(1), INJECT);
+	/*
+	 * The memop injected a program exception and the test needs to check the
+	 * Translation-Exception Identification (TEID). It is necessary to run
+	 * the guest in order to be able to read the TEID from guest memory.
+	 * Set the guest program new PSW, so the guest state is not clobbered.
+	 */
+	prefix = t.run->s.regs.prefix;
+	psw[0] = t.run->psw_mask;
+	psw[1] = t.run->psw_addr;
+	MOP(t.vm, ABSOLUTE, WRITE, psw, sizeof(psw), GADDR(prefix + 464));
+	HOST_SYNC(t.vcpu, STAGE_IDLED);
+	MOP(t.vm, ABSOLUTE, READ, &teid, sizeof(teid), GADDR(prefix + 168));
+	/* Bits 56, 60, 61 form a code, 0 being the only one allowing for termination */
+	ASSERT_EQ(teid & teid_mask, 0);
 
 	kvm_vm_free(t.kvm_vm);
 }
···
 		test_copy_key_fetch_prot();
 		test_copy_key_fetch_prot_override();
 		test_errors_key();
+		test_termination();
 		test_errors_key_storage_prot_override();
 		test_errors_key_fetch_prot_override_not_enabled();
 		test_errors_key_fetch_prot_override_enabled();

tools/testing/selftests/kvm/steal_time.c

···
 
 static int64_t smccc(uint32_t func, uint64_t arg)
 {
-	unsigned long ret;
+	struct arm_smccc_res res;
 
-	asm volatile(
-		"mov	w0, %w1\n"
-		"mov	x1, %2\n"
-		"hvc	#0\n"
-		"mov	%0, x0\n"
-	: "=r" (ret) : "r" (func), "r" (arg) :
-	  "x0", "x1", "x2", "x3");
-
-	return ret;
+	smccc_hvc(func, arg, 0, 0, 0, 0, 0, 0, &res);
+	return res.a0;
 }
 
 static void check_status(struct st_time *st)

tools/testing/selftests/kvm/x86_64/fix_hypercall_test.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2020, Google LLC.
+ *
+ * Tests for KVM paravirtual feature disablement
+ */
+#include <asm/kvm_para.h>
+#include <linux/kvm_para.h>
+#include <linux/stringify.h>
+#include <stdint.h>
+
+#include "apic.h"
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#define VCPU_ID 0
+
+static bool ud_expected;
+
+static void guest_ud_handler(struct ex_regs *regs)
+{
+	GUEST_ASSERT(ud_expected);
+	GUEST_DONE();
+}
+
+extern unsigned char svm_hypercall_insn;
+static uint64_t svm_do_sched_yield(uint8_t apic_id)
+{
+	uint64_t ret;
+
+	asm volatile("mov %1, %%rax\n\t"
+		     "mov %2, %%rbx\n\t"
+		     "svm_hypercall_insn:\n\t"
+		     "vmmcall\n\t"
+		     "mov %%rax, %0\n\t"
+		     : "=r"(ret)
+		     : "r"((uint64_t)KVM_HC_SCHED_YIELD), "r"((uint64_t)apic_id)
+		     : "rax", "rbx", "memory");
+
+	return ret;
+}
+
+extern unsigned char vmx_hypercall_insn;
+static uint64_t vmx_do_sched_yield(uint8_t apic_id)
+{
+	uint64_t ret;
+
+	asm volatile("mov %1, %%rax\n\t"
+		     "mov %2, %%rbx\n\t"
+		     "vmx_hypercall_insn:\n\t"
+		     "vmcall\n\t"
+		     "mov %%rax, %0\n\t"
+		     : "=r"(ret)
+		     : "r"((uint64_t)KVM_HC_SCHED_YIELD), "r"((uint64_t)apic_id)
+		     : "rax", "rbx", "memory");
+
+	return ret;
+}
+
+static void assert_hypercall_insn(unsigned char *exp_insn, unsigned char *obs_insn)
+{
+	uint32_t exp = 0, obs = 0;
+
+	memcpy(&exp, exp_insn, sizeof(exp));
+	memcpy(&obs, obs_insn, sizeof(obs));
+
+	GUEST_ASSERT_EQ(exp, obs);
+}
+
+static void guest_main(void)
+{
+	unsigned char *native_hypercall_insn, *hypercall_insn;
+	uint8_t apic_id;
+
+	apic_id = GET_APIC_ID_FIELD(xapic_read_reg(APIC_ID));
+
+	if (is_intel_cpu()) {
+		native_hypercall_insn = &vmx_hypercall_insn;
+		hypercall_insn = &svm_hypercall_insn;
+		svm_do_sched_yield(apic_id);
+	} else if (is_amd_cpu()) {
+		native_hypercall_insn = &svm_hypercall_insn;
+		hypercall_insn = &vmx_hypercall_insn;
+		vmx_do_sched_yield(apic_id);
+	} else {
+		GUEST_ASSERT(0);
+		/* unreachable */
+		return;
+	}
+
+	GUEST_ASSERT(!ud_expected);
+	assert_hypercall_insn(native_hypercall_insn, hypercall_insn);
+	GUEST_DONE();
+}
+
+static void setup_ud_vector(struct kvm_vm *vm)
+{
+	vm_init_descriptor_tables(vm);
+	vcpu_init_descriptor_tables(vm, VCPU_ID);
+	vm_install_exception_handler(vm, UD_VECTOR, guest_ud_handler);
+}
+
+static void enter_guest(struct kvm_vm *vm)
+{
+	struct kvm_run *run;
+	struct ucall uc;
+
+	run = vcpu_state(vm, VCPU_ID);
+
+	vcpu_run(vm, VCPU_ID);
+	switch (get_ucall(vm, VCPU_ID, &uc)) {
+	case UCALL_SYNC:
+		pr_info("%s: %016lx\n", (const char *)uc.args[2], uc.args[3]);
+		break;
+	case UCALL_DONE:
+		return;
+	case UCALL_ABORT:
+		TEST_FAIL("%s at %s:%ld", (const char *)uc.args[0], __FILE__, uc.args[1]);
+	default:
+		TEST_FAIL("Unhandled ucall: %ld\nexit_reason: %u (%s)",
+			  uc.cmd, run->exit_reason, exit_reason_str(run->exit_reason));
+	}
+}
+
+static void test_fix_hypercall(void)
+{
+	struct kvm_vm *vm;
+
+	vm = vm_create_default(VCPU_ID, 0, guest_main);
+	setup_ud_vector(vm);
+
+	ud_expected = false;
+	sync_global_to_guest(vm, ud_expected);
+
+	virt_pg_map(vm, APIC_DEFAULT_GPA, APIC_DEFAULT_GPA);
+
+	enter_guest(vm);
+}
+
+static void test_fix_hypercall_disabled(void)
+{
+	struct kvm_enable_cap cap = {0};
+	struct kvm_vm *vm;
+
+	vm = vm_create_default(VCPU_ID, 0, guest_main);
+	setup_ud_vector(vm);
+
+	cap.cap = KVM_CAP_DISABLE_QUIRKS2;
+	cap.args[0] = KVM_X86_QUIRK_FIX_HYPERCALL_INSN;
+	vm_enable_cap(vm, &cap);
+
+	ud_expected = true;
+	sync_global_to_guest(vm, ud_expected);
+
+	virt_pg_map(vm, APIC_DEFAULT_GPA, APIC_DEFAULT_GPA);
+
+	enter_guest(vm);
+}
+
+int main(void)
+{
+	if (!(kvm_check_cap(KVM_CAP_DISABLE_QUIRKS2) & KVM_X86_QUIRK_FIX_HYPERCALL_INSN)) {
+		print_skip("KVM_X86_QUIRK_HYPERCALL_INSN not supported");
+		exit(KSFT_SKIP);
+	}
+
+	test_fix_hypercall();
+	test_fix_hypercall_disabled();
+}

tools/testing/selftests/kvm/x86_64/tsc_scaling_sync.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * svm_vmcall_test
+ *
+ * Copyright © 2021 Amazon.com, Inc. or its affiliates.
+ *
+ * Xen shared_info / pvclock testing
+ */
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#include <stdint.h>
+#include <time.h>
+#include <sched.h>
+#include <signal.h>
+#include <pthread.h>
+
+#define NR_TEST_VCPUS 20
+
+static struct kvm_vm *vm;
+pthread_spinlock_t create_lock;
+
+#define TEST_TSC_KHZ    2345678UL
+#define TEST_TSC_OFFSET 200000000
+
+uint64_t tsc_sync;
+static void guest_code(void)
+{
+	uint64_t start_tsc, local_tsc, tmp;
+
+	start_tsc = rdtsc();
+	do {
+		tmp = READ_ONCE(tsc_sync);
+		local_tsc = rdtsc();
+		WRITE_ONCE(tsc_sync, local_tsc);
+		if (unlikely(local_tsc < tmp))
+			GUEST_SYNC_ARGS(0, local_tsc, tmp, 0, 0);
+
+	} while (local_tsc - start_tsc < 5000 * TEST_TSC_KHZ);
+
+	GUEST_DONE();
+}
+
+
+static void *run_vcpu(void *_cpu_nr)
+{
+	unsigned long cpu = (unsigned long)_cpu_nr;
+	unsigned long failures = 0;
+	static bool first_cpu_done;
+
+	/* The kernel is fine, but vm_vcpu_add_default() needs locking */
+	pthread_spin_lock(&create_lock);
+
+	vm_vcpu_add_default(vm, cpu, guest_code);
+
+	if (!first_cpu_done) {
+		first_cpu_done = true;
+		vcpu_set_msr(vm, cpu, MSR_IA32_TSC, TEST_TSC_OFFSET);
+	}
+
+	pthread_spin_unlock(&create_lock);
+
+	for (;;) {
+		volatile struct kvm_run *run = vcpu_state(vm, cpu);
+                struct ucall uc;
+
+                vcpu_run(vm, cpu);
+                TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
+                            "Got exit_reason other than KVM_EXIT_IO: %u (%s)\n",
+                            run->exit_reason,
+                            exit_reason_str(run->exit_reason));
+
+                switch (get_ucall(vm, cpu, &uc)) {
+                case UCALL_DONE:
+			goto out;
+
+                case UCALL_SYNC:
+			printf("Guest %ld sync %lx %lx %ld\n", cpu, uc.args[2], uc.args[3], uc.args[2] - uc.args[3]);
+			failures++;
+			break;
+
+                default:
+                        TEST_FAIL("Unknown ucall %lu", uc.cmd);
+		}
+	}
+ out:
+	return (void *)failures;
+}
+
+int main(int argc, char *argv[])
+{
+        if (!kvm_check_cap(KVM_CAP_VM_TSC_CONTROL)) {
+		print_skip("KVM_CAP_VM_TSC_CONTROL not available");
+		exit(KSFT_SKIP);
+	}
+
+	vm = vm_create_default_with_vcpus(0, DEFAULT_STACK_PGS * NR_TEST_VCPUS, 0, guest_code, NULL);
+	vm_ioctl(vm, KVM_SET_TSC_KHZ, (void *) TEST_TSC_KHZ);
+
+	pthread_spin_init(&create_lock, PTHREAD_PROCESS_PRIVATE);
+	pthread_t cpu_threads[NR_TEST_VCPUS];
+	unsigned long cpu;
+	for (cpu = 0; cpu < NR_TEST_VCPUS; cpu++)
+		pthread_create(&cpu_threads[cpu], NULL, run_vcpu, (void *)cpu);
+
+	unsigned long failures = 0;
+	for (cpu = 0; cpu < NR_TEST_VCPUS; cpu++) {
+		void *this_cpu_failures;
+		pthread_join(cpu_threads[cpu], &this_cpu_failures);
+		failures += (unsigned long)this_cpu_failures;
+	}
+
+	TEST_ASSERT(!failures, "TSC sync failed");
+	pthread_spin_destroy(&create_lock);
+	kvm_vm_free(vm);
+	return 0;
+}

tools/testing/selftests/kvm/x86_64/vmx_pmu_msrs_test.c tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c

···
 // SPDX-License-Identifier: GPL-2.0
 /*
- * VMX-pmu related msrs test
+ * Test for VMX-pmu perf capability msr
  *
  * Copyright (C) 2021 Intel Corporation
  *
- * Test to check the effect of various CPUID settings
- * on the MSR_IA32_PERF_CAPABILITIES MSR, and check that
- * whatever we write with KVM_SET_MSR is _not_ modified
- * in the guest and test it can be retrieved with KVM_GET_MSR.
- *
- * Test to check that invalid LBR formats are rejected.
+ * Test to check the effect of various CPUID settings on
+ * MSR_IA32_PERF_CAPABILITIES MSR, and check that what
+ * we write with KVM_SET_MSR is _not_ modified by the guest
+ * and check it can be retrieved with KVM_GET_MSR, also test
+ * the invalid LBR formats are rejected.
  */
 
 #define _GNU_SOURCE /* for program_invocation_short_name */
···
 	ASSERT_EQ(vcpu_get_msr(vm, VCPU_ID, MSR_IA32_PERF_CAPABILITIES), (u64)host_cap.lbr_format);
 
 	/* testcase 3, check invalid LBR format is rejected */
-	ret = _vcpu_set_msr(vm, 0, MSR_IA32_PERF_CAPABILITIES, PMU_CAP_LBR_FMT);
+	/* Note, on Arch LBR capable platforms, LBR_FMT in perf capability msr is 0x3f,
+	 * to avoid the failure, use a true invalid format 0x30 for the test. */
+	ret = _vcpu_set_msr(vm, 0, MSR_IA32_PERF_CAPABILITIES, 0x30);
 	TEST_ASSERT(ret == 0, "Bad PERF_CAPABILITIES didn't fail.");
 
+	printf("Completed perf capability tests.\n");
 	kvm_vm_free(vm);
 }

tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c

···
 
 #define EVTCHN_VECTOR	0x10
 
+#define EVTCHN_TEST1 15
+#define EVTCHN_TEST2 66
+#define EVTCHN_TIMER 13
+
 static struct kvm_vm *vm;
 
 #define XEN_HYPERCALL_MSR	0x40000000
 
 #define MIN_STEAL_TIME		50000
+
+#define __HYPERVISOR_set_timer_op	15
+#define __HYPERVISOR_sched_op		29
+#define __HYPERVISOR_event_channel_op	32
+
+#define SCHEDOP_poll			3
+
+#define EVTCHNOP_send			4
+
+#define EVTCHNSTAT_interdomain		2
+
+struct evtchn_send {
+	u32 port;
+};
+
+struct sched_poll {
+	u32 *ports;
+	unsigned int nr_ports;
+	u64 timeout;
+};
 
 struct pvclock_vcpu_time_info {
 	u32   version;
···
 	struct kvm_irq_routing_entry entries[2];
 } irq_routes;
 
+bool guest_saw_irq;
+
 static void evtchn_handler(struct ex_regs *regs)
 {
 	struct vcpu_info *vi = (void *)VCPU_INFO_VADDR;
 	vi->evtchn_upcall_pending = 0;
 	vi->evtchn_pending_sel = 0;
+	guest_saw_irq = true;
 
 	GUEST_SYNC(0x20);
+}
+
+static void guest_wait_for_irq(void)
+{
+	while (!guest_saw_irq)
+		__asm__ __volatile__ ("rep nop" : : : "memory");
+	guest_saw_irq = false;
 }
 
 static void guest_code(void)
···
 
 	/* Trigger an interrupt injection */
 	GUEST_SYNC(0);
+
+	guest_wait_for_irq();
 
 	/* Test having the host set runstates manually */
 	GUEST_SYNC(RUNSTATE_runnable);
···
 	/* Now deliver an *unmasked* interrupt */
 	GUEST_SYNC(8);
 
-	while (!si->evtchn_pending[1])
-		__asm__ __volatile__ ("rep nop" : : : "memory");
+	guest_wait_for_irq();
 
 	/* Change memslots and deliver an interrupt */
 	GUEST_SYNC(9);
 
-	for (;;)
-		__asm__ __volatile__ ("rep nop" : : : "memory");
+	guest_wait_for_irq();
+
+	/* Deliver event channel with KVM_XEN_HVM_EVTCHN_SEND */
+	GUEST_SYNC(10);
+
+	guest_wait_for_irq();
+
+	GUEST_SYNC(11);
+
+	/* Our turn. Deliver event channel (to ourselves) with
+	 * EVTCHNOP_send hypercall. */
+	unsigned long rax;
+	struct evtchn_send s = { .port = 127 };
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_event_channel_op),
+			      "D" (EVTCHNOP_send),
+			      "S" (&s));
+
+	GUEST_ASSERT(rax == 0);
+
+	guest_wait_for_irq();
+
+	GUEST_SYNC(12);
+
+	/* Deliver "outbound" event channel to an eventfd which
+	 * happens to be one of our own irqfds. */
+	s.port = 197;
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_event_channel_op),
+			      "D" (EVTCHNOP_send),
+			      "S" (&s));
+
+	GUEST_ASSERT(rax == 0);
+
+	guest_wait_for_irq();
+
+	GUEST_SYNC(13);
+
+	/* Set a timer 100ms in the future. */
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_set_timer_op),
+			      "D" (rs->state_entry_time + 100000000));
+	GUEST_ASSERT(rax == 0);
+
+	GUEST_SYNC(14);
+
+	/* Now wait for the timer */
+	guest_wait_for_irq();
+
+	GUEST_SYNC(15);
+
+	/* The host has 'restored' the timer. Just wait for it. */
+	guest_wait_for_irq();
+
+	GUEST_SYNC(16);
+
+	/* Poll for an event channel port which is already set */
+	u32 ports[1] = { EVTCHN_TIMER };
+	struct sched_poll p = {
+		.ports = ports,
+		.nr_ports = 1,
+		.timeout = 0,
+	};
+
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_sched_op),
+			      "D" (SCHEDOP_poll),
+			      "S" (&p));
+
+	GUEST_ASSERT(rax == 0);
+
+	GUEST_SYNC(17);
+
+	/* Poll for an unset port and wait for the timeout. */
+	p.timeout = 100000000;
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_sched_op),
+			      "D" (SCHEDOP_poll),
+			      "S" (&p));
+
+	GUEST_ASSERT(rax == 0);
+
+	GUEST_SYNC(18);
+
+	/* A timer will wake the masked port we're waiting on, while we poll */
+	p.timeout = 0;
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_sched_op),
+			      "D" (SCHEDOP_poll),
+			      "S" (&p));
+
+	GUEST_ASSERT(rax == 0);
+
+	GUEST_SYNC(19);
+
+	/* A timer wake an *unmasked* port which should wake us with an
+	 * actual interrupt, while we're polling on a different port. */
+	ports[0]++;
+	p.timeout = 0;
+	__asm__ __volatile__ ("vmcall" :
+			      "=a" (rax) :
+			      "a" (__HYPERVISOR_sched_op),
+			      "D" (SCHEDOP_poll),
+			      "S" (&p));
+
+	GUEST_ASSERT(rax == 0);
+
+	guest_wait_for_irq();
+
+	GUEST_SYNC(20);
+
+	/* Timer should have fired already */
+	guest_wait_for_irq();
+
+	GUEST_SYNC(21);
 }
 
 static int cmp_timespec(struct timespec *a, struct timespec *b)
···
 	else
 		return 0;
 }
+struct vcpu_info *vinfo;
 
 static void handle_alrm(int sig)
 {
+	if (vinfo)
+		printf("evtchn_upcall_pending 0x%x\n", vinfo->evtchn_upcall_pending);
+	vcpu_dump(stdout, vm, VCPU_ID, 0);
 	TEST_FAIL("IRQ delivery timed out");
 }
 
···
 
 	bool do_runstate_tests = !!(xen_caps & KVM_XEN_HVM_CONFIG_RUNSTATE);
 	bool do_eventfd_tests = !!(xen_caps & KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL);
+	bool do_evtchn_tests = do_eventfd_tests && !!(xen_caps & KVM_XEN_HVM_CONFIG_EVTCHN_SEND);
 
 	clock_gettime(CLOCK_REALTIME, &min_ts);
 
···
 		.flags = KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL,
 		.msr = XEN_HYPERCALL_MSR,
 	};
+
+	/* Let the kernel know that we *will* use it for sending all
+	 * event channels, which lets it intercept SCHEDOP_poll */
+	if (do_evtchn_tests)
+		hvmc.flags |= KVM_XEN_HVM_CONFIG_EVTCHN_SEND;
+
 	vm_ioctl(vm, KVM_XEN_HVM_CONFIG, &hvmc);
 
 	struct kvm_xen_hvm_attr lm = {
···
 
 		/* Unexpected, but not a KVM failure */
 		if (irq_fd[0] == -1 || irq_fd[1] == -1)
-			do_eventfd_tests = false;
+			do_evtchn_tests = do_eventfd_tests = false;
 	}
 
 	if (do_eventfd_tests) {
···
 
 		irq_routes.entries[0].gsi = 32;
 		irq_routes.entries[0].type = KVM_IRQ_ROUTING_XEN_EVTCHN;
-		irq_routes.entries[0].u.xen_evtchn.port = 15;
+		irq_routes.entries[0].u.xen_evtchn.port = EVTCHN_TEST1;
 		irq_routes.entries[0].u.xen_evtchn.vcpu = VCPU_ID;
 		irq_routes.entries[0].u.xen_evtchn.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL;
 
 		irq_routes.entries[1].gsi = 33;
 		irq_routes.entries[1].type = KVM_IRQ_ROUTING_XEN_EVTCHN;
-		irq_routes.entries[1].u.xen_evtchn.port = 66;
+		irq_routes.entries[1].u.xen_evtchn.port = EVTCHN_TEST2;
 		irq_routes.entries[1].u.xen_evtchn.vcpu = VCPU_ID;
 		irq_routes.entries[1].u.xen_evtchn.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL;
 
···
 		sigaction(SIGALRM, &sa, NULL);
 	}
 
-	struct vcpu_info *vinfo = addr_gpa2hva(vm, VCPU_INFO_VADDR);
+	struct kvm_xen_vcpu_attr tmr = {
+		.type = KVM_XEN_VCPU_ATTR_TYPE_TIMER,
+		.u.timer.port = EVTCHN_TIMER,
+		.u.timer.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL,
+		.u.timer.expires_ns = 0
+	};
+
+	if (do_evtchn_tests) {
+		struct kvm_xen_hvm_attr inj = {
+			.type = KVM_XEN_ATTR_TYPE_EVTCHN,
+			.u.evtchn.send_port = 127,
+			.u.evtchn.type = EVTCHNSTAT_interdomain,
+			.u.evtchn.flags = 0,
+			.u.evtchn.deliver.port.port = EVTCHN_TEST1,
+			.u.evtchn.deliver.port.vcpu = VCPU_ID + 1,
+			.u.evtchn.deliver.port.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL,
+		};
+		vm_ioctl(vm, KVM_XEN_HVM_SET_ATTR, &inj);
+
+		/* Test migration to a different vCPU */
+		inj.u.evtchn.flags = KVM_XEN_EVTCHN_UPDATE;
+		inj.u.evtchn.deliver.port.vcpu = VCPU_ID;
+		vm_ioctl(vm, KVM_XEN_HVM_SET_ATTR, &inj);
+
+		inj.u.evtchn.send_port = 197;
+		inj.u.evtchn.deliver.eventfd.port = 0;
+		inj.u.evtchn.deliver.eventfd.fd = irq_fd[1];
+		inj.u.evtchn.flags = 0;
+		vm_ioctl(vm, KVM_XEN_HVM_SET_ATTR, &inj);
+
+		vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &tmr);
+	}
+	vinfo = addr_gpa2hva(vm, VCPU_INFO_VADDR);
 	vinfo->evtchn_upcall_pending = 0;
 
 	struct vcpu_runstate_info *rs = addr_gpa2hva(vm, RUNSTATE_ADDR);
···
 					goto done;
 				if (verbose)
 					printf("Testing masked event channel\n");
-				shinfo->evtchn_mask[0] = 0x8000;
+				shinfo->evtchn_mask[0] = 1UL << EVTCHN_TEST1;
 				eventfd_write(irq_fd[0], 1UL);
 				alarm(1);
 				break;
···
 				break;
 
 			case 9:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				shinfo->evtchn_pending[1] = 0;
 				if (verbose)
 					printf("Testing event channel after memslot change\n");
 				vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS,
···
 				alarm(1);
 				break;
 
+			case 10:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				if (!do_evtchn_tests)
+					goto done;
+
+				shinfo->evtchn_pending[0] = 0;
+				if (verbose)
+					printf("Testing injection with KVM_XEN_HVM_EVTCHN_SEND\n");
+
+				struct kvm_irq_routing_xen_evtchn e;
+				e.port = EVTCHN_TEST2;
+				e.vcpu = VCPU_ID;
+				e.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL;
+
+				vm_ioctl(vm, KVM_XEN_HVM_EVTCHN_SEND, &e);
+				evtchn_irq_expected = true;
+				alarm(1);
+				break;
+
+			case 11:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				shinfo->evtchn_pending[1] = 0;
+
+				if (verbose)
+					printf("Testing guest EVTCHNOP_send direct to evtchn\n");
+				evtchn_irq_expected = true;
+				alarm(1);
+				break;
+
+			case 12:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				shinfo->evtchn_pending[0] = 0;
+
+				if (verbose)
+					printf("Testing guest EVTCHNOP_send to eventfd\n");
+				evtchn_irq_expected = true;
+				alarm(1);
+				break;
+
+			case 13:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				shinfo->evtchn_pending[1] = 0;
+
+				if (verbose)
+					printf("Testing guest oneshot timer\n");
+				break;
+
+			case 14:
+				memset(&tmr, 0, sizeof(tmr));
+				tmr.type = KVM_XEN_VCPU_ATTR_TYPE_TIMER;
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_GET_ATTR, &tmr);
+				TEST_ASSERT(tmr.u.timer.port == EVTCHN_TIMER,
+					    "Timer port not returned");
+				TEST_ASSERT(tmr.u.timer.priority == KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL,
+					    "Timer priority not returned");
+				TEST_ASSERT(tmr.u.timer.expires_ns > rs->state_entry_time,
+					    "Timer expiry not returned");
+				evtchn_irq_expected = true;
+				alarm(1);
+				break;
+
+			case 15:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				shinfo->evtchn_pending[0] = 0;
+
+				if (verbose)
+					printf("Testing restored oneshot timer\n");
+
+				tmr.u.timer.expires_ns = rs->state_entry_time + 100000000,
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &tmr);
+				evtchn_irq_expected = true;
+				alarm(1);
+				break;
+
+			case 16:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+
+				if (verbose)
+					printf("Testing SCHEDOP_poll with already pending event\n");
+				shinfo->evtchn_pending[0] = shinfo->evtchn_mask[0] = 1UL << EVTCHN_TIMER;
+				alarm(1);
+				break;
+
+			case 17:
+				if (verbose)
+					printf("Testing SCHEDOP_poll timeout\n");
+				shinfo->evtchn_pending[0] = 0;
+				alarm(1);
+				break;
+
+			case 18:
+				if (verbose)
+					printf("Testing SCHEDOP_poll wake on masked event\n");
+
+				tmr.u.timer.expires_ns = rs->state_entry_time + 100000000,
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &tmr);
+				alarm(1);
+				break;
+
+			case 19:
+				shinfo->evtchn_pending[0] = shinfo->evtchn_mask[0] = 0;
+				if (verbose)
+					printf("Testing SCHEDOP_poll wake on unmasked event\n");
+
+				evtchn_irq_expected = true;
+				tmr.u.timer.expires_ns = rs->state_entry_time + 100000000;
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &tmr);
+
+				/* Read it back and check the pending time is reported correctly */
+				tmr.u.timer.expires_ns = 0;
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_GET_ATTR, &tmr);
+				TEST_ASSERT(tmr.u.timer.expires_ns == rs->state_entry_time + 100000000,
+					    "Timer not reported pending");
+				alarm(1);
+				break;
+
+			case 20:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				/* Read timer and check it is no longer pending */
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_GET_ATTR, &tmr);
+				TEST_ASSERT(!tmr.u.timer.expires_ns, "Timer still reported pending");
+
+				shinfo->evtchn_pending[0] = 0;
+				if (verbose)
+					printf("Testing timer in the past\n");
+
+				evtchn_irq_expected = true;
+				tmr.u.timer.expires_ns = rs->state_entry_time - 100000000ULL;
+				vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &tmr);
+				alarm(1);
+				break;
+
+			case 21:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+				goto done;
+
 			case 0x20:
 				TEST_ASSERT(evtchn_irq_expected, "Unexpected event channel IRQ");
 				evtchn_irq_expected = false;
-				if (shinfo->evtchn_pending[1] &&
-				    shinfo->evtchn_pending[0])
-					goto done;
 				break;
 			}
 			break;
···
 	}
 
  done:
+	alarm(0);
 	clock_gettime(CLOCK_REALTIME, &max_ts);
 
 	/*

virt/kvm/kvm_main.c

···
 	spin_lock_init(&kvm->gpc_lock);
 
 	INIT_LIST_HEAD(&kvm->devices);
+	kvm->max_vcpus = KVM_MAX_VCPUS;
 
 	BUILD_BUG_ON(KVM_MEM_SLOTS_NUM > SHRT_MAX);
 
···
 		return -EINVAL;
 
 	mutex_lock(&kvm->lock);
-	if (kvm->created_vcpus == KVM_MAX_VCPUS) {
+	if (kvm->created_vcpus >= kvm->max_vcpus) {
 		mutex_unlock(&kvm->lock);
 		return -EINVAL;
 	}