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

Pull KVM updates from Paolo Bonzini:
"ARM:

- More progress on the protected VM front, now with the full fixed
feature set as well as the limitation of some hypercalls after
initialisation.

- Cleanup of the RAZ/WI sysreg handling, which was pointlessly
complicated

- Fixes for the vgic placement in the IPA space, together with a
bunch of selftests

- More memcg accounting of the memory allocated on behalf of a guest

- Timer and vgic selftests

- Workarounds for the Apple M1 broken vgic implementation

- KConfig cleanups

- New kvmarm.mode=none option, for those who really dislike us

RISC-V:

- New KVM port.

x86:

- New API to control TSC offset from userspace

- TSC scaling for nested hypervisors on SVM

- Switch masterclock protection from raw_spin_lock to seqcount

- Clean up function prototypes in the page fault code and avoid
repeated memslot lookups

- Convey the exit reason to userspace on emulation failure

- Configure time between NX page recovery iterations

- Expose Predictive Store Forwarding Disable CPUID leaf

- Allocate page tracking data structures lazily (if the i915 KVM-GT
functionality is not compiled in)

- Cleanups, fixes and optimizations for the shadow MMU code

s390:

- SIGP Fixes

- initial preparations for lazy destroy of secure VMs

- storage key improvements/fixes

- Log the guest CPNC

Starting from this release, KVM-PPC patches will come from Michael
Ellerman's PPC tree"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (227 commits)
RISC-V: KVM: fix boolreturn.cocci warnings
RISC-V: KVM: remove unneeded semicolon
RISC-V: KVM: Fix GPA passed to __kvm_riscv_hfence_gvma_xyz() functions
RISC-V: KVM: Factor-out FP virtualization into separate sources
KVM: s390: add debug statement for diag 318 CPNC data
KVM: s390: pv: properly handle page flags for protected guests
KVM: s390: Fix handle_sske page fault handling
KVM: x86: SGX must obey the KVM_INTERNAL_ERROR_EMULATION protocol
KVM: x86: On emulation failure, convey the exit reason, etc. to userspace
KVM: x86: Get exit_reason as part of kvm_x86_ops.get_exit_info
KVM: x86: Clarify the kvm_run.emulation_failure structure layout
KVM: s390: Add a routine for setting userspace CPU state
KVM: s390: Simplify SIGP Set Arch handling
KVM: s390: pv: avoid stalls when making pages secure
KVM: s390: pv: avoid stalls for kvm_s390_pv_init_vm
KVM: s390: pv: avoid double free of sida page
KVM: s390: pv: add macros for UVC CC values
s390/mm: optimize reset_guest_reference_bit()
s390/mm: optimize set_guest_storage_key()
s390/mm: no need for pte_alloc_map_lock() if we know the pmd is present
...

Documentation/admin-guide/kernel-parameters.txt

···
 			[KVM] Controls how many 4KiB pages are periodically zapped
 			back to huge pages.  0 disables the recovery, otherwise if
 			the value is N KVM will zap 1/Nth of the 4KiB pages every
-			minute.  The default is 60.
+			period (see below).  The default is 60.
+
+	kvm.nx_huge_pages_recovery_period_ms=
+			[KVM] Controls the time period at which KVM zaps 4KiB pages
+			back to huge pages. If the value is a non-zero N, KVM will
+			zap a portion (see ratio above) of the pages every N msecs.
+			If the value is 0 (the default), KVM will pick a period based
+			on the ratio, such that a page is zapped after 1 hour on average.
 
 	kvm-amd.nested=	[KVM,AMD] Allow nested virtualization in KVM/SVM.
 			Default is 1 (enabled)
···
 	kvm-arm.mode=
 			[KVM,ARM] Select one of KVM/arm64's modes of operation.
 
+			none: Forcefully disable KVM.
+
 			nvhe: Standard nVHE-based mode, without support for
 			      protected guests.
 
···
 				   state is kept private from the host.
 				   Not valid if the kernel is running in EL2.
 
-			Defaults to VHE/nVHE based on hardware support.
+			Defaults to VHE/nVHE based on hardware support. Setting
+			mode to "protected" will disable kexec and hibernation
+			for the host.
 
 	kvm-arm.vgic_v3_group0_trap=
 			[KVM,ARM] Trap guest accesses to GICv3 group-0

Documentation/virt/kvm/api.rst

···
 ------------------
 
 :Capability: basic
-:Architectures: x86, ppc, mips
+:Architectures: x86, ppc, mips, riscv
 :Type: vcpu ioctl
 :Parameters: struct kvm_interrupt (in)
 :Returns: 0 on success, negative on failure.
···
 
 Queues an external interrupt to be injected into the virtual CPU. A negative
 interrupt number dequeues the interrupt.
+
+This is an asynchronous vcpu ioctl and can be invoked from any thread.
+
+RISC-V:
+^^^^^^^
+
+Queues an external interrupt to be injected into the virutal CPU. This ioctl
+is overloaded with 2 different irq values:
+
+a) KVM_INTERRUPT_SET
+
+   This sets external interrupt for a virtual CPU and it will receive
+   once it is ready.
+
+b) KVM_INTERRUPT_UNSET
+
+   This clears pending external interrupt for a virtual CPU.
 
 This is an asynchronous vcpu ioctl and can be invoked from any thread.
 
···
 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the
 set of bits that KVM can return in struct kvm_clock_data's flag member.
 
-The only flag defined now is KVM_CLOCK_TSC_STABLE.  If set, the returned
-value is the exact kvmclock value seen by all VCPUs at the instant
-when KVM_GET_CLOCK was called.  If clear, the returned value is simply
-CLOCK_MONOTONIC plus a constant offset; the offset can be modified
-with KVM_SET_CLOCK.  KVM will try to make all VCPUs follow this clock,
-but the exact value read by each VCPU could differ, because the host
-TSC is not stable.
+The following flags are defined:
+
+KVM_CLOCK_TSC_STABLE
+  If set, the returned value is the exact kvmclock
+  value seen by all VCPUs at the instant when KVM_GET_CLOCK was called.
+  If clear, the returned value is simply CLOCK_MONOTONIC plus a constant
+  offset; the offset can be modified with KVM_SET_CLOCK.  KVM will try
+  to make all VCPUs follow this clock, but the exact value read by each
+  VCPU could differ, because the host TSC is not stable.
+
+KVM_CLOCK_REALTIME
+  If set, the `realtime` field in the kvm_clock_data
+  structure is populated with the value of the host's real time
+  clocksource at the instant when KVM_GET_CLOCK was called. If clear,
+  the `realtime` field does not contain a value.
+
+KVM_CLOCK_HOST_TSC
+  If set, the `host_tsc` field in the kvm_clock_data
+  structure is populated with the value of the host's timestamp counter (TSC)
+  at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field
+  does not contain a value.
 
 ::
 
   struct kvm_clock_data {
 	__u64 clock;  /* kvmclock current value */
 	__u32 flags;
-	__u32 pad[9];
+	__u32 pad0;
+	__u64 realtime;
+	__u64 host_tsc;
+	__u32 pad[4];
   };
 
 
···
 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
 such as migration.
 
+The following flags can be passed:
+
+KVM_CLOCK_REALTIME
+  If set, KVM will compare the value of the `realtime` field
+  with the value of the host's real time clocksource at the instant when
+  KVM_SET_CLOCK was called. The difference in elapsed time is added to the final
+  kvmclock value that will be provided to guests.
+
+Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored.
+
 ::
 
   struct kvm_clock_data {
 	__u64 clock;  /* kvmclock current value */
 	__u32 flags;
-	__u32 pad[9];
+	__u32 pad0;
+	__u64 realtime;
+	__u64 host_tsc;
+	__u32 pad[4];
   };
 
 
···
 ---------------------
 
 :Capability: KVM_CAP_MP_STATE
-:Architectures: x86, s390, arm, arm64
+:Architectures: x86, s390, arm, arm64, riscv
 :Type: vcpu ioctl
 :Parameters: struct kvm_mp_state (out)
 :Returns: 0 on success; -1 on error
···
 Possible values are:
 
    ==========================    ===============================================
-   KVM_MP_STATE_RUNNABLE         the vcpu is currently running [x86,arm/arm64]
+   KVM_MP_STATE_RUNNABLE         the vcpu is currently running
+                                 [x86,arm/arm64,riscv]
    KVM_MP_STATE_UNINITIALIZED    the vcpu is an application processor (AP)
                                  which has not yet received an INIT signal [x86]
    KVM_MP_STATE_INIT_RECEIVED    the vcpu has received an INIT signal, and is
···
                                  is waiting for an interrupt [x86]
    KVM_MP_STATE_SIPI_RECEIVED    the vcpu has just received a SIPI (vector
                                  accessible via KVM_GET_VCPU_EVENTS) [x86]
-   KVM_MP_STATE_STOPPED          the vcpu is stopped [s390,arm/arm64]
+   KVM_MP_STATE_STOPPED          the vcpu is stopped [s390,arm/arm64,riscv]
    KVM_MP_STATE_CHECK_STOP       the vcpu is in a special error state [s390]
    KVM_MP_STATE_OPERATING        the vcpu is operating (running or halted)
                                  [s390]
···
 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
 these architectures.
 
-For arm/arm64:
-^^^^^^^^^^^^^^
+For arm/arm64/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.
···
 ---------------------
 
 :Capability: KVM_CAP_MP_STATE
-:Architectures: x86, s390, arm, arm64
+:Architectures: x86, s390, arm, arm64, riscv
 :Type: vcpu ioctl
 :Parameters: struct kvm_mp_state (in)
 :Returns: 0 on success; -1 on error
···
 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
 these architectures.
 
-For arm/arm64:
-^^^^^^^^^^^^^^
+For arm/arm64/riscv:
+^^^^^^^^^^^^^^^^^^^^
 
 The only states that are valid are KVM_MP_STATE_STOPPED and
 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
···
 following id bit patterns::
 
   0x7020 0000 0003 02 <0:3> <reg:5>
+
+RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of
+that is the register group type.
+
+RISC-V config registers are meant for configuring a Guest VCPU and it has
+the following id bit patterns::
+
+  0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host)
+  0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host)
+
+Following are the RISC-V config registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x80x0 0000 0100 0000 isa       ISA feature bitmap of Guest VCPU
+======================= ========= =============================================
+
+The isa config register can be read anytime but can only be written before
+a Guest VCPU runs. It will have ISA feature bits matching underlying host
+set by default.
+
+RISC-V core registers represent the general excution state of a Guest VCPU
+and it has the following id bit patterns::
+
+  0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host)
+  0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host)
+
+Following are the RISC-V core registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x80x0 0000 0200 0000 regs.pc   Program counter
+  0x80x0 0000 0200 0001 regs.ra   Return address
+  0x80x0 0000 0200 0002 regs.sp   Stack pointer
+  0x80x0 0000 0200 0003 regs.gp   Global pointer
+  0x80x0 0000 0200 0004 regs.tp   Task pointer
+  0x80x0 0000 0200 0005 regs.t0   Caller saved register 0
+  0x80x0 0000 0200 0006 regs.t1   Caller saved register 1
+  0x80x0 0000 0200 0007 regs.t2   Caller saved register 2
+  0x80x0 0000 0200 0008 regs.s0   Callee saved register 0
+  0x80x0 0000 0200 0009 regs.s1   Callee saved register 1
+  0x80x0 0000 0200 000a regs.a0   Function argument (or return value) 0
+  0x80x0 0000 0200 000b regs.a1   Function argument (or return value) 1
+  0x80x0 0000 0200 000c regs.a2   Function argument 2
+  0x80x0 0000 0200 000d regs.a3   Function argument 3
+  0x80x0 0000 0200 000e regs.a4   Function argument 4
+  0x80x0 0000 0200 000f regs.a5   Function argument 5
+  0x80x0 0000 0200 0010 regs.a6   Function argument 6
+  0x80x0 0000 0200 0011 regs.a7   Function argument 7
+  0x80x0 0000 0200 0012 regs.s2   Callee saved register 2
+  0x80x0 0000 0200 0013 regs.s3   Callee saved register 3
+  0x80x0 0000 0200 0014 regs.s4   Callee saved register 4
+  0x80x0 0000 0200 0015 regs.s5   Callee saved register 5
+  0x80x0 0000 0200 0016 regs.s6   Callee saved register 6
+  0x80x0 0000 0200 0017 regs.s7   Callee saved register 7
+  0x80x0 0000 0200 0018 regs.s8   Callee saved register 8
+  0x80x0 0000 0200 0019 regs.s9   Callee saved register 9
+  0x80x0 0000 0200 001a regs.s10  Callee saved register 10
+  0x80x0 0000 0200 001b regs.s11  Callee saved register 11
+  0x80x0 0000 0200 001c regs.t3   Caller saved register 3
+  0x80x0 0000 0200 001d regs.t4   Caller saved register 4
+  0x80x0 0000 0200 001e regs.t5   Caller saved register 5
+  0x80x0 0000 0200 001f regs.t6   Caller saved register 6
+  0x80x0 0000 0200 0020 mode      Privilege mode (1 = S-mode or 0 = U-mode)
+======================= ========= =============================================
+
+RISC-V csr registers represent the supervisor mode control/status registers
+of a Guest VCPU and it has the following id bit patterns::
+
+  0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host)
+  0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host)
+
+Following are the RISC-V csr registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x80x0 0000 0300 0000 sstatus   Supervisor status
+  0x80x0 0000 0300 0001 sie       Supervisor interrupt enable
+  0x80x0 0000 0300 0002 stvec     Supervisor trap vector base
+  0x80x0 0000 0300 0003 sscratch  Supervisor scratch register
+  0x80x0 0000 0300 0004 sepc      Supervisor exception program counter
+  0x80x0 0000 0300 0005 scause    Supervisor trap cause
+  0x80x0 0000 0300 0006 stval     Supervisor bad address or instruction
+  0x80x0 0000 0300 0007 sip       Supervisor interrupt pending
+  0x80x0 0000 0300 0008 satp      Supervisor address translation and protection
+======================= ========= =============================================
+
+RISC-V timer registers represent the timer state of a Guest VCPU and it has
+the following id bit patterns::
+
+  0x8030 0000 04 <index into the kvm_riscv_timer struct:24>
+
+Following are the RISC-V timer registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x8030 0000 0400 0000 frequency Time base frequency (read-only)
+  0x8030 0000 0400 0001 time      Time value visible to Guest
+  0x8030 0000 0400 0002 compare   Time compare programmed by Guest
+  0x8030 0000 0400 0003 state     Time compare state (1 = ON or 0 = OFF)
+======================= ========= =============================================
+
+RISC-V F-extension registers represent the single precision floating point
+state of a Guest VCPU and it has the following id bit patterns::
+
+  0x8020 0000 05 <index into the __riscv_f_ext_state struct:24>
+
+Following are the RISC-V F-extension registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x8020 0000 0500 0000 f[0]      Floating point register 0
+  ...
+  0x8020 0000 0500 001f f[31]     Floating point register 31
+  0x8020 0000 0500 0020 fcsr      Floating point control and status register
+======================= ========= =============================================
+
+RISC-V D-extension registers represent the double precision floating point
+state of a Guest VCPU and it has the following id bit patterns::
+
+  0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr)
+  0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr)
+
+Following are the RISC-V D-extension registers:
+
+======================= ========= =============================================
+    Encoding            Register  Description
+======================= ========= =============================================
+  0x8030 0000 0600 0000 f[0]      Floating point register 0
+  ...
+  0x8030 0000 0600 001f f[31]     Floating point register 31
+  0x8020 0000 0600 0020 fcsr      Floating point control and status register
+======================= ========= =============================================
 
 
 4.69 KVM_GET_ONE_REG
···
   - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall.
     Userspace is expected to place the hypercall result into the appropriate
     field before invoking KVM_RUN again.
+
+::
+
+		/* KVM_EXIT_RISCV_SBI */
+		struct {
+			unsigned long extension_id;
+			unsigned long function_id;
+			unsigned long args[6];
+			unsigned long ret[2];
+		} riscv_sbi;
+If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has
+done a SBI call which is not handled by KVM RISC-V kernel module. The details
+of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The
+'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the
+'function_id' field represents function ID of given SBI extension. The 'args'
+array field of 'riscv_sbi' represents parameters for the SBI call and 'ret'
+array field represents return values. The userspace should update the return
+values of SBI call before resuming the VCPU. For more details on RISC-V SBI
+spec refer, https://github.com/riscv/riscv-sbi-doc.
 
 ::
 

Documentation/virt/kvm/devices/vcpu.rst

···
 base address must be 64 byte aligned and exist within a valid guest memory
 region. See Documentation/virt/kvm/arm/pvtime.rst for more information
 including the layout of the stolen time structure.
+
+4. GROUP: KVM_VCPU_TSC_CTRL
+===========================
+
+:Architectures: x86
+
+4.1 ATTRIBUTE: KVM_VCPU_TSC_OFFSET
+
+:Parameters: 64-bit unsigned TSC offset
+
+Returns:
+
+	 ======= ======================================
+	 -EFAULT Error reading/writing the provided
+		 parameter address.
+	 -ENXIO  Attribute not supported
+	 ======= ======================================
+
+Specifies the guest's TSC offset relative to the host's TSC. The guest's
+TSC is then derived by the following equation:
+
+  guest_tsc = host_tsc + KVM_VCPU_TSC_OFFSET
+
+This attribute is useful to adjust the guest's TSC on live migration,
+so that the TSC counts the time during which the VM was paused. The
+following describes a possible algorithm to use for this purpose.
+
+From the source VMM process:
+
+1. Invoke the KVM_GET_CLOCK ioctl to record the host TSC (tsc_src),
+   kvmclock nanoseconds (guest_src), and host CLOCK_REALTIME nanoseconds
+   (host_src).
+
+2. Read the KVM_VCPU_TSC_OFFSET attribute for every vCPU to record the
+   guest TSC offset (ofs_src[i]).
+
+3. Invoke the KVM_GET_TSC_KHZ ioctl to record the frequency of the
+   guest's TSC (freq).
+
+From the destination VMM process:
+
+4. Invoke the KVM_SET_CLOCK ioctl, providing the source nanoseconds from
+   kvmclock (guest_src) and CLOCK_REALTIME (host_src) in their respective
+   fields.  Ensure that the KVM_CLOCK_REALTIME flag is set in the provided
+   structure.
+
+   KVM will advance the VM's kvmclock to account for elapsed time since
+   recording the clock values.  Note that this will cause problems in
+   the guest (e.g., timeouts) unless CLOCK_REALTIME is synchronized
+   between the source and destination, and a reasonably short time passes
+   between the source pausing the VMs and the destination executing
+   steps 4-7.
+
+5. Invoke the KVM_GET_CLOCK ioctl to record the host TSC (tsc_dest) and
+   kvmclock nanoseconds (guest_dest).
+
+6. Adjust the guest TSC offsets for every vCPU to account for (1) time
+   elapsed since recording state and (2) difference in TSCs between the
+   source and destination machine:
+
+   ofs_dst[i] = ofs_src[i] -
+     (guest_src - guest_dest) * freq +
+     (tsc_src - tsc_dest)
+
+   ("ofs[i] + tsc - guest * freq" is the guest TSC value corresponding to
+   a time of 0 in kvmclock.  The above formula ensures that it is the
+   same on the destination as it was on the source).
+
+7. Write the KVM_VCPU_TSC_OFFSET attribute for every vCPU with the
+   respective value derived in the previous step.

Documentation/virt/kvm/devices/xics.rst

···
   Errors:
 
     =======  ==========================================
-    -EINVAL  Value greater than KVM_MAX_VCPU_ID.
+    -EINVAL  Value greater than KVM_MAX_VCPU_IDS.
     -EFAULT  Invalid user pointer for attr->addr.
     -EBUSY   A vcpu is already connected to the device.
     =======  ==========================================

Documentation/virt/kvm/devices/xive.rst

···
     Errors:
 
       =======  ==========================================
-      -EINVAL  Value greater than KVM_MAX_VCPU_ID.
+      -EINVAL  Value greater than KVM_MAX_VCPU_IDS.
       -EFAULT  Invalid user pointer for attr->addr.
       -EBUSY   A vCPU is already connected to the device.
       =======  ==========================================

MAINTAINERS

···
 F:	arch/powerpc/kernel/kvm*
 F:	arch/powerpc/kvm/
 
+KERNEL VIRTUAL MACHINE FOR RISC-V (KVM/riscv)
+M:	Anup Patel <[e-mail]>
+R:	Atish Patra <[e-mail]>
+L:	[e-mail]
+L:	[e-mail]
+L:	[e-mail]
+S:	Maintained
+T:	git git://github.com/kvm-riscv/linux.git
+F:	arch/riscv/include/asm/kvm*
+F:	arch/riscv/include/uapi/asm/kvm*
+F:	arch/riscv/kvm/
+
 KERNEL VIRTUAL MACHINE for s390 (KVM/s390)
 M:	Christian Borntraeger <[e-mail]>
 M:	Janosch Frank <[e-mail]>

arch/arm64/Kconfig

···
 	select HAVE_GCC_PLUGINS
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS
 	select HAVE_IRQ_TIME_ACCOUNTING
+	select HAVE_KVM
 	select HAVE_NMI
 	select HAVE_PATA_PLATFORM
 	select HAVE_PERF_EVENTS

arch/arm64/include/asm/kvm_arm.h

···
 #define MDCR_EL2_HPMFZO		(UL(1) << 29)
 #define MDCR_EL2_MTPME		(UL(1) << 28)
 #define MDCR_EL2_TDCC		(UL(1) << 27)
+#define MDCR_EL2_HLP		(UL(1) << 26)
 #define MDCR_EL2_HCCD		(UL(1) << 23)
 #define MDCR_EL2_TTRF		(UL(1) << 19)
 #define MDCR_EL2_HPMD		(UL(1) << 17)

arch/arm64/include/asm/kvm_asm.h

···
 #define KVM_HOST_SMCCC_FUNC(name) KVM_HOST_SMCCC_ID(__KVM_HOST_SMCCC_FUNC_##name)
 
 #define __KVM_HOST_SMCCC_FUNC___kvm_hyp_init			0
-#define __KVM_HOST_SMCCC_FUNC___kvm_vcpu_run			1
-#define __KVM_HOST_SMCCC_FUNC___kvm_flush_vm_context		2
-#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid_ipa		3
-#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid		4
-#define __KVM_HOST_SMCCC_FUNC___kvm_flush_cpu_context		5
-#define __KVM_HOST_SMCCC_FUNC___kvm_timer_set_cntvoff		6
-#define __KVM_HOST_SMCCC_FUNC___kvm_enable_ssbs			7
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_get_gic_config		8
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_read_vmcr		9
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_write_vmcr		10
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_init_lrs		11
-#define __KVM_HOST_SMCCC_FUNC___kvm_get_mdcr_el2		12
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_save_aprs		13
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_restore_aprs		14
-#define __KVM_HOST_SMCCC_FUNC___pkvm_init			15
-#define __KVM_HOST_SMCCC_FUNC___pkvm_host_share_hyp		16
-#define __KVM_HOST_SMCCC_FUNC___pkvm_create_private_mapping	17
-#define __KVM_HOST_SMCCC_FUNC___pkvm_cpu_set_vector		18
-#define __KVM_HOST_SMCCC_FUNC___pkvm_prot_finalize		19
-#define __KVM_HOST_SMCCC_FUNC___kvm_adjust_pc			20
 
 #ifndef __ASSEMBLY__
 
 #include <linux/mm.h>
+
+enum __kvm_host_smccc_func {
+	/* Hypercalls available only prior to pKVM finalisation */
+	/* __KVM_HOST_SMCCC_FUNC___kvm_hyp_init */
+	__KVM_HOST_SMCCC_FUNC___kvm_get_mdcr_el2 = __KVM_HOST_SMCCC_FUNC___kvm_hyp_init + 1,
+	__KVM_HOST_SMCCC_FUNC___pkvm_init,
+	__KVM_HOST_SMCCC_FUNC___pkvm_create_private_mapping,
+	__KVM_HOST_SMCCC_FUNC___pkvm_cpu_set_vector,
+	__KVM_HOST_SMCCC_FUNC___kvm_enable_ssbs,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_init_lrs,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_get_gic_config,
+	__KVM_HOST_SMCCC_FUNC___pkvm_prot_finalize,
+
+	/* Hypercalls available after pKVM finalisation */
+	__KVM_HOST_SMCCC_FUNC___pkvm_host_share_hyp,
+	__KVM_HOST_SMCCC_FUNC___kvm_adjust_pc,
+	__KVM_HOST_SMCCC_FUNC___kvm_vcpu_run,
+	__KVM_HOST_SMCCC_FUNC___kvm_flush_vm_context,
+	__KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid_ipa,
+	__KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid,
+	__KVM_HOST_SMCCC_FUNC___kvm_flush_cpu_context,
+	__KVM_HOST_SMCCC_FUNC___kvm_timer_set_cntvoff,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_read_vmcr,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_write_vmcr,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_save_aprs,
+	__KVM_HOST_SMCCC_FUNC___vgic_v3_restore_aprs,
+	__KVM_HOST_SMCCC_FUNC___pkvm_vcpu_init_traps,
+};
 
 #define DECLARE_KVM_VHE_SYM(sym)	extern char sym[]
 #define DECLARE_KVM_NVHE_SYM(sym)	extern char kvm_nvhe_sym(sym)[]

arch/arm64/include/asm/kvm_emulate.h

···
 	if (vcpu_mode_is_32bit(vcpu))
 		return !!(*vcpu_cpsr(vcpu) & PSR_AA32_E_BIT);
 
-	return !!(vcpu_read_sys_reg(vcpu, SCTLR_EL1) & (1 << 25));
+	if (vcpu_mode_priv(vcpu))
+		return !!(vcpu_read_sys_reg(vcpu, SCTLR_EL1) & SCTLR_ELx_EE);
+	else
+		return !!(vcpu_read_sys_reg(vcpu, SCTLR_EL1) & SCTLR_EL1_E0E);
 }
 
 static inline unsigned long vcpu_data_guest_to_host(struct kvm_vcpu *vcpu,

arch/arm64/include/asm/kvm_host.h

···
 enum kvm_mode {
 	KVM_MODE_DEFAULT,
 	KVM_MODE_PROTECTED,
+	KVM_MODE_NONE,
 };
 enum kvm_mode kvm_get_mode(void);
 
···
 
 #define __KVM_HAVE_ARCH_VM_ALLOC
 struct kvm *kvm_arch_alloc_vm(void);
-void kvm_arch_free_vm(struct kvm *kvm);
 
 int kvm_arm_setup_stage2(struct kvm *kvm, unsigned long type);
 
···
 {
 	return false;
 }
+
+void kvm_init_protected_traps(struct kvm_vcpu *vcpu);
 
 int kvm_arm_vcpu_finalize(struct kvm_vcpu *vcpu, int feature);
 bool kvm_arm_vcpu_is_finalized(struct kvm_vcpu *vcpu);

arch/arm64/include/asm/kvm_hyp.h

···
 void __noreturn __host_enter(struct kvm_cpu_context *host_ctxt);
 #endif
 
+extern u64 kvm_nvhe_sym(id_aa64pfr0_el1_sys_val);
+extern u64 kvm_nvhe_sym(id_aa64pfr1_el1_sys_val);
+extern u64 kvm_nvhe_sym(id_aa64isar0_el1_sys_val);
+extern u64 kvm_nvhe_sym(id_aa64isar1_el1_sys_val);
 extern u64 kvm_nvhe_sym(id_aa64mmfr0_el1_sys_val);
 extern u64 kvm_nvhe_sym(id_aa64mmfr1_el1_sys_val);
+extern u64 kvm_nvhe_sym(id_aa64mmfr2_el1_sys_val);
 
 #endif /* __ARM64_KVM_HYP_H__ */

arch/arm64/include/asm/sysreg.h

···
 #define ICH_HCR_TC		(1 << 10)
 #define ICH_HCR_TALL0		(1 << 11)
 #define ICH_HCR_TALL1		(1 << 12)
+#define ICH_HCR_TDIR		(1 << 14)
 #define ICH_HCR_EOIcount_SHIFT	27
 #define ICH_HCR_EOIcount_MASK	(0x1f << ICH_HCR_EOIcount_SHIFT)
 
···
 #define ICH_VTR_SEIS_MASK	(1 << ICH_VTR_SEIS_SHIFT)
 #define ICH_VTR_A3V_SHIFT	21
 #define ICH_VTR_A3V_MASK	(1 << ICH_VTR_A3V_SHIFT)
+#define ICH_VTR_TDS_SHIFT	19
+#define ICH_VTR_TDS_MASK	(1 << ICH_VTR_TDS_SHIFT)
 
 #define ARM64_FEATURE_FIELD_BITS	4
 

arch/arm64/kernel/smp.c

···
 {
 	bool smp_spin_tables = (num_possible_cpus() > 1 && !have_cpu_die());
 
-	return !!cpus_stuck_in_kernel || smp_spin_tables;
+	return !!cpus_stuck_in_kernel || smp_spin_tables ||
+		is_protected_kvm_enabled();
 }

arch/arm64/kvm/Kconfig

···
 #
 
 source "virt/lib/Kconfig"
+source "virt/kvm/Kconfig"
 
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
···
 
 menuconfig KVM
 	bool "Kernel-based Virtual Machine (KVM) support"
-	depends on OF
+	depends on HAVE_KVM
 	select MMU_NOTIFIER
 	select PREEMPT_NOTIFIERS
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
···
 
 	  If unsure, say N.
 
-if KVM
-
-source "virt/kvm/Kconfig"
-
 config NVHE_EL2_DEBUG
 	bool "Debug mode for non-VHE EL2 object"
+	depends on KVM
 	help
 	  Say Y here to enable the debug mode for the non-VHE KVM EL2 object.
 	  Failure reports will BUG() in the hypervisor. This is intended for
 	  local EL2 hypervisor development.
 
 	  If unsure, say N.
-
-endif # KVM
 
 endif # VIRTUALIZATION

arch/arm64/kvm/arm.c

···
 
 struct kvm *kvm_arch_alloc_vm(void)
 {
-	if (!has_vhe())
-		return kzalloc(sizeof(struct kvm), GFP_KERNEL);
+	size_t sz = sizeof(struct kvm);
 
-	return vzalloc(sizeof(struct kvm));
-}
-
-void kvm_arch_free_vm(struct kvm *kvm)
-{
 	if (!has_vhe())
-		kfree(kvm);
-	else
-		vfree(kvm);
+		return kzalloc(sz, GFP_KERNEL_ACCOUNT);
+
+	return __vmalloc(sz, GFP_KERNEL_ACCOUNT | __GFP_HIGHMEM | __GFP_ZERO);
 }
 
 int kvm_arch_vcpu_precreate(struct kvm *kvm, unsigned int id)
···
 		return ret;
 
 	ret = kvm_arm_pmu_v3_enable(vcpu);
+
+	/*
+	 * Initialize traps for protected VMs.
+	 * NOTE: Move to run in EL2 directly, rather than via a hypercall, once
+	 * the code is in place for first run initialization at EL2.
+	 */
+	if (kvm_vm_is_protected(kvm))
+		kvm_call_hyp_nvhe(__pkvm_vcpu_init_traps, vcpu);
 
 	return ret;
 }
···
 		kvm_call_hyp_nvhe(__pkvm_cpu_set_vector, data->slot);
 }
 
-static void cpu_hyp_reinit(void)
+static void cpu_hyp_init_context(void)
 {
 	kvm_init_host_cpu_context(&this_cpu_ptr_hyp_sym(kvm_host_data)->host_ctxt);
 
-	cpu_hyp_reset();
+	if (!is_kernel_in_hyp_mode())
+		cpu_init_hyp_mode();
+}
+
+static void cpu_hyp_init_features(void)
+{
+	cpu_set_hyp_vector();
+	kvm_arm_init_debug();
 
 	if (is_kernel_in_hyp_mode())
 		kvm_timer_init_vhe();
-	else
-		cpu_init_hyp_mode();
-
-	cpu_set_hyp_vector();
-
-	kvm_arm_init_debug();
 
 	if (vgic_present)
 		kvm_vgic_init_cpu_hardware();
+}
+
+static void cpu_hyp_reinit(void)
+{
+	cpu_hyp_reset();
+	cpu_hyp_init_context();
+	cpu_hyp_init_features();
 }
 
 static void _kvm_arch_hardware_enable(void *discard)
···
 	int ret;
 
 	preempt_disable();
-	hyp_install_host_vector();
+	cpu_hyp_init_context();
 	ret = kvm_call_hyp_nvhe(__pkvm_init, hyp_mem_base, hyp_mem_size,
 				num_possible_cpus(), kern_hyp_va(per_cpu_base),
 				hyp_va_bits);
+	cpu_hyp_init_features();
+
+	/*
+	 * The stub hypercalls are now disabled, so set our local flag to
+	 * prevent a later re-init attempt in kvm_arch_hardware_enable().
+	 */
+	__this_cpu_write(kvm_arm_hardware_enabled, 1);
 	preempt_enable();
 
 	return ret;
···
 	void *addr = phys_to_virt(hyp_mem_base);
 	int ret;
 
+	kvm_nvhe_sym(id_aa64pfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1);
+	kvm_nvhe_sym(id_aa64pfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR1_EL1);
+	kvm_nvhe_sym(id_aa64isar0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR0_EL1);
+	kvm_nvhe_sym(id_aa64isar1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR1_EL1);
 	kvm_nvhe_sym(id_aa64mmfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1);
 	kvm_nvhe_sym(id_aa64mmfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1);
+	kvm_nvhe_sym(id_aa64mmfr2_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR2_EL1);
 
 	ret = create_hyp_mappings(addr, addr + hyp_mem_size, PAGE_HYP);
 	if (ret)
···
 	return err;
 }
 
-static void _kvm_host_prot_finalize(void *discard)
+static void _kvm_host_prot_finalize(void *arg)
 {
-	WARN_ON(kvm_call_hyp_nvhe(__pkvm_prot_finalize));
+	int *err = arg;
+
+	if (WARN_ON(kvm_call_hyp_nvhe(__pkvm_prot_finalize)))
+		WRITE_ONCE(*err, -EINVAL);
+}
+
+static int pkvm_drop_host_privileges(void)
+{
+	int ret = 0;
+
+	/*
+	 * Flip the static key upfront as that may no longer be possible
+	 * once the host stage 2 is installed.
+	 */
+	static_branch_enable(&kvm_protected_mode_initialized);
+	on_each_cpu(_kvm_host_prot_finalize, &ret, 1);
+	return ret;
 }
 
 static int finalize_hyp_mode(void)
···
 	 * None of other sections should ever be introspected.
 	 */
 	kmemleak_free_part(__hyp_bss_start, __hyp_bss_end - __hyp_bss_start);
-
-	/*
-	 * Flip the static key upfront as that may no longer be possible
-	 * once the host stage 2 is installed.
-	 */
-	static_branch_enable(&kvm_protected_mode_initialized);
-	on_each_cpu(_kvm_host_prot_finalize, NULL, 1);
-
-	return 0;
+	return pkvm_drop_host_privileges();
 }
 
 struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr)
···
 
 	if (!is_hyp_mode_available()) {
 		kvm_info("HYP mode not available\n");
+		return -ENODEV;
+	}
+
+	if (kvm_get_mode() == KVM_MODE_NONE) {
+		kvm_info("KVM disabled from command line\n");
 		return -ENODEV;
 	}
 
···
 		return 0;
 	}
 
-	if (strcmp(arg, "nvhe") == 0 && !WARN_ON(is_kernel_in_hyp_mode()))
+	if (strcmp(arg, "nvhe") == 0 && !WARN_ON(is_kernel_in_hyp_mode())) {
+		kvm_mode = KVM_MODE_DEFAULT;
 		return 0;
+	}
+
+	if (strcmp(arg, "none") == 0) {
+		kvm_mode = KVM_MODE_NONE;
+		return 0;
+	}
 
 	return -EINVAL;
 }

arch/arm64/kvm/hyp/include/hyp/fault.h

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015 - ARM Ltd
+ * Author: Marc Zyngier <[e-mail]>
+ */
+
+#ifndef __ARM64_KVM_HYP_FAULT_H__
+#define __ARM64_KVM_HYP_FAULT_H__
+
+#include <asm/kvm_asm.h>
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_hyp.h>
+#include <asm/kvm_mmu.h>
+
+static inline bool __translate_far_to_hpfar(u64 far, u64 *hpfar)
+{
+	u64 par, tmp;
+
+	/*
+	 * Resolve the IPA the hard way using the guest VA.
+	 *
+	 * Stage-1 translation already validated the memory access
+	 * rights. As such, we can use the EL1 translation regime, and
+	 * don't have to distinguish between EL0 and EL1 access.
+	 *
+	 * We do need to save/restore PAR_EL1 though, as we haven't
+	 * saved the guest context yet, and we may return early...
+	 */
+	par = read_sysreg_par();
+	if (!__kvm_at("s1e1r", far))
+		tmp = read_sysreg_par();
+	else
+		tmp = SYS_PAR_EL1_F; /* back to the guest */
+	write_sysreg(par, par_el1);
+
+	if (unlikely(tmp & SYS_PAR_EL1_F))
+		return false; /* Translation failed, back to guest */
+
+	/* Convert PAR to HPFAR format */
+	*hpfar = PAR_TO_HPFAR(tmp);
+	return true;
+}
+
+static inline bool __get_fault_info(u64 esr, struct kvm_vcpu_fault_info *fault)
+{
+	u64 hpfar, far;
+
+	far = read_sysreg_el2(SYS_FAR);
+
+	/*
+	 * The HPFAR can be invalid if the stage 2 fault did not
+	 * happen during a stage 1 page table walk (the ESR_EL2.S1PTW
+	 * bit is clear) and one of the two following cases are true:
+	 *   1. The fault was due to a permission fault
+	 *   2. The processor carries errata 834220
+	 *
+	 * Therefore, for all non S1PTW faults where we either have a
+	 * permission fault or the errata workaround is enabled, we
+	 * resolve the IPA using the AT instruction.
+	 */
+	if (!(esr & ESR_ELx_S1PTW) &&
+	    (cpus_have_final_cap(ARM64_WORKAROUND_834220) ||
+	     (esr & ESR_ELx_FSC_TYPE) == FSC_PERM)) {
+		if (!__translate_far_to_hpfar(far, &hpfar))
+			return false;
+	} else {
+		hpfar = read_sysreg(hpfar_el2);
+	}
+
+	fault->far_el2 = far;
+	fault->hpfar_el2 = hpfar;
+	return true;
+}
+
+#endif

arch/arm64/kvm/hyp/include/hyp/switch.h

···
 #define __ARM64_KVM_HYP_SWITCH_H__
 
 #include <hyp/adjust_pc.h>
+#include <hyp/fault.h>
 
 #include <linux/arm-smccc.h>
 #include <linux/kvm_host.h>
···
 	}
 }
 
-static inline bool __translate_far_to_hpfar(u64 far, u64 *hpfar)
-{
-	u64 par, tmp;
-
-	/*
-	 * Resolve the IPA the hard way using the guest VA.
-	 *
-	 * Stage-1 translation already validated the memory access
-	 * rights. As such, we can use the EL1 translation regime, and
-	 * don't have to distinguish between EL0 and EL1 access.
-	 *
-	 * We do need to save/restore PAR_EL1 though, as we haven't
-	 * saved the guest context yet, and we may return early...
-	 */
-	par = read_sysreg_par();
-	if (!__kvm_at("s1e1r", far))
-		tmp = read_sysreg_par();
-	else
-		tmp = SYS_PAR_EL1_F; /* back to the guest */
-	write_sysreg(par, par_el1);
-
-	if (unlikely(tmp & SYS_PAR_EL1_F))
-		return false; /* Translation failed, back to guest */
-
-	/* Convert PAR to HPFAR format */
-	*hpfar = PAR_TO_HPFAR(tmp);
-	return true;
-}
-
-static inline bool __get_fault_info(u64 esr, struct kvm_vcpu_fault_info *fault)
-{
-	u64 hpfar, far;
-
-	far = read_sysreg_el2(SYS_FAR);
-
-	/*
-	 * The HPFAR can be invalid if the stage 2 fault did not
-	 * happen during a stage 1 page table walk (the ESR_EL2.S1PTW
-	 * bit is clear) and one of the two following cases are true:
-	 *   1. The fault was due to a permission fault
-	 *   2. The processor carries errata 834220
-	 *
-	 * Therefore, for all non S1PTW faults where we either have a
-	 * permission fault or the errata workaround is enabled, we
-	 * resolve the IPA using the AT instruction.
-	 */
-	if (!(esr & ESR_ELx_S1PTW) &&
-	    (cpus_have_final_cap(ARM64_WORKAROUND_834220) ||
-	     (esr & ESR_ELx_FSC_TYPE) == FSC_PERM)) {
-		if (!__translate_far_to_hpfar(far, &hpfar))
-			return false;
-	} else {
-		hpfar = read_sysreg(hpfar_el2);
-	}
-
-	fault->far_el2 = far;
-	fault->hpfar_el2 = hpfar;
-	return true;
-}
-
 static inline bool __populate_fault_info(struct kvm_vcpu *vcpu)
 {
-	u8 ec;
-	u64 esr;
-
-	esr = vcpu->arch.fault.esr_el2;
-	ec = ESR_ELx_EC(esr);
-
-	if (ec != ESR_ELx_EC_DABT_LOW && ec != ESR_ELx_EC_IABT_LOW)
-		return true;
-
-	return __get_fault_info(esr, &vcpu->arch.fault);
+	return __get_fault_info(vcpu->arch.fault.esr_el2, &vcpu->arch.fault);
 }
 
 static inline void __hyp_sve_save_host(struct kvm_vcpu *vcpu)
···
 	write_sysreg_el1(__vcpu_sys_reg(vcpu, ZCR_EL1), SYS_ZCR);
 }
 
-/* Check for an FPSIMD/SVE trap and handle as appropriate */
-static inline bool __hyp_handle_fpsimd(struct kvm_vcpu *vcpu)
+/*
+ * We trap the first access to the FP/SIMD to save the host context and
+ * restore the guest context lazily.
+ * If FP/SIMD is not implemented, handle the trap and inject an undefined
+ * instruction exception to the guest. Similarly for trapped SVE accesses.
+ */
+static bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code)
 {
 	bool sve_guest, sve_host;
 	u8 esr_ec;
···
 	}
 
 	esr_ec = kvm_vcpu_trap_get_class(vcpu);
-	if (esr_ec != ESR_ELx_EC_FP_ASIMD &&
-	    esr_ec != ESR_ELx_EC_SVE)
-		return false;
 
 	/* Don't handle SVE traps for non-SVE vcpus here: */
 	if (!sve_guest && esr_ec != ESR_ELx_EC_FP_ASIMD)
···
 
 static inline bool esr_is_ptrauth_trap(u32 esr)
 {
-	u32 ec = ESR_ELx_EC(esr);
-
-	if (ec == ESR_ELx_EC_PAC)
-		return true;
-
-	if (ec != ESR_ELx_EC_SYS64)
-		return false;
-
 	switch (esr_sys64_to_sysreg(esr)) {
 	case SYS_APIAKEYLO_EL1:
 	case SYS_APIAKEYHI_EL1:
···
 
 DECLARE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt);
 
-static inline bool __hyp_handle_ptrauth(struct kvm_vcpu *vcpu)
+static bool kvm_hyp_handle_ptrauth(struct kvm_vcpu *vcpu, u64 *exit_code)
 {
 	struct kvm_cpu_context *ctxt;
 	u64 val;
 
-	if (!vcpu_has_ptrauth(vcpu) ||
-	    !esr_is_ptrauth_trap(kvm_vcpu_get_esr(vcpu)))
+	if (!vcpu_has_ptrauth(vcpu))
 		return false;
 
 	ctxt = this_cpu_ptr(&kvm_hyp_ctxt);
···
 	write_sysreg(val, hcr_el2);
 
 	return true;
+}
+
+static bool kvm_hyp_handle_sysreg(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	if (cpus_have_final_cap(ARM64_WORKAROUND_CAVIUM_TX2_219_TVM) &&
+	    handle_tx2_tvm(vcpu))
+		return true;
+
+	if (static_branch_unlikely(&vgic_v3_cpuif_trap) &&
+	    __vgic_v3_perform_cpuif_access(vcpu) == 1)
+		return true;
+
+	if (esr_is_ptrauth_trap(kvm_vcpu_get_esr(vcpu)))
+		return kvm_hyp_handle_ptrauth(vcpu, exit_code);
+
+	return false;
+}
+
+static bool kvm_hyp_handle_cp15_32(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	if (static_branch_unlikely(&vgic_v3_cpuif_trap) &&
+	    __vgic_v3_perform_cpuif_access(vcpu) == 1)
+		return true;
+
+	return false;
+}
+
+static bool kvm_hyp_handle_iabt_low(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	if (!__populate_fault_info(vcpu))
+		return true;
+
+	return false;
+}
+
+static bool kvm_hyp_handle_dabt_low(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	if (!__populate_fault_info(vcpu))
+		return true;
+
+	if (static_branch_unlikely(&vgic_v2_cpuif_trap)) {
+		bool valid;
+
+		valid = kvm_vcpu_trap_get_fault_type(vcpu) == FSC_FAULT &&
+			kvm_vcpu_dabt_isvalid(vcpu) &&
+			!kvm_vcpu_abt_issea(vcpu) &&
+			!kvm_vcpu_abt_iss1tw(vcpu);
+
+		if (valid) {
+			int ret = __vgic_v2_perform_cpuif_access(vcpu);
+
+			if (ret == 1)
+				return true;
+
+			/* Promote an illegal access to an SError.*/
+			if (ret == -1)
+				*exit_code = ARM_EXCEPTION_EL1_SERROR;
+		}
+	}
+
+	return false;
+}
+
+typedef bool (*exit_handler_fn)(struct kvm_vcpu *, u64 *);
+
+static const exit_handler_fn *kvm_get_exit_handler_array(struct kvm_vcpu *vcpu);
+
+/*
+ * Allow the hypervisor to handle the exit with an exit handler if it has one.
+ *
+ * Returns true if the hypervisor handled the exit, and control should go back
+ * to the guest, or false if it hasn't.
+ */
+static inline bool kvm_hyp_handle_exit(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	const exit_handler_fn *handlers = kvm_get_exit_handler_array(vcpu);
+	exit_handler_fn fn;
+
+	fn = handlers[kvm_vcpu_trap_get_class(vcpu)];
+
+	if (fn)
+		return fn(vcpu, exit_code);
+
+	return false;
 }
 
 /*
···
 	if (*exit_code != ARM_EXCEPTION_TRAP)
 		goto exit;
 
-	if (cpus_have_final_cap(ARM64_WORKAROUND_CAVIUM_TX2_219_TVM) &&
-	    kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_SYS64 &&
-	    handle_tx2_tvm(vcpu))
+	/* Check if there's an exit handler and allow it to handle the exit. */
+	if (kvm_hyp_handle_exit(vcpu, exit_code))
 		goto guest;
-
-	/*
-	 * We trap the first access to the FP/SIMD to save the host context
-	 * and restore the guest context lazily.
-	 * If FP/SIMD is not implemented, handle the trap and inject an
-	 * undefined instruction exception to the guest.
-	 * Similarly for trapped SVE accesses.
-	 */
-	if (__hyp_handle_fpsimd(vcpu))
-		goto guest;
-
-	if (__hyp_handle_ptrauth(vcpu))
-		goto guest;
-
-	if (!__populate_fault_info(vcpu))
-		goto guest;
-
-	if (static_branch_unlikely(&vgic_v2_cpuif_trap)) {
-		bool valid;
-
-		valid = kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_DABT_LOW &&
-			kvm_vcpu_trap_get_fault_type(vcpu) == FSC_FAULT &&
-			kvm_vcpu_dabt_isvalid(vcpu) &&
-			!kvm_vcpu_abt_issea(vcpu) &&
-			!kvm_vcpu_abt_iss1tw(vcpu);
-
-		if (valid) {
-			int ret = __vgic_v2_perform_cpuif_access(vcpu);
-
-			if (ret == 1)
-				goto guest;
-
-			/* Promote an illegal access to an SError.*/
-			if (ret == -1)
-				*exit_code = ARM_EXCEPTION_EL1_SERROR;
-
-			goto exit;
-		}
-	}
-
-	if (static_branch_unlikely(&vgic_v3_cpuif_trap) &&
-	    (kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_SYS64 ||
-	     kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_CP15_32)) {
-		int ret = __vgic_v3_perform_cpuif_access(vcpu);
-
-		if (ret == 1)
-			goto guest;
-	}
-
 exit:
 	/* Return to the host kernel and handle the exit */
 	return false;

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

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2021 Google LLC
+ * Author: Fuad Tabba <[e-mail]>
+ */
+
+#ifndef __ARM64_KVM_FIXED_CONFIG_H__
+#define __ARM64_KVM_FIXED_CONFIG_H__
+
+#include <asm/sysreg.h>
+
+/*
+ * This file contains definitions for features to be allowed or restricted for
+ * guest virtual machines, depending on the mode KVM is running in and on the
+ * type of guest that is running.
+ *
+ * The ALLOW masks represent a bitmask of feature fields that are allowed
+ * without any restrictions as long as they are supported by the system.
+ *
+ * The RESTRICT_UNSIGNED masks, if present, represent unsigned fields for
+ * features that are restricted to support at most the specified feature.
+ *
+ * If a feature field is not present in either, than it is not supported.
+ *
+ * The approach taken for protected VMs is to allow features that are:
+ * - Needed by common Linux distributions (e.g., floating point)
+ * - Trivial to support, e.g., supporting the feature does not introduce or
+ * require tracking of additional state in KVM
+ * - Cannot be trapped or prevent the guest from using anyway
+ */
+
+/*
+ * Allow for protected VMs:
+ * - Floating-point and Advanced SIMD
+ * - Data Independent Timing
+ */
+#define PVM_ID_AA64PFR0_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64PFR0_FP) | \
+	ARM64_FEATURE_MASK(ID_AA64PFR0_ASIMD) | \
+	ARM64_FEATURE_MASK(ID_AA64PFR0_DIT) \
+	)
+
+/*
+ * Restrict to the following *unsigned* features for protected VMs:
+ * - AArch64 guests only (no support for AArch32 guests):
+ *	AArch32 adds complexity in trap handling, emulation, condition codes,
+ *	etc...
+ * - RAS (v1)
+ *	Supported by KVM
+ */
+#define PVM_ID_AA64PFR0_RESTRICT_UNSIGNED (\
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL0), ID_AA64PFR0_ELx_64BIT_ONLY) | \
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1), ID_AA64PFR0_ELx_64BIT_ONLY) | \
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL2), ID_AA64PFR0_ELx_64BIT_ONLY) | \
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_EL3), ID_AA64PFR0_ELx_64BIT_ONLY) | \
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_RAS), ID_AA64PFR0_RAS_V1) \
+	)
+
+/*
+ * Allow for protected VMs:
+ * - Branch Target Identification
+ * - Speculative Store Bypassing
+ */
+#define PVM_ID_AA64PFR1_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64PFR1_BT) | \
+	ARM64_FEATURE_MASK(ID_AA64PFR1_SSBS) \
+	)
+
+/*
+ * Allow for protected VMs:
+ * - Mixed-endian
+ * - Distinction between Secure and Non-secure Memory
+ * - Mixed-endian at EL0 only
+ * - Non-context synchronizing exception entry and exit
+ */
+#define PVM_ID_AA64MMFR0_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64MMFR0_BIGENDEL) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR0_SNSMEM) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR0_BIGENDEL0) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR0_EXS) \
+	)
+
+/*
+ * Restrict to the following *unsigned* features for protected VMs:
+ * - 40-bit IPA
+ * - 16-bit ASID
+ */
+#define PVM_ID_AA64MMFR0_RESTRICT_UNSIGNED (\
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64MMFR0_PARANGE), ID_AA64MMFR0_PARANGE_40) | \
+	FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64MMFR0_ASID), ID_AA64MMFR0_ASID_16) \
+	)
+
+/*
+ * Allow for protected VMs:
+ * - Hardware translation table updates to Access flag and Dirty state
+ * - Number of VMID bits from CPU
+ * - Hierarchical Permission Disables
+ * - Privileged Access Never
+ * - SError interrupt exceptions from speculative reads
+ * - Enhanced Translation Synchronization
+ */
+#define PVM_ID_AA64MMFR1_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_HADBS) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_VMIDBITS) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_HPD) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_PAN) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_SPECSEI) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR1_ETS) \
+	)
+
+/*
+ * Allow for protected VMs:
+ * - Common not Private translations
+ * - User Access Override
+ * - IESB bit in the SCTLR_ELx registers
+ * - Unaligned single-copy atomicity and atomic functions
+ * - ESR_ELx.EC value on an exception by read access to feature ID space
+ * - TTL field in address operations.
+ * - Break-before-make sequences when changing translation block size
+ * - E0PDx mechanism
+ */
+#define PVM_ID_AA64MMFR2_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_CNP) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_UAO) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_IESB) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_AT) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_IDS) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_TTL) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_BBM) | \
+	ARM64_FEATURE_MASK(ID_AA64MMFR2_E0PD) \
+	)
+
+/*
+ * No support for Scalable Vectors for protected VMs:
+ *	Requires additional support from KVM, e.g., context-switching and
+ *	trapping at EL2
+ */
+#define PVM_ID_AA64ZFR0_ALLOW (0ULL)
+
+/*
+ * No support for debug, including breakpoints, and watchpoints for protected
+ * VMs:
+ *	The Arm architecture mandates support for at least the Armv8 debug
+ *	architecture, which would include at least 2 hardware breakpoints and
+ *	watchpoints. Providing that support to protected guests adds
+ *	considerable state and complexity. Therefore, the reserved value of 0 is
+ *	used for debug-related fields.
+ */
+#define PVM_ID_AA64DFR0_ALLOW (0ULL)
+#define PVM_ID_AA64DFR1_ALLOW (0ULL)
+
+/*
+ * No support for implementation defined features.
+ */
+#define PVM_ID_AA64AFR0_ALLOW (0ULL)
+#define PVM_ID_AA64AFR1_ALLOW (0ULL)
+
+/*
+ * No restrictions on instructions implemented in AArch64.
+ */
+#define PVM_ID_AA64ISAR0_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_AES) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_SHA1) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_SHA2) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_CRC32) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_ATOMICS) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_RDM) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_SHA3) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_SM3) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_SM4) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_DP) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_FHM) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_TS) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_TLB) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR0_RNDR) \
+	)
+
+#define PVM_ID_AA64ISAR1_ALLOW (\
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_DPB) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_APA) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_API) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_JSCVT) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_FCMA) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_LRCPC) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_GPA) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_GPI) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_FRINTTS) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_SB) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_SPECRES) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_BF16) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_DGH) | \
+	ARM64_FEATURE_MASK(ID_AA64ISAR1_I8MM) \
+	)
+
+u64 pvm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id);
+bool kvm_handle_pvm_sysreg(struct kvm_vcpu *vcpu, u64 *exit_code);
+bool kvm_handle_pvm_restricted(struct kvm_vcpu *vcpu, u64 *exit_code);
+int kvm_check_pvm_sysreg_table(void);
+
+#endif /* __ARM64_KVM_FIXED_CONFIG_H__ */

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

···
 #define DECLARE_REG(type, name, ctxt, reg)	\
 				type name = (type)cpu_reg(ctxt, (reg))
 
+void __pkvm_vcpu_init_traps(struct kvm_vcpu *vcpu);
+
 #endif /* __ARM64_KVM_NVHE_TRAP_HANDLER_H__ */

arch/arm64/kvm/hyp/nvhe/Makefile

···
 
 obj-y := timer-sr.o sysreg-sr.o debug-sr.o switch.o tlb.o hyp-init.o host.o \
 	 hyp-main.o hyp-smp.o psci-relay.o early_alloc.o stub.o page_alloc.o \
-	 cache.o setup.o mm.o mem_protect.o
+	 cache.o setup.o mm.o mem_protect.o sys_regs.o pkvm.o
 obj-y += ../vgic-v3-sr.o ../aarch32.o ../vgic-v2-cpuif-proxy.o ../entry.o \
 	 ../fpsimd.o ../hyp-entry.o ../exception.o ../pgtable.o
 obj-y += $(lib-objs)

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

···
 	b	__host_enter_for_panic
 SYM_FUNC_END(__hyp_do_panic)
 
-.macro host_el1_sync_vect
-	.align 7
-.L__vect_start\@:
-	stp	x0, x1, [sp, #-16]!
-	mrs	x0, esr_el2
-	lsr	x0, x0, #ESR_ELx_EC_SHIFT
-	cmp	x0, #ESR_ELx_EC_HVC64
-	b.ne	__host_exit
-
+SYM_FUNC_START(__host_hvc)
 	ldp	x0, x1, [sp]		// Don't fixup the stack yet
+
+	/* No stub for you, sonny Jim */
+alternative_if ARM64_KVM_PROTECTED_MODE
+	b	__host_exit
+alternative_else_nop_endif
 
 	/* Check for a stub HVC call */
 	cmp	x0, #HVC_STUB_HCALL_NR
···
 	ldr	x5, =__kvm_handle_stub_hvc
 	hyp_pa	x5, x6
 	br	x5
+SYM_FUNC_END(__host_hvc)
+
+.macro host_el1_sync_vect
+	.align 7
+.L__vect_start\@:
+	stp	x0, x1, [sp, #-16]!
+	mrs	x0, esr_el2
+	lsr	x0, x0, #ESR_ELx_EC_SHIFT
+	cmp	x0, #ESR_ELx_EC_HVC64
+	b.eq	__host_hvc
+	b	__host_exit
 .L__vect_end\@:
 .if ((.L__vect_end\@ - .L__vect_start\@) > 0x80)
 	.error "host_el1_sync_vect larger than vector entry"

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

···
  * Author: Andrew Scull <[e-mail]>
  */
 
-#include <hyp/switch.h>
+#include <hyp/adjust_pc.h>
 
 #include <asm/pgtable-types.h>
 #include <asm/kvm_asm.h>
···
 {
 	cpu_reg(host_ctxt, 1) = __pkvm_prot_finalize();
 }
+
+static void handle___pkvm_vcpu_init_traps(struct kvm_cpu_context *host_ctxt)
+{
+	DECLARE_REG(struct kvm_vcpu *, vcpu, host_ctxt, 1);
+
+	__pkvm_vcpu_init_traps(kern_hyp_va(vcpu));
+}
+
 typedef void (*hcall_t)(struct kvm_cpu_context *);
 
 #define HANDLE_FUNC(x)	[__KVM_HOST_SMCCC_FUNC_##x] = (hcall_t)handle_##x
 
 static const hcall_t host_hcall[] = {
-	HANDLE_FUNC(__kvm_vcpu_run),
+	/* ___kvm_hyp_init */
+	HANDLE_FUNC(__kvm_get_mdcr_el2),
+	HANDLE_FUNC(__pkvm_init),
+	HANDLE_FUNC(__pkvm_create_private_mapping),
+	HANDLE_FUNC(__pkvm_cpu_set_vector),
+	HANDLE_FUNC(__kvm_enable_ssbs),
+	HANDLE_FUNC(__vgic_v3_init_lrs),
+	HANDLE_FUNC(__vgic_v3_get_gic_config),
+	HANDLE_FUNC(__pkvm_prot_finalize),
+
+	HANDLE_FUNC(__pkvm_host_share_hyp),
 	HANDLE_FUNC(__kvm_adjust_pc),
+	HANDLE_FUNC(__kvm_vcpu_run),
 	HANDLE_FUNC(__kvm_flush_vm_context),
 	HANDLE_FUNC(__kvm_tlb_flush_vmid_ipa),
 	HANDLE_FUNC(__kvm_tlb_flush_vmid),
 	HANDLE_FUNC(__kvm_flush_cpu_context),
 	HANDLE_FUNC(__kvm_timer_set_cntvoff),
-	HANDLE_FUNC(__kvm_enable_ssbs),
-	HANDLE_FUNC(__vgic_v3_get_gic_config),
 	HANDLE_FUNC(__vgic_v3_read_vmcr),
 	HANDLE_FUNC(__vgic_v3_write_vmcr),
-	HANDLE_FUNC(__vgic_v3_init_lrs),
-	HANDLE_FUNC(__kvm_get_mdcr_el2),
 	HANDLE_FUNC(__vgic_v3_save_aprs),
 	HANDLE_FUNC(__vgic_v3_restore_aprs),
-	HANDLE_FUNC(__pkvm_init),
-	HANDLE_FUNC(__pkvm_cpu_set_vector),
-	HANDLE_FUNC(__pkvm_host_share_hyp),
-	HANDLE_FUNC(__pkvm_create_private_mapping),
-	HANDLE_FUNC(__pkvm_prot_finalize),
+	HANDLE_FUNC(__pkvm_vcpu_init_traps),
 };
 
 static void handle_host_hcall(struct kvm_cpu_context *host_ctxt)
 {
 	DECLARE_REG(unsigned long, id, host_ctxt, 0);
+	unsigned long hcall_min = 0;
 	hcall_t hfn;
+
+	/*
+	 * If pKVM has been initialised then reject any calls to the
+	 * early "privileged" hypercalls. Note that we cannot reject
+	 * calls to __pkvm_prot_finalize for two reasons: (1) The static
+	 * key used to determine initialisation must be toggled prior to
+	 * finalisation and (2) finalisation is performed on a per-CPU
+	 * basis. This is all fine, however, since __pkvm_prot_finalize
+	 * returns -EPERM after the first call for a given CPU.
+	 */
+	if (static_branch_unlikely(&kvm_protected_mode_initialized))
+		hcall_min = __KVM_HOST_SMCCC_FUNC___pkvm_prot_finalize;
 
 	id -= KVM_HOST_SMCCC_ID(0);
 
-	if (unlikely(id >= ARRAY_SIZE(host_hcall)))
+	if (unlikely(id < hcall_min || id >= ARRAY_SIZE(host_hcall)))
 		goto inval;
 
 	hfn = host_hcall[id];

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

···
 #include <asm/kvm_pgtable.h>
 #include <asm/stage2_pgtable.h>
 
-#include <hyp/switch.h>
+#include <hyp/fault.h>
 
 #include <nvhe/gfp.h>
 #include <nvhe/memory.h>
···
 struct host_kvm host_kvm;
 
 static struct hyp_pool host_s2_pool;
-
-/*
- * Copies of the host's CPU features registers holding sanitized values.
- */
-u64 id_aa64mmfr0_el1_sys_val;
-u64 id_aa64mmfr1_el1_sys_val;
 
 const u8 pkvm_hyp_id = 1;
 
···
 {
 	struct kvm_s2_mmu *mmu = &host_kvm.arch.mmu;
 	struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params);
+
+	if (params->hcr_el2 & HCR_VM)
+		return -EPERM;
 
 	params->vttbr = kvm_get_vttbr(mmu);
 	params->vtcr = host_kvm.arch.vtcr;

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

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021 Google LLC
+ * Author: Fuad Tabba <[e-mail]>
+ */
+
+#include <linux/kvm_host.h>
+#include <linux/mm.h>
+#include <nvhe/fixed_config.h>
+#include <nvhe/trap_handler.h>
+
+/*
+ * Set trap register values based on features in ID_AA64PFR0.
+ */
+static void pvm_init_traps_aa64pfr0(struct kvm_vcpu *vcpu)
+{
+	const u64 feature_ids = pvm_read_id_reg(vcpu, SYS_ID_AA64PFR0_EL1);
+	u64 hcr_set = HCR_RW;
+	u64 hcr_clear = 0;
+	u64 cptr_set = 0;
+
+	/* Protected KVM does not support AArch32 guests. */
+	BUILD_BUG_ON(FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_EL0),
+		PVM_ID_AA64PFR0_RESTRICT_UNSIGNED) != ID_AA64PFR0_ELx_64BIT_ONLY);
+	BUILD_BUG_ON(FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1),
+		PVM_ID_AA64PFR0_RESTRICT_UNSIGNED) != ID_AA64PFR0_ELx_64BIT_ONLY);
+
+	/*
+	 * Linux guests assume support for floating-point and Advanced SIMD. Do
+	 * not change the trapping behavior for these from the KVM default.
+	 */
+	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));
+
+	/* Trap RAS unless all current versions are supported */
+	if (FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_RAS), feature_ids) <
+	    ID_AA64PFR0_RAS_V1P1) {
+		hcr_set |= HCR_TERR | HCR_TEA;
+		hcr_clear |= HCR_FIEN;
+	}
+
+	/* Trap AMU */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_AMU), feature_ids)) {
+		hcr_clear |= HCR_AMVOFFEN;
+		cptr_set |= CPTR_EL2_TAM;
+	}
+
+	/* Trap SVE */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_SVE), feature_ids))
+		cptr_set |= CPTR_EL2_TZ;
+
+	vcpu->arch.hcr_el2 |= hcr_set;
+	vcpu->arch.hcr_el2 &= ~hcr_clear;
+	vcpu->arch.cptr_el2 |= cptr_set;
+}
+
+/*
+ * Set trap register values based on features in ID_AA64PFR1.
+ */
+static void pvm_init_traps_aa64pfr1(struct kvm_vcpu *vcpu)
+{
+	const u64 feature_ids = pvm_read_id_reg(vcpu, SYS_ID_AA64PFR1_EL1);
+	u64 hcr_set = 0;
+	u64 hcr_clear = 0;
+
+	/* Memory Tagging: Trap and Treat as Untagged if not supported. */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR1_MTE), feature_ids)) {
+		hcr_set |= HCR_TID5;
+		hcr_clear |= HCR_DCT | HCR_ATA;
+	}
+
+	vcpu->arch.hcr_el2 |= hcr_set;
+	vcpu->arch.hcr_el2 &= ~hcr_clear;
+}
+
+/*
+ * Set trap register values based on features in ID_AA64DFR0.
+ */
+static void pvm_init_traps_aa64dfr0(struct kvm_vcpu *vcpu)
+{
+	const u64 feature_ids = pvm_read_id_reg(vcpu, SYS_ID_AA64DFR0_EL1);
+	u64 mdcr_set = 0;
+	u64 mdcr_clear = 0;
+	u64 cptr_set = 0;
+
+	/* Trap/constrain PMU */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_PMUVER), feature_ids)) {
+		mdcr_set |= MDCR_EL2_TPM | MDCR_EL2_TPMCR;
+		mdcr_clear |= MDCR_EL2_HPME | MDCR_EL2_MTPME |
+			      MDCR_EL2_HPMN_MASK;
+	}
+
+	/* Trap Debug */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_DEBUGVER), feature_ids))
+		mdcr_set |= MDCR_EL2_TDRA | MDCR_EL2_TDA | MDCR_EL2_TDE;
+
+	/* Trap OS Double Lock */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_DOUBLELOCK), feature_ids))
+		mdcr_set |= MDCR_EL2_TDOSA;
+
+	/* Trap SPE */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_PMSVER), feature_ids)) {
+		mdcr_set |= MDCR_EL2_TPMS;
+		mdcr_clear |= MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT;
+	}
+
+	/* Trap Trace Filter */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_TRACE_FILT), feature_ids))
+		mdcr_set |= MDCR_EL2_TTRF;
+
+	/* Trap Trace */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_TRACEVER), feature_ids))
+		cptr_set |= CPTR_EL2_TTA;
+
+	vcpu->arch.mdcr_el2 |= mdcr_set;
+	vcpu->arch.mdcr_el2 &= ~mdcr_clear;
+	vcpu->arch.cptr_el2 |= cptr_set;
+}
+
+/*
+ * Set trap register values based on features in ID_AA64MMFR0.
+ */
+static void pvm_init_traps_aa64mmfr0(struct kvm_vcpu *vcpu)
+{
+	const u64 feature_ids = pvm_read_id_reg(vcpu, SYS_ID_AA64MMFR0_EL1);
+	u64 mdcr_set = 0;
+
+	/* Trap Debug Communications Channel registers */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64MMFR0_FGT), feature_ids))
+		mdcr_set |= MDCR_EL2_TDCC;
+
+	vcpu->arch.mdcr_el2 |= mdcr_set;
+}
+
+/*
+ * Set trap register values based on features in ID_AA64MMFR1.
+ */
+static void pvm_init_traps_aa64mmfr1(struct kvm_vcpu *vcpu)
+{
+	const u64 feature_ids = pvm_read_id_reg(vcpu, SYS_ID_AA64MMFR1_EL1);
+	u64 hcr_set = 0;
+
+	/* Trap LOR */
+	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64MMFR1_LOR), feature_ids))
+		hcr_set |= HCR_TLOR;
+
+	vcpu->arch.hcr_el2 |= hcr_set;
+}
+
+/*
+ * Set baseline trap register values.
+ */
+static void pvm_init_trap_regs(struct kvm_vcpu *vcpu)
+{
+	const u64 hcr_trap_feat_regs = HCR_TID3;
+	const u64 hcr_trap_impdef = HCR_TACR | HCR_TIDCP | HCR_TID1;
+
+	/*
+	 * Always trap:
+	 * - Feature id registers: to control features exposed to guests
+	 * - Implementation-defined features
+	 */
+	vcpu->arch.hcr_el2 |= hcr_trap_feat_regs | hcr_trap_impdef;
+
+	/* Clear res0 and set res1 bits to trap potential new features. */
+	vcpu->arch.hcr_el2 &= ~(HCR_RES0);
+	vcpu->arch.mdcr_el2 &= ~(MDCR_EL2_RES0);
+	vcpu->arch.cptr_el2 |= CPTR_NVHE_EL2_RES1;
+	vcpu->arch.cptr_el2 &= ~(CPTR_NVHE_EL2_RES0);
+}
+
+/*
+ * Initialize trap register values for protected VMs.
+ */
+void __pkvm_vcpu_init_traps(struct kvm_vcpu *vcpu)
+{
+	pvm_init_trap_regs(vcpu);
+	pvm_init_traps_aa64pfr0(vcpu);
+	pvm_init_traps_aa64pfr1(vcpu);
+	pvm_init_traps_aa64dfr0(vcpu);
+	pvm_init_traps_aa64mmfr0(vcpu);
+	pvm_init_traps_aa64mmfr1(vcpu);
+}

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

···
 #include <asm/kvm_pgtable.h>
 
 #include <nvhe/early_alloc.h>
+#include <nvhe/fixed_config.h>
 #include <nvhe/gfp.h>
 #include <nvhe/memory.h>
 #include <nvhe/mem_protect.h>
···
 	void *virt = hyp_phys_to_virt(phys);
 	void (*fn)(phys_addr_t params_pa, void *finalize_fn_va);
 	int ret;
+
+	BUG_ON(kvm_check_pvm_sysreg_table());
 
 	if (!PAGE_ALIGNED(phys) || !PAGE_ALIGNED(size))
 		return -EINVAL;

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

···
 #include <asm/processor.h>
 #include <asm/thread_info.h>
 
+#include <nvhe/fixed_config.h>
 #include <nvhe/mem_protect.h>
 
 /* Non-VHE specific context */
···
 		write_sysreg(pmu->events_host, pmcntenset_el0);
 }
 
+/**
+ * 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
+ * back to the guest, or false if it hasn't.
+ */
+static bool kvm_handle_pvm_sys64(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	/*
+	 * Make sure we handle the exit for workarounds and ptrauth
+	 * before the pKVM handling, as the latter could decide to
+	 * UNDEF.
+	 */
+	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_CP15_32]		= kvm_hyp_handle_cp15_32,
+	[ESR_ELx_EC_SYS64]		= kvm_hyp_handle_sysreg,
+	[ESR_ELx_EC_SVE]		= kvm_hyp_handle_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,
+};
+
+static const exit_handler_fn pvm_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_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,
+};
+
+static const exit_handler_fn *kvm_get_exit_handler_array(struct kvm_vcpu *vcpu)
+{
+	if (unlikely(kvm_vm_is_protected(kern_hyp_va(vcpu->kvm))))
+		return pvm_exit_handlers;
+
+	return hyp_exit_handlers;
+}
+
+/*
+ * Some guests (e.g., protected VMs) are not be allowed to run in AArch32.
+ * The ARMv8 architecture does not give the hypervisor a mechanism to prevent a
+ * guest from dropping to AArch32 EL0 if implemented by the CPU. If the
+ * hypervisor spots a guest in such a state ensure it is handled, and don't
+ * trust the host to spot or fix it.  The check below is based on the one in
+ * kvm_arch_vcpu_ioctl_run().
+ *
+ * Returns false if the guest ran in AArch32 when it shouldn't have, and
+ * thus should exit to the host, or true if a the guest run loop can continue.
+ */
+static bool handle_aarch32_guest(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	struct kvm *kvm = kern_hyp_va(vcpu->kvm);
+
+	if (kvm_vm_is_protected(kvm) && vcpu_mode_is_32bit(vcpu)) {
+		/*
+		 * As we have caught the guest red-handed, decide that it isn't
+		 * fit for purpose anymore by making the vcpu invalid. The VMM
+		 * can try and fix it by re-initializing the vcpu with
+		 * KVM_ARM_VCPU_INIT, however, this is likely not possible for
+		 * protected VMs.
+		 */
+		vcpu->arch.target = -1;
+		*exit_code &= BIT(ARM_EXIT_WITH_SERROR_BIT);
+		*exit_code |= ARM_EXCEPTION_IL;
+		return false;
+	}
+
+	return true;
+}
+
 /* Switch to the guest for legacy non-VHE systems */
 int __kvm_vcpu_run(struct kvm_vcpu *vcpu)
 {
···
 	do {
 		/* Jump in the fire! */
 		exit_code = __guest_enter(vcpu);
+
+		if (unlikely(!handle_aarch32_guest(vcpu, &exit_code)))
+			break;
 
 		/* And we're baaack! */
 	} while (fixup_guest_exit(vcpu, &exit_code));

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

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021 Google LLC
+ * Author: Fuad Tabba <[e-mail]>
+ */
+
+#include <linux/irqchip/arm-gic-v3.h>
+
+#include <asm/kvm_asm.h>
+#include <asm/kvm_mmu.h>
+
+#include <hyp/adjust_pc.h>
+
+#include <nvhe/fixed_config.h>
+
+#include "../../sys_regs.h"
+
+/*
+ * Copies of the host's CPU features registers holding sanitized values at hyp.
+ */
+u64 id_aa64pfr0_el1_sys_val;
+u64 id_aa64pfr1_el1_sys_val;
+u64 id_aa64isar0_el1_sys_val;
+u64 id_aa64isar1_el1_sys_val;
+u64 id_aa64mmfr0_el1_sys_val;
+u64 id_aa64mmfr1_el1_sys_val;
+u64 id_aa64mmfr2_el1_sys_val;
+
+/*
+ * Inject an unknown/undefined exception to an AArch64 guest while most of its
+ * sysregs are live.
+ */
+static void inject_undef64(struct kvm_vcpu *vcpu)
+{
+	u32 esr = (ESR_ELx_EC_UNKNOWN << ESR_ELx_EC_SHIFT);
+
+	*vcpu_pc(vcpu) = read_sysreg_el2(SYS_ELR);
+	*vcpu_cpsr(vcpu) = read_sysreg_el2(SYS_SPSR);
+
+	vcpu->arch.flags |= (KVM_ARM64_EXCEPT_AA64_EL1 |
+			     KVM_ARM64_EXCEPT_AA64_ELx_SYNC |
+			     KVM_ARM64_PENDING_EXCEPTION);
+
+	__kvm_adjust_pc(vcpu);
+
+	write_sysreg_el1(esr, SYS_ESR);
+	write_sysreg_el1(read_sysreg_el2(SYS_ELR), SYS_ELR);
+	write_sysreg_el2(*vcpu_pc(vcpu), SYS_ELR);
+	write_sysreg_el2(*vcpu_cpsr(vcpu), SYS_SPSR);
+}
+
+/*
+ * Returns the restricted features values of the feature register based on the
+ * limitations in restrict_fields.
+ * A feature id field value of 0b0000 does not impose any restrictions.
+ * Note: Use only for unsigned feature field values.
+ */
+static u64 get_restricted_features_unsigned(u64 sys_reg_val,
+					    u64 restrict_fields)
+{
+	u64 value = 0UL;
+	u64 mask = GENMASK_ULL(ARM64_FEATURE_FIELD_BITS - 1, 0);
+
+	/*
+	 * According to the Arm Architecture Reference Manual, feature fields
+	 * use increasing values to indicate increases in functionality.
+	 * Iterate over the restricted feature fields and calculate the minimum
+	 * unsigned value between the one supported by the system, and what the
+	 * value is being restricted to.
+	 */
+	while (sys_reg_val && restrict_fields) {
+		value |= min(sys_reg_val & mask, restrict_fields & mask);
+		sys_reg_val &= ~mask;
+		restrict_fields &= ~mask;
+		mask <<= ARM64_FEATURE_FIELD_BITS;
+	}
+
+	return value;
+}
+
+/*
+ * Functions that return the value of feature id registers for protected VMs
+ * based on allowed features, system features, and KVM support.
+ */
+
+static u64 get_pvm_id_aa64pfr0(const struct kvm_vcpu *vcpu)
+{
+	const struct kvm *kvm = (const struct kvm *)kern_hyp_va(vcpu->kvm);
+	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);
+
+	/* Spectre and Meltdown mitigation in KVM */
+	set_mask |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_CSV2),
+			       (u64)kvm->arch.pfr0_csv2);
+	set_mask |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_CSV3),
+			       (u64)kvm->arch.pfr0_csv3);
+
+	return (id_aa64pfr0_el1_sys_val & allow_mask) | set_mask;
+}
+
+static u64 get_pvm_id_aa64pfr1(const struct kvm_vcpu *vcpu)
+{
+	const struct kvm *kvm = (const struct kvm *)kern_hyp_va(vcpu->kvm);
+	u64 allow_mask = PVM_ID_AA64PFR1_ALLOW;
+
+	if (!kvm_has_mte(kvm))
+		allow_mask &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_MTE);
+
+	return id_aa64pfr1_el1_sys_val & allow_mask;
+}
+
+static u64 get_pvm_id_aa64zfr0(const struct kvm_vcpu *vcpu)
+{
+	/*
+	 * No support for Scalable Vectors, therefore, hyp has no sanitized
+	 * copy of the feature id register.
+	 */
+	BUILD_BUG_ON(PVM_ID_AA64ZFR0_ALLOW != 0ULL);
+	return 0;
+}
+
+static u64 get_pvm_id_aa64dfr0(const struct kvm_vcpu *vcpu)
+{
+	/*
+	 * No support for debug, including breakpoints, and watchpoints,
+	 * therefore, pKVM has no sanitized copy of the feature id register.
+	 */
+	BUILD_BUG_ON(PVM_ID_AA64DFR0_ALLOW != 0ULL);
+	return 0;
+}
+
+static u64 get_pvm_id_aa64dfr1(const struct kvm_vcpu *vcpu)
+{
+	/*
+	 * No support for debug, therefore, hyp has no sanitized copy of the
+	 * feature id register.
+	 */
+	BUILD_BUG_ON(PVM_ID_AA64DFR1_ALLOW != 0ULL);
+	return 0;
+}
+
+static u64 get_pvm_id_aa64afr0(const struct kvm_vcpu *vcpu)
+{
+	/*
+	 * No support for implementation defined features, therefore, hyp has no
+	 * sanitized copy of the feature id register.
+	 */
+	BUILD_BUG_ON(PVM_ID_AA64AFR0_ALLOW != 0ULL);
+	return 0;
+}
+
+static u64 get_pvm_id_aa64afr1(const struct kvm_vcpu *vcpu)
+{
+	/*
+	 * No support for implementation defined features, therefore, hyp has no
+	 * sanitized copy of the feature id register.
+	 */
+	BUILD_BUG_ON(PVM_ID_AA64AFR1_ALLOW != 0ULL);
+	return 0;
+}
+
+static u64 get_pvm_id_aa64isar0(const struct kvm_vcpu *vcpu)
+{
+	return id_aa64isar0_el1_sys_val & PVM_ID_AA64ISAR0_ALLOW;
+}
+
+static u64 get_pvm_id_aa64isar1(const struct kvm_vcpu *vcpu)
+{
+	u64 allow_mask = PVM_ID_AA64ISAR1_ALLOW;
+
+	if (!vcpu_has_ptrauth(vcpu))
+		allow_mask &= ~(ARM64_FEATURE_MASK(ID_AA64ISAR1_APA) |
+				ARM64_FEATURE_MASK(ID_AA64ISAR1_API) |
+				ARM64_FEATURE_MASK(ID_AA64ISAR1_GPA) |
+				ARM64_FEATURE_MASK(ID_AA64ISAR1_GPI));
+
+	return id_aa64isar1_el1_sys_val & allow_mask;
+}
+
+static u64 get_pvm_id_aa64mmfr0(const struct kvm_vcpu *vcpu)
+{
+	u64 set_mask;
+
+	set_mask = get_restricted_features_unsigned(id_aa64mmfr0_el1_sys_val,
+		PVM_ID_AA64MMFR0_RESTRICT_UNSIGNED);
+
+	return (id_aa64mmfr0_el1_sys_val & PVM_ID_AA64MMFR0_ALLOW) | set_mask;
+}
+
+static u64 get_pvm_id_aa64mmfr1(const struct kvm_vcpu *vcpu)
+{
+	return id_aa64mmfr1_el1_sys_val & PVM_ID_AA64MMFR1_ALLOW;
+}
+
+static u64 get_pvm_id_aa64mmfr2(const struct kvm_vcpu *vcpu)
+{
+	return id_aa64mmfr2_el1_sys_val & PVM_ID_AA64MMFR2_ALLOW;
+}
+
+/* Read a sanitized cpufeature ID register by its encoding */
+u64 pvm_read_id_reg(const struct kvm_vcpu *vcpu, u32 id)
+{
+	switch (id) {
+	case SYS_ID_AA64PFR0_EL1:
+		return get_pvm_id_aa64pfr0(vcpu);
+	case SYS_ID_AA64PFR1_EL1:
+		return get_pvm_id_aa64pfr1(vcpu);
+	case SYS_ID_AA64ZFR0_EL1:
+		return get_pvm_id_aa64zfr0(vcpu);
+	case SYS_ID_AA64DFR0_EL1:
+		return get_pvm_id_aa64dfr0(vcpu);
+	case SYS_ID_AA64DFR1_EL1:
+		return get_pvm_id_aa64dfr1(vcpu);
+	case SYS_ID_AA64AFR0_EL1:
+		return get_pvm_id_aa64afr0(vcpu);
+	case SYS_ID_AA64AFR1_EL1:
+		return get_pvm_id_aa64afr1(vcpu);
+	case SYS_ID_AA64ISAR0_EL1:
+		return get_pvm_id_aa64isar0(vcpu);
+	case SYS_ID_AA64ISAR1_EL1:
+		return get_pvm_id_aa64isar1(vcpu);
+	case SYS_ID_AA64MMFR0_EL1:
+		return get_pvm_id_aa64mmfr0(vcpu);
+	case SYS_ID_AA64MMFR1_EL1:
+		return get_pvm_id_aa64mmfr1(vcpu);
+	case SYS_ID_AA64MMFR2_EL1:
+		return get_pvm_id_aa64mmfr2(vcpu);
+	default:
+		/*
+		 * Should never happen because all cases are covered in
+		 * pvm_sys_reg_descs[].
+		 */
+		WARN_ON(1);
+		break;
+	}
+
+	return 0;
+}
+
+static u64 read_id_reg(const struct kvm_vcpu *vcpu,
+		       struct sys_reg_desc const *r)
+{
+	return pvm_read_id_reg(vcpu, reg_to_encoding(r));
+}
+
+/* Handler to RAZ/WI sysregs */
+static bool pvm_access_raz_wi(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+			      const struct sys_reg_desc *r)
+{
+	if (!p->is_write)
+		p->regval = 0;
+
+	return true;
+}
+
+/*
+ * Accessor for AArch32 feature id registers.
+ *
+ * The value of these registers is "unknown" according to the spec if AArch32
+ * isn't supported.
+ */
+static bool pvm_access_id_aarch32(struct kvm_vcpu *vcpu,
+				  struct sys_reg_params *p,
+				  const struct sys_reg_desc *r)
+{
+	if (p->is_write) {
+		inject_undef64(vcpu);
+		return false;
+	}
+
+	/*
+	 * No support for AArch32 guests, therefore, pKVM has no sanitized copy
+	 * of AArch32 feature id registers.
+	 */
+	BUILD_BUG_ON(FIELD_GET(ARM64_FEATURE_MASK(ID_AA64PFR0_EL1),
+		     PVM_ID_AA64PFR0_RESTRICT_UNSIGNED) > ID_AA64PFR0_ELx_64BIT_ONLY);
+
+	return pvm_access_raz_wi(vcpu, p, r);
+}
+
+/*
+ * Accessor for AArch64 feature id registers.
+ *
+ * If access is allowed, set the regval to the protected VM's view of the
+ * register and return true.
+ * Otherwise, inject an undefined exception and return false.
+ */
+static bool pvm_access_id_aarch64(struct kvm_vcpu *vcpu,
+				  struct sys_reg_params *p,
+				  const struct sys_reg_desc *r)
+{
+	if (p->is_write) {
+		inject_undef64(vcpu);
+		return false;
+	}
+
+	p->regval = read_id_reg(vcpu, r);
+	return true;
+}
+
+static bool pvm_gic_read_sre(struct kvm_vcpu *vcpu,
+			     struct sys_reg_params *p,
+			     const struct sys_reg_desc *r)
+{
+	/* pVMs only support GICv3. 'nuf said. */
+	if (!p->is_write)
+		p->regval = ICC_SRE_EL1_DIB | ICC_SRE_EL1_DFB | ICC_SRE_EL1_SRE;
+
+	return true;
+}
+
+/* Mark the specified system register as an AArch32 feature id register. */
+#define AARCH32(REG) { SYS_DESC(REG), .access = pvm_access_id_aarch32 }
+
+/* Mark the specified system register as an AArch64 feature id register. */
+#define AARCH64(REG) { SYS_DESC(REG), .access = pvm_access_id_aarch64 }
+
+/* Mark the specified system register as Read-As-Zero/Write-Ignored */
+#define RAZ_WI(REG) { SYS_DESC(REG), .access = pvm_access_raz_wi }
+
+/* Mark the specified system register as not being handled in hyp. */
+#define HOST_HANDLED(REG) { SYS_DESC(REG), .access = NULL }
+
+/*
+ * Architected system registers.
+ * Important: Must be sorted ascending by Op0, Op1, CRn, CRm, Op2
+ *
+ * NOTE: Anything not explicitly listed here is *restricted by default*, i.e.,
+ * it will lead to injecting an exception into the guest.
+ */
+static const struct sys_reg_desc pvm_sys_reg_descs[] = {
+	/* Cache maintenance by set/way operations are restricted. */
+
+	/* Debug and Trace Registers are restricted. */
+
+	/* AArch64 mappings of the AArch32 ID registers */
+	/* CRm=1 */
+	AARCH32(SYS_ID_PFR0_EL1),
+	AARCH32(SYS_ID_PFR1_EL1),
+	AARCH32(SYS_ID_DFR0_EL1),
+	AARCH32(SYS_ID_AFR0_EL1),
+	AARCH32(SYS_ID_MMFR0_EL1),
+	AARCH32(SYS_ID_MMFR1_EL1),
+	AARCH32(SYS_ID_MMFR2_EL1),
+	AARCH32(SYS_ID_MMFR3_EL1),
+
+	/* CRm=2 */
+	AARCH32(SYS_ID_ISAR0_EL1),
+	AARCH32(SYS_ID_ISAR1_EL1),
+	AARCH32(SYS_ID_ISAR2_EL1),
+	AARCH32(SYS_ID_ISAR3_EL1),
+	AARCH32(SYS_ID_ISAR4_EL1),
+	AARCH32(SYS_ID_ISAR5_EL1),
+	AARCH32(SYS_ID_MMFR4_EL1),
+	AARCH32(SYS_ID_ISAR6_EL1),
+
+	/* CRm=3 */
+	AARCH32(SYS_MVFR0_EL1),
+	AARCH32(SYS_MVFR1_EL1),
+	AARCH32(SYS_MVFR2_EL1),
+	AARCH32(SYS_ID_PFR2_EL1),
+	AARCH32(SYS_ID_DFR1_EL1),
+	AARCH32(SYS_ID_MMFR5_EL1),
+
+	/* AArch64 ID registers */
+	/* CRm=4 */
+	AARCH64(SYS_ID_AA64PFR0_EL1),
+	AARCH64(SYS_ID_AA64PFR1_EL1),
+	AARCH64(SYS_ID_AA64ZFR0_EL1),
+	AARCH64(SYS_ID_AA64DFR0_EL1),
+	AARCH64(SYS_ID_AA64DFR1_EL1),
+	AARCH64(SYS_ID_AA64AFR0_EL1),
+	AARCH64(SYS_ID_AA64AFR1_EL1),
+	AARCH64(SYS_ID_AA64ISAR0_EL1),
+	AARCH64(SYS_ID_AA64ISAR1_EL1),
+	AARCH64(SYS_ID_AA64MMFR0_EL1),
+	AARCH64(SYS_ID_AA64MMFR1_EL1),
+	AARCH64(SYS_ID_AA64MMFR2_EL1),
+
+	/* Scalable Vector Registers are restricted. */
+
+	RAZ_WI(SYS_ERRIDR_EL1),
+	RAZ_WI(SYS_ERRSELR_EL1),
+	RAZ_WI(SYS_ERXFR_EL1),
+	RAZ_WI(SYS_ERXCTLR_EL1),
+	RAZ_WI(SYS_ERXSTATUS_EL1),
+	RAZ_WI(SYS_ERXADDR_EL1),
+	RAZ_WI(SYS_ERXMISC0_EL1),
+	RAZ_WI(SYS_ERXMISC1_EL1),
+
+	/* Performance Monitoring Registers are restricted. */
+
+	/* Limited Ordering Regions Registers are restricted. */
+
+	HOST_HANDLED(SYS_ICC_SGI1R_EL1),
+	HOST_HANDLED(SYS_ICC_ASGI1R_EL1),
+	HOST_HANDLED(SYS_ICC_SGI0R_EL1),
+	{ SYS_DESC(SYS_ICC_SRE_EL1), .access = pvm_gic_read_sre, },
+
+	HOST_HANDLED(SYS_CCSIDR_EL1),
+	HOST_HANDLED(SYS_CLIDR_EL1),
+	HOST_HANDLED(SYS_CSSELR_EL1),
+	HOST_HANDLED(SYS_CTR_EL0),
+
+	/* Performance Monitoring Registers are restricted. */
+
+	/* Activity Monitoring Registers are restricted. */
+
+	HOST_HANDLED(SYS_CNTP_TVAL_EL0),
+	HOST_HANDLED(SYS_CNTP_CTL_EL0),
+	HOST_HANDLED(SYS_CNTP_CVAL_EL0),
+
+	/* Performance Monitoring Registers are restricted. */
+};
+
+/*
+ * Checks that the sysreg table is unique and in-order.
+ *
+ * Returns 0 if the table is consistent, or 1 otherwise.
+ */
+int kvm_check_pvm_sysreg_table(void)
+{
+	unsigned int i;
+
+	for (i = 1; i < ARRAY_SIZE(pvm_sys_reg_descs); i++) {
+		if (cmp_sys_reg(&pvm_sys_reg_descs[i-1], &pvm_sys_reg_descs[i]) >= 0)
+			return 1;
+	}
+
+	return 0;
+}
+
+/*
+ * Handler for protected VM MSR, MRS or System instruction execution.
+ *
+ * Returns true if the hypervisor has handled the exit, and control should go
+ * back to the guest, or false if it hasn't, to be handled by the host.
+ */
+bool kvm_handle_pvm_sysreg(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	const struct sys_reg_desc *r;
+	struct sys_reg_params params;
+	unsigned long esr = kvm_vcpu_get_esr(vcpu);
+	int Rt = kvm_vcpu_sys_get_rt(vcpu);
+
+	params = esr_sys64_to_params(esr);
+	params.regval = vcpu_get_reg(vcpu, Rt);
+
+	r = find_reg(&params, pvm_sys_reg_descs, ARRAY_SIZE(pvm_sys_reg_descs));
+
+	/* Undefined (RESTRICTED). */
+	if (r == NULL) {
+		inject_undef64(vcpu);
+		return true;
+	}
+
+	/* Handled by the host (HOST_HANDLED) */
+	if (r->access == NULL)
+		return false;
+
+	/* Handled by hyp: skip instruction if instructed to do so. */
+	if (r->access(vcpu, &params, r))
+		__kvm_skip_instr(vcpu);
+
+	if (!params.is_write)
+		vcpu_set_reg(vcpu, Rt, params.regval);
+
+	return true;
+}
+
+/**
+ * Handler for protected VM restricted exceptions.
+ *
+ * Inject an undefined exception into the guest and return true to indicate that
+ * the hypervisor has handled the exit, and control should go back to the guest.
+ */
+bool kvm_handle_pvm_restricted(struct kvm_vcpu *vcpu, u64 *exit_code)
+{
+	inject_undef64(vcpu);
+	return true;
+}

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

···
 		goto spurious;
 
 	lr_val &= ~ICH_LR_STATE;
-	/* No active state for LPIs */
-	if ((lr_val & ICH_LR_VIRTUAL_ID_MASK) <= VGIC_MAX_SPI)
-		lr_val |= ICH_LR_ACTIVE_BIT;
+	lr_val |= ICH_LR_ACTIVE_BIT;
 	__gic_v3_set_lr(lr_val, lr);
 	__vgic_v3_set_active_priority(lr_prio, vmcr, grp);
 	vcpu_set_reg(vcpu, rt, lr_val & ICH_LR_VIRTUAL_ID_MASK);
···
 	/* Drop priority in any case */
 	act_prio = __vgic_v3_clear_highest_active_priority();
 
-	/* If EOIing an LPI, no deactivate to be performed */
-	if (vid >= VGIC_MIN_LPI)
-		return;
-
-	/* EOImode == 1, nothing to be done here */
-	if (vmcr & ICH_VMCR_EOIM_MASK)
-		return;
-
 	lr = __vgic_v3_find_active_lr(vcpu, vid, &lr_val);
 	if (lr == -1) {
-		__vgic_v3_bump_eoicount();
+		/* Do not bump EOIcount for LPIs that aren't in the LRs */
+		if (!(vid >= VGIC_MIN_LPI))
+			__vgic_v3_bump_eoicount();
 		return;
 	}
+
+	/* EOImode == 1 and not an LPI, nothing to be done here */
+	if ((vmcr & ICH_VMCR_EOIM_MASK) && !(vid >= VGIC_MIN_LPI))
+		return;
 
 	lr_prio = (lr_val & ICH_LR_PRIORITY_MASK) >> ICH_LR_PRIORITY_SHIFT;
 
···
 	val = ((vtr >> 29) & 7) << ICC_CTLR_EL1_PRI_BITS_SHIFT;
 	/* IDbits */
 	val |= ((vtr >> 23) & 7) << ICC_CTLR_EL1_ID_BITS_SHIFT;
-	/* SEIS */
-	val |= ((vtr >> 22) & 1) << ICC_CTLR_EL1_SEIS_SHIFT;
 	/* A3V */
 	val |= ((vtr >> 21) & 1) << ICC_CTLR_EL1_A3V_SHIFT;
 	/* EOImode */

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

···
 	__deactivate_traps_common(vcpu);
 }
 
+static const exit_handler_fn hyp_exit_handlers[] = {
+	[0 ... ESR_ELx_EC_MAX]		= NULL,
+	[ESR_ELx_EC_CP15_32]		= kvm_hyp_handle_cp15_32,
+	[ESR_ELx_EC_SYS64]		= kvm_hyp_handle_sysreg,
+	[ESR_ELx_EC_SVE]		= kvm_hyp_handle_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,
+};
+
+static const exit_handler_fn *kvm_get_exit_handler_array(struct kvm_vcpu *vcpu)
+{
+	return hyp_exit_handlers;
+}
+
 /* Switch to the guest for VHE systems running in EL2 */
 static int __kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu)
 {

arch/arm64/kvm/mmu.c

···
 		return -EINVAL;
 	}
 
-	pgt = kzalloc(sizeof(*pgt), GFP_KERNEL);
+	pgt = kzalloc(sizeof(*pgt), GFP_KERNEL_ACCOUNT);
 	if (!pgt)
 		return -ENOMEM;
 

arch/arm64/kvm/pmu-emul.c

···
 		mutex_lock(&vcpu->kvm->lock);
 
 		if (!vcpu->kvm->arch.pmu_filter) {
-			vcpu->kvm->arch.pmu_filter = bitmap_alloc(nr_events, GFP_KERNEL);
+			vcpu->kvm->arch.pmu_filter = bitmap_alloc(nr_events, GFP_KERNEL_ACCOUNT);
 			if (!vcpu->kvm->arch.pmu_filter) {
 				mutex_unlock(&vcpu->kvm->lock);
 				return -ENOMEM;

arch/arm64/kvm/reset.c

···
 		    vl > SVE_VL_ARCH_MAX))
 		return -EIO;
 
-	buf = kzalloc(SVE_SIG_REGS_SIZE(sve_vq_from_vl(vl)), GFP_KERNEL);
+	buf = kzalloc(SVE_SIG_REGS_SIZE(sve_vq_from_vl(vl)), GFP_KERNEL_ACCOUNT);
 	if (!buf)
 		return -ENOMEM;
 

arch/arm64/kvm/sys_regs.c

···
 		struct sys_reg_desc const *r, bool raz)
 {
 	u32 id = reg_to_encoding(r);
-	u64 val = raz ? 0 : read_sanitised_ftr_reg(id);
+	u64 val;
+
+	if (raz)
+		return 0;
+
+	val = read_sanitised_ftr_reg(id);
 
 	switch (id) {
 	case SYS_ID_AA64PFR0_EL1:
···
 		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_CSV2), (u64)vcpu->kvm->arch.pfr0_csv2);
 		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_CSV3);
 		val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_CSV3), (u64)vcpu->kvm->arch.pfr0_csv3);
+		if (irqchip_in_kernel(vcpu->kvm) &&
+		    vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+			val &= ~ARM64_FEATURE_MASK(ID_AA64PFR0_GIC);
+			val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR0_GIC), 1);
+		}
 		break;
 	case SYS_ID_AA64PFR1_EL1:
-		val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_MTE);
-		if (kvm_has_mte(vcpu->kvm)) {
-			u64 pfr, mte;
-
-			pfr = read_sanitised_ftr_reg(SYS_ID_AA64PFR1_EL1);
-			mte = cpuid_feature_extract_unsigned_field(pfr, ID_AA64PFR1_MTE_SHIFT);
-			val |= FIELD_PREP(ARM64_FEATURE_MASK(ID_AA64PFR1_MTE), mte);
-		}
+		if (!kvm_has_mte(vcpu->kvm))
+			val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_MTE);
 		break;
 	case SYS_ID_AA64ISAR1_EL1:
 		if (!vcpu_has_ptrauth(vcpu))
···
 	return __set_id_reg(vcpu, rd, uaddr, raz);
 }
 
-static int get_raz_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
-			  const struct kvm_one_reg *reg, void __user *uaddr)
-{
-	return __get_id_reg(vcpu, rd, uaddr, true);
-}
-
 static int set_raz_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
 			  const struct kvm_one_reg *reg, void __user *uaddr)
 {
 	return __set_id_reg(vcpu, rd, uaddr, true);
+}
+
+static int get_raz_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
+		       const struct kvm_one_reg *reg, void __user *uaddr)
+{
+	const u64 id = sys_reg_to_index(rd);
+	const u64 val = 0;
+
+	return reg_to_user(uaddr, &val, id);
 }
 
 static int set_wi_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
···
 #define ID_UNALLOCATED(crm, op2) {			\
 	Op0(3), Op1(0), CRn(0), CRm(crm), Op2(op2),	\
 	.access = access_raz_id_reg,			\
-	.get_user = get_raz_id_reg,			\
+	.get_user = get_raz_reg,			\
 	.set_user = set_raz_id_reg,			\
 }
 
···
 #define ID_HIDDEN(name) {			\
 	SYS_DESC(SYS_##name),			\
 	.access = access_raz_id_reg,		\
-	.get_user = get_raz_id_reg,		\
+	.get_user = get_raz_reg,		\
 	.set_user = set_raz_id_reg,		\
 }
 
···
 	 * previously (and pointlessly) advertised in the past...
 	 */
 	{ PMU_SYS_REG(SYS_PMSWINC_EL0),
-	  .get_user = get_raz_id_reg, .set_user = set_wi_reg,
+	  .get_user = get_raz_reg, .set_user = set_wi_reg,
 	  .access = access_pmswinc, .reset = NULL },
 	{ PMU_SYS_REG(SYS_PMSELR_EL0),
 	  .access = access_pmselr, .reset = reset_pmselr, .reg = PMSELR_EL0 },

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

···
 	struct kvm_vcpu *vcpu0 = kvm_get_vcpu(kvm, 0);
 	int i;
 
-	dist->spis = kcalloc(nr_spis, sizeof(struct vgic_irq), GFP_KERNEL);
+	dist->spis = kcalloc(nr_spis, sizeof(struct vgic_irq), GFP_KERNEL_ACCOUNT);
 	if (!dist->spis)
 		return  -ENOMEM;
 

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

···
 	u32 nr = dist->nr_spis;
 	int i, ret;
 
-	entries = kcalloc(nr, sizeof(*entries), GFP_KERNEL);
+	entries = kcalloc(nr, sizeof(*entries), GFP_KERNEL_ACCOUNT);
 	if (!entries)
 		return -ENOMEM;
 

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

···
 	if (irq)
 		return irq;
 
-	irq = kzalloc(sizeof(struct vgic_irq), GFP_KERNEL);
+	irq = kzalloc(sizeof(struct vgic_irq), GFP_KERNEL_ACCOUNT);
 	if (!irq)
 		return ERR_PTR(-ENOMEM);
 
···
 	 * we must be careful not to overrun the array.
 	 */
 	irq_count = READ_ONCE(dist->lpi_list_count);
-	intids = kmalloc_array(irq_count, sizeof(intids[0]), GFP_KERNEL);
+	intids = kmalloc_array(irq_count, sizeof(intids[0]), GFP_KERNEL_ACCOUNT);
 	if (!intids)
 		return -ENOMEM;
 
···
 	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);
+	collection = kzalloc(sizeof(*collection), GFP_KERNEL_ACCOUNT);
 	if (!collection)
 		return -ENOMEM;
 
···
 {
 	struct its_ite *ite;
 
-	ite = kzalloc(sizeof(*ite), GFP_KERNEL);
+	ite = kzalloc(sizeof(*ite), GFP_KERNEL_ACCOUNT);
 	if (!ite)
 		return ERR_PTR(-ENOMEM);
 
···
 {
 	struct its_device *device;
 
-	device = kzalloc(sizeof(*device), GFP_KERNEL);
+	device = kzalloc(sizeof(*device), GFP_KERNEL_ACCOUNT);
 	if (!device)
 		return ERR_PTR(-ENOMEM);
 
···
 		struct vgic_translation_cache_entry *cte;
 
 		/* An allocation failure is not fatal */
-		cte = kzalloc(sizeof(*cte), GFP_KERNEL);
+		cte = kzalloc(sizeof(*cte), GFP_KERNEL_ACCOUNT);
 		if (WARN_ON(!cte))
 			break;
 
···
 	if (type != KVM_DEV_TYPE_ARM_VGIC_ITS)
 		return -ENODEV;
 
-	its = kzalloc(sizeof(struct vgic_its), GFP_KERNEL);
+	its = kzalloc(sizeof(struct vgic_its), GFP_KERNEL_ACCOUNT);
 	if (!its)
 		return -ENOMEM;
 
···
 		if (copy_from_user(&addr, uaddr, sizeof(addr)))
 			return -EFAULT;
 
-		ret = vgic_check_ioaddr(dev->kvm, &its->vgic_its_base,
-					addr, SZ_64K);
+		ret = vgic_check_iorange(dev->kvm, its->vgic_its_base,
+					 addr, SZ_64K, KVM_VGIC_V3_ITS_SIZE);
 		if (ret)
 			return ret;
 

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

···
 
 /* common helpers */
 
-int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
-		      phys_addr_t addr, phys_addr_t alignment)
+int vgic_check_iorange(struct kvm *kvm, phys_addr_t ioaddr,
+		       phys_addr_t addr, phys_addr_t alignment,
+		       phys_addr_t size)
 {
-	if (addr & ~kvm_phys_mask(kvm))
-		return -E2BIG;
+	if (!IS_VGIC_ADDR_UNDEF(ioaddr))
+		return -EEXIST;
 
-	if (!IS_ALIGNED(addr, alignment))
+	if (!IS_ALIGNED(addr, alignment) || !IS_ALIGNED(size, alignment))
 		return -EINVAL;
 
-	if (!IS_VGIC_ADDR_UNDEF(*ioaddr))
-		return -EEXIST;
+	if (addr + size < addr)
+		return -EINVAL;
+
+	if (addr & ~kvm_phys_mask(kvm) || addr + size > kvm_phys_size(kvm))
+		return -E2BIG;
 
 	return 0;
 }
···
 {
 	int r = 0;
 	struct vgic_dist *vgic = &kvm->arch.vgic;
-	phys_addr_t *addr_ptr, alignment;
+	phys_addr_t *addr_ptr, alignment, size;
 	u64 undef_value = VGIC_ADDR_UNDEF;
 
 	mutex_lock(&kvm->lock);
···
 		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V2);
 		addr_ptr = &vgic->vgic_dist_base;
 		alignment = SZ_4K;
+		size = KVM_VGIC_V2_DIST_SIZE;
 		break;
 	case KVM_VGIC_V2_ADDR_TYPE_CPU:
 		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V2);
 		addr_ptr = &vgic->vgic_cpu_base;
 		alignment = SZ_4K;
+		size = KVM_VGIC_V2_CPU_SIZE;
 		break;
 	case KVM_VGIC_V3_ADDR_TYPE_DIST:
 		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V3);
 		addr_ptr = &vgic->vgic_dist_base;
 		alignment = SZ_64K;
+		size = KVM_VGIC_V3_DIST_SIZE;
 		break;
 	case KVM_VGIC_V3_ADDR_TYPE_REDIST: {
 		struct vgic_redist_region *rdreg;
···
 		goto out;
 
 	if (write) {
-		r = vgic_check_ioaddr(kvm, addr_ptr, *addr, alignment);
+		r = vgic_check_iorange(kvm, *addr_ptr, *addr, alignment, size);
 		if (!r)
 			*addr_ptr = *addr;
 	} else {

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

···
 	struct vgic_dist *d = &kvm->arch.vgic;
 	struct vgic_redist_region *rdreg;
 	struct list_head *rd_regions = &d->rd_regions;
-	size_t size = count * KVM_VGIC_V3_REDIST_SIZE;
+	int nr_vcpus = atomic_read(&kvm->online_vcpus);
+	size_t size = count ? count * KVM_VGIC_V3_REDIST_SIZE
+			    : nr_vcpus * KVM_VGIC_V3_REDIST_SIZE;
 	int ret;
 
 	/* cross the end of memory ? */
···
 	if (vgic_v3_rdist_overlap(kvm, base, size))
 		return -EINVAL;
 
-	rdreg = kzalloc(sizeof(*rdreg), GFP_KERNEL);
+	rdreg = kzalloc(sizeof(*rdreg), GFP_KERNEL_ACCOUNT);
 	if (!rdreg)
 		return -ENOMEM;
 
 	rdreg->base = VGIC_ADDR_UNDEF;
 
-	ret = vgic_check_ioaddr(kvm, &rdreg->base, base, SZ_64K);
+	ret = vgic_check_iorange(kvm, rdreg->base, base, SZ_64K, size);
 	if (ret)
 		goto free;
 

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

···
 static bool group0_trap;
 static bool group1_trap;
 static bool common_trap;
+static bool dir_trap;
 static bool gicv4_enable;
 
 void vgic_v3_set_underflow(struct kvm_vcpu *vcpu)
···
 		vgic_v3->vgic_hcr |= ICH_HCR_TALL1;
 	if (common_trap)
 		vgic_v3->vgic_hcr |= ICH_HCR_TC;
+	if (dir_trap)
+		vgic_v3->vgic_hcr |= ICH_HCR_TDIR;
 }
 
 int vgic_v3_lpi_sync_pending_status(struct kvm *kvm, struct vgic_irq *irq)
···
 		return false;
 
 	list_for_each_entry(rdreg, &d->rd_regions, list) {
-		if (rdreg->base + vgic_v3_rd_region_size(kvm, rdreg) <
-			rdreg->base)
+		size_t sz = vgic_v3_rd_region_size(kvm, rdreg);
+
+		if (vgic_check_iorange(kvm, VGIC_ADDR_UNDEF,
+				       rdreg->base, SZ_64K, sz))
 			return false;
 	}
 
···
 		group1_trap = true;
 	}
 
-	if (group0_trap || group1_trap || common_trap) {
-		kvm_info("GICv3 sysreg trapping enabled ([%s%s%s], reduced performance)\n",
+	if (kvm_vgic_global_state.ich_vtr_el2 & ICH_VTR_SEIS_MASK) {
+		kvm_info("GICv3 with locally generated SEI\n");
+
+		group0_trap = true;
+		group1_trap = true;
+		if (ich_vtr_el2 & ICH_VTR_TDS_MASK)
+			dir_trap = true;
+		else
+			common_trap = true;
+	}
+
+	if (group0_trap || group1_trap || common_trap | dir_trap) {
+		kvm_info("GICv3 sysreg trapping enabled ([%s%s%s%s], reduced performance)\n",
 			 group0_trap ? "G0" : "",
 			 group1_trap ? "G1" : "",
-			 common_trap ? "C"  : "");
+			 common_trap ? "C"  : "",
+			 dir_trap    ? "D"  : "");
 		static_branch_enable(&vgic_v3_cpuif_trap);
 	}
 

arch/arm64/kvm/vgic/vgic-v4.c

···
 	nr_vcpus = atomic_read(&kvm->online_vcpus);
 
 	dist->its_vm.vpes = kcalloc(nr_vcpus, sizeof(*dist->its_vm.vpes),
-				    GFP_KERNEL);
+				    GFP_KERNEL_ACCOUNT);
 	if (!dist->its_vm.vpes)
 		return -ENOMEM;
 

arch/arm64/kvm/vgic/vgic.h

···
 void vgic_irq_handle_resampling(struct vgic_irq *irq,
 				bool lr_deactivated, bool lr_pending);
 
-int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
-		      phys_addr_t addr, phys_addr_t alignment);
+int vgic_check_iorange(struct kvm *kvm, phys_addr_t ioaddr,
+		       phys_addr_t addr, phys_addr_t alignment,
+		       phys_addr_t size);
 
 void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu);
 void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr);

arch/mips/kvm/mips.c

···
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 	case KVM_CAP_MIPS_FPU:
 		/* We don't handle systems with inconsistent cpu_has_fpu */

arch/powerpc/include/asm/kvm_book3s.h

···
 #define SPLIT_HACK_OFFS			0xfb000000
 
 /*
- * This packs a VCPU ID from the [0..KVM_MAX_VCPU_ID) space down to the
+ * This packs a VCPU ID from the [0..KVM_MAX_VCPU_IDS) space down to the
  * [0..KVM_MAX_VCPUS) space, using knowledge of the guest's core stride
  * (but not its actual threading mode, which is not available) to avoid
  * collisions.

arch/powerpc/include/asm/kvm_host.h

···
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 #include <asm/kvm_book3s_asm.h>		/* for MAX_SMT_THREADS */
-#define KVM_MAX_VCPU_ID		(MAX_SMT_THREADS * KVM_MAX_VCORES)
+#define KVM_MAX_VCPU_IDS	(MAX_SMT_THREADS * KVM_MAX_VCORES)
 #define KVM_MAX_NESTED_GUESTS	KVMPPC_NR_LPIDS
 
 #else
-#define KVM_MAX_VCPU_ID		KVM_MAX_VCPUS
+#define KVM_MAX_VCPU_IDS	KVM_MAX_VCPUS
 #endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 
 #define __KVM_HAVE_ARCH_INTC_INITIALIZED

arch/powerpc/kvm/book3s_xive.c

···
 
 	pr_devel("%s nr_servers=%u\n", __func__, nr_servers);
 
-	if (!nr_servers || nr_servers > KVM_MAX_VCPU_ID)
+	if (!nr_servers || nr_servers > KVM_MAX_VCPU_IDS)
 		return -EINVAL;
 
 	mutex_lock(&xive->lock);

arch/powerpc/kvm/powerpc.c

···
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 #ifdef CONFIG_PPC_BOOK3S_64
 	case KVM_CAP_PPC_GET_SMMU_INFO:

arch/riscv/Kconfig

···
 source "kernel/power/Kconfig"
 
 endmenu
+
+source "arch/riscv/kvm/Kconfig"

arch/riscv/Makefile

···
 head-y := arch/riscv/kernel/head.o
 
 core-$(CONFIG_RISCV_ERRATA_ALTERNATIVE) += arch/riscv/errata/
+core-$(CONFIG_KVM) += arch/riscv/kvm/
 
 libs-y += arch/riscv/lib/
 libs-$(CONFIG_EFI_STUB) += $(objtree)/drivers/firmware/efi/libstub/lib.a

arch/riscv/include/asm/csr.h

···
 
 /* Interrupt causes (minus the high bit) */
 #define IRQ_S_SOFT		1
+#define IRQ_VS_SOFT		2
 #define IRQ_M_SOFT		3
 #define IRQ_S_TIMER		5
+#define IRQ_VS_TIMER		6
 #define IRQ_M_TIMER		7
 #define IRQ_S_EXT		9
+#define IRQ_VS_EXT		10
 #define IRQ_M_EXT		11
 
 /* Exception causes */
 #define EXC_INST_MISALIGNED	0
 #define EXC_INST_ACCESS		1
+#define EXC_INST_ILLEGAL	2
 #define EXC_BREAKPOINT		3
 #define EXC_LOAD_ACCESS		5
 #define EXC_STORE_ACCESS	7
 #define EXC_SYSCALL		8
+#define EXC_HYPERVISOR_SYSCALL	9
+#define EXC_SUPERVISOR_SYSCALL	10
 #define EXC_INST_PAGE_FAULT	12
 #define EXC_LOAD_PAGE_FAULT	13
 #define EXC_STORE_PAGE_FAULT	15
+#define EXC_INST_GUEST_PAGE_FAULT	20
+#define EXC_LOAD_GUEST_PAGE_FAULT	21
+#define EXC_VIRTUAL_INST_FAULT		22
+#define EXC_STORE_GUEST_PAGE_FAULT	23
 
 /* PMP configuration */
 #define PMP_R			0x01
···
 #define PMP_A_NA4		0x10
 #define PMP_A_NAPOT		0x18
 #define PMP_L			0x80
+
+/* HSTATUS flags */
+#ifdef CONFIG_64BIT
+#define HSTATUS_VSXL		_AC(0x300000000, UL)
+#define HSTATUS_VSXL_SHIFT	32
+#endif
+#define HSTATUS_VTSR		_AC(0x00400000, UL)
+#define HSTATUS_VTW		_AC(0x00200000, UL)
+#define HSTATUS_VTVM		_AC(0x00100000, UL)
+#define HSTATUS_VGEIN		_AC(0x0003f000, UL)
+#define HSTATUS_VGEIN_SHIFT	12
+#define HSTATUS_HU		_AC(0x00000200, UL)
+#define HSTATUS_SPVP		_AC(0x00000100, UL)
+#define HSTATUS_SPV		_AC(0x00000080, UL)
+#define HSTATUS_GVA		_AC(0x00000040, UL)
+#define HSTATUS_VSBE		_AC(0x00000020, UL)
+
+/* HGATP flags */
+#define HGATP_MODE_OFF		_AC(0, UL)
+#define HGATP_MODE_SV32X4	_AC(1, UL)
+#define HGATP_MODE_SV39X4	_AC(8, UL)
+#define HGATP_MODE_SV48X4	_AC(9, UL)
+
+#define HGATP32_MODE_SHIFT	31
+#define HGATP32_VMID_SHIFT	22
+#define HGATP32_VMID_MASK	_AC(0x1FC00000, UL)
+#define HGATP32_PPN		_AC(0x003FFFFF, UL)
+
+#define HGATP64_MODE_SHIFT	60
+#define HGATP64_VMID_SHIFT	44
+#define HGATP64_VMID_MASK	_AC(0x03FFF00000000000, UL)
+#define HGATP64_PPN		_AC(0x00000FFFFFFFFFFF, UL)
+
+#define HGATP_PAGE_SHIFT	12
+
+#ifdef CONFIG_64BIT
+#define HGATP_PPN		HGATP64_PPN
+#define HGATP_VMID_SHIFT	HGATP64_VMID_SHIFT
+#define HGATP_VMID_MASK		HGATP64_VMID_MASK
+#define HGATP_MODE_SHIFT	HGATP64_MODE_SHIFT
+#else
+#define HGATP_PPN		HGATP32_PPN
+#define HGATP_VMID_SHIFT	HGATP32_VMID_SHIFT
+#define HGATP_VMID_MASK		HGATP32_VMID_MASK
+#define HGATP_MODE_SHIFT	HGATP32_MODE_SHIFT
+#endif
+
+/* VSIP & HVIP relation */
+#define VSIP_TO_HVIP_SHIFT	(IRQ_VS_SOFT - IRQ_S_SOFT)
+#define VSIP_VALID_MASK		((_AC(1, UL) << IRQ_S_SOFT) | \
+				 (_AC(1, UL) << IRQ_S_TIMER) | \
+				 (_AC(1, UL) << IRQ_S_EXT))
 
 /* symbolic CSR names: */
 #define CSR_CYCLE		0xc00
···
 #define CSR_STVAL		0x143
 #define CSR_SIP			0x144
 #define CSR_SATP		0x180
+
+#define CSR_VSSTATUS		0x200
+#define CSR_VSIE		0x204
+#define CSR_VSTVEC		0x205
+#define CSR_VSSCRATCH		0x240
+#define CSR_VSEPC		0x241
+#define CSR_VSCAUSE		0x242
+#define CSR_VSTVAL		0x243
+#define CSR_VSIP		0x244
+#define CSR_VSATP		0x280
+
+#define CSR_HSTATUS		0x600
+#define CSR_HEDELEG		0x602
+#define CSR_HIDELEG		0x603
+#define CSR_HIE			0x604
+#define CSR_HTIMEDELTA		0x605
+#define CSR_HCOUNTEREN		0x606
+#define CSR_HGEIE		0x607
+#define CSR_HTIMEDELTAH		0x615
+#define CSR_HTVAL		0x643
+#define CSR_HIP			0x644
+#define CSR_HVIP		0x645
+#define CSR_HTINST		0x64a
+#define CSR_HGATP		0x680
+#define CSR_HGEIP		0xe12
 
 #define CSR_MSTATUS		0x300
 #define CSR_MISA		0x301

arch/riscv/include/asm/kvm_host.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#ifndef __RISCV_KVM_HOST_H__
+#define __RISCV_KVM_HOST_H__
+
+#include <linux/types.h>
+#include <linux/kvm.h>
+#include <linux/kvm_types.h>
+#include <asm/kvm_vcpu_fp.h>
+#include <asm/kvm_vcpu_timer.h>
+
+#ifdef CONFIG_64BIT
+#define KVM_MAX_VCPUS			(1U << 16)
+#else
+#define KVM_MAX_VCPUS			(1U << 9)
+#endif
+
+#define KVM_HALT_POLL_NS_DEFAULT	500000
+
+#define KVM_VCPU_MAX_FEATURES		0
+
+#define KVM_REQ_SLEEP \
+	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)
+
+struct kvm_vm_stat {
+	struct kvm_vm_stat_generic generic;
+};
+
+struct kvm_vcpu_stat {
+	struct kvm_vcpu_stat_generic generic;
+	u64 ecall_exit_stat;
+	u64 wfi_exit_stat;
+	u64 mmio_exit_user;
+	u64 mmio_exit_kernel;
+	u64 exits;
+};
+
+struct kvm_arch_memory_slot {
+};
+
+struct kvm_vmid {
+	/*
+	 * Writes to vmid_version and vmid happen with vmid_lock held
+	 * whereas reads happen without any lock held.
+	 */
+	unsigned long vmid_version;
+	unsigned long vmid;
+};
+
+struct kvm_arch {
+	/* stage2 vmid */
+	struct kvm_vmid vmid;
+
+	/* stage2 page table */
+	pgd_t *pgd;
+	phys_addr_t pgd_phys;
+
+	/* Guest Timer */
+	struct kvm_guest_timer timer;
+};
+
+struct kvm_mmio_decode {
+	unsigned long insn;
+	int insn_len;
+	int len;
+	int shift;
+	int return_handled;
+};
+
+struct kvm_sbi_context {
+	int return_handled;
+};
+
+#define KVM_MMU_PAGE_CACHE_NR_OBJS	32
+
+struct kvm_mmu_page_cache {
+	int nobjs;
+	void *objects[KVM_MMU_PAGE_CACHE_NR_OBJS];
+};
+
+struct kvm_cpu_trap {
+	unsigned long sepc;
+	unsigned long scause;
+	unsigned long stval;
+	unsigned long htval;
+	unsigned long htinst;
+};
+
+struct kvm_cpu_context {
+	unsigned long zero;
+	unsigned long ra;
+	unsigned long sp;
+	unsigned long gp;
+	unsigned long tp;
+	unsigned long t0;
+	unsigned long t1;
+	unsigned long t2;
+	unsigned long s0;
+	unsigned long s1;
+	unsigned long a0;
+	unsigned long a1;
+	unsigned long a2;
+	unsigned long a3;
+	unsigned long a4;
+	unsigned long a5;
+	unsigned long a6;
+	unsigned long a7;
+	unsigned long s2;
+	unsigned long s3;
+	unsigned long s4;
+	unsigned long s5;
+	unsigned long s6;
+	unsigned long s7;
+	unsigned long s8;
+	unsigned long s9;
+	unsigned long s10;
+	unsigned long s11;
+	unsigned long t3;
+	unsigned long t4;
+	unsigned long t5;
+	unsigned long t6;
+	unsigned long sepc;
+	unsigned long sstatus;
+	unsigned long hstatus;
+	union __riscv_fp_state fp;
+};
+
+struct kvm_vcpu_csr {
+	unsigned long vsstatus;
+	unsigned long vsie;
+	unsigned long vstvec;
+	unsigned long vsscratch;
+	unsigned long vsepc;
+	unsigned long vscause;
+	unsigned long vstval;
+	unsigned long hvip;
+	unsigned long vsatp;
+	unsigned long scounteren;
+};
+
+struct kvm_vcpu_arch {
+	/* VCPU ran at least once */
+	bool ran_atleast_once;
+
+	/* ISA feature bits (similar to MISA) */
+	unsigned long isa;
+
+	/* SSCRATCH, STVEC, and SCOUNTEREN of Host */
+	unsigned long host_sscratch;
+	unsigned long host_stvec;
+	unsigned long host_scounteren;
+
+	/* CPU context of Host */
+	struct kvm_cpu_context host_context;
+
+	/* CPU context of Guest VCPU */
+	struct kvm_cpu_context guest_context;
+
+	/* CPU CSR context of Guest VCPU */
+	struct kvm_vcpu_csr guest_csr;
+
+	/* CPU context upon Guest VCPU reset */
+	struct kvm_cpu_context guest_reset_context;
+
+	/* CPU CSR context upon Guest VCPU reset */
+	struct kvm_vcpu_csr guest_reset_csr;
+
+	/*
+	 * VCPU interrupts
+	 *
+	 * We have a lockless approach for tracking pending VCPU interrupts
+	 * implemented using atomic bitops. The irqs_pending bitmap represent
+	 * pending interrupts whereas irqs_pending_mask represent bits changed
+	 * in irqs_pending. Our approach is modeled around multiple producer
+	 * and single consumer problem where the consumer is the VCPU itself.
+	 */
+	unsigned long irqs_pending;
+	unsigned long irqs_pending_mask;
+
+	/* VCPU Timer */
+	struct kvm_vcpu_timer timer;
+
+	/* MMIO instruction details */
+	struct kvm_mmio_decode mmio_decode;
+
+	/* SBI context */
+	struct kvm_sbi_context sbi_context;
+
+	/* Cache pages needed to program page tables with spinlock held */
+	struct kvm_mmu_page_cache mmu_page_cache;
+
+	/* VCPU power-off state */
+	bool power_off;
+
+	/* Don't run the VCPU (blocked) */
+	bool pause;
+
+	/* SRCU lock index for in-kernel run loop */
+	int srcu_idx;
+};
+
+static inline void kvm_arch_hardware_unsetup(void) {}
+static inline void kvm_arch_sync_events(struct kvm *kvm) {}
+static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
+static inline void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu) {}
+
+#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);
+
+int kvm_riscv_stage2_map(struct kvm_vcpu *vcpu,
+			 struct kvm_memory_slot *memslot,
+			 gpa_t gpa, unsigned long hva, bool is_write);
+void kvm_riscv_stage2_flush_cache(struct kvm_vcpu *vcpu);
+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);
+
+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_unpriv_trap(void);
+
+unsigned long kvm_riscv_vcpu_unpriv_read(struct kvm_vcpu *vcpu,
+					 bool read_insn,
+					 unsigned long guest_addr,
+					 struct kvm_cpu_trap *trap);
+void kvm_riscv_vcpu_trap_redirect(struct kvm_vcpu *vcpu,
+				  struct kvm_cpu_trap *trap);
+int kvm_riscv_vcpu_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
+int kvm_riscv_vcpu_exit(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			struct kvm_cpu_trap *trap);
+
+void __kvm_riscv_switch_to(struct kvm_vcpu_arch *vcpu_arch);
+
+int kvm_riscv_vcpu_set_interrupt(struct kvm_vcpu *vcpu, unsigned int irq);
+int kvm_riscv_vcpu_unset_interrupt(struct kvm_vcpu *vcpu, unsigned int irq);
+void kvm_riscv_vcpu_flush_interrupts(struct kvm_vcpu *vcpu);
+void kvm_riscv_vcpu_sync_interrupts(struct kvm_vcpu *vcpu);
+bool kvm_riscv_vcpu_has_interrupts(struct kvm_vcpu *vcpu, unsigned long mask);
+void kvm_riscv_vcpu_power_off(struct kvm_vcpu *vcpu);
+void kvm_riscv_vcpu_power_on(struct kvm_vcpu *vcpu);
+
+int kvm_riscv_vcpu_sbi_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
+int kvm_riscv_vcpu_sbi_ecall(struct kvm_vcpu *vcpu, struct kvm_run *run);
+
+#endif /* __RISCV_KVM_HOST_H__ */

arch/riscv/include/asm/kvm_types.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_RISCV_KVM_TYPES_H
+#define _ASM_RISCV_KVM_TYPES_H
+
+#define KVM_ARCH_NR_OBJS_PER_MEMORY_CACHE 40
+
+#endif /* _ASM_RISCV_KVM_TYPES_H */

arch/riscv/include/asm/kvm_vcpu_fp.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2021 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Atish Patra <[e-mail]>
+ *     Anup Patel <[e-mail]>
+ */
+
+#ifndef __KVM_VCPU_RISCV_FP_H
+#define __KVM_VCPU_RISCV_FP_H
+
+#include <linux/types.h>
+
+struct kvm_cpu_context;
+
+#ifdef CONFIG_FPU
+void __kvm_riscv_fp_f_save(struct kvm_cpu_context *context);
+void __kvm_riscv_fp_f_restore(struct kvm_cpu_context *context);
+void __kvm_riscv_fp_d_save(struct kvm_cpu_context *context);
+void __kvm_riscv_fp_d_restore(struct kvm_cpu_context *context);
+
+void kvm_riscv_vcpu_fp_reset(struct kvm_vcpu *vcpu);
+void kvm_riscv_vcpu_guest_fp_save(struct kvm_cpu_context *cntx,
+				  unsigned long isa);
+void kvm_riscv_vcpu_guest_fp_restore(struct kvm_cpu_context *cntx,
+				     unsigned long isa);
+void kvm_riscv_vcpu_host_fp_save(struct kvm_cpu_context *cntx);
+void kvm_riscv_vcpu_host_fp_restore(struct kvm_cpu_context *cntx);
+#else
+static inline void kvm_riscv_vcpu_fp_reset(struct kvm_vcpu *vcpu)
+{
+}
+static inline void kvm_riscv_vcpu_guest_fp_save(struct kvm_cpu_context *cntx,
+						unsigned long isa)
+{
+}
+static inline void kvm_riscv_vcpu_guest_fp_restore(
+					struct kvm_cpu_context *cntx,
+					unsigned long isa)
+{
+}
+static inline void kvm_riscv_vcpu_host_fp_save(struct kvm_cpu_context *cntx)
+{
+}
+static inline void kvm_riscv_vcpu_host_fp_restore(
+					struct kvm_cpu_context *cntx)
+{
+}
+#endif
+
+int kvm_riscv_vcpu_get_reg_fp(struct kvm_vcpu *vcpu,
+			      const struct kvm_one_reg *reg,
+			      unsigned long rtype);
+int kvm_riscv_vcpu_set_reg_fp(struct kvm_vcpu *vcpu,
+			      const struct kvm_one_reg *reg,
+			      unsigned long rtype);
+
+#endif

arch/riscv/include/asm/kvm_vcpu_timer.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *	Atish Patra <[e-mail]>
+ */
+
+#ifndef __KVM_VCPU_RISCV_TIMER_H
+#define __KVM_VCPU_RISCV_TIMER_H
+
+#include <linux/hrtimer.h>
+
+struct kvm_guest_timer {
+	/* Mult & Shift values to get nanoseconds from cycles */
+	u32 nsec_mult;
+	u32 nsec_shift;
+	/* Time delta value */
+	u64 time_delta;
+};
+
+struct kvm_vcpu_timer {
+	/* Flag for whether init is done */
+	bool init_done;
+	/* Flag for whether timer event is configured */
+	bool next_set;
+	/* Next timer event cycles */
+	u64 next_cycles;
+	/* Underlying hrtimer instance */
+	struct hrtimer hrt;
+};
+
+int kvm_riscv_vcpu_timer_next_event(struct kvm_vcpu *vcpu, u64 ncycles);
+int kvm_riscv_vcpu_get_reg_timer(struct kvm_vcpu *vcpu,
+				 const struct kvm_one_reg *reg);
+int kvm_riscv_vcpu_set_reg_timer(struct kvm_vcpu *vcpu,
+				 const struct kvm_one_reg *reg);
+int kvm_riscv_vcpu_timer_init(struct kvm_vcpu *vcpu);
+int kvm_riscv_vcpu_timer_deinit(struct kvm_vcpu *vcpu);
+int kvm_riscv_vcpu_timer_reset(struct kvm_vcpu *vcpu);
+void kvm_riscv_vcpu_timer_restore(struct kvm_vcpu *vcpu);
+int kvm_riscv_guest_timer_init(struct kvm *kvm);
+
+#endif

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

···
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#ifndef __LINUX_KVM_RISCV_H
+#define __LINUX_KVM_RISCV_H
+
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+
+#define __KVM_HAVE_READONLY_MEM
+
+#define KVM_COALESCED_MMIO_PAGE_OFFSET 1
+
+#define KVM_INTERRUPT_SET	-1U
+#define KVM_INTERRUPT_UNSET	-2U
+
+/* for KVM_GET_REGS and KVM_SET_REGS */
+struct kvm_regs {
+};
+
+/* for KVM_GET_FPU and KVM_SET_FPU */
+struct kvm_fpu {
+};
+
+/* KVM Debug exit structure */
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
+/* definition of registers in kvm_run */
+struct kvm_sync_regs {
+};
+
+/* for KVM_GET_SREGS and KVM_SET_SREGS */
+struct kvm_sregs {
+};
+
+/* CONFIG registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_config {
+	unsigned long isa;
+};
+
+/* CORE registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_core {
+	struct user_regs_struct regs;
+	unsigned long mode;
+};
+
+/* Possible privilege modes for kvm_riscv_core */
+#define KVM_RISCV_MODE_S	1
+#define KVM_RISCV_MODE_U	0
+
+/* CSR registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_csr {
+	unsigned long sstatus;
+	unsigned long sie;
+	unsigned long stvec;
+	unsigned long sscratch;
+	unsigned long sepc;
+	unsigned long scause;
+	unsigned long stval;
+	unsigned long sip;
+	unsigned long satp;
+	unsigned long scounteren;
+};
+
+/* TIMER registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_timer {
+	__u64 frequency;
+	__u64 time;
+	__u64 compare;
+	__u64 state;
+};
+
+/* Possible states for kvm_riscv_timer */
+#define KVM_RISCV_TIMER_STATE_OFF	0
+#define KVM_RISCV_TIMER_STATE_ON	1
+
+#define KVM_REG_SIZE(id)		\
+	(1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
+
+/* If you need to interpret the index values, here is the key: */
+#define KVM_REG_RISCV_TYPE_MASK		0x00000000FF000000
+#define KVM_REG_RISCV_TYPE_SHIFT	24
+
+/* Config registers are mapped as type 1 */
+#define KVM_REG_RISCV_CONFIG		(0x01 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CONFIG_REG(name)	\
+	(offsetof(struct kvm_riscv_config, name) / sizeof(unsigned long))
+
+/* Core registers are mapped as type 2 */
+#define KVM_REG_RISCV_CORE		(0x02 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CORE_REG(name)	\
+		(offsetof(struct kvm_riscv_core, name) / sizeof(unsigned long))
+
+/* Control and status registers are mapped as type 3 */
+#define KVM_REG_RISCV_CSR		(0x03 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CSR_REG(name)	\
+		(offsetof(struct kvm_riscv_csr, name) / sizeof(unsigned long))
+
+/* Timer registers are mapped as type 4 */
+#define KVM_REG_RISCV_TIMER		(0x04 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_TIMER_REG(name)	\
+		(offsetof(struct kvm_riscv_timer, name) / sizeof(__u64))
+
+/* F extension registers are mapped as type 5 */
+#define KVM_REG_RISCV_FP_F		(0x05 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_FP_F_REG(name)	\
+		(offsetof(struct __riscv_f_ext_state, name) / sizeof(__u32))
+
+/* D extension registers are mapped as type 6 */
+#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))
+
+#endif
+
+#endif /* __LINUX_KVM_RISCV_H */

arch/riscv/kernel/asm-offsets.c

···
 #define GENERATING_ASM_OFFSETS
 
 #include <linux/kbuild.h>
+#include <linux/mm.h>
 #include <linux/sched.h>
+#include <asm/kvm_host.h>
 #include <asm/thread_info.h>
 #include <asm/ptrace.h>
 
···
 	OFFSET(PT_STATUS, pt_regs, status);
 	OFFSET(PT_BADADDR, pt_regs, badaddr);
 	OFFSET(PT_CAUSE, pt_regs, cause);
+
+	OFFSET(KVM_ARCH_GUEST_ZERO, kvm_vcpu_arch, guest_context.zero);
+	OFFSET(KVM_ARCH_GUEST_RA, kvm_vcpu_arch, guest_context.ra);
+	OFFSET(KVM_ARCH_GUEST_SP, kvm_vcpu_arch, guest_context.sp);
+	OFFSET(KVM_ARCH_GUEST_GP, kvm_vcpu_arch, guest_context.gp);
+	OFFSET(KVM_ARCH_GUEST_TP, kvm_vcpu_arch, guest_context.tp);
+	OFFSET(KVM_ARCH_GUEST_T0, kvm_vcpu_arch, guest_context.t0);
+	OFFSET(KVM_ARCH_GUEST_T1, kvm_vcpu_arch, guest_context.t1);
+	OFFSET(KVM_ARCH_GUEST_T2, kvm_vcpu_arch, guest_context.t2);
+	OFFSET(KVM_ARCH_GUEST_S0, kvm_vcpu_arch, guest_context.s0);
+	OFFSET(KVM_ARCH_GUEST_S1, kvm_vcpu_arch, guest_context.s1);
+	OFFSET(KVM_ARCH_GUEST_A0, kvm_vcpu_arch, guest_context.a0);
+	OFFSET(KVM_ARCH_GUEST_A1, kvm_vcpu_arch, guest_context.a1);
+	OFFSET(KVM_ARCH_GUEST_A2, kvm_vcpu_arch, guest_context.a2);
+	OFFSET(KVM_ARCH_GUEST_A3, kvm_vcpu_arch, guest_context.a3);
+	OFFSET(KVM_ARCH_GUEST_A4, kvm_vcpu_arch, guest_context.a4);
+	OFFSET(KVM_ARCH_GUEST_A5, kvm_vcpu_arch, guest_context.a5);
+	OFFSET(KVM_ARCH_GUEST_A6, kvm_vcpu_arch, guest_context.a6);
+	OFFSET(KVM_ARCH_GUEST_A7, kvm_vcpu_arch, guest_context.a7);
+	OFFSET(KVM_ARCH_GUEST_S2, kvm_vcpu_arch, guest_context.s2);
+	OFFSET(KVM_ARCH_GUEST_S3, kvm_vcpu_arch, guest_context.s3);
+	OFFSET(KVM_ARCH_GUEST_S4, kvm_vcpu_arch, guest_context.s4);
+	OFFSET(KVM_ARCH_GUEST_S5, kvm_vcpu_arch, guest_context.s5);
+	OFFSET(KVM_ARCH_GUEST_S6, kvm_vcpu_arch, guest_context.s6);
+	OFFSET(KVM_ARCH_GUEST_S7, kvm_vcpu_arch, guest_context.s7);
+	OFFSET(KVM_ARCH_GUEST_S8, kvm_vcpu_arch, guest_context.s8);
+	OFFSET(KVM_ARCH_GUEST_S9, kvm_vcpu_arch, guest_context.s9);
+	OFFSET(KVM_ARCH_GUEST_S10, kvm_vcpu_arch, guest_context.s10);
+	OFFSET(KVM_ARCH_GUEST_S11, kvm_vcpu_arch, guest_context.s11);
+	OFFSET(KVM_ARCH_GUEST_T3, kvm_vcpu_arch, guest_context.t3);
+	OFFSET(KVM_ARCH_GUEST_T4, kvm_vcpu_arch, guest_context.t4);
+	OFFSET(KVM_ARCH_GUEST_T5, kvm_vcpu_arch, guest_context.t5);
+	OFFSET(KVM_ARCH_GUEST_T6, kvm_vcpu_arch, guest_context.t6);
+	OFFSET(KVM_ARCH_GUEST_SEPC, kvm_vcpu_arch, guest_context.sepc);
+	OFFSET(KVM_ARCH_GUEST_SSTATUS, kvm_vcpu_arch, guest_context.sstatus);
+	OFFSET(KVM_ARCH_GUEST_HSTATUS, kvm_vcpu_arch, guest_context.hstatus);
+	OFFSET(KVM_ARCH_GUEST_SCOUNTEREN, kvm_vcpu_arch, guest_csr.scounteren);
+
+	OFFSET(KVM_ARCH_HOST_ZERO, kvm_vcpu_arch, host_context.zero);
+	OFFSET(KVM_ARCH_HOST_RA, kvm_vcpu_arch, host_context.ra);
+	OFFSET(KVM_ARCH_HOST_SP, kvm_vcpu_arch, host_context.sp);
+	OFFSET(KVM_ARCH_HOST_GP, kvm_vcpu_arch, host_context.gp);
+	OFFSET(KVM_ARCH_HOST_TP, kvm_vcpu_arch, host_context.tp);
+	OFFSET(KVM_ARCH_HOST_T0, kvm_vcpu_arch, host_context.t0);
+	OFFSET(KVM_ARCH_HOST_T1, kvm_vcpu_arch, host_context.t1);
+	OFFSET(KVM_ARCH_HOST_T2, kvm_vcpu_arch, host_context.t2);
+	OFFSET(KVM_ARCH_HOST_S0, kvm_vcpu_arch, host_context.s0);
+	OFFSET(KVM_ARCH_HOST_S1, kvm_vcpu_arch, host_context.s1);
+	OFFSET(KVM_ARCH_HOST_A0, kvm_vcpu_arch, host_context.a0);
+	OFFSET(KVM_ARCH_HOST_A1, kvm_vcpu_arch, host_context.a1);
+	OFFSET(KVM_ARCH_HOST_A2, kvm_vcpu_arch, host_context.a2);
+	OFFSET(KVM_ARCH_HOST_A3, kvm_vcpu_arch, host_context.a3);
+	OFFSET(KVM_ARCH_HOST_A4, kvm_vcpu_arch, host_context.a4);
+	OFFSET(KVM_ARCH_HOST_A5, kvm_vcpu_arch, host_context.a5);
+	OFFSET(KVM_ARCH_HOST_A6, kvm_vcpu_arch, host_context.a6);
+	OFFSET(KVM_ARCH_HOST_A7, kvm_vcpu_arch, host_context.a7);
+	OFFSET(KVM_ARCH_HOST_S2, kvm_vcpu_arch, host_context.s2);
+	OFFSET(KVM_ARCH_HOST_S3, kvm_vcpu_arch, host_context.s3);
+	OFFSET(KVM_ARCH_HOST_S4, kvm_vcpu_arch, host_context.s4);
+	OFFSET(KVM_ARCH_HOST_S5, kvm_vcpu_arch, host_context.s5);
+	OFFSET(KVM_ARCH_HOST_S6, kvm_vcpu_arch, host_context.s6);
+	OFFSET(KVM_ARCH_HOST_S7, kvm_vcpu_arch, host_context.s7);
+	OFFSET(KVM_ARCH_HOST_S8, kvm_vcpu_arch, host_context.s8);
+	OFFSET(KVM_ARCH_HOST_S9, kvm_vcpu_arch, host_context.s9);
+	OFFSET(KVM_ARCH_HOST_S10, kvm_vcpu_arch, host_context.s10);
+	OFFSET(KVM_ARCH_HOST_S11, kvm_vcpu_arch, host_context.s11);
+	OFFSET(KVM_ARCH_HOST_T3, kvm_vcpu_arch, host_context.t3);
+	OFFSET(KVM_ARCH_HOST_T4, kvm_vcpu_arch, host_context.t4);
+	OFFSET(KVM_ARCH_HOST_T5, kvm_vcpu_arch, host_context.t5);
+	OFFSET(KVM_ARCH_HOST_T6, kvm_vcpu_arch, host_context.t6);
+	OFFSET(KVM_ARCH_HOST_SEPC, kvm_vcpu_arch, host_context.sepc);
+	OFFSET(KVM_ARCH_HOST_SSTATUS, kvm_vcpu_arch, host_context.sstatus);
+	OFFSET(KVM_ARCH_HOST_HSTATUS, kvm_vcpu_arch, host_context.hstatus);
+	OFFSET(KVM_ARCH_HOST_SSCRATCH, kvm_vcpu_arch, host_sscratch);
+	OFFSET(KVM_ARCH_HOST_STVEC, kvm_vcpu_arch, host_stvec);
+	OFFSET(KVM_ARCH_HOST_SCOUNTEREN, kvm_vcpu_arch, host_scounteren);
+
+	OFFSET(KVM_ARCH_TRAP_SEPC, kvm_cpu_trap, sepc);
+	OFFSET(KVM_ARCH_TRAP_SCAUSE, kvm_cpu_trap, scause);
+	OFFSET(KVM_ARCH_TRAP_STVAL, kvm_cpu_trap, stval);
+	OFFSET(KVM_ARCH_TRAP_HTVAL, kvm_cpu_trap, htval);
+	OFFSET(KVM_ARCH_TRAP_HTINST, kvm_cpu_trap, htinst);
+
+	/* F extension */
+
+	OFFSET(KVM_ARCH_FP_F_F0, kvm_cpu_context, fp.f.f[0]);
+	OFFSET(KVM_ARCH_FP_F_F1, kvm_cpu_context, fp.f.f[1]);
+	OFFSET(KVM_ARCH_FP_F_F2, kvm_cpu_context, fp.f.f[2]);
+	OFFSET(KVM_ARCH_FP_F_F3, kvm_cpu_context, fp.f.f[3]);
+	OFFSET(KVM_ARCH_FP_F_F4, kvm_cpu_context, fp.f.f[4]);
+	OFFSET(KVM_ARCH_FP_F_F5, kvm_cpu_context, fp.f.f[5]);
+	OFFSET(KVM_ARCH_FP_F_F6, kvm_cpu_context, fp.f.f[6]);
+	OFFSET(KVM_ARCH_FP_F_F7, kvm_cpu_context, fp.f.f[7]);
+	OFFSET(KVM_ARCH_FP_F_F8, kvm_cpu_context, fp.f.f[8]);
+	OFFSET(KVM_ARCH_FP_F_F9, kvm_cpu_context, fp.f.f[9]);
+	OFFSET(KVM_ARCH_FP_F_F10, kvm_cpu_context, fp.f.f[10]);
+	OFFSET(KVM_ARCH_FP_F_F11, kvm_cpu_context, fp.f.f[11]);
+	OFFSET(KVM_ARCH_FP_F_F12, kvm_cpu_context, fp.f.f[12]);
+	OFFSET(KVM_ARCH_FP_F_F13, kvm_cpu_context, fp.f.f[13]);
+	OFFSET(KVM_ARCH_FP_F_F14, kvm_cpu_context, fp.f.f[14]);
+	OFFSET(KVM_ARCH_FP_F_F15, kvm_cpu_context, fp.f.f[15]);
+	OFFSET(KVM_ARCH_FP_F_F16, kvm_cpu_context, fp.f.f[16]);
+	OFFSET(KVM_ARCH_FP_F_F17, kvm_cpu_context, fp.f.f[17]);
+	OFFSET(KVM_ARCH_FP_F_F18, kvm_cpu_context, fp.f.f[18]);
+	OFFSET(KVM_ARCH_FP_F_F19, kvm_cpu_context, fp.f.f[19]);
+	OFFSET(KVM_ARCH_FP_F_F20, kvm_cpu_context, fp.f.f[20]);
+	OFFSET(KVM_ARCH_FP_F_F21, kvm_cpu_context, fp.f.f[21]);
+	OFFSET(KVM_ARCH_FP_F_F22, kvm_cpu_context, fp.f.f[22]);
+	OFFSET(KVM_ARCH_FP_F_F23, kvm_cpu_context, fp.f.f[23]);
+	OFFSET(KVM_ARCH_FP_F_F24, kvm_cpu_context, fp.f.f[24]);
+	OFFSET(KVM_ARCH_FP_F_F25, kvm_cpu_context, fp.f.f[25]);
+	OFFSET(KVM_ARCH_FP_F_F26, kvm_cpu_context, fp.f.f[26]);
+	OFFSET(KVM_ARCH_FP_F_F27, kvm_cpu_context, fp.f.f[27]);
+	OFFSET(KVM_ARCH_FP_F_F28, kvm_cpu_context, fp.f.f[28]);
+	OFFSET(KVM_ARCH_FP_F_F29, kvm_cpu_context, fp.f.f[29]);
+	OFFSET(KVM_ARCH_FP_F_F30, kvm_cpu_context, fp.f.f[30]);
+	OFFSET(KVM_ARCH_FP_F_F31, kvm_cpu_context, fp.f.f[31]);
+	OFFSET(KVM_ARCH_FP_F_FCSR, kvm_cpu_context, fp.f.fcsr);
+
+	/* D extension */
+
+	OFFSET(KVM_ARCH_FP_D_F0, kvm_cpu_context, fp.d.f[0]);
+	OFFSET(KVM_ARCH_FP_D_F1, kvm_cpu_context, fp.d.f[1]);
+	OFFSET(KVM_ARCH_FP_D_F2, kvm_cpu_context, fp.d.f[2]);
+	OFFSET(KVM_ARCH_FP_D_F3, kvm_cpu_context, fp.d.f[3]);
+	OFFSET(KVM_ARCH_FP_D_F4, kvm_cpu_context, fp.d.f[4]);
+	OFFSET(KVM_ARCH_FP_D_F5, kvm_cpu_context, fp.d.f[5]);
+	OFFSET(KVM_ARCH_FP_D_F6, kvm_cpu_context, fp.d.f[6]);
+	OFFSET(KVM_ARCH_FP_D_F7, kvm_cpu_context, fp.d.f[7]);
+	OFFSET(KVM_ARCH_FP_D_F8, kvm_cpu_context, fp.d.f[8]);
+	OFFSET(KVM_ARCH_FP_D_F9, kvm_cpu_context, fp.d.f[9]);
+	OFFSET(KVM_ARCH_FP_D_F10, kvm_cpu_context, fp.d.f[10]);
+	OFFSET(KVM_ARCH_FP_D_F11, kvm_cpu_context, fp.d.f[11]);
+	OFFSET(KVM_ARCH_FP_D_F12, kvm_cpu_context, fp.d.f[12]);
+	OFFSET(KVM_ARCH_FP_D_F13, kvm_cpu_context, fp.d.f[13]);
+	OFFSET(KVM_ARCH_FP_D_F14, kvm_cpu_context, fp.d.f[14]);
+	OFFSET(KVM_ARCH_FP_D_F15, kvm_cpu_context, fp.d.f[15]);
+	OFFSET(KVM_ARCH_FP_D_F16, kvm_cpu_context, fp.d.f[16]);
+	OFFSET(KVM_ARCH_FP_D_F17, kvm_cpu_context, fp.d.f[17]);
+	OFFSET(KVM_ARCH_FP_D_F18, kvm_cpu_context, fp.d.f[18]);
+	OFFSET(KVM_ARCH_FP_D_F19, kvm_cpu_context, fp.d.f[19]);
+	OFFSET(KVM_ARCH_FP_D_F20, kvm_cpu_context, fp.d.f[20]);
+	OFFSET(KVM_ARCH_FP_D_F21, kvm_cpu_context, fp.d.f[21]);
+	OFFSET(KVM_ARCH_FP_D_F22, kvm_cpu_context, fp.d.f[22]);
+	OFFSET(KVM_ARCH_FP_D_F23, kvm_cpu_context, fp.d.f[23]);
+	OFFSET(KVM_ARCH_FP_D_F24, kvm_cpu_context, fp.d.f[24]);
+	OFFSET(KVM_ARCH_FP_D_F25, kvm_cpu_context, fp.d.f[25]);
+	OFFSET(KVM_ARCH_FP_D_F26, kvm_cpu_context, fp.d.f[26]);
+	OFFSET(KVM_ARCH_FP_D_F27, kvm_cpu_context, fp.d.f[27]);
+	OFFSET(KVM_ARCH_FP_D_F28, kvm_cpu_context, fp.d.f[28]);
+	OFFSET(KVM_ARCH_FP_D_F29, kvm_cpu_context, fp.d.f[29]);
+	OFFSET(KVM_ARCH_FP_D_F30, kvm_cpu_context, fp.d.f[30]);
+	OFFSET(KVM_ARCH_FP_D_F31, kvm_cpu_context, fp.d.f[31]);
+	OFFSET(KVM_ARCH_FP_D_FCSR, kvm_cpu_context, fp.d.fcsr);
 
 	/*
 	 * THREAD_{F,X}* might be larger than a S-type offset can handle, but

arch/riscv/kvm/Kconfig

···
+# SPDX-License-Identifier: GPL-2.0
+#
+# KVM configuration
+#
+
+source "virt/kvm/Kconfig"
+
+menuconfig VIRTUALIZATION
+	bool "Virtualization"
+	help
+	  Say Y here to get to see options for using your Linux host to run
+	  other operating systems inside virtual machines (guests).
+	  This option alone does not add any kernel code.
+
+	  If you say N, all options in this submenu will be skipped and
+	  disabled.
+
+if VIRTUALIZATION
+
+config KVM
+	tristate "Kernel-based Virtual Machine (KVM) support (EXPERIMENTAL)"
+	depends on RISCV_SBI && MMU
+	select MMU_NOTIFIER
+	select PREEMPT_NOTIFIERS
+	select KVM_MMIO
+	select KVM_GENERIC_DIRTYLOG_READ_PROTECT
+	select HAVE_KVM_VCPU_ASYNC_IOCTL
+	select HAVE_KVM_EVENTFD
+	select SRCU
+	help
+	  Support hosting virtualized guest machines.
+
+	  If unsure, say N.
+
+endif # VIRTUALIZATION

arch/riscv/kvm/Makefile

···
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for RISC-V KVM support
+#
+
+ccflags-y += -I $(srctree)/$(src)
+
+KVM := ../../../virt/kvm
+
+obj-$(CONFIG_KVM) += kvm.o
+
+kvm-y += $(KVM)/kvm_main.o
+kvm-y += $(KVM)/coalesced_mmio.o
+kvm-y += $(KVM)/binary_stats.o
+kvm-y += $(KVM)/eventfd.o
+kvm-y += main.o
+kvm-y += vm.o
+kvm-y += vmid.o
+kvm-y += tlb.o
+kvm-y += mmu.o
+kvm-y += vcpu.o
+kvm-y += vcpu_exit.o
+kvm-y += vcpu_fp.o
+kvm-y += vcpu_switch.o
+kvm-y += vcpu_sbi.o
+kvm-y += vcpu_timer.o

arch/riscv/kvm/main.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/kvm_host.h>
+#include <asm/csr.h>
+#include <asm/hwcap.h>
+#include <asm/sbi.h>
+
+long kvm_arch_dev_ioctl(struct file *filp,
+			unsigned int ioctl, unsigned long arg)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_check_processor_compat(void *opaque)
+{
+	return 0;
+}
+
+int kvm_arch_hardware_setup(void *opaque)
+{
+	return 0;
+}
+
+int kvm_arch_hardware_enable(void)
+{
+	unsigned long hideleg, hedeleg;
+
+	hedeleg = 0;
+	hedeleg |= (1UL << EXC_INST_MISALIGNED);
+	hedeleg |= (1UL << EXC_BREAKPOINT);
+	hedeleg |= (1UL << EXC_SYSCALL);
+	hedeleg |= (1UL << EXC_INST_PAGE_FAULT);
+	hedeleg |= (1UL << EXC_LOAD_PAGE_FAULT);
+	hedeleg |= (1UL << EXC_STORE_PAGE_FAULT);
+	csr_write(CSR_HEDELEG, hedeleg);
+
+	hideleg = 0;
+	hideleg |= (1UL << IRQ_VS_SOFT);
+	hideleg |= (1UL << IRQ_VS_TIMER);
+	hideleg |= (1UL << IRQ_VS_EXT);
+	csr_write(CSR_HIDELEG, hideleg);
+
+	csr_write(CSR_HCOUNTEREN, -1UL);
+
+	csr_write(CSR_HVIP, 0);
+
+	return 0;
+}
+
+void kvm_arch_hardware_disable(void)
+{
+	csr_write(CSR_HEDELEG, 0);
+	csr_write(CSR_HIDELEG, 0);
+}
+
+int kvm_arch_init(void *opaque)
+{
+	const char *str;
+
+	if (!riscv_isa_extension_available(NULL, h)) {
+		kvm_info("hypervisor extension not available\n");
+		return -ENODEV;
+	}
+
+	if (sbi_spec_is_0_1()) {
+		kvm_info("require SBI v0.2 or higher\n");
+		return -ENODEV;
+	}
+
+	if (sbi_probe_extension(SBI_EXT_RFENCE) <= 0) {
+		kvm_info("require SBI RFENCE extension\n");
+		return -ENODEV;
+	}
+
+	kvm_riscv_stage2_mode_detect();
+
+	kvm_riscv_stage2_vmid_detect();
+
+	kvm_info("hypervisor extension available\n");
+
+	switch (kvm_riscv_stage2_mode()) {
+	case HGATP_MODE_SV32X4:
+		str = "Sv32x4";
+		break;
+	case HGATP_MODE_SV39X4:
+		str = "Sv39x4";
+		break;
+	case HGATP_MODE_SV48X4:
+		str = "Sv48x4";
+		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());
+
+	return 0;
+}
+
+void kvm_arch_exit(void)
+{
+}
+
+static int riscv_kvm_init(void)
+{
+	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
+}
+module_init(riscv_kvm_init);

arch/riscv/kvm/mmu.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/hugetlb.h>
+#include <linux/module.h>
+#include <linux/uaccess.h>
+#include <linux/vmalloc.h>
+#include <linux/kvm_host.h>
+#include <linux/sched/signal.h>
+#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
+#else
+static unsigned long stage2_mode = (HGATP_MODE_SV32X4 << HGATP_MODE_SHIFT);
+static unsigned long stage2_pgd_levels = 2;
+#define stage2_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 stage2_pte_leaf(__ptep)	\
+	(pte_val(*(__ptep)) & (_PAGE_READ | _PAGE_WRITE | _PAGE_EXEC))
+
+static inline unsigned long stage2_pte_index(gpa_t addr, u32 level)
+{
+	unsigned long mask;
+	unsigned long shift = HGATP_PAGE_SHIFT + (stage2_index_bits * level);
+
+	if (level == (stage2_pgd_levels - 1))
+		mask = (PTRS_PER_PTE * (1UL << stage2_pgd_xbits)) - 1;
+	else
+		mask = PTRS_PER_PTE - 1;
+
+	return (addr >> shift) & mask;
+}
+
+static inline unsigned long stage2_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)
+{
+	u32 i;
+	unsigned long psz = 1UL << 12;
+
+	for (i = 0; i < stage2_pgd_levels; i++) {
+		if (page_size == (psz << (i * stage2_index_bits))) {
+			*out_level = i;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+static int stage2_level_to_page_size(u32 level, unsigned long *out_pgsize)
+{
+	if (stage2_pgd_levels < level)
+		return -EINVAL;
+
+	*out_pgsize = 1UL << (12 + (level * stage2_index_bits));
+
+	return 0;
+}
+
+static int stage2_cache_topup(struct kvm_mmu_page_cache *pcache,
+			      int min, int max)
+{
+	void *page;
+
+	BUG_ON(max > KVM_MMU_PAGE_CACHE_NR_OBJS);
+	if (pcache->nobjs >= min)
+		return 0;
+	while (pcache->nobjs < max) {
+		page = (void *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+		if (!page)
+			return -ENOMEM;
+		pcache->objects[pcache->nobjs++] = page;
+	}
+
+	return 0;
+}
+
+static void stage2_cache_flush(struct kvm_mmu_page_cache *pcache)
+{
+	while (pcache && pcache->nobjs)
+		free_page((unsigned long)pcache->objects[--pcache->nobjs]);
+}
+
+static void *stage2_cache_alloc(struct kvm_mmu_page_cache *pcache)
+{
+	void *p;
+
+	if (!pcache)
+		return NULL;
+
+	BUG_ON(!pcache->nobjs);
+	p = pcache->objects[--pcache->nobjs];
+
+	return p;
+}
+
+static bool stage2_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;
+
+	*ptep_level = current_level;
+	ptep = (pte_t *)kvm->arch.pgd;
+	ptep = &ptep[stage2_pte_index(addr, current_level)];
+	while (ptep && pte_val(*ptep)) {
+		if (stage2_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)];
+		} else {
+			ptep = NULL;
+		}
+	}
+
+	return false;
+}
+
+static void stage2_remote_tlb_flush(struct kvm *kvm, u32 level, gpa_t addr)
+{
+	struct cpumask hmask;
+	unsigned long size = PAGE_SIZE;
+	struct kvm_vmid *vmid = &kvm->arch.vmid;
+
+	if (stage2_level_to_page_size(level, &size))
+		return;
+	addr &= ~(size - 1);
+
+	/*
+	 * TODO: Instead of cpu_online_mask, we should only target CPUs
+	 * where the Guest/VM is running.
+	 */
+	preempt_disable();
+	riscv_cpuid_to_hartid_mask(cpu_online_mask, &hmask);
+	sbi_remote_hfence_gvma_vmid(cpumask_bits(&hmask), addr, size,
+				    READ_ONCE(vmid->vmid));
+	preempt_enable();
+}
+
+static int stage2_set_pte(struct kvm *kvm, u32 level,
+			   struct kvm_mmu_page_cache *pcache,
+			   gpa_t addr, const pte_t *new_pte)
+{
+	u32 current_level = stage2_pgd_levels - 1;
+	pte_t *next_ptep = (pte_t *)kvm->arch.pgd;
+	pte_t *ptep = &next_ptep[stage2_pte_index(addr, current_level)];
+
+	if (current_level < level)
+		return -EINVAL;
+
+	while (current_level != level) {
+		if (stage2_pte_leaf(ptep))
+			return -EEXIST;
+
+		if (!pte_val(*ptep)) {
+			next_ptep = stage2_cache_alloc(pcache);
+			if (!next_ptep)
+				return -ENOMEM;
+			*ptep = pfn_pte(PFN_DOWN(__pa(next_ptep)),
+					__pgprot(_PAGE_TABLE));
+		} else {
+			if (stage2_pte_leaf(ptep))
+				return -EEXIST;
+			next_ptep = (pte_t *)stage2_pte_page_vaddr(*ptep);
+		}
+
+		current_level--;
+		ptep = &next_ptep[stage2_pte_index(addr, current_level)];
+	}
+
+	*ptep = *new_pte;
+	if (stage2_pte_leaf(ptep))
+		stage2_remote_tlb_flush(kvm, current_level, addr);
+
+	return 0;
+}
+
+static int stage2_map_page(struct kvm *kvm,
+			   struct kvm_mmu_page_cache *pcache,
+			   gpa_t gpa, phys_addr_t hpa,
+			   unsigned long page_size,
+			   bool page_rdonly, bool page_exec)
+{
+	int ret;
+	u32 level = 0;
+	pte_t new_pte;
+	pgprot_t prot;
+
+	ret = stage2_page_size_to_level(page_size, &level);
+	if (ret)
+		return ret;
+
+	/*
+	 * A RISC-V implementation can choose to either:
+	 * 1) Update 'A' and 'D' PTE bits in hardware
+	 * 2) Generate page fault when 'A' and/or 'D' bits are not set
+	 *    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
+	 * mapping. To support KVM dirty page logging with both options
+	 * mentioned above, we will write-protect stage2 PTEs to track
+	 * dirty pages.
+	 */
+
+	if (page_exec) {
+		if (page_rdonly)
+			prot = PAGE_READ_EXEC;
+		else
+			prot = PAGE_WRITE_EXEC;
+	} else {
+		if (page_rdonly)
+			prot = PAGE_READ;
+		else
+			prot = PAGE_WRITE;
+	}
+	new_pte = pfn_pte(PFN_DOWN(hpa), prot);
+	new_pte = pte_mkdirty(new_pte);
+
+	return stage2_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 */
+};
+
+static void stage2_op_pte(struct kvm *kvm, gpa_t addr,
+			  pte_t *ptep, u32 ptep_level, enum stage2_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);
+	if (ret)
+		return;
+
+	BUG_ON(addr & (page_size - 1));
+
+	if (!pte_val(*ptep))
+		return;
+
+	if (ptep_level && !stage2_pte_leaf(ptep)) {
+		next_ptep = (pte_t *)stage2_pte_page_vaddr(*ptep);
+		next_ptep_level = ptep_level - 1;
+		ret = stage2_level_to_page_size(next_ptep_level,
+						&next_page_size);
+		if (ret)
+			return;
+
+		if (op == STAGE2_OP_CLEAR)
+			set_pte(ptep, __pte(0));
+		for (i = 0; i < PTRS_PER_PTE; i++)
+			stage2_op_pte(kvm, addr + i * next_page_size,
+					&next_ptep[i], next_ptep_level, op);
+		if (op == STAGE2_OP_CLEAR)
+			put_page(virt_to_page(next_ptep));
+	} else {
+		if (op == STAGE2_OP_CLEAR)
+			set_pte(ptep, __pte(0));
+		else if (op == STAGE2_OP_WP)
+			set_pte(ptep, __pte(pte_val(*ptep) & ~_PAGE_WRITE));
+		stage2_remote_tlb_flush(kvm, ptep_level, addr);
+	}
+}
+
+static void stage2_unmap_range(struct kvm *kvm, gpa_t start,
+			       gpa_t size, bool may_block)
+{
+	int ret;
+	pte_t *ptep;
+	u32 ptep_level;
+	bool found_leaf;
+	unsigned long page_size;
+	gpa_t addr = start, end = start + size;
+
+	while (addr < end) {
+		found_leaf = stage2_get_leaf_entry(kvm, addr,
+						   &ptep, &ptep_level);
+		ret = stage2_level_to_page_size(ptep_level, &page_size);
+		if (ret)
+			break;
+
+		if (!found_leaf)
+			goto next;
+
+		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
+			stage2_op_pte(kvm, addr, ptep,
+				      ptep_level, STAGE2_OP_CLEAR);
+
+next:
+		addr += page_size;
+
+		/*
+		 * If the range is too large, release the kvm->mmu_lock
+		 * to prevent starvation and lockup detector warnings.
+		 */
+		if (may_block && addr < end)
+			cond_resched_lock(&kvm->mmu_lock);
+	}
+}
+
+static void stage2_wp_range(struct kvm *kvm, gpa_t start, gpa_t end)
+{
+	int ret;
+	pte_t *ptep;
+	u32 ptep_level;
+	bool found_leaf;
+	gpa_t addr = start;
+	unsigned long page_size;
+
+	while (addr < end) {
+		found_leaf = stage2_get_leaf_entry(kvm, addr,
+						   &ptep, &ptep_level);
+		ret = stage2_level_to_page_size(ptep_level, &page_size);
+		if (ret)
+			break;
+
+		if (!found_leaf)
+			goto next;
+
+		if (!(addr & (page_size - 1)) && ((end - addr) >= page_size))
+			stage2_op_pte(kvm, addr, ptep,
+				      ptep_level, STAGE2_OP_WP);
+
+next:
+		addr += page_size;
+	}
+}
+
+static void stage2_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 start = memslot->base_gfn << PAGE_SHIFT;
+	phys_addr_t end = (memslot->base_gfn + memslot->npages) << PAGE_SHIFT;
+
+	spin_lock(&kvm->mmu_lock);
+	stage2_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,
+			  unsigned long size, bool writable)
+{
+	pte_t pte;
+	int ret = 0;
+	unsigned long pfn;
+	phys_addr_t addr, end;
+	struct kvm_mmu_page_cache pcache = { 0, };
+
+	end = (gpa + size + PAGE_SIZE - 1) & PAGE_MASK;
+	pfn = __phys_to_pfn(hpa);
+
+	for (addr = gpa; addr < end; addr += PAGE_SIZE) {
+		pte = pfn_pte(pfn, PAGE_KERNEL);
+
+		if (!writable)
+			pte = pte_wrprotect(pte);
+
+		ret = stage2_cache_topup(&pcache,
+					 stage2_pgd_levels,
+					 KVM_MMU_PAGE_CACHE_NR_OBJS);
+		if (ret)
+			goto out;
+
+		spin_lock(&kvm->mmu_lock);
+		ret = stage2_set_pte(kvm, 0, &pcache, addr, &pte);
+		spin_unlock(&kvm->mmu_lock);
+		if (ret)
+			goto out;
+
+		pfn++;
+	}
+
+out:
+	stage2_cache_flush(&pcache);
+	return ret;
+}
+
+void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm,
+					     struct kvm_memory_slot *slot,
+					     gfn_t gfn_offset,
+					     unsigned long mask)
+{
+	phys_addr_t base_gfn = slot->base_gfn + gfn_offset;
+	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);
+}
+
+void kvm_arch_sync_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot)
+{
+}
+
+void kvm_arch_flush_remote_tlbs_memslot(struct kvm *kvm,
+					const struct kvm_memory_slot *memslot)
+{
+	kvm_flush_remote_tlbs(kvm);
+}
+
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free)
+{
+}
+
+void kvm_arch_memslots_updated(struct kvm *kvm, u64 gen)
+{
+}
+
+void kvm_arch_flush_shadow_all(struct kvm *kvm)
+{
+	kvm_riscv_stage2_free_pgd(kvm);
+}
+
+void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
+				   struct kvm_memory_slot *slot)
+{
+}
+
+void kvm_arch_commit_memory_region(struct kvm *kvm,
+				const struct kvm_userspace_memory_region *mem,
+				struct kvm_memory_slot *old,
+				const struct kvm_memory_slot *new,
+				enum kvm_mr_change change)
+{
+	/*
+	 * At this point memslot has been committed and there is an
+	 * allocated dirty_bitmap[], dirty pages will be tracked while
+	 * the memory slot is write protected.
+	 */
+	if (change != KVM_MR_DELETE && mem->flags & KVM_MEM_LOG_DIRTY_PAGES)
+		stage2_wp_memory_region(kvm, mem->slot);
+}
+
+int kvm_arch_prepare_memory_region(struct kvm *kvm,
+				struct kvm_memory_slot *memslot,
+				const struct kvm_userspace_memory_region *mem,
+				enum kvm_mr_change change)
+{
+	hva_t hva = mem->userspace_addr;
+	hva_t reg_end = hva + mem->memory_size;
+	bool writable = !(mem->flags & KVM_MEM_READONLY);
+	int ret = 0;
+
+	if (change != KVM_MR_CREATE && change != KVM_MR_MOVE &&
+			change != KVM_MR_FLAGS_ONLY)
+		return 0;
+
+	/*
+	 * Prevent userspace from creating a memory region outside of the GPA
+	 * space addressable by the KVM guest GPA space.
+	 */
+	if ((memslot->base_gfn + memslot->npages) >=
+	    (stage2_gpa_size >> PAGE_SHIFT))
+		return -EFAULT;
+
+	mmap_read_lock(current->mm);
+
+	/*
+	 * A memory region could potentially cover multiple VMAs, and
+	 * any holes between them, so iterate over all of them to find
+	 * out if we can map any of them right now.
+	 *
+	 *     +--------------------------------------------+
+	 * +---------------+----------------+   +----------------+
+	 * |   : VMA 1     |      VMA 2     |   |    VMA 3  :    |
+	 * +---------------+----------------+   +----------------+
+	 *     |               memory region                |
+	 *     +--------------------------------------------+
+	 */
+	do {
+		struct vm_area_struct *vma = find_vma(current->mm, hva);
+		hva_t vm_start, vm_end;
+
+		if (!vma || vma->vm_start >= reg_end)
+			break;
+
+		/*
+		 * Mapping a read-only VMA is only allowed if the
+		 * memory region is configured as read-only.
+		 */
+		if (writable && !(vma->vm_flags & VM_WRITE)) {
+			ret = -EPERM;
+			break;
+		}
+
+		/* Take the intersection of this VMA with the memory region */
+		vm_start = max(hva, vma->vm_start);
+		vm_end = min(reg_end, vma->vm_end);
+
+		if (vma->vm_flags & VM_PFNMAP) {
+			gpa_t gpa = mem->guest_phys_addr +
+				    (vm_start - mem->userspace_addr);
+			phys_addr_t pa;
+
+			pa = (phys_addr_t)vma->vm_pgoff << PAGE_SHIFT;
+			pa += vm_start - vma->vm_start;
+
+			/* IO region dirty page logging not allowed */
+			if (memslot->flags & KVM_MEM_LOG_DIRTY_PAGES) {
+				ret = -EINVAL;
+				goto out;
+			}
+
+			ret = stage2_ioremap(kvm, gpa, pa,
+					     vm_end - vm_start, writable);
+			if (ret)
+				break;
+		}
+		hva = vm_end;
+	} while (hva < reg_end);
+
+	if (change == KVM_MR_FLAGS_ONLY)
+		goto out;
+
+	spin_lock(&kvm->mmu_lock);
+	if (ret)
+		stage2_unmap_range(kvm, mem->guest_phys_addr,
+				   mem->memory_size, false);
+	spin_unlock(&kvm->mmu_lock);
+
+out:
+	mmap_read_unlock(current->mm);
+	return ret;
+}
+
+bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range)
+{
+	if (!kvm->arch.pgd)
+		return false;
+
+	stage2_unmap_range(kvm, range->start << PAGE_SHIFT,
+			   (range->end - range->start) << PAGE_SHIFT,
+			   range->may_block);
+	return false;
+}
+
+bool kvm_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range)
+{
+	int ret;
+	kvm_pfn_t pfn = pte_pfn(range->pte);
+
+	if (!kvm->arch.pgd)
+		return false;
+
+	WARN_ON(range->end - range->start != 1);
+
+	ret = stage2_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);
+		return true;
+	}
+
+	return false;
+}
+
+bool kvm_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range)
+{
+	pte_t *ptep;
+	u32 ptep_level = 0;
+	u64 size = (range->end - range->start) << PAGE_SHIFT;
+
+	if (!kvm->arch.pgd)
+		return false;
+
+	WARN_ON(size != PAGE_SIZE && size != PMD_SIZE && size != PGDIR_SIZE);
+
+	if (!stage2_get_leaf_entry(kvm, range->start << PAGE_SHIFT,
+				   &ptep, &ptep_level))
+		return false;
+
+	return ptep_test_and_clear_young(NULL, 0, ptep);
+}
+
+bool kvm_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range)
+{
+	pte_t *ptep;
+	u32 ptep_level = 0;
+	u64 size = (range->end - range->start) << PAGE_SHIFT;
+
+	if (!kvm->arch.pgd)
+		return false;
+
+	WARN_ON(size != PAGE_SIZE && size != PMD_SIZE && size != PGDIR_SIZE);
+
+	if (!stage2_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,
+			 struct kvm_memory_slot *memslot,
+			 gpa_t gpa, unsigned long hva, bool is_write)
+{
+	int ret;
+	kvm_pfn_t hfn;
+	bool writeable;
+	short vma_pageshift;
+	gfn_t gfn = gpa >> PAGE_SHIFT;
+	struct vm_area_struct *vma;
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_mmu_page_cache *pcache = &vcpu->arch.mmu_page_cache;
+	bool logging = (memslot->dirty_bitmap &&
+			!(memslot->flags & KVM_MEM_READONLY)) ? true : false;
+	unsigned long vma_pagesize, mmu_seq;
+
+	mmap_read_lock(current->mm);
+
+	vma = find_vma_intersection(current->mm, hva, hva + 1);
+	if (unlikely(!vma)) {
+		kvm_err("Failed to find VMA for hva 0x%lx\n", hva);
+		mmap_read_unlock(current->mm);
+		return -EFAULT;
+	}
+
+	if (is_vm_hugetlb_page(vma))
+		vma_pageshift = huge_page_shift(hstate_vma(vma));
+	else
+		vma_pageshift = PAGE_SHIFT;
+	vma_pagesize = 1ULL << vma_pageshift;
+	if (logging || (vma->vm_flags & VM_PFNMAP))
+		vma_pagesize = PAGE_SIZE;
+
+	if (vma_pagesize == PMD_SIZE || vma_pagesize == PGDIR_SIZE)
+		gfn = (gpa & huge_page_mask(hstate_vma(vma))) >> PAGE_SHIFT;
+
+	mmap_read_unlock(current->mm);
+
+	if (vma_pagesize != PGDIR_SIZE &&
+	    vma_pagesize != PMD_SIZE &&
+	    vma_pagesize != PAGE_SIZE) {
+		kvm_err("Invalid VMA page size 0x%lx\n", vma_pagesize);
+		return -EFAULT;
+	}
+
+	/* We need minimum second+third level pages */
+	ret = stage2_cache_topup(pcache, stage2_pgd_levels,
+				 KVM_MMU_PAGE_CACHE_NR_OBJS);
+	if (ret) {
+		kvm_err("Failed to topup stage2 cache\n");
+		return ret;
+	}
+
+	mmu_seq = kvm->mmu_notifier_seq;
+
+	hfn = gfn_to_pfn_prot(kvm, gfn, is_write, &writeable);
+	if (hfn == KVM_PFN_ERR_HWPOISON) {
+		send_sig_mceerr(BUS_MCEERR_AR, (void __user *)hva,
+				vma_pageshift, current);
+		return 0;
+	}
+	if (is_error_noslot_pfn(hfn))
+		return -EFAULT;
+
+	/*
+	 * If logging is active then we allow writable pages only
+	 * for write faults.
+	 */
+	if (logging && !is_write)
+		writeable = false;
+
+	spin_lock(&kvm->mmu_lock);
+
+	if (mmu_notifier_retry(kvm, mmu_seq))
+		goto out_unlock;
+
+	if (writeable) {
+		kvm_set_pfn_dirty(hfn);
+		mark_page_dirty(kvm, gfn);
+		ret = stage2_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
+				      vma_pagesize, false, true);
+	} else {
+		ret = stage2_map_page(kvm, pcache, gpa, hfn << PAGE_SHIFT,
+				      vma_pagesize, true, true);
+	}
+
+	if (ret)
+		kvm_err("Failed to map in stage2\n");
+
+out_unlock:
+	spin_unlock(&kvm->mmu_lock);
+	kvm_set_pfn_accessed(hfn);
+	kvm_release_pfn_clean(hfn);
+	return ret;
+}
+
+void kvm_riscv_stage2_flush_cache(struct kvm_vcpu *vcpu)
+{
+	stage2_cache_flush(&vcpu->arch.mmu_page_cache);
+}
+
+int kvm_riscv_stage2_alloc_pgd(struct kvm *kvm)
+{
+	struct page *pgd_page;
+
+	if (kvm->arch.pgd != NULL) {
+		kvm_err("kvm_arch already initialized?\n");
+		return -EINVAL;
+	}
+
+	pgd_page = alloc_pages(GFP_KERNEL | __GFP_ZERO,
+				get_order(stage2_pgd_size));
+	if (!pgd_page)
+		return -ENOMEM;
+	kvm->arch.pgd = page_to_virt(pgd_page);
+	kvm->arch.pgd_phys = page_to_phys(pgd_page);
+
+	return 0;
+}
+
+void kvm_riscv_stage2_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);
+		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));
+}
+
+void kvm_riscv_stage2_update_hgatp(struct kvm_vcpu *vcpu)
+{
+	unsigned long hgatp = stage2_mode;
+	struct kvm_arch *k = &vcpu->kvm->arch;
+
+	hgatp |= (READ_ONCE(k->vmid.vmid) << HGATP_VMID_SHIFT) &
+		 HGATP_VMID_MASK;
+	hgatp |= (k->pgd_phys >> PAGE_SHIFT) & HGATP_PPN;
+
+	csr_write(CSR_HGATP, hgatp);
+
+	if (!kvm_riscv_stage2_vmid_bits())
+		__kvm_riscv_hfence_gvma_all();
+}
+
+void kvm_riscv_stage2_mode_detect(void)
+{
+#ifdef CONFIG_64BIT
+	/* Try Sv48x4 stage2 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;
+	}
+	csr_write(CSR_HGATP, 0);
+
+	__kvm_riscv_hfence_gvma_all();
+#endif
+}
+
+unsigned long kvm_riscv_stage2_mode(void)
+{
+	return stage2_mode >> HGATP_MODE_SHIFT;
+}

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/vcpu.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kdebug.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <linux/uaccess.h>
+#include <linux/vmalloc.h>
+#include <linux/sched/signal.h>
+#include <linux/fs.h>
+#include <linux/kvm_host.h>
+#include <asm/csr.h>
+#include <asm/hwcap.h>
+
+const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = {
+	KVM_GENERIC_VCPU_STATS(),
+	STATS_DESC_COUNTER(VCPU, ecall_exit_stat),
+	STATS_DESC_COUNTER(VCPU, wfi_exit_stat),
+	STATS_DESC_COUNTER(VCPU, mmio_exit_user),
+	STATS_DESC_COUNTER(VCPU, mmio_exit_kernel),
+	STATS_DESC_COUNTER(VCPU, exits)
+};
+
+const struct kvm_stats_header kvm_vcpu_stats_header = {
+	.name_size = KVM_STATS_NAME_SIZE,
+	.num_desc = ARRAY_SIZE(kvm_vcpu_stats_desc),
+	.id_offset = sizeof(struct kvm_stats_header),
+	.desc_offset = sizeof(struct kvm_stats_header) + KVM_STATS_NAME_SIZE,
+	.data_offset = sizeof(struct kvm_stats_header) + KVM_STATS_NAME_SIZE +
+		       sizeof(kvm_vcpu_stats_desc),
+};
+
+#define KVM_RISCV_ISA_ALLOWED	(riscv_isa_extension_mask(a) | \
+				 riscv_isa_extension_mask(c) | \
+				 riscv_isa_extension_mask(d) | \
+				 riscv_isa_extension_mask(f) | \
+				 riscv_isa_extension_mask(i) | \
+				 riscv_isa_extension_mask(m) | \
+				 riscv_isa_extension_mask(s) | \
+				 riscv_isa_extension_mask(u))
+
+static void kvm_riscv_reset_vcpu(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+	struct kvm_vcpu_csr *reset_csr = &vcpu->arch.guest_reset_csr;
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+	struct kvm_cpu_context *reset_cntx = &vcpu->arch.guest_reset_context;
+
+	memcpy(csr, reset_csr, sizeof(*csr));
+
+	memcpy(cntx, reset_cntx, sizeof(*cntx));
+
+	kvm_riscv_vcpu_fp_reset(vcpu);
+
+	kvm_riscv_vcpu_timer_reset(vcpu);
+
+	WRITE_ONCE(vcpu->arch.irqs_pending, 0);
+	WRITE_ONCE(vcpu->arch.irqs_pending_mask, 0);
+}
+
+int kvm_arch_vcpu_precreate(struct kvm *kvm, unsigned int id)
+{
+	return 0;
+}
+
+int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
+{
+	struct kvm_cpu_context *cntx;
+
+	/* Mark this VCPU never ran */
+	vcpu->arch.ran_atleast_once = false;
+
+	/* Setup ISA features available to VCPU */
+	vcpu->arch.isa = riscv_isa_extension_base(NULL) & KVM_RISCV_ISA_ALLOWED;
+
+	/* Setup reset state of shadow SSTATUS and HSTATUS CSRs */
+	cntx = &vcpu->arch.guest_reset_context;
+	cntx->sstatus = SR_SPP | SR_SPIE;
+	cntx->hstatus = 0;
+	cntx->hstatus |= HSTATUS_VTW;
+	cntx->hstatus |= HSTATUS_SPVP;
+	cntx->hstatus |= HSTATUS_SPV;
+
+	/* Setup VCPU timer */
+	kvm_riscv_vcpu_timer_init(vcpu);
+
+	/* Reset VCPU */
+	kvm_riscv_reset_vcpu(vcpu);
+
+	return 0;
+}
+
+void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
+{
+}
+
+void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	/* Cleanup VCPU timer */
+	kvm_riscv_vcpu_timer_deinit(vcpu);
+
+	/* Flush the pages pre-allocated for Stage2 page table mappings */
+	kvm_riscv_stage2_flush_cache(vcpu);
+}
+
+int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
+{
+	return kvm_riscv_vcpu_has_interrupts(vcpu, 1UL << IRQ_VS_TIMER);
+}
+
+void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
+{
+}
+
+void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu)
+{
+}
+
+int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
+{
+	return (kvm_riscv_vcpu_has_interrupts(vcpu, -1UL) &&
+		!vcpu->arch.power_off && !vcpu->arch.pause);
+}
+
+int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
+{
+	return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE;
+}
+
+bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
+{
+	return (vcpu->arch.guest_context.sstatus & SR_SPP) ? true : false;
+}
+
+vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
+{
+	return VM_FAULT_SIGBUS;
+}
+
+static int kvm_riscv_vcpu_get_reg_config(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_CONFIG);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+
+	switch (reg_num) {
+	case KVM_REG_RISCV_CONFIG_REG(isa):
+		reg_val = vcpu->arch.isa;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_set_reg_config(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_CONFIG);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+
+	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	switch (reg_num) {
+	case KVM_REG_RISCV_CONFIG_REG(isa):
+		if (!vcpu->arch.ran_atleast_once) {
+			vcpu->arch.isa = reg_val;
+			vcpu->arch.isa &= riscv_isa_extension_base(NULL);
+			vcpu->arch.isa &= KVM_RISCV_ISA_ALLOWED;
+			kvm_riscv_vcpu_fp_reset(vcpu);
+		} else {
+			return -EOPNOTSUPP;
+		}
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_get_reg_core(struct kvm_vcpu *vcpu,
+				       const struct kvm_one_reg *reg)
+{
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+	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_CORE);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_core) / sizeof(unsigned long))
+		return -EINVAL;
+
+	if (reg_num == KVM_REG_RISCV_CORE_REG(regs.pc))
+		reg_val = cntx->sepc;
+	else if (KVM_REG_RISCV_CORE_REG(regs.pc) < reg_num &&
+		 reg_num <= KVM_REG_RISCV_CORE_REG(regs.t6))
+		reg_val = ((unsigned long *)cntx)[reg_num];
+	else if (reg_num == KVM_REG_RISCV_CORE_REG(mode))
+		reg_val = (cntx->sstatus & SR_SPP) ?
+				KVM_RISCV_MODE_S : KVM_RISCV_MODE_U;
+	else
+		return -EINVAL;
+
+	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_set_reg_core(struct kvm_vcpu *vcpu,
+				       const struct kvm_one_reg *reg)
+{
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+	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_CORE);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_core) / sizeof(unsigned long))
+		return -EINVAL;
+
+	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	if (reg_num == KVM_REG_RISCV_CORE_REG(regs.pc))
+		cntx->sepc = reg_val;
+	else if (KVM_REG_RISCV_CORE_REG(regs.pc) < reg_num &&
+		 reg_num <= KVM_REG_RISCV_CORE_REG(regs.t6))
+		((unsigned long *)cntx)[reg_num] = reg_val;
+	else if (reg_num == KVM_REG_RISCV_CORE_REG(mode)) {
+		if (reg_val == KVM_RISCV_MODE_S)
+			cntx->sstatus |= SR_SPP;
+		else
+			cntx->sstatus &= ~SR_SPP;
+	} else
+		return -EINVAL;
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_get_reg_csr(struct kvm_vcpu *vcpu,
+				      const struct kvm_one_reg *reg)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+	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_CSR);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_csr) / sizeof(unsigned long))
+		return -EINVAL;
+
+	if (reg_num == KVM_REG_RISCV_CSR_REG(sip)) {
+		kvm_riscv_vcpu_flush_interrupts(vcpu);
+		reg_val = (csr->hvip >> VSIP_TO_HVIP_SHIFT) & VSIP_VALID_MASK;
+	} else
+		reg_val = ((unsigned long *)csr)[reg_num];
+
+	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_set_reg_csr(struct kvm_vcpu *vcpu,
+				      const struct kvm_one_reg *reg)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+	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_CSR);
+	unsigned long reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(unsigned long))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_csr) / sizeof(unsigned long))
+		return -EINVAL;
+
+	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	if (reg_num == KVM_REG_RISCV_CSR_REG(sip)) {
+		reg_val &= VSIP_VALID_MASK;
+		reg_val <<= VSIP_TO_HVIP_SHIFT;
+	}
+
+	((unsigned long *)csr)[reg_num] = reg_val;
+
+	if (reg_num == KVM_REG_RISCV_CSR_REG(sip))
+		WRITE_ONCE(vcpu->arch.irqs_pending_mask, 0);
+
+	return 0;
+}
+
+static int kvm_riscv_vcpu_set_reg(struct kvm_vcpu *vcpu,
+				  const struct kvm_one_reg *reg)
+{
+	if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CONFIG)
+		return kvm_riscv_vcpu_set_reg_config(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CORE)
+		return kvm_riscv_vcpu_set_reg_core(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CSR)
+		return kvm_riscv_vcpu_set_reg_csr(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_TIMER)
+		return kvm_riscv_vcpu_set_reg_timer(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_FP_F)
+		return kvm_riscv_vcpu_set_reg_fp(vcpu, reg,
+						 KVM_REG_RISCV_FP_F);
+	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);
+
+	return -EINVAL;
+}
+
+static int kvm_riscv_vcpu_get_reg(struct kvm_vcpu *vcpu,
+				  const struct kvm_one_reg *reg)
+{
+	if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CONFIG)
+		return kvm_riscv_vcpu_get_reg_config(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CORE)
+		return kvm_riscv_vcpu_get_reg_core(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_CSR)
+		return kvm_riscv_vcpu_get_reg_csr(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_TIMER)
+		return kvm_riscv_vcpu_get_reg_timer(vcpu, reg);
+	else if ((reg->id & KVM_REG_RISCV_TYPE_MASK) == KVM_REG_RISCV_FP_F)
+		return kvm_riscv_vcpu_get_reg_fp(vcpu, reg,
+						 KVM_REG_RISCV_FP_F);
+	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);
+
+	return -EINVAL;
+}
+
+long kvm_arch_vcpu_async_ioctl(struct file *filp,
+			       unsigned int ioctl, unsigned long arg)
+{
+	struct kvm_vcpu *vcpu = filp->private_data;
+	void __user *argp = (void __user *)arg;
+
+	if (ioctl == KVM_INTERRUPT) {
+		struct kvm_interrupt irq;
+
+		if (copy_from_user(&irq, argp, sizeof(irq)))
+			return -EFAULT;
+
+		if (irq.irq == KVM_INTERRUPT_SET)
+			return kvm_riscv_vcpu_set_interrupt(vcpu, IRQ_VS_EXT);
+		else
+			return kvm_riscv_vcpu_unset_interrupt(vcpu, IRQ_VS_EXT);
+	}
+
+	return -ENOIOCTLCMD;
+}
+
+long kvm_arch_vcpu_ioctl(struct file *filp,
+			 unsigned int ioctl, unsigned long arg)
+{
+	struct kvm_vcpu *vcpu = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	long r = -EINVAL;
+
+	switch (ioctl) {
+	case KVM_SET_ONE_REG:
+	case KVM_GET_ONE_REG: {
+		struct kvm_one_reg reg;
+
+		r = -EFAULT;
+		if (copy_from_user(&reg, argp, sizeof(reg)))
+			break;
+
+		if (ioctl == KVM_SET_ONE_REG)
+			r = kvm_riscv_vcpu_set_reg(vcpu, &reg);
+		else
+			r = kvm_riscv_vcpu_get_reg(vcpu, &reg);
+		break;
+	}
+	default:
+		break;
+	}
+
+	return r;
+}
+
+int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
+				  struct kvm_translation *tr)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	return -EINVAL;
+}
+
+int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
+{
+	return -EINVAL;
+}
+
+void kvm_riscv_vcpu_flush_interrupts(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+	unsigned long mask, val;
+
+	if (READ_ONCE(vcpu->arch.irqs_pending_mask)) {
+		mask = xchg_acquire(&vcpu->arch.irqs_pending_mask, 0);
+		val = READ_ONCE(vcpu->arch.irqs_pending) & mask;
+
+		csr->hvip &= ~mask;
+		csr->hvip |= val;
+	}
+}
+
+void kvm_riscv_vcpu_sync_interrupts(struct kvm_vcpu *vcpu)
+{
+	unsigned long hvip;
+	struct kvm_vcpu_arch *v = &vcpu->arch;
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+
+	/* Read current HVIP and VSIE CSRs */
+	csr->vsie = csr_read(CSR_VSIE);
+
+	/* Sync-up HVIP.VSSIP bit changes does by Guest */
+	hvip = csr_read(CSR_HVIP);
+	if ((csr->hvip ^ hvip) & (1UL << IRQ_VS_SOFT)) {
+		if (hvip & (1UL << IRQ_VS_SOFT)) {
+			if (!test_and_set_bit(IRQ_VS_SOFT,
+					      &v->irqs_pending_mask))
+				set_bit(IRQ_VS_SOFT, &v->irqs_pending);
+		} else {
+			if (!test_and_set_bit(IRQ_VS_SOFT,
+					      &v->irqs_pending_mask))
+				clear_bit(IRQ_VS_SOFT, &v->irqs_pending);
+		}
+	}
+}
+
+int kvm_riscv_vcpu_set_interrupt(struct kvm_vcpu *vcpu, unsigned int irq)
+{
+	if (irq != IRQ_VS_SOFT &&
+	    irq != IRQ_VS_TIMER &&
+	    irq != IRQ_VS_EXT)
+		return -EINVAL;
+
+	set_bit(irq, &vcpu->arch.irqs_pending);
+	smp_mb__before_atomic();
+	set_bit(irq, &vcpu->arch.irqs_pending_mask);
+
+	kvm_vcpu_kick(vcpu);
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_unset_interrupt(struct kvm_vcpu *vcpu, unsigned int irq)
+{
+	if (irq != IRQ_VS_SOFT &&
+	    irq != IRQ_VS_TIMER &&
+	    irq != IRQ_VS_EXT)
+		return -EINVAL;
+
+	clear_bit(irq, &vcpu->arch.irqs_pending);
+	smp_mb__before_atomic();
+	set_bit(irq, &vcpu->arch.irqs_pending_mask);
+
+	return 0;
+}
+
+bool kvm_riscv_vcpu_has_interrupts(struct kvm_vcpu *vcpu, unsigned long mask)
+{
+	unsigned long ie = ((vcpu->arch.guest_csr.vsie & VSIP_VALID_MASK)
+			    << VSIP_TO_HVIP_SHIFT) & mask;
+
+	return (READ_ONCE(vcpu->arch.irqs_pending) & ie) ? true : false;
+}
+
+void kvm_riscv_vcpu_power_off(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.power_off = true;
+	kvm_make_request(KVM_REQ_SLEEP, vcpu);
+	kvm_vcpu_kick(vcpu);
+}
+
+void kvm_riscv_vcpu_power_on(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.power_off = false;
+	kvm_vcpu_wake_up(vcpu);
+}
+
+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;
+
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
+				    struct kvm_mp_state *mp_state)
+{
+	int ret = 0;
+
+	switch (mp_state->mp_state) {
+	case KVM_MP_STATE_RUNNABLE:
+		vcpu->arch.power_off = false;
+		break;
+	case KVM_MP_STATE_STOPPED:
+		kvm_riscv_vcpu_power_off(vcpu);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
+{
+	/* TODO; To be implemented later. */
+	return -EINVAL;
+}
+
+void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+
+	csr_write(CSR_VSSTATUS, csr->vsstatus);
+	csr_write(CSR_VSIE, csr->vsie);
+	csr_write(CSR_VSTVEC, csr->vstvec);
+	csr_write(CSR_VSSCRATCH, csr->vsscratch);
+	csr_write(CSR_VSEPC, csr->vsepc);
+	csr_write(CSR_VSCAUSE, csr->vscause);
+	csr_write(CSR_VSTVAL, csr->vstval);
+	csr_write(CSR_HVIP, csr->hvip);
+	csr_write(CSR_VSATP, csr->vsatp);
+
+	kvm_riscv_stage2_update_hgatp(vcpu);
+
+	kvm_riscv_vcpu_timer_restore(vcpu);
+
+	kvm_riscv_vcpu_host_fp_save(&vcpu->arch.host_context);
+	kvm_riscv_vcpu_guest_fp_restore(&vcpu->arch.guest_context,
+					vcpu->arch.isa);
+
+	vcpu->cpu = cpu;
+}
+
+void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+
+	vcpu->cpu = -1;
+
+	kvm_riscv_vcpu_guest_fp_save(&vcpu->arch.guest_context,
+				     vcpu->arch.isa);
+	kvm_riscv_vcpu_host_fp_restore(&vcpu->arch.host_context);
+
+	csr_write(CSR_HGATP, 0);
+
+	csr->vsstatus = csr_read(CSR_VSSTATUS);
+	csr->vsie = csr_read(CSR_VSIE);
+	csr->vstvec = csr_read(CSR_VSTVEC);
+	csr->vsscratch = csr_read(CSR_VSSCRATCH);
+	csr->vsepc = csr_read(CSR_VSEPC);
+	csr->vscause = csr_read(CSR_VSCAUSE);
+	csr->vstval = csr_read(CSR_VSTVAL);
+	csr->hvip = csr_read(CSR_HVIP);
+	csr->vsatp = csr_read(CSR_VSATP);
+}
+
+static void kvm_riscv_check_vcpu_requests(struct kvm_vcpu *vcpu)
+{
+	struct rcuwait *wait = kvm_arch_vcpu_get_wait(vcpu);
+
+	if (kvm_request_pending(vcpu)) {
+		if (kvm_check_request(KVM_REQ_SLEEP, vcpu)) {
+			rcuwait_wait_event(wait,
+				(!vcpu->arch.power_off) && (!vcpu->arch.pause),
+				TASK_INTERRUPTIBLE);
+
+			if (vcpu->arch.power_off || vcpu->arch.pause) {
+				/*
+				 * Awaken to handle a signal, request to
+				 * sleep again later.
+				 */
+				kvm_make_request(KVM_REQ_SLEEP, vcpu);
+			}
+		}
+
+		if (kvm_check_request(KVM_REQ_VCPU_RESET, vcpu))
+			kvm_riscv_reset_vcpu(vcpu);
+
+		if (kvm_check_request(KVM_REQ_UPDATE_HGATP, vcpu))
+			kvm_riscv_stage2_update_hgatp(vcpu);
+
+		if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu))
+			__kvm_riscv_hfence_gvma_all();
+	}
+}
+
+static void kvm_riscv_update_hvip(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
+
+	csr_write(CSR_HVIP, csr->hvip);
+}
+
+int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
+{
+	int ret;
+	struct kvm_cpu_trap trap;
+	struct kvm_run *run = vcpu->run;
+
+	/* Mark this VCPU ran at least once */
+	vcpu->arch.ran_atleast_once = true;
+
+	vcpu->arch.srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+
+	/* Process MMIO value returned from user-space */
+	if (run->exit_reason == KVM_EXIT_MMIO) {
+		ret = kvm_riscv_vcpu_mmio_return(vcpu, vcpu->run);
+		if (ret) {
+			srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+			return ret;
+		}
+	}
+
+	/* Process SBI value returned from user-space */
+	if (run->exit_reason == KVM_EXIT_RISCV_SBI) {
+		ret = kvm_riscv_vcpu_sbi_return(vcpu, vcpu->run);
+		if (ret) {
+			srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+			return ret;
+		}
+	}
+
+	if (run->immediate_exit) {
+		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+		return -EINTR;
+	}
+
+	vcpu_load(vcpu);
+
+	kvm_sigset_activate(vcpu);
+
+	ret = 1;
+	run->exit_reason = KVM_EXIT_UNKNOWN;
+	while (ret > 0) {
+		/* Check conditions before entering the guest */
+		cond_resched();
+
+		kvm_riscv_stage2_vmid_update(vcpu);
+
+		kvm_riscv_check_vcpu_requests(vcpu);
+
+		preempt_disable();
+
+		local_irq_disable();
+
+		/*
+		 * Exit if we have a signal pending so that we can deliver
+		 * the signal to user space.
+		 */
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			run->exit_reason = KVM_EXIT_INTR;
+		}
+
+		/*
+		 * Ensure we set mode to IN_GUEST_MODE after we disable
+		 * interrupts and before the final VCPU requests check.
+		 * See the comment in kvm_vcpu_exiting_guest_mode() and
+		 * Documentation/virtual/kvm/vcpu-requests.rst
+		 */
+		vcpu->mode = IN_GUEST_MODE;
+
+		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+		smp_mb__after_srcu_read_unlock();
+
+		/*
+		 * We might have got VCPU interrupts updated asynchronously
+		 * so update it in HW.
+		 */
+		kvm_riscv_vcpu_flush_interrupts(vcpu);
+
+		/* Update HVIP CSR for current CPU */
+		kvm_riscv_update_hvip(vcpu);
+
+		if (ret <= 0 ||
+		    kvm_riscv_stage2_vmid_ver_changed(&vcpu->kvm->arch.vmid) ||
+		    kvm_request_pending(vcpu)) {
+			vcpu->mode = OUTSIDE_GUEST_MODE;
+			local_irq_enable();
+			preempt_enable();
+			vcpu->arch.srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+			continue;
+		}
+
+		guest_enter_irqoff();
+
+		__kvm_riscv_switch_to(&vcpu->arch);
+
+		vcpu->mode = OUTSIDE_GUEST_MODE;
+		vcpu->stat.exits++;
+
+		/*
+		 * Save SCAUSE, STVAL, HTVAL, and HTINST because we might
+		 * get an interrupt between __kvm_riscv_switch_to() and
+		 * local_irq_enable() which can potentially change CSRs.
+		 */
+		trap.sepc = vcpu->arch.guest_context.sepc;
+		trap.scause = csr_read(CSR_SCAUSE);
+		trap.stval = csr_read(CSR_STVAL);
+		trap.htval = csr_read(CSR_HTVAL);
+		trap.htinst = csr_read(CSR_HTINST);
+
+		/* Syncup interrupts state with HW */
+		kvm_riscv_vcpu_sync_interrupts(vcpu);
+
+		/*
+		 * We may have taken a host interrupt in VS/VU-mode (i.e.
+		 * while executing the guest). This interrupt is still
+		 * pending, as we haven't serviced it yet!
+		 *
+		 * We're now back in HS-mode with interrupts disabled
+		 * so enabling the interrupts now will have the effect
+		 * of taking the interrupt again, in HS-mode this time.
+		 */
+		local_irq_enable();
+
+		/*
+		 * We do local_irq_enable() before calling guest_exit() so
+		 * that if a timer interrupt hits while running the guest
+		 * we account that tick as being spent in the guest. We
+		 * enable preemption after calling guest_exit() so that if
+		 * we get preempted we make sure ticks after that is not
+		 * counted as guest time.
+		 */
+		guest_exit();
+
+		preempt_enable();
+
+		vcpu->arch.srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+
+		ret = kvm_riscv_vcpu_exit(vcpu, run, &trap);
+	}
+
+	kvm_sigset_deactivate(vcpu);
+
+	vcpu_put(vcpu);
+
+	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+
+	return ret;
+}

arch/riscv/kvm/vcpu_exit.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kvm_host.h>
+#include <asm/csr.h>
+
+#define INSN_OPCODE_MASK	0x007c
+#define INSN_OPCODE_SHIFT	2
+#define INSN_OPCODE_SYSTEM	28
+
+#define INSN_MASK_WFI		0xffffffff
+#define INSN_MATCH_WFI		0x10500073
+
+#define INSN_MATCH_LB		0x3
+#define INSN_MASK_LB		0x707f
+#define INSN_MATCH_LH		0x1003
+#define INSN_MASK_LH		0x707f
+#define INSN_MATCH_LW		0x2003
+#define INSN_MASK_LW		0x707f
+#define INSN_MATCH_LD		0x3003
+#define INSN_MASK_LD		0x707f
+#define INSN_MATCH_LBU		0x4003
+#define INSN_MASK_LBU		0x707f
+#define INSN_MATCH_LHU		0x5003
+#define INSN_MASK_LHU		0x707f
+#define INSN_MATCH_LWU		0x6003
+#define INSN_MASK_LWU		0x707f
+#define INSN_MATCH_SB		0x23
+#define INSN_MASK_SB		0x707f
+#define INSN_MATCH_SH		0x1023
+#define INSN_MASK_SH		0x707f
+#define INSN_MATCH_SW		0x2023
+#define INSN_MASK_SW		0x707f
+#define INSN_MATCH_SD		0x3023
+#define INSN_MASK_SD		0x707f
+
+#define INSN_MATCH_C_LD		0x6000
+#define INSN_MASK_C_LD		0xe003
+#define INSN_MATCH_C_SD		0xe000
+#define INSN_MASK_C_SD		0xe003
+#define INSN_MATCH_C_LW		0x4000
+#define INSN_MASK_C_LW		0xe003
+#define INSN_MATCH_C_SW		0xc000
+#define INSN_MASK_C_SW		0xe003
+#define INSN_MATCH_C_LDSP	0x6002
+#define INSN_MASK_C_LDSP	0xe003
+#define INSN_MATCH_C_SDSP	0xe002
+#define INSN_MASK_C_SDSP	0xe003
+#define INSN_MATCH_C_LWSP	0x4002
+#define INSN_MASK_C_LWSP	0xe003
+#define INSN_MATCH_C_SWSP	0xc002
+#define INSN_MASK_C_SWSP	0xe003
+
+#define INSN_16BIT_MASK		0x3
+
+#define INSN_IS_16BIT(insn)	(((insn) & INSN_16BIT_MASK) != INSN_16BIT_MASK)
+
+#define INSN_LEN(insn)		(INSN_IS_16BIT(insn) ? 2 : 4)
+
+#ifdef CONFIG_64BIT
+#define LOG_REGBYTES		3
+#else
+#define LOG_REGBYTES		2
+#endif
+#define REGBYTES		(1 << LOG_REGBYTES)
+
+#define SH_RD			7
+#define SH_RS1			15
+#define SH_RS2			20
+#define SH_RS2C			2
+
+#define RV_X(x, s, n)		(((x) >> (s)) & ((1 << (n)) - 1))
+#define RVC_LW_IMM(x)		((RV_X(x, 6, 1) << 2) | \
+				 (RV_X(x, 10, 3) << 3) | \
+				 (RV_X(x, 5, 1) << 6))
+#define RVC_LD_IMM(x)		((RV_X(x, 10, 3) << 3) | \
+				 (RV_X(x, 5, 2) << 6))
+#define RVC_LWSP_IMM(x)		((RV_X(x, 4, 3) << 2) | \
+				 (RV_X(x, 12, 1) << 5) | \
+				 (RV_X(x, 2, 2) << 6))
+#define RVC_LDSP_IMM(x)		((RV_X(x, 5, 2) << 3) | \
+				 (RV_X(x, 12, 1) << 5) | \
+				 (RV_X(x, 2, 3) << 6))
+#define RVC_SWSP_IMM(x)		((RV_X(x, 9, 4) << 2) | \
+				 (RV_X(x, 7, 2) << 6))
+#define RVC_SDSP_IMM(x)		((RV_X(x, 10, 3) << 3) | \
+				 (RV_X(x, 7, 3) << 6))
+#define RVC_RS1S(insn)		(8 + RV_X(insn, SH_RD, 3))
+#define RVC_RS2S(insn)		(8 + RV_X(insn, SH_RS2C, 3))
+#define RVC_RS2(insn)		RV_X(insn, SH_RS2C, 5)
+
+#define SHIFT_RIGHT(x, y)		\
+	((y) < 0 ? ((x) << -(y)) : ((x) >> (y)))
+
+#define REG_MASK			\
+	((1 << (5 + LOG_REGBYTES)) - (1 << LOG_REGBYTES))
+
+#define REG_OFFSET(insn, pos)		\
+	(SHIFT_RIGHT((insn), (pos) - LOG_REGBYTES) & REG_MASK)
+
+#define REG_PTR(insn, pos, regs)	\
+	((ulong *)((ulong)(regs) + REG_OFFSET(insn, pos)))
+
+#define GET_RM(insn)		(((insn) >> 12) & 7)
+
+#define GET_RS1(insn, regs)	(*REG_PTR(insn, SH_RS1, regs))
+#define GET_RS2(insn, regs)	(*REG_PTR(insn, SH_RS2, regs))
+#define GET_RS1S(insn, regs)	(*REG_PTR(RVC_RS1S(insn), 0, regs))
+#define GET_RS2S(insn, regs)	(*REG_PTR(RVC_RS2S(insn), 0, regs))
+#define GET_RS2C(insn, regs)	(*REG_PTR(insn, SH_RS2C, regs))
+#define GET_SP(regs)		(*REG_PTR(2, 0, regs))
+#define SET_RD(insn, regs, val)	(*REG_PTR(insn, SH_RD, regs) = (val))
+#define IMM_I(insn)		((s32)(insn) >> 20)
+#define IMM_S(insn)		(((s32)(insn) >> 25 << 5) | \
+				 (s32)(((insn) >> 7) & 0x1f))
+#define MASK_FUNCT3		0x7000
+
+static int truly_illegal_insn(struct kvm_vcpu *vcpu,
+			      struct kvm_run *run,
+			      ulong insn)
+{
+	struct kvm_cpu_trap utrap = { 0 };
+
+	/* Redirect trap to Guest VCPU */
+	utrap.sepc = vcpu->arch.guest_context.sepc;
+	utrap.scause = EXC_INST_ILLEGAL;
+	utrap.stval = insn;
+	kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+
+	return 1;
+}
+
+static int system_opcode_insn(struct kvm_vcpu *vcpu,
+			      struct kvm_run *run,
+			      ulong insn)
+{
+	if ((insn & INSN_MASK_WFI) == INSN_MATCH_WFI) {
+		vcpu->stat.wfi_exit_stat++;
+		if (!kvm_arch_vcpu_runnable(vcpu)) {
+			srcu_read_unlock(&vcpu->kvm->srcu, vcpu->arch.srcu_idx);
+			kvm_vcpu_block(vcpu);
+			vcpu->arch.srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+			kvm_clear_request(KVM_REQ_UNHALT, vcpu);
+		}
+		vcpu->arch.guest_context.sepc += INSN_LEN(insn);
+		return 1;
+	}
+
+	return truly_illegal_insn(vcpu, run, insn);
+}
+
+static int virtual_inst_fault(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			      struct kvm_cpu_trap *trap)
+{
+	unsigned long insn = trap->stval;
+	struct kvm_cpu_trap utrap = { 0 };
+	struct kvm_cpu_context *ct;
+
+	if (unlikely(INSN_IS_16BIT(insn))) {
+		if (insn == 0) {
+			ct = &vcpu->arch.guest_context;
+			insn = kvm_riscv_vcpu_unpriv_read(vcpu, true,
+							  ct->sepc,
+							  &utrap);
+			if (utrap.scause) {
+				utrap.sepc = ct->sepc;
+				kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+				return 1;
+			}
+		}
+		if (INSN_IS_16BIT(insn))
+			return truly_illegal_insn(vcpu, run, insn);
+	}
+
+	switch ((insn & INSN_OPCODE_MASK) >> INSN_OPCODE_SHIFT) {
+	case INSN_OPCODE_SYSTEM:
+		return system_opcode_insn(vcpu, run, insn);
+	default:
+		return truly_illegal_insn(vcpu, run, insn);
+	}
+}
+
+static int emulate_load(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			unsigned long fault_addr, unsigned long htinst)
+{
+	u8 data_buf[8];
+	unsigned long insn;
+	int shift = 0, len = 0, insn_len = 0;
+	struct kvm_cpu_trap utrap = { 0 };
+	struct kvm_cpu_context *ct = &vcpu->arch.guest_context;
+
+	/* Determine trapped instruction */
+	if (htinst & 0x1) {
+		/*
+		 * Bit[0] == 1 implies trapped instruction value is
+		 * transformed instruction or custom instruction.
+		 */
+		insn = htinst | INSN_16BIT_MASK;
+		insn_len = (htinst & BIT(1)) ? INSN_LEN(insn) : 2;
+	} else {
+		/*
+		 * Bit[0] == 0 implies trapped instruction value is
+		 * zero or special value.
+		 */
+		insn = kvm_riscv_vcpu_unpriv_read(vcpu, true, ct->sepc,
+						  &utrap);
+		if (utrap.scause) {
+			/* Redirect trap if we failed to read instruction */
+			utrap.sepc = ct->sepc;
+			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+			return 1;
+		}
+		insn_len = INSN_LEN(insn);
+	}
+
+	/* Decode length of MMIO and shift */
+	if ((insn & INSN_MASK_LW) == INSN_MATCH_LW) {
+		len = 4;
+		shift = 8 * (sizeof(ulong) - len);
+	} else if ((insn & INSN_MASK_LB) == INSN_MATCH_LB) {
+		len = 1;
+		shift = 8 * (sizeof(ulong) - len);
+	} else if ((insn & INSN_MASK_LBU) == INSN_MATCH_LBU) {
+		len = 1;
+		shift = 8 * (sizeof(ulong) - len);
+#ifdef CONFIG_64BIT
+	} else if ((insn & INSN_MASK_LD) == INSN_MATCH_LD) {
+		len = 8;
+		shift = 8 * (sizeof(ulong) - len);
+	} else if ((insn & INSN_MASK_LWU) == INSN_MATCH_LWU) {
+		len = 4;
+#endif
+	} else if ((insn & INSN_MASK_LH) == INSN_MATCH_LH) {
+		len = 2;
+		shift = 8 * (sizeof(ulong) - len);
+	} else if ((insn & INSN_MASK_LHU) == INSN_MATCH_LHU) {
+		len = 2;
+#ifdef CONFIG_64BIT
+	} else if ((insn & INSN_MASK_C_LD) == INSN_MATCH_C_LD) {
+		len = 8;
+		shift = 8 * (sizeof(ulong) - len);
+		insn = RVC_RS2S(insn) << SH_RD;
+	} else if ((insn & INSN_MASK_C_LDSP) == INSN_MATCH_C_LDSP &&
+		   ((insn >> SH_RD) & 0x1f)) {
+		len = 8;
+		shift = 8 * (sizeof(ulong) - len);
+#endif
+	} else if ((insn & INSN_MASK_C_LW) == INSN_MATCH_C_LW) {
+		len = 4;
+		shift = 8 * (sizeof(ulong) - len);
+		insn = RVC_RS2S(insn) << SH_RD;
+	} else if ((insn & INSN_MASK_C_LWSP) == INSN_MATCH_C_LWSP &&
+		   ((insn >> SH_RD) & 0x1f)) {
+		len = 4;
+		shift = 8 * (sizeof(ulong) - len);
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	/* Fault address should be aligned to length of MMIO */
+	if (fault_addr & (len - 1))
+		return -EIO;
+
+	/* Save instruction decode info */
+	vcpu->arch.mmio_decode.insn = insn;
+	vcpu->arch.mmio_decode.insn_len = insn_len;
+	vcpu->arch.mmio_decode.shift = shift;
+	vcpu->arch.mmio_decode.len = len;
+	vcpu->arch.mmio_decode.return_handled = 0;
+
+	/* Update MMIO details in kvm_run struct */
+	run->mmio.is_write = false;
+	run->mmio.phys_addr = fault_addr;
+	run->mmio.len = len;
+
+	/* Try to handle MMIO access in the kernel */
+	if (!kvm_io_bus_read(vcpu, KVM_MMIO_BUS, fault_addr, len, data_buf)) {
+		/* Successfully handled MMIO access in the kernel so resume */
+		memcpy(run->mmio.data, data_buf, len);
+		vcpu->stat.mmio_exit_kernel++;
+		kvm_riscv_vcpu_mmio_return(vcpu, run);
+		return 1;
+	}
+
+	/* Exit to userspace for MMIO emulation */
+	vcpu->stat.mmio_exit_user++;
+	run->exit_reason = KVM_EXIT_MMIO;
+
+	return 0;
+}
+
+static int emulate_store(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			 unsigned long fault_addr, unsigned long htinst)
+{
+	u8 data8;
+	u16 data16;
+	u32 data32;
+	u64 data64;
+	ulong data;
+	unsigned long insn;
+	int len = 0, insn_len = 0;
+	struct kvm_cpu_trap utrap = { 0 };
+	struct kvm_cpu_context *ct = &vcpu->arch.guest_context;
+
+	/* Determine trapped instruction */
+	if (htinst & 0x1) {
+		/*
+		 * Bit[0] == 1 implies trapped instruction value is
+		 * transformed instruction or custom instruction.
+		 */
+		insn = htinst | INSN_16BIT_MASK;
+		insn_len = (htinst & BIT(1)) ? INSN_LEN(insn) : 2;
+	} else {
+		/*
+		 * Bit[0] == 0 implies trapped instruction value is
+		 * zero or special value.
+		 */
+		insn = kvm_riscv_vcpu_unpriv_read(vcpu, true, ct->sepc,
+						  &utrap);
+		if (utrap.scause) {
+			/* Redirect trap if we failed to read instruction */
+			utrap.sepc = ct->sepc;
+			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+			return 1;
+		}
+		insn_len = INSN_LEN(insn);
+	}
+
+	data = GET_RS2(insn, &vcpu->arch.guest_context);
+	data8 = data16 = data32 = data64 = data;
+
+	if ((insn & INSN_MASK_SW) == INSN_MATCH_SW) {
+		len = 4;
+	} else if ((insn & INSN_MASK_SB) == INSN_MATCH_SB) {
+		len = 1;
+#ifdef CONFIG_64BIT
+	} else if ((insn & INSN_MASK_SD) == INSN_MATCH_SD) {
+		len = 8;
+#endif
+	} else if ((insn & INSN_MASK_SH) == INSN_MATCH_SH) {
+		len = 2;
+#ifdef CONFIG_64BIT
+	} else if ((insn & INSN_MASK_C_SD) == INSN_MATCH_C_SD) {
+		len = 8;
+		data64 = GET_RS2S(insn, &vcpu->arch.guest_context);
+	} else if ((insn & INSN_MASK_C_SDSP) == INSN_MATCH_C_SDSP &&
+		   ((insn >> SH_RD) & 0x1f)) {
+		len = 8;
+		data64 = GET_RS2C(insn, &vcpu->arch.guest_context);
+#endif
+	} else if ((insn & INSN_MASK_C_SW) == INSN_MATCH_C_SW) {
+		len = 4;
+		data32 = GET_RS2S(insn, &vcpu->arch.guest_context);
+	} else if ((insn & INSN_MASK_C_SWSP) == INSN_MATCH_C_SWSP &&
+		   ((insn >> SH_RD) & 0x1f)) {
+		len = 4;
+		data32 = GET_RS2C(insn, &vcpu->arch.guest_context);
+	} else {
+		return -EOPNOTSUPP;
+	}
+
+	/* Fault address should be aligned to length of MMIO */
+	if (fault_addr & (len - 1))
+		return -EIO;
+
+	/* Save instruction decode info */
+	vcpu->arch.mmio_decode.insn = insn;
+	vcpu->arch.mmio_decode.insn_len = insn_len;
+	vcpu->arch.mmio_decode.shift = 0;
+	vcpu->arch.mmio_decode.len = len;
+	vcpu->arch.mmio_decode.return_handled = 0;
+
+	/* Copy data to kvm_run instance */
+	switch (len) {
+	case 1:
+		*((u8 *)run->mmio.data) = data8;
+		break;
+	case 2:
+		*((u16 *)run->mmio.data) = data16;
+		break;
+	case 4:
+		*((u32 *)run->mmio.data) = data32;
+		break;
+	case 8:
+		*((u64 *)run->mmio.data) = data64;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	/* Update MMIO details in kvm_run struct */
+	run->mmio.is_write = true;
+	run->mmio.phys_addr = fault_addr;
+	run->mmio.len = len;
+
+	/* Try to handle MMIO access in the kernel */
+	if (!kvm_io_bus_write(vcpu, KVM_MMIO_BUS,
+			      fault_addr, len, run->mmio.data)) {
+		/* Successfully handled MMIO access in the kernel so resume */
+		vcpu->stat.mmio_exit_kernel++;
+		kvm_riscv_vcpu_mmio_return(vcpu, run);
+		return 1;
+	}
+
+	/* Exit to userspace for MMIO emulation */
+	vcpu->stat.mmio_exit_user++;
+	run->exit_reason = KVM_EXIT_MMIO;
+
+	return 0;
+}
+
+static int stage2_page_fault(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			     struct kvm_cpu_trap *trap)
+{
+	struct kvm_memory_slot *memslot;
+	unsigned long hva, fault_addr;
+	bool writeable;
+	gfn_t gfn;
+	int ret;
+
+	fault_addr = (trap->htval << 2) | (trap->stval & 0x3);
+	gfn = fault_addr >> PAGE_SHIFT;
+	memslot = gfn_to_memslot(vcpu->kvm, gfn);
+	hva = gfn_to_hva_memslot_prot(memslot, gfn, &writeable);
+
+	if (kvm_is_error_hva(hva) ||
+	    (trap->scause == EXC_STORE_GUEST_PAGE_FAULT && !writeable)) {
+		switch (trap->scause) {
+		case EXC_LOAD_GUEST_PAGE_FAULT:
+			return emulate_load(vcpu, run, fault_addr,
+					    trap->htinst);
+		case EXC_STORE_GUEST_PAGE_FAULT:
+			return emulate_store(vcpu, run, fault_addr,
+					     trap->htinst);
+		default:
+			return -EOPNOTSUPP;
+		};
+	}
+
+	ret = kvm_riscv_stage2_map(vcpu, memslot, fault_addr, hva,
+		(trap->scause == EXC_STORE_GUEST_PAGE_FAULT) ? true : false);
+	if (ret < 0)
+		return ret;
+
+	return 1;
+}
+
+/**
+ * kvm_riscv_vcpu_unpriv_read -- Read machine word from Guest memory
+ *
+ * @vcpu: The VCPU pointer
+ * @read_insn: Flag representing whether we are reading instruction
+ * @guest_addr: Guest address to read
+ * @trap: Output pointer to trap details
+ */
+unsigned long kvm_riscv_vcpu_unpriv_read(struct kvm_vcpu *vcpu,
+					 bool read_insn,
+					 unsigned long guest_addr,
+					 struct kvm_cpu_trap *trap)
+{
+	register unsigned long taddr asm("a0") = (unsigned long)trap;
+	register unsigned long ttmp asm("a1");
+	register unsigned long val asm("t0");
+	register unsigned long tmp asm("t1");
+	register unsigned long addr asm("t2") = guest_addr;
+	unsigned long flags;
+	unsigned long old_stvec, old_hstatus;
+
+	local_irq_save(flags);
+
+	old_hstatus = csr_swap(CSR_HSTATUS, vcpu->arch.guest_context.hstatus);
+	old_stvec = csr_swap(CSR_STVEC, (ulong)&__kvm_riscv_unpriv_trap);
+
+	if (read_insn) {
+		/*
+		 * HLVX.HU instruction
+		 * 0110010 00011 rs1 100 rd 1110011
+		 */
+		asm volatile ("\n"
+			".option push\n"
+			".option norvc\n"
+			"add %[ttmp], %[taddr], 0\n"
+			/*
+			 * HLVX.HU %[val], (%[addr])
+			 * HLVX.HU t0, (t2)
+			 * 0110010 00011 00111 100 00101 1110011
+			 */
+			".word 0x6433c2f3\n"
+			"andi %[tmp], %[val], 3\n"
+			"addi %[tmp], %[tmp], -3\n"
+			"bne %[tmp], zero, 2f\n"
+			"addi %[addr], %[addr], 2\n"
+			/*
+			 * HLVX.HU %[tmp], (%[addr])
+			 * HLVX.HU t1, (t2)
+			 * 0110010 00011 00111 100 00110 1110011
+			 */
+			".word 0x6433c373\n"
+			"sll %[tmp], %[tmp], 16\n"
+			"add %[val], %[val], %[tmp]\n"
+			"2:\n"
+			".option pop"
+		: [val] "=&r" (val), [tmp] "=&r" (tmp),
+		  [taddr] "+&r" (taddr), [ttmp] "+&r" (ttmp),
+		  [addr] "+&r" (addr) : : "memory");
+
+		if (trap->scause == EXC_LOAD_PAGE_FAULT)
+			trap->scause = EXC_INST_PAGE_FAULT;
+	} else {
+		/*
+		 * HLV.D instruction
+		 * 0110110 00000 rs1 100 rd 1110011
+		 *
+		 * HLV.W instruction
+		 * 0110100 00000 rs1 100 rd 1110011
+		 */
+		asm volatile ("\n"
+			".option push\n"
+			".option norvc\n"
+			"add %[ttmp], %[taddr], 0\n"
+#ifdef CONFIG_64BIT
+			/*
+			 * HLV.D %[val], (%[addr])
+			 * HLV.D t0, (t2)
+			 * 0110110 00000 00111 100 00101 1110011
+			 */
+			".word 0x6c03c2f3\n"
+#else
+			/*
+			 * HLV.W %[val], (%[addr])
+			 * HLV.W t0, (t2)
+			 * 0110100 00000 00111 100 00101 1110011
+			 */
+			".word 0x6803c2f3\n"
+#endif
+			".option pop"
+		: [val] "=&r" (val),
+		  [taddr] "+&r" (taddr), [ttmp] "+&r" (ttmp)
+		: [addr] "r" (addr) : "memory");
+	}
+
+	csr_write(CSR_STVEC, old_stvec);
+	csr_write(CSR_HSTATUS, old_hstatus);
+
+	local_irq_restore(flags);
+
+	return val;
+}
+
+/**
+ * kvm_riscv_vcpu_trap_redirect -- Redirect trap to Guest
+ *
+ * @vcpu: The VCPU pointer
+ * @trap: Trap details
+ */
+void kvm_riscv_vcpu_trap_redirect(struct kvm_vcpu *vcpu,
+				  struct kvm_cpu_trap *trap)
+{
+	unsigned long vsstatus = csr_read(CSR_VSSTATUS);
+
+	/* Change Guest SSTATUS.SPP bit */
+	vsstatus &= ~SR_SPP;
+	if (vcpu->arch.guest_context.sstatus & SR_SPP)
+		vsstatus |= SR_SPP;
+
+	/* Change Guest SSTATUS.SPIE bit */
+	vsstatus &= ~SR_SPIE;
+	if (vsstatus & SR_SIE)
+		vsstatus |= SR_SPIE;
+
+	/* Clear Guest SSTATUS.SIE bit */
+	vsstatus &= ~SR_SIE;
+
+	/* Update Guest SSTATUS */
+	csr_write(CSR_VSSTATUS, vsstatus);
+
+	/* Update Guest SCAUSE, STVAL, and SEPC */
+	csr_write(CSR_VSCAUSE, trap->scause);
+	csr_write(CSR_VSTVAL, trap->stval);
+	csr_write(CSR_VSEPC, trap->sepc);
+
+	/* Set Guest PC to Guest exception vector */
+	vcpu->arch.guest_context.sepc = csr_read(CSR_VSTVEC);
+}
+
+/**
+ * kvm_riscv_vcpu_mmio_return -- Handle MMIO loads after user space emulation
+ *			     or in-kernel IO emulation
+ *
+ * @vcpu: The VCPU pointer
+ * @run:  The VCPU run struct containing the mmio data
+ */
+int kvm_riscv_vcpu_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	u8 data8;
+	u16 data16;
+	u32 data32;
+	u64 data64;
+	ulong insn;
+	int len, shift;
+
+	if (vcpu->arch.mmio_decode.return_handled)
+		return 0;
+
+	vcpu->arch.mmio_decode.return_handled = 1;
+	insn = vcpu->arch.mmio_decode.insn;
+
+	if (run->mmio.is_write)
+		goto done;
+
+	len = vcpu->arch.mmio_decode.len;
+	shift = vcpu->arch.mmio_decode.shift;
+
+	switch (len) {
+	case 1:
+		data8 = *((u8 *)run->mmio.data);
+		SET_RD(insn, &vcpu->arch.guest_context,
+			(ulong)data8 << shift >> shift);
+		break;
+	case 2:
+		data16 = *((u16 *)run->mmio.data);
+		SET_RD(insn, &vcpu->arch.guest_context,
+			(ulong)data16 << shift >> shift);
+		break;
+	case 4:
+		data32 = *((u32 *)run->mmio.data);
+		SET_RD(insn, &vcpu->arch.guest_context,
+			(ulong)data32 << shift >> shift);
+		break;
+	case 8:
+		data64 = *((u64 *)run->mmio.data);
+		SET_RD(insn, &vcpu->arch.guest_context,
+			(ulong)data64 << shift >> shift);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+done:
+	/* Move to next instruction */
+	vcpu->arch.guest_context.sepc += vcpu->arch.mmio_decode.insn_len;
+
+	return 0;
+}
+
+/*
+ * Return > 0 to return to guest, < 0 on error, 0 (and set exit_reason) on
+ * proper exit to userspace.
+ */
+int kvm_riscv_vcpu_exit(struct kvm_vcpu *vcpu, struct kvm_run *run,
+			struct kvm_cpu_trap *trap)
+{
+	int ret;
+
+	/* If we got host interrupt then do nothing */
+	if (trap->scause & CAUSE_IRQ_FLAG)
+		return 1;
+
+	/* Handle guest traps */
+	ret = -EFAULT;
+	run->exit_reason = KVM_EXIT_UNKNOWN;
+	switch (trap->scause) {
+	case EXC_VIRTUAL_INST_FAULT:
+		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)
+			ret = virtual_inst_fault(vcpu, run, trap);
+		break;
+	case EXC_INST_GUEST_PAGE_FAULT:
+	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);
+		break;
+	case EXC_SUPERVISOR_SYSCALL:
+		if (vcpu->arch.guest_context.hstatus & HSTATUS_SPV)
+			ret = kvm_riscv_vcpu_sbi_ecall(vcpu, run);
+		break;
+	default:
+		break;
+	}
+
+	/* Print details in-case of error */
+	if (ret < 0) {
+		kvm_err("VCPU exit error %d\n", ret);
+		kvm_err("SEPC=0x%lx SSTATUS=0x%lx HSTATUS=0x%lx\n",
+			vcpu->arch.guest_context.sepc,
+			vcpu->arch.guest_context.sstatus,
+			vcpu->arch.guest_context.hstatus);
+		kvm_err("SCAUSE=0x%lx STVAL=0x%lx HTVAL=0x%lx HTINST=0x%lx\n",
+			trap->scause, trap->stval, trap->htval, trap->htinst);
+	}
+
+	return ret;
+}

arch/riscv/kvm/vcpu_fp.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2021 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Atish Patra <[e-mail]>
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kvm_host.h>
+#include <linux/uaccess.h>
+
+#ifdef CONFIG_FPU
+void kvm_riscv_vcpu_fp_reset(struct kvm_vcpu *vcpu)
+{
+	unsigned long isa = vcpu->arch.isa;
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+
+	cntx->sstatus &= ~SR_FS;
+	if (riscv_isa_extension_available(&isa, f) ||
+	    riscv_isa_extension_available(&isa, d))
+		cntx->sstatus |= SR_FS_INITIAL;
+	else
+		cntx->sstatus |= SR_FS_OFF;
+}
+
+void kvm_riscv_vcpu_fp_clean(struct kvm_cpu_context *cntx)
+{
+	cntx->sstatus &= ~SR_FS;
+	cntx->sstatus |= SR_FS_CLEAN;
+}
+
+void kvm_riscv_vcpu_guest_fp_save(struct kvm_cpu_context *cntx,
+				  unsigned long isa)
+{
+	if ((cntx->sstatus & SR_FS) == SR_FS_DIRTY) {
+		if (riscv_isa_extension_available(&isa, d))
+			__kvm_riscv_fp_d_save(cntx);
+		else if (riscv_isa_extension_available(&isa, f))
+			__kvm_riscv_fp_f_save(cntx);
+		kvm_riscv_vcpu_fp_clean(cntx);
+	}
+}
+
+void kvm_riscv_vcpu_guest_fp_restore(struct kvm_cpu_context *cntx,
+				     unsigned long isa)
+{
+	if ((cntx->sstatus & SR_FS) != SR_FS_OFF) {
+		if (riscv_isa_extension_available(&isa, d))
+			__kvm_riscv_fp_d_restore(cntx);
+		else if (riscv_isa_extension_available(&isa, f))
+			__kvm_riscv_fp_f_restore(cntx);
+		kvm_riscv_vcpu_fp_clean(cntx);
+	}
+}
+
+void kvm_riscv_vcpu_host_fp_save(struct kvm_cpu_context *cntx)
+{
+	/* No need to check host sstatus as it can be modified outside */
+	if (riscv_isa_extension_available(NULL, d))
+		__kvm_riscv_fp_d_save(cntx);
+	else if (riscv_isa_extension_available(NULL, f))
+		__kvm_riscv_fp_f_save(cntx);
+}
+
+void kvm_riscv_vcpu_host_fp_restore(struct kvm_cpu_context *cntx)
+{
+	if (riscv_isa_extension_available(NULL, d))
+		__kvm_riscv_fp_d_restore(cntx);
+	else if (riscv_isa_extension_available(NULL, f))
+		__kvm_riscv_fp_f_restore(cntx);
+}
+#endif
+
+int kvm_riscv_vcpu_get_reg_fp(struct kvm_vcpu *vcpu,
+			      const struct kvm_one_reg *reg,
+			      unsigned long rtype)
+{
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+	unsigned long isa = vcpu->arch.isa;
+	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 |
+					    rtype);
+	void *reg_val;
+
+	if ((rtype == KVM_REG_RISCV_FP_F) &&
+	    riscv_isa_extension_available(&isa, f)) {
+		if (KVM_REG_SIZE(reg->id) != sizeof(u32))
+			return -EINVAL;
+		if (reg_num == KVM_REG_RISCV_FP_F_REG(fcsr))
+			reg_val = &cntx->fp.f.fcsr;
+		else if ((KVM_REG_RISCV_FP_F_REG(f[0]) <= reg_num) &&
+			  reg_num <= KVM_REG_RISCV_FP_F_REG(f[31]))
+			reg_val = &cntx->fp.f.f[reg_num];
+		else
+			return -EINVAL;
+	} else if ((rtype == KVM_REG_RISCV_FP_D) &&
+		   riscv_isa_extension_available(&isa, d)) {
+		if (reg_num == KVM_REG_RISCV_FP_D_REG(fcsr)) {
+			if (KVM_REG_SIZE(reg->id) != sizeof(u32))
+				return -EINVAL;
+			reg_val = &cntx->fp.d.fcsr;
+		} else if ((KVM_REG_RISCV_FP_D_REG(f[0]) <= reg_num) &&
+			   reg_num <= KVM_REG_RISCV_FP_D_REG(f[31])) {
+			if (KVM_REG_SIZE(reg->id) != sizeof(u64))
+				return -EINVAL;
+			reg_val = &cntx->fp.d.f[reg_num];
+		} else
+			return -EINVAL;
+	} else
+		return -EINVAL;
+
+	if (copy_to_user(uaddr, reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_set_reg_fp(struct kvm_vcpu *vcpu,
+			      const struct kvm_one_reg *reg,
+			      unsigned long rtype)
+{
+	struct kvm_cpu_context *cntx = &vcpu->arch.guest_context;
+	unsigned long isa = vcpu->arch.isa;
+	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 |
+					    rtype);
+	void *reg_val;
+
+	if ((rtype == KVM_REG_RISCV_FP_F) &&
+	    riscv_isa_extension_available(&isa, f)) {
+		if (KVM_REG_SIZE(reg->id) != sizeof(u32))
+			return -EINVAL;
+		if (reg_num == KVM_REG_RISCV_FP_F_REG(fcsr))
+			reg_val = &cntx->fp.f.fcsr;
+		else if ((KVM_REG_RISCV_FP_F_REG(f[0]) <= reg_num) &&
+			  reg_num <= KVM_REG_RISCV_FP_F_REG(f[31]))
+			reg_val = &cntx->fp.f.f[reg_num];
+		else
+			return -EINVAL;
+	} else if ((rtype == KVM_REG_RISCV_FP_D) &&
+		   riscv_isa_extension_available(&isa, d)) {
+		if (reg_num == KVM_REG_RISCV_FP_D_REG(fcsr)) {
+			if (KVM_REG_SIZE(reg->id) != sizeof(u32))
+				return -EINVAL;
+			reg_val = &cntx->fp.d.fcsr;
+		} else if ((KVM_REG_RISCV_FP_D_REG(f[0]) <= reg_num) &&
+			   reg_num <= KVM_REG_RISCV_FP_D_REG(f[31])) {
+			if (KVM_REG_SIZE(reg->id) != sizeof(u64))
+				return -EINVAL;
+			reg_val = &cntx->fp.d.f[reg_num];
+		} else
+			return -EINVAL;
+	} else
+		return -EINVAL;
+
+	if (copy_from_user(reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}

arch/riscv/kvm/vcpu_sbi.c

···
+// SPDX-License-Identifier: GPL-2.0
+/**
+ * Copyright (c) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Atish Patra <[e-mail]>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kvm_host.h>
+#include <asm/csr.h>
+#include <asm/sbi.h>
+#include <asm/kvm_vcpu_timer.h>
+
+#define SBI_VERSION_MAJOR			0
+#define SBI_VERSION_MINOR			1
+
+static void kvm_riscv_vcpu_sbi_forward(struct kvm_vcpu *vcpu,
+				       struct kvm_run *run)
+{
+	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
+
+	vcpu->arch.sbi_context.return_handled = 0;
+	vcpu->stat.ecall_exit_stat++;
+	run->exit_reason = KVM_EXIT_RISCV_SBI;
+	run->riscv_sbi.extension_id = cp->a7;
+	run->riscv_sbi.function_id = cp->a6;
+	run->riscv_sbi.args[0] = cp->a0;
+	run->riscv_sbi.args[1] = cp->a1;
+	run->riscv_sbi.args[2] = cp->a2;
+	run->riscv_sbi.args[3] = cp->a3;
+	run->riscv_sbi.args[4] = cp->a4;
+	run->riscv_sbi.args[5] = cp->a5;
+	run->riscv_sbi.ret[0] = cp->a0;
+	run->riscv_sbi.ret[1] = cp->a1;
+}
+
+int kvm_riscv_vcpu_sbi_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
+
+	/* Handle SBI return only once */
+	if (vcpu->arch.sbi_context.return_handled)
+		return 0;
+	vcpu->arch.sbi_context.return_handled = 1;
+
+	/* Update return values */
+	cp->a0 = run->riscv_sbi.ret[0];
+	cp->a1 = run->riscv_sbi.ret[1];
+
+	/* Move to next instruction */
+	vcpu->arch.guest_context.sepc += 4;
+
+	return 0;
+}
+
+#ifdef CONFIG_RISCV_SBI_V01
+
+static void kvm_sbi_system_shutdown(struct kvm_vcpu *vcpu,
+				    struct kvm_run *run, u32 type)
+{
+	int i;
+	struct kvm_vcpu *tmp;
+
+	kvm_for_each_vcpu(i, tmp, vcpu->kvm)
+		tmp->arch.power_off = true;
+	kvm_make_all_cpus_request(vcpu->kvm, KVM_REQ_SLEEP);
+
+	memset(&run->system_event, 0, sizeof(run->system_event));
+	run->system_event.type = type;
+	run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
+}
+
+int kvm_riscv_vcpu_sbi_ecall(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	ulong hmask;
+	int i, ret = 1;
+	u64 next_cycle;
+	struct kvm_vcpu *rvcpu;
+	bool next_sepc = true;
+	struct cpumask cm, hm;
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_cpu_trap utrap = { 0 };
+	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
+
+	if (!cp)
+		return -EINVAL;
+
+	switch (cp->a7) {
+	case SBI_EXT_0_1_CONSOLE_GETCHAR:
+	case SBI_EXT_0_1_CONSOLE_PUTCHAR:
+		/*
+		 * The CONSOLE_GETCHAR/CONSOLE_PUTCHAR SBI calls cannot be
+		 * handled in kernel so we forward these to user-space
+		 */
+		kvm_riscv_vcpu_sbi_forward(vcpu, run);
+		next_sepc = false;
+		ret = 0;
+		break;
+	case SBI_EXT_0_1_SET_TIMER:
+#if __riscv_xlen == 32
+		next_cycle = ((u64)cp->a1 << 32) | (u64)cp->a0;
+#else
+		next_cycle = (u64)cp->a0;
+#endif
+		kvm_riscv_vcpu_timer_next_event(vcpu, next_cycle);
+		break;
+	case SBI_EXT_0_1_CLEAR_IPI:
+		kvm_riscv_vcpu_unset_interrupt(vcpu, IRQ_VS_SOFT);
+		break;
+	case SBI_EXT_0_1_SEND_IPI:
+		if (cp->a0)
+			hmask = kvm_riscv_vcpu_unpriv_read(vcpu, false, cp->a0,
+							   &utrap);
+		else
+			hmask = (1UL << atomic_read(&kvm->online_vcpus)) - 1;
+		if (utrap.scause) {
+			utrap.sepc = cp->sepc;
+			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+			next_sepc = false;
+			break;
+		}
+		for_each_set_bit(i, &hmask, BITS_PER_LONG) {
+			rvcpu = kvm_get_vcpu_by_id(vcpu->kvm, i);
+			kvm_riscv_vcpu_set_interrupt(rvcpu, IRQ_VS_SOFT);
+		}
+		break;
+	case SBI_EXT_0_1_SHUTDOWN:
+		kvm_sbi_system_shutdown(vcpu, run, KVM_SYSTEM_EVENT_SHUTDOWN);
+		next_sepc = false;
+		ret = 0;
+		break;
+	case SBI_EXT_0_1_REMOTE_FENCE_I:
+	case SBI_EXT_0_1_REMOTE_SFENCE_VMA:
+	case SBI_EXT_0_1_REMOTE_SFENCE_VMA_ASID:
+		if (cp->a0)
+			hmask = kvm_riscv_vcpu_unpriv_read(vcpu, false, cp->a0,
+							   &utrap);
+		else
+			hmask = (1UL << atomic_read(&kvm->online_vcpus)) - 1;
+		if (utrap.scause) {
+			utrap.sepc = cp->sepc;
+			kvm_riscv_vcpu_trap_redirect(vcpu, &utrap);
+			next_sepc = false;
+			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);
+		}
+		riscv_cpuid_to_hartid_mask(&cm, &hm);
+		if (cp->a7 == SBI_EXT_0_1_REMOTE_FENCE_I)
+			sbi_remote_fence_i(cpumask_bits(&hm));
+		else if (cp->a7 == SBI_EXT_0_1_REMOTE_SFENCE_VMA)
+			sbi_remote_hfence_vvma(cpumask_bits(&hm),
+						cp->a1, cp->a2);
+		else
+			sbi_remote_hfence_vvma_asid(cpumask_bits(&hm),
+						cp->a1, cp->a2, cp->a3);
+		break;
+	default:
+		/* Return error for unsupported SBI calls */
+		cp->a0 = SBI_ERR_NOT_SUPPORTED;
+		break;
+	}
+
+	if (next_sepc)
+		cp->sepc += 4;
+
+	return ret;
+}
+
+#else
+
+int kvm_riscv_vcpu_sbi_ecall(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	kvm_riscv_vcpu_sbi_forward(vcpu, run);
+	return 0;
+}
+
+#endif

arch/riscv/kvm/vcpu_switch.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>
+#include <asm/asm-offsets.h>
+#include <asm/csr.h>
+
+	.text
+	.altmacro
+	.option norelax
+
+ENTRY(__kvm_riscv_switch_to)
+	/* Save Host GPRs (except A0 and T0-T6) */
+	REG_S	ra, (KVM_ARCH_HOST_RA)(a0)
+	REG_S	sp, (KVM_ARCH_HOST_SP)(a0)
+	REG_S	gp, (KVM_ARCH_HOST_GP)(a0)
+	REG_S	tp, (KVM_ARCH_HOST_TP)(a0)
+	REG_S	s0, (KVM_ARCH_HOST_S0)(a0)
+	REG_S	s1, (KVM_ARCH_HOST_S1)(a0)
+	REG_S	a1, (KVM_ARCH_HOST_A1)(a0)
+	REG_S	a2, (KVM_ARCH_HOST_A2)(a0)
+	REG_S	a3, (KVM_ARCH_HOST_A3)(a0)
+	REG_S	a4, (KVM_ARCH_HOST_A4)(a0)
+	REG_S	a5, (KVM_ARCH_HOST_A5)(a0)
+	REG_S	a6, (KVM_ARCH_HOST_A6)(a0)
+	REG_S	a7, (KVM_ARCH_HOST_A7)(a0)
+	REG_S	s2, (KVM_ARCH_HOST_S2)(a0)
+	REG_S	s3, (KVM_ARCH_HOST_S3)(a0)
+	REG_S	s4, (KVM_ARCH_HOST_S4)(a0)
+	REG_S	s5, (KVM_ARCH_HOST_S5)(a0)
+	REG_S	s6, (KVM_ARCH_HOST_S6)(a0)
+	REG_S	s7, (KVM_ARCH_HOST_S7)(a0)
+	REG_S	s8, (KVM_ARCH_HOST_S8)(a0)
+	REG_S	s9, (KVM_ARCH_HOST_S9)(a0)
+	REG_S	s10, (KVM_ARCH_HOST_S10)(a0)
+	REG_S	s11, (KVM_ARCH_HOST_S11)(a0)
+
+	/* Save Host and Restore Guest SSTATUS */
+	REG_L	t0, (KVM_ARCH_GUEST_SSTATUS)(a0)
+	csrrw	t0, CSR_SSTATUS, t0
+	REG_S	t0, (KVM_ARCH_HOST_SSTATUS)(a0)
+
+	/* Save Host and Restore Guest HSTATUS */
+	REG_L	t1, (KVM_ARCH_GUEST_HSTATUS)(a0)
+	csrrw	t1, CSR_HSTATUS, t1
+	REG_S	t1, (KVM_ARCH_HOST_HSTATUS)(a0)
+
+	/* Save Host and Restore Guest SCOUNTEREN */
+	REG_L	t2, (KVM_ARCH_GUEST_SCOUNTEREN)(a0)
+	csrrw	t2, CSR_SCOUNTEREN, t2
+	REG_S	t2, (KVM_ARCH_HOST_SCOUNTEREN)(a0)
+
+	/* Save Host SSCRATCH and change it to struct kvm_vcpu_arch pointer */
+	csrrw	t3, CSR_SSCRATCH, a0
+	REG_S	t3, (KVM_ARCH_HOST_SSCRATCH)(a0)
+
+	/* Save Host STVEC and change it to return path */
+	la	t4, __kvm_switch_return
+	csrrw	t4, CSR_STVEC, t4
+	REG_S	t4, (KVM_ARCH_HOST_STVEC)(a0)
+
+	/* Restore Guest SEPC */
+	REG_L	t0, (KVM_ARCH_GUEST_SEPC)(a0)
+	csrw	CSR_SEPC, t0
+
+	/* Restore Guest GPRs (except A0) */
+	REG_L	ra, (KVM_ARCH_GUEST_RA)(a0)
+	REG_L	sp, (KVM_ARCH_GUEST_SP)(a0)
+	REG_L	gp, (KVM_ARCH_GUEST_GP)(a0)
+	REG_L	tp, (KVM_ARCH_GUEST_TP)(a0)
+	REG_L	t0, (KVM_ARCH_GUEST_T0)(a0)
+	REG_L	t1, (KVM_ARCH_GUEST_T1)(a0)
+	REG_L	t2, (KVM_ARCH_GUEST_T2)(a0)
+	REG_L	s0, (KVM_ARCH_GUEST_S0)(a0)
+	REG_L	s1, (KVM_ARCH_GUEST_S1)(a0)
+	REG_L	a1, (KVM_ARCH_GUEST_A1)(a0)
+	REG_L	a2, (KVM_ARCH_GUEST_A2)(a0)
+	REG_L	a3, (KVM_ARCH_GUEST_A3)(a0)
+	REG_L	a4, (KVM_ARCH_GUEST_A4)(a0)
+	REG_L	a5, (KVM_ARCH_GUEST_A5)(a0)
+	REG_L	a6, (KVM_ARCH_GUEST_A6)(a0)
+	REG_L	a7, (KVM_ARCH_GUEST_A7)(a0)
+	REG_L	s2, (KVM_ARCH_GUEST_S2)(a0)
+	REG_L	s3, (KVM_ARCH_GUEST_S3)(a0)
+	REG_L	s4, (KVM_ARCH_GUEST_S4)(a0)
+	REG_L	s5, (KVM_ARCH_GUEST_S5)(a0)
+	REG_L	s6, (KVM_ARCH_GUEST_S6)(a0)
+	REG_L	s7, (KVM_ARCH_GUEST_S7)(a0)
+	REG_L	s8, (KVM_ARCH_GUEST_S8)(a0)
+	REG_L	s9, (KVM_ARCH_GUEST_S9)(a0)
+	REG_L	s10, (KVM_ARCH_GUEST_S10)(a0)
+	REG_L	s11, (KVM_ARCH_GUEST_S11)(a0)
+	REG_L	t3, (KVM_ARCH_GUEST_T3)(a0)
+	REG_L	t4, (KVM_ARCH_GUEST_T4)(a0)
+	REG_L	t5, (KVM_ARCH_GUEST_T5)(a0)
+	REG_L	t6, (KVM_ARCH_GUEST_T6)(a0)
+
+	/* Restore Guest A0 */
+	REG_L	a0, (KVM_ARCH_GUEST_A0)(a0)
+
+	/* Resume Guest */
+	sret
+
+	/* Back to Host */
+	.align 2
+__kvm_switch_return:
+	/* Swap Guest A0 with SSCRATCH */
+	csrrw	a0, CSR_SSCRATCH, a0
+
+	/* Save Guest GPRs (except A0) */
+	REG_S	ra, (KVM_ARCH_GUEST_RA)(a0)
+	REG_S	sp, (KVM_ARCH_GUEST_SP)(a0)
+	REG_S	gp, (KVM_ARCH_GUEST_GP)(a0)
+	REG_S	tp, (KVM_ARCH_GUEST_TP)(a0)
+	REG_S	t0, (KVM_ARCH_GUEST_T0)(a0)
+	REG_S	t1, (KVM_ARCH_GUEST_T1)(a0)
+	REG_S	t2, (KVM_ARCH_GUEST_T2)(a0)
+	REG_S	s0, (KVM_ARCH_GUEST_S0)(a0)
+	REG_S	s1, (KVM_ARCH_GUEST_S1)(a0)
+	REG_S	a1, (KVM_ARCH_GUEST_A1)(a0)
+	REG_S	a2, (KVM_ARCH_GUEST_A2)(a0)
+	REG_S	a3, (KVM_ARCH_GUEST_A3)(a0)
+	REG_S	a4, (KVM_ARCH_GUEST_A4)(a0)
+	REG_S	a5, (KVM_ARCH_GUEST_A5)(a0)
+	REG_S	a6, (KVM_ARCH_GUEST_A6)(a0)
+	REG_S	a7, (KVM_ARCH_GUEST_A7)(a0)
+	REG_S	s2, (KVM_ARCH_GUEST_S2)(a0)
+	REG_S	s3, (KVM_ARCH_GUEST_S3)(a0)
+	REG_S	s4, (KVM_ARCH_GUEST_S4)(a0)
+	REG_S	s5, (KVM_ARCH_GUEST_S5)(a0)
+	REG_S	s6, (KVM_ARCH_GUEST_S6)(a0)
+	REG_S	s7, (KVM_ARCH_GUEST_S7)(a0)
+	REG_S	s8, (KVM_ARCH_GUEST_S8)(a0)
+	REG_S	s9, (KVM_ARCH_GUEST_S9)(a0)
+	REG_S	s10, (KVM_ARCH_GUEST_S10)(a0)
+	REG_S	s11, (KVM_ARCH_GUEST_S11)(a0)
+	REG_S	t3, (KVM_ARCH_GUEST_T3)(a0)
+	REG_S	t4, (KVM_ARCH_GUEST_T4)(a0)
+	REG_S	t5, (KVM_ARCH_GUEST_T5)(a0)
+	REG_S	t6, (KVM_ARCH_GUEST_T6)(a0)
+
+	/* Save Guest SEPC */
+	csrr	t0, CSR_SEPC
+	REG_S	t0, (KVM_ARCH_GUEST_SEPC)(a0)
+
+	/* Restore Host STVEC */
+	REG_L	t1, (KVM_ARCH_HOST_STVEC)(a0)
+	csrw	CSR_STVEC, t1
+
+	/* Save Guest A0 and Restore Host SSCRATCH */
+	REG_L	t2, (KVM_ARCH_HOST_SSCRATCH)(a0)
+	csrrw	t2, CSR_SSCRATCH, t2
+	REG_S	t2, (KVM_ARCH_GUEST_A0)(a0)
+
+	/* Save Guest and Restore Host SCOUNTEREN */
+	REG_L	t3, (KVM_ARCH_HOST_SCOUNTEREN)(a0)
+	csrrw	t3, CSR_SCOUNTEREN, t3
+	REG_S	t3, (KVM_ARCH_GUEST_SCOUNTEREN)(a0)
+
+	/* Save Guest and Restore Host HSTATUS */
+	REG_L	t4, (KVM_ARCH_HOST_HSTATUS)(a0)
+	csrrw	t4, CSR_HSTATUS, t4
+	REG_S	t4, (KVM_ARCH_GUEST_HSTATUS)(a0)
+
+	/* Save Guest and Restore Host SSTATUS */
+	REG_L	t5, (KVM_ARCH_HOST_SSTATUS)(a0)
+	csrrw	t5, CSR_SSTATUS, t5
+	REG_S	t5, (KVM_ARCH_GUEST_SSTATUS)(a0)
+
+	/* Restore Host GPRs (except A0 and T0-T6) */
+	REG_L	ra, (KVM_ARCH_HOST_RA)(a0)
+	REG_L	sp, (KVM_ARCH_HOST_SP)(a0)
+	REG_L	gp, (KVM_ARCH_HOST_GP)(a0)
+	REG_L	tp, (KVM_ARCH_HOST_TP)(a0)
+	REG_L	s0, (KVM_ARCH_HOST_S0)(a0)
+	REG_L	s1, (KVM_ARCH_HOST_S1)(a0)
+	REG_L	a1, (KVM_ARCH_HOST_A1)(a0)
+	REG_L	a2, (KVM_ARCH_HOST_A2)(a0)
+	REG_L	a3, (KVM_ARCH_HOST_A3)(a0)
+	REG_L	a4, (KVM_ARCH_HOST_A4)(a0)
+	REG_L	a5, (KVM_ARCH_HOST_A5)(a0)
+	REG_L	a6, (KVM_ARCH_HOST_A6)(a0)
+	REG_L	a7, (KVM_ARCH_HOST_A7)(a0)
+	REG_L	s2, (KVM_ARCH_HOST_S2)(a0)
+	REG_L	s3, (KVM_ARCH_HOST_S3)(a0)
+	REG_L	s4, (KVM_ARCH_HOST_S4)(a0)
+	REG_L	s5, (KVM_ARCH_HOST_S5)(a0)
+	REG_L	s6, (KVM_ARCH_HOST_S6)(a0)
+	REG_L	s7, (KVM_ARCH_HOST_S7)(a0)
+	REG_L	s8, (KVM_ARCH_HOST_S8)(a0)
+	REG_L	s9, (KVM_ARCH_HOST_S9)(a0)
+	REG_L	s10, (KVM_ARCH_HOST_S10)(a0)
+	REG_L	s11, (KVM_ARCH_HOST_S11)(a0)
+
+	/* Return to C code */
+	ret
+ENDPROC(__kvm_riscv_switch_to)
+
+ENTRY(__kvm_riscv_unpriv_trap)
+	/*
+	 * We assume that faulting unpriv load/store instruction is
+	 * 4-byte long and blindly increment SEPC by 4.
+	 *
+	 * The trap details will be saved at address pointed by 'A0'
+	 * register and we use 'A1' register as temporary.
+	 */
+	csrr	a1, CSR_SEPC
+	REG_S	a1, (KVM_ARCH_TRAP_SEPC)(a0)
+	addi	a1, a1, 4
+	csrw	CSR_SEPC, a1
+	csrr	a1, CSR_SCAUSE
+	REG_S	a1, (KVM_ARCH_TRAP_SCAUSE)(a0)
+	csrr	a1, CSR_STVAL
+	REG_S	a1, (KVM_ARCH_TRAP_STVAL)(a0)
+	csrr	a1, CSR_HTVAL
+	REG_S	a1, (KVM_ARCH_TRAP_HTVAL)(a0)
+	csrr	a1, CSR_HTINST
+	REG_S	a1, (KVM_ARCH_TRAP_HTINST)(a0)
+	sret
+ENDPROC(__kvm_riscv_unpriv_trap)
+
+#ifdef	CONFIG_FPU
+	.align 3
+	.global __kvm_riscv_fp_f_save
+__kvm_riscv_fp_f_save:
+	csrr t2, CSR_SSTATUS
+	li t1, SR_FS
+	csrs CSR_SSTATUS, t1
+	frcsr t0
+	fsw f0,  KVM_ARCH_FP_F_F0(a0)
+	fsw f1,  KVM_ARCH_FP_F_F1(a0)
+	fsw f2,  KVM_ARCH_FP_F_F2(a0)
+	fsw f3,  KVM_ARCH_FP_F_F3(a0)
+	fsw f4,  KVM_ARCH_FP_F_F4(a0)
+	fsw f5,  KVM_ARCH_FP_F_F5(a0)
+	fsw f6,  KVM_ARCH_FP_F_F6(a0)
+	fsw f7,  KVM_ARCH_FP_F_F7(a0)
+	fsw f8,  KVM_ARCH_FP_F_F8(a0)
+	fsw f9,  KVM_ARCH_FP_F_F9(a0)
+	fsw f10, KVM_ARCH_FP_F_F10(a0)
+	fsw f11, KVM_ARCH_FP_F_F11(a0)
+	fsw f12, KVM_ARCH_FP_F_F12(a0)
+	fsw f13, KVM_ARCH_FP_F_F13(a0)
+	fsw f14, KVM_ARCH_FP_F_F14(a0)
+	fsw f15, KVM_ARCH_FP_F_F15(a0)
+	fsw f16, KVM_ARCH_FP_F_F16(a0)
+	fsw f17, KVM_ARCH_FP_F_F17(a0)
+	fsw f18, KVM_ARCH_FP_F_F18(a0)
+	fsw f19, KVM_ARCH_FP_F_F19(a0)
+	fsw f20, KVM_ARCH_FP_F_F20(a0)
+	fsw f21, KVM_ARCH_FP_F_F21(a0)
+	fsw f22, KVM_ARCH_FP_F_F22(a0)
+	fsw f23, KVM_ARCH_FP_F_F23(a0)
+	fsw f24, KVM_ARCH_FP_F_F24(a0)
+	fsw f25, KVM_ARCH_FP_F_F25(a0)
+	fsw f26, KVM_ARCH_FP_F_F26(a0)
+	fsw f27, KVM_ARCH_FP_F_F27(a0)
+	fsw f28, KVM_ARCH_FP_F_F28(a0)
+	fsw f29, KVM_ARCH_FP_F_F29(a0)
+	fsw f30, KVM_ARCH_FP_F_F30(a0)
+	fsw f31, KVM_ARCH_FP_F_F31(a0)
+	sw t0, KVM_ARCH_FP_F_FCSR(a0)
+	csrw CSR_SSTATUS, t2
+	ret
+
+	.align 3
+	.global __kvm_riscv_fp_d_save
+__kvm_riscv_fp_d_save:
+	csrr t2, CSR_SSTATUS
+	li t1, SR_FS
+	csrs CSR_SSTATUS, t1
+	frcsr t0
+	fsd f0,  KVM_ARCH_FP_D_F0(a0)
+	fsd f1,  KVM_ARCH_FP_D_F1(a0)
+	fsd f2,  KVM_ARCH_FP_D_F2(a0)
+	fsd f3,  KVM_ARCH_FP_D_F3(a0)
+	fsd f4,  KVM_ARCH_FP_D_F4(a0)
+	fsd f5,  KVM_ARCH_FP_D_F5(a0)
+	fsd f6,  KVM_ARCH_FP_D_F6(a0)
+	fsd f7,  KVM_ARCH_FP_D_F7(a0)
+	fsd f8,  KVM_ARCH_FP_D_F8(a0)
+	fsd f9,  KVM_ARCH_FP_D_F9(a0)
+	fsd f10, KVM_ARCH_FP_D_F10(a0)
+	fsd f11, KVM_ARCH_FP_D_F11(a0)
+	fsd f12, KVM_ARCH_FP_D_F12(a0)
+	fsd f13, KVM_ARCH_FP_D_F13(a0)
+	fsd f14, KVM_ARCH_FP_D_F14(a0)
+	fsd f15, KVM_ARCH_FP_D_F15(a0)
+	fsd f16, KVM_ARCH_FP_D_F16(a0)
+	fsd f17, KVM_ARCH_FP_D_F17(a0)
+	fsd f18, KVM_ARCH_FP_D_F18(a0)
+	fsd f19, KVM_ARCH_FP_D_F19(a0)
+	fsd f20, KVM_ARCH_FP_D_F20(a0)
+	fsd f21, KVM_ARCH_FP_D_F21(a0)
+	fsd f22, KVM_ARCH_FP_D_F22(a0)
+	fsd f23, KVM_ARCH_FP_D_F23(a0)
+	fsd f24, KVM_ARCH_FP_D_F24(a0)
+	fsd f25, KVM_ARCH_FP_D_F25(a0)
+	fsd f26, KVM_ARCH_FP_D_F26(a0)
+	fsd f27, KVM_ARCH_FP_D_F27(a0)
+	fsd f28, KVM_ARCH_FP_D_F28(a0)
+	fsd f29, KVM_ARCH_FP_D_F29(a0)
+	fsd f30, KVM_ARCH_FP_D_F30(a0)
+	fsd f31, KVM_ARCH_FP_D_F31(a0)
+	sw t0, KVM_ARCH_FP_D_FCSR(a0)
+	csrw CSR_SSTATUS, t2
+	ret
+
+	.align 3
+	.global __kvm_riscv_fp_f_restore
+__kvm_riscv_fp_f_restore:
+	csrr t2, CSR_SSTATUS
+	li t1, SR_FS
+	lw t0, KVM_ARCH_FP_F_FCSR(a0)
+	csrs CSR_SSTATUS, t1
+	flw f0,  KVM_ARCH_FP_F_F0(a0)
+	flw f1,  KVM_ARCH_FP_F_F1(a0)
+	flw f2,  KVM_ARCH_FP_F_F2(a0)
+	flw f3,  KVM_ARCH_FP_F_F3(a0)
+	flw f4,  KVM_ARCH_FP_F_F4(a0)
+	flw f5,  KVM_ARCH_FP_F_F5(a0)
+	flw f6,  KVM_ARCH_FP_F_F6(a0)
+	flw f7,  KVM_ARCH_FP_F_F7(a0)
+	flw f8,  KVM_ARCH_FP_F_F8(a0)
+	flw f9,  KVM_ARCH_FP_F_F9(a0)
+	flw f10, KVM_ARCH_FP_F_F10(a0)
+	flw f11, KVM_ARCH_FP_F_F11(a0)
+	flw f12, KVM_ARCH_FP_F_F12(a0)
+	flw f13, KVM_ARCH_FP_F_F13(a0)
+	flw f14, KVM_ARCH_FP_F_F14(a0)
+	flw f15, KVM_ARCH_FP_F_F15(a0)
+	flw f16, KVM_ARCH_FP_F_F16(a0)
+	flw f17, KVM_ARCH_FP_F_F17(a0)
+	flw f18, KVM_ARCH_FP_F_F18(a0)
+	flw f19, KVM_ARCH_FP_F_F19(a0)
+	flw f20, KVM_ARCH_FP_F_F20(a0)
+	flw f21, KVM_ARCH_FP_F_F21(a0)
+	flw f22, KVM_ARCH_FP_F_F22(a0)
+	flw f23, KVM_ARCH_FP_F_F23(a0)
+	flw f24, KVM_ARCH_FP_F_F24(a0)
+	flw f25, KVM_ARCH_FP_F_F25(a0)
+	flw f26, KVM_ARCH_FP_F_F26(a0)
+	flw f27, KVM_ARCH_FP_F_F27(a0)
+	flw f28, KVM_ARCH_FP_F_F28(a0)
+	flw f29, KVM_ARCH_FP_F_F29(a0)
+	flw f30, KVM_ARCH_FP_F_F30(a0)
+	flw f31, KVM_ARCH_FP_F_F31(a0)
+	fscsr t0
+	csrw CSR_SSTATUS, t2
+	ret
+
+	.align 3
+	.global __kvm_riscv_fp_d_restore
+__kvm_riscv_fp_d_restore:
+	csrr t2, CSR_SSTATUS
+	li t1, SR_FS
+	lw t0, KVM_ARCH_FP_D_FCSR(a0)
+	csrs CSR_SSTATUS, t1
+	fld f0,  KVM_ARCH_FP_D_F0(a0)
+	fld f1,  KVM_ARCH_FP_D_F1(a0)
+	fld f2,  KVM_ARCH_FP_D_F2(a0)
+	fld f3,  KVM_ARCH_FP_D_F3(a0)
+	fld f4,  KVM_ARCH_FP_D_F4(a0)
+	fld f5,  KVM_ARCH_FP_D_F5(a0)
+	fld f6,  KVM_ARCH_FP_D_F6(a0)
+	fld f7,  KVM_ARCH_FP_D_F7(a0)
+	fld f8,  KVM_ARCH_FP_D_F8(a0)
+	fld f9,  KVM_ARCH_FP_D_F9(a0)
+	fld f10, KVM_ARCH_FP_D_F10(a0)
+	fld f11, KVM_ARCH_FP_D_F11(a0)
+	fld f12, KVM_ARCH_FP_D_F12(a0)
+	fld f13, KVM_ARCH_FP_D_F13(a0)
+	fld f14, KVM_ARCH_FP_D_F14(a0)
+	fld f15, KVM_ARCH_FP_D_F15(a0)
+	fld f16, KVM_ARCH_FP_D_F16(a0)
+	fld f17, KVM_ARCH_FP_D_F17(a0)
+	fld f18, KVM_ARCH_FP_D_F18(a0)
+	fld f19, KVM_ARCH_FP_D_F19(a0)
+	fld f20, KVM_ARCH_FP_D_F20(a0)
+	fld f21, KVM_ARCH_FP_D_F21(a0)
+	fld f22, KVM_ARCH_FP_D_F22(a0)
+	fld f23, KVM_ARCH_FP_D_F23(a0)
+	fld f24, KVM_ARCH_FP_D_F24(a0)
+	fld f25, KVM_ARCH_FP_D_F25(a0)
+	fld f26, KVM_ARCH_FP_D_F26(a0)
+	fld f27, KVM_ARCH_FP_D_F27(a0)
+	fld f28, KVM_ARCH_FP_D_F28(a0)
+	fld f29, KVM_ARCH_FP_D_F29(a0)
+	fld f30, KVM_ARCH_FP_D_F30(a0)
+	fld f31, KVM_ARCH_FP_D_F31(a0)
+	fscsr t0
+	csrw CSR_SSTATUS, t2
+	ret
+#endif

arch/riscv/kvm/vcpu_timer.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Atish Patra <[e-mail]>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/kvm_host.h>
+#include <linux/uaccess.h>
+#include <clocksource/timer-riscv.h>
+#include <asm/csr.h>
+#include <asm/delay.h>
+#include <asm/kvm_vcpu_timer.h>
+
+static u64 kvm_riscv_current_cycles(struct kvm_guest_timer *gt)
+{
+	return get_cycles64() + gt->time_delta;
+}
+
+static u64 kvm_riscv_delta_cycles2ns(u64 cycles,
+				     struct kvm_guest_timer *gt,
+				     struct kvm_vcpu_timer *t)
+{
+	unsigned long flags;
+	u64 cycles_now, cycles_delta, delta_ns;
+
+	local_irq_save(flags);
+	cycles_now = kvm_riscv_current_cycles(gt);
+	if (cycles_now < cycles)
+		cycles_delta = cycles - cycles_now;
+	else
+		cycles_delta = 0;
+	delta_ns = (cycles_delta * gt->nsec_mult) >> gt->nsec_shift;
+	local_irq_restore(flags);
+
+	return delta_ns;
+}
+
+static enum hrtimer_restart kvm_riscv_vcpu_hrtimer_expired(struct hrtimer *h)
+{
+	u64 delta_ns;
+	struct kvm_vcpu_timer *t = container_of(h, struct kvm_vcpu_timer, hrt);
+	struct kvm_vcpu *vcpu = container_of(t, struct kvm_vcpu, arch.timer);
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+
+	if (kvm_riscv_current_cycles(gt) < t->next_cycles) {
+		delta_ns = kvm_riscv_delta_cycles2ns(t->next_cycles, gt, t);
+		hrtimer_forward_now(&t->hrt, ktime_set(0, delta_ns));
+		return HRTIMER_RESTART;
+	}
+
+	t->next_set = false;
+	kvm_riscv_vcpu_set_interrupt(vcpu, IRQ_VS_TIMER);
+
+	return HRTIMER_NORESTART;
+}
+
+static int kvm_riscv_vcpu_timer_cancel(struct kvm_vcpu_timer *t)
+{
+	if (!t->init_done || !t->next_set)
+		return -EINVAL;
+
+	hrtimer_cancel(&t->hrt);
+	t->next_set = false;
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_timer_next_event(struct kvm_vcpu *vcpu, u64 ncycles)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+	u64 delta_ns;
+
+	if (!t->init_done)
+		return -EINVAL;
+
+	kvm_riscv_vcpu_unset_interrupt(vcpu, IRQ_VS_TIMER);
+
+	delta_ns = kvm_riscv_delta_cycles2ns(ncycles, gt, t);
+	t->next_cycles = ncycles;
+	hrtimer_start(&t->hrt, ktime_set(0, delta_ns), HRTIMER_MODE_REL);
+	t->next_set = true;
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_get_reg_timer(struct kvm_vcpu *vcpu,
+				 const struct kvm_one_reg *reg)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+	u64 __user *uaddr = (u64 __user *)(unsigned long)reg->addr;
+	unsigned long reg_num = reg->id & ~(KVM_REG_ARCH_MASK |
+					    KVM_REG_SIZE_MASK |
+					    KVM_REG_RISCV_TIMER);
+	u64 reg_val;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(u64))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_timer) / sizeof(u64))
+		return -EINVAL;
+
+	switch (reg_num) {
+	case KVM_REG_RISCV_TIMER_REG(frequency):
+		reg_val = riscv_timebase;
+		break;
+	case KVM_REG_RISCV_TIMER_REG(time):
+		reg_val = kvm_riscv_current_cycles(gt);
+		break;
+	case KVM_REG_RISCV_TIMER_REG(compare):
+		reg_val = t->next_cycles;
+		break;
+	case KVM_REG_RISCV_TIMER_REG(state):
+		reg_val = (t->next_set) ? KVM_RISCV_TIMER_STATE_ON :
+					  KVM_RISCV_TIMER_STATE_OFF;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_set_reg_timer(struct kvm_vcpu *vcpu,
+				 const struct kvm_one_reg *reg)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+	u64 __user *uaddr = (u64 __user *)(unsigned long)reg->addr;
+	unsigned long reg_num = reg->id & ~(KVM_REG_ARCH_MASK |
+					    KVM_REG_SIZE_MASK |
+					    KVM_REG_RISCV_TIMER);
+	u64 reg_val;
+	int ret = 0;
+
+	if (KVM_REG_SIZE(reg->id) != sizeof(u64))
+		return -EINVAL;
+	if (reg_num >= sizeof(struct kvm_riscv_timer) / sizeof(u64))
+		return -EINVAL;
+
+	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	switch (reg_num) {
+	case KVM_REG_RISCV_TIMER_REG(frequency):
+		ret = -EOPNOTSUPP;
+		break;
+	case KVM_REG_RISCV_TIMER_REG(time):
+		gt->time_delta = reg_val - get_cycles64();
+		break;
+	case KVM_REG_RISCV_TIMER_REG(compare):
+		t->next_cycles = reg_val;
+		break;
+	case KVM_REG_RISCV_TIMER_REG(state):
+		if (reg_val == KVM_RISCV_TIMER_STATE_ON)
+			ret = kvm_riscv_vcpu_timer_next_event(vcpu, reg_val);
+		else
+			ret = kvm_riscv_vcpu_timer_cancel(t);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+int kvm_riscv_vcpu_timer_init(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	if (t->init_done)
+		return -EINVAL;
+
+	hrtimer_init(&t->hrt, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	t->hrt.function = kvm_riscv_vcpu_hrtimer_expired;
+	t->init_done = true;
+	t->next_set = false;
+
+	return 0;
+}
+
+int kvm_riscv_vcpu_timer_deinit(struct kvm_vcpu *vcpu)
+{
+	int ret;
+
+	ret = kvm_riscv_vcpu_timer_cancel(&vcpu->arch.timer);
+	vcpu->arch.timer.init_done = false;
+
+	return ret;
+}
+
+int kvm_riscv_vcpu_timer_reset(struct kvm_vcpu *vcpu)
+{
+	return kvm_riscv_vcpu_timer_cancel(&vcpu->arch.timer);
+}
+
+void kvm_riscv_vcpu_timer_restore(struct kvm_vcpu *vcpu)
+{
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+
+#ifdef CONFIG_64BIT
+	csr_write(CSR_HTIMEDELTA, gt->time_delta);
+#else
+	csr_write(CSR_HTIMEDELTA, (u32)(gt->time_delta));
+	csr_write(CSR_HTIMEDELTAH, (u32)(gt->time_delta >> 32));
+#endif
+}
+
+int kvm_riscv_guest_timer_init(struct kvm *kvm)
+{
+	struct kvm_guest_timer *gt = &kvm->arch.timer;
+
+	riscv_cs_get_mult_shift(&gt->nsec_mult, &gt->nsec_shift);
+	gt->time_delta = -get_cycles64();
+
+	return 0;
+}

arch/riscv/kvm/vm.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/uaccess.h>
+#include <linux/kvm_host.h>
+
+const struct _kvm_stats_desc kvm_vm_stats_desc[] = {
+	KVM_GENERIC_VM_STATS()
+};
+static_assert(ARRAY_SIZE(kvm_vm_stats_desc) ==
+		sizeof(struct kvm_vm_stat) / sizeof(u64));
+
+const struct kvm_stats_header kvm_vm_stats_header = {
+	.name_size = KVM_STATS_NAME_SIZE,
+	.num_desc = ARRAY_SIZE(kvm_vm_stats_desc),
+	.id_offset =  sizeof(struct kvm_stats_header),
+	.desc_offset = sizeof(struct kvm_stats_header) + KVM_STATS_NAME_SIZE,
+	.data_offset = sizeof(struct kvm_stats_header) + KVM_STATS_NAME_SIZE +
+		       sizeof(kvm_vm_stats_desc),
+};
+
+int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
+{
+	int r;
+
+	r = kvm_riscv_stage2_alloc_pgd(kvm);
+	if (r)
+		return r;
+
+	r = kvm_riscv_stage2_vmid_init(kvm);
+	if (r) {
+		kvm_riscv_stage2_free_pgd(kvm);
+		return r;
+	}
+
+	return kvm_riscv_guest_timer_init(kvm);
+}
+
+void kvm_arch_destroy_vm(struct kvm *kvm)
+{
+	int i;
+
+	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+		if (kvm->vcpus[i]) {
+			kvm_vcpu_destroy(kvm->vcpus[i]);
+			kvm->vcpus[i] = NULL;
+		}
+	}
+	atomic_set(&kvm->online_vcpus, 0);
+}
+
+int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
+{
+	int r;
+
+	switch (ext) {
+	case KVM_CAP_IOEVENTFD:
+	case KVM_CAP_DEVICE_CTRL:
+	case KVM_CAP_USER_MEMORY:
+	case KVM_CAP_SYNC_MMU:
+	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
+	case KVM_CAP_ONE_REG:
+	case KVM_CAP_READONLY_MEM:
+	case KVM_CAP_MP_STATE:
+	case KVM_CAP_IMMEDIATE_EXIT:
+		r = 1;
+		break;
+	case KVM_CAP_NR_VCPUS:
+		r = num_online_cpus();
+		break;
+	case KVM_CAP_MAX_VCPUS:
+		r = KVM_MAX_VCPUS;
+		break;
+	case KVM_CAP_NR_MEMSLOTS:
+		r = KVM_USER_MEM_SLOTS;
+		break;
+	default:
+		r = 0;
+		break;
+	}
+
+	return r;
+}
+
+long kvm_arch_vm_ioctl(struct file *filp,
+		       unsigned int ioctl, unsigned long arg)
+{
+	return -EINVAL;
+}

arch/riscv/kvm/vmid.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *     Anup Patel <[e-mail]>
+ */
+
+#include <linux/bitops.h>
+#include <linux/cpumask.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/module.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)
+{
+	unsigned long old;
+
+	/* Figure-out number of VMID bits in HW */
+	old = csr_read(CSR_HGATP);
+	csr_write(CSR_HGATP, old | HGATP_VMID_MASK);
+	vmid_bits = csr_read(CSR_HGATP);
+	vmid_bits = (vmid_bits & HGATP_VMID_MASK) >> HGATP_VMID_SHIFT;
+	vmid_bits = fls_long(vmid_bits);
+	csr_write(CSR_HGATP, old);
+
+	/* We polluted local TLB so flush all guest TLB */
+	__kvm_riscv_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)
+{
+	return vmid_bits;
+}
+
+int kvm_riscv_stage2_vmid_init(struct kvm *kvm)
+{
+	/* Mark the initial VMID and VMID version invalid */
+	kvm->arch.vmid.vmid_version = 0;
+	kvm->arch.vmid.vmid = 0;
+
+	return 0;
+}
+
+bool kvm_riscv_stage2_vmid_ver_changed(struct kvm_vmid *vmid)
+{
+	if (!vmid_bits)
+		return false;
+
+	return unlikely(READ_ONCE(vmid->vmid_version) !=
+			READ_ONCE(vmid_version));
+}
+
+void kvm_riscv_stage2_vmid_update(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_vcpu *v;
+	struct cpumask hmask;
+	struct kvm_vmid *vmid = &vcpu->kvm->arch.vmid;
+
+	if (!kvm_riscv_stage2_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)) {
+		spin_unlock(&vmid_lock);
+		return;
+	}
+
+	/* First user of a new VMID version? */
+	if (unlikely(vmid_next == 0)) {
+		WRITE_ONCE(vmid_version, READ_ONCE(vmid_version) + 1);
+		vmid_next = 1;
+
+		/*
+		 * We ran out of VMIDs so we increment vmid_version and
+		 * start assigning VMIDs from 1.
+		 *
+		 * This also means existing VMIDs assignement to all Guest
+		 * 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
+		 * 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.
+		 */
+		riscv_cpuid_to_hartid_mask(cpu_online_mask, &hmask);
+		sbi_remote_hfence_gvma(cpumask_bits(&hmask), 0, 0);
+	}
+
+	vmid->vmid = vmid_next;
+	vmid_next++;
+	vmid_next &= (1 << vmid_bits) - 1;
+
+	WRITE_ONCE(vmid->vmid_version, READ_ONCE(vmid_version));
+
+	spin_unlock(&vmid_lock);
+
+	/* Request stage2 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/pgtable.h

···
 	pte_t res;
 
 	res = ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID));
+	/* At this point the reference through the mapping is still present */
 	if (mm_is_protected(mm) && pte_present(res))
-		uv_convert_from_secure(pte_val(res) & PAGE_MASK);
+		uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK);
 	return res;
 }
 
···
 	pte_t res;
 
 	res = ptep_xchg_direct(vma->vm_mm, addr, ptep, __pte(_PAGE_INVALID));
+	/* At this point the reference through the mapping is still present */
 	if (mm_is_protected(vma->vm_mm) && pte_present(res))
-		uv_convert_from_secure(pte_val(res) & PAGE_MASK);
+		uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK);
 	return res;
 }
 
···
 	} else {
 		res = ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID));
 	}
+	/* At this point the reference through the mapping is still present */
 	if (mm_is_protected(mm) && pte_present(res))
-		uv_convert_from_secure(pte_val(res) & PAGE_MASK);
+		uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK);
 	return res;
 }
 

arch/s390/include/asm/uv.h

···
 #include <asm/page.h>
 #include <asm/gmap.h>
 
+#define UVC_CC_OK	0
+#define UVC_CC_ERROR	1
+#define UVC_CC_BUSY	2
+#define UVC_CC_PARTIAL	3
+
 #define UVC_RC_EXECUTED		0x0001
 #define UVC_RC_INV_CMD		0x0002
 #define UVC_RC_INV_STATE	0x0003
···
 }
 
 int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb);
-int uv_destroy_page(unsigned long paddr);
+int uv_destroy_owned_page(unsigned long paddr);
 int uv_convert_from_secure(unsigned long paddr);
+int uv_convert_owned_from_secure(unsigned long paddr);
 int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr);
 
 void setup_uv(void);
···
 #define is_prot_virt_host() 0
 static inline void setup_uv(void) {}
 
-static inline int uv_destroy_page(unsigned long paddr)
+static inline int uv_destroy_owned_page(unsigned long paddr)
 {
 	return 0;
 }
 
 static inline int uv_convert_from_secure(unsigned long paddr)
+{
+	return 0;
+}
+
+static inline int uv_convert_owned_from_secure(unsigned long paddr)
 {
 	return 0;
 }

arch/s390/kernel/uv.c

···
  *
  * @paddr: Absolute host address of page to be destroyed
  */
-int uv_destroy_page(unsigned long paddr)
+static int uv_destroy_page(unsigned long paddr)
 {
 	struct uv_cb_cfs uvcb = {
 		.header.cmd = UVC_CMD_DESTR_SEC_STOR,
···
 }
 
 /*
+ * The caller must already hold a reference to the page
+ */
+int uv_destroy_owned_page(unsigned long paddr)
+{
+	struct page *page = phys_to_page(paddr);
+	int rc;
+
+	get_page(page);
+	rc = uv_destroy_page(paddr);
+	if (!rc)
+		clear_bit(PG_arch_1, &page->flags);
+	put_page(page);
+	return rc;
+}
+
+/*
  * Requests the Ultravisor to encrypt a guest page and make it
  * accessible to the host for paging (export).
  *
···
 	if (uv_call(0, (u64)&uvcb))
 		return -EINVAL;
 	return 0;
+}
+
+/*
+ * The caller must already hold a reference to the page
+ */
+int uv_convert_owned_from_secure(unsigned long paddr)
+{
+	struct page *page = phys_to_page(paddr);
+	int rc;
+
+	get_page(page);
+	rc = uv_convert_from_secure(paddr);
+	if (!rc)
+		clear_bit(PG_arch_1, &page->flags);
+	put_page(page);
+	return rc;
 }
 
 /*
···
 {
 	pte_t entry = READ_ONCE(*ptep);
 	struct page *page;
-	int expected, rc = 0;
+	int expected, cc = 0;
 
 	if (!pte_present(entry))
 		return -ENXIO;
···
 	if (!page_ref_freeze(page, expected))
 		return -EBUSY;
 	set_bit(PG_arch_1, &page->flags);
-	rc = uv_call(0, (u64)uvcb);
+	/*
+	 * If the UVC does not succeed or fail immediately, we don't want to
+	 * loop for long, or we might get stall notifications.
+	 * On the other hand, this is a complex scenario and we are holding a lot of
+	 * locks, so we can't easily sleep and reschedule. We try only once,
+	 * and if the UVC returned busy or partial completion, we return
+	 * -EAGAIN and we let the callers deal with it.
+	 */
+	cc = __uv_call(0, (u64)uvcb);
 	page_ref_unfreeze(page, expected);
-	/* Return -ENXIO if the page was not mapped, -EINVAL otherwise */
-	if (rc)
-		rc = uvcb->rc == 0x10a ? -ENXIO : -EINVAL;
-	return rc;
+	/*
+	 * Return -ENXIO if the page was not mapped, -EINVAL for other errors.
+	 * If busy or partially completed, return -EAGAIN.
+	 */
+	if (cc == UVC_CC_OK)
+		return 0;
+	else if (cc == UVC_CC_BUSY || cc == UVC_CC_PARTIAL)
+		return -EAGAIN;
+	return uvcb->rc == 0x10a ? -ENXIO : -EINVAL;
 }
 
 /*
···
 	uaddr = __gmap_translate(gmap, gaddr);
 	if (IS_ERR_VALUE(uaddr))
 		goto out;
-	vma = find_vma(gmap->mm, uaddr);
+	vma = vma_lookup(gmap->mm, uaddr);
 	if (!vma)
 		goto out;
 	/*
···
 	mmap_read_unlock(gmap->mm);
 
 	if (rc == -EAGAIN) {
+		/*
+		 * If we are here because the UVC returned busy or partial
+		 * completion, this is just a useless check, but it is safe.
+		 */
 		wait_on_page_writeback(page);
 	} else if (rc == -EBUSY) {
 		/*

arch/s390/kvm/intercept.c

···
 	 */
 	if (rc == -EINVAL)
 		return 0;
+	/*
+	 * If we got -EAGAIN here, we simply return it. It will eventually
+	 * get propagated all the way to userspace, which should then try
+	 * again.
+	 */
 	return rc;
 }
 

arch/s390/kvm/kvm-s390.c

···
 	case KVM_S390_PV_COMMAND: {
 		struct kvm_pv_cmd args;
 
-		/* protvirt means user sigp */
-		kvm->arch.user_cpu_state_ctrl = 1;
+		/* protvirt means user cpu state */
+		kvm_s390_set_user_cpu_state_ctrl(kvm);
 		r = 0;
 		if (!is_prot_virt_host()) {
 			r = -EINVAL;
···
 	vcpu_load(vcpu);
 
 	/* user space knows about this interface - let it control the state */
-	vcpu->kvm->arch.user_cpu_state_ctrl = 1;
+	kvm_s390_set_user_cpu_state_ctrl(vcpu->kvm);
 
 	switch (mp_state->mp_state) {
 	case KVM_MP_STATE_STOPPED:
···
 	if (kvm_run->kvm_dirty_regs & KVM_SYNC_DIAG318) {
 		vcpu->arch.diag318_info.val = kvm_run->s.regs.diag318;
 		vcpu->arch.sie_block->cpnc = vcpu->arch.diag318_info.cpnc;
+		VCPU_EVENT(vcpu, 3, "setting cpnc to %d", vcpu->arch.diag318_info.cpnc);
 	}
 	/*
 	 * If userspace sets the riccb (e.g. after migration) to a valid state,

arch/s390/kvm/kvm-s390.h

···
 	return kvm->arch.user_cpu_state_ctrl != 0;
 }
 
+static inline void kvm_s390_set_user_cpu_state_ctrl(struct kvm *kvm)
+{
+	if (kvm->arch.user_cpu_state_ctrl)
+		return;
+
+	VM_EVENT(kvm, 3, "%s", "ENABLE: Userspace CPU state control");
+	kvm->arch.user_cpu_state_ctrl = 1;
+}
+
 /* implemented in pv.c */
 int kvm_s390_pv_destroy_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc);
 int kvm_s390_pv_create_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc);

arch/s390/kvm/priv.c

···
 		mmap_read_unlock(current->mm);
 		if (rc == -EFAULT)
 			return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+		if (rc == -EAGAIN)
+			continue;
 		if (rc < 0)
 			return rc;
 		start += PAGE_SIZE;

arch/s390/kvm/pv.c

···
 
 int kvm_s390_pv_destroy_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc)
 {
-	int cc = 0;
+	int cc;
 
-	if (kvm_s390_pv_cpu_get_handle(vcpu)) {
-		cc = uv_cmd_nodata(kvm_s390_pv_cpu_get_handle(vcpu),
-				   UVC_CMD_DESTROY_SEC_CPU, rc, rrc);
+	if (!kvm_s390_pv_cpu_get_handle(vcpu))
+		return 0;
 
-		KVM_UV_EVENT(vcpu->kvm, 3,
-			     "PROTVIRT DESTROY VCPU %d: rc %x rrc %x",
-			     vcpu->vcpu_id, *rc, *rrc);
-		WARN_ONCE(cc, "protvirt destroy cpu failed rc %x rrc %x",
-			  *rc, *rrc);
-	}
+	cc = uv_cmd_nodata(kvm_s390_pv_cpu_get_handle(vcpu), UVC_CMD_DESTROY_SEC_CPU, rc, rrc);
+
+	KVM_UV_EVENT(vcpu->kvm, 3, "PROTVIRT DESTROY VCPU %d: rc %x rrc %x",
+		     vcpu->vcpu_id, *rc, *rrc);
+	WARN_ONCE(cc, "protvirt destroy cpu failed rc %x rrc %x", *rc, *rrc);
+
 	/* Intended memory leak for something that should never happen. */
 	if (!cc)
 		free_pages(vcpu->arch.pv.stor_base,
···
 	uvcb.conf_base_stor_origin = (u64)kvm->arch.pv.stor_base;
 	uvcb.conf_virt_stor_origin = (u64)kvm->arch.pv.stor_var;
 
-	cc = uv_call(0, (u64)&uvcb);
+	cc = uv_call_sched(0, (u64)&uvcb);
 	*rc = uvcb.header.rc;
 	*rrc = uvcb.header.rrc;
 	KVM_UV_EVENT(kvm, 3, "PROTVIRT CREATE VM: handle %llx len %llx rc %x rrc %x",

arch/s390/kvm/sigp.c

···
 static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter,
 			   u64 *status_reg)
 {
-	unsigned int i;
-	struct kvm_vcpu *v;
-	bool all_stopped = true;
-
-	kvm_for_each_vcpu(i, v, vcpu->kvm) {
-		if (v == vcpu)
-			continue;
-		if (!is_vcpu_stopped(v))
-			all_stopped = false;
-	}
-
 	*status_reg &= 0xffffffff00000000UL;
 
 	/* Reject set arch order, with czam we're always in z/Arch mode. */
-	*status_reg |= (all_stopped ? SIGP_STATUS_INVALID_PARAMETER :
-					SIGP_STATUS_INCORRECT_STATE);
+	*status_reg |= SIGP_STATUS_INVALID_PARAMETER;
 	return SIGP_CC_STATUS_STORED;
 }
 

arch/s390/mm/gmap.c

···
  */
 void __gmap_zap(struct gmap *gmap, unsigned long gaddr)
 {
+	struct vm_area_struct *vma;
 	unsigned long vmaddr;
 	spinlock_t *ptl;
 	pte_t *ptep;
···
 						   gaddr >> PMD_SHIFT);
 	if (vmaddr) {
 		vmaddr |= gaddr & ~PMD_MASK;
+
+		vma = vma_lookup(gmap->mm, vmaddr);
+		if (!vma || is_vm_hugetlb_page(vma))
+			return;
+
 		/* Get pointer to the page table entry */
 		ptep = get_locked_pte(gmap->mm, vmaddr, &ptl);
-		if (likely(ptep))
+		if (likely(ptep)) {
 			ptep_zap_unused(gmap->mm, vmaddr, ptep, 0);
-		pte_unmap_unlock(ptep, ptl);
+			pte_unmap_unlock(ptep, ptl);
+		}
 	}
 }
 EXPORT_SYMBOL_GPL(__gmap_zap);
···
 {
 	pte_t pte = READ_ONCE(*ptep);
 
+	/* There is a reference through the mapping */
 	if (pte_present(pte))
-		WARN_ON_ONCE(uv_destroy_page(pte_val(pte) & PAGE_MASK));
+		WARN_ON_ONCE(uv_destroy_owned_page(pte_val(pte) & PAGE_MASK));
+
 	return 0;
 }
 

arch/s390/mm/pgtable.c

···
 }
 
 #ifdef CONFIG_PGSTE
-static pmd_t *pmd_alloc_map(struct mm_struct *mm, unsigned long addr)
+static int pmd_lookup(struct mm_struct *mm, unsigned long addr, pmd_t **pmdp)
 {
+	struct vm_area_struct *vma;
 	pgd_t *pgd;
 	p4d_t *p4d;
 	pud_t *pud;
-	pmd_t *pmd;
+
+	/* We need a valid VMA, otherwise this is clearly a fault. */
+	vma = vma_lookup(mm, addr);
+	if (!vma)
+		return -EFAULT;
 
 	pgd = pgd_offset(mm, addr);
-	p4d = p4d_alloc(mm, pgd, addr);
-	if (!p4d)
-		return NULL;
-	pud = pud_alloc(mm, p4d, addr);
-	if (!pud)
-		return NULL;
-	pmd = pmd_alloc(mm, pud, addr);
-	return pmd;
+	if (!pgd_present(*pgd))
+		return -ENOENT;
+
+	p4d = p4d_offset(pgd, addr);
+	if (!p4d_present(*p4d))
+		return -ENOENT;
+
+	pud = pud_offset(p4d, addr);
+	if (!pud_present(*pud))
+		return -ENOENT;
+
+	/* Large PUDs are not supported yet. */
+	if (pud_large(*pud))
+		return -EFAULT;
+
+	*pmdp = pmd_offset(pud, addr);
+	return 0;
 }
 #endif
 
···
 	pmd_t *pmdp;
 	pte_t *ptep;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	/*
+	 * If we don't have a PTE table and if there is no huge page mapped,
+	 * we can ignore attempts to set the key to 0, because it already is 0.
+	 */
+	switch (pmd_lookup(mm, addr, &pmdp)) {
+	case -ENOENT:
+		return key ? -EFAULT : 0;
+	case 0:
+		break;
+	default:
 		return -EFAULT;
+	}
 
 	ptl = pmd_lock(mm, pmdp);
 	if (!pmd_present(*pmdp)) {
 		spin_unlock(ptl);
-		return -EFAULT;
+		return key ? -EFAULT : 0;
 	}
 
 	if (pmd_large(*pmdp)) {
···
 	}
 	spin_unlock(ptl);
 
-	ptep = pte_alloc_map_lock(mm, pmdp, addr, &ptl);
-	if (unlikely(!ptep))
-		return -EFAULT;
-
+	ptep = pte_offset_map_lock(mm, pmdp, addr, &ptl);
 	new = old = pgste_get_lock(ptep);
 	pgste_val(new) &= ~(PGSTE_GR_BIT | PGSTE_GC_BIT |
 			    PGSTE_ACC_BITS | PGSTE_FP_BIT);
···
 	pte_t *ptep;
 	int cc = 0;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	/*
+	 * If we don't have a PTE table and if there is no huge page mapped,
+	 * the storage key is 0 and there is nothing for us to do.
+	 */
+	switch (pmd_lookup(mm, addr, &pmdp)) {
+	case -ENOENT:
+		return 0;
+	case 0:
+		break;
+	default:
 		return -EFAULT;
+	}
 
 	ptl = pmd_lock(mm, pmdp);
 	if (!pmd_present(*pmdp)) {
 		spin_unlock(ptl);
-		return -EFAULT;
+		return 0;
 	}
 
 	if (pmd_large(*pmdp)) {
···
 	}
 	spin_unlock(ptl);
 
-	ptep = pte_alloc_map_lock(mm, pmdp, addr, &ptl);
-	if (unlikely(!ptep))
-		return -EFAULT;
-
+	ptep = pte_offset_map_lock(mm, pmdp, addr, &ptl);
 	new = old = pgste_get_lock(ptep);
 	/* Reset guest reference bit only */
 	pgste_val(new) &= ~PGSTE_GR_BIT;
···
 	pmd_t *pmdp;
 	pte_t *ptep;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	/*
+	 * If we don't have a PTE table and if there is no huge page mapped,
+	 * the storage key is 0.
+	 */
+	*key = 0;
+
+	switch (pmd_lookup(mm, addr, &pmdp)) {
+	case -ENOENT:
+		return 0;
+	case 0:
+		break;
+	default:
 		return -EFAULT;
+	}
 
 	ptl = pmd_lock(mm, pmdp);
 	if (!pmd_present(*pmdp)) {
-		/* Not yet mapped memory has a zero key */
 		spin_unlock(ptl);
-		*key = 0;
 		return 0;
 	}
 
···
 	}
 	spin_unlock(ptl);
 
-	ptep = pte_alloc_map_lock(mm, pmdp, addr, &ptl);
-	if (unlikely(!ptep))
-		return -EFAULT;
-
+	ptep = pte_offset_map_lock(mm, pmdp, addr, &ptl);
 	pgste = pgste_get_lock(ptep);
 	*key = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56;
 	paddr = pte_val(*ptep) & PAGE_MASK;
···
 int pgste_perform_essa(struct mm_struct *mm, unsigned long hva, int orc,
 			unsigned long *oldpte, unsigned long *oldpgste)
 {
+	struct vm_area_struct *vma;
 	unsigned long pgstev;
 	spinlock_t *ptl;
 	pgste_t pgste;
···
 	WARN_ON_ONCE(orc > ESSA_MAX);
 	if (unlikely(orc > ESSA_MAX))
 		return -EINVAL;
+
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;
···
 int set_pgste_bits(struct mm_struct *mm, unsigned long hva,
 			unsigned long bits, unsigned long value)
 {
+	struct vm_area_struct *vma;
 	spinlock_t *ptl;
 	pgste_t new;
 	pte_t *ptep;
 
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;
···
  */
 int get_pgste(struct mm_struct *mm, unsigned long hva, unsigned long *pgstep)
 {
+	struct vm_area_struct *vma;
 	spinlock_t *ptl;
 	pte_t *ptep;
 
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;

arch/x86/include/asm/kvm_host.h

···
  * so ratio of 4 should be enough.
  */
 #define KVM_VCPU_ID_RATIO 4
-#define KVM_MAX_VCPU_ID (KVM_MAX_VCPUS * KVM_VCPU_ID_RATIO)
+#define KVM_MAX_VCPU_IDS (KVM_MAX_VCPUS * KVM_VCPU_ID_RATIO)
 
 /* memory slots that are not exposed to userspace */
 #define KVM_PRIVATE_MEM_SLOTS 3
···
 #define KVM_HAVE_MMU_RWLOCK
 
 struct kvm_mmu_page;
+struct kvm_page_fault;
 
 /*
  * x86 supports 4 paging modes (5-level 64-bit, 4-level 64-bit, 3-level 32-bit,
···
 struct kvm_mmu {
 	unsigned long (*get_guest_pgd)(struct kvm_vcpu *vcpu);
 	u64 (*get_pdptr)(struct kvm_vcpu *vcpu, int index);
-	int (*page_fault)(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u32 err,
-			  bool prefault);
+	int (*page_fault)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
 	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
 				  struct x86_exception *fault);
 	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gpa_t gva_or_gpa,
···
 	u64 fixed_ctr_ctrl;
 	u64 global_ctrl;
 	u64 global_status;
-	u64 global_ovf_ctrl;
 	u64 counter_bitmask[2];
 	u64 global_ctrl_mask;
 	u64 global_ovf_ctrl_mask;
···
 	struct kvm_hyperv_exit exit;
 	struct kvm_vcpu_hv_stimer stimer[HV_SYNIC_STIMER_COUNT];
 	DECLARE_BITMAP(stimer_pending_bitmap, HV_SYNIC_STIMER_COUNT);
-	cpumask_t tlb_flush;
 	bool enforce_cpuid;
 	struct {
 		u32 features_eax; /* HYPERV_CPUID_FEATURES.EAX */
···
 	atomic_t apic_map_dirty;
 
 	/* Protects apic_access_memslot_enabled and apicv_inhibit_reasons */
-	struct mutex apicv_update_lock;
+	struct rw_semaphore apicv_update_lock;
 
 	bool apic_access_memslot_enabled;
 	unsigned long apicv_inhibit_reasons;
···
 
 	unsigned long irq_sources_bitmap;
 	s64 kvmclock_offset;
+
+	/*
+	 * This also protects nr_vcpus_matched_tsc which is read from a
+	 * preemption-disabled region, so it must be a raw spinlock.
+	 */
 	raw_spinlock_t tsc_write_lock;
 	u64 last_tsc_nsec;
 	u64 last_tsc_write;
 	u32 last_tsc_khz;
+	u64 last_tsc_offset;
 	u64 cur_tsc_nsec;
 	u64 cur_tsc_write;
 	u64 cur_tsc_offset;
 	u64 cur_tsc_generation;
 	int nr_vcpus_matched_tsc;
 
-	raw_spinlock_t pvclock_gtod_sync_lock;
+	seqcount_raw_spinlock_t pvclock_sc;
 	bool use_master_clock;
 	u64 master_kernel_ns;
 	u64 master_cycle_now;
···
 #endif /* CONFIG_X86_64 */
 
 	/*
-	 * If set, rmaps have been allocated for all memslots and should be
-	 * allocated for any newly created or modified memslots.
+	 * If set, at least one shadow root has been allocated. This flag
+	 * is used as one input when determining whether certain memslot
+	 * related allocations are necessary.
 	 */
-	bool memslots_have_rmaps;
+	bool shadow_root_allocated;
 
 #if IS_ENABLED(CONFIG_HYPERV)
 	hpa_t	hv_root_tdp;
···
 }
 
 struct kvm_x86_ops {
+	const char *name;
+
 	int (*hardware_enable)(void);
 	void (*hardware_disable)(void);
 	void (*hardware_unsetup)(void);
···
 	void (*write_tsc_multiplier)(struct kvm_vcpu *vcpu, u64 multiplier);
 
 	/*
-	 * Retrieve somewhat arbitrary exit information.  Intended to be used
-	 * only from within tracepoints to avoid VMREADs when tracing is off.
+	 * Retrieve somewhat arbitrary exit information.  Intended to
+	 * be used only from within tracepoints or error paths.
 	 */
-	void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2,
+	void (*get_exit_info)(struct kvm_vcpu *vcpu, u32 *reason,
+			      u64 *info1, u64 *info2,
 			      u32 *exit_int_info, u32 *exit_int_info_err_code);
 
 	int (*check_intercept)(struct kvm_vcpu *vcpu,
···
 {
 	return __vmalloc(kvm_x86_ops.vm_size, GFP_KERNEL_ACCOUNT | __GFP_ZERO);
 }
+
+#define __KVM_HAVE_ARCH_VM_FREE
 void kvm_arch_free_vm(struct kvm *kvm);
 
 #define __KVM_HAVE_ARCH_FLUSH_REMOTE_TLB
···
 int kvm_emulate_instruction(struct kvm_vcpu *vcpu, int emulation_type);
 int kvm_emulate_instruction_from_buffer(struct kvm_vcpu *vcpu,
 					void *insn, int insn_len);
+void __kvm_prepare_emulation_failure_exit(struct kvm_vcpu *vcpu,
+					  u64 *data, u8 ndata);
+void kvm_prepare_emulation_failure_exit(struct kvm_vcpu *vcpu);
 
 void kvm_enable_efer_bits(u64);
 bool kvm_valid_efer(struct kvm_vcpu *vcpu, u64 efer);
···
 void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
 bool kvm_inject_emulated_page_fault(struct kvm_vcpu *vcpu,
 				    struct x86_exception *fault);
-int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
-			    gfn_t gfn, void *data, int offset, int len,
-			    u32 access);
 bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
 bool kvm_require_dr(struct kvm_vcpu *vcpu, int dr);
 
···
 unsigned long kvm_get_linear_rip(struct kvm_vcpu *vcpu);
 bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip);
 
-void kvm_make_mclock_inprogress_request(struct kvm *kvm);
 void kvm_make_scan_ioapic_request(struct kvm *kvm);
 void kvm_make_scan_ioapic_request_mask(struct kvm *kvm,
 				       unsigned long *vcpu_bitmap);
···
 
 int kvm_cpu_dirty_log_size(void);
 
-int alloc_all_memslots_rmaps(struct kvm *kvm);
+int memslot_rmap_alloc(struct kvm_memory_slot *slot, unsigned long npages);
+
+#define KVM_CLOCK_VALID_FLAGS						\
+	(KVM_CLOCK_TSC_STABLE | KVM_CLOCK_REALTIME | KVM_CLOCK_HOST_TSC)
 
 #endif /* _ASM_X86_KVM_HOST_H */

arch/x86/include/asm/kvm_page_track.h

···
 int kvm_page_track_init(struct kvm *kvm);
 void kvm_page_track_cleanup(struct kvm *kvm);
 
+bool kvm_page_track_write_tracking_enabled(struct kvm *kvm);
+int kvm_page_track_write_tracking_alloc(struct kvm_memory_slot *slot);
+
 void kvm_page_track_free_memslot(struct kvm_memory_slot *slot);
-int kvm_page_track_create_memslot(struct kvm_memory_slot *slot,
+int kvm_page_track_create_memslot(struct kvm *kvm,
+				  struct kvm_memory_slot *slot,
 				  unsigned long npages);
 
 void kvm_slot_page_track_add_page(struct kvm *kvm,
···
 void kvm_slot_page_track_remove_page(struct kvm *kvm,
 				     struct kvm_memory_slot *slot, gfn_t gfn,
 				     enum kvm_page_track_mode mode);
-bool kvm_page_track_is_active(struct kvm_vcpu *vcpu, gfn_t gfn,
-			      enum kvm_page_track_mode mode);
+bool kvm_slot_page_track_is_active(struct kvm_vcpu *vcpu,
+				   struct kvm_memory_slot *slot, gfn_t gfn,
+				   enum kvm_page_track_mode mode);
 
 void
 kvm_page_track_register_notifier(struct kvm *kvm,

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

···
 #define KVM_PMU_EVENT_ALLOW 0
 #define KVM_PMU_EVENT_DENY 1
 
+/* for KVM_{GET,SET,HAS}_DEVICE_ATTR */
+#define KVM_VCPU_TSC_CTRL 0 /* control group for the timestamp counter (TSC) */
+#define   KVM_VCPU_TSC_OFFSET 0 /* attribute for the TSC offset */
+
 #endif /* _ASM_X86_KVM_H */

arch/x86/kernel/irq.c

···
 {
 	if (handler)
 		kvm_posted_intr_wakeup_handler = handler;
-	else
+	else {
 		kvm_posted_intr_wakeup_handler = dummy_handler;
+		synchronize_rcu();
+	}
 }
 EXPORT_SYMBOL_GPL(kvm_set_posted_intr_wakeup_handler);
 

arch/x86/kvm/Kconfig

···
 	 This option adds a R/W kVM module parameter 'mmu_audit', which allows
 	 auditing of KVM MMU events at runtime.
 
+config KVM_EXTERNAL_WRITE_TRACKING
+	bool
+
 endif # VIRTUALIZATION

arch/x86/kvm/cpuid.c

···
 	return ret;
 }
 
+/*
+ * This one is tied to SSB in the user API, and not
+ * visible in /proc/cpuinfo.
+ */
+#define KVM_X86_FEATURE_PSFD		(13*32+28) /* Predictive Store Forwarding Disable */
+
 #define F feature_bit
 #define SF(name) (boot_cpu_has(X86_FEATURE_##name) ? F(name) : 0)
+
 
 static inline struct kvm_cpuid_entry2 *cpuid_entry2_find(
 	struct kvm_cpuid_entry2 *entries, int nent, u32 function, u32 index)
···
 	kvm_cpu_cap_mask(CPUID_8000_0008_EBX,
 		F(CLZERO) | F(XSAVEERPTR) |
 		F(WBNOINVD) | F(AMD_IBPB) | F(AMD_IBRS) | F(AMD_SSBD) | F(VIRT_SSBD) |
-		F(AMD_SSB_NO) | F(AMD_STIBP) | F(AMD_STIBP_ALWAYS_ON)
+		F(AMD_SSB_NO) | F(AMD_STIBP) | F(AMD_STIBP_ALWAYS_ON) |
+		__feature_bit(KVM_X86_FEATURE_PSFD)
 	);
 
 	/*

arch/x86/kvm/emulate.c

···
 	if (enable_vmware_backdoor && is_vmware_backdoor_pmc(rcx))
 		return X86EMUL_CONTINUE;
 
+	/*
+	 * If CR4.PCE is set, the SDM requires CPL=0 or CR0.PE=0.  The CR0.PE
+	 * check however is unnecessary because CPL is always 0 outside
+	 * protected mode.
+	 */
 	if ((!(cr4 & X86_CR4_PCE) && ctxt->ops->cpl(ctxt)) ||
 	    ctxt->ops->check_pmc(ctxt, rcx))
 		return emulate_gp(ctxt, 0);

arch/x86/kvm/hyperv.c

···
 	if (!!auto_eoi_old == !!auto_eoi_new)
 		return;
 
-	mutex_lock(&vcpu->kvm->arch.apicv_update_lock);
+	down_write(&vcpu->kvm->arch.apicv_update_lock);
 
 	if (auto_eoi_new)
 		hv->synic_auto_eoi_used++;
···
 				   !hv->synic_auto_eoi_used,
 				   APICV_INHIBIT_REASON_HYPERV);
 
-	mutex_unlock(&vcpu->kvm->arch.apicv_update_lock);
+	up_write(&vcpu->kvm->arch.apicv_update_lock);
 }
 
 static int synic_set_sint(struct kvm_vcpu_hv_synic *synic, int sint,
···
 	int i;
 	gpa_t gpa;
 	struct kvm *kvm = vcpu->kvm;
-	struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
 	struct hv_tlb_flush_ex flush_ex;
 	struct hv_tlb_flush flush;
 	u64 vp_bitmap[KVM_HV_MAX_SPARSE_VCPU_SET_BITS];
···
 		}
 	}
 
-	cpumask_clear(&hv_vcpu->tlb_flush);
-
-	vcpu_mask = all_cpus ? NULL :
-		sparse_set_to_vcpu_mask(kvm, sparse_banks, valid_bank_mask,
-					vp_bitmap, vcpu_bitmap);
-
 	/*
 	 * vcpu->arch.cr3 may not be up-to-date for running vCPUs so we can't
 	 * analyze it here, flush TLB regardless of the specified address space.
 	 */
-	kvm_make_vcpus_request_mask(kvm, KVM_REQ_TLB_FLUSH_GUEST,
-				    NULL, vcpu_mask, &hv_vcpu->tlb_flush);
+	if (all_cpus) {
+		kvm_make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH_GUEST);
+	} else {
+		vcpu_mask = sparse_set_to_vcpu_mask(kvm, sparse_banks, valid_bank_mask,
+						    vp_bitmap, vcpu_bitmap);
+
+		kvm_make_vcpus_request_mask(kvm, KVM_REQ_TLB_FLUSH_GUEST,
+					    vcpu_mask);
+	}
 
 ret_success:
 	/* We always do full TLB flush, set 'Reps completed' = 'Rep Count' */

arch/x86/kvm/ioapic.c

···
 static void rtc_irq_eoi_tracking_reset(struct kvm_ioapic *ioapic)
 {
 	ioapic->rtc_status.pending_eoi = 0;
-	bitmap_zero(ioapic->rtc_status.dest_map.map, KVM_MAX_VCPU_ID + 1);
+	bitmap_zero(ioapic->rtc_status.dest_map.map, KVM_MAX_VCPU_IDS);
 }
 
 static void kvm_rtc_eoi_tracking_restore_all(struct kvm_ioapic *ioapic);

arch/x86/kvm/ioapic.h

···
 
 struct dest_map {
 	/* vcpu bitmap where IRQ has been sent */
-	DECLARE_BITMAP(map, KVM_MAX_VCPU_ID + 1);
+	DECLARE_BITMAP(map, KVM_MAX_VCPU_IDS);
 
 	/*
 	 * Vector sent to a given vcpu, only valid when
 	 * the vcpu's bit in map is set
 	 */
-	u8 vectors[KVM_MAX_VCPU_ID + 1];
+	u8 vectors[KVM_MAX_VCPU_IDS];
 };
 
 

arch/x86/kvm/mmu.h

···
 #define PT32_ROOT_LEVEL 2
 #define PT32E_ROOT_LEVEL 3
 
-#define KVM_MMU_CR4_ROLE_BITS (X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE | \
-			       X86_CR4_SMEP | X86_CR4_SMAP | X86_CR4_PKE | \
-			       X86_CR4_LA57)
+#define KVM_MMU_CR4_ROLE_BITS (X86_CR4_PSE | X86_CR4_PAE | X86_CR4_LA57 | \
+			       X86_CR4_SMEP | X86_CR4_SMAP | X86_CR4_PKE)
 
 #define KVM_MMU_CR0_ROLE_BITS (X86_CR0_PG | X86_CR0_WP)
 
···
 int kvm_mmu_load(struct kvm_vcpu *vcpu);
 void kvm_mmu_unload(struct kvm_vcpu *vcpu);
 void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
+void kvm_mmu_sync_prev_roots(struct kvm_vcpu *vcpu);
 
 static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
 {
···
 					  vcpu->arch.mmu->shadow_root_level);
 }
 
-int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-		       bool prefault);
+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 prefault)
+					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 (likely(vcpu->arch.mmu->page_fault == kvm_tdp_page_fault))
-		return kvm_tdp_page_fault(vcpu, cr2_or_gpa, err, prefault);
+	if (fault.is_tdp)
+		return kvm_tdp_page_fault(vcpu, &fault);
 #endif
-	return vcpu->arch.mmu->page_fault(vcpu, cr2_or_gpa, err, prefault);
+	return vcpu->arch.mmu->page_fault(vcpu, &fault);
 }
 
 /*
···
 int kvm_mmu_post_init_vm(struct kvm *kvm);
 void kvm_mmu_pre_destroy_vm(struct kvm *kvm);
 
-static inline bool kvm_memslots_have_rmaps(struct kvm *kvm)
+static inline bool kvm_shadow_root_allocated(struct kvm *kvm)
 {
 	/*
-	 * Read memslot_have_rmaps before rmap pointers.  Hence, threads reading
-	 * memslots_have_rmaps in any lock context are guaranteed to see the
-	 * pointers.  Pairs with smp_store_release in alloc_all_memslots_rmaps.
+	 * Read shadow_root_allocated before related pointers. Hence, threads
+	 * reading shadow_root_allocated in any lock context are guaranteed to
+	 * see the pointers. Pairs with smp_store_release in
+	 * mmu_first_shadow_root_alloc.
 	 */
-	return smp_load_acquire(&kvm->arch.memslots_have_rmaps);
+	return smp_load_acquire(&kvm->arch.shadow_root_allocated);
+}
+
+#ifdef CONFIG_X86_64
+static inline bool is_tdp_mmu_enabled(struct kvm *kvm) { return kvm->arch.tdp_mmu_enabled; }
+#else
+static inline bool is_tdp_mmu_enabled(struct kvm *kvm) { return false; }
+#endif
+
+static inline bool kvm_memslots_have_rmaps(struct kvm *kvm)
+{
+	return !is_tdp_mmu_enabled(kvm) || kvm_shadow_root_allocated(kvm);
 }
 
 static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)

arch/x86/kvm/mmu/mmu.c

···
 extern bool itlb_multihit_kvm_mitigation;
 
 int __read_mostly nx_huge_pages = -1;
+static uint __read_mostly nx_huge_pages_recovery_period_ms;
 #ifdef CONFIG_PREEMPT_RT
 /* Recovery can cause latency spikes, disable it for PREEMPT_RT.  */
 static uint __read_mostly nx_huge_pages_recovery_ratio = 0;
···
 #endif
 
 static int set_nx_huge_pages(const char *val, const struct kernel_param *kp);
-static int set_nx_huge_pages_recovery_ratio(const char *val, const struct kernel_param *kp);
+static int set_nx_huge_pages_recovery_param(const char *val, const struct kernel_param *kp);
 
 static const struct kernel_param_ops nx_huge_pages_ops = {
 	.set = set_nx_huge_pages,
 	.get = param_get_bool,
 };
 
-static const struct kernel_param_ops nx_huge_pages_recovery_ratio_ops = {
-	.set = set_nx_huge_pages_recovery_ratio,
+static const struct kernel_param_ops nx_huge_pages_recovery_param_ops = {
+	.set = set_nx_huge_pages_recovery_param,
 	.get = param_get_uint,
 };
 
 module_param_cb(nx_huge_pages, &nx_huge_pages_ops, &nx_huge_pages, 0644);
 __MODULE_PARM_TYPE(nx_huge_pages, "bool");
-module_param_cb(nx_huge_pages_recovery_ratio, &nx_huge_pages_recovery_ratio_ops,
+module_param_cb(nx_huge_pages_recovery_ratio, &nx_huge_pages_recovery_param_ops,
 		&nx_huge_pages_recovery_ratio, 0644);
 __MODULE_PARM_TYPE(nx_huge_pages_recovery_ratio, "uint");
+module_param_cb(nx_huge_pages_recovery_period_ms, &nx_huge_pages_recovery_param_ops,
+		&nx_huge_pages_recovery_period_ms, 0644);
+__MODULE_PARM_TYPE(nx_huge_pages_recovery_period_ms, "uint");
 
 static bool __read_mostly force_flush_and_sync_on_reuse;
 module_param_named(flush_on_reuse, force_flush_and_sync_on_reuse, bool, 0644);
···
 	return kvm_mmu_memory_cache_nr_free_objects(mc);
 }
 
-static int rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
-{
-	struct kvm_memory_slot *slot;
-	struct kvm_mmu_page *sp;
-	struct kvm_rmap_head *rmap_head;
-
-	sp = sptep_to_sp(spte);
-	kvm_mmu_page_set_gfn(sp, spte - sp->spt, gfn);
-	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
-	rmap_head = gfn_to_rmap(gfn, sp->role.level, slot);
-	return pte_list_add(vcpu, spte, rmap_head);
-}
-
-
 static void rmap_remove(struct kvm *kvm, u64 *spte)
 {
 	struct kvm_memslots *slots;
···
 	gfn = kvm_mmu_page_get_gfn(sp, spte - sp->spt);
 
 	/*
-	 * Unlike rmap_add and rmap_recycle, rmap_remove does not run in the
-	 * context of a vCPU so have to determine which memslots to use based
-	 * on context information in sp->role.
+	 * Unlike rmap_add, rmap_remove does not run in the context of a vCPU
+	 * so we have to determine which memslots to use based on context
+	 * information in sp->role.
 	 */
 	slots = kvm_memslots_for_spte_role(kvm, sp->role);
 
···
 
 #define RMAP_RECYCLE_THRESHOLD 1000
 
-static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
+static void rmap_add(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot,
+		     u64 *spte, gfn_t gfn)
 {
-	struct kvm_memory_slot *slot;
-	struct kvm_rmap_head *rmap_head;
 	struct kvm_mmu_page *sp;
+	struct kvm_rmap_head *rmap_head;
+	int rmap_count;
 
 	sp = sptep_to_sp(spte);
-	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
+	kvm_mmu_page_set_gfn(sp, spte - sp->spt, gfn);
 	rmap_head = gfn_to_rmap(gfn, sp->role.level, slot);
+	rmap_count = pte_list_add(vcpu, spte, rmap_head);
 
-	kvm_unmap_rmapp(vcpu->kvm, rmap_head, NULL, gfn, sp->role.level, __pte(0));
-	kvm_flush_remote_tlbs_with_address(vcpu->kvm, sp->gfn,
-			KVM_PAGES_PER_HPAGE(sp->role.level));
+	if (rmap_count > RMAP_RECYCLE_THRESHOLD) {
+		kvm_unmap_rmapp(vcpu->kvm, rmap_head, NULL, gfn, sp->role.level, __pte(0));
+		kvm_flush_remote_tlbs_with_address(
+				vcpu->kvm, sp->gfn, KVM_PAGES_PER_HPAGE(sp->role.level));
+	}
 }
 
 bool kvm_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range)
···
 static int nonpaging_sync_page(struct kvm_vcpu *vcpu,
 			       struct kvm_mmu_page *sp)
 {
-	return 0;
+	return -1;
 }
 
 #define KVM_PAGE_ARRAY_NR 16
···
 static bool kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 			 struct list_head *invalid_list)
 {
-	if (vcpu->arch.mmu->sync_page(vcpu, sp) == 0) {
+	int ret = vcpu->arch.mmu->sync_page(vcpu, sp);
+
+	if (ret < 0) {
 		kvm_mmu_prepare_zap_page(vcpu->kvm, sp, invalid_list);
 		return false;
 	}
 
-	return true;
+	return !!ret;
 }
 
 static bool kvm_mmu_remote_flush_or_zap(struct kvm *kvm,
···
 	else
 		kvm_flush_remote_tlbs(kvm);
 	return true;
-}
-
-static void kvm_mmu_flush_or_zap(struct kvm_vcpu *vcpu,
-				 struct list_head *invalid_list,
-				 bool remote_flush, bool local_flush)
-{
-	if (kvm_mmu_remote_flush_or_zap(vcpu->kvm, invalid_list, remote_flush))
-		return;
-
-	if (local_flush)
-		kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
 }
 
 #ifdef CONFIG_KVM_MMU_AUDIT
···
 			protected |= rmap_write_protect(vcpu, sp->gfn);
 
 		if (protected) {
-			kvm_flush_remote_tlbs(vcpu->kvm);
+			kvm_mmu_remote_flush_or_zap(vcpu->kvm, &invalid_list, true);
 			flush = false;
 		}
 
···
 			mmu_pages_clear_parents(&parents);
 		}
 		if (need_resched() || rwlock_needbreak(&vcpu->kvm->mmu_lock)) {
-			kvm_mmu_flush_or_zap(vcpu, &invalid_list, false, flush);
+			kvm_mmu_remote_flush_or_zap(vcpu->kvm, &invalid_list, flush);
 			if (!can_yield) {
 				kvm_make_request(KVM_REQ_MMU_SYNC, vcpu);
 				return -EINTR;
···
 		}
 	}
 
-	kvm_mmu_flush_or_zap(vcpu, &invalid_list, false, flush);
+	kvm_mmu_remote_flush_or_zap(vcpu->kvm, &invalid_list, flush);
 	return 0;
 }
 
···
 				break;
 
 			WARN_ON(!list_empty(&invalid_list));
-			kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
+			kvm_flush_remote_tlbs(vcpu->kvm);
 		}
 
 		__clear_sp_write_flooding_count(sp);
···
 static void __shadow_walk_next(struct kvm_shadow_walk_iterator *iterator,
 			       u64 spte)
 {
-	if (is_last_spte(spte, iterator->level)) {
+	if (!is_shadow_present_pte(spte) || is_last_spte(spte, iterator->level)) {
 		iterator->level = 0;
 		return;
 	}
···
  * were marked unsync (or if there is no shadow page), -EPERM if the SPTE must
  * be write-protected.
  */
-int mmu_try_to_unsync_pages(struct kvm_vcpu *vcpu, gfn_t gfn, bool can_unsync)
+int mmu_try_to_unsync_pages(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot,
+			    gfn_t gfn, bool can_unsync, bool prefetch)
 {
 	struct kvm_mmu_page *sp;
 	bool locked = false;
···
 	 * track machinery is used to write-protect upper-level shadow pages,
 	 * i.e. this guards the role.level == 4K assertion below!
 	 */
-	if (kvm_page_track_is_active(vcpu, gfn, KVM_PAGE_TRACK_WRITE))
+	if (kvm_slot_page_track_is_active(vcpu, slot, gfn, KVM_PAGE_TRACK_WRITE))
 		return -EPERM;
 
 	/*
···
 
 		if (sp->unsync)
 			continue;
+
+		if (prefetch)
+			return -EEXIST;
 
 		/*
 		 * TDP MMU page faults require an additional spinlock as they
···
 	 *     (sp->unsync = true)
 	 *
 	 * The write barrier below ensures that 1.1 happens before 1.2 and thus
-	 * the situation in 2.4 does not arise. The implicit barrier in 2.2
-	 * pairs with this write barrier.
+	 * the situation in 2.4 does not arise.  It pairs with the read barrier
+	 * in is_unsync_root(), placed between 2.1's load of SPTE.W and 2.3.
 	 */
 	smp_wmb();
 
 	return 0;
 }
 
-static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
-		    unsigned int pte_access, int level,
-		    gfn_t gfn, kvm_pfn_t pfn, bool speculative,
-		    bool can_unsync, bool host_writable)
+static int mmu_set_spte(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot,
+			u64 *sptep, unsigned int pte_access, gfn_t gfn,
+			kvm_pfn_t pfn, struct kvm_page_fault *fault)
 {
-	u64 spte;
-	struct kvm_mmu_page *sp;
-	int ret;
-
-	sp = sptep_to_sp(sptep);
-
-	ret = make_spte(vcpu, pte_access, level, gfn, pfn, *sptep, speculative,
-			can_unsync, host_writable, sp_ad_disabled(sp), &spte);
-
-	if (spte & PT_WRITABLE_MASK)
-		kvm_vcpu_mark_page_dirty(vcpu, gfn);
-
-	if (*sptep == spte)
-		ret |= SET_SPTE_SPURIOUS;
-	else if (mmu_spte_update(sptep, spte))
-		ret |= SET_SPTE_NEED_REMOTE_TLB_FLUSH;
-	return ret;
-}
-
-static int mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
-			unsigned int pte_access, bool write_fault, int level,
-			gfn_t gfn, kvm_pfn_t pfn, bool speculative,
-			bool host_writable)
-{
+	struct kvm_mmu_page *sp = sptep_to_sp(sptep);
+	int level = sp->role.level;
 	int was_rmapped = 0;
-	int rmap_count;
-	int set_spte_ret;
 	int ret = RET_PF_FIXED;
 	bool flush = false;
+	bool wrprot;
+	u64 spte;
+
+	/* Prefetching always gets a writable pfn.  */
+	bool host_writable = !fault || fault->map_writable;
+	bool prefetch = !fault || fault->prefetch;
+	bool write_fault = fault && fault->write;
 
 	pgprintk("%s: spte %llx write_fault %d gfn %llx\n", __func__,
 		 *sptep, write_fault, gfn);
···
 			was_rmapped = 1;
 	}
 
-	set_spte_ret = set_spte(vcpu, sptep, pte_access, level, gfn, pfn,
-				speculative, true, host_writable);
-	if (set_spte_ret & SET_SPTE_WRITE_PROTECTED_PT) {
-		if (write_fault)
-			ret = RET_PF_EMULATE;
-		kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
+	wrprot = make_spte(vcpu, sp, slot, pte_access, gfn, pfn, *sptep, prefetch,
+			   true, host_writable, &spte);
+
+	if (*sptep == spte) {
+		ret = RET_PF_SPURIOUS;
+	} else {
+		trace_kvm_mmu_set_spte(level, gfn, sptep);
+		flush |= mmu_spte_update(sptep, spte);
 	}
 
-	if (set_spte_ret & SET_SPTE_NEED_REMOTE_TLB_FLUSH || flush)
+	if (wrprot) {
+		if (write_fault)
+			ret = RET_PF_EMULATE;
+	}
+
+	if (flush)
 		kvm_flush_remote_tlbs_with_address(vcpu->kvm, gfn,
 				KVM_PAGES_PER_HPAGE(level));
 
-	/*
-	 * The fault is fully spurious if and only if the new SPTE and old SPTE
-	 * are identical, and emulation is not required.
-	 */
-	if ((set_spte_ret & SET_SPTE_SPURIOUS) && ret == RET_PF_FIXED) {
-		WARN_ON_ONCE(!was_rmapped);
-		return RET_PF_SPURIOUS;
-	}
-
 	pgprintk("%s: setting spte %llx\n", __func__, *sptep);
-	trace_kvm_mmu_set_spte(level, gfn, sptep);
 
 	if (!was_rmapped) {
+		WARN_ON_ONCE(ret == RET_PF_SPURIOUS);
 		kvm_update_page_stats(vcpu->kvm, level, 1);
-		rmap_count = rmap_add(vcpu, sptep, gfn);
-		if (rmap_count > RMAP_RECYCLE_THRESHOLD)
-			rmap_recycle(vcpu, sptep, gfn);
+		rmap_add(vcpu, slot, sptep, gfn);
 	}
 
 	return ret;
-}
-
-static kvm_pfn_t pte_prefetch_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn,
-				     bool no_dirty_log)
-{
-	struct kvm_memory_slot *slot;
-
-	slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, no_dirty_log);
-	if (!slot)
-		return KVM_PFN_ERR_FAULT;
-
-	return gfn_to_pfn_memslot_atomic(slot, gfn);
 }
 
 static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu,
···
 		return -1;
 
 	for (i = 0; i < ret; i++, gfn++, start++) {
-		mmu_set_spte(vcpu, start, access, false, sp->role.level, gfn,
-			     page_to_pfn(pages[i]), true, true);
+		mmu_set_spte(vcpu, slot, start, access, gfn,
+			     page_to_pfn(pages[i]), NULL);
 		put_page(pages[i]);
 	}
 
···
 			if (!start)
 				continue;
 			if (direct_pte_prefetch_many(vcpu, sp, start, spte) < 0)
-				break;
+				return;
 			start = NULL;
 		} else if (!start)
 			start = spte;
 	}
+	if (start)
+		direct_pte_prefetch_many(vcpu, sp, start, spte);
 }
 
 static void direct_pte_prefetch(struct kvm_vcpu *vcpu, u64 *sptep)
···
 	return min(host_level, max_level);
 }
 
-int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn,
-			    int max_level, kvm_pfn_t *pfnp,
-			    bool huge_page_disallowed, int *req_level)
+void kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
-	struct kvm_memory_slot *slot;
-	kvm_pfn_t pfn = *pfnp;
+	struct kvm_memory_slot *slot = fault->slot;
 	kvm_pfn_t mask;
-	int level;
 
-	*req_level = PG_LEVEL_4K;
+	fault->huge_page_disallowed = fault->exec && fault->nx_huge_page_workaround_enabled;
 
-	if (unlikely(max_level == PG_LEVEL_4K))
-		return PG_LEVEL_4K;
+	if (unlikely(fault->max_level == PG_LEVEL_4K))
+		return;
 
-	if (is_error_noslot_pfn(pfn) || kvm_is_reserved_pfn(pfn))
-		return PG_LEVEL_4K;
+	if (is_error_noslot_pfn(fault->pfn) || kvm_is_reserved_pfn(fault->pfn))
+		return;
 
-	slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, true);
-	if (!slot)
-		return PG_LEVEL_4K;
+	if (kvm_slot_dirty_track_enabled(slot))
+		return;
 
 	/*
 	 * Enforce the iTLB multihit workaround after capturing the requested
 	 * level, which will be used to do precise, accurate accounting.
 	 */
-	*req_level = level = kvm_mmu_max_mapping_level(vcpu->kvm, slot, gfn, pfn, max_level);
-	if (level == PG_LEVEL_4K || huge_page_disallowed)
-		return PG_LEVEL_4K;
+	fault->req_level = kvm_mmu_max_mapping_level(vcpu->kvm, slot,
+						     fault->gfn, fault->pfn,
+						     fault->max_level);
+	if (fault->req_level == PG_LEVEL_4K || fault->huge_page_disallowed)
+		return;
 
 	/*
 	 * mmu_notifier_retry() was successful and mmu_lock is held, so
 	 * the pmd can't be split from under us.
 	 */
-	mask = KVM_PAGES_PER_HPAGE(level) - 1;
-	VM_BUG_ON((gfn & mask) != (pfn & mask));
-	*pfnp = pfn & ~mask;
-
-	return level;
+	fault->goal_level = fault->req_level;
+	mask = KVM_PAGES_PER_HPAGE(fault->goal_level) - 1;
+	VM_BUG_ON((fault->gfn & mask) != (fault->pfn & mask));
+	fault->pfn &= ~mask;
 }
 
-void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level,
-				kvm_pfn_t *pfnp, int *goal_levelp)
+void disallowed_hugepage_adjust(struct kvm_page_fault *fault, u64 spte, int cur_level)
 {
-	int level = *goal_levelp;
-
-	if (cur_level == level && level > PG_LEVEL_4K &&
+	if (cur_level > PG_LEVEL_4K &&
+	    cur_level == fault->goal_level &&
 	    is_shadow_present_pte(spte) &&
 	    !is_large_pte(spte)) {
 		/*
···
 		 * patching back for them into pfn the next 9 bits of
 		 * the address.
 		 */
-		u64 page_mask = KVM_PAGES_PER_HPAGE(level) -
-				KVM_PAGES_PER_HPAGE(level - 1);
-		*pfnp |= gfn & page_mask;
-		(*goal_levelp)--;
+		u64 page_mask = KVM_PAGES_PER_HPAGE(cur_level) -
+				KVM_PAGES_PER_HPAGE(cur_level - 1);
+		fault->pfn |= fault->gfn & page_mask;
+		fault->goal_level--;
 	}
 }
 
-static int __direct_map(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-			int map_writable, int max_level, kvm_pfn_t pfn,
-			bool prefault, bool is_tdp)
+static int __direct_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
-	bool nx_huge_page_workaround_enabled = is_nx_huge_page_enabled();
-	bool write = error_code & PFERR_WRITE_MASK;
-	bool exec = error_code & PFERR_FETCH_MASK;
-	bool huge_page_disallowed = exec && nx_huge_page_workaround_enabled;
 	struct kvm_shadow_walk_iterator it;
 	struct kvm_mmu_page *sp;
-	int level, req_level, ret;
-	gfn_t gfn = gpa >> PAGE_SHIFT;
-	gfn_t base_gfn = gfn;
+	int ret;
+	gfn_t base_gfn = fault->gfn;
 
-	level = kvm_mmu_hugepage_adjust(vcpu, gfn, max_level, &pfn,
-					huge_page_disallowed, &req_level);
+	kvm_mmu_hugepage_adjust(vcpu, fault);
 
-	trace_kvm_mmu_spte_requested(gpa, level, pfn);
-	for_each_shadow_entry(vcpu, gpa, it) {
+	trace_kvm_mmu_spte_requested(fault);
+	for_each_shadow_entry(vcpu, fault->addr, it) {
 		/*
 		 * We cannot overwrite existing page tables with an NX
 		 * large page, as the leaf could be executable.
 		 */
-		if (nx_huge_page_workaround_enabled)
-			disallowed_hugepage_adjust(*it.sptep, gfn, it.level,
-						   &pfn, &level);
+		if (fault->nx_huge_page_workaround_enabled)
+			disallowed_hugepage_adjust(fault, *it.sptep, it.level);
 
-		base_gfn = gfn & ~(KVM_PAGES_PER_HPAGE(it.level) - 1);
-		if (it.level == level)
+		base_gfn = fault->gfn & ~(KVM_PAGES_PER_HPAGE(it.level) - 1);
+		if (it.level == fault->goal_level)
 			break;
 
 		drop_large_spte(vcpu, it.sptep);
···
 				      it.level - 1, true, ACC_ALL);
 
 		link_shadow_page(vcpu, it.sptep, sp);
-		if (is_tdp && huge_page_disallowed &&
-		    req_level >= it.level)
+		if (fault->is_tdp && fault->huge_page_disallowed &&
+		    fault->req_level >= it.level)
 			account_huge_nx_page(vcpu->kvm, sp);
 	}
 
-	ret = mmu_set_spte(vcpu, it.sptep, ACC_ALL,
-			   write, level, base_gfn, pfn, prefault,
-			   map_writable);
+	if (WARN_ON_ONCE(it.level != fault->goal_level))
+		return -EFAULT;
+
+	ret = mmu_set_spte(vcpu, fault->slot, it.sptep, ACC_ALL,
+			   base_gfn, fault->pfn, fault);
 	if (ret == RET_PF_SPURIOUS)
 		return ret;
 
···
 	return -EFAULT;
 }
 
-static bool handle_abnormal_pfn(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn,
-				kvm_pfn_t pfn, unsigned int access,
-				int *ret_val)
+static bool handle_abnormal_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
+				unsigned int access, int *ret_val)
 {
 	/* The pfn is invalid, report the error! */
-	if (unlikely(is_error_pfn(pfn))) {
-		*ret_val = kvm_handle_bad_page(vcpu, gfn, pfn);
+	if (unlikely(is_error_pfn(fault->pfn))) {
+		*ret_val = kvm_handle_bad_page(vcpu, fault->gfn, fault->pfn);
 		return true;
 	}
 
-	if (unlikely(is_noslot_pfn(pfn))) {
-		vcpu_cache_mmio_info(vcpu, gva, gfn,
+	if (unlikely(!fault->slot)) {
+		gva_t gva = fault->is_tdp ? 0 : fault->addr;
+
+		vcpu_cache_mmio_info(vcpu, gva, fault->gfn,
 				     access & shadow_mmio_access_mask);
 		/*
 		 * If MMIO caching is disabled, emulate immediately without
···
 	return false;
 }
 
-static bool page_fault_can_be_fast(u32 error_code)
+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.
 	 */
-	if (unlikely(error_code & PFERR_RSVD_MASK))
+	if (fault->rsvd)
 		return false;
 
 	/* See if the page fault is due to an NX violation */
-	if (unlikely(((error_code & (PFERR_FETCH_MASK | PFERR_PRESENT_MASK))
-		      == (PFERR_FETCH_MASK | PFERR_PRESENT_MASK))))
+	if (unlikely(fault->exec && fault->present))
 		return false;
 
 	/*
···
 	 * accesses to a present page.
 	 */
 
-	return shadow_acc_track_mask != 0 ||
-	       ((error_code & (PFERR_WRITE_MASK | PFERR_PRESENT_MASK))
-		== (PFERR_WRITE_MASK | PFERR_PRESENT_MASK));
+	return shadow_acc_track_mask != 0 || (fault->write && fault->present);
 }
 
 /*
···
  * someone else modified the SPTE from its original value.
  */
 static bool
-fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
+fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
 			u64 *sptep, u64 old_spte, u64 new_spte)
 {
-	gfn_t gfn;
-
-	WARN_ON(!sp->role.direct);
-
 	/*
 	 * Theoretically we could also set dirty bit (and flush TLB) here in
 	 * order to eliminate unnecessary PML logging. See comments in
···
 	if (cmpxchg64(sptep, old_spte, new_spte) != old_spte)
 		return false;
 
-	if (is_writable_pte(new_spte) && !is_writable_pte(old_spte)) {
-		/*
-		 * The gfn of direct spte is stable since it is
-		 * calculated by sp->gfn.
-		 */
-		gfn = kvm_mmu_page_get_gfn(sp, sptep - sp->spt);
-		kvm_vcpu_mark_page_dirty(vcpu, gfn);
-	}
+	if (is_writable_pte(new_spte) && !is_writable_pte(old_spte))
+		mark_page_dirty_in_slot(vcpu->kvm, fault->slot, fault->gfn);
 
 	return true;
 }
 
-static bool is_access_allowed(u32 fault_err_code, u64 spte)
+static bool is_access_allowed(struct kvm_page_fault *fault, u64 spte)
 {
-	if (fault_err_code & PFERR_FETCH_MASK)
+	if (fault->exec)
 		return is_executable_pte(spte);
 
-	if (fault_err_code & PFERR_WRITE_MASK)
+	if (fault->write)
 		return is_writable_pte(spte);
 
 	/* Fault was on Read access */
···
 	for_each_shadow_entry_lockless(vcpu, gpa, iterator, old_spte) {
 		sptep = iterator.sptep;
 		*spte = old_spte;
-
-		if (!is_shadow_present_pte(old_spte))
-			break;
 	}
 
 	return sptep;
···
 /*
  * Returns one of RET_PF_INVALID, RET_PF_FIXED or RET_PF_SPURIOUS.
  */
-static int fast_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code)
+static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
 	struct kvm_mmu_page *sp;
 	int ret = RET_PF_INVALID;
···
 	u64 *sptep = NULL;
 	uint retry_count = 0;
 
-	if (!page_fault_can_be_fast(error_code))
+	if (!page_fault_can_be_fast(fault))
 		return ret;
 
 	walk_shadow_page_lockless_begin(vcpu);
···
 		u64 new_spte;
 
 		if (is_tdp_mmu(vcpu->arch.mmu))
-			sptep = kvm_tdp_mmu_fast_pf_get_last_sptep(vcpu, gpa, &spte);
+			sptep = kvm_tdp_mmu_fast_pf_get_last_sptep(vcpu, fault->addr, &spte);
 		else
-			sptep = fast_pf_get_last_sptep(vcpu, gpa, &spte);
+			sptep = fast_pf_get_last_sptep(vcpu, fault->addr, &spte);
 
 		if (!is_shadow_present_pte(spte))
 			break;
···
 		 * Need not check the access of upper level table entries since
 		 * they are always ACC_ALL.
 		 */
-		if (is_access_allowed(error_code, spte)) {
+		if (is_access_allowed(fault, spte)) {
 			ret = RET_PF_SPURIOUS;
 			break;
 		}
···
 		 * be removed in the fast path only if the SPTE was
 		 * write-protected for dirty-logging or access tracking.
 		 */
-		if ((error_code & PFERR_WRITE_MASK) &&
+		if (fault->write &&
 		    spte_can_locklessly_be_made_writable(spte)) {
 			new_spte |= PT_WRITABLE_MASK;
 
···
 
 		/* Verify that the fault can be handled in the fast path */
 		if (new_spte == spte ||
-		    !is_access_allowed(error_code, new_spte))
+		    !is_access_allowed(fault, new_spte))
 			break;
 
 		/*
···
 		 * since the gfn is not stable for indirect shadow page. See
 		 * Documentation/virt/kvm/locking.rst to get more detail.
 		 */
-		if (fast_pf_fix_direct_spte(vcpu, sp, sptep, spte, new_spte)) {
+		if (fast_pf_fix_direct_spte(vcpu, fault, sptep, spte, new_spte)) {
 			ret = RET_PF_FIXED;
 			break;
 		}
···
 
 	} while (true);
 
-	trace_fast_page_fault(vcpu, gpa, error_code, sptep, spte, ret);
+	trace_fast_page_fault(vcpu, fault, sptep, spte, ret);
 	walk_shadow_page_lockless_end(vcpu);
 
 	return ret;
···
 	return r;
 }
 
+static int mmu_first_shadow_root_alloc(struct kvm *kvm)
+{
+	struct kvm_memslots *slots;
+	struct kvm_memory_slot *slot;
+	int r = 0, i;
+
+	/*
+	 * Check if this is the first shadow root being allocated before
+	 * taking the lock.
+	 */
+	if (kvm_shadow_root_allocated(kvm))
+		return 0;
+
+	mutex_lock(&kvm->slots_arch_lock);
+
+	/* Recheck, under the lock, whether this is the first shadow root. */
+	if (kvm_shadow_root_allocated(kvm))
+		goto out_unlock;
+
+	/*
+	 * Check if anything actually needs to be allocated, e.g. all metadata
+	 * will be allocated upfront if TDP is disabled.
+	 */
+	if (kvm_memslots_have_rmaps(kvm) &&
+	    kvm_page_track_write_tracking_enabled(kvm))
+		goto out_success;
+
+	for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) {
+		slots = __kvm_memslots(kvm, i);
+		kvm_for_each_memslot(slot, slots) {
+			/*
+			 * Both of these functions are no-ops if the target is
+			 * already allocated, so unconditionally calling both
+			 * is safe.  Intentionally do NOT free allocations on
+			 * failure to avoid having to track which allocations
+			 * were made now versus when the memslot was created.
+			 * The metadata is guaranteed to be freed when the slot
+			 * is freed, and will be kept/used if userspace retries
+			 * KVM_RUN instead of killing the VM.
+			 */
+			r = memslot_rmap_alloc(slot, slot->npages);
+			if (r)
+				goto out_unlock;
+			r = kvm_page_track_write_tracking_alloc(slot);
+			if (r)
+				goto out_unlock;
+		}
+	}
+
+	/*
+	 * Ensure that shadow_root_allocated becomes true strictly after
+	 * all the related pointers are set.
+	 */
+out_success:
+	smp_store_release(&kvm->arch.shadow_root_allocated, true);
+
+out_unlock:
+	mutex_unlock(&kvm->slots_arch_lock);
+	return r;
+}
+
 static int mmu_alloc_shadow_roots(struct kvm_vcpu *vcpu)
 {
 	struct kvm_mmu *mmu = vcpu->arch.mmu;
···
 		}
 	}
 
-	r = alloc_all_memslots_rmaps(vcpu->kvm);
+	r = mmu_first_shadow_root_alloc(vcpu->kvm);
 	if (r)
 		return r;
 
···
 #endif
 }
 
+static bool is_unsync_root(hpa_t root)
+{
+	struct kvm_mmu_page *sp;
+
+	if (!VALID_PAGE(root))
+		return false;
+
+	/*
+	 * The read barrier orders the CPU's read of SPTE.W during the page table
+	 * walk before the reads of sp->unsync/sp->unsync_children here.
+	 *
+	 * Even if another CPU was marking the SP as unsync-ed simultaneously,
+	 * any guest page table changes are not guaranteed to be visible anyway
+	 * until this VCPU issues a TLB flush strictly after those changes are
+	 * made.  We only need to ensure that the other CPU sets these flags
+	 * before any actual changes to the page tables are made.  The comments
+	 * in mmu_try_to_unsync_pages() describe what could go wrong if this
+	 * requirement isn't satisfied.
+	 */
+	smp_rmb();
+	sp = to_shadow_page(root);
+	if (sp->unsync || sp->unsync_children)
+		return true;
+
+	return false;
+}
+
 void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu)
 {
 	int i;
···
 		hpa_t root = vcpu->arch.mmu->root_hpa;
 		sp = to_shadow_page(root);
 
-		/*
-		 * Even if another CPU was marking the SP as unsync-ed
-		 * simultaneously, any guest page table changes are not
-		 * guaranteed to be visible anyway until this VCPU issues a TLB
-		 * flush strictly after those changes are made. We only need to
-		 * ensure that the other CPU sets these flags before any actual
-		 * changes to the page tables are made. The comments in
-		 * mmu_try_to_unsync_pages() describe what could go wrong if
-		 * this requirement isn't satisfied.
-		 */
-		if (!smp_load_acquire(&sp->unsync) &&
-		    !smp_load_acquire(&sp->unsync_children))
+		if (!is_unsync_root(root))
 			return;
 
 		write_lock(&vcpu->kvm->mmu_lock);
···
 
 	kvm_mmu_audit(vcpu, AUDIT_POST_SYNC);
 	write_unlock(&vcpu->kvm->mmu_lock);
+}
+
+void kvm_mmu_sync_prev_roots(struct kvm_vcpu *vcpu)
+{
+	unsigned long roots_to_free = 0;
+	int i;
+
+	for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++)
+		if (is_unsync_root(vcpu->arch.mmu->prev_roots[i].hpa))
+			roots_to_free |= KVM_MMU_ROOT_PREVIOUS(i);
+
+	/* sync prev_roots by simply freeing them */
+	kvm_mmu_free_roots(vcpu, vcpu->arch.mmu, roots_to_free);
 }
 
 static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gpa_t vaddr,
···
 		spte = mmu_spte_get_lockless(iterator.sptep);
 
 		sptes[leaf] = spte;
-
-		if (!is_shadow_present_pte(spte))
-			break;
 	}
 
 	return leaf;
···
 }
 
 static bool page_fault_handle_page_track(struct kvm_vcpu *vcpu,
-					 u32 error_code, gfn_t gfn)
+					 struct kvm_page_fault *fault)
 {
-	if (unlikely(error_code & PFERR_RSVD_MASK))
+	if (unlikely(fault->rsvd))
 		return false;
 
-	if (!(error_code & PFERR_PRESENT_MASK) ||
-	      !(error_code & PFERR_WRITE_MASK))
+	if (!fault->present || !fault->write)
 		return false;
 
 	/*
 	 * guest is writing the page which is write tracked which can
 	 * not be fixed by page fault handler.
 	 */
-	if (kvm_page_track_is_active(vcpu, gfn, KVM_PAGE_TRACK_WRITE))
+	if (kvm_slot_page_track_is_active(vcpu, fault->slot, fault->gfn, KVM_PAGE_TRACK_WRITE))
 		return true;
 
 	return false;
···
 	u64 spte;
 
 	walk_shadow_page_lockless_begin(vcpu);
-	for_each_shadow_entry_lockless(vcpu, addr, iterator, spte) {
+	for_each_shadow_entry_lockless(vcpu, addr, iterator, spte)
 		clear_sp_write_flooding_count(iterator.sptep);
-		if (!is_shadow_present_pte(spte))
-			break;
-	}
 	walk_shadow_page_lockless_end(vcpu);
 }
 
···
 				  kvm_vcpu_gfn_to_hva(vcpu, gfn), &arch);
 }
 
-static bool kvm_faultin_pfn(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
-			 gpa_t cr2_or_gpa, kvm_pfn_t *pfn, hva_t *hva,
-			 bool write, bool *writable, int *r)
+static bool kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, int *r)
 {
-	struct kvm_memory_slot *slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
+	struct kvm_memory_slot *slot = fault->slot;
 	bool async;
 
 	/*
···
 	if (!kvm_is_visible_memslot(slot)) {
 		/* Don't expose private memslots to L2. */
 		if (is_guest_mode(vcpu)) {
-			*pfn = KVM_PFN_NOSLOT;
-			*writable = false;
+			fault->slot = NULL;
+			fault->pfn = KVM_PFN_NOSLOT;
+			fault->map_writable = false;
 			return false;
 		}
 		/*
···
 	}
 
 	async = false;
-	*pfn = __gfn_to_pfn_memslot(slot, gfn, false, &async,
-				    write, writable, hva);
+	fault->pfn = __gfn_to_pfn_memslot(slot, fault->gfn, false, &async,
+					  fault->write, &fault->map_writable,
+					  &fault->hva);
 	if (!async)
 		return false; /* *pfn has correct page already */
 
-	if (!prefault && kvm_can_do_async_pf(vcpu)) {
-		trace_kvm_try_async_get_page(cr2_or_gpa, gfn);
-		if (kvm_find_async_pf_gfn(vcpu, gfn)) {
-			trace_kvm_async_pf_doublefault(cr2_or_gpa, gfn);
+	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, cr2_or_gpa, gfn))
+		} else if (kvm_arch_setup_async_pf(vcpu, fault->addr, fault->gfn))
 			goto out_retry;
 	}
 
-	*pfn = __gfn_to_pfn_memslot(slot, gfn, false, NULL,
-				    write, writable, hva);
+	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;
 }
 
-static int direct_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-			     bool prefault, int max_level, bool is_tdp)
+static int direct_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
 	bool is_tdp_mmu_fault = is_tdp_mmu(vcpu->arch.mmu);
-	bool write = error_code & PFERR_WRITE_MASK;
-	bool map_writable;
 
-	gfn_t gfn = gpa >> PAGE_SHIFT;
 	unsigned long mmu_seq;
-	kvm_pfn_t pfn;
-	hva_t hva;
 	int r;
 
-	if (page_fault_handle_page_track(vcpu, error_code, gfn))
+	fault->gfn = fault->addr >> PAGE_SHIFT;
+	fault->slot = kvm_vcpu_gfn_to_memslot(vcpu, fault->gfn);
+
+	if (page_fault_handle_page_track(vcpu, fault))
 		return RET_PF_EMULATE;
 
-	r = fast_page_fault(vcpu, gpa, error_code);
+	r = fast_page_fault(vcpu, fault);
 	if (r != RET_PF_INVALID)
 		return r;
 
···
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (kvm_faultin_pfn(vcpu, prefault, gfn, gpa, &pfn, &hva,
-			 write, &map_writable, &r))
+	if (kvm_faultin_pfn(vcpu, fault, &r))
 		return r;
 
-	if (handle_abnormal_pfn(vcpu, is_tdp ? 0 : gpa, gfn, pfn, ACC_ALL, &r))
+	if (handle_abnormal_pfn(vcpu, fault, ACC_ALL, &r))
 		return r;
 
 	r = RET_PF_RETRY;
···
 	else
 		write_lock(&vcpu->kvm->mmu_lock);
 
-	if (!is_noslot_pfn(pfn) && mmu_notifier_retry_hva(vcpu->kvm, mmu_seq, hva))
+	if (fault->slot && mmu_notifier_retry_hva(vcpu->kvm, mmu_seq, fault->hva))
 		goto out_unlock;
 	r = make_mmu_pages_available(vcpu);
 	if (r)
 		goto out_unlock;
 
 	if (is_tdp_mmu_fault)
-		r = kvm_tdp_mmu_map(vcpu, gpa, error_code, map_writable, max_level,
-				    pfn, prefault);
+		r = kvm_tdp_mmu_map(vcpu, fault);
 	else
-		r = __direct_map(vcpu, gpa, error_code, map_writable, max_level, pfn,
-				 prefault, is_tdp);
+		r = __direct_map(vcpu, fault);
 
 out_unlock:
 	if (is_tdp_mmu_fault)
 		read_unlock(&vcpu->kvm->mmu_lock);
 	else
 		write_unlock(&vcpu->kvm->mmu_lock);
-	kvm_release_pfn_clean(pfn);
+	kvm_release_pfn_clean(fault->pfn);
 	return r;
 }
 
-static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa,
-				u32 error_code, bool prefault)
+static int nonpaging_page_fault(struct kvm_vcpu *vcpu,
+				struct kvm_page_fault *fault)
 {
-	pgprintk("%s: gva %lx error %x\n", __func__, gpa, error_code);
+	pgprintk("%s: gva %lx error %x\n", __func__, fault->addr, fault->error_code);
 
 	/* This path builds a PAE pagetable, we can map 2mb pages at maximum. */
-	return direct_page_fault(vcpu, gpa & PAGE_MASK, error_code, prefault,
-				 PG_LEVEL_2M, false);
+	fault->max_level = PG_LEVEL_2M;
+	return direct_page_fault(vcpu, fault);
 }
 
 int kvm_handle_page_fault(struct kvm_vcpu *vcpu, u64 error_code,
···
 }
 EXPORT_SYMBOL_GPL(kvm_handle_page_fault);
 
-int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-		       bool prefault)
+int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
-	int max_level;
-
-	for (max_level = KVM_MAX_HUGEPAGE_LEVEL;
-	     max_level > PG_LEVEL_4K;
-	     max_level--) {
-		int page_num = KVM_PAGES_PER_HPAGE(max_level);
-		gfn_t base = (gpa >> PAGE_SHIFT) & ~(page_num - 1);
+	while (fault->max_level > PG_LEVEL_4K) {
+		int page_num = KVM_PAGES_PER_HPAGE(fault->max_level);
+		gfn_t base = (fault->addr >> PAGE_SHIFT) & ~(page_num - 1);
 
 		if (kvm_mtrr_check_gfn_range_consistency(vcpu, base, page_num))
 			break;
+
+		--fault->max_level;
 	}
 
-	return direct_page_fault(vcpu, gpa, error_code, prefault,
-				 max_level, true);
+	return direct_page_fault(vcpu, fault);
 }
 
 static void nonpaging_init_context(struct kvm_mmu *context)
···
 }
 
 static bool sync_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, gfn_t gfn,
-			   unsigned int access, int *nr_present)
+			   unsigned int access)
 {
 	if (unlikely(is_mmio_spte(*sptep))) {
 		if (gfn != get_mmio_spte_gfn(*sptep)) {
···
 			return true;
 		}
 
-		(*nr_present)++;
 		mark_mmio_spte(vcpu, sptep, gfn, access);
 		return true;
 	}
···
 	LIST_HEAD(invalid_list);
 	u64 entry, gentry, *spte;
 	int npte;
-	bool remote_flush, local_flush;
+	bool flush = false;
 
 	/*
 	 * If we don't have indirect shadow pages, it means no page is
···
 	 */
 	if (!READ_ONCE(vcpu->kvm->arch.indirect_shadow_pages))
 		return;
-
-	remote_flush = local_flush = false;
 
 	pgprintk("%s: gpa %llx bytes %d\n", __func__, gpa, bytes);
 
···
 		if (!spte)
 			continue;
 
-		local_flush = true;
 		while (npte--) {
 			entry = *spte;
 			mmu_page_zap_pte(vcpu->kvm, sp, spte, NULL);
 			if (gentry && sp->role.level != PG_LEVEL_4K)
 				++vcpu->kvm->stat.mmu_pde_zapped;
 			if (need_remote_flush(entry, *spte))
-				remote_flush = true;
+				flush = true;
 			++spte;
 		}
 	}
-	kvm_mmu_flush_or_zap(vcpu, &invalid_list, remote_flush, local_flush);
+	kvm_mmu_remote_flush_or_zap(vcpu->kvm, &invalid_list, flush);
 	kvm_mmu_audit(vcpu, AUDIT_POST_PTE_WRITE);
 	write_unlock(&vcpu->kvm->mmu_lock);
 }
···
 }
 
 static __always_inline bool
-slot_handle_leaf(struct kvm *kvm, const struct kvm_memory_slot *memslot,
-		 slot_level_handler fn, bool flush_on_yield)
+slot_handle_level_4k(struct kvm *kvm, const struct kvm_memory_slot *memslot,
+		     slot_level_handler fn, bool flush_on_yield)
 {
 	return slot_handle_level(kvm, memslot, fn, PG_LEVEL_4K,
 				 PG_LEVEL_4K, flush_on_yield);
···
 
 	spin_lock_init(&kvm->arch.mmu_unsync_pages_lock);
 
-	if (!kvm_mmu_init_tdp_mmu(kvm))
-		/*
-		 * No smp_load/store wrappers needed here as we are in
-		 * VM init and there cannot be any memslots / other threads
-		 * accessing this struct kvm yet.
-		 */
-		kvm->arch.memslots_have_rmaps = true;
+	kvm_mmu_init_tdp_mmu(kvm);
 
 	node->track_write = kvm_mmu_pte_write;
 	node->track_flush_slot = kvm_mmu_invalidate_zap_pages_in_memslot;
···
 	kvm_mmu_uninit_tdp_mmu(kvm);
 }
 
+static bool __kvm_zap_rmaps(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_end)
+{
+	const struct kvm_memory_slot *memslot;
+	struct kvm_memslots *slots;
+	bool flush = false;
+	gfn_t start, end;
+	int i;
+
+	if (!kvm_memslots_have_rmaps(kvm))
+		return flush;
+
+	for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) {
+		slots = __kvm_memslots(kvm, i);
+		kvm_for_each_memslot(memslot, slots) {
+			start = max(gfn_start, memslot->base_gfn);
+			end = min(gfn_end, memslot->base_gfn + memslot->npages);
+			if (start >= end)
+				continue;
+
+			flush = slot_handle_level_range(kvm, memslot, kvm_zap_rmapp,
+							PG_LEVEL_4K, KVM_MAX_HUGEPAGE_LEVEL,
+							start, end - 1, true, flush);
+		}
+	}
+
+	return flush;
+}
+
 /*
  * Invalidate (zap) SPTEs that cover GFNs from gfn_start and up to gfn_end
  * (not including it)
  */
 void kvm_zap_gfn_range(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_end)
 {
-	struct kvm_memslots *slots;
-	struct kvm_memory_slot *memslot;
+	bool flush;
 	int i;
-	bool flush = false;
 
 	write_lock(&kvm->mmu_lock);
 
 	kvm_inc_notifier_count(kvm, gfn_start, gfn_end);
 
-	if (kvm_memslots_have_rmaps(kvm)) {
-		for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) {
-			slots = __kvm_memslots(kvm, i);
-			kvm_for_each_memslot(memslot, slots) {
-				gfn_t start, end;
-
-				start = max(gfn_start, memslot->base_gfn);
-				end = min(gfn_end, memslot->base_gfn + memslot->npages);
-				if (start >= end)
-					continue;
-
-				flush = slot_handle_level_range(kvm,
-						(const struct kvm_memory_slot *) memslot,
-						kvm_zap_rmapp, PG_LEVEL_4K,
-						KVM_MAX_HUGEPAGE_LEVEL, start,
-						end - 1, true, flush);
-			}
-		}
-		if (flush)
-			kvm_flush_remote_tlbs_with_address(kvm, gfn_start,
-							   gfn_end - gfn_start);
-	}
+	flush = __kvm_zap_rmaps(kvm, gfn_start, gfn_end);
 
 	if (is_tdp_mmu_enabled(kvm)) {
 		for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++)
 			flush = kvm_tdp_mmu_zap_gfn_range(kvm, i, gfn_start,
 							  gfn_end, flush);
-		if (flush)
-			kvm_flush_remote_tlbs_with_address(kvm, gfn_start,
-							   gfn_end - gfn_start);
 	}
 
 	if (flush)
-		kvm_flush_remote_tlbs_with_address(kvm, gfn_start, gfn_end);
+		kvm_flush_remote_tlbs_with_address(kvm, gfn_start,
+						   gfn_end - gfn_start);
 
 	kvm_dec_notifier_count(kvm, gfn_start, gfn_end);
 
···
 
 	if (kvm_memslots_have_rmaps(kvm)) {
 		write_lock(&kvm->mmu_lock);
-		flush = slot_handle_leaf(kvm, slot, kvm_mmu_zap_collapsible_spte, true);
+		/*
+		 * Zap only 4k SPTEs since the legacy MMU only supports dirty
+		 * logging at a 4k granularity and never creates collapsible
+		 * 2m SPTEs during dirty logging.
+		 */
+		flush = slot_handle_level_4k(kvm, slot, kvm_mmu_zap_collapsible_spte, true);
 		if (flush)
 			kvm_arch_flush_remote_tlbs_memslot(kvm, slot);
 		write_unlock(&kvm->mmu_lock);
···
 
 	if (kvm_memslots_have_rmaps(kvm)) {
 		write_lock(&kvm->mmu_lock);
-		flush = slot_handle_leaf(kvm, memslot, __rmap_clear_dirty,
-					 false);
+		/*
+		 * Clear dirty bits only on 4k SPTEs since the legacy MMU only
+		 * support dirty logging at a 4k granularity.
+		 */
+		flush = slot_handle_level_4k(kvm, memslot, __rmap_clear_dirty, false);
 		write_unlock(&kvm->mmu_lock);
 	}
 
···
 	mmu_audit_disable();
 }
 
-static int set_nx_huge_pages_recovery_ratio(const char *val, const struct kernel_param *kp)
+static int set_nx_huge_pages_recovery_param(const char *val, const struct kernel_param *kp)
 {
-	unsigned int old_val;
+	bool was_recovery_enabled, is_recovery_enabled;
+	uint old_period, new_period;
 	int err;
 
-	old_val = nx_huge_pages_recovery_ratio;
+	was_recovery_enabled = nx_huge_pages_recovery_ratio;
+	old_period = nx_huge_pages_recovery_period_ms;
+
 	err = param_set_uint(val, kp);
 	if (err)
 		return err;
 
-	if (READ_ONCE(nx_huge_pages) &&
-	    !old_val && nx_huge_pages_recovery_ratio) {
+	is_recovery_enabled = nx_huge_pages_recovery_ratio;
+	new_period = nx_huge_pages_recovery_period_ms;
+
+	if (READ_ONCE(nx_huge_pages) && is_recovery_enabled &&
+	    (!was_recovery_enabled || old_period > new_period)) {
 		struct kvm *kvm;
 
 		mutex_lock(&kvm_lock);
···
 
 static long get_nx_lpage_recovery_timeout(u64 start_time)
 {
-	return READ_ONCE(nx_huge_pages) && READ_ONCE(nx_huge_pages_recovery_ratio)
-		? start_time + 60 * HZ - get_jiffies_64()
+	uint ratio = READ_ONCE(nx_huge_pages_recovery_ratio);
+	uint period = READ_ONCE(nx_huge_pages_recovery_period_ms);
+
+	if (!period && ratio) {
+		/* Make sure the period is not less than one second.  */
+		ratio = min(ratio, 3600u);
+		period = 60 * 60 * 1000 / ratio;
+	}
+
+	return READ_ONCE(nx_huge_pages) && ratio
+		? start_time + msecs_to_jiffies(period) - get_jiffies_64()
 		: MAX_SCHEDULE_TIMEOUT;
 }
 

arch/x86/kvm/mmu/mmu_internal.h

···
 	       kvm_x86_ops.cpu_dirty_log_size;
 }
 
-extern int nx_huge_pages;
-static inline bool is_nx_huge_page_enabled(void)
-{
-	return READ_ONCE(nx_huge_pages);
-}
-
-int mmu_try_to_unsync_pages(struct kvm_vcpu *vcpu, gfn_t gfn, bool can_unsync);
+int mmu_try_to_unsync_pages(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot,
+			    gfn_t gfn, bool can_unsync, bool prefetch);
 
 void kvm_mmu_gfn_disallow_lpage(const struct kvm_memory_slot *slot, gfn_t gfn);
 void kvm_mmu_gfn_allow_lpage(const struct kvm_memory_slot *slot, gfn_t gfn);
···
 	RET_PF_SPURIOUS,
 };
 
-/* Bits which may be returned by set_spte() */
-#define SET_SPTE_WRITE_PROTECTED_PT	BIT(0)
-#define SET_SPTE_NEED_REMOTE_TLB_FLUSH	BIT(1)
-#define SET_SPTE_SPURIOUS		BIT(2)
-
 int kvm_mmu_max_mapping_level(struct kvm *kvm,
 			      const struct kvm_memory_slot *slot, gfn_t gfn,
 			      kvm_pfn_t pfn, int max_level);
-int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn,
-			    int max_level, kvm_pfn_t *pfnp,
-			    bool huge_page_disallowed, int *req_level);
-void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level,
-				kvm_pfn_t *pfnp, int *goal_levelp);
+void kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
+void disallowed_hugepage_adjust(struct kvm_page_fault *fault, u64 spte, int cur_level);
 
 void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc);
 

arch/x86/kvm/mmu/mmutrace.h

···
 
 TRACE_EVENT(
 	fast_page_fault,
-	TP_PROTO(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u32 error_code,
+	TP_PROTO(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
 		 u64 *sptep, u64 old_spte, int ret),
-	TP_ARGS(vcpu, cr2_or_gpa, error_code, sptep, old_spte, ret),
+	TP_ARGS(vcpu, fault, sptep, old_spte, ret),
 
 	TP_STRUCT__entry(
 		__field(int, vcpu_id)
···
 
 	TP_fast_assign(
 		__entry->vcpu_id = vcpu->vcpu_id;
-		__entry->cr2_or_gpa = cr2_or_gpa;
-		__entry->error_code = error_code;
+		__entry->cr2_or_gpa = fault->addr;
+		__entry->error_code = fault->error_code;
 		__entry->sptep = sptep;
 		__entry->old_spte = old_spte;
 		__entry->new_spte = *sptep;
···
 
 TRACE_EVENT(
 	kvm_mmu_spte_requested,
-	TP_PROTO(gpa_t addr, int level, kvm_pfn_t pfn),
-	TP_ARGS(addr, level, pfn),
+	TP_PROTO(struct kvm_page_fault *fault),
+	TP_ARGS(fault),
 
 	TP_STRUCT__entry(
 		__field(u64, gfn)
···
 	),
 
 	TP_fast_assign(
-		__entry->gfn = addr >> PAGE_SHIFT;
-		__entry->pfn = pfn | (__entry->gfn & (KVM_PAGES_PER_HPAGE(level) - 1));
-		__entry->level = level;
+		__entry->gfn = fault->gfn;
+		__entry->pfn = fault->pfn | (fault->gfn & (KVM_PAGES_PER_HPAGE(fault->goal_level) - 1));
+		__entry->level = fault->goal_level;
 	),
 
 	TP_printk("gfn %llx pfn %llx level %d",

arch/x86/kvm/mmu/page_track.c

···
 #include "mmu.h"
 #include "mmu_internal.h"
 
+bool kvm_page_track_write_tracking_enabled(struct kvm *kvm)
+{
+	return IS_ENABLED(CONFIG_KVM_EXTERNAL_WRITE_TRACKING) ||
+	       !tdp_enabled || kvm_shadow_root_allocated(kvm);
+}
+
 void kvm_page_track_free_memslot(struct kvm_memory_slot *slot)
 {
 	int i;
···
 	}
 }
 
-int kvm_page_track_create_memslot(struct kvm_memory_slot *slot,
+int kvm_page_track_create_memslot(struct kvm *kvm,
+				  struct kvm_memory_slot *slot,
 				  unsigned long npages)
 {
-	int  i;
+	int i;
 
 	for (i = 0; i < KVM_PAGE_TRACK_MAX; i++) {
+		if (i == KVM_PAGE_TRACK_WRITE &&
+		    !kvm_page_track_write_tracking_enabled(kvm))
+			continue;
+
 		slot->arch.gfn_track[i] =
 			kvcalloc(npages, sizeof(*slot->arch.gfn_track[i]),
 				 GFP_KERNEL_ACCOUNT);
···
 		return false;
 
 	return true;
+}
+
+int kvm_page_track_write_tracking_alloc(struct kvm_memory_slot *slot)
+{
+	unsigned short *gfn_track;
+
+	if (slot->arch.gfn_track[KVM_PAGE_TRACK_WRITE])
+		return 0;
+
+	gfn_track = kvcalloc(slot->npages, sizeof(*gfn_track), GFP_KERNEL_ACCOUNT);
+	if (gfn_track == NULL)
+		return -ENOMEM;
+
+	slot->arch.gfn_track[KVM_PAGE_TRACK_WRITE] = gfn_track;
+	return 0;
 }
 
 static void update_gfn_track(struct kvm_memory_slot *slot, gfn_t gfn,
···
 	if (WARN_ON(!page_track_mode_is_valid(mode)))
 		return;
 
+	if (WARN_ON(mode == KVM_PAGE_TRACK_WRITE &&
+		    !kvm_page_track_write_tracking_enabled(kvm)))
+		return;
+
 	update_gfn_track(slot, gfn, mode, 1);
 
 	/*
···
 	if (WARN_ON(!page_track_mode_is_valid(mode)))
 		return;
 
+	if (WARN_ON(mode == KVM_PAGE_TRACK_WRITE &&
+		    !kvm_page_track_write_tracking_enabled(kvm)))
+		return;
+
 	update_gfn_track(slot, gfn, mode, -1);
 
 	/*
···
 /*
  * check if the corresponding access on the specified guest page is tracked.
  */
-bool kvm_page_track_is_active(struct kvm_vcpu *vcpu, gfn_t gfn,
-			      enum kvm_page_track_mode mode)
+bool kvm_slot_page_track_is_active(struct kvm_vcpu *vcpu,
+				   struct kvm_memory_slot *slot, gfn_t gfn,
+				   enum kvm_page_track_mode mode)
 {
-	struct kvm_memory_slot *slot;
 	int index;
 
 	if (WARN_ON(!page_track_mode_is_valid(mode)))
 		return false;
 
-	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
 	if (!slot)
+		return false;
+
+	if (mode == KVM_PAGE_TRACK_WRITE &&
+	    !kvm_page_track_write_tracking_enabled(vcpu->kvm))
 		return false;
 
 	index = gfn_to_index(gfn, slot->base_gfn, PG_LEVEL_4K);

arch/x86/kvm/mmu/paging_tmpl.h

···
 FNAME(prefetch_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 		     u64 *spte, pt_element_t gpte, bool no_dirty_log)
 {
+	struct kvm_memory_slot *slot;
 	unsigned pte_access;
 	gfn_t gfn;
 	kvm_pfn_t pfn;
···
 	gfn = gpte_to_gfn(gpte);
 	pte_access = sp->role.access & FNAME(gpte_access)(gpte);
 	FNAME(protect_clean_gpte)(vcpu->arch.mmu, &pte_access, gpte);
-	pfn = pte_prefetch_gfn_to_pfn(vcpu, gfn,
+
+	slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn,
 			no_dirty_log && (pte_access & ACC_WRITE_MASK));
+	if (!slot)
+		return false;
+
+	pfn = gfn_to_pfn_memslot_atomic(slot, gfn);
 	if (is_error_pfn(pfn))
 		return false;
 
-	/*
-	 * we call mmu_set_spte() with host_writable = true because
-	 * pte_prefetch_gfn_to_pfn always gets a writable pfn.
-	 */
-	mmu_set_spte(vcpu, spte, pte_access, false, PG_LEVEL_4K, gfn, pfn,
-		     true, true);
-
+	mmu_set_spte(vcpu, slot, spte, pte_access, gfn, pfn, NULL);
 	kvm_release_pfn_clean(pfn);
 	return true;
-}
-
-static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
-			      u64 *spte, const void *pte)
-{
-	pt_element_t gpte = *(const pt_element_t *)pte;
-
-	FNAME(prefetch_gpte)(vcpu, sp, spte, gpte, false);
 }
 
 static bool FNAME(gpte_changed)(struct kvm_vcpu *vcpu,
···
  * If the guest tries to write a write-protected page, we need to
  * emulate this operation, return 1 to indicate this case.
  */
-static int FNAME(fetch)(struct kvm_vcpu *vcpu, gpa_t addr,
-			 struct guest_walker *gw, u32 error_code,
-			 int max_level, kvm_pfn_t pfn, bool map_writable,
-			 bool prefault)
+static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault,
+			 struct guest_walker *gw)
 {
-	bool nx_huge_page_workaround_enabled = is_nx_huge_page_enabled();
-	bool write_fault = error_code & PFERR_WRITE_MASK;
-	bool exec = error_code & PFERR_FETCH_MASK;
-	bool huge_page_disallowed = exec && nx_huge_page_workaround_enabled;
 	struct kvm_mmu_page *sp = NULL;
 	struct kvm_shadow_walk_iterator it;
 	unsigned int direct_access, access;
-	int top_level, level, req_level, ret;
-	gfn_t base_gfn = gw->gfn;
+	int top_level, ret;
+	gfn_t base_gfn = fault->gfn;
 
+	WARN_ON_ONCE(gw->gfn != base_gfn);
 	direct_access = gw->pte_access;
 
 	top_level = vcpu->arch.mmu->root_level;
···
 	if (WARN_ON(!VALID_PAGE(vcpu->arch.mmu->root_hpa)))
 		goto out_gpte_changed;
 
-	for (shadow_walk_init(&it, vcpu, addr);
+	for (shadow_walk_init(&it, vcpu, fault->addr);
 	     shadow_walk_okay(&it) && it.level > gw->level;
 	     shadow_walk_next(&it)) {
 		gfn_t table_gfn;
···
 		if (!is_shadow_present_pte(*it.sptep)) {
 			table_gfn = gw->table_gfn[it.level - 2];
 			access = gw->pt_access[it.level - 2];
-			sp = kvm_mmu_get_page(vcpu, table_gfn, addr,
+			sp = kvm_mmu_get_page(vcpu, table_gfn, fault->addr,
 					      it.level-1, false, access);
 			/*
 			 * We must synchronize the pagetable before linking it
···
 			link_shadow_page(vcpu, it.sptep, sp);
 	}
 
-	level = kvm_mmu_hugepage_adjust(vcpu, gw->gfn, max_level, &pfn,
-					huge_page_disallowed, &req_level);
+	kvm_mmu_hugepage_adjust(vcpu, fault);
 
-	trace_kvm_mmu_spte_requested(addr, gw->level, pfn);
+	trace_kvm_mmu_spte_requested(fault);
 
 	for (; shadow_walk_okay(&it); shadow_walk_next(&it)) {
 		clear_sp_write_flooding_count(it.sptep);
···
 		 * We cannot overwrite existing page tables with an NX
 		 * large page, as the leaf could be executable.
 		 */
-		if (nx_huge_page_workaround_enabled)
-			disallowed_hugepage_adjust(*it.sptep, gw->gfn, it.level,
-						   &pfn, &level);
+		if (fault->nx_huge_page_workaround_enabled)
+			disallowed_hugepage_adjust(fault, *it.sptep, it.level);
 
-		base_gfn = gw->gfn & ~(KVM_PAGES_PER_HPAGE(it.level) - 1);
-		if (it.level == level)
+		base_gfn = fault->gfn & ~(KVM_PAGES_PER_HPAGE(it.level) - 1);
+		if (it.level == fault->goal_level)
 			break;
 
 		validate_direct_spte(vcpu, it.sptep, direct_access);
···
 		drop_large_spte(vcpu, it.sptep);
 
 		if (!is_shadow_present_pte(*it.sptep)) {
-			sp = kvm_mmu_get_page(vcpu, base_gfn, addr,
+			sp = kvm_mmu_get_page(vcpu, base_gfn, fault->addr,
 					      it.level - 1, true, direct_access);
 			link_shadow_page(vcpu, it.sptep, sp);
-			if (huge_page_disallowed && req_level >= it.level)
+			if (fault->huge_page_disallowed &&
+			    fault->req_level >= it.level)
 				account_huge_nx_page(vcpu->kvm, sp);
 		}
 	}
 
-	ret = mmu_set_spte(vcpu, it.sptep, gw->pte_access, write_fault,
-			   it.level, base_gfn, pfn, prefault, map_writable);
+	if (WARN_ON_ONCE(it.level != fault->goal_level))
+		return -EFAULT;
+
+	ret = mmu_set_spte(vcpu, fault->slot, it.sptep, gw->pte_access,
+			   base_gfn, fault->pfn, fault);
 	if (ret == RET_PF_SPURIOUS)
 		return ret;
 
···
  *  Returns: 1 if we need to emulate the instruction, 0 otherwise, or
  *           a negative value on error.
  */
-static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gpa_t addr, u32 error_code,
-			     bool prefault)
+static int FNAME(page_fault)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
-	bool write_fault = error_code & PFERR_WRITE_MASK;
-	bool user_fault = error_code & PFERR_USER_MASK;
 	struct guest_walker walker;
 	int r;
-	kvm_pfn_t pfn;
-	hva_t hva;
 	unsigned long mmu_seq;
-	bool map_writable, is_self_change_mapping;
-	int max_level;
+	bool is_self_change_mapping;
 
-	pgprintk("%s: addr %lx err %x\n", __func__, addr, error_code);
-
-	/*
-	 * If PFEC.RSVD is set, this is a shadow page fault.
-	 * The bit needs to be cleared before walking guest page tables.
-	 */
-	error_code &= ~PFERR_RSVD_MASK;
+	pgprintk("%s: addr %lx err %x\n", __func__, fault->addr, fault->error_code);
+	WARN_ON_ONCE(fault->is_tdp);
 
 	/*
 	 * Look up the guest pte for the faulting address.
+	 * If PFEC.RSVD is set, this is a shadow page fault.
+	 * The bit needs to be cleared before walking guest page tables.
 	 */
-	r = FNAME(walk_addr)(&walker, vcpu, addr, error_code);
+	r = FNAME(walk_addr)(&walker, vcpu, fault->addr,
+			     fault->error_code & ~PFERR_RSVD_MASK);
 
 	/*
 	 * The page is not mapped by the guest.  Let the guest handle it.
 	 */
 	if (!r) {
 		pgprintk("%s: guest page fault\n", __func__);
-		if (!prefault)
+		if (!fault->prefetch)
 			kvm_inject_emulated_page_fault(vcpu, &walker.fault);
 
 		return RET_PF_RETRY;
 	}
 
-	if (page_fault_handle_page_track(vcpu, error_code, walker.gfn)) {
-		shadow_page_table_clear_flood(vcpu, addr);
+	fault->gfn = walker.gfn;
+	fault->slot = kvm_vcpu_gfn_to_memslot(vcpu, fault->gfn);
+
+	if (page_fault_handle_page_track(vcpu, fault)) {
+		shadow_page_table_clear_flood(vcpu, fault->addr);
 		return RET_PF_EMULATE;
 	}
 
···
 	vcpu->arch.write_fault_to_shadow_pgtable = false;
 
 	is_self_change_mapping = FNAME(is_self_change_mapping)(vcpu,
-	      &walker, user_fault, &vcpu->arch.write_fault_to_shadow_pgtable);
+	      &walker, fault->user, &vcpu->arch.write_fault_to_shadow_pgtable);
 
 	if (is_self_change_mapping)
-		max_level = PG_LEVEL_4K;
+		fault->max_level = PG_LEVEL_4K;
 	else
-		max_level = walker.level;
+		fault->max_level = walker.level;
 
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (kvm_faultin_pfn(vcpu, prefault, walker.gfn, addr, &pfn, &hva,
-			 write_fault, &map_writable, &r))
+	if (kvm_faultin_pfn(vcpu, fault, &r))
 		return r;
 
-	if (handle_abnormal_pfn(vcpu, addr, walker.gfn, pfn, walker.pte_access, &r))
+	if (handle_abnormal_pfn(vcpu, fault, walker.pte_access, &r))
 		return r;
 
 	/*
 	 * Do not change pte_access if the pfn is a mmio page, otherwise
 	 * we will cache the incorrect access into mmio spte.
 	 */
-	if (write_fault && !(walker.pte_access & ACC_WRITE_MASK) &&
-	    !is_cr0_wp(vcpu->arch.mmu) && !user_fault && !is_noslot_pfn(pfn)) {
+	if (fault->write && !(walker.pte_access & ACC_WRITE_MASK) &&
+	    !is_cr0_wp(vcpu->arch.mmu) && !fault->user && fault->slot) {
 		walker.pte_access |= ACC_WRITE_MASK;
 		walker.pte_access &= ~ACC_USER_MASK;
 
···
 
 	r = RET_PF_RETRY;
 	write_lock(&vcpu->kvm->mmu_lock);
-	if (!is_noslot_pfn(pfn) && mmu_notifier_retry_hva(vcpu->kvm, mmu_seq, hva))
+	if (fault->slot && mmu_notifier_retry_hva(vcpu->kvm, mmu_seq, fault->hva))
 		goto out_unlock;
 
 	kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT);
 	r = make_mmu_pages_available(vcpu);
 	if (r)
 		goto out_unlock;
-	r = FNAME(fetch)(vcpu, addr, &walker, error_code, max_level, pfn,
-			 map_writable, prefault);
+	r = FNAME(fetch)(vcpu, fault, &walker);
 	kvm_mmu_audit(vcpu, AUDIT_POST_PAGE_FAULT);
 
 out_unlock:
 	write_unlock(&vcpu->kvm->mmu_lock);
-	kvm_release_pfn_clean(pfn);
+	kvm_release_pfn_clean(fault->pfn);
 	return r;
 }
 
···
 						       sizeof(pt_element_t)))
 				break;
 
-			FNAME(update_pte)(vcpu, sp, sptep, &gpte);
+			FNAME(prefetch_gpte)(vcpu, sp, sptep, gpte, false);
 		}
 
-		if (!is_shadow_present_pte(*sptep) || !sp->unsync_children)
+		if (!sp->unsync_children)
 			break;
 	}
 	write_unlock(&vcpu->kvm->mmu_lock);
···
  * Using the cached information from sp->gfns is safe because:
  * - The spte has a reference to the struct page, so the pfn for a given gfn
  *   can't change unless all sptes pointing to it are nuked first.
+ *
+ * Returns
+ * < 0: the sp should be zapped
+ *   0: the sp is synced and no tlb flushing is required
+ * > 0: the sp is synced and tlb flushing is required
  */
 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;
-	int i, nr_present = 0;
+	int i;
 	bool host_writable;
 	gpa_t first_pte_gpa;
-	int set_spte_ret = 0;
+	bool flush = false;
 
 	/*
 	 * Ignore various flags when verifying that it's safe to sync a shadow
···
 	 */
 	if (WARN_ON_ONCE(sp->role.direct ||
 			 (sp->role.word ^ mmu_role.word) & ~sync_role_ign.word))
-		return 0;
+		return -1;
 
 	first_pte_gpa = FNAME(get_level1_sp_gpa)(sp);
 
 	for (i = 0; i < PT64_ENT_PER_PAGE; i++) {
+		u64 *sptep, spte;
+		struct kvm_memory_slot *slot;
 		unsigned pte_access;
 		pt_element_t gpte;
 		gpa_t pte_gpa;
···
 
 		if (kvm_vcpu_read_guest_atomic(vcpu, pte_gpa, &gpte,
 					       sizeof(pt_element_t)))
-			return 0;
+			return -1;
 
 		if (FNAME(prefetch_invalid_gpte)(vcpu, sp, &sp->spt[i], gpte)) {
-			set_spte_ret |= SET_SPTE_NEED_REMOTE_TLB_FLUSH;
+			flush = true;
 			continue;
 		}
 
···
 		pte_access &= FNAME(gpte_access)(gpte);
 		FNAME(protect_clean_gpte)(vcpu->arch.mmu, &pte_access, gpte);
 
-		if (sync_mmio_spte(vcpu, &sp->spt[i], gfn, pte_access,
-		      &nr_present))
+		if (sync_mmio_spte(vcpu, &sp->spt[i], gfn, pte_access))
 			continue;
 
 		if (gfn != sp->gfns[i]) {
 			drop_spte(vcpu->kvm, &sp->spt[i]);
-			set_spte_ret |= SET_SPTE_NEED_REMOTE_TLB_FLUSH;
+			flush = true;
 			continue;
 		}
 
-		nr_present++;
+		sptep = &sp->spt[i];
+		spte = *sptep;
+		host_writable = spte & shadow_host_writable_mask;
+		slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
+		make_spte(vcpu, sp, slot, pte_access, gfn,
+			  spte_to_pfn(spte), spte, true, false,
+			  host_writable, &spte);
 
-		host_writable = sp->spt[i] & shadow_host_writable_mask;
-
-		set_spte_ret |= set_spte(vcpu, &sp->spt[i],
-					 pte_access, PG_LEVEL_4K,
-					 gfn, spte_to_pfn(sp->spt[i]),
-					 true, false, host_writable);
+		flush |= mmu_spte_update(sptep, spte);
 	}
 
-	if (set_spte_ret & SET_SPTE_NEED_REMOTE_TLB_FLUSH)
-		kvm_flush_remote_tlbs(vcpu->kvm);
-
-	return nr_present;
+	return flush;
 }
 
 #undef pt_element_t

arch/x86/kvm/mmu/spte.c

···
 				     E820_TYPE_RAM);
 }
 
-int make_spte(struct kvm_vcpu *vcpu, unsigned int pte_access, int level,
-		     gfn_t gfn, kvm_pfn_t pfn, u64 old_spte, bool speculative,
-		     bool can_unsync, bool host_writable, bool ad_disabled,
-		     u64 *new_spte)
+bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
+	       struct kvm_memory_slot *slot,
+	       unsigned int pte_access, gfn_t gfn, kvm_pfn_t pfn,
+	       u64 old_spte, bool prefetch, bool can_unsync,
+	       bool host_writable, u64 *new_spte)
 {
+	int level = sp->role.level;
 	u64 spte = SPTE_MMU_PRESENT_MASK;
-	int ret = 0;
+	bool wrprot = false;
 
-	if (ad_disabled)
+	if (sp->role.ad_disabled)
 		spte |= SPTE_TDP_AD_DISABLED_MASK;
 	else if (kvm_vcpu_ad_need_write_protect(vcpu))
 		spte |= SPTE_TDP_AD_WRPROT_ONLY_MASK;
···
 	 * read access.  See FNAME(gpte_access) in paging_tmpl.h.
 	 */
 	spte |= shadow_present_mask;
-	if (!speculative)
+	if (!prefetch)
 		spte |= spte_shadow_accessed_mask(spte);
 
 	if (level > PG_LEVEL_4K && (pte_access & ACC_EXEC_MASK) &&
···
 		 * is responsibility of kvm_mmu_get_page / kvm_mmu_sync_roots.
 		 * Same reasoning can be applied to dirty page accounting.
 		 */
-		if (!can_unsync && is_writable_pte(old_spte))
+		if (is_writable_pte(old_spte))
 			goto out;
 
 		/*
···
 		 * e.g. it's write-tracked (upper-level SPs) or has one or more
 		 * shadow pages and unsync'ing pages is not allowed.
 		 */
-		if (mmu_try_to_unsync_pages(vcpu, gfn, can_unsync)) {
+		if (mmu_try_to_unsync_pages(vcpu, slot, gfn, can_unsync, prefetch)) {
 			pgprintk("%s: found shadow page for %llx, marking ro\n",
 				 __func__, gfn);
-			ret |= SET_SPTE_WRITE_PROTECTED_PT;
+			wrprot = true;
 			pte_access &= ~ACC_WRITE_MASK;
 			spte &= ~(PT_WRITABLE_MASK | shadow_mmu_writable_mask);
 		}
···
 	if (pte_access & ACC_WRITE_MASK)
 		spte |= spte_shadow_dirty_mask(spte);
 
-	if (speculative)
+out:
+	if (prefetch)
 		spte = mark_spte_for_access_track(spte);
 
-out:
 	WARN_ONCE(is_rsvd_spte(&vcpu->arch.mmu->shadow_zero_check, spte, level),
 		  "spte = 0x%llx, level = %d, rsvd bits = 0x%llx", spte, level,
 		  get_rsvd_bits(&vcpu->arch.mmu->shadow_zero_check, spte, level));
 
+	if ((spte & PT_WRITABLE_MASK) && kvm_slot_dirty_track_enabled(slot)) {
+		/* Enforced by kvm_mmu_hugepage_adjust. */
+		WARN_ON(level > PG_LEVEL_4K);
+		mark_page_dirty_in_slot(vcpu->kvm, slot, gfn);
+	}
+
 	*new_spte = spte;
-	return ret;
+	return wrprot;
 }
 
 u64 make_nonleaf_spte(u64 *child_pt, bool ad_disabled)

arch/x86/kvm/mmu/spte.h

···
 static __always_inline bool is_rsvd_spte(struct rsvd_bits_validate *rsvd_check,
 					 u64 spte, int level)
 {
-	/*
-	 * Use a bitwise-OR instead of a logical-OR to aggregate the reserved
-	 * bits and EPT's invalid memtype/XWR checks to avoid an extra Jcc
-	 * (this is extremely unlikely to be short-circuited as true).
-	 */
-	return __is_bad_mt_xwr(rsvd_check, spte) |
+	return __is_bad_mt_xwr(rsvd_check, spte) ||
 	       __is_rsvd_bits_set(rsvd_check, spte, level);
 }
 
···
 	return gen;
 }
 
-/* Bits which may be returned by set_spte() */
-#define SET_SPTE_WRITE_PROTECTED_PT    BIT(0)
-#define SET_SPTE_NEED_REMOTE_TLB_FLUSH BIT(1)
-#define SET_SPTE_SPURIOUS              BIT(2)
-
-int make_spte(struct kvm_vcpu *vcpu, unsigned int pte_access, int level,
-		     gfn_t gfn, kvm_pfn_t pfn, u64 old_spte, bool speculative,
-		     bool can_unsync, bool host_writable, bool ad_disabled,
-		     u64 *new_spte);
+bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
+	       struct kvm_memory_slot *slot,
+	       unsigned int pte_access, gfn_t gfn, kvm_pfn_t pfn,
+	       u64 old_spte, bool prefetch, bool can_unsync,
+	       bool host_writable, u64 *new_spte);
 u64 make_nonleaf_spte(u64 *child_pt, bool ad_disabled);
 u64 make_mmio_spte(struct kvm_vcpu *vcpu, u64 gfn, unsigned int access);
 u64 mark_spte_for_access_track(u64 spte);

arch/x86/kvm/mmu/tdp_mmu.c

···
 	role.direct = true;
 	role.gpte_is_8_bytes = true;
 	role.access = ACC_ALL;
+	role.ad_disabled = !shadow_accessed_mask;
 
 	return role;
 }
···
 }
 
 /*
- * tdp_mmu_set_spte_atomic_no_dirty_log - Set a TDP MMU SPTE atomically
- * and handle the associated bookkeeping, but do not mark the page dirty
+ * tdp_mmu_set_spte_atomic - Set a TDP MMU SPTE atomically
+ * and handle the associated bookkeeping.  Do not mark the page dirty
  * in KVM's dirty bitmaps.
  *
  * @kvm: kvm instance
···
  * Returns: true if the SPTE was set, false if it was not. If false is returned,
  *	    this function will have no side-effects.
  */
-static inline bool tdp_mmu_set_spte_atomic_no_dirty_log(struct kvm *kvm,
-							struct tdp_iter *iter,
-							u64 new_spte)
+static inline bool tdp_mmu_set_spte_atomic(struct kvm *kvm,
+					   struct tdp_iter *iter,
+					   u64 new_spte)
 {
 	lockdep_assert_held_read(&kvm->mmu_lock);
 
···
 	return true;
 }
 
-/*
- * tdp_mmu_map_set_spte_atomic - Set a leaf TDP MMU SPTE atomically to resolve a
- * TDP page fault.
- *
- * @vcpu: The vcpu instance that took the TDP page fault.
- * @iter: a tdp_iter instance currently on the SPTE that should be set
- * @new_spte: The value the SPTE should be set to
- *
- * Returns: true if the SPTE was set, false if it was not. If false is returned,
- *	    this function will have no side-effects.
- */
-static inline bool tdp_mmu_map_set_spte_atomic(struct kvm_vcpu *vcpu,
-					       struct tdp_iter *iter,
-					       u64 new_spte)
-{
-	struct kvm *kvm = vcpu->kvm;
-
-	if (!tdp_mmu_set_spte_atomic_no_dirty_log(kvm, iter, new_spte))
-		return false;
-
-	/*
-	 * Use kvm_vcpu_gfn_to_memslot() instead of going through
-	 * handle_changed_spte_dirty_log() to leverage vcpu->last_used_slot.
-	 */
-	if (is_writable_pte(new_spte)) {
-		struct kvm_memory_slot *slot = kvm_vcpu_gfn_to_memslot(vcpu, iter->gfn);
-
-		if (slot && kvm_slot_dirty_track_enabled(slot)) {
-			/* Enforced by kvm_mmu_hugepage_adjust. */
-			WARN_ON_ONCE(iter->level > PG_LEVEL_4K);
-			mark_page_dirty_in_slot(kvm, slot, iter->gfn);
-		}
-	}
-
-	return true;
-}
-
 static inline bool tdp_mmu_zap_spte_atomic(struct kvm *kvm,
 					   struct tdp_iter *iter)
 {
···
 	 * immediately installing a present entry in its place
 	 * before the TLBs are flushed.
 	 */
-	if (!tdp_mmu_set_spte_atomic_no_dirty_log(kvm, iter, REMOVED_SPTE))
+	if (!tdp_mmu_set_spte_atomic(kvm, iter, REMOVED_SPTE))
 		return false;
 
 	kvm_flush_remote_tlbs_with_address(kvm, iter->gfn,
···
  * Installs a last-level SPTE to handle a TDP page fault.
  * (NPT/EPT violation/misconfiguration)
  */
-static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu, int write,
-					  int map_writable,
-					  struct tdp_iter *iter,
-					  kvm_pfn_t pfn, bool prefault)
+static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu,
+					  struct kvm_page_fault *fault,
+					  struct tdp_iter *iter)
 {
+	struct kvm_mmu_page *sp = sptep_to_sp(iter->sptep);
 	u64 new_spte;
 	int ret = RET_PF_FIXED;
-	int make_spte_ret = 0;
+	bool wrprot = false;
 
-	if (unlikely(is_noslot_pfn(pfn)))
+	WARN_ON(sp->role.level != fault->goal_level);
+	if (unlikely(!fault->slot))
 		new_spte = make_mmio_spte(vcpu, iter->gfn, ACC_ALL);
 	else
-		make_spte_ret = make_spte(vcpu, ACC_ALL, iter->level, iter->gfn,
-					 pfn, iter->old_spte, prefault, true,
-					 map_writable, !shadow_accessed_mask,
-					 &new_spte);
+		wrprot = make_spte(vcpu, sp, fault->slot, ACC_ALL, iter->gfn,
+					 fault->pfn, iter->old_spte, fault->prefetch, true,
+					 fault->map_writable, &new_spte);
 
 	if (new_spte == iter->old_spte)
 		ret = RET_PF_SPURIOUS;
-	else if (!tdp_mmu_map_set_spte_atomic(vcpu, iter, new_spte))
+	else if (!tdp_mmu_set_spte_atomic(vcpu->kvm, iter, new_spte))
 		return RET_PF_RETRY;
 
 	/*
···
 	 * protected, emulation is needed. If the emulation was skipped,
 	 * the vCPU would have the same fault again.
 	 */
-	if (make_spte_ret & SET_SPTE_WRITE_PROTECTED_PT) {
-		if (write)
+	if (wrprot) {
+		if (fault->write)
 			ret = RET_PF_EMULATE;
-		kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
 	}
 
 	/* If a MMIO SPTE is installed, the MMIO will need to be emulated. */
···
  * Handle a TDP page fault (NPT/EPT violation/misconfiguration) by installing
  * page tables and SPTEs to translate the faulting guest physical address.
  */
-int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-		    int map_writable, int max_level, kvm_pfn_t pfn,
-		    bool prefault)
+int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
-	bool nx_huge_page_workaround_enabled = is_nx_huge_page_enabled();
-	bool write = error_code & PFERR_WRITE_MASK;
-	bool exec = error_code & PFERR_FETCH_MASK;
-	bool huge_page_disallowed = exec && nx_huge_page_workaround_enabled;
 	struct kvm_mmu *mmu = vcpu->arch.mmu;
 	struct tdp_iter iter;
 	struct kvm_mmu_page *sp;
 	u64 *child_pt;
 	u64 new_spte;
 	int ret;
-	gfn_t gfn = gpa >> PAGE_SHIFT;
-	int level;
-	int req_level;
 
-	level = kvm_mmu_hugepage_adjust(vcpu, gfn, max_level, &pfn,
-					huge_page_disallowed, &req_level);
+	kvm_mmu_hugepage_adjust(vcpu, fault);
 
-	trace_kvm_mmu_spte_requested(gpa, level, pfn);
+	trace_kvm_mmu_spte_requested(fault);
 
 	rcu_read_lock();
 
-	tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) {
-		if (nx_huge_page_workaround_enabled)
-			disallowed_hugepage_adjust(iter.old_spte, gfn,
-						   iter.level, &pfn, &level);
+	tdp_mmu_for_each_pte(iter, mmu, fault->gfn, fault->gfn + 1) {
+		if (fault->nx_huge_page_workaround_enabled)
+			disallowed_hugepage_adjust(fault, iter.old_spte, iter.level);
 
-		if (iter.level == level)
+		if (iter.level == fault->goal_level)
 			break;
 
 		/*
···
 			new_spte = make_nonleaf_spte(child_pt,
 						     !shadow_accessed_mask);
 
-			if (tdp_mmu_set_spte_atomic_no_dirty_log(vcpu->kvm, &iter, new_spte)) {
+			if (tdp_mmu_set_spte_atomic(vcpu->kvm, &iter, new_spte)) {
 				tdp_mmu_link_page(vcpu->kvm, sp,
-						  huge_page_disallowed &&
-						  req_level >= iter.level);
+						  fault->huge_page_disallowed &&
+						  fault->req_level >= iter.level);
 
 				trace_kvm_mmu_get_page(sp, true);
 			} else {
···
 		}
 	}
 
-	if (iter.level != level) {
+	if (iter.level != fault->goal_level) {
 		rcu_read_unlock();
 		return RET_PF_RETRY;
 	}
 
-	ret = tdp_mmu_map_handle_target_level(vcpu, write, map_writable, &iter,
-					      pfn, prefault);
+	ret = tdp_mmu_map_handle_target_level(vcpu, fault, &iter);
 	rcu_read_unlock();
 
 	return ret;
···
 
 		new_spte = iter.old_spte & ~PT_WRITABLE_MASK;
 
-		if (!tdp_mmu_set_spte_atomic_no_dirty_log(kvm, &iter,
-							  new_spte)) {
+		if (!tdp_mmu_set_spte_atomic(kvm, &iter, new_spte)) {
 			/*
 			 * The iter must explicitly re-read the SPTE because
 			 * the atomic cmpxchg failed.
···
 				continue;
 		}
 
-		if (!tdp_mmu_set_spte_atomic_no_dirty_log(kvm, &iter,
-							  new_spte)) {
+		if (!tdp_mmu_set_spte_atomic(kvm, &iter, new_spte)) {
 			/*
 			 * The iter must explicitly re-read the SPTE because
 			 * the atomic cmpxchg failed.

arch/x86/kvm/mmu/tdp_mmu.h

···
 void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm);
 void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm);
 
-int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
-		    int map_writable, int max_level, kvm_pfn_t pfn,
-		    bool prefault);
+int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
 
 bool kvm_tdp_mmu_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range,
 				 bool flush);
···
 #ifdef CONFIG_X86_64
 bool kvm_mmu_init_tdp_mmu(struct kvm *kvm);
 void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm);
-static inline bool is_tdp_mmu_enabled(struct kvm *kvm) { return kvm->arch.tdp_mmu_enabled; }
 static inline bool is_tdp_mmu_page(struct kvm_mmu_page *sp) { return sp->tdp_mmu_page; }
 
 static inline bool is_tdp_mmu(struct kvm_mmu *mmu)
···
 #else
 static inline bool kvm_mmu_init_tdp_mmu(struct kvm *kvm) { return false; }
 static inline void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm) {}
-static inline bool is_tdp_mmu_enabled(struct kvm *kvm) { return false; }
 static inline bool is_tdp_mmu_page(struct kvm_mmu_page *sp) { return false; }
 static inline bool is_tdp_mmu(struct kvm_mmu *mmu) { return false; }
 #endif

arch/x86/kvm/svm/nested.c

···
 	    kvm_vcpu_is_legal_gpa(vcpu, addr + size - 1);
 }
 
+static bool nested_svm_check_tlb_ctl(struct kvm_vcpu *vcpu, u8 tlb_ctl)
+{
+	/* Nested FLUSHBYASID is not supported yet.  */
+	switch(tlb_ctl) {
+		case TLB_CONTROL_DO_NOTHING:
+		case TLB_CONTROL_FLUSH_ALL_ASID:
+			return true;
+		default:
+			return false;
+	}
+}
+
 static bool nested_vmcb_check_controls(struct kvm_vcpu *vcpu,
 				       struct vmcb_control_area *control)
 {
···
 		return false;
 	if (CC(!nested_svm_check_bitmap_pa(vcpu, control->iopm_base_pa,
 					   IOPM_SIZE)))
+		return false;
+
+	if (CC(!nested_svm_check_tlb_ctl(vcpu, control->tlb_ctl)))
 		return false;
 
 	return true;
···
 	if (nested_npt_enabled(svm))
 		nested_svm_init_mmu_context(vcpu);
 
-	svm->vmcb->control.tsc_offset = vcpu->arch.tsc_offset =
-		vcpu->arch.l1_tsc_offset + svm->nested.ctl.tsc_offset;
+	vcpu->arch.tsc_offset = kvm_calc_nested_tsc_offset(
+			vcpu->arch.l1_tsc_offset,
+			svm->nested.ctl.tsc_offset,
+			svm->tsc_ratio_msr);
+
+	svm->vmcb->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             =
 		(svm->nested.ctl.int_ctl & int_ctl_vmcb12_bits) |
···
 	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;
-
-	svm->vmcb->control.pause_filter_count  = svm->nested.ctl.pause_filter_count;
-	svm->vmcb->control.pause_filter_thresh = svm->nested.ctl.pause_filter_thresh;
 
 	nested_svm_transition_tlb_flush(vcpu);
 
···
 	vmcb12->control.event_inj         = svm->nested.ctl.event_inj;
 	vmcb12->control.event_inj_err     = svm->nested.ctl.event_inj_err;
 
-	vmcb12->control.pause_filter_count =
-		svm->vmcb->control.pause_filter_count;
-	vmcb12->control.pause_filter_thresh =
-		svm->vmcb->control.pause_filter_thresh;
-
 	nested_svm_copy_common_state(svm->nested.vmcb02.ptr, svm->vmcb01.ptr);
 
 	svm_switch_vmcb(svm, &svm->vmcb01);
···
 	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 (svm->tsc_ratio_msr != kvm_default_tsc_scaling_ratio) {
+		WARN_ON(!svm->tsc_scaling_enabled);
+		vcpu->arch.tsc_scaling_ratio = vcpu->arch.l1_tsc_scaling_ratio;
+		svm_write_tsc_multiplier(vcpu, vcpu->arch.tsc_scaling_ratio);
 	}
 
 	svm->nested.ctl.nested_cr3 = 0;
···
 	}
 
 	return NESTED_EXIT_CONTINUE;
+}
+
+void nested_svm_update_tsc_ratio_msr(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	vcpu->arch.tsc_scaling_ratio =
+		kvm_calc_nested_tsc_multiplier(vcpu->arch.l1_tsc_scaling_ratio,
+					       svm->tsc_ratio_msr);
+	svm_write_tsc_multiplier(vcpu, vcpu->arch.tsc_scaling_ratio);
 }
 
 static int svm_get_nested_state(struct kvm_vcpu *vcpu,

arch/x86/kvm/svm/sev.c

···
 	set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1);
 }
 
-void sev_es_create_vcpu(struct vcpu_svm *svm)
+void sev_es_vcpu_reset(struct vcpu_svm *svm)
 {
 	/*
-	 * Set the GHCB MSR value as per the GHCB specification when creating
-	 * a vCPU for an SEV-ES guest.
+	 * Set the GHCB MSR value as per the GHCB specification when emulating
+	 * vCPU RESET for an SEV-ES guest.
 	 */
 	set_ghcb_msr(svm, GHCB_MSR_SEV_INFO(GHCB_VERSION_MAX,
 					    GHCB_VERSION_MIN,

arch/x86/kvm/svm/svm.c

···
 static int vgif = true;
 module_param(vgif, int, 0444);
 
+/* enable/disable LBR virtualization */
+static int lbrv = true;
+module_param(lbrv, int, 0444);
+
+static int tsc_scaling = true;
+module_param(tsc_scaling, int, 0444);
+
 /*
  * enable / disable AVIC.  Because the defaults differ for APICv
  * support between VMX and SVM we cannot use module_param_named.
···
 static void svm_hardware_disable(void)
 {
 	/* Make sure we clean up behind us */
-	if (static_cpu_has(X86_FEATURE_TSCRATEMSR))
+	if (tsc_scaling)
 		wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT);
 
 	cpu_svm_disable();
···
 	wrmsrl(MSR_VM_HSAVE_PA, __sme_page_pa(sd->save_area));
 
 	if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) {
+		/*
+		 * Set the default value, even if we don't use TSC scaling
+		 * to avoid having stale value in the msr
+		 */
 		wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT);
 		__this_cpu_write(current_tsc_ratio, TSC_RATIO_DEFAULT);
 	}
···
 		if (npt_enabled)
 			kvm_cpu_cap_set(X86_FEATURE_NPT);
 
+		if (tsc_scaling)
+			kvm_cpu_cap_set(X86_FEATURE_TSCRATEMSR);
+
 		/* Nested VM can receive #VMEXIT instead of triggering #GP */
 		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
 	}
···
 	if (boot_cpu_has(X86_FEATURE_FXSR_OPT))
 		kvm_enable_efer_bits(EFER_FFXSR);
 
-	if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) {
-		kvm_has_tsc_control = true;
-		kvm_max_tsc_scaling_ratio = TSC_RATIO_MAX;
-		kvm_tsc_scaling_ratio_frac_bits = 32;
+	if (tsc_scaling) {
+		if (!boot_cpu_has(X86_FEATURE_TSCRATEMSR)) {
+			tsc_scaling = false;
+		} else {
+			pr_info("TSC scaling supported\n");
+			kvm_has_tsc_control = true;
+			kvm_max_tsc_scaling_ratio = TSC_RATIO_MAX;
+			kvm_tsc_scaling_ratio_frac_bits = 32;
+		}
 	}
 
 	tsc_aux_uret_slot = kvm_add_user_return_msr(MSR_TSC_AUX);
···
 			pr_info("Virtual GIF supported\n");
 	}
 
+	if (lbrv) {
+		if (!boot_cpu_has(X86_FEATURE_LBRV))
+			lbrv = false;
+		else
+			pr_info("LBR virtualization supported\n");
+	}
+
 	svm_set_cpu_caps();
 
 	/*
···
 
 static u64 svm_get_l2_tsc_multiplier(struct kvm_vcpu *vcpu)
 {
-	return kvm_default_tsc_scaling_ratio;
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	return svm->tsc_ratio_msr;
 }
 
 static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
···
 	vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
 }
 
-static void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu, u64 multiplier)
+void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu, u64 multiplier)
 {
 	wrmsrl(MSR_AMD64_TSC_RATIO, multiplier);
 }
···
 			svm_clr_intercept(svm, INTERCEPT_RDTSCP);
 		else
 			svm_set_intercept(svm, INTERCEPT_RDTSCP);
+	}
+}
+
+static inline void init_vmcb_after_set_cpuid(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	if (guest_cpuid_is_intel(vcpu)) {
+		/*
+		 * We must intercept SYSENTER_EIP and SYSENTER_ESP
+		 * accesses because the processor only stores 32 bits.
+		 * For the same reason we cannot use virtual VMLOAD/VMSAVE.
+		 */
+		svm_set_intercept(svm, INTERCEPT_VMLOAD);
+		svm_set_intercept(svm, INTERCEPT_VMSAVE);
+		svm->vmcb->control.virt_ext &= ~VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
+
+		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 0, 0);
+		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 0, 0);
+	} else {
+		/*
+		 * If hardware supports Virtual VMLOAD VMSAVE then enable it
+		 * in VMCB and clear intercepts to avoid #VMEXIT.
+		 */
+		if (vls) {
+			svm_clr_intercept(svm, INTERCEPT_VMLOAD);
+			svm_clr_intercept(svm, INTERCEPT_VMSAVE);
+			svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
+		}
+		/* No need to intercept these MSRs */
+		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 1, 1);
+		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 1, 1);
 	}
 }
 
···
 	}
 
 	svm_hv_init_vmcb(svm->vmcb);
+	init_vmcb_after_set_cpuid(vcpu);
 
 	vmcb_mark_all_dirty(svm->vmcb);
 
 	enable_gif(svm);
+}
 
+static void __svm_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	svm_vcpu_init_msrpm(vcpu, svm->msrpm);
+
+	svm_init_osvw(vcpu);
+	vcpu->arch.microcode_version = 0x01000065;
+	svm->tsc_ratio_msr = kvm_default_tsc_scaling_ratio;
+
+	if (sev_es_guest(vcpu->kvm))
+		sev_es_vcpu_reset(svm);
 }
 
 static void svm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
···
 	svm->virt_spec_ctrl = 0;
 
 	init_vmcb(vcpu);
+
+	if (!init_event)
+		__svm_vcpu_reset(vcpu);
 }
 
 void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb)
···
 
 	svm->vmcb01.ptr = page_address(vmcb01_page);
 	svm->vmcb01.pa = __sme_set(page_to_pfn(vmcb01_page) << PAGE_SHIFT);
+	svm_switch_vmcb(svm, &svm->vmcb01);
 
 	if (vmsa_page)
 		svm->vmsa = page_address(vmsa_page);
 
 	svm->guest_state_loaded = false;
-
-	svm_switch_vmcb(svm, &svm->vmcb01);
-	init_vmcb(vcpu);
-
-	svm_vcpu_init_msrpm(vcpu, svm->msrpm);
-
-	svm_init_osvw(vcpu);
-	vcpu->arch.microcode_version = 0x01000065;
-
-	if (sev_es_guest(vcpu->kvm))
-		/* Perform SEV-ES specific VMCB creation updates */
-		sev_es_create_vcpu(svm);
 
 	return 0;
 
···
 		vmsave(__sme_page_pa(sd->save_area));
 	}
 
-	if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) {
+	if (tsc_scaling) {
 		u64 tsc_ratio = vcpu->arch.tsc_scaling_ratio;
 		if (tsc_ratio != __this_cpu_read(current_tsc_ratio)) {
 			__this_cpu_write(current_tsc_ratio, tsc_ratio);
···
 	struct vcpu_svm *svm = to_svm(vcpu);
 
 	switch (msr_info->index) {
+	case MSR_AMD64_TSC_RATIO:
+		if (!msr_info->host_initiated && !svm->tsc_scaling_enabled)
+			return 1;
+		msr_info->data = svm->tsc_ratio_msr;
+		break;
 	case MSR_STAR:
 		msr_info->data = svm->vmcb01.ptr->save.star;
 		break;
···
 	u32 ecx = msr->index;
 	u64 data = msr->data;
 	switch (ecx) {
+	case MSR_AMD64_TSC_RATIO:
+		if (!msr->host_initiated && !svm->tsc_scaling_enabled)
+			return 1;
+
+		if (data & TSC_RATIO_RSVD)
+			return 1;
+
+		svm->tsc_ratio_msr = data;
+
+		if (svm->tsc_scaling_enabled && is_guest_mode(vcpu))
+			nested_svm_update_tsc_ratio_msr(vcpu);
+
+		break;
 	case MSR_IA32_CR_PAT:
 		if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data))
 			return 1;
···
 		svm->tsc_aux = data;
 		break;
 	case MSR_IA32_DEBUGCTLMSR:
-		if (!boot_cpu_has(X86_FEATURE_LBRV)) {
+		if (!lbrv) {
 			vcpu_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTL 0x%llx, nop\n",
 				    __func__, data);
 			break;
···
 	return svm_exit_handlers[exit_code](vcpu);
 }
 
-static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2,
+static void svm_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason,
+			      u64 *info1, u64 *info2,
 			      u32 *intr_info, u32 *error_code)
 {
 	struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control;
 
+	*reason = control->exit_code;
 	*info1 = control->exit_info_1;
 	*info2 = control->exit_info_2;
 	*intr_info = control->exit_int_info;
···
 	struct kvm_run *kvm_run = vcpu->run;
 	u32 exit_code = svm->vmcb->control.exit_code;
 
-	trace_kvm_exit(exit_code, vcpu, KVM_ISA_SVM);
+	trace_kvm_exit(vcpu, KVM_ISA_SVM);
 
 	/* SEV-ES guests must use the CR write traps to track CR registers. */
 	if (!sev_es_guest(vcpu->kvm)) {
···
 	if (is_guest_mode(vcpu)) {
 		int vmexit;
 
-		trace_kvm_nested_vmexit(exit_code, vcpu, KVM_ISA_SVM);
+		trace_kvm_nested_vmexit(vcpu, KVM_ISA_SVM);
 
 		vmexit = nested_svm_exit_special(svm);
 
···
 
 	pre_svm_run(vcpu);
 
-	WARN_ON_ONCE(kvm_apicv_activated(vcpu->kvm) != kvm_vcpu_apicv_active(vcpu));
-
 	sync_lapic_to_cr8(vcpu);
 
 	if (unlikely(svm->asid != svm->vmcb->control.asid)) {
···
 	svm->nrips_enabled = kvm_cpu_cap_has(X86_FEATURE_NRIPS) &&
 			     guest_cpuid_has(vcpu, X86_FEATURE_NRIPS);
 
+	svm->tsc_scaling_enabled = tsc_scaling && guest_cpuid_has(vcpu, X86_FEATURE_TSCRATEMSR);
+
 	svm_recalc_instruction_intercepts(vcpu, svm);
 
 	/* For sev guests, the memory encryption bit is not reserved in CR3.  */
···
 			kvm_request_apicv_update(vcpu->kvm, false,
 						 APICV_INHIBIT_REASON_NESTED);
 	}
-
-	if (guest_cpuid_is_intel(vcpu)) {
-		/*
-		 * We must intercept SYSENTER_EIP and SYSENTER_ESP
-		 * accesses because the processor only stores 32 bits.
-		 * For the same reason we cannot use virtual VMLOAD/VMSAVE.
-		 */
-		svm_set_intercept(svm, INTERCEPT_VMLOAD);
-		svm_set_intercept(svm, INTERCEPT_VMSAVE);
-		svm->vmcb->control.virt_ext &= ~VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
-
-		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 0, 0);
-		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 0, 0);
-	} else {
-		/*
-		 * If hardware supports Virtual VMLOAD VMSAVE then enable it
-		 * in VMCB and clear intercepts to avoid #VMEXIT.
-		 */
-		if (vls) {
-			svm_clr_intercept(svm, INTERCEPT_VMLOAD);
-			svm_clr_intercept(svm, INTERCEPT_VMSAVE);
-			svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
-		}
-		/* No need to intercept these MSRs */
-		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 1, 1);
-		set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 1, 1);
-	}
+	init_vmcb_after_set_cpuid(vcpu);
 }
 
 static bool svm_has_wbinvd_exit(void)
···
 }
 
 static struct kvm_x86_ops svm_x86_ops __initdata = {
+	.name = "kvm_amd",
+
 	.hardware_unsetup = svm_hardware_teardown,
 	.hardware_enable = svm_hardware_enable,
 	.hardware_disable = svm_hardware_disable,

arch/x86/kvm/svm/svm.h

···
 	u64 next_rip;
 
 	u64 spec_ctrl;
+
+	u64 tsc_ratio_msr;
 	/*
 	 * Contains guest-controlled bits of VIRT_SPEC_CTRL, which will be
 	 * translated into the appropriate L2_CFG bits on the host to
···
 	unsigned long int3_rip;
 
 	/* cached guest cpuid flags for faster access */
-	bool nrips_enabled	: 1;
+	bool nrips_enabled                : 1;
+	bool tsc_scaling_enabled          : 1;
 
 	u32 ldr_reg;
 	u32 dfr_reg;
···
 int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
 			       bool has_error_code, u32 error_code);
 int nested_svm_exit_special(struct vcpu_svm *svm);
+void nested_svm_update_tsc_ratio_msr(struct kvm_vcpu *vcpu);
+void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu, u64 multiplier);
 void nested_load_control_from_vmcb12(struct vcpu_svm *svm,
 				     struct vmcb_control_area *control);
 void nested_sync_control_from_vmcb02(struct vcpu_svm *svm);
···
 int sev_handle_vmgexit(struct kvm_vcpu *vcpu);
 int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in);
 void sev_es_init_vmcb(struct vcpu_svm *svm);
-void sev_es_create_vcpu(struct vcpu_svm *svm);
+void sev_es_vcpu_reset(struct vcpu_svm *svm);
 void sev_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector);
 void sev_es_prepare_guest_switch(struct vcpu_svm *svm, unsigned int cpu);
 void sev_es_unmap_ghcb(struct vcpu_svm *svm);

arch/x86/kvm/trace.h

···
 
 #define TRACE_EVENT_KVM_EXIT(name)					     \
 TRACE_EVENT(name,							     \
-	TP_PROTO(unsigned int exit_reason, struct kvm_vcpu *vcpu, u32 isa),  \
-	TP_ARGS(exit_reason, vcpu, isa),				     \
+	TP_PROTO(struct kvm_vcpu *vcpu, u32 isa),			     \
+	TP_ARGS(vcpu, isa),						     \
 									     \
 	TP_STRUCT__entry(						     \
 		__field(	unsigned int,	exit_reason	)	     \
···
 	),								     \
 									     \
 	TP_fast_assign(							     \
-		__entry->exit_reason	= exit_reason;			     \
 		__entry->guest_rip	= kvm_rip_read(vcpu);		     \
 		__entry->isa            = isa;				     \
 		__entry->vcpu_id        = vcpu->vcpu_id;		     \
-		static_call(kvm_x86_get_exit_info)(vcpu, &__entry->info1,    \
+		static_call(kvm_x86_get_exit_info)(vcpu,		     \
+					  &__entry->exit_reason,	     \
+					  &__entry->info1,		     \
 					  &__entry->info2,		     \
 					  &__entry->intr_info,		     \
 					  &__entry->error_code);	     \

arch/x86/kvm/vmx/nested.c

···
 	 * failValid writes the error number to the current VMCS, which
 	 * can't be done if there isn't a current VMCS.
 	 */
-	if (vmx->nested.current_vmptr == -1ull &&
+	if (vmx->nested.current_vmptr == INVALID_GPA &&
 	    !evmptr_is_valid(vmx->nested.hv_evmcs_vmptr))
 		return nested_vmx_failInvalid(vcpu);
 
···
 static void vmx_disable_shadow_vmcs(struct vcpu_vmx *vmx)
 {
 	secondary_exec_controls_clearbit(vmx, SECONDARY_EXEC_SHADOW_VMCS);
-	vmcs_write64(VMCS_LINK_POINTER, -1ull);
+	vmcs_write64(VMCS_LINK_POINTER, INVALID_GPA);
 	vmx->nested.need_vmcs12_to_shadow_sync = false;
 }
 
···
 
 	vmx->nested.vmxon = false;
 	vmx->nested.smm.vmxon = false;
+	vmx->nested.vmxon_ptr = INVALID_GPA;
 	free_vpid(vmx->nested.vpid02);
 	vmx->nested.posted_intr_nv = -1;
-	vmx->nested.current_vmptr = -1ull;
+	vmx->nested.current_vmptr = INVALID_GPA;
 	if (enable_shadow_vmcs) {
 		vmx_disable_shadow_vmcs(vmx);
 		vmcs_clear(vmx->vmcs01.shadow_vmcs);
···
 	struct vmcs12 *shadow;
 
 	if (!nested_cpu_has_shadow_vmcs(vmcs12) ||
-	    vmcs12->vmcs_link_pointer == -1ull)
+	    vmcs12->vmcs_link_pointer == INVALID_GPA)
 		return;
 
 	shadow = get_shadow_vmcs12(vcpu);
···
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
 	if (!nested_cpu_has_shadow_vmcs(vmcs12) ||
-	    vmcs12->vmcs_link_pointer == -1ull)
+	    vmcs12->vmcs_link_pointer == INVALID_GPA)
 		return;
 
 	kvm_write_guest(vmx->vcpu.kvm, vmcs12->vmcs_link_pointer,
···
 	}
 
 	if (unlikely(evmcs_gpa != vmx->nested.hv_evmcs_vmptr)) {
-		vmx->nested.current_vmptr = -1ull;
+		vmx->nested.current_vmptr = INVALID_GPA;
 
 		nested_release_evmcs(vcpu);
 
···
 	}
 
 	if (cpu_has_vmx_encls_vmexit())
-		vmcs_write64(ENCLS_EXITING_BITMAP, -1ull);
+		vmcs_write64(ENCLS_EXITING_BITMAP, INVALID_GPA);
 
 	/*
 	 * Set the MSR load/store lists to match L0's settings.  Only the
···
 {
 	prepare_vmcs02_constant_state(vmx);
 
-	vmcs_write64(VMCS_LINK_POINTER, -1ull);
+	vmcs_write64(VMCS_LINK_POINTER, INVALID_GPA);
 
 	if (enable_vpid) {
 		if (nested_cpu_has_vpid(vmcs12) && vmx->nested.vpid02)
···
 	struct vmcs12 *shadow;
 	struct kvm_host_map map;
 
-	if (vmcs12->vmcs_link_pointer == -1ull)
+	if (vmcs12->vmcs_link_pointer == INVALID_GPA)
 		return 0;
 
 	if (CC(!page_address_valid(vcpu, vmcs12->vmcs_link_pointer)))
···
 			 * Write an illegal value to VIRTUAL_APIC_PAGE_ADDR to
 			 * force VM-Entry to fail.
 			 */
-			vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, -1ull);
+			vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, INVALID_GPA);
 		}
 	}
 
···
 	}
 
 	if (CC(!evmptr_is_valid(vmx->nested.hv_evmcs_vmptr) &&
-	       vmx->nested.current_vmptr == -1ull))
+	       vmx->nested.current_vmptr == INVALID_GPA))
 		return nested_vmx_failInvalid(vcpu);
 
 	vmcs12 = get_vmcs12(vcpu);
···
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
-	if (vmx->nested.current_vmptr == -1ull)
+	if (vmx->nested.current_vmptr == INVALID_GPA)
 		return;
 
 	copy_vmcs02_to_vmcs12_rare(vcpu, get_vmcs12(vcpu));
···
 
 	kvm_mmu_free_roots(vcpu, &vcpu->arch.guest_mmu, KVM_MMU_ROOTS_ALL);
 
-	vmx->nested.current_vmptr = -1ull;
+	vmx->nested.current_vmptr = INVALID_GPA;
 }
 
 /* Emulate the VMXOFF instruction */
···
 		return 1;
 
 	/*
-	 * In VMX non-root operation, when the VMCS-link pointer is -1ull,
+	 * In VMX non-root operation, when the VMCS-link pointer is INVALID_GPA,
 	 * any VMREAD sets the ALU flags for VMfailInvalid.
 	 */
-	if (vmx->nested.current_vmptr == -1ull ||
+	if (vmx->nested.current_vmptr == INVALID_GPA ||
 	    (is_guest_mode(vcpu) &&
-	     get_vmcs12(vcpu)->vmcs_link_pointer == -1ull))
+	     get_vmcs12(vcpu)->vmcs_link_pointer == INVALID_GPA))
 		return nested_vmx_failInvalid(vcpu);
 
 	/* Decode instruction info and find the field to read */
···
 		return 1;
 
 	/*
-	 * In VMX non-root operation, when the VMCS-link pointer is -1ull,
+	 * In VMX non-root operation, when the VMCS-link pointer is INVALID_GPA,
 	 * any VMWRITE sets the ALU flags for VMfailInvalid.
 	 */
-	if (vmx->nested.current_vmptr == -1ull ||
+	if (vmx->nested.current_vmptr == INVALID_GPA ||
 	    (is_guest_mode(vcpu) &&
-	     get_vmcs12(vcpu)->vmcs_link_pointer == -1ull))
+	     get_vmcs12(vcpu)->vmcs_link_pointer == INVALID_GPA))
 		return nested_vmx_failInvalid(vcpu);
 
 	if (instr_info & BIT(10))
···
 	gpa_t bitmap, last_bitmap;
 	u8 b;
 
-	last_bitmap = (gpa_t)-1;
+	last_bitmap = INVALID_GPA;
 	b = -1;
 
 	while (size > 0) {
···
 		goto reflect_vmexit;
 	}
 
-	trace_kvm_nested_vmexit(exit_reason.full, vcpu, KVM_ISA_VMX);
+	trace_kvm_nested_vmexit(vcpu, KVM_ISA_VMX);
 
 	/* If L0 (KVM) wants the exit, it trumps L1's desires. */
 	if (nested_vmx_l0_wants_exit(vcpu, exit_reason))
···
 		.format = KVM_STATE_NESTED_FORMAT_VMX,
 		.size = sizeof(kvm_state),
 		.hdr.vmx.flags = 0,
-		.hdr.vmx.vmxon_pa = -1ull,
-		.hdr.vmx.vmcs12_pa = -1ull,
+		.hdr.vmx.vmxon_pa = INVALID_GPA,
+		.hdr.vmx.vmcs12_pa = INVALID_GPA,
 		.hdr.vmx.preemption_timer_deadline = 0,
 	};
 	struct kvm_vmx_nested_state_data __user *user_vmx_nested_state =
···
 
 			if (is_guest_mode(vcpu) &&
 			    nested_cpu_has_shadow_vmcs(vmcs12) &&
-			    vmcs12->vmcs_link_pointer != -1ull)
+			    vmcs12->vmcs_link_pointer != INVALID_GPA)
 				kvm_state.size += sizeof(user_vmx_nested_state->shadow_vmcs12);
 		}
 
···
 		return -EFAULT;
 
 	if (nested_cpu_has_shadow_vmcs(vmcs12) &&
-	    vmcs12->vmcs_link_pointer != -1ull) {
+	    vmcs12->vmcs_link_pointer != INVALID_GPA) {
 		if (copy_to_user(user_vmx_nested_state->shadow_vmcs12,
 				 get_shadow_vmcs12(vcpu), VMCS12_SIZE))
 			return -EFAULT;
···
 	if (kvm_state->format != KVM_STATE_NESTED_FORMAT_VMX)
 		return -EINVAL;
 
-	if (kvm_state->hdr.vmx.vmxon_pa == -1ull) {
+	if (kvm_state->hdr.vmx.vmxon_pa == INVALID_GPA) {
 		if (kvm_state->hdr.vmx.smm.flags)
 			return -EINVAL;
 
-		if (kvm_state->hdr.vmx.vmcs12_pa != -1ull)
+		if (kvm_state->hdr.vmx.vmcs12_pa != INVALID_GPA)
 			return -EINVAL;
 
 		/*
···
 
 	vmx_leave_nested(vcpu);
 
-	if (kvm_state->hdr.vmx.vmxon_pa == -1ull)
+	if (kvm_state->hdr.vmx.vmxon_pa == INVALID_GPA)
 		return 0;
 
 	vmx->nested.vmxon_ptr = kvm_state->hdr.vmx.vmxon_pa;
···
 		/* See vmx_has_valid_vmcs12.  */
 		if ((kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE) ||
 		    (kvm_state->flags & KVM_STATE_NESTED_EVMCS) ||
-		    (kvm_state->hdr.vmx.vmcs12_pa != -1ull))
+		    (kvm_state->hdr.vmx.vmcs12_pa != INVALID_GPA))
 			return -EINVAL;
 		else
 			return 0;
 	}
 
-	if (kvm_state->hdr.vmx.vmcs12_pa != -1ull) {
+	if (kvm_state->hdr.vmx.vmcs12_pa != INVALID_GPA) {
 		if (kvm_state->hdr.vmx.vmcs12_pa == kvm_state->hdr.vmx.vmxon_pa ||
 		    !page_address_valid(vcpu, kvm_state->hdr.vmx.vmcs12_pa))
 			return -EINVAL;
···
 
 	ret = -EINVAL;
 	if (nested_cpu_has_shadow_vmcs(vmcs12) &&
-	    vmcs12->vmcs_link_pointer != -1ull) {
+	    vmcs12->vmcs_link_pointer != INVALID_GPA) {
 		struct vmcs12 *shadow_vmcs12 = get_shadow_vmcs12(vcpu);
 
 		if (kvm_state->size <

arch/x86/kvm/vmx/pmu_intel.c

···
 		msr_info->data = pmu->global_ctrl;
 		return 0;
 	case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
-		msr_info->data = pmu->global_ovf_ctrl;
+		msr_info->data = 0;
 		return 0;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
···
 		if (!(data & pmu->global_ovf_ctrl_mask)) {
 			if (!msr_info->host_initiated)
 				pmu->global_status &= ~data;
-			pmu->global_ovf_ctrl = data;
 			return 0;
 		}
 		break;
···
 		pmc->counter = 0;
 	}
 
-	pmu->fixed_ctr_ctrl = pmu->global_ctrl = pmu->global_status =
-		pmu->global_ovf_ctrl = 0;
+	pmu->fixed_ctr_ctrl = pmu->global_ctrl = pmu->global_status = 0;
 
 	intel_pmu_release_guest_lbr_event(vcpu);
 }

arch/x86/kvm/vmx/sgx.c

···
 static void sgx_handle_emulation_failure(struct kvm_vcpu *vcpu, u64 addr,
 					 unsigned int size)
 {
-	vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-	vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
-	vcpu->run->internal.ndata = 2;
-	vcpu->run->internal.data[0] = addr;
-	vcpu->run->internal.data[1] = size;
+	uint64_t data[2] = { addr, size };
+
+	__kvm_prepare_emulation_failure_exit(vcpu, data, ARRAY_SIZE(data));
 }
 
 static int sgx_read_hva(struct kvm_vcpu *vcpu, unsigned long hva, void *data,
···
 	 * but the error code isn't (yet) plumbed through the ENCLS helpers.
 	 */
 	if (trapnr == PF_VECTOR && !boot_cpu_has(X86_FEATURE_SGX2)) {
-		vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-		vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
-		vcpu->run->internal.ndata = 0;
+		kvm_prepare_emulation_failure_exit(vcpu);
 		return 0;
 	}
 
···
 	sgx_12_0 = kvm_find_cpuid_entry(vcpu, 0x12, 0);
 	sgx_12_1 = kvm_find_cpuid_entry(vcpu, 0x12, 1);
 	if (!sgx_12_0 || !sgx_12_1) {
-		vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-		vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
-		vcpu->run->internal.ndata = 0;
+		kvm_prepare_emulation_failure_exit(vcpu);
 		return 0;
 	}
 

arch/x86/kvm/vmx/vmx.c

···
 	rdmsrl(MSR_IA32_RTIT_CTL, vmx->pt_desc.host.ctl);
 	if (vmx->pt_desc.guest.ctl & RTIT_CTL_TRACEEN) {
 		wrmsrl(MSR_IA32_RTIT_CTL, 0);
-		pt_save_msr(&vmx->pt_desc.host, vmx->pt_desc.addr_range);
-		pt_load_msr(&vmx->pt_desc.guest, vmx->pt_desc.addr_range);
+		pt_save_msr(&vmx->pt_desc.host, vmx->pt_desc.num_address_ranges);
+		pt_load_msr(&vmx->pt_desc.guest, vmx->pt_desc.num_address_ranges);
 	}
 }
 
···
 		return;
 
 	if (vmx->pt_desc.guest.ctl & RTIT_CTL_TRACEEN) {
-		pt_save_msr(&vmx->pt_desc.guest, vmx->pt_desc.addr_range);
-		pt_load_msr(&vmx->pt_desc.host, vmx->pt_desc.addr_range);
+		pt_save_msr(&vmx->pt_desc.guest, vmx->pt_desc.num_address_ranges);
+		pt_load_msr(&vmx->pt_desc.host, vmx->pt_desc.num_address_ranges);
 	}
 
-	/* Reload host state (IA32_RTIT_CTL will be cleared on VM exit). */
-	wrmsrl(MSR_IA32_RTIT_CTL, vmx->pt_desc.host.ctl);
+	/*
+	 * KVM requires VM_EXIT_CLEAR_IA32_RTIT_CTL to expose PT to the guest,
+	 * i.e. RTIT_CTL is always cleared on VM-Exit.  Restore it if necessary.
+	 */
+	if (vmx->pt_desc.host.ctl)
+		wrmsrl(MSR_IA32_RTIT_CTL, vmx->pt_desc.host.ctl);
 }
 
 void vmx_set_host_fs_gs(struct vmcs_host_state *host, u16 fs_sel, u16 gs_sel,
···
 	 * cause a #GP fault.
 	 */
 	value = (data & RTIT_CTL_ADDR0) >> RTIT_CTL_ADDR0_OFFSET;
-	if ((value && (vmx->pt_desc.addr_range < 1)) || (value > 2))
+	if ((value && (vmx->pt_desc.num_address_ranges < 1)) || (value > 2))
 		return 1;
 	value = (data & RTIT_CTL_ADDR1) >> RTIT_CTL_ADDR1_OFFSET;
-	if ((value && (vmx->pt_desc.addr_range < 2)) || (value > 2))
+	if ((value && (vmx->pt_desc.num_address_ranges < 2)) || (value > 2))
 		return 1;
 	value = (data & RTIT_CTL_ADDR2) >> RTIT_CTL_ADDR2_OFFSET;
-	if ((value && (vmx->pt_desc.addr_range < 3)) || (value > 2))
+	if ((value && (vmx->pt_desc.num_address_ranges < 3)) || (value > 2))
 		return 1;
 	value = (data & RTIT_CTL_ADDR3) >> RTIT_CTL_ADDR3_OFFSET;
-	if ((value && (vmx->pt_desc.addr_range < 4)) || (value > 2))
+	if ((value && (vmx->pt_desc.num_address_ranges < 4)) || (value > 2))
 		return 1;
 
 	return 0;
···
 	case MSR_IA32_RTIT_ADDR0_A ... MSR_IA32_RTIT_ADDR3_B:
 		index = msr_info->index - MSR_IA32_RTIT_ADDR0_A;
 		if (!vmx_pt_mode_is_host_guest() ||
-			(index >= 2 * intel_pt_validate_cap(vmx->pt_desc.caps,
-					PT_CAP_num_address_ranges)))
+		    (index >= 2 * vmx->pt_desc.num_address_ranges))
 			return 1;
 		if (index % 2)
 			msr_info->data = vmx->pt_desc.guest.addr_b[index / 2];
···
 		if (!pt_can_write_msr(vmx))
 			return 1;
 		index = msr_info->index - MSR_IA32_RTIT_ADDR0_A;
-		if (index >= 2 * intel_pt_validate_cap(vmx->pt_desc.caps,
-						       PT_CAP_num_address_ranges))
+		if (index >= 2 * vmx->pt_desc.num_address_ranges)
 			return 1;
 		if (is_noncanonical_address(data, vcpu))
 			return 1;
···
 	vmx_set_intercept_for_msr(vcpu, MSR_IA32_RTIT_OUTPUT_BASE, MSR_TYPE_RW, flag);
 	vmx_set_intercept_for_msr(vcpu, MSR_IA32_RTIT_OUTPUT_MASK, MSR_TYPE_RW, flag);
 	vmx_set_intercept_for_msr(vcpu, MSR_IA32_RTIT_CR3_MATCH, MSR_TYPE_RW, flag);
-	for (i = 0; i < vmx->pt_desc.addr_range; i++) {
+	for (i = 0; i < vmx->pt_desc.num_address_ranges; i++) {
 		vmx_set_intercept_for_msr(vcpu, MSR_IA32_RTIT_ADDR0_A + i * 2, MSR_TYPE_RW, flag);
 		vmx_set_intercept_for_msr(vcpu, MSR_IA32_RTIT_ADDR0_B + i * 2, MSR_TYPE_RW, flag);
 	}
···
 
 #define VMX_XSS_EXIT_BITMAP 0
 
-/*
- * Noting that the initialization of Guest-state Area of VMCS is in
- * vmx_vcpu_reset().
- */
 static void init_vmcs(struct vcpu_vmx *vmx)
 {
 	if (nested)
···
 	if (cpu_has_vmx_msr_bitmap())
 		vmcs_write64(MSR_BITMAP, __pa(vmx->vmcs01.msr_bitmap));
 
-	vmcs_write64(VMCS_LINK_POINTER, -1ull); /* 22.3.1.5 */
+	vmcs_write64(VMCS_LINK_POINTER, INVALID_GPA); /* 22.3.1.5 */
 
 	/* Control */
 	pin_controls_set(vmx, vmx_pin_based_exec_ctrl(vmx));
···
 	vmx_setup_uret_msrs(vmx);
 }
 
+static void __vmx_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	init_vmcs(vmx);
+
+	if (nested)
+		memcpy(&vmx->nested.msrs, &vmcs_config.nested, sizeof(vmx->nested.msrs));
+
+	vcpu_setup_sgx_lepubkeyhash(vcpu);
+
+	vmx->nested.posted_intr_nv = -1;
+	vmx->nested.vmxon_ptr = INVALID_GPA;
+	vmx->nested.current_vmptr = INVALID_GPA;
+	vmx->nested.hv_evmcs_vmptr = EVMPTR_INVALID;
+
+	vcpu->arch.microcode_version = 0x100000000ULL;
+	vmx->msr_ia32_feature_control_valid_bits = FEAT_CTL_LOCKED;
+
+	/*
+	 * Enforce invariant: pi_desc.nv is always either POSTED_INTR_VECTOR
+	 * or POSTED_INTR_WAKEUP_VECTOR.
+	 */
+	vmx->pi_desc.nv = POSTED_INTR_VECTOR;
+	vmx->pi_desc.sn = 1;
+}
+
 static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+	if (!init_event)
+		__vmx_vcpu_reset(vcpu);
 
 	vmx->rmode.vm86_active = 0;
 	vmx->spec_ctrl = 0;
···
 	kvm_set_cr8(vcpu, 0);
 
 	vmx_segment_cache_clear(vmx);
+	kvm_register_mark_available(vcpu, VCPU_EXREG_SEGMENTS);
 
 	seg_setup(VCPU_SREG_CS);
 	vmcs_write16(GUEST_CS_SELECTOR, 0xf000);
···
 
 		if (vmx->emulation_required && !vmx->rmode.vm86_active &&
 		    vcpu->arch.exception.pending) {
-			vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-			vcpu->run->internal.suberror =
-						KVM_INTERNAL_ERROR_EMULATION;
-			vcpu->run->internal.ndata = 0;
+			kvm_prepare_emulation_failure_exit(vcpu);
 			return 0;
 		}
 
···
 static const int kvm_vmx_max_exit_handlers =
 	ARRAY_SIZE(kvm_vmx_exit_handlers);
 
-static void vmx_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2,
+static void vmx_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason,
+			      u64 *info1, u64 *info2,
 			      u32 *intr_info, u32 *error_code)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
+	*reason = vmx->exit_reason.full;
 	*info1 = vmx_get_exit_qual(vcpu);
 	if (!(vmx->exit_reason.failed_vmentry)) {
 		*info2 = vmx->idt_vectoring_info;
···
 	case MSR_IA32_VMX_BASIC ... MSR_IA32_VMX_VMFUNC:
 		return nested;
 	case MSR_AMD64_VIRT_SPEC_CTRL:
+	case MSR_AMD64_TSC_RATIO:
 		/* This is AMD only.  */
 		return false;
 	default:
···
 	if (likely(!vmx->exit_reason.failed_vmentry))
 		vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);
 
-	trace_kvm_exit(vmx->exit_reason.full, vcpu, KVM_ISA_VMX);
+	trace_kvm_exit(vcpu, KVM_ISA_VMX);
 
 	if (unlikely(vmx->exit_reason.failed_vmentry))
 		return EXIT_FASTPATH_NONE;
···
 {
 	struct vmx_uret_msr *tsx_ctrl;
 	struct vcpu_vmx *vmx;
-	int i, cpu, err;
+	int i, err;
 
 	BUILD_BUG_ON(offsetof(struct vcpu_vmx, vcpu) != 0);
 	vmx = to_vmx(vcpu);
···
 			goto free_vpid;
 	}
 
-	for (i = 0; i < kvm_nr_uret_msrs; ++i) {
-		vmx->guest_uret_msrs[i].data = 0;
+	for (i = 0; i < kvm_nr_uret_msrs; ++i)
 		vmx->guest_uret_msrs[i].mask = -1ull;
-	}
 	if (boot_cpu_has(X86_FEATURE_RTM)) {
 		/*
 		 * TSX_CTRL_CPUID_CLEAR is handled in the CPUID interception.
···
 	}
 
 	vmx->loaded_vmcs = &vmx->vmcs01;
-	cpu = get_cpu();
-	vmx_vcpu_load(vcpu, cpu);
-	vcpu->cpu = cpu;
-	init_vmcs(vmx);
-	vmx_vcpu_put(vcpu);
-	put_cpu();
+
 	if (cpu_need_virtualize_apic_accesses(vcpu)) {
 		err = alloc_apic_access_page(vcpu->kvm);
 		if (err)
···
 		if (err)
 			goto free_vmcs;
 	}
-
-	if (nested)
-		memcpy(&vmx->nested.msrs, &vmcs_config.nested, sizeof(vmx->nested.msrs));
-	else
-		memset(&vmx->nested.msrs, 0, sizeof(vmx->nested.msrs));
-
-	vcpu_setup_sgx_lepubkeyhash(vcpu);
-
-	vmx->nested.posted_intr_nv = -1;
-	vmx->nested.current_vmptr = -1ull;
-	vmx->nested.hv_evmcs_vmptr = EVMPTR_INVALID;
-
-	vcpu->arch.microcode_version = 0x100000000ULL;
-	vmx->msr_ia32_feature_control_valid_bits = FEAT_CTL_LOCKED;
-
-	/*
-	 * Enforce invariant: pi_desc.nv is always either POSTED_INTR_VECTOR
-	 * or POSTED_INTR_WAKEUP_VECTOR.
-	 */
-	vmx->pi_desc.nv = POSTED_INTR_VECTOR;
-	vmx->pi_desc.sn = 1;
 
 	return 0;
 
···
 	}
 
 	/* Get the number of configurable Address Ranges for filtering */
-	vmx->pt_desc.addr_range = intel_pt_validate_cap(vmx->pt_desc.caps,
+	vmx->pt_desc.num_address_ranges = intel_pt_validate_cap(vmx->pt_desc.caps,
 						PT_CAP_num_address_ranges);
 
 	/* Initialize and clear the no dependency bits */
 	vmx->pt_desc.ctl_bitmask = ~(RTIT_CTL_TRACEEN | RTIT_CTL_OS |
-			RTIT_CTL_USR | RTIT_CTL_TSC_EN | RTIT_CTL_DISRETC);
+			RTIT_CTL_USR | RTIT_CTL_TSC_EN | RTIT_CTL_DISRETC |
+			RTIT_CTL_BRANCH_EN);
 
 	/*
 	 * If CPUID.(EAX=14H,ECX=0):EBX[0]=1 CR3Filter can be set otherwise
···
 				RTIT_CTL_CYC_THRESH | RTIT_CTL_PSB_FREQ);
 
 	/*
-	 * If CPUID.(EAX=14H,ECX=0):EBX[3]=1 MTCEn BranchEn and
-	 * MTCFreq can be set
+	 * If CPUID.(EAX=14H,ECX=0):EBX[3]=1 MTCEn and MTCFreq can be set
 	 */
 	if (intel_pt_validate_cap(vmx->pt_desc.caps, PT_CAP_mtc))
 		vmx->pt_desc.ctl_bitmask &= ~(RTIT_CTL_MTC_EN |
-				RTIT_CTL_BRANCH_EN | RTIT_CTL_MTC_RANGE);
+					      RTIT_CTL_MTC_RANGE);
 
 	/* If CPUID.(EAX=14H,ECX=0):EBX[4]=1 FUPonPTW and PTWEn can be set */
 	if (intel_pt_validate_cap(vmx->pt_desc.caps, PT_CAP_ptwrite))
···
 		vmx->pt_desc.ctl_bitmask &= ~RTIT_CTL_FABRIC_EN;
 
 	/* unmask address range configure area */
-	for (i = 0; i < vmx->pt_desc.addr_range; i++)
+	for (i = 0; i < vmx->pt_desc.num_address_ranges; i++)
 		vmx->pt_desc.ctl_bitmask &= ~(0xfULL << (32 + i * 4));
 }
 
···
 
 static void hardware_unsetup(void)
 {
+	kvm_set_posted_intr_wakeup_handler(NULL);
+
 	if (nested)
 		nested_vmx_hardware_unsetup();
 
···
 }
 
 static struct kvm_x86_ops vmx_x86_ops __initdata = {
+	.name = "kvm_intel",
+
 	.hardware_unsetup = hardware_unsetup,
 
 	.hardware_enable = hardware_enable,
···
 		vmx_x86_ops.request_immediate_exit = __kvm_request_immediate_exit;
 	}
 
-	kvm_set_posted_intr_wakeup_handler(pi_wakeup_handler);
-
 	kvm_mce_cap_supported |= MCG_LMCE_P;
 
 	if (pt_mode != PT_MODE_SYSTEM && pt_mode != PT_MODE_HOST_GUEST)
···
 	r = alloc_kvm_area();
 	if (r)
 		nested_vmx_hardware_unsetup();
+
+	kvm_set_posted_intr_wakeup_handler(pi_wakeup_handler);
+
 	return r;
 }
 

arch/x86/kvm/vmx/vmx.h

···
 
 struct pt_desc {
 	u64 ctl_bitmask;
-	u32 addr_range;
+	u32 num_address_ranges;
 	u32 caps[PT_CPUID_REGS_NUM * PT_CPUID_LEAVES];
 	struct pt_ctx host;
 	struct pt_ctx guest;

arch/x86/kvm/x86.c

···
 }
 EXPORT_SYMBOL_GPL(kvm_require_dr);
 
-/*
- * This function will be used to read from the physical memory of the currently
- * running guest. The difference to kvm_vcpu_read_guest_page is that this function
- * can read from guest physical or from the guest's guest physical memory.
- */
-int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
-			    gfn_t ngfn, void *data, int offset, int len,
-			    u32 access)
-{
-	struct x86_exception exception;
-	gfn_t real_gfn;
-	gpa_t ngpa;
-
-	ngpa     = gfn_to_gpa(ngfn);
-	real_gfn = mmu->translate_gpa(vcpu, ngpa, access, &exception);
-	if (real_gfn == UNMAPPED_GVA)
-		return -EFAULT;
-
-	real_gfn = gpa_to_gfn(real_gfn);
-
-	return kvm_vcpu_read_guest_page(vcpu, real_gfn, data, offset, len);
-}
-EXPORT_SYMBOL_GPL(kvm_read_guest_page_mmu);
-
 static inline u64 pdptr_rsvd_bits(struct kvm_vcpu *vcpu)
 {
 	return vcpu->arch.reserved_gpa_bits | rsvd_bits(5, 8) | rsvd_bits(1, 2);
···
 int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3)
 {
 	gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT;
-	unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2;
+	gpa_t real_gpa;
 	int i;
 	int ret;
 	u64 pdpte[ARRAY_SIZE(mmu->pdptrs)];
 
-	ret = kvm_read_guest_page_mmu(vcpu, mmu, pdpt_gfn, pdpte,
-				      offset * sizeof(u64), sizeof(pdpte),
-				      PFERR_USER_MASK|PFERR_WRITE_MASK);
-	if (ret < 0) {
-		ret = 0;
-		goto out;
-	}
+	/*
+	 * If the MMU is nested, CR3 holds an L2 GPA and needs to be translated
+	 * to an L1 GPA.
+	 */
+	real_gpa = mmu->translate_gpa(vcpu, gfn_to_gpa(pdpt_gfn),
+				      PFERR_USER_MASK | PFERR_WRITE_MASK, NULL);
+	if (real_gpa == UNMAPPED_GVA)
+		return 0;
+
+	/* Note the offset, PDPTRs are 32 byte aligned when using PAE paging. */
+	ret = kvm_vcpu_read_guest_page(vcpu, gpa_to_gfn(real_gpa), pdpte,
+				       cr3 & GENMASK(11, 5), sizeof(pdpte));
+	if (ret < 0)
+		return 0;
+
 	for (i = 0; i < ARRAY_SIZE(pdpte); ++i) {
 		if ((pdpte[i] & PT_PRESENT_MASK) &&
 		    (pdpte[i] & pdptr_rsvd_bits(vcpu))) {
-			ret = 0;
-			goto out;
+			return 0;
 		}
 	}
-	ret = 1;
 
 	memcpy(mmu->pdptrs, pdpte, sizeof(mmu->pdptrs));
 	kvm_register_mark_dirty(vcpu, VCPU_EXREG_PDPTR);
 	vcpu->arch.pdptrs_from_userspace = false;
 
-out:
-
-	return ret;
+	return 1;
 }
 EXPORT_SYMBOL_GPL(load_pdptrs);
 
···
 	/*
 	 * Do not allow the guest to set bits that we do not support
 	 * saving.  However, xcr0 bit 0 is always set, even if the
-	 * emulated CPU does not support XSAVE (see fx_init).
+	 * emulated CPU does not support XSAVE (see kvm_vcpu_reset()).
 	 */
 	valid_bits = vcpu->arch.guest_supported_xcr0 | XFEATURE_MASK_FP;
 	if (xcr0 & ~valid_bits)
···
 
 void kvm_post_set_cr4(struct kvm_vcpu *vcpu, unsigned long old_cr4, unsigned long cr4)
 {
-	if (((cr4 ^ old_cr4) & KVM_MMU_CR4_ROLE_BITS) ||
-	    (!(cr4 & X86_CR4_PCIDE) && (old_cr4 & X86_CR4_PCIDE)))
+	/*
+	 * If any role bit is changed, the MMU needs to be reset.
+	 *
+	 * If CR4.PCIDE is changed 1 -> 0, the guest TLB must be flushed.
+	 * If CR4.PCIDE is changed 0 -> 1, there is no need to flush the TLB
+	 * according to the SDM; however, stale prev_roots could be reused
+	 * incorrectly in the future after a MOV to CR3 with NOFLUSH=1, so we
+	 * free them all.  KVM_REQ_MMU_RELOAD is fit for the both cases; it
+	 * is slow, but changing CR4.PCIDE is a rare case.
+	 *
+	 * If CR4.PGE is changed, the guest TLB must be flushed.
+	 *
+	 * Note: resetting MMU is a superset of KVM_REQ_MMU_RELOAD and
+	 * KVM_REQ_MMU_RELOAD is a superset of KVM_REQ_TLB_FLUSH_GUEST, hence
+	 * the usage of "else if".
+	 */
+	if ((cr4 ^ old_cr4) & KVM_MMU_CR4_ROLE_BITS)
 		kvm_mmu_reset_context(vcpu);
+	else if ((cr4 ^ old_cr4) & X86_CR4_PCIDE)
+		kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
+	else if ((cr4 ^ old_cr4) & X86_CR4_PGE)
+		kvm_make_request(KVM_REQ_TLB_FLUSH_GUEST, vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_post_set_cr4);
 
···
 	int i;
 
 	/*
+	 * MOV CR3 and INVPCID are usually not intercepted when using TDP, but
+	 * this is reachable when running EPT=1 and unrestricted_guest=0,  and
+	 * also via the emulator.  KVM's TDP page tables are not in the scope of
+	 * the invalidation, but the guest's TLB entries need to be flushed as
+	 * the CPU may have cached entries in its TLB for the target PCID.
+	 */
+	if (unlikely(tdp_enabled)) {
+		kvm_make_request(KVM_REQ_TLB_FLUSH_GUEST, vcpu);
+		return;
+	}
+
+	/*
 	 * If neither the current CR3 nor any of the prev_roots use the given
 	 * PCID, then nothing needs to be done here because a resync will
 	 * happen anyway before switching to any other CR3.
···
 		kvm_make_request(KVM_REQ_MMU_SYNC, vcpu);
 		kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
 	}
+
+	/*
+	 * If PCID is disabled, there is no need to free prev_roots even if the
+	 * PCIDs for them are also 0, because MOV to CR3 always flushes the TLB
+	 * with PCIDE=0.
+	 */
+	if (!kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
+		return;
 
 	for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++)
 		if (kvm_get_pcid(vcpu, mmu->prev_roots[i].pgd) == pcid)
···
 	MSR_PLATFORM_INFO,
 	MSR_MISC_FEATURES_ENABLES,
 	MSR_AMD64_VIRT_SPEC_CTRL,
+	MSR_AMD64_TSC_RATIO,
 	MSR_IA32_POWER_CTL,
 	MSR_IA32_UCODE_REV,
 
···
 	return check_tsc_unstable();
 }
 
+/*
+ * Infers attempts to synchronize the guest's tsc from host writes. Sets the
+ * offset for the vcpu and tracks the TSC matching generation that the vcpu
+ * participates in.
+ */
+static void __kvm_synchronize_tsc(struct kvm_vcpu *vcpu, u64 offset, u64 tsc,
+				  u64 ns, bool matched)
+{
+	struct kvm *kvm = vcpu->kvm;
+
+	lockdep_assert_held(&kvm->arch.tsc_write_lock);
+
+	/*
+	 * We also track th most recent recorded KHZ, write and time to
+	 * allow the matching interval to be extended at each write.
+	 */
+	kvm->arch.last_tsc_nsec = ns;
+	kvm->arch.last_tsc_write = tsc;
+	kvm->arch.last_tsc_khz = vcpu->arch.virtual_tsc_khz;
+	kvm->arch.last_tsc_offset = offset;
+
+	vcpu->arch.last_guest_tsc = tsc;
+
+	kvm_vcpu_write_tsc_offset(vcpu, offset);
+
+	if (!matched) {
+		/*
+		 * We split periods of matched TSC writes into generations.
+		 * For each generation, we track the original measured
+		 * nanosecond time, offset, and write, so if TSCs are in
+		 * sync, we can match exact offset, and if not, we can match
+		 * exact software computation in compute_guest_tsc()
+		 *
+		 * These values are tracked in kvm->arch.cur_xxx variables.
+		 */
+		kvm->arch.cur_tsc_generation++;
+		kvm->arch.cur_tsc_nsec = ns;
+		kvm->arch.cur_tsc_write = tsc;
+		kvm->arch.cur_tsc_offset = offset;
+		kvm->arch.nr_vcpus_matched_tsc = 0;
+	} else if (vcpu->arch.this_tsc_generation != kvm->arch.cur_tsc_generation) {
+		kvm->arch.nr_vcpus_matched_tsc++;
+	}
+
+	/* Keep track of which generation this VCPU has synchronized to */
+	vcpu->arch.this_tsc_generation = kvm->arch.cur_tsc_generation;
+	vcpu->arch.this_tsc_nsec = kvm->arch.cur_tsc_nsec;
+	vcpu->arch.this_tsc_write = kvm->arch.cur_tsc_write;
+
+	kvm_track_tsc_matching(vcpu);
+}
+
 static void kvm_synchronize_tsc(struct kvm_vcpu *vcpu, u64 data)
 {
 	struct kvm *kvm = vcpu->kvm;
 	u64 offset, ns, elapsed;
 	unsigned long flags;
-	bool matched;
-	bool already_matched;
+	bool matched = false;
 	bool synchronizing = false;
 
 	raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags);
···
 			offset = kvm_compute_l1_tsc_offset(vcpu, data);
 		}
 		matched = true;
-		already_matched = (vcpu->arch.this_tsc_generation == kvm->arch.cur_tsc_generation);
-	} else {
-		/*
-		 * We split periods of matched TSC writes into generations.
-		 * For each generation, we track the original measured
-		 * nanosecond time, offset, and write, so if TSCs are in
-		 * sync, we can match exact offset, and if not, we can match
-		 * exact software computation in compute_guest_tsc()
-		 *
-		 * These values are tracked in kvm->arch.cur_xxx variables.
-		 */
-		kvm->arch.cur_tsc_generation++;
-		kvm->arch.cur_tsc_nsec = ns;
-		kvm->arch.cur_tsc_write = data;
-		kvm->arch.cur_tsc_offset = offset;
-		matched = false;
 	}
 
-	/*
-	 * We also track th most recent recorded KHZ, write and time to
-	 * allow the matching interval to be extended at each write.
-	 */
-	kvm->arch.last_tsc_nsec = ns;
-	kvm->arch.last_tsc_write = data;
-	kvm->arch.last_tsc_khz = vcpu->arch.virtual_tsc_khz;
-
-	vcpu->arch.last_guest_tsc = data;
-
-	/* Keep track of which generation this VCPU has synchronized to */
-	vcpu->arch.this_tsc_generation = kvm->arch.cur_tsc_generation;
-	vcpu->arch.this_tsc_nsec = kvm->arch.cur_tsc_nsec;
-	vcpu->arch.this_tsc_write = kvm->arch.cur_tsc_write;
-
-	kvm_vcpu_write_tsc_offset(vcpu, offset);
+	__kvm_synchronize_tsc(vcpu, offset, data, ns, matched);
 	raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags);
-
-	raw_spin_lock_irqsave(&kvm->arch.pvclock_gtod_sync_lock, flags);
-	if (!matched) {
-		kvm->arch.nr_vcpus_matched_tsc = 0;
-	} else if (!already_matched) {
-		kvm->arch.nr_vcpus_matched_tsc++;
-	}
-
-	kvm_track_tsc_matching(vcpu);
-	raw_spin_unlock_irqrestore(&kvm->arch.pvclock_gtod_sync_lock, flags);
 }
 
 static inline void adjust_tsc_offset_guest(struct kvm_vcpu *vcpu,
···
 	int vclock_mode;
 	bool host_tsc_clocksource, vcpus_matched;
 
+	lockdep_assert_held(&kvm->arch.tsc_write_lock);
 	vcpus_matched = (ka->nr_vcpus_matched_tsc + 1 ==
 			atomic_read(&kvm->online_vcpus));
 
···
 #endif
 }
 
-void kvm_make_mclock_inprogress_request(struct kvm *kvm)
+static void kvm_make_mclock_inprogress_request(struct kvm *kvm)
 {
 	kvm_make_all_cpus_request(kvm, KVM_REQ_MCLOCK_INPROGRESS);
 }
 
-static void kvm_gen_update_masterclock(struct kvm *kvm)
+static void __kvm_start_pvclock_update(struct kvm *kvm)
 {
-#ifdef CONFIG_X86_64
-	int i;
-	struct kvm_vcpu *vcpu;
-	struct kvm_arch *ka = &kvm->arch;
-	unsigned long flags;
+	raw_spin_lock_irq(&kvm->arch.tsc_write_lock);
+	write_seqcount_begin(&kvm->arch.pvclock_sc);
+}
 
-	kvm_hv_invalidate_tsc_page(kvm);
-
+static void kvm_start_pvclock_update(struct kvm *kvm)
+{
 	kvm_make_mclock_inprogress_request(kvm);
 
 	/* no guest entries from this point */
-	raw_spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
-	pvclock_update_vm_gtod_copy(kvm);
-	raw_spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
+	__kvm_start_pvclock_update(kvm);
+}
 
+static void kvm_end_pvclock_update(struct kvm *kvm)
+{
+	struct kvm_arch *ka = &kvm->arch;
+	struct kvm_vcpu *vcpu;
+	int i;
+
+	write_seqcount_end(&ka->pvclock_sc);
+	raw_spin_unlock_irq(&ka->tsc_write_lock);
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 
 	/* guest entries allowed */
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		kvm_clear_request(KVM_REQ_MCLOCK_INPROGRESS, vcpu);
-#endif
 }
 
-u64 get_kvmclock_ns(struct kvm *kvm)
+static void kvm_update_masterclock(struct kvm *kvm)
+{
+	kvm_hv_invalidate_tsc_page(kvm);
+	kvm_start_pvclock_update(kvm);
+	pvclock_update_vm_gtod_copy(kvm);
+	kvm_end_pvclock_update(kvm);
+}
+
+/* Called within read_seqcount_begin/retry for kvm->pvclock_sc.  */
+static void __get_kvmclock(struct kvm *kvm, struct kvm_clock_data *data)
 {
 	struct kvm_arch *ka = &kvm->arch;
 	struct pvclock_vcpu_time_info hv_clock;
-	unsigned long flags;
-	u64 ret;
-
-	raw_spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
-	if (!ka->use_master_clock) {
-		raw_spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
-		return get_kvmclock_base_ns() + ka->kvmclock_offset;
-	}
-
-	hv_clock.tsc_timestamp = ka->master_cycle_now;
-	hv_clock.system_time = ka->master_kernel_ns + ka->kvmclock_offset;
-	raw_spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
 
 	/* both __this_cpu_read() and rdtsc() should be on the same cpu */
 	get_cpu();
 
-	if (__this_cpu_read(cpu_tsc_khz)) {
+	data->flags = 0;
+	if (ka->use_master_clock && __this_cpu_read(cpu_tsc_khz)) {
+#ifdef CONFIG_X86_64
+		struct timespec64 ts;
+
+		if (kvm_get_walltime_and_clockread(&ts, &data->host_tsc)) {
+			data->realtime = ts.tv_nsec + NSEC_PER_SEC * ts.tv_sec;
+			data->flags |= KVM_CLOCK_REALTIME | KVM_CLOCK_HOST_TSC;
+		} else
+#endif
+		data->host_tsc = rdtsc();
+
+		data->flags |= KVM_CLOCK_TSC_STABLE;
+		hv_clock.tsc_timestamp = ka->master_cycle_now;
+		hv_clock.system_time = ka->master_kernel_ns + ka->kvmclock_offset;
 		kvm_get_time_scale(NSEC_PER_SEC, __this_cpu_read(cpu_tsc_khz) * 1000LL,
 				   &hv_clock.tsc_shift,
 				   &hv_clock.tsc_to_system_mul);
-		ret = __pvclock_read_cycles(&hv_clock, rdtsc());
-	} else
-		ret = get_kvmclock_base_ns() + ka->kvmclock_offset;
+		data->clock = __pvclock_read_cycles(&hv_clock, data->host_tsc);
+	} else {
+		data->clock = get_kvmclock_base_ns() + ka->kvmclock_offset;
+	}
 
 	put_cpu();
+}
 
-	return ret;
+static void get_kvmclock(struct kvm *kvm, struct kvm_clock_data *data)
+{
+	struct kvm_arch *ka = &kvm->arch;
+	unsigned seq;
+
+	do {
+		seq = read_seqcount_begin(&ka->pvclock_sc);
+		__get_kvmclock(kvm, data);
+	} while (read_seqcount_retry(&ka->pvclock_sc, seq));
+}
+
+u64 get_kvmclock_ns(struct kvm *kvm)
+{
+	struct kvm_clock_data data;
+
+	get_kvmclock(kvm, &data);
+	return data.clock;
 }
 
 static void kvm_setup_pvclock_page(struct kvm_vcpu *v,
···
 static int kvm_guest_time_update(struct kvm_vcpu *v)
 {
 	unsigned long flags, tgt_tsc_khz;
+	unsigned seq;
 	struct kvm_vcpu_arch *vcpu = &v->arch;
 	struct kvm_arch *ka = &v->kvm->arch;
 	s64 kernel_ns;
···
 	 * If the host uses TSC clock, then passthrough TSC as stable
 	 * to the guest.
 	 */
-	raw_spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
-	use_master_clock = ka->use_master_clock;
-	if (use_master_clock) {
-		host_tsc = ka->master_cycle_now;
-		kernel_ns = ka->master_kernel_ns;
-	}
-	raw_spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
+	do {
+		seq = read_seqcount_begin(&ka->pvclock_sc);
+		use_master_clock = ka->use_master_clock;
+		if (use_master_clock) {
+			host_tsc = ka->master_cycle_now;
+			kernel_ns = ka->master_kernel_ns;
+		}
+	} while (read_seqcount_retry(&ka->pvclock_sc, seq));
 
 	/* Keep irq disabled to prevent changes to the clock */
 	local_irq_save(flags);
···
 	++vcpu->stat.tlb_flush;
 
 	if (!tdp_enabled) {
-               /*
+		/*
 		 * A TLB flush on behalf of the guest is equivalent to
 		 * INVPCID(all), toggling CR4.PGE, etc., which requires
-		 * a forced sync of the shadow page tables.  Unload the
-		 * entire MMU here and the subsequent load will sync the
-		 * shadow page tables, and also flush the TLB.
+		 * a forced sync of the shadow page tables.  Ensure all the
+		 * roots are synced and the guest TLB in hardware is clean.
 		 */
-		kvm_mmu_unload(vcpu);
-		return;
+		kvm_mmu_sync_roots(vcpu);
+		kvm_mmu_sync_prev_roots(vcpu);
 	}
 
 	static_call(kvm_x86_tlb_flush_guest)(vcpu);
···
 	case KVM_CAP_VM_COPY_ENC_CONTEXT_FROM:
 	case KVM_CAP_SREGS2:
 	case KVM_CAP_EXIT_ON_EMULATION_FAILURE:
+	case KVM_CAP_VCPU_ATTRIBUTES:
 		r = 1;
 		break;
 	case KVM_CAP_EXIT_HYPERCALL:
···
 		r = KVM_SYNC_X86_VALID_FIELDS;
 		break;
 	case KVM_CAP_ADJUST_CLOCK:
-		r = KVM_CLOCK_TSC_STABLE;
+		r = KVM_CLOCK_VALID_FLAGS;
 		break;
 	case KVM_CAP_X86_DISABLE_EXITS:
 		r |=  KVM_X86_DISABLE_EXITS_HLT | KVM_X86_DISABLE_EXITS_PAUSE |
···
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 	case KVM_CAP_PV_MMU:	/* obsolete */
 		r = 0;
···
 	return 0;
 }
 
+static int kvm_arch_tsc_has_attr(struct kvm_vcpu *vcpu,
+				 struct kvm_device_attr *attr)
+{
+	int r;
+
+	switch (attr->attr) {
+	case KVM_VCPU_TSC_OFFSET:
+		r = 0;
+		break;
+	default:
+		r = -ENXIO;
+	}
+
+	return r;
+}
+
+static int kvm_arch_tsc_get_attr(struct kvm_vcpu *vcpu,
+				 struct kvm_device_attr *attr)
+{
+	u64 __user *uaddr = (u64 __user *)(unsigned long)attr->addr;
+	int r;
+
+	if ((u64)(unsigned long)uaddr != attr->addr)
+		return -EFAULT;
+
+	switch (attr->attr) {
+	case KVM_VCPU_TSC_OFFSET:
+		r = -EFAULT;
+		if (put_user(vcpu->arch.l1_tsc_offset, uaddr))
+			break;
+		r = 0;
+		break;
+	default:
+		r = -ENXIO;
+	}
+
+	return r;
+}
+
+static int kvm_arch_tsc_set_attr(struct kvm_vcpu *vcpu,
+				 struct kvm_device_attr *attr)
+{
+	u64 __user *uaddr = (u64 __user *)(unsigned long)attr->addr;
+	struct kvm *kvm = vcpu->kvm;
+	int r;
+
+	if ((u64)(unsigned long)uaddr != attr->addr)
+		return -EFAULT;
+
+	switch (attr->attr) {
+	case KVM_VCPU_TSC_OFFSET: {
+		u64 offset, tsc, ns;
+		unsigned long flags;
+		bool matched;
+
+		r = -EFAULT;
+		if (get_user(offset, uaddr))
+			break;
+
+		raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags);
+
+		matched = (vcpu->arch.virtual_tsc_khz &&
+			   kvm->arch.last_tsc_khz == vcpu->arch.virtual_tsc_khz &&
+			   kvm->arch.last_tsc_offset == offset);
+
+		tsc = kvm_scale_tsc(vcpu, rdtsc(), vcpu->arch.l1_tsc_scaling_ratio) + offset;
+		ns = get_kvmclock_base_ns();
+
+		__kvm_synchronize_tsc(vcpu, offset, tsc, ns, matched);
+		raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags);
+
+		r = 0;
+		break;
+	}
+	default:
+		r = -ENXIO;
+	}
+
+	return r;
+}
+
+static int kvm_vcpu_ioctl_device_attr(struct kvm_vcpu *vcpu,
+				      unsigned int ioctl,
+				      void __user *argp)
+{
+	struct kvm_device_attr attr;
+	int r;
+
+	if (copy_from_user(&attr, argp, sizeof(attr)))
+		return -EFAULT;
+
+	if (attr.group != KVM_VCPU_TSC_CTRL)
+		return -ENXIO;
+
+	switch (ioctl) {
+	case KVM_HAS_DEVICE_ATTR:
+		r = kvm_arch_tsc_has_attr(vcpu, &attr);
+		break;
+	case KVM_GET_DEVICE_ATTR:
+		r = kvm_arch_tsc_get_attr(vcpu, &attr);
+		break;
+	case KVM_SET_DEVICE_ATTR:
+		r = kvm_arch_tsc_set_attr(vcpu, &attr);
+		break;
+	}
+
+	return r;
+}
+
 static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
 				     struct kvm_enable_cap *cap)
 {
···
 		r = __set_sregs2(vcpu, u.sregs2);
 		break;
 	}
+	case KVM_HAS_DEVICE_ATTR:
+	case KVM_GET_DEVICE_ATTR:
+	case KVM_SET_DEVICE_ATTR:
+		r = kvm_vcpu_ioctl_device_attr(vcpu, ioctl, argp);
+		break;
 	default:
 		r = -EINVAL;
 	}
···
 }
 #endif /* CONFIG_HAVE_KVM_PM_NOTIFIER */
 
+static int kvm_vm_ioctl_get_clock(struct kvm *kvm, void __user *argp)
+{
+	struct kvm_clock_data data = { 0 };
+
+	get_kvmclock(kvm, &data);
+	if (copy_to_user(argp, &data, sizeof(data)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int kvm_vm_ioctl_set_clock(struct kvm *kvm, void __user *argp)
+{
+	struct kvm_arch *ka = &kvm->arch;
+	struct kvm_clock_data data;
+	u64 now_raw_ns;
+
+	if (copy_from_user(&data, argp, sizeof(data)))
+		return -EFAULT;
+
+	/*
+	 * Only KVM_CLOCK_REALTIME is used, but allow passing the
+	 * result of KVM_GET_CLOCK back to KVM_SET_CLOCK.
+	 */
+	if (data.flags & ~KVM_CLOCK_VALID_FLAGS)
+		return -EINVAL;
+
+	kvm_hv_invalidate_tsc_page(kvm);
+	kvm_start_pvclock_update(kvm);
+	pvclock_update_vm_gtod_copy(kvm);
+
+	/*
+	 * This pairs with kvm_guest_time_update(): when masterclock is
+	 * in use, we use master_kernel_ns + kvmclock_offset to set
+	 * unsigned 'system_time' so if we use get_kvmclock_ns() (which
+	 * is slightly ahead) here we risk going negative on unsigned
+	 * 'system_time' when 'data.clock' is very small.
+	 */
+	if (data.flags & KVM_CLOCK_REALTIME) {
+		u64 now_real_ns = ktime_get_real_ns();
+
+		/*
+		 * Avoid stepping the kvmclock backwards.
+		 */
+		if (now_real_ns > data.realtime)
+			data.clock += now_real_ns - data.realtime;
+	}
+
+	if (ka->use_master_clock)
+		now_raw_ns = ka->master_kernel_ns;
+	else
+		now_raw_ns = get_kvmclock_base_ns();
+	ka->kvmclock_offset = data.clock - now_raw_ns;
+	kvm_end_pvclock_update(kvm);
+	return 0;
+}
+
 long kvm_arch_vm_ioctl(struct file *filp,
 		       unsigned int ioctl, unsigned long arg)
 {
···
 		break;
 	}
 #endif
-	case KVM_SET_CLOCK: {
-		struct kvm_arch *ka = &kvm->arch;
-		struct kvm_clock_data user_ns;
-		u64 now_ns;
-
-		r = -EFAULT;
-		if (copy_from_user(&user_ns, argp, sizeof(user_ns)))
-			goto out;
-
-		r = -EINVAL;
-		if (user_ns.flags)
-			goto out;
-
-		r = 0;
-		/*
-		 * TODO: userspace has to take care of races with VCPU_RUN, so
-		 * kvm_gen_update_masterclock() can be cut down to locked
-		 * pvclock_update_vm_gtod_copy().
-		 */
-		kvm_gen_update_masterclock(kvm);
-
-		/*
-		 * This pairs with kvm_guest_time_update(): when masterclock is
-		 * in use, we use master_kernel_ns + kvmclock_offset to set
-		 * unsigned 'system_time' so if we use get_kvmclock_ns() (which
-		 * is slightly ahead) here we risk going negative on unsigned
-		 * 'system_time' when 'user_ns.clock' is very small.
-		 */
-		raw_spin_lock_irq(&ka->pvclock_gtod_sync_lock);
-		if (kvm->arch.use_master_clock)
-			now_ns = ka->master_kernel_ns;
-		else
-			now_ns = get_kvmclock_base_ns();
-		ka->kvmclock_offset = user_ns.clock - now_ns;
-		raw_spin_unlock_irq(&ka->pvclock_gtod_sync_lock);
-
-		kvm_make_all_cpus_request(kvm, KVM_REQ_CLOCK_UPDATE);
+	case KVM_SET_CLOCK:
+		r = kvm_vm_ioctl_set_clock(kvm, argp);
 		break;
-	}
-	case KVM_GET_CLOCK: {
-		struct kvm_clock_data user_ns;
-		u64 now_ns;
-
-		now_ns = get_kvmclock_ns(kvm);
-		user_ns.clock = now_ns;
-		user_ns.flags = kvm->arch.use_master_clock ? KVM_CLOCK_TSC_STABLE : 0;
-		memset(&user_ns.pad, 0, sizeof(user_ns.pad));
-
-		r = -EFAULT;
-		if (copy_to_user(argp, &user_ns, sizeof(user_ns)))
-			goto out;
-		r = 0;
+	case KVM_GET_CLOCK:
+		r = kvm_vm_ioctl_get_clock(kvm, argp);
 		break;
-	}
 	case KVM_MEMORY_ENCRYPT_OP: {
 		r = -ENOTTY;
 		if (kvm_x86_ops.mem_enc_op)
···
 }
 EXPORT_SYMBOL_GPL(kvm_inject_realmode_interrupt);
 
-static void prepare_emulation_failure_exit(struct kvm_vcpu *vcpu)
+static void prepare_emulation_failure_exit(struct kvm_vcpu *vcpu, u64 *data,
+					   u8 ndata, u8 *insn_bytes, u8 insn_size)
 {
-	struct x86_emulate_ctxt *ctxt = vcpu->arch.emulate_ctxt;
-	u32 insn_size = ctxt->fetch.end - ctxt->fetch.data;
 	struct kvm_run *run = vcpu->run;
+	u64 info[5];
+	u8 info_start;
+
+	/*
+	 * Zero the whole array used to retrieve the exit info, as casting to
+	 * u32 for select entries will leave some chunks uninitialized.
+	 */
+	memset(&info, 0, sizeof(info));
+
+	static_call(kvm_x86_get_exit_info)(vcpu, (u32 *)&info[0], &info[1],
+					   &info[2], (u32 *)&info[3],
+					   (u32 *)&info[4]);
 
 	run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
 	run->emulation_failure.suberror = KVM_INTERNAL_ERROR_EMULATION;
-	run->emulation_failure.ndata = 0;
+
+	/*
+	 * There's currently space for 13 entries, but 5 are used for the exit
+	 * reason and info.  Restrict to 4 to reduce the maintenance burden
+	 * when expanding kvm_run.emulation_failure in the future.
+	 */
+	if (WARN_ON_ONCE(ndata > 4))
+		ndata = 4;
+
+	/* Always include the flags as a 'data' entry. */
+	info_start = 1;
 	run->emulation_failure.flags = 0;
 
 	if (insn_size) {
-		run->emulation_failure.ndata = 3;
+		BUILD_BUG_ON((sizeof(run->emulation_failure.insn_size) +
+			      sizeof(run->emulation_failure.insn_bytes) != 16));
+		info_start += 2;
 		run->emulation_failure.flags |=
 			KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES;
 		run->emulation_failure.insn_size = insn_size;
 		memset(run->emulation_failure.insn_bytes, 0x90,
 		       sizeof(run->emulation_failure.insn_bytes));
-		memcpy(run->emulation_failure.insn_bytes,
-		       ctxt->fetch.data, insn_size);
+		memcpy(run->emulation_failure.insn_bytes, insn_bytes, insn_size);
 	}
+
+	memcpy(&run->internal.data[info_start], info, sizeof(info));
+	memcpy(&run->internal.data[info_start + ARRAY_SIZE(info)], data,
+	       ndata * sizeof(data[0]));
+
+	run->emulation_failure.ndata = info_start + ARRAY_SIZE(info) + ndata;
 }
+
+static void prepare_emulation_ctxt_failure_exit(struct kvm_vcpu *vcpu)
+{
+	struct x86_emulate_ctxt *ctxt = vcpu->arch.emulate_ctxt;
+
+	prepare_emulation_failure_exit(vcpu, NULL, 0, ctxt->fetch.data,
+				       ctxt->fetch.end - ctxt->fetch.data);
+}
+
+void __kvm_prepare_emulation_failure_exit(struct kvm_vcpu *vcpu, u64 *data,
+					  u8 ndata)
+{
+	prepare_emulation_failure_exit(vcpu, data, ndata, NULL, 0);
+}
+EXPORT_SYMBOL_GPL(__kvm_prepare_emulation_failure_exit);
+
+void kvm_prepare_emulation_failure_exit(struct kvm_vcpu *vcpu)
+{
+	__kvm_prepare_emulation_failure_exit(vcpu, NULL, 0);
+}
+EXPORT_SYMBOL_GPL(kvm_prepare_emulation_failure_exit);
 
 static int handle_emulation_failure(struct kvm_vcpu *vcpu, int emulation_type)
 {
···
 
 	if (kvm->arch.exit_on_emulation_error ||
 	    (emulation_type & EMULTYPE_SKIP)) {
-		prepare_emulation_failure_exit(vcpu);
+		prepare_emulation_ctxt_failure_exit(vcpu);
 		return 0;
 	}
 
 	kvm_queue_exception(vcpu, UD_VECTOR);
 
 	if (!is_guest_mode(vcpu) && static_call(kvm_x86_get_cpl)(vcpu) == 0) {
-		vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-		vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
-		vcpu->run->internal.ndata = 0;
+		prepare_emulation_ctxt_failure_exit(vcpu);
 		return 0;
 	}
 
···
 static void kvm_hyperv_tsc_notifier(void)
 {
 	struct kvm *kvm;
-	struct kvm_vcpu *vcpu;
 	int cpu;
-	unsigned long flags;
 
 	mutex_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_make_mclock_inprogress_request(kvm);
 
+	/* no guest entries from this point */
 	hyperv_stop_tsc_emulation();
 
 	/* TSC frequency always matches when on Hyper-V */
···
 	kvm_max_guest_tsc_khz = tsc_khz;
 
 	list_for_each_entry(kvm, &vm_list, vm_list) {
-		struct kvm_arch *ka = &kvm->arch;
-
-		raw_spin_lock_irqsave(&ka->pvclock_gtod_sync_lock, flags);
+		__kvm_start_pvclock_update(kvm);
 		pvclock_update_vm_gtod_copy(kvm);
-		raw_spin_unlock_irqrestore(&ka->pvclock_gtod_sync_lock, flags);
-
-		kvm_for_each_vcpu(cpu, vcpu, kvm)
-			kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
-
-		kvm_for_each_vcpu(cpu, vcpu, kvm)
-			kvm_clear_request(KVM_REQ_MCLOCK_INPROGRESS, vcpu);
+		kvm_end_pvclock_update(kvm);
 	}
+
 	mutex_unlock(&kvm_lock);
 }
 #endif
···
 	int r;
 
 	if (kvm_x86_ops.hardware_enable) {
-		printk(KERN_ERR "kvm: already loaded the other module\n");
+		pr_err("kvm: already loaded vendor module '%s'\n", kvm_x86_ops.name);
 		r = -EEXIST;
 		goto out;
 	}
 
 	if (!ops->cpu_has_kvm_support()) {
-		pr_err_ratelimited("kvm: no hardware support\n");
+		pr_err_ratelimited("kvm: no hardware support for '%s'\n",
+				   ops->runtime_ops->name);
 		r = -EOPNOTSUPP;
 		goto out;
 	}
 	if (ops->disabled_by_bios()) {
-		pr_err_ratelimited("kvm: disabled by bios\n");
+		pr_err_ratelimited("kvm: support for '%s' disabled by bios\n",
+				   ops->runtime_ops->name);
 		r = -EOPNOTSUPP;
 		goto out;
 	}
···
 
 static void kvm_apicv_init(struct kvm *kvm)
 {
-	mutex_init(&kvm->arch.apicv_update_lock);
+	init_rwsem(&kvm->arch.apicv_update_lock);
 
 	if (enable_apicv)
 		clear_bit(APICV_INHIBIT_REASON_DISABLE,
···
 void kvm_make_scan_ioapic_request_mask(struct kvm *kvm,
 				       unsigned long *vcpu_bitmap)
 {
-	cpumask_var_t cpus;
-
-	zalloc_cpumask_var(&cpus, GFP_ATOMIC);
-
-	kvm_make_vcpus_request_mask(kvm, KVM_REQ_SCAN_IOAPIC,
-				    NULL, vcpu_bitmap, cpus);
-
-	free_cpumask_var(cpus);
+	kvm_make_vcpus_request_mask(kvm, KVM_REQ_SCAN_IOAPIC, vcpu_bitmap);
 }
 
 void kvm_make_scan_ioapic_request(struct kvm *kvm)
···
 	if (!lapic_in_kernel(vcpu))
 		return;
 
-	mutex_lock(&vcpu->kvm->arch.apicv_update_lock);
+	down_read(&vcpu->kvm->arch.apicv_update_lock);
 
 	activate = kvm_apicv_activated(vcpu->kvm);
 	if (vcpu->arch.apicv_active == activate)
···
 		kvm_make_request(KVM_REQ_EVENT, vcpu);
 
 out:
-	mutex_unlock(&vcpu->kvm->arch.apicv_update_lock);
+	up_read(&vcpu->kvm->arch.apicv_update_lock);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_update_apicv);
 
 void __kvm_request_apicv_update(struct kvm *kvm, bool activate, ulong bit)
 {
 	unsigned long old, new;
+
+	lockdep_assert_held_write(&kvm->arch.apicv_update_lock);
 
 	if (!kvm_x86_ops.check_apicv_inhibit_reasons ||
 	    !static_call(kvm_x86_check_apicv_inhibit_reasons)(bit))
···
 
 	if (!!old != !!new) {
 		trace_kvm_apicv_update_request(activate, bit);
+		/*
+		 * Kick all vCPUs before setting apicv_inhibit_reasons to avoid
+		 * false positives in the sanity check WARN in svm_vcpu_run().
+		 * This task will wait for all vCPUs to ack the kick IRQ before
+		 * updating apicv_inhibit_reasons, and all other vCPUs will
+		 * block on acquiring apicv_update_lock so that vCPUs can't
+		 * redo svm_vcpu_run() without seeing the new inhibit state.
+		 *
+		 * Note, holding apicv_update_lock and taking it in the read
+		 * side (handling the request) also prevents other vCPUs from
+		 * servicing the request with a stale apicv_inhibit_reasons.
+		 */
 		kvm_make_all_cpus_request(kvm, KVM_REQ_APICV_UPDATE);
 		kvm->arch.apicv_inhibit_reasons = new;
 		if (new) {
···
 
 void kvm_request_apicv_update(struct kvm *kvm, bool activate, ulong bit)
 {
-	mutex_lock(&kvm->arch.apicv_update_lock);
+	down_write(&kvm->arch.apicv_update_lock);
 	__kvm_request_apicv_update(kvm, activate, bit);
-	mutex_unlock(&kvm->arch.apicv_update_lock);
+	up_write(&kvm->arch.apicv_update_lock);
 }
 EXPORT_SYMBOL_GPL(kvm_request_apicv_update);
 
···
 		if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu))
 			__kvm_migrate_timers(vcpu);
 		if (kvm_check_request(KVM_REQ_MASTERCLOCK_UPDATE, vcpu))
-			kvm_gen_update_masterclock(vcpu->kvm);
+			kvm_update_masterclock(vcpu->kvm);
 		if (kvm_check_request(KVM_REQ_GLOBAL_CLOCK_UPDATE, vcpu))
 			kvm_gen_kvmclock_update(vcpu);
 		if (kvm_check_request(KVM_REQ_CLOCK_UPDATE, vcpu)) {
···
 	}
 
 	for (;;) {
+		/*
+		 * Assert that vCPU vs. VM APICv state is consistent.  An APICv
+		 * update must kick and wait for all vCPUs before toggling the
+		 * 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));
+
 		exit_fastpath = static_call(kvm_x86_run)(vcpu);
 		if (likely(exit_fastpath != EXIT_FASTPATH_REENTER_GUEST))
 			break;
···
 	return 0;
 }
 
-static void fx_init(struct kvm_vcpu *vcpu)
-{
-	/*
-	 * Ensure guest xcr0 is valid for loading
-	 */
-	vcpu->arch.xcr0 = XFEATURE_MASK_FP;
-
-	vcpu->arch.cr0 |= X86_CR0_ET;
-}
-
 int kvm_arch_vcpu_precreate(struct kvm *kvm, unsigned int id)
 {
 	if (kvm_check_tsc_unstable() && atomic_read(&kvm->online_vcpus) != 0)
···
 		pr_err("kvm: failed to allocate vcpu's fpu\n");
 		goto free_emulate_ctxt;
 	}
-
-	fx_init(vcpu);
 
 	vcpu->arch.maxphyaddr = cpuid_query_maxphyaddr(vcpu);
 	vcpu->arch.reserved_gpa_bits = kvm_vcpu_reserved_gpa_bits_raw(vcpu);
···
 
 void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 {
+	struct kvm_cpuid_entry2 *cpuid_0x1;
 	unsigned long old_cr0 = kvm_read_cr0(vcpu);
 	unsigned long new_cr0;
-	u32 eax, dummy;
+
+	/*
+	 * Several of the "set" flows, e.g. ->set_cr0(), read other registers
+	 * to handle side effects.  RESET emulation hits those flows and relies
+	 * on emulated/virtualized registers, including those that are loaded
+	 * into hardware, to be zeroed at vCPU creation.  Use CRs as a sentinel
+	 * to detect improper or missing initialization.
+	 */
+	WARN_ON_ONCE(!init_event &&
+		     (old_cr0 || kvm_read_cr3(vcpu) || kvm_read_cr4(vcpu)));
 
 	kvm_lapic_reset(vcpu, init_event);
 
···
 		vcpu->arch.xcr0 = XFEATURE_MASK_FP;
 	}
 
+	/* All GPRs except RDX (handled below) are zeroed on RESET/INIT. */
 	memset(vcpu->arch.regs, 0, sizeof(vcpu->arch.regs));
-	vcpu->arch.regs_avail = ~0;
-	vcpu->arch.regs_dirty = ~0;
+	kvm_register_mark_dirty(vcpu, VCPU_REGS_RSP);
 
 	/*
 	 * Fall back to KVM's default Family/Model/Stepping of 0x600 (P6/Athlon)
 	 * if no CPUID match is found.  Note, it's impossible to get a match at
 	 * RESET since KVM emulates RESET before exposing the vCPU to userspace,
-	 * i.e. it'simpossible for kvm_cpuid() to find a valid entry on RESET.
-	 * But, go through the motions in case that's ever remedied.
+	 * i.e. it's impossible for kvm_find_cpuid_entry() to find a valid entry
+	 * on RESET.  But, go through the motions in case that's ever remedied.
 	 */
-	eax = 1;
-	if (!kvm_cpuid(vcpu, &eax, &dummy, &dummy, &dummy, true))
-		eax = 0x600;
-	kvm_rdx_write(vcpu, eax);
+	cpuid_0x1 = kvm_find_cpuid_entry(vcpu, 1, 0);
+	kvm_rdx_write(vcpu, cpuid_0x1 ? cpuid_0x1->eax : 0x600);
 
 	vcpu->arch.ia32_xss = 0;
 
···
 void kvm_arch_free_vm(struct kvm *kvm)
 {
 	kfree(to_kvm_hv(kvm)->hv_pa_pg);
-	vfree(kvm);
+	__kvm_arch_free_vm(kvm);
 }
 
 
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
 	int ret;
+	unsigned long flags;
 
 	if (type)
 		return -EINVAL;
···
 
 	raw_spin_lock_init(&kvm->arch.tsc_write_lock);
 	mutex_init(&kvm->arch.apic_map_lock);
-	raw_spin_lock_init(&kvm->arch.pvclock_gtod_sync_lock);
-
+	seqcount_raw_spinlock_init(&kvm->arch.pvclock_sc, &kvm->arch.tsc_write_lock);
 	kvm->arch.kvmclock_offset = -get_kvmclock_base_ns();
+
+	raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags);
 	pvclock_update_vm_gtod_copy(kvm);
+	raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags);
 
 	kvm->arch.guest_can_read_msr_platform_info = true;
 
···
 	kvm_page_track_free_memslot(slot);
 }
 
-static int memslot_rmap_alloc(struct kvm_memory_slot *slot,
-			      unsigned long npages)
+int memslot_rmap_alloc(struct kvm_memory_slot *slot, unsigned long npages)
 {
 	const int sz = sizeof(*slot->arch.rmap[0]);
 	int i;
···
 		}
 	}
 
-	return 0;
-}
-
-int alloc_all_memslots_rmaps(struct kvm *kvm)
-{
-	struct kvm_memslots *slots;
-	struct kvm_memory_slot *slot;
-	int r, i;
-
-	/*
-	 * Check if memslots alreday have rmaps early before acquiring
-	 * the slots_arch_lock below.
-	 */
-	if (kvm_memslots_have_rmaps(kvm))
-		return 0;
-
-	mutex_lock(&kvm->slots_arch_lock);
-
-	/*
-	 * Read memslots_have_rmaps again, under the slots arch lock,
-	 * before allocating the rmaps
-	 */
-	if (kvm_memslots_have_rmaps(kvm)) {
-		mutex_unlock(&kvm->slots_arch_lock);
-		return 0;
-	}
-
-	for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) {
-		slots = __kvm_memslots(kvm, i);
-		kvm_for_each_memslot(slot, slots) {
-			r = memslot_rmap_alloc(slot, slot->npages);
-			if (r) {
-				mutex_unlock(&kvm->slots_arch_lock);
-				return r;
-			}
-		}
-	}
-
-	/*
-	 * Ensure that memslots_have_rmaps becomes true strictly after
-	 * all the rmap pointers are set.
-	 */
-	smp_store_release(&kvm->arch.memslots_have_rmaps, true);
-	mutex_unlock(&kvm->slots_arch_lock);
 	return 0;
 }
 
···
 		}
 	}
 
-	if (kvm_page_track_create_memslot(slot, npages))
+	if (kvm_page_track_create_memslot(kvm, slot, npages))
 		goto out_free;
 
 	return 0;
···
 	return static_call(kvm_x86_update_pi_irte)(kvm, host_irq, guest_irq, set);
 }
 
+bool kvm_arch_irqfd_route_changed(struct kvm_kernel_irq_routing_entry *old,
+				  struct kvm_kernel_irq_routing_entry *new)
+{
+	if (new->type != KVM_IRQ_ROUTING_MSI)
+		return true;
+
+	return !!memcmp(&old->msi, &new->msi, sizeof(new->msi));
+}
+
 bool kvm_vector_hashing_enabled(void)
 {
 	return vector_hashing;
···
 	 * doesn't seem to be a real use-case behind such requests, just return
 	 * KVM_EXIT_INTERNAL_ERROR for now.
 	 */
-	vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
-	vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
-	vcpu->run->internal.ndata = 0;
+	kvm_prepare_emulation_failure_exit(vcpu);
 
 	return 0;
 }

arch/x86/kvm/x86.h

···
 
 extern int pi_inject_timer;
 
-extern struct static_key kvm_no_apic_vcpu;
-
 extern bool report_ignored_msrs;
 
 static inline u64 nsec_to_cycles(struct kvm_vcpu *vcpu, u64 nsec)

drivers/clocksource/timer-riscv.c

···
 #include <linux/delay.h>
 #include <linux/irq.h>
 #include <linux/irqdomain.h>
+#include <linux/module.h>
 #include <linux/sched_clock.h>
 #include <linux/io-64-nonatomic-lo-hi.h>
 #include <linux/interrupt.h>
 #include <linux/of_irq.h>
+#include <clocksource/timer-riscv.h>
 #include <asm/smp.h>
 #include <asm/sbi.h>
 #include <asm/timex.h>
···
 	disable_percpu_irq(riscv_clock_event_irq);
 	return 0;
 }
+
+void riscv_cs_get_mult_shift(u32 *mult, u32 *shift)
+{
+	*mult = riscv_clocksource.mult;
+	*shift = riscv_clocksource.shift;
+}
+EXPORT_SYMBOL_GPL(riscv_cs_get_mult_shift);
 
 /* called directly from the low-level interrupt handler */
 static irqreturn_t riscv_timer_interrupt(int irq, void *dev_id)

drivers/gpu/drm/i915/Kconfig

···
 	depends on DRM_I915_GVT
 	depends on KVM
 	depends on VFIO_MDEV
+	select KVM_EXTERNAL_WRITE_TRACKING
 	default n
 	help
 	  Choose this option if you want to enable KVMGT support for

include/clocksource/timer-riscv.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ *	Atish Patra <[e-mail]>
+ */
+
+#ifndef __TIMER_RISCV_H
+#define __TIMER_RISCV_H
+
+#include <linux/types.h>
+
+extern void riscv_cs_get_mult_shift(u32 *mult, u32 *shift);
+
+#endif

include/linux/kvm_host.h

···
 #include <asm/kvm_host.h>
 #include <linux/kvm_dirty_ring.h>
 
-#ifndef KVM_MAX_VCPU_ID
-#define KVM_MAX_VCPU_ID KVM_MAX_VCPUS
+#ifndef KVM_MAX_VCPU_IDS
+#define KVM_MAX_VCPU_IDS KVM_MAX_VCPUS
 #endif
 
 /*
···
 #define KVM_ARCH_REQ(nr)           KVM_ARCH_REQ_FLAGS(nr, 0)
 
 bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
-				 struct kvm_vcpu *except,
-				 unsigned long *vcpu_bitmap, cpumask_var_t tmp);
+				 unsigned long *vcpu_bitmap);
 bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req);
 bool kvm_make_all_cpus_request_except(struct kvm *kvm, unsigned int req,
 				      struct kvm_vcpu *except);
···
 {
 	return kzalloc(sizeof(struct kvm), GFP_KERNEL);
 }
+#endif
 
+static inline void __kvm_arch_free_vm(struct kvm *kvm)
+{
+	kvfree(kvm);
+}
+
+#ifndef __KVM_HAVE_ARCH_VM_FREE
 static inline void kvm_arch_free_vm(struct kvm *kvm)
 {
-	kfree(kvm);
+	__kvm_arch_free_vm(kvm);
 }
 #endif
 
···
 void kvm_arch_irq_bypass_start(struct irq_bypass_consumer *);
 int kvm_arch_update_irqfd_routing(struct kvm *kvm, unsigned int host_irq,
 				  uint32_t guest_irq, bool set);
+bool kvm_arch_irqfd_route_changed(struct kvm_kernel_irq_routing_entry *,
+				  struct kvm_kernel_irq_routing_entry *);
 #endif /* CONFIG_HAVE_KVM_IRQ_BYPASS */
 
 #ifdef CONFIG_HAVE_KVM_INVALID_WAKEUPS

include/uapi/linux/kvm.h

···
 #define KVM_EXIT_AP_RESET_HOLD    32
 #define KVM_EXIT_X86_BUS_LOCK     33
 #define KVM_EXIT_XEN              34
+#define KVM_EXIT_RISCV_SBI        35
 
 /* For KVM_EXIT_INTERNAL_ERROR */
 /* Emulate instruction failed. */
···
 		 * "ndata" is correct, that new fields are enumerated in "flags",
 		 * and that each flag enumerates fields that are 64-bit aligned
 		 * and sized (so that ndata+internal.data[] is valid/accurate).
+		 *
+		 * Space beyond the defined fields may be used to store arbitrary
+		 * debug information relating to the emulation failure. It is
+		 * accounted for in "ndata" but the format is unspecified and is
+		 * not represented in "flags". Any such information is *not* ABI!
 		 */
 		struct {
 			__u32 suberror;
 			__u32 ndata;
 			__u64 flags;
-			__u8  insn_size;
-			__u8  insn_bytes[15];
+			union {
+				struct {
+					__u8  insn_size;
+					__u8  insn_bytes[15];
+				};
+			};
+			/* Arbitrary debug data may follow. */
 		} emulation_failure;
 		/* KVM_EXIT_OSI */
 		struct {
···
 		} msr;
 		/* KVM_EXIT_XEN */
 		struct kvm_xen_exit xen;
+		/* KVM_EXIT_RISCV_SBI */
+		struct {
+			unsigned long extension_id;
+			unsigned long function_id;
+			unsigned long args[6];
+			unsigned long ret[2];
+		} riscv_sbi;
 		/* Fix the size of the union. */
 		char padding[256];
 	};
···
 
 /* Do not use 1, KVM_CHECK_EXTENSION returned it before we had flags.  */
 #define KVM_CLOCK_TSC_STABLE		2
+#define KVM_CLOCK_REALTIME		(1 << 2)
+#define KVM_CLOCK_HOST_TSC		(1 << 3)
 
 struct kvm_clock_data {
 	__u64 clock;
 	__u32 flags;
-	__u32 pad[9];
+	__u32 pad0;
+	__u64 realtime;
+	__u64 host_tsc;
+	__u32 pad[4];
 };
 
 /* For KVM_CAP_SW_TLB */

tools/arch/arm64/include/asm/sysreg.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Macros for accessing system registers with older binutils.
+ *
+ * Copyright (C) 2014 ARM Ltd.
+ * Author: Catalin Marinas <[e-mail]>
+ */
+
+#ifndef __ASM_SYSREG_H
+#define __ASM_SYSREG_H
+
+#include <linux/bits.h>
+#include <linux/stringify.h>
+
+/*
+ * ARMv8 ARM reserves the following encoding for system registers:
+ * (Ref: ARMv8 ARM, Section: "System instruction class encoding overview",
+ *  C5.2, version:ARM DDI 0487A.f)
+ *	[20-19] : Op0
+ *	[18-16] : Op1
+ *	[15-12] : CRn
+ *	[11-8]  : CRm
+ *	[7-5]   : Op2
+ */
+#define Op0_shift	19
+#define Op0_mask	0x3
+#define Op1_shift	16
+#define Op1_mask	0x7
+#define CRn_shift	12
+#define CRn_mask	0xf
+#define CRm_shift	8
+#define CRm_mask	0xf
+#define Op2_shift	5
+#define Op2_mask	0x7
+
+#define sys_reg(op0, op1, crn, crm, op2) \
+	(((op0) << Op0_shift) | ((op1) << Op1_shift) | \
+	 ((crn) << CRn_shift) | ((crm) << CRm_shift) | \
+	 ((op2) << Op2_shift))
+
+#define sys_insn	sys_reg
+
+#define sys_reg_Op0(id)	(((id) >> Op0_shift) & Op0_mask)
+#define sys_reg_Op1(id)	(((id) >> Op1_shift) & Op1_mask)
+#define sys_reg_CRn(id)	(((id) >> CRn_shift) & CRn_mask)
+#define sys_reg_CRm(id)	(((id) >> CRm_shift) & CRm_mask)
+#define sys_reg_Op2(id)	(((id) >> Op2_shift) & Op2_mask)
+
+#ifndef CONFIG_BROKEN_GAS_INST
+
+#ifdef __ASSEMBLY__
+// The space separator is omitted so that __emit_inst(x) can be parsed as
+// either an assembler directive or an assembler macro argument.
+#define __emit_inst(x)			.inst(x)
+#else
+#define __emit_inst(x)			".inst " __stringify((x)) "\n\t"
+#endif
+
+#else  /* CONFIG_BROKEN_GAS_INST */
+
+#ifndef CONFIG_CPU_BIG_ENDIAN
+#define __INSTR_BSWAP(x)		(x)
+#else  /* CONFIG_CPU_BIG_ENDIAN */
+#define __INSTR_BSWAP(x)		((((x) << 24) & 0xff000000)	| \
+					 (((x) <<  8) & 0x00ff0000)	| \
+					 (((x) >>  8) & 0x0000ff00)	| \
+					 (((x) >> 24) & 0x000000ff))
+#endif	/* CONFIG_CPU_BIG_ENDIAN */
+
+#ifdef __ASSEMBLY__
+#define __emit_inst(x)			.long __INSTR_BSWAP(x)
+#else  /* __ASSEMBLY__ */
+#define __emit_inst(x)			".long " __stringify(__INSTR_BSWAP(x)) "\n\t"
+#endif	/* __ASSEMBLY__ */
+
+#endif	/* CONFIG_BROKEN_GAS_INST */
+
+/*
+ * Instructions for modifying PSTATE fields.
+ * As per Arm ARM for v8-A, Section "C.5.1.3 op0 == 0b00, architectural hints,
+ * barriers and CLREX, and PSTATE access", ARM DDI 0487 C.a, system instructions
+ * for accessing PSTATE fields have the following encoding:
+ *	Op0 = 0, CRn = 4
+ *	Op1, Op2 encodes the PSTATE field modified and defines the constraints.
+ *	CRm = Imm4 for the instruction.
+ *	Rt = 0x1f
+ */
+#define pstate_field(op1, op2)		((op1) << Op1_shift | (op2) << Op2_shift)
+#define PSTATE_Imm_shift		CRm_shift
+
+#define PSTATE_PAN			pstate_field(0, 4)
+#define PSTATE_UAO			pstate_field(0, 3)
+#define PSTATE_SSBS			pstate_field(3, 1)
+#define PSTATE_TCO			pstate_field(3, 4)
+
+#define SET_PSTATE_PAN(x)		__emit_inst(0xd500401f | PSTATE_PAN | ((!!x) << PSTATE_Imm_shift))
+#define SET_PSTATE_UAO(x)		__emit_inst(0xd500401f | PSTATE_UAO | ((!!x) << PSTATE_Imm_shift))
+#define SET_PSTATE_SSBS(x)		__emit_inst(0xd500401f | PSTATE_SSBS | ((!!x) << PSTATE_Imm_shift))
+#define SET_PSTATE_TCO(x)		__emit_inst(0xd500401f | PSTATE_TCO | ((!!x) << PSTATE_Imm_shift))
+
+#define set_pstate_pan(x)		asm volatile(SET_PSTATE_PAN(x))
+#define set_pstate_uao(x)		asm volatile(SET_PSTATE_UAO(x))
+#define set_pstate_ssbs(x)		asm volatile(SET_PSTATE_SSBS(x))
+
+#define __SYS_BARRIER_INSN(CRm, op2, Rt) \
+	__emit_inst(0xd5000000 | sys_insn(0, 3, 3, (CRm), (op2)) | ((Rt) & 0x1f))
+
+#define SB_BARRIER_INSN			__SYS_BARRIER_INSN(0, 7, 31)
+
+#define SYS_DC_ISW			sys_insn(1, 0, 7, 6, 2)
+#define SYS_DC_CSW			sys_insn(1, 0, 7, 10, 2)
+#define SYS_DC_CISW			sys_insn(1, 0, 7, 14, 2)
+
+/*
+ * System registers, organised loosely by encoding but grouped together
+ * where the architected name contains an index. e.g. ID_MMFR<n>_EL1.
+ */
+#define SYS_OSDTRRX_EL1			sys_reg(2, 0, 0, 0, 2)
+#define SYS_MDCCINT_EL1			sys_reg(2, 0, 0, 2, 0)
+#define SYS_MDSCR_EL1			sys_reg(2, 0, 0, 2, 2)
+#define SYS_OSDTRTX_EL1			sys_reg(2, 0, 0, 3, 2)
+#define SYS_OSECCR_EL1			sys_reg(2, 0, 0, 6, 2)
+#define SYS_DBGBVRn_EL1(n)		sys_reg(2, 0, 0, n, 4)
+#define SYS_DBGBCRn_EL1(n)		sys_reg(2, 0, 0, n, 5)
+#define SYS_DBGWVRn_EL1(n)		sys_reg(2, 0, 0, n, 6)
+#define SYS_DBGWCRn_EL1(n)		sys_reg(2, 0, 0, n, 7)
+#define SYS_MDRAR_EL1			sys_reg(2, 0, 1, 0, 0)
+#define SYS_OSLAR_EL1			sys_reg(2, 0, 1, 0, 4)
+#define SYS_OSLSR_EL1			sys_reg(2, 0, 1, 1, 4)
+#define SYS_OSDLR_EL1			sys_reg(2, 0, 1, 3, 4)
+#define SYS_DBGPRCR_EL1			sys_reg(2, 0, 1, 4, 4)
+#define SYS_DBGCLAIMSET_EL1		sys_reg(2, 0, 7, 8, 6)
+#define SYS_DBGCLAIMCLR_EL1		sys_reg(2, 0, 7, 9, 6)
+#define SYS_DBGAUTHSTATUS_EL1		sys_reg(2, 0, 7, 14, 6)
+#define SYS_MDCCSR_EL0			sys_reg(2, 3, 0, 1, 0)
+#define SYS_DBGDTR_EL0			sys_reg(2, 3, 0, 4, 0)
+#define SYS_DBGDTRRX_EL0		sys_reg(2, 3, 0, 5, 0)
+#define SYS_DBGDTRTX_EL0		sys_reg(2, 3, 0, 5, 0)
+#define SYS_DBGVCR32_EL2		sys_reg(2, 4, 0, 7, 0)
+
+#define SYS_MIDR_EL1			sys_reg(3, 0, 0, 0, 0)
+#define SYS_MPIDR_EL1			sys_reg(3, 0, 0, 0, 5)
+#define SYS_REVIDR_EL1			sys_reg(3, 0, 0, 0, 6)
+
+#define SYS_ID_PFR0_EL1			sys_reg(3, 0, 0, 1, 0)
+#define SYS_ID_PFR1_EL1			sys_reg(3, 0, 0, 1, 1)
+#define SYS_ID_PFR2_EL1			sys_reg(3, 0, 0, 3, 4)
+#define SYS_ID_DFR0_EL1			sys_reg(3, 0, 0, 1, 2)
+#define SYS_ID_DFR1_EL1			sys_reg(3, 0, 0, 3, 5)
+#define SYS_ID_AFR0_EL1			sys_reg(3, 0, 0, 1, 3)
+#define SYS_ID_MMFR0_EL1		sys_reg(3, 0, 0, 1, 4)
+#define SYS_ID_MMFR1_EL1		sys_reg(3, 0, 0, 1, 5)
+#define SYS_ID_MMFR2_EL1		sys_reg(3, 0, 0, 1, 6)
+#define SYS_ID_MMFR3_EL1		sys_reg(3, 0, 0, 1, 7)
+#define SYS_ID_MMFR4_EL1		sys_reg(3, 0, 0, 2, 6)
+#define SYS_ID_MMFR5_EL1		sys_reg(3, 0, 0, 3, 6)
+
+#define SYS_ID_ISAR0_EL1		sys_reg(3, 0, 0, 2, 0)
+#define SYS_ID_ISAR1_EL1		sys_reg(3, 0, 0, 2, 1)
+#define SYS_ID_ISAR2_EL1		sys_reg(3, 0, 0, 2, 2)
+#define SYS_ID_ISAR3_EL1		sys_reg(3, 0, 0, 2, 3)
+#define SYS_ID_ISAR4_EL1		sys_reg(3, 0, 0, 2, 4)
+#define SYS_ID_ISAR5_EL1		sys_reg(3, 0, 0, 2, 5)
+#define SYS_ID_ISAR6_EL1		sys_reg(3, 0, 0, 2, 7)
+
+#define SYS_MVFR0_EL1			sys_reg(3, 0, 0, 3, 0)
+#define SYS_MVFR1_EL1			sys_reg(3, 0, 0, 3, 1)
+#define SYS_MVFR2_EL1			sys_reg(3, 0, 0, 3, 2)
+
+#define SYS_ID_AA64PFR0_EL1		sys_reg(3, 0, 0, 4, 0)
+#define SYS_ID_AA64PFR1_EL1		sys_reg(3, 0, 0, 4, 1)
+#define SYS_ID_AA64ZFR0_EL1		sys_reg(3, 0, 0, 4, 4)
+
+#define SYS_ID_AA64DFR0_EL1		sys_reg(3, 0, 0, 5, 0)
+#define SYS_ID_AA64DFR1_EL1		sys_reg(3, 0, 0, 5, 1)
+
+#define SYS_ID_AA64AFR0_EL1		sys_reg(3, 0, 0, 5, 4)
+#define SYS_ID_AA64AFR1_EL1		sys_reg(3, 0, 0, 5, 5)
+
+#define SYS_ID_AA64ISAR0_EL1		sys_reg(3, 0, 0, 6, 0)
+#define SYS_ID_AA64ISAR1_EL1		sys_reg(3, 0, 0, 6, 1)
+
+#define SYS_ID_AA64MMFR0_EL1		sys_reg(3, 0, 0, 7, 0)
+#define SYS_ID_AA64MMFR1_EL1		sys_reg(3, 0, 0, 7, 1)
+#define SYS_ID_AA64MMFR2_EL1		sys_reg(3, 0, 0, 7, 2)
+
+#define SYS_SCTLR_EL1			sys_reg(3, 0, 1, 0, 0)
+#define SYS_ACTLR_EL1			sys_reg(3, 0, 1, 0, 1)
+#define SYS_CPACR_EL1			sys_reg(3, 0, 1, 0, 2)
+#define SYS_RGSR_EL1			sys_reg(3, 0, 1, 0, 5)
+#define SYS_GCR_EL1			sys_reg(3, 0, 1, 0, 6)
+
+#define SYS_ZCR_EL1			sys_reg(3, 0, 1, 2, 0)
+#define SYS_TRFCR_EL1			sys_reg(3, 0, 1, 2, 1)
+
+#define SYS_TTBR0_EL1			sys_reg(3, 0, 2, 0, 0)
+#define SYS_TTBR1_EL1			sys_reg(3, 0, 2, 0, 1)
+#define SYS_TCR_EL1			sys_reg(3, 0, 2, 0, 2)
+
+#define SYS_APIAKEYLO_EL1		sys_reg(3, 0, 2, 1, 0)
+#define SYS_APIAKEYHI_EL1		sys_reg(3, 0, 2, 1, 1)
+#define SYS_APIBKEYLO_EL1		sys_reg(3, 0, 2, 1, 2)
+#define SYS_APIBKEYHI_EL1		sys_reg(3, 0, 2, 1, 3)
+
+#define SYS_APDAKEYLO_EL1		sys_reg(3, 0, 2, 2, 0)
+#define SYS_APDAKEYHI_EL1		sys_reg(3, 0, 2, 2, 1)
+#define SYS_APDBKEYLO_EL1		sys_reg(3, 0, 2, 2, 2)
+#define SYS_APDBKEYHI_EL1		sys_reg(3, 0, 2, 2, 3)
+
+#define SYS_APGAKEYLO_EL1		sys_reg(3, 0, 2, 3, 0)
+#define SYS_APGAKEYHI_EL1		sys_reg(3, 0, 2, 3, 1)
+
+#define SYS_SPSR_EL1			sys_reg(3, 0, 4, 0, 0)
+#define SYS_ELR_EL1			sys_reg(3, 0, 4, 0, 1)
+
+#define SYS_ICC_PMR_EL1			sys_reg(3, 0, 4, 6, 0)
+
+#define SYS_AFSR0_EL1			sys_reg(3, 0, 5, 1, 0)
+#define SYS_AFSR1_EL1			sys_reg(3, 0, 5, 1, 1)
+#define SYS_ESR_EL1			sys_reg(3, 0, 5, 2, 0)
+
+#define SYS_ERRIDR_EL1			sys_reg(3, 0, 5, 3, 0)
+#define SYS_ERRSELR_EL1			sys_reg(3, 0, 5, 3, 1)
+#define SYS_ERXFR_EL1			sys_reg(3, 0, 5, 4, 0)
+#define SYS_ERXCTLR_EL1			sys_reg(3, 0, 5, 4, 1)
+#define SYS_ERXSTATUS_EL1		sys_reg(3, 0, 5, 4, 2)
+#define SYS_ERXADDR_EL1			sys_reg(3, 0, 5, 4, 3)
+#define SYS_ERXMISC0_EL1		sys_reg(3, 0, 5, 5, 0)
+#define SYS_ERXMISC1_EL1		sys_reg(3, 0, 5, 5, 1)
+#define SYS_TFSR_EL1			sys_reg(3, 0, 5, 6, 0)
+#define SYS_TFSRE0_EL1			sys_reg(3, 0, 5, 6, 1)
+
+#define SYS_FAR_EL1			sys_reg(3, 0, 6, 0, 0)
+#define SYS_PAR_EL1			sys_reg(3, 0, 7, 4, 0)
+
+#define SYS_PAR_EL1_F			BIT(0)
+#define SYS_PAR_EL1_FST			GENMASK(6, 1)
+
+/*** Statistical Profiling Extension ***/
+/* ID registers */
+#define SYS_PMSIDR_EL1			sys_reg(3, 0, 9, 9, 7)
+#define SYS_PMSIDR_EL1_FE_SHIFT		0
+#define SYS_PMSIDR_EL1_FT_SHIFT		1
+#define SYS_PMSIDR_EL1_FL_SHIFT		2
+#define SYS_PMSIDR_EL1_ARCHINST_SHIFT	3
+#define SYS_PMSIDR_EL1_LDS_SHIFT	4
+#define SYS_PMSIDR_EL1_ERND_SHIFT	5
+#define SYS_PMSIDR_EL1_INTERVAL_SHIFT	8
+#define SYS_PMSIDR_EL1_INTERVAL_MASK	0xfUL
+#define SYS_PMSIDR_EL1_MAXSIZE_SHIFT	12
+#define SYS_PMSIDR_EL1_MAXSIZE_MASK	0xfUL
+#define SYS_PMSIDR_EL1_COUNTSIZE_SHIFT	16
+#define SYS_PMSIDR_EL1_COUNTSIZE_MASK	0xfUL
+
+#define SYS_PMBIDR_EL1			sys_reg(3, 0, 9, 10, 7)
+#define SYS_PMBIDR_EL1_ALIGN_SHIFT	0
+#define SYS_PMBIDR_EL1_ALIGN_MASK	0xfU
+#define SYS_PMBIDR_EL1_P_SHIFT		4
+#define SYS_PMBIDR_EL1_F_SHIFT		5
+
+/* Sampling controls */
+#define SYS_PMSCR_EL1			sys_reg(3, 0, 9, 9, 0)
+#define SYS_PMSCR_EL1_E0SPE_SHIFT	0
+#define SYS_PMSCR_EL1_E1SPE_SHIFT	1
+#define SYS_PMSCR_EL1_CX_SHIFT		3
+#define SYS_PMSCR_EL1_PA_SHIFT		4
+#define SYS_PMSCR_EL1_TS_SHIFT		5
+#define SYS_PMSCR_EL1_PCT_SHIFT		6
+
+#define SYS_PMSCR_EL2			sys_reg(3, 4, 9, 9, 0)
+#define SYS_PMSCR_EL2_E0HSPE_SHIFT	0
+#define SYS_PMSCR_EL2_E2SPE_SHIFT	1
+#define SYS_PMSCR_EL2_CX_SHIFT		3
+#define SYS_PMSCR_EL2_PA_SHIFT		4
+#define SYS_PMSCR_EL2_TS_SHIFT		5
+#define SYS_PMSCR_EL2_PCT_SHIFT		6
+
+#define SYS_PMSICR_EL1			sys_reg(3, 0, 9, 9, 2)
+
+#define SYS_PMSIRR_EL1			sys_reg(3, 0, 9, 9, 3)
+#define SYS_PMSIRR_EL1_RND_SHIFT	0
+#define SYS_PMSIRR_EL1_INTERVAL_SHIFT	8
+#define SYS_PMSIRR_EL1_INTERVAL_MASK	0xffffffUL
+
+/* Filtering controls */
+#define SYS_PMSNEVFR_EL1		sys_reg(3, 0, 9, 9, 1)
+
+#define SYS_PMSFCR_EL1			sys_reg(3, 0, 9, 9, 4)
+#define SYS_PMSFCR_EL1_FE_SHIFT		0
+#define SYS_PMSFCR_EL1_FT_SHIFT		1
+#define SYS_PMSFCR_EL1_FL_SHIFT		2
+#define SYS_PMSFCR_EL1_B_SHIFT		16
+#define SYS_PMSFCR_EL1_LD_SHIFT		17
+#define SYS_PMSFCR_EL1_ST_SHIFT		18
+
+#define SYS_PMSEVFR_EL1			sys_reg(3, 0, 9, 9, 5)
+#define SYS_PMSEVFR_EL1_RES0_8_2	\
+	(GENMASK_ULL(47, 32) | GENMASK_ULL(23, 16) | GENMASK_ULL(11, 8) |\
+	 BIT_ULL(6) | BIT_ULL(4) | BIT_ULL(2) | BIT_ULL(0))
+#define SYS_PMSEVFR_EL1_RES0_8_3	\
+	(SYS_PMSEVFR_EL1_RES0_8_2 & ~(BIT_ULL(18) | BIT_ULL(17) | BIT_ULL(11)))
+
+#define SYS_PMSLATFR_EL1		sys_reg(3, 0, 9, 9, 6)
+#define SYS_PMSLATFR_EL1_MINLAT_SHIFT	0
+
+/* Buffer controls */
+#define SYS_PMBLIMITR_EL1		sys_reg(3, 0, 9, 10, 0)
+#define SYS_PMBLIMITR_EL1_E_SHIFT	0
+#define SYS_PMBLIMITR_EL1_FM_SHIFT	1
+#define SYS_PMBLIMITR_EL1_FM_MASK	0x3UL
+#define SYS_PMBLIMITR_EL1_FM_STOP_IRQ	(0 << SYS_PMBLIMITR_EL1_FM_SHIFT)
+
+#define SYS_PMBPTR_EL1			sys_reg(3, 0, 9, 10, 1)
+
+/* Buffer error reporting */
+#define SYS_PMBSR_EL1			sys_reg(3, 0, 9, 10, 3)
+#define SYS_PMBSR_EL1_COLL_SHIFT	16
+#define SYS_PMBSR_EL1_S_SHIFT		17
+#define SYS_PMBSR_EL1_EA_SHIFT		18
+#define SYS_PMBSR_EL1_DL_SHIFT		19
+#define SYS_PMBSR_EL1_EC_SHIFT		26
+#define SYS_PMBSR_EL1_EC_MASK		0x3fUL
+
+#define SYS_PMBSR_EL1_EC_BUF		(0x0UL << SYS_PMBSR_EL1_EC_SHIFT)
+#define SYS_PMBSR_EL1_EC_FAULT_S1	(0x24UL << SYS_PMBSR_EL1_EC_SHIFT)
+#define SYS_PMBSR_EL1_EC_FAULT_S2	(0x25UL << SYS_PMBSR_EL1_EC_SHIFT)
+
+#define SYS_PMBSR_EL1_FAULT_FSC_SHIFT	0
+#define SYS_PMBSR_EL1_FAULT_FSC_MASK	0x3fUL
+
+#define SYS_PMBSR_EL1_BUF_BSC_SHIFT	0
+#define SYS_PMBSR_EL1_BUF_BSC_MASK	0x3fUL
+
+#define SYS_PMBSR_EL1_BUF_BSC_FULL	(0x1UL << SYS_PMBSR_EL1_BUF_BSC_SHIFT)
+
+/*** End of Statistical Profiling Extension ***/
+
+/*
+ * TRBE Registers
+ */
+#define SYS_TRBLIMITR_EL1		sys_reg(3, 0, 9, 11, 0)
+#define SYS_TRBPTR_EL1			sys_reg(3, 0, 9, 11, 1)
+#define SYS_TRBBASER_EL1		sys_reg(3, 0, 9, 11, 2)
+#define SYS_TRBSR_EL1			sys_reg(3, 0, 9, 11, 3)
+#define SYS_TRBMAR_EL1			sys_reg(3, 0, 9, 11, 4)
+#define SYS_TRBTRG_EL1			sys_reg(3, 0, 9, 11, 6)
+#define SYS_TRBIDR_EL1			sys_reg(3, 0, 9, 11, 7)
+
+#define TRBLIMITR_LIMIT_MASK		GENMASK_ULL(51, 0)
+#define TRBLIMITR_LIMIT_SHIFT		12
+#define TRBLIMITR_NVM			BIT(5)
+#define TRBLIMITR_TRIG_MODE_MASK	GENMASK(1, 0)
+#define TRBLIMITR_TRIG_MODE_SHIFT	3
+#define TRBLIMITR_FILL_MODE_MASK	GENMASK(1, 0)
+#define TRBLIMITR_FILL_MODE_SHIFT	1
+#define TRBLIMITR_ENABLE		BIT(0)
+#define TRBPTR_PTR_MASK			GENMASK_ULL(63, 0)
+#define TRBPTR_PTR_SHIFT		0
+#define TRBBASER_BASE_MASK		GENMASK_ULL(51, 0)
+#define TRBBASER_BASE_SHIFT		12
+#define TRBSR_EC_MASK			GENMASK(5, 0)
+#define TRBSR_EC_SHIFT			26
+#define TRBSR_IRQ			BIT(22)
+#define TRBSR_TRG			BIT(21)
+#define TRBSR_WRAP			BIT(20)
+#define TRBSR_ABORT			BIT(18)
+#define TRBSR_STOP			BIT(17)
+#define TRBSR_MSS_MASK			GENMASK(15, 0)
+#define TRBSR_MSS_SHIFT			0
+#define TRBSR_BSC_MASK			GENMASK(5, 0)
+#define TRBSR_BSC_SHIFT			0
+#define TRBSR_FSC_MASK			GENMASK(5, 0)
+#define TRBSR_FSC_SHIFT			0
+#define TRBMAR_SHARE_MASK		GENMASK(1, 0)
+#define TRBMAR_SHARE_SHIFT		8
+#define TRBMAR_OUTER_MASK		GENMASK(3, 0)
+#define TRBMAR_OUTER_SHIFT		4
+#define TRBMAR_INNER_MASK		GENMASK(3, 0)
+#define TRBMAR_INNER_SHIFT		0
+#define TRBTRG_TRG_MASK			GENMASK(31, 0)
+#define TRBTRG_TRG_SHIFT		0
+#define TRBIDR_FLAG			BIT(5)
+#define TRBIDR_PROG			BIT(4)
+#define TRBIDR_ALIGN_MASK		GENMASK(3, 0)
+#define TRBIDR_ALIGN_SHIFT		0
+
+#define SYS_PMINTENSET_EL1		sys_reg(3, 0, 9, 14, 1)
+#define SYS_PMINTENCLR_EL1		sys_reg(3, 0, 9, 14, 2)
+
+#define SYS_PMMIR_EL1			sys_reg(3, 0, 9, 14, 6)
+
+#define SYS_MAIR_EL1			sys_reg(3, 0, 10, 2, 0)
+#define SYS_AMAIR_EL1			sys_reg(3, 0, 10, 3, 0)
+
+#define SYS_LORSA_EL1			sys_reg(3, 0, 10, 4, 0)
+#define SYS_LOREA_EL1			sys_reg(3, 0, 10, 4, 1)
+#define SYS_LORN_EL1			sys_reg(3, 0, 10, 4, 2)
+#define SYS_LORC_EL1			sys_reg(3, 0, 10, 4, 3)
+#define SYS_LORID_EL1			sys_reg(3, 0, 10, 4, 7)
+
+#define SYS_VBAR_EL1			sys_reg(3, 0, 12, 0, 0)
+#define SYS_DISR_EL1			sys_reg(3, 0, 12, 1, 1)
+
+#define SYS_ICC_IAR0_EL1		sys_reg(3, 0, 12, 8, 0)
+#define SYS_ICC_EOIR0_EL1		sys_reg(3, 0, 12, 8, 1)
+#define SYS_ICC_HPPIR0_EL1		sys_reg(3, 0, 12, 8, 2)
+#define SYS_ICC_BPR0_EL1		sys_reg(3, 0, 12, 8, 3)
+#define SYS_ICC_AP0Rn_EL1(n)		sys_reg(3, 0, 12, 8, 4 | n)
+#define SYS_ICC_AP0R0_EL1		SYS_ICC_AP0Rn_EL1(0)
+#define SYS_ICC_AP0R1_EL1		SYS_ICC_AP0Rn_EL1(1)
+#define SYS_ICC_AP0R2_EL1		SYS_ICC_AP0Rn_EL1(2)
+#define SYS_ICC_AP0R3_EL1		SYS_ICC_AP0Rn_EL1(3)
+#define SYS_ICC_AP1Rn_EL1(n)		sys_reg(3, 0, 12, 9, n)
+#define SYS_ICC_AP1R0_EL1		SYS_ICC_AP1Rn_EL1(0)
+#define SYS_ICC_AP1R1_EL1		SYS_ICC_AP1Rn_EL1(1)
+#define SYS_ICC_AP1R2_EL1		SYS_ICC_AP1Rn_EL1(2)
+#define SYS_ICC_AP1R3_EL1		SYS_ICC_AP1Rn_EL1(3)
+#define SYS_ICC_DIR_EL1			sys_reg(3, 0, 12, 11, 1)
+#define SYS_ICC_RPR_EL1			sys_reg(3, 0, 12, 11, 3)
+#define SYS_ICC_SGI1R_EL1		sys_reg(3, 0, 12, 11, 5)
+#define SYS_ICC_ASGI1R_EL1		sys_reg(3, 0, 12, 11, 6)
+#define SYS_ICC_SGI0R_EL1		sys_reg(3, 0, 12, 11, 7)
+#define SYS_ICC_IAR1_EL1		sys_reg(3, 0, 12, 12, 0)
+#define SYS_ICC_EOIR1_EL1		sys_reg(3, 0, 12, 12, 1)
+#define SYS_ICC_HPPIR1_EL1		sys_reg(3, 0, 12, 12, 2)
+#define SYS_ICC_BPR1_EL1		sys_reg(3, 0, 12, 12, 3)
+#define SYS_ICC_CTLR_EL1		sys_reg(3, 0, 12, 12, 4)
+#define SYS_ICC_SRE_EL1			sys_reg(3, 0, 12, 12, 5)
+#define SYS_ICC_IGRPEN0_EL1		sys_reg(3, 0, 12, 12, 6)
+#define SYS_ICC_IGRPEN1_EL1		sys_reg(3, 0, 12, 12, 7)
+
+#define SYS_CONTEXTIDR_EL1		sys_reg(3, 0, 13, 0, 1)
+#define SYS_TPIDR_EL1			sys_reg(3, 0, 13, 0, 4)
+
+#define SYS_SCXTNUM_EL1			sys_reg(3, 0, 13, 0, 7)
+
+#define SYS_CNTKCTL_EL1			sys_reg(3, 0, 14, 1, 0)
+
+#define SYS_CCSIDR_EL1			sys_reg(3, 1, 0, 0, 0)
+#define SYS_CLIDR_EL1			sys_reg(3, 1, 0, 0, 1)
+#define SYS_GMID_EL1			sys_reg(3, 1, 0, 0, 4)
+#define SYS_AIDR_EL1			sys_reg(3, 1, 0, 0, 7)
+
+#define SYS_CSSELR_EL1			sys_reg(3, 2, 0, 0, 0)
+
+#define SYS_CTR_EL0			sys_reg(3, 3, 0, 0, 1)
+#define SYS_DCZID_EL0			sys_reg(3, 3, 0, 0, 7)
+
+#define SYS_RNDR_EL0			sys_reg(3, 3, 2, 4, 0)
+#define SYS_RNDRRS_EL0			sys_reg(3, 3, 2, 4, 1)
+
+#define SYS_PMCR_EL0			sys_reg(3, 3, 9, 12, 0)
+#define SYS_PMCNTENSET_EL0		sys_reg(3, 3, 9, 12, 1)
+#define SYS_PMCNTENCLR_EL0		sys_reg(3, 3, 9, 12, 2)
+#define SYS_PMOVSCLR_EL0		sys_reg(3, 3, 9, 12, 3)
+#define SYS_PMSWINC_EL0			sys_reg(3, 3, 9, 12, 4)
+#define SYS_PMSELR_EL0			sys_reg(3, 3, 9, 12, 5)
+#define SYS_PMCEID0_EL0			sys_reg(3, 3, 9, 12, 6)
+#define SYS_PMCEID1_EL0			sys_reg(3, 3, 9, 12, 7)
+#define SYS_PMCCNTR_EL0			sys_reg(3, 3, 9, 13, 0)
+#define SYS_PMXEVTYPER_EL0		sys_reg(3, 3, 9, 13, 1)
+#define SYS_PMXEVCNTR_EL0		sys_reg(3, 3, 9, 13, 2)
+#define SYS_PMUSERENR_EL0		sys_reg(3, 3, 9, 14, 0)
+#define SYS_PMOVSSET_EL0		sys_reg(3, 3, 9, 14, 3)
+
+#define SYS_TPIDR_EL0			sys_reg(3, 3, 13, 0, 2)
+#define SYS_TPIDRRO_EL0			sys_reg(3, 3, 13, 0, 3)
+
+#define SYS_SCXTNUM_EL0			sys_reg(3, 3, 13, 0, 7)
+
+/* Definitions for system register interface to AMU for ARMv8.4 onwards */
+#define SYS_AM_EL0(crm, op2)		sys_reg(3, 3, 13, (crm), (op2))
+#define SYS_AMCR_EL0			SYS_AM_EL0(2, 0)
+#define SYS_AMCFGR_EL0			SYS_AM_EL0(2, 1)
+#define SYS_AMCGCR_EL0			SYS_AM_EL0(2, 2)
+#define SYS_AMUSERENR_EL0		SYS_AM_EL0(2, 3)
+#define SYS_AMCNTENCLR0_EL0		SYS_AM_EL0(2, 4)
+#define SYS_AMCNTENSET0_EL0		SYS_AM_EL0(2, 5)
+#define SYS_AMCNTENCLR1_EL0		SYS_AM_EL0(3, 0)
+#define SYS_AMCNTENSET1_EL0		SYS_AM_EL0(3, 1)
+
+/*
+ * Group 0 of activity monitors (architected):
+ *                op0  op1  CRn   CRm       op2
+ * Counter:       11   011  1101  010:n<3>  n<2:0>
+ * Type:          11   011  1101  011:n<3>  n<2:0>
+ * n: 0-15
+ *
+ * Group 1 of activity monitors (auxiliary):
+ *                op0  op1  CRn   CRm       op2
+ * Counter:       11   011  1101  110:n<3>  n<2:0>
+ * Type:          11   011  1101  111:n<3>  n<2:0>
+ * n: 0-15
+ */
+
+#define SYS_AMEVCNTR0_EL0(n)		SYS_AM_EL0(4 + ((n) >> 3), (n) & 7)
+#define SYS_AMEVTYPER0_EL0(n)		SYS_AM_EL0(6 + ((n) >> 3), (n) & 7)
+#define SYS_AMEVCNTR1_EL0(n)		SYS_AM_EL0(12 + ((n) >> 3), (n) & 7)
+#define SYS_AMEVTYPER1_EL0(n)		SYS_AM_EL0(14 + ((n) >> 3), (n) & 7)
+
+/* AMU v1: Fixed (architecturally defined) activity monitors */
+#define SYS_AMEVCNTR0_CORE_EL0		SYS_AMEVCNTR0_EL0(0)
+#define SYS_AMEVCNTR0_CONST_EL0		SYS_AMEVCNTR0_EL0(1)
+#define SYS_AMEVCNTR0_INST_RET_EL0	SYS_AMEVCNTR0_EL0(2)
+#define SYS_AMEVCNTR0_MEM_STALL		SYS_AMEVCNTR0_EL0(3)
+
+#define SYS_CNTFRQ_EL0			sys_reg(3, 3, 14, 0, 0)
+
+#define SYS_CNTP_TVAL_EL0		sys_reg(3, 3, 14, 2, 0)
+#define SYS_CNTP_CTL_EL0		sys_reg(3, 3, 14, 2, 1)
+#define SYS_CNTP_CVAL_EL0		sys_reg(3, 3, 14, 2, 2)
+
+#define SYS_CNTV_CTL_EL0		sys_reg(3, 3, 14, 3, 1)
+#define SYS_CNTV_CVAL_EL0		sys_reg(3, 3, 14, 3, 2)
+
+#define SYS_AARCH32_CNTP_TVAL		sys_reg(0, 0, 14, 2, 0)
+#define SYS_AARCH32_CNTP_CTL		sys_reg(0, 0, 14, 2, 1)
+#define SYS_AARCH32_CNTP_CVAL		sys_reg(0, 2, 0, 14, 0)
+
+#define __PMEV_op2(n)			((n) & 0x7)
+#define __CNTR_CRm(n)			(0x8 | (((n) >> 3) & 0x3))
+#define SYS_PMEVCNTRn_EL0(n)		sys_reg(3, 3, 14, __CNTR_CRm(n), __PMEV_op2(n))
+#define __TYPER_CRm(n)			(0xc | (((n) >> 3) & 0x3))
+#define SYS_PMEVTYPERn_EL0(n)		sys_reg(3, 3, 14, __TYPER_CRm(n), __PMEV_op2(n))
+
+#define SYS_PMCCFILTR_EL0		sys_reg(3, 3, 14, 15, 7)
+
+#define SYS_SCTLR_EL2			sys_reg(3, 4, 1, 0, 0)
+#define SYS_HFGRTR_EL2			sys_reg(3, 4, 1, 1, 4)
+#define SYS_HFGWTR_EL2			sys_reg(3, 4, 1, 1, 5)
+#define SYS_HFGITR_EL2			sys_reg(3, 4, 1, 1, 6)
+#define SYS_ZCR_EL2			sys_reg(3, 4, 1, 2, 0)
+#define SYS_TRFCR_EL2			sys_reg(3, 4, 1, 2, 1)
+#define SYS_DACR32_EL2			sys_reg(3, 4, 3, 0, 0)
+#define SYS_HDFGRTR_EL2			sys_reg(3, 4, 3, 1, 4)
+#define SYS_HDFGWTR_EL2			sys_reg(3, 4, 3, 1, 5)
+#define SYS_HAFGRTR_EL2			sys_reg(3, 4, 3, 1, 6)
+#define SYS_SPSR_EL2			sys_reg(3, 4, 4, 0, 0)
+#define SYS_ELR_EL2			sys_reg(3, 4, 4, 0, 1)
+#define SYS_IFSR32_EL2			sys_reg(3, 4, 5, 0, 1)
+#define SYS_ESR_EL2			sys_reg(3, 4, 5, 2, 0)
+#define SYS_VSESR_EL2			sys_reg(3, 4, 5, 2, 3)
+#define SYS_FPEXC32_EL2			sys_reg(3, 4, 5, 3, 0)
+#define SYS_TFSR_EL2			sys_reg(3, 4, 5, 6, 0)
+#define SYS_FAR_EL2			sys_reg(3, 4, 6, 0, 0)
+
+#define SYS_VDISR_EL2			sys_reg(3, 4, 12, 1,  1)
+#define __SYS__AP0Rx_EL2(x)		sys_reg(3, 4, 12, 8, x)
+#define SYS_ICH_AP0R0_EL2		__SYS__AP0Rx_EL2(0)
+#define SYS_ICH_AP0R1_EL2		__SYS__AP0Rx_EL2(1)
+#define SYS_ICH_AP0R2_EL2		__SYS__AP0Rx_EL2(2)
+#define SYS_ICH_AP0R3_EL2		__SYS__AP0Rx_EL2(3)
+
+#define __SYS__AP1Rx_EL2(x)		sys_reg(3, 4, 12, 9, x)
+#define SYS_ICH_AP1R0_EL2		__SYS__AP1Rx_EL2(0)
+#define SYS_ICH_AP1R1_EL2		__SYS__AP1Rx_EL2(1)
+#define SYS_ICH_AP1R2_EL2		__SYS__AP1Rx_EL2(2)
+#define SYS_ICH_AP1R3_EL2		__SYS__AP1Rx_EL2(3)
+
+#define SYS_ICH_VSEIR_EL2		sys_reg(3, 4, 12, 9, 4)
+#define SYS_ICC_SRE_EL2			sys_reg(3, 4, 12, 9, 5)
+#define SYS_ICH_HCR_EL2			sys_reg(3, 4, 12, 11, 0)
+#define SYS_ICH_VTR_EL2			sys_reg(3, 4, 12, 11, 1)
+#define SYS_ICH_MISR_EL2		sys_reg(3, 4, 12, 11, 2)
+#define SYS_ICH_EISR_EL2		sys_reg(3, 4, 12, 11, 3)
+#define SYS_ICH_ELRSR_EL2		sys_reg(3, 4, 12, 11, 5)
+#define SYS_ICH_VMCR_EL2		sys_reg(3, 4, 12, 11, 7)
+
+#define __SYS__LR0_EL2(x)		sys_reg(3, 4, 12, 12, x)
+#define SYS_ICH_LR0_EL2			__SYS__LR0_EL2(0)
+#define SYS_ICH_LR1_EL2			__SYS__LR0_EL2(1)
+#define SYS_ICH_LR2_EL2			__SYS__LR0_EL2(2)
+#define SYS_ICH_LR3_EL2			__SYS__LR0_EL2(3)
+#define SYS_ICH_LR4_EL2			__SYS__LR0_EL2(4)
+#define SYS_ICH_LR5_EL2			__SYS__LR0_EL2(5)
+#define SYS_ICH_LR6_EL2			__SYS__LR0_EL2(6)
+#define SYS_ICH_LR7_EL2			__SYS__LR0_EL2(7)
+
+#define __SYS__LR8_EL2(x)		sys_reg(3, 4, 12, 13, x)
+#define SYS_ICH_LR8_EL2			__SYS__LR8_EL2(0)
+#define SYS_ICH_LR9_EL2			__SYS__LR8_EL2(1)
+#define SYS_ICH_LR10_EL2		__SYS__LR8_EL2(2)
+#define SYS_ICH_LR11_EL2		__SYS__LR8_EL2(3)
+#define SYS_ICH_LR12_EL2		__SYS__LR8_EL2(4)
+#define SYS_ICH_LR13_EL2		__SYS__LR8_EL2(5)
+#define SYS_ICH_LR14_EL2		__SYS__LR8_EL2(6)
+#define SYS_ICH_LR15_EL2		__SYS__LR8_EL2(7)
+
+/* VHE encodings for architectural EL0/1 system registers */
+#define SYS_SCTLR_EL12			sys_reg(3, 5, 1, 0, 0)
+#define SYS_CPACR_EL12			sys_reg(3, 5, 1, 0, 2)
+#define SYS_ZCR_EL12			sys_reg(3, 5, 1, 2, 0)
+#define SYS_TTBR0_EL12			sys_reg(3, 5, 2, 0, 0)
+#define SYS_TTBR1_EL12			sys_reg(3, 5, 2, 0, 1)
+#define SYS_TCR_EL12			sys_reg(3, 5, 2, 0, 2)
+#define SYS_SPSR_EL12			sys_reg(3, 5, 4, 0, 0)
+#define SYS_ELR_EL12			sys_reg(3, 5, 4, 0, 1)
+#define SYS_AFSR0_EL12			sys_reg(3, 5, 5, 1, 0)
+#define SYS_AFSR1_EL12			sys_reg(3, 5, 5, 1, 1)
+#define SYS_ESR_EL12			sys_reg(3, 5, 5, 2, 0)
+#define SYS_TFSR_EL12			sys_reg(3, 5, 5, 6, 0)
+#define SYS_FAR_EL12			sys_reg(3, 5, 6, 0, 0)
+#define SYS_MAIR_EL12			sys_reg(3, 5, 10, 2, 0)
+#define SYS_AMAIR_EL12			sys_reg(3, 5, 10, 3, 0)
+#define SYS_VBAR_EL12			sys_reg(3, 5, 12, 0, 0)
+#define SYS_CONTEXTIDR_EL12		sys_reg(3, 5, 13, 0, 1)
+#define SYS_CNTKCTL_EL12		sys_reg(3, 5, 14, 1, 0)
+#define SYS_CNTP_TVAL_EL02		sys_reg(3, 5, 14, 2, 0)
+#define SYS_CNTP_CTL_EL02		sys_reg(3, 5, 14, 2, 1)
+#define SYS_CNTP_CVAL_EL02		sys_reg(3, 5, 14, 2, 2)
+#define SYS_CNTV_TVAL_EL02		sys_reg(3, 5, 14, 3, 0)
+#define SYS_CNTV_CTL_EL02		sys_reg(3, 5, 14, 3, 1)
+#define SYS_CNTV_CVAL_EL02		sys_reg(3, 5, 14, 3, 2)
+
+/* Common SCTLR_ELx flags. */
+#define SCTLR_ELx_DSSBS	(BIT(44))
+#define SCTLR_ELx_ATA	(BIT(43))
+
+#define SCTLR_ELx_TCF_SHIFT	40
+#define SCTLR_ELx_TCF_NONE	(UL(0x0) << SCTLR_ELx_TCF_SHIFT)
+#define SCTLR_ELx_TCF_SYNC	(UL(0x1) << SCTLR_ELx_TCF_SHIFT)
+#define SCTLR_ELx_TCF_ASYNC	(UL(0x2) << SCTLR_ELx_TCF_SHIFT)
+#define SCTLR_ELx_TCF_MASK	(UL(0x3) << SCTLR_ELx_TCF_SHIFT)
+
+#define SCTLR_ELx_ENIA_SHIFT	31
+
+#define SCTLR_ELx_ITFSB	(BIT(37))
+#define SCTLR_ELx_ENIA	(BIT(SCTLR_ELx_ENIA_SHIFT))
+#define SCTLR_ELx_ENIB	(BIT(30))
+#define SCTLR_ELx_ENDA	(BIT(27))
+#define SCTLR_ELx_EE    (BIT(25))
+#define SCTLR_ELx_IESB	(BIT(21))
+#define SCTLR_ELx_WXN	(BIT(19))
+#define SCTLR_ELx_ENDB	(BIT(13))
+#define SCTLR_ELx_I	(BIT(12))
+#define SCTLR_ELx_SA	(BIT(3))
+#define SCTLR_ELx_C	(BIT(2))
+#define SCTLR_ELx_A	(BIT(1))
+#define SCTLR_ELx_M	(BIT(0))
+
+/* SCTLR_EL2 specific flags. */
+#define SCTLR_EL2_RES1	((BIT(4))  | (BIT(5))  | (BIT(11)) | (BIT(16)) | \
+			 (BIT(18)) | (BIT(22)) | (BIT(23)) | (BIT(28)) | \
+			 (BIT(29)))
+
+#ifdef CONFIG_CPU_BIG_ENDIAN
+#define ENDIAN_SET_EL2		SCTLR_ELx_EE
+#else
+#define ENDIAN_SET_EL2		0
+#endif
+
+#define INIT_SCTLR_EL2_MMU_ON						\
+	(SCTLR_ELx_M  | SCTLR_ELx_C | SCTLR_ELx_SA | SCTLR_ELx_I |	\
+	 SCTLR_ELx_IESB | SCTLR_ELx_WXN | ENDIAN_SET_EL2 |		\
+	 SCTLR_ELx_ITFSB | SCTLR_EL2_RES1)
+
+#define INIT_SCTLR_EL2_MMU_OFF \
+	(SCTLR_EL2_RES1 | ENDIAN_SET_EL2)
+
+/* SCTLR_EL1 specific flags. */
+#define SCTLR_EL1_EPAN		(BIT(57))
+#define SCTLR_EL1_ATA0		(BIT(42))
+
+#define SCTLR_EL1_TCF0_SHIFT	38
+#define SCTLR_EL1_TCF0_NONE	(UL(0x0) << SCTLR_EL1_TCF0_SHIFT)
+#define SCTLR_EL1_TCF0_SYNC	(UL(0x1) << SCTLR_EL1_TCF0_SHIFT)
+#define SCTLR_EL1_TCF0_ASYNC	(UL(0x2) << SCTLR_EL1_TCF0_SHIFT)
+#define SCTLR_EL1_TCF0_MASK	(UL(0x3) << SCTLR_EL1_TCF0_SHIFT)
+
+#define SCTLR_EL1_BT1		(BIT(36))
+#define SCTLR_EL1_BT0		(BIT(35))
+#define SCTLR_EL1_UCI		(BIT(26))
+#define SCTLR_EL1_E0E		(BIT(24))
+#define SCTLR_EL1_SPAN		(BIT(23))
+#define SCTLR_EL1_NTWE		(BIT(18))
+#define SCTLR_EL1_NTWI		(BIT(16))
+#define SCTLR_EL1_UCT		(BIT(15))
+#define SCTLR_EL1_DZE		(BIT(14))
+#define SCTLR_EL1_UMA		(BIT(9))
+#define SCTLR_EL1_SED		(BIT(8))
+#define SCTLR_EL1_ITD		(BIT(7))
+#define SCTLR_EL1_CP15BEN	(BIT(5))
+#define SCTLR_EL1_SA0		(BIT(4))
+
+#define SCTLR_EL1_RES1	((BIT(11)) | (BIT(20)) | (BIT(22)) | (BIT(28)) | \
+			 (BIT(29)))
+
+#ifdef CONFIG_CPU_BIG_ENDIAN
+#define ENDIAN_SET_EL1		(SCTLR_EL1_E0E | SCTLR_ELx_EE)
+#else
+#define ENDIAN_SET_EL1		0
+#endif
+
+#define INIT_SCTLR_EL1_MMU_OFF \
+	(ENDIAN_SET_EL1 | SCTLR_EL1_RES1)
+
+#define INIT_SCTLR_EL1_MMU_ON \
+	(SCTLR_ELx_M    | SCTLR_ELx_C    | SCTLR_ELx_SA   | SCTLR_EL1_SA0   | \
+	 SCTLR_EL1_SED  | SCTLR_ELx_I    | SCTLR_EL1_DZE  | SCTLR_EL1_UCT   | \
+	 SCTLR_EL1_NTWE | SCTLR_ELx_IESB | SCTLR_EL1_SPAN | SCTLR_ELx_ITFSB | \
+	 SCTLR_ELx_ATA  | SCTLR_EL1_ATA0 | ENDIAN_SET_EL1 | SCTLR_EL1_UCI   | \
+	 SCTLR_EL1_EPAN | SCTLR_EL1_RES1)
+
+/* MAIR_ELx memory attributes (used by Linux) */
+#define MAIR_ATTR_DEVICE_nGnRnE		UL(0x00)
+#define MAIR_ATTR_DEVICE_nGnRE		UL(0x04)
+#define MAIR_ATTR_NORMAL_NC		UL(0x44)
+#define MAIR_ATTR_NORMAL_TAGGED		UL(0xf0)
+#define MAIR_ATTR_NORMAL		UL(0xff)
+#define MAIR_ATTR_MASK			UL(0xff)
+
+/* Position the attr at the correct index */
+#define MAIR_ATTRIDX(attr, idx)		((attr) << ((idx) * 8))
+
+/* id_aa64isar0 */
+#define ID_AA64ISAR0_RNDR_SHIFT		60
+#define ID_AA64ISAR0_TLB_SHIFT		56
+#define ID_AA64ISAR0_TS_SHIFT		52
+#define ID_AA64ISAR0_FHM_SHIFT		48
+#define ID_AA64ISAR0_DP_SHIFT		44
+#define ID_AA64ISAR0_SM4_SHIFT		40
+#define ID_AA64ISAR0_SM3_SHIFT		36
+#define ID_AA64ISAR0_SHA3_SHIFT		32
+#define ID_AA64ISAR0_RDM_SHIFT		28
+#define ID_AA64ISAR0_ATOMICS_SHIFT	20
+#define ID_AA64ISAR0_CRC32_SHIFT	16
+#define ID_AA64ISAR0_SHA2_SHIFT		12
+#define ID_AA64ISAR0_SHA1_SHIFT		8
+#define ID_AA64ISAR0_AES_SHIFT		4
+
+#define ID_AA64ISAR0_TLB_RANGE_NI	0x0
+#define ID_AA64ISAR0_TLB_RANGE		0x2
+
+/* id_aa64isar1 */
+#define ID_AA64ISAR1_I8MM_SHIFT		52
+#define ID_AA64ISAR1_DGH_SHIFT		48
+#define ID_AA64ISAR1_BF16_SHIFT		44
+#define ID_AA64ISAR1_SPECRES_SHIFT	40
+#define ID_AA64ISAR1_SB_SHIFT		36
+#define ID_AA64ISAR1_FRINTTS_SHIFT	32
+#define ID_AA64ISAR1_GPI_SHIFT		28
+#define ID_AA64ISAR1_GPA_SHIFT		24
+#define ID_AA64ISAR1_LRCPC_SHIFT	20
+#define ID_AA64ISAR1_FCMA_SHIFT		16
+#define ID_AA64ISAR1_JSCVT_SHIFT	12
+#define ID_AA64ISAR1_API_SHIFT		8
+#define ID_AA64ISAR1_APA_SHIFT		4
+#define ID_AA64ISAR1_DPB_SHIFT		0
+
+#define ID_AA64ISAR1_APA_NI			0x0
+#define ID_AA64ISAR1_APA_ARCHITECTED		0x1
+#define ID_AA64ISAR1_APA_ARCH_EPAC		0x2
+#define ID_AA64ISAR1_APA_ARCH_EPAC2		0x3
+#define ID_AA64ISAR1_APA_ARCH_EPAC2_FPAC	0x4
+#define ID_AA64ISAR1_APA_ARCH_EPAC2_FPAC_CMB	0x5
+#define ID_AA64ISAR1_API_NI			0x0
+#define ID_AA64ISAR1_API_IMP_DEF		0x1
+#define ID_AA64ISAR1_API_IMP_DEF_EPAC		0x2
+#define ID_AA64ISAR1_API_IMP_DEF_EPAC2		0x3
+#define ID_AA64ISAR1_API_IMP_DEF_EPAC2_FPAC	0x4
+#define ID_AA64ISAR1_API_IMP_DEF_EPAC2_FPAC_CMB	0x5
+#define ID_AA64ISAR1_GPA_NI			0x0
+#define ID_AA64ISAR1_GPA_ARCHITECTED		0x1
+#define ID_AA64ISAR1_GPI_NI			0x0
+#define ID_AA64ISAR1_GPI_IMP_DEF		0x1
+
+/* id_aa64pfr0 */
+#define ID_AA64PFR0_CSV3_SHIFT		60
+#define ID_AA64PFR0_CSV2_SHIFT		56
+#define ID_AA64PFR0_DIT_SHIFT		48
+#define ID_AA64PFR0_AMU_SHIFT		44
+#define ID_AA64PFR0_MPAM_SHIFT		40
+#define ID_AA64PFR0_SEL2_SHIFT		36
+#define ID_AA64PFR0_SVE_SHIFT		32
+#define ID_AA64PFR0_RAS_SHIFT		28
+#define ID_AA64PFR0_GIC_SHIFT		24
+#define ID_AA64PFR0_ASIMD_SHIFT		20
+#define ID_AA64PFR0_FP_SHIFT		16
+#define ID_AA64PFR0_EL3_SHIFT		12
+#define ID_AA64PFR0_EL2_SHIFT		8
+#define ID_AA64PFR0_EL1_SHIFT		4
+#define ID_AA64PFR0_EL0_SHIFT		0
+
+#define ID_AA64PFR0_AMU			0x1
+#define ID_AA64PFR0_SVE			0x1
+#define ID_AA64PFR0_RAS_V1		0x1
+#define ID_AA64PFR0_RAS_V1P1		0x2
+#define ID_AA64PFR0_FP_NI		0xf
+#define ID_AA64PFR0_FP_SUPPORTED	0x0
+#define ID_AA64PFR0_ASIMD_NI		0xf
+#define ID_AA64PFR0_ASIMD_SUPPORTED	0x0
+#define ID_AA64PFR0_ELx_64BIT_ONLY	0x1
+#define ID_AA64PFR0_ELx_32BIT_64BIT	0x2
+
+/* id_aa64pfr1 */
+#define ID_AA64PFR1_MPAMFRAC_SHIFT	16
+#define ID_AA64PFR1_RASFRAC_SHIFT	12
+#define ID_AA64PFR1_MTE_SHIFT		8
+#define ID_AA64PFR1_SSBS_SHIFT		4
+#define ID_AA64PFR1_BT_SHIFT		0
+
+#define ID_AA64PFR1_SSBS_PSTATE_NI	0
+#define ID_AA64PFR1_SSBS_PSTATE_ONLY	1
+#define ID_AA64PFR1_SSBS_PSTATE_INSNS	2
+#define ID_AA64PFR1_BT_BTI		0x1
+
+#define ID_AA64PFR1_MTE_NI		0x0
+#define ID_AA64PFR1_MTE_EL0		0x1
+#define ID_AA64PFR1_MTE			0x2
+
+/* id_aa64zfr0 */
+#define ID_AA64ZFR0_F64MM_SHIFT		56
+#define ID_AA64ZFR0_F32MM_SHIFT		52
+#define ID_AA64ZFR0_I8MM_SHIFT		44
+#define ID_AA64ZFR0_SM4_SHIFT		40
+#define ID_AA64ZFR0_SHA3_SHIFT		32
+#define ID_AA64ZFR0_BF16_SHIFT		20
+#define ID_AA64ZFR0_BITPERM_SHIFT	16
+#define ID_AA64ZFR0_AES_SHIFT		4
+#define ID_AA64ZFR0_SVEVER_SHIFT	0
+
+#define ID_AA64ZFR0_F64MM		0x1
+#define ID_AA64ZFR0_F32MM		0x1
+#define ID_AA64ZFR0_I8MM		0x1
+#define ID_AA64ZFR0_BF16		0x1
+#define ID_AA64ZFR0_SM4			0x1
+#define ID_AA64ZFR0_SHA3		0x1
+#define ID_AA64ZFR0_BITPERM		0x1
+#define ID_AA64ZFR0_AES			0x1
+#define ID_AA64ZFR0_AES_PMULL		0x2
+#define ID_AA64ZFR0_SVEVER_SVE2		0x1
+
+/* id_aa64mmfr0 */
+#define ID_AA64MMFR0_ECV_SHIFT		60
+#define ID_AA64MMFR0_FGT_SHIFT		56
+#define ID_AA64MMFR0_EXS_SHIFT		44
+#define ID_AA64MMFR0_TGRAN4_2_SHIFT	40
+#define ID_AA64MMFR0_TGRAN64_2_SHIFT	36
+#define ID_AA64MMFR0_TGRAN16_2_SHIFT	32
+#define ID_AA64MMFR0_TGRAN4_SHIFT	28
+#define ID_AA64MMFR0_TGRAN64_SHIFT	24
+#define ID_AA64MMFR0_TGRAN16_SHIFT	20
+#define ID_AA64MMFR0_BIGENDEL0_SHIFT	16
+#define ID_AA64MMFR0_SNSMEM_SHIFT	12
+#define ID_AA64MMFR0_BIGENDEL_SHIFT	8
+#define ID_AA64MMFR0_ASID_SHIFT		4
+#define ID_AA64MMFR0_PARANGE_SHIFT	0
+
+#define ID_AA64MMFR0_ASID_8		0x0
+#define ID_AA64MMFR0_ASID_16		0x2
+
+#define ID_AA64MMFR0_TGRAN4_NI			0xf
+#define ID_AA64MMFR0_TGRAN4_SUPPORTED_MIN	0x0
+#define ID_AA64MMFR0_TGRAN4_SUPPORTED_MAX	0x7
+#define ID_AA64MMFR0_TGRAN64_NI			0xf
+#define ID_AA64MMFR0_TGRAN64_SUPPORTED_MIN	0x0
+#define ID_AA64MMFR0_TGRAN64_SUPPORTED_MAX	0x7
+#define ID_AA64MMFR0_TGRAN16_NI			0x0
+#define ID_AA64MMFR0_TGRAN16_SUPPORTED_MIN	0x1
+#define ID_AA64MMFR0_TGRAN16_SUPPORTED_MAX	0xf
+
+#define ID_AA64MMFR0_PARANGE_32		0x0
+#define ID_AA64MMFR0_PARANGE_36		0x1
+#define ID_AA64MMFR0_PARANGE_40		0x2
+#define ID_AA64MMFR0_PARANGE_42		0x3
+#define ID_AA64MMFR0_PARANGE_44		0x4
+#define ID_AA64MMFR0_PARANGE_48		0x5
+#define ID_AA64MMFR0_PARANGE_52		0x6
+
+#define ARM64_MIN_PARANGE_BITS		32
+
+#define ID_AA64MMFR0_TGRAN_2_SUPPORTED_DEFAULT	0x0
+#define ID_AA64MMFR0_TGRAN_2_SUPPORTED_NONE	0x1
+#define ID_AA64MMFR0_TGRAN_2_SUPPORTED_MIN	0x2
+#define ID_AA64MMFR0_TGRAN_2_SUPPORTED_MAX	0x7
+
+#ifdef CONFIG_ARM64_PA_BITS_52
+#define ID_AA64MMFR0_PARANGE_MAX	ID_AA64MMFR0_PARANGE_52
+#else
+#define ID_AA64MMFR0_PARANGE_MAX	ID_AA64MMFR0_PARANGE_48
+#endif
+
+/* id_aa64mmfr1 */
+#define ID_AA64MMFR1_ETS_SHIFT		36
+#define ID_AA64MMFR1_TWED_SHIFT		32
+#define ID_AA64MMFR1_XNX_SHIFT		28
+#define ID_AA64MMFR1_SPECSEI_SHIFT	24
+#define ID_AA64MMFR1_PAN_SHIFT		20
+#define ID_AA64MMFR1_LOR_SHIFT		16
+#define ID_AA64MMFR1_HPD_SHIFT		12
+#define ID_AA64MMFR1_VHE_SHIFT		8
+#define ID_AA64MMFR1_VMIDBITS_SHIFT	4
+#define ID_AA64MMFR1_HADBS_SHIFT	0
+
+#define ID_AA64MMFR1_VMIDBITS_8		0
+#define ID_AA64MMFR1_VMIDBITS_16	2
+
+/* id_aa64mmfr2 */
+#define ID_AA64MMFR2_E0PD_SHIFT		60
+#define ID_AA64MMFR2_EVT_SHIFT		56
+#define ID_AA64MMFR2_BBM_SHIFT		52
+#define ID_AA64MMFR2_TTL_SHIFT		48
+#define ID_AA64MMFR2_FWB_SHIFT		40
+#define ID_AA64MMFR2_IDS_SHIFT		36
+#define ID_AA64MMFR2_AT_SHIFT		32
+#define ID_AA64MMFR2_ST_SHIFT		28
+#define ID_AA64MMFR2_NV_SHIFT		24
+#define ID_AA64MMFR2_CCIDX_SHIFT	20
+#define ID_AA64MMFR2_LVA_SHIFT		16
+#define ID_AA64MMFR2_IESB_SHIFT		12
+#define ID_AA64MMFR2_LSM_SHIFT		8
+#define ID_AA64MMFR2_UAO_SHIFT		4
+#define ID_AA64MMFR2_CNP_SHIFT		0
+
+/* id_aa64dfr0 */
+#define ID_AA64DFR0_MTPMU_SHIFT		48
+#define ID_AA64DFR0_TRBE_SHIFT		44
+#define ID_AA64DFR0_TRACE_FILT_SHIFT	40
+#define ID_AA64DFR0_DOUBLELOCK_SHIFT	36
+#define ID_AA64DFR0_PMSVER_SHIFT	32
+#define ID_AA64DFR0_CTX_CMPS_SHIFT	28
+#define ID_AA64DFR0_WRPS_SHIFT		20
+#define ID_AA64DFR0_BRPS_SHIFT		12
+#define ID_AA64DFR0_PMUVER_SHIFT	8
+#define ID_AA64DFR0_TRACEVER_SHIFT	4
+#define ID_AA64DFR0_DEBUGVER_SHIFT	0
+
+#define ID_AA64DFR0_PMUVER_8_0		0x1
+#define ID_AA64DFR0_PMUVER_8_1		0x4
+#define ID_AA64DFR0_PMUVER_8_4		0x5
+#define ID_AA64DFR0_PMUVER_8_5		0x6
+#define ID_AA64DFR0_PMUVER_IMP_DEF	0xf
+
+#define ID_AA64DFR0_PMSVER_8_2		0x1
+#define ID_AA64DFR0_PMSVER_8_3		0x2
+
+#define ID_DFR0_PERFMON_SHIFT		24
+
+#define ID_DFR0_PERFMON_8_0		0x3
+#define ID_DFR0_PERFMON_8_1		0x4
+#define ID_DFR0_PERFMON_8_4		0x5
+#define ID_DFR0_PERFMON_8_5		0x6
+
+#define ID_ISAR4_SWP_FRAC_SHIFT		28
+#define ID_ISAR4_PSR_M_SHIFT		24
+#define ID_ISAR4_SYNCH_PRIM_FRAC_SHIFT	20
+#define ID_ISAR4_BARRIER_SHIFT		16
+#define ID_ISAR4_SMC_SHIFT		12
+#define ID_ISAR4_WRITEBACK_SHIFT	8
+#define ID_ISAR4_WITHSHIFTS_SHIFT	4
+#define ID_ISAR4_UNPRIV_SHIFT		0
+
+#define ID_DFR1_MTPMU_SHIFT		0
+
+#define ID_ISAR0_DIVIDE_SHIFT		24
+#define ID_ISAR0_DEBUG_SHIFT		20
+#define ID_ISAR0_COPROC_SHIFT		16
+#define ID_ISAR0_CMPBRANCH_SHIFT	12
+#define ID_ISAR0_BITFIELD_SHIFT		8
+#define ID_ISAR0_BITCOUNT_SHIFT		4
+#define ID_ISAR0_SWAP_SHIFT		0
+
+#define ID_ISAR5_RDM_SHIFT		24
+#define ID_ISAR5_CRC32_SHIFT		16
+#define ID_ISAR5_SHA2_SHIFT		12
+#define ID_ISAR5_SHA1_SHIFT		8
+#define ID_ISAR5_AES_SHIFT		4
+#define ID_ISAR5_SEVL_SHIFT		0
+
+#define ID_ISAR6_I8MM_SHIFT		24
+#define ID_ISAR6_BF16_SHIFT		20
+#define ID_ISAR6_SPECRES_SHIFT		16
+#define ID_ISAR6_SB_SHIFT		12
+#define ID_ISAR6_FHM_SHIFT		8
+#define ID_ISAR6_DP_SHIFT		4
+#define ID_ISAR6_JSCVT_SHIFT		0
+
+#define ID_MMFR0_INNERSHR_SHIFT		28
+#define ID_MMFR0_FCSE_SHIFT		24
+#define ID_MMFR0_AUXREG_SHIFT		20
+#define ID_MMFR0_TCM_SHIFT		16
+#define ID_MMFR0_SHARELVL_SHIFT		12
+#define ID_MMFR0_OUTERSHR_SHIFT		8
+#define ID_MMFR0_PMSA_SHIFT		4
+#define ID_MMFR0_VMSA_SHIFT		0
+
+#define ID_MMFR4_EVT_SHIFT		28
+#define ID_MMFR4_CCIDX_SHIFT		24
+#define ID_MMFR4_LSM_SHIFT		20
+#define ID_MMFR4_HPDS_SHIFT		16
+#define ID_MMFR4_CNP_SHIFT		12
+#define ID_MMFR4_XNX_SHIFT		8
+#define ID_MMFR4_AC2_SHIFT		4
+#define ID_MMFR4_SPECSEI_SHIFT		0
+
+#define ID_MMFR5_ETS_SHIFT		0
+
+#define ID_PFR0_DIT_SHIFT		24
+#define ID_PFR0_CSV2_SHIFT		16
+#define ID_PFR0_STATE3_SHIFT		12
+#define ID_PFR0_STATE2_SHIFT		8
+#define ID_PFR0_STATE1_SHIFT		4
+#define ID_PFR0_STATE0_SHIFT		0
+
+#define ID_DFR0_PERFMON_SHIFT		24
+#define ID_DFR0_MPROFDBG_SHIFT		20
+#define ID_DFR0_MMAPTRC_SHIFT		16
+#define ID_DFR0_COPTRC_SHIFT		12
+#define ID_DFR0_MMAPDBG_SHIFT		8
+#define ID_DFR0_COPSDBG_SHIFT		4
+#define ID_DFR0_COPDBG_SHIFT		0
+
+#define ID_PFR2_SSBS_SHIFT		4
+#define ID_PFR2_CSV3_SHIFT		0
+
+#define MVFR0_FPROUND_SHIFT		28
+#define MVFR0_FPSHVEC_SHIFT		24
+#define MVFR0_FPSQRT_SHIFT		20
+#define MVFR0_FPDIVIDE_SHIFT		16
+#define MVFR0_FPTRAP_SHIFT		12
+#define MVFR0_FPDP_SHIFT		8
+#define MVFR0_FPSP_SHIFT		4
+#define MVFR0_SIMD_SHIFT		0
+
+#define MVFR1_SIMDFMAC_SHIFT		28
+#define MVFR1_FPHP_SHIFT		24
+#define MVFR1_SIMDHP_SHIFT		20
+#define MVFR1_SIMDSP_SHIFT		16
+#define MVFR1_SIMDINT_SHIFT		12
+#define MVFR1_SIMDLS_SHIFT		8
+#define MVFR1_FPDNAN_SHIFT		4
+#define MVFR1_FPFTZ_SHIFT		0
+
+#define ID_PFR1_GIC_SHIFT		28
+#define ID_PFR1_VIRT_FRAC_SHIFT		24
+#define ID_PFR1_SEC_FRAC_SHIFT		20
+#define ID_PFR1_GENTIMER_SHIFT		16
+#define ID_PFR1_VIRTUALIZATION_SHIFT	12
+#define ID_PFR1_MPROGMOD_SHIFT		8
+#define ID_PFR1_SECURITY_SHIFT		4
+#define ID_PFR1_PROGMOD_SHIFT		0
+
+#if defined(CONFIG_ARM64_4K_PAGES)
+#define ID_AA64MMFR0_TGRAN_SHIFT		ID_AA64MMFR0_TGRAN4_SHIFT
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MIN	ID_AA64MMFR0_TGRAN4_SUPPORTED_MIN
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MAX	ID_AA64MMFR0_TGRAN4_SUPPORTED_MAX
+#define ID_AA64MMFR0_TGRAN_2_SHIFT		ID_AA64MMFR0_TGRAN4_2_SHIFT
+#elif defined(CONFIG_ARM64_16K_PAGES)
+#define ID_AA64MMFR0_TGRAN_SHIFT		ID_AA64MMFR0_TGRAN16_SHIFT
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MIN	ID_AA64MMFR0_TGRAN16_SUPPORTED_MIN
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MAX	ID_AA64MMFR0_TGRAN16_SUPPORTED_MAX
+#define ID_AA64MMFR0_TGRAN_2_SHIFT		ID_AA64MMFR0_TGRAN16_2_SHIFT
+#elif defined(CONFIG_ARM64_64K_PAGES)
+#define ID_AA64MMFR0_TGRAN_SHIFT		ID_AA64MMFR0_TGRAN64_SHIFT
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MIN	ID_AA64MMFR0_TGRAN64_SUPPORTED_MIN
+#define ID_AA64MMFR0_TGRAN_SUPPORTED_MAX	ID_AA64MMFR0_TGRAN64_SUPPORTED_MAX
+#define ID_AA64MMFR0_TGRAN_2_SHIFT		ID_AA64MMFR0_TGRAN64_2_SHIFT
+#endif
+
+#define MVFR2_FPMISC_SHIFT		4
+#define MVFR2_SIMDMISC_SHIFT		0
+
+#define DCZID_DZP_SHIFT			4
+#define DCZID_BS_SHIFT			0
+
+/*
+ * The ZCR_ELx_LEN_* definitions intentionally include bits [8:4] which
+ * are reserved by the SVE architecture for future expansion of the LEN
+ * field, with compatible semantics.
+ */
+#define ZCR_ELx_LEN_SHIFT	0
+#define ZCR_ELx_LEN_SIZE	9
+#define ZCR_ELx_LEN_MASK	0x1ff
+
+#define CPACR_EL1_ZEN_EL1EN	(BIT(16)) /* enable EL1 access */
+#define CPACR_EL1_ZEN_EL0EN	(BIT(17)) /* enable EL0 access, if EL1EN set */
+#define CPACR_EL1_ZEN		(CPACR_EL1_ZEN_EL1EN | CPACR_EL1_ZEN_EL0EN)
+
+/* TCR EL1 Bit Definitions */
+#define SYS_TCR_EL1_TCMA1	(BIT(58))
+#define SYS_TCR_EL1_TCMA0	(BIT(57))
+
+/* GCR_EL1 Definitions */
+#define SYS_GCR_EL1_RRND	(BIT(16))
+#define SYS_GCR_EL1_EXCL_MASK	0xffffUL
+
+/* RGSR_EL1 Definitions */
+#define SYS_RGSR_EL1_TAG_MASK	0xfUL
+#define SYS_RGSR_EL1_SEED_SHIFT	8
+#define SYS_RGSR_EL1_SEED_MASK	0xffffUL
+
+/* GMID_EL1 field definitions */
+#define SYS_GMID_EL1_BS_SHIFT	0
+#define SYS_GMID_EL1_BS_SIZE	4
+
+/* TFSR{,E0}_EL1 bit definitions */
+#define SYS_TFSR_EL1_TF0_SHIFT	0
+#define SYS_TFSR_EL1_TF1_SHIFT	1
+#define SYS_TFSR_EL1_TF0	(UL(1) << SYS_TFSR_EL1_TF0_SHIFT)
+#define SYS_TFSR_EL1_TF1	(UL(1) << SYS_TFSR_EL1_TF1_SHIFT)
+
+/* Safe value for MPIDR_EL1: Bit31:RES1, Bit30:U:0, Bit24:MT:0 */
+#define SYS_MPIDR_SAFE_VAL	(BIT(31))
+
+#define TRFCR_ELx_TS_SHIFT		5
+#define TRFCR_ELx_TS_VIRTUAL		((0x1UL) << TRFCR_ELx_TS_SHIFT)
+#define TRFCR_ELx_TS_GUEST_PHYSICAL	((0x2UL) << TRFCR_ELx_TS_SHIFT)
+#define TRFCR_ELx_TS_PHYSICAL		((0x3UL) << TRFCR_ELx_TS_SHIFT)
+#define TRFCR_EL2_CX			BIT(3)
+#define TRFCR_ELx_ExTRE			BIT(1)
+#define TRFCR_ELx_E0TRE			BIT(0)
+
+
+/* GIC Hypervisor interface registers */
+/* ICH_MISR_EL2 bit definitions */
+#define ICH_MISR_EOI		(1 << 0)
+#define ICH_MISR_U		(1 << 1)
+
+/* ICH_LR*_EL2 bit definitions */
+#define ICH_LR_VIRTUAL_ID_MASK	((1ULL << 32) - 1)
+
+#define ICH_LR_EOI		(1ULL << 41)
+#define ICH_LR_GROUP		(1ULL << 60)
+#define ICH_LR_HW		(1ULL << 61)
+#define ICH_LR_STATE		(3ULL << 62)
+#define ICH_LR_PENDING_BIT	(1ULL << 62)
+#define ICH_LR_ACTIVE_BIT	(1ULL << 63)
+#define ICH_LR_PHYS_ID_SHIFT	32
+#define ICH_LR_PHYS_ID_MASK	(0x3ffULL << ICH_LR_PHYS_ID_SHIFT)
+#define ICH_LR_PRIORITY_SHIFT	48
+#define ICH_LR_PRIORITY_MASK	(0xffULL << ICH_LR_PRIORITY_SHIFT)
+
+/* ICH_HCR_EL2 bit definitions */
+#define ICH_HCR_EN		(1 << 0)
+#define ICH_HCR_UIE		(1 << 1)
+#define ICH_HCR_NPIE		(1 << 3)
+#define ICH_HCR_TC		(1 << 10)
+#define ICH_HCR_TALL0		(1 << 11)
+#define ICH_HCR_TALL1		(1 << 12)
+#define ICH_HCR_EOIcount_SHIFT	27
+#define ICH_HCR_EOIcount_MASK	(0x1f << ICH_HCR_EOIcount_SHIFT)
+
+/* ICH_VMCR_EL2 bit definitions */
+#define ICH_VMCR_ACK_CTL_SHIFT	2
+#define ICH_VMCR_ACK_CTL_MASK	(1 << ICH_VMCR_ACK_CTL_SHIFT)
+#define ICH_VMCR_FIQ_EN_SHIFT	3
+#define ICH_VMCR_FIQ_EN_MASK	(1 << ICH_VMCR_FIQ_EN_SHIFT)
+#define ICH_VMCR_CBPR_SHIFT	4
+#define ICH_VMCR_CBPR_MASK	(1 << ICH_VMCR_CBPR_SHIFT)
+#define ICH_VMCR_EOIM_SHIFT	9
+#define ICH_VMCR_EOIM_MASK	(1 << ICH_VMCR_EOIM_SHIFT)
+#define ICH_VMCR_BPR1_SHIFT	18
+#define ICH_VMCR_BPR1_MASK	(7 << ICH_VMCR_BPR1_SHIFT)
+#define ICH_VMCR_BPR0_SHIFT	21
+#define ICH_VMCR_BPR0_MASK	(7 << ICH_VMCR_BPR0_SHIFT)
+#define ICH_VMCR_PMR_SHIFT	24
+#define ICH_VMCR_PMR_MASK	(0xffUL << ICH_VMCR_PMR_SHIFT)
+#define ICH_VMCR_ENG0_SHIFT	0
+#define ICH_VMCR_ENG0_MASK	(1 << ICH_VMCR_ENG0_SHIFT)
+#define ICH_VMCR_ENG1_SHIFT	1
+#define ICH_VMCR_ENG1_MASK	(1 << ICH_VMCR_ENG1_SHIFT)
+
+/* ICH_VTR_EL2 bit definitions */
+#define ICH_VTR_PRI_BITS_SHIFT	29
+#define ICH_VTR_PRI_BITS_MASK	(7 << ICH_VTR_PRI_BITS_SHIFT)
+#define ICH_VTR_ID_BITS_SHIFT	23
+#define ICH_VTR_ID_BITS_MASK	(7 << ICH_VTR_ID_BITS_SHIFT)
+#define ICH_VTR_SEIS_SHIFT	22
+#define ICH_VTR_SEIS_MASK	(1 << ICH_VTR_SEIS_SHIFT)
+#define ICH_VTR_A3V_SHIFT	21
+#define ICH_VTR_A3V_MASK	(1 << ICH_VTR_A3V_SHIFT)
+
+#define ARM64_FEATURE_FIELD_BITS	4
+
+/* Create a mask for the feature bits of the specified feature. */
+#define ARM64_FEATURE_MASK(x)	(GENMASK_ULL(x##_SHIFT + ARM64_FEATURE_FIELD_BITS - 1, x##_SHIFT))
+
+#ifdef __ASSEMBLY__
+
+	.irp	num,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30
+	.equ	.L__reg_num_x\num, \num
+	.endr
+	.equ	.L__reg_num_xzr, 31
+
+	.macro	mrs_s, rt, sreg
+	 __emit_inst(0xd5200000|(\sreg)|(.L__reg_num_\rt))
+	.endm
+
+	.macro	msr_s, sreg, rt
+	__emit_inst(0xd5000000|(\sreg)|(.L__reg_num_\rt))
+	.endm
+
+#else
+
+#include <linux/build_bug.h>
+#include <linux/types.h>
+#include <asm/alternative.h>
+
+#define __DEFINE_MRS_MSR_S_REGNUM				\
+"	.irp	num,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30\n" \
+"	.equ	.L__reg_num_x\\num, \\num\n"			\
+"	.endr\n"						\
+"	.equ	.L__reg_num_xzr, 31\n"
+
+#define DEFINE_MRS_S						\
+	__DEFINE_MRS_MSR_S_REGNUM				\
+"	.macro	mrs_s, rt, sreg\n"				\
+	__emit_inst(0xd5200000|(\\sreg)|(.L__reg_num_\\rt))	\
+"	.endm\n"
+
+#define DEFINE_MSR_S						\
+	__DEFINE_MRS_MSR_S_REGNUM				\
+"	.macro	msr_s, sreg, rt\n"				\
+	__emit_inst(0xd5000000|(\\sreg)|(.L__reg_num_\\rt))	\
+"	.endm\n"
+
+#define UNDEFINE_MRS_S						\
+"	.purgem	mrs_s\n"
+
+#define UNDEFINE_MSR_S						\
+"	.purgem	msr_s\n"
+
+#define __mrs_s(v, r)						\
+	DEFINE_MRS_S						\
+"	mrs_s " v ", " __stringify(r) "\n"			\
+	UNDEFINE_MRS_S
+
+#define __msr_s(r, v)						\
+	DEFINE_MSR_S						\
+"	msr_s " __stringify(r) ", " v "\n"			\
+	UNDEFINE_MSR_S
+
+/*
+ * Unlike read_cpuid, calls to read_sysreg are never expected to be
+ * optimized away or replaced with synthetic values.
+ */
+#define read_sysreg(r) ({					\
+	u64 __val;						\
+	asm volatile("mrs %0, " __stringify(r) : "=r" (__val));	\
+	__val;							\
+})
+
+/*
+ * The "Z" constraint normally means a zero immediate, but when combined with
+ * the "%x0" template means XZR.
+ */
+#define write_sysreg(v, r) do {					\
+	u64 __val = (u64)(v);					\
+	asm volatile("msr " __stringify(r) ", %x0"		\
+		     : : "rZ" (__val));				\
+} while (0)
+
+/*
+ * For registers without architectural names, or simply unsupported by
+ * GAS.
+ */
+#define read_sysreg_s(r) ({						\
+	u64 __val;							\
+	asm volatile(__mrs_s("%0", r) : "=r" (__val));			\
+	__val;								\
+})
+
+#define write_sysreg_s(v, r) do {					\
+	u64 __val = (u64)(v);						\
+	asm volatile(__msr_s(r, "%x0") : : "rZ" (__val));		\
+} while (0)
+
+/*
+ * Modify bits in a sysreg. Bits in the clear mask are zeroed, then bits in the
+ * set mask are set. Other bits are left as-is.
+ */
+#define sysreg_clear_set(sysreg, clear, set) do {			\
+	u64 __scs_val = read_sysreg(sysreg);				\
+	u64 __scs_new = (__scs_val & ~(u64)(clear)) | (set);		\
+	if (__scs_new != __scs_val)					\
+		write_sysreg(__scs_new, sysreg);			\
+} while (0)
+
+#define sysreg_clear_set_s(sysreg, clear, set) do {			\
+	u64 __scs_val = read_sysreg_s(sysreg);				\
+	u64 __scs_new = (__scs_val & ~(u64)(clear)) | (set);		\
+	if (__scs_new != __scs_val)					\
+		write_sysreg_s(__scs_new, sysreg);			\
+} while (0)
+
+#define read_sysreg_par() ({						\
+	u64 par;							\
+	asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412));	\
+	par = read_sysreg(par_el1);					\
+	asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412));	\
+	par;								\
+})
+
+#endif
+
+#endif	/* __ASM_SYSREG_H */

tools/arch/x86/include/asm/pvclock-abi.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_PVCLOCK_ABI_H
+#define _ASM_X86_PVCLOCK_ABI_H
+#ifndef __ASSEMBLY__
+
+/*
+ * These structs MUST NOT be changed.
+ * They are the ABI between hypervisor and guest OS.
+ * Both Xen and KVM are using this.
+ *
+ * pvclock_vcpu_time_info holds the system time and the tsc timestamp
+ * of the last update. So the guest can use the tsc delta to get a
+ * more precise system time.  There is one per virtual cpu.
+ *
+ * pvclock_wall_clock references the point in time when the system
+ * time was zero (usually boot time), thus the guest calculates the
+ * current wall clock by adding the system time.
+ *
+ * Protocol for the "version" fields is: hypervisor raises it (making
+ * it uneven) before it starts updating the fields and raises it again
+ * (making it even) when it is done.  Thus the guest can make sure the
+ * time values it got are consistent by checking the version before
+ * and after reading them.
+ */
+
+struct pvclock_vcpu_time_info {
+	u32   version;
+	u32   pad0;
+	u64   tsc_timestamp;
+	u64   system_time;
+	u32   tsc_to_system_mul;
+	s8    tsc_shift;
+	u8    flags;
+	u8    pad[2];
+} __attribute__((__packed__)); /* 32 bytes */
+
+struct pvclock_wall_clock {
+	u32   version;
+	u32   sec;
+	u32   nsec;
+} __attribute__((__packed__));
+
+#define PVCLOCK_TSC_STABLE_BIT	(1 << 0)
+#define PVCLOCK_GUEST_STOPPED	(1 << 1)
+/* PVCLOCK_COUNTS_FROM_ZERO broke ABI and can't be used anymore. */
+#define PVCLOCK_COUNTS_FROM_ZERO (1 << 2)
+#endif /* __ASSEMBLY__ */
+#endif /* _ASM_X86_PVCLOCK_ABI_H */

tools/arch/x86/include/asm/pvclock.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_PVCLOCK_H
+#define _ASM_X86_PVCLOCK_H
+
+#include <asm/barrier.h>
+#include <asm/pvclock-abi.h>
+
+/* some helper functions for xen and kvm pv clock sources */
+u64 pvclock_clocksource_read(struct pvclock_vcpu_time_info *src);
+u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src);
+void pvclock_set_flags(u8 flags);
+unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src);
+void pvclock_resume(void);
+
+void pvclock_touch_watchdogs(void);
+
+static __always_inline
+unsigned pvclock_read_begin(const struct pvclock_vcpu_time_info *src)
+{
+	unsigned version = src->version & ~1;
+	/* Make sure that the version is read before the data. */
+	rmb();
+	return version;
+}
+
+static __always_inline
+bool pvclock_read_retry(const struct pvclock_vcpu_time_info *src,
+			unsigned version)
+{
+	/* Make sure that the version is re-read after the data. */
+	rmb();
+	return version != src->version;
+}
+
+/*
+ * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
+ * yielding a 64-bit result.
+ */
+static inline u64 pvclock_scale_delta(u64 delta, u32 mul_frac, int shift)
+{
+	u64 product;
+#ifdef __i386__
+	u32 tmp1, tmp2;
+#else
+	unsigned long tmp;
+#endif
+
+	if (shift < 0)
+		delta >>= -shift;
+	else
+		delta <<= shift;
+
+#ifdef __i386__
+	__asm__ (
+		"mul  %5       ; "
+		"mov  %4,%%eax ; "
+		"mov  %%edx,%4 ; "
+		"mul  %5       ; "
+		"xor  %5,%5    ; "
+		"add  %4,%%eax ; "
+		"adc  %5,%%edx ; "
+		: "=A" (product), "=r" (tmp1), "=r" (tmp2)
+		: "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
+#elif defined(__x86_64__)
+	__asm__ (
+		"mulq %[mul_frac] ; shrd $32, %[hi], %[lo]"
+		: [lo]"=a"(product),
+		  [hi]"=d"(tmp)
+		: "0"(delta),
+		  [mul_frac]"rm"((u64)mul_frac));
+#else
+#error implement me!
+#endif
+
+	return product;
+}
+
+static __always_inline
+u64 __pvclock_read_cycles(const struct pvclock_vcpu_time_info *src, u64 tsc)
+{
+	u64 delta = tsc - src->tsc_timestamp;
+	u64 offset = pvclock_scale_delta(delta, src->tsc_to_system_mul,
+					     src->tsc_shift);
+	return src->system_time + offset;
+}
+
+struct pvclock_vsyscall_time_info {
+	struct pvclock_vcpu_time_info pvti;
+} __attribute__((__aligned__(64)));
+
+#define PVTI_SIZE sizeof(struct pvclock_vsyscall_time_info)
+
+#ifdef CONFIG_PARAVIRT_CLOCK
+void pvclock_set_pvti_cpu0_va(struct pvclock_vsyscall_time_info *pvti);
+struct pvclock_vsyscall_time_info *pvclock_get_pvti_cpu0_va(void);
+#else
+static inline struct pvclock_vsyscall_time_info *pvclock_get_pvti_cpu0_va(void)
+{
+	return NULL;
+}
+#endif
+
+#endif /* _ASM_X86_PVCLOCK_H */

tools/testing/selftests/kvm/.gitignore

···
 # SPDX-License-Identifier: GPL-2.0-only
+/aarch64/arch_timer
 /aarch64/debug-exceptions
 /aarch64/get-reg-list
 /aarch64/psci_cpu_on_test
···
 /x86_64/emulator_error_test
 /x86_64/get_cpuid_test
 /x86_64/get_msr_index_features
+/x86_64/kvm_clock_test
 /x86_64/kvm_pv_test
 /x86_64/hyperv_clock
 /x86_64/hyperv_cpuid
···
 /set_memory_region_test
 /steal_time
 /kvm_binary_stats_test
+/system_counter_offset_test

tools/testing/selftests/kvm/Makefile

···
 
 LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/rbtree.c lib/sparsebit.c lib/test_util.c lib/guest_modes.c lib/perf_test_util.c
 LIBKVM_x86_64 = lib/x86_64/apic.c lib/x86_64/processor.c lib/x86_64/vmx.c lib/x86_64/svm.c lib/x86_64/ucall.c lib/x86_64/handlers.S
-LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c lib/aarch64/handlers.S
+LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c lib/aarch64/handlers.S lib/aarch64/spinlock.c lib/aarch64/gic.c lib/aarch64/gic_v3.c lib/aarch64/vgic.c
 LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c
 
 TEST_GEN_PROGS_x86_64 = x86_64/cr4_cpuid_sync_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/kvm_clock_test
 TEST_GEN_PROGS_x86_64 += x86_64/kvm_pv_test
 TEST_GEN_PROGS_x86_64 += x86_64/mmio_warning_test
 TEST_GEN_PROGS_x86_64 += x86_64/mmu_role_test
···
 TEST_GEN_PROGS_x86_64 += set_memory_region_test
 TEST_GEN_PROGS_x86_64 += steal_time
 TEST_GEN_PROGS_x86_64 += kvm_binary_stats_test
+TEST_GEN_PROGS_x86_64 += system_counter_offset_test
 
+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 += dirty_log_perf_test
 TEST_GEN_PROGS_aarch64 += kvm_create_max_vcpus
 TEST_GEN_PROGS_aarch64 += kvm_page_table_test
+TEST_GEN_PROGS_aarch64 += memslot_modification_stress_test
+TEST_GEN_PROGS_aarch64 += memslot_perf_test
 TEST_GEN_PROGS_aarch64 += rseq_test
 TEST_GEN_PROGS_aarch64 += set_memory_region_test
 TEST_GEN_PROGS_aarch64 += steal_time

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

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * arch_timer.c - Tests the aarch64 timer IRQ functionality
+ *
+ * The test validates both the virtual and physical timer IRQs using
+ * CVAL and TVAL registers. This consitutes the four stages in the test.
+ * The guest's main thread configures the timer interrupt for a stage
+ * and waits for it to fire, with a timeout equal to the timer period.
+ * It asserts that the timeout doesn't exceed the timer period.
+ *
+ * On the other hand, upon receipt of an interrupt, the guest's interrupt
+ * handler validates the interrupt by checking if the architectural state
+ * is in compliance with the specifications.
+ *
+ * The test provides command-line options to configure the timer's
+ * period (-p), number of vCPUs (-n), and iterations per stage (-i).
+ * To stress-test the timer stack even more, an option to migrate the
+ * vCPUs across pCPUs (-m), at a particular rate, is also provided.
+ *
+ * Copyright (c) 2021, Google LLC.
+ */
+
+#define _GNU_SOURCE
+
+#include <stdlib.h>
+#include <pthread.h>
+#include <linux/kvm.h>
+#include <linux/sizes.h>
+#include <linux/bitmap.h>
+#include <sys/sysinfo.h>
+
+#include "kvm_util.h"
+#include "processor.h"
+#include "delay.h"
+#include "arch_timer.h"
+#include "gic.h"
+#include "vgic.h"
+
+#define NR_VCPUS_DEF			4
+#define NR_TEST_ITERS_DEF		5
+#define TIMER_TEST_PERIOD_MS_DEF	10
+#define TIMER_TEST_ERR_MARGIN_US	100
+#define TIMER_TEST_MIGRATION_FREQ_MS	2
+
+struct test_args {
+	int nr_vcpus;
+	int nr_iter;
+	int timer_period_ms;
+	int migration_freq_ms;
+};
+
+static struct test_args test_args = {
+	.nr_vcpus = NR_VCPUS_DEF,
+	.nr_iter = NR_TEST_ITERS_DEF,
+	.timer_period_ms = TIMER_TEST_PERIOD_MS_DEF,
+	.migration_freq_ms = TIMER_TEST_MIGRATION_FREQ_MS,
+};
+
+#define msecs_to_usecs(msec)		((msec) * 1000LL)
+
+#define GICD_BASE_GPA			0x8000000ULL
+#define GICR_BASE_GPA			0x80A0000ULL
+
+enum guest_stage {
+	GUEST_STAGE_VTIMER_CVAL = 1,
+	GUEST_STAGE_VTIMER_TVAL,
+	GUEST_STAGE_PTIMER_CVAL,
+	GUEST_STAGE_PTIMER_TVAL,
+	GUEST_STAGE_MAX,
+};
+
+/* Shared variables between host and guest */
+struct test_vcpu_shared_data {
+	int nr_iter;
+	enum guest_stage guest_stage;
+	uint64_t xcnt;
+};
+
+struct test_vcpu {
+	uint32_t vcpuid;
+	pthread_t pt_vcpu_run;
+	struct kvm_vm *vm;
+};
+
+static struct test_vcpu test_vcpu[KVM_MAX_VCPUS];
+static struct test_vcpu_shared_data vcpu_shared_data[KVM_MAX_VCPUS];
+
+static int vtimer_irq, ptimer_irq;
+
+static unsigned long *vcpu_done_map;
+static pthread_mutex_t vcpu_done_map_lock;
+
+static void
+guest_configure_timer_action(struct test_vcpu_shared_data *shared_data)
+{
+	switch (shared_data->guest_stage) {
+	case GUEST_STAGE_VTIMER_CVAL:
+		timer_set_next_cval_ms(VIRTUAL, test_args.timer_period_ms);
+		shared_data->xcnt = timer_get_cntct(VIRTUAL);
+		timer_set_ctl(VIRTUAL, CTL_ENABLE);
+		break;
+	case GUEST_STAGE_VTIMER_TVAL:
+		timer_set_next_tval_ms(VIRTUAL, test_args.timer_period_ms);
+		shared_data->xcnt = timer_get_cntct(VIRTUAL);
+		timer_set_ctl(VIRTUAL, CTL_ENABLE);
+		break;
+	case GUEST_STAGE_PTIMER_CVAL:
+		timer_set_next_cval_ms(PHYSICAL, test_args.timer_period_ms);
+		shared_data->xcnt = timer_get_cntct(PHYSICAL);
+		timer_set_ctl(PHYSICAL, CTL_ENABLE);
+		break;
+	case GUEST_STAGE_PTIMER_TVAL:
+		timer_set_next_tval_ms(PHYSICAL, test_args.timer_period_ms);
+		shared_data->xcnt = timer_get_cntct(PHYSICAL);
+		timer_set_ctl(PHYSICAL, CTL_ENABLE);
+		break;
+	default:
+		GUEST_ASSERT(0);
+	}
+}
+
+static void guest_validate_irq(unsigned int intid,
+				struct test_vcpu_shared_data *shared_data)
+{
+	enum guest_stage stage = shared_data->guest_stage;
+	uint64_t xcnt = 0, xcnt_diff_us, cval = 0;
+	unsigned long xctl = 0;
+	unsigned int timer_irq = 0;
+
+	if (stage == GUEST_STAGE_VTIMER_CVAL ||
+		stage == GUEST_STAGE_VTIMER_TVAL) {
+		xctl = timer_get_ctl(VIRTUAL);
+		timer_set_ctl(VIRTUAL, CTL_IMASK);
+		xcnt = timer_get_cntct(VIRTUAL);
+		cval = timer_get_cval(VIRTUAL);
+		timer_irq = vtimer_irq;
+	} else if (stage == GUEST_STAGE_PTIMER_CVAL ||
+		stage == GUEST_STAGE_PTIMER_TVAL) {
+		xctl = timer_get_ctl(PHYSICAL);
+		timer_set_ctl(PHYSICAL, CTL_IMASK);
+		xcnt = timer_get_cntct(PHYSICAL);
+		cval = timer_get_cval(PHYSICAL);
+		timer_irq = ptimer_irq;
+	} else {
+		GUEST_ASSERT(0);
+	}
+
+	xcnt_diff_us = cycles_to_usec(xcnt - shared_data->xcnt);
+
+	/* Make sure we are dealing with the correct timer IRQ */
+	GUEST_ASSERT_2(intid == timer_irq, intid, timer_irq);
+
+	/* Basic 'timer condition met' check */
+	GUEST_ASSERT_3(xcnt >= cval, xcnt, cval, xcnt_diff_us);
+	GUEST_ASSERT_1(xctl & CTL_ISTATUS, xctl);
+}
+
+static void guest_irq_handler(struct ex_regs *regs)
+{
+	unsigned int intid = gic_get_and_ack_irq();
+	uint32_t cpu = guest_get_vcpuid();
+	struct test_vcpu_shared_data *shared_data = &vcpu_shared_data[cpu];
+
+	guest_validate_irq(intid, shared_data);
+
+	WRITE_ONCE(shared_data->nr_iter, shared_data->nr_iter + 1);
+
+	gic_set_eoi(intid);
+}
+
+static void guest_run_stage(struct test_vcpu_shared_data *shared_data,
+				enum guest_stage stage)
+{
+	uint32_t irq_iter, config_iter;
+
+	shared_data->guest_stage = stage;
+	shared_data->nr_iter = 0;
+
+	for (config_iter = 0; config_iter < test_args.nr_iter; config_iter++) {
+		/* Setup the next interrupt */
+		guest_configure_timer_action(shared_data);
+
+		/* Setup a timeout for the interrupt to arrive */
+		udelay(msecs_to_usecs(test_args.timer_period_ms) +
+			TIMER_TEST_ERR_MARGIN_US);
+
+		irq_iter = READ_ONCE(shared_data->nr_iter);
+		GUEST_ASSERT_2(config_iter + 1 == irq_iter,
+				config_iter + 1, irq_iter);
+	}
+}
+
+static void guest_code(void)
+{
+	uint32_t cpu = guest_get_vcpuid();
+	struct test_vcpu_shared_data *shared_data = &vcpu_shared_data[cpu];
+
+	local_irq_disable();
+
+	gic_init(GIC_V3, test_args.nr_vcpus,
+		(void *)GICD_BASE_GPA, (void *)GICR_BASE_GPA);
+
+	timer_set_ctl(VIRTUAL, CTL_IMASK);
+	timer_set_ctl(PHYSICAL, CTL_IMASK);
+
+	gic_irq_enable(vtimer_irq);
+	gic_irq_enable(ptimer_irq);
+	local_irq_enable();
+
+	guest_run_stage(shared_data, GUEST_STAGE_VTIMER_CVAL);
+	guest_run_stage(shared_data, GUEST_STAGE_VTIMER_TVAL);
+	guest_run_stage(shared_data, GUEST_STAGE_PTIMER_CVAL);
+	guest_run_stage(shared_data, GUEST_STAGE_PTIMER_TVAL);
+
+	GUEST_DONE();
+}
+
+static void *test_vcpu_run(void *arg)
+{
+	struct ucall uc;
+	struct test_vcpu *vcpu = arg;
+	struct kvm_vm *vm = vcpu->vm;
+	uint32_t vcpuid = vcpu->vcpuid;
+	struct test_vcpu_shared_data *shared_data = &vcpu_shared_data[vcpuid];
+
+	vcpu_run(vm, vcpuid);
+
+	/* Currently, any exit from guest is an indication of completion */
+	pthread_mutex_lock(&vcpu_done_map_lock);
+	set_bit(vcpuid, vcpu_done_map);
+	pthread_mutex_unlock(&vcpu_done_map_lock);
+
+	switch (get_ucall(vm, vcpuid, &uc)) {
+	case UCALL_SYNC:
+	case UCALL_DONE:
+		break;
+	case UCALL_ABORT:
+		sync_global_from_guest(vm, *shared_data);
+		TEST_FAIL("%s at %s:%ld\n\tvalues: %lu, %lu; %lu, vcpu: %u; stage: %u; iter: %u",
+			(const char *)uc.args[0], __FILE__, uc.args[1],
+			uc.args[2], uc.args[3], uc.args[4], vcpuid,
+			shared_data->guest_stage, shared_data->nr_iter);
+		break;
+	default:
+		TEST_FAIL("Unexpected guest exit\n");
+	}
+
+	return NULL;
+}
+
+static uint32_t test_get_pcpu(void)
+{
+	uint32_t pcpu;
+	unsigned int nproc_conf;
+	cpu_set_t online_cpuset;
+
+	nproc_conf = get_nprocs_conf();
+	sched_getaffinity(0, sizeof(cpu_set_t), &online_cpuset);
+
+	/* Randomly find an available pCPU to place a vCPU on */
+	do {
+		pcpu = rand() % nproc_conf;
+	} while (!CPU_ISSET(pcpu, &online_cpuset));
+
+	return pcpu;
+}
+
+static int test_migrate_vcpu(struct test_vcpu *vcpu)
+{
+	int ret;
+	cpu_set_t cpuset;
+	uint32_t new_pcpu = test_get_pcpu();
+
+	CPU_ZERO(&cpuset);
+	CPU_SET(new_pcpu, &cpuset);
+
+	pr_debug("Migrating vCPU: %u to pCPU: %u\n", vcpu->vcpuid, new_pcpu);
+
+	ret = pthread_setaffinity_np(vcpu->pt_vcpu_run,
+					sizeof(cpuset), &cpuset);
+
+	/* Allow the error where the vCPU thread is already finished */
+	TEST_ASSERT(ret == 0 || ret == ESRCH,
+			"Failed to migrate the vCPU:%u to pCPU: %u; ret: %d\n",
+			vcpu->vcpuid, new_pcpu, ret);
+
+	return ret;
+}
+
+static void *test_vcpu_migration(void *arg)
+{
+	unsigned int i, n_done;
+	bool vcpu_done;
+
+	do {
+		usleep(msecs_to_usecs(test_args.migration_freq_ms));
+
+		for (n_done = 0, i = 0; i < test_args.nr_vcpus; i++) {
+			pthread_mutex_lock(&vcpu_done_map_lock);
+			vcpu_done = test_bit(i, vcpu_done_map);
+			pthread_mutex_unlock(&vcpu_done_map_lock);
+
+			if (vcpu_done) {
+				n_done++;
+				continue;
+			}
+
+			test_migrate_vcpu(&test_vcpu[i]);
+		}
+	} while (test_args.nr_vcpus != n_done);
+
+	return NULL;
+}
+
+static void test_run(struct kvm_vm *vm)
+{
+	int i, ret;
+	pthread_t pt_vcpu_migration;
+
+	pthread_mutex_init(&vcpu_done_map_lock, NULL);
+	vcpu_done_map = bitmap_zalloc(test_args.nr_vcpus);
+	TEST_ASSERT(vcpu_done_map, "Failed to allocate vcpu done bitmap\n");
+
+	for (i = 0; i < test_args.nr_vcpus; i++) {
+		ret = pthread_create(&test_vcpu[i].pt_vcpu_run, NULL,
+				test_vcpu_run, &test_vcpu[i]);
+		TEST_ASSERT(!ret, "Failed to create vCPU-%d pthread\n", i);
+	}
+
+	/* Spawn a thread to control the vCPU migrations */
+	if (test_args.migration_freq_ms) {
+		srand(time(NULL));
+
+		ret = pthread_create(&pt_vcpu_migration, NULL,
+					test_vcpu_migration, NULL);
+		TEST_ASSERT(!ret, "Failed to create the migration pthread\n");
+	}
+
+
+	for (i = 0; i < test_args.nr_vcpus; i++)
+		pthread_join(test_vcpu[i].pt_vcpu_run, NULL);
+
+	if (test_args.migration_freq_ms)
+		pthread_join(pt_vcpu_migration, NULL);
+
+	bitmap_free(vcpu_done_map);
+}
+
+static void test_init_timer_irq(struct kvm_vm *vm)
+{
+	/* Timer initid should be same for all the vCPUs, so query only vCPU-0 */
+	int vcpu0_fd = vcpu_get_fd(vm, 0);
+
+	kvm_device_access(vcpu0_fd, KVM_ARM_VCPU_TIMER_CTRL,
+			KVM_ARM_VCPU_TIMER_IRQ_PTIMER, &ptimer_irq, false);
+	kvm_device_access(vcpu0_fd, KVM_ARM_VCPU_TIMER_CTRL,
+			KVM_ARM_VCPU_TIMER_IRQ_VTIMER, &vtimer_irq, false);
+
+	sync_global_to_guest(vm, ptimer_irq);
+	sync_global_to_guest(vm, vtimer_irq);
+
+	pr_debug("ptimer_irq: %d; vtimer_irq: %d\n", ptimer_irq, vtimer_irq);
+}
+
+static struct kvm_vm *test_vm_create(void)
+{
+	struct kvm_vm *vm;
+	unsigned int i;
+	int nr_vcpus = test_args.nr_vcpus;
+
+	vm = vm_create_default_with_vcpus(nr_vcpus, 0, 0, guest_code, NULL);
+
+	vm_init_descriptor_tables(vm);
+	vm_install_exception_handler(vm, VECTOR_IRQ_CURRENT, guest_irq_handler);
+
+	for (i = 0; i < nr_vcpus; i++) {
+		vcpu_init_descriptor_tables(vm, i);
+
+		test_vcpu[i].vcpuid = i;
+		test_vcpu[i].vm = vm;
+	}
+
+	ucall_init(vm, NULL);
+	test_init_timer_irq(vm);
+	vgic_v3_setup(vm, nr_vcpus, GICD_BASE_GPA, GICR_BASE_GPA);
+
+	/* Make all the test's cmdline args visible to the guest */
+	sync_global_to_guest(vm, test_args);
+
+	return vm;
+}
+
+static void test_print_help(char *name)
+{
+	pr_info("Usage: %s [-h] [-n nr_vcpus] [-i iterations] [-p timer_period_ms]\n",
+		name);
+	pr_info("\t-n: Number of vCPUs to configure (default: %u; max: %u)\n",
+		NR_VCPUS_DEF, KVM_MAX_VCPUS);
+	pr_info("\t-i: Number of iterations per stage (default: %u)\n",
+		NR_TEST_ITERS_DEF);
+	pr_info("\t-p: Periodicity (in ms) of the guest timer (default: %u)\n",
+		TIMER_TEST_PERIOD_MS_DEF);
+	pr_info("\t-m: Frequency (in ms) of vCPUs to migrate to different pCPU. 0 to turn off (default: %u)\n",
+		TIMER_TEST_MIGRATION_FREQ_MS);
+	pr_info("\t-h: print this help screen\n");
+}
+
+static bool parse_args(int argc, char *argv[])
+{
+	int opt;
+
+	while ((opt = getopt(argc, argv, "hn:i:p:m:")) != -1) {
+		switch (opt) {
+		case 'n':
+			test_args.nr_vcpus = atoi(optarg);
+			if (test_args.nr_vcpus <= 0) {
+				pr_info("Positive value needed for -n\n");
+				goto err;
+			} else if (test_args.nr_vcpus > KVM_MAX_VCPUS) {
+				pr_info("Max allowed vCPUs: %u\n",
+					KVM_MAX_VCPUS);
+				goto err;
+			}
+			break;
+		case 'i':
+			test_args.nr_iter = atoi(optarg);
+			if (test_args.nr_iter <= 0) {
+				pr_info("Positive value needed for -i\n");
+				goto err;
+			}
+			break;
+		case 'p':
+			test_args.timer_period_ms = atoi(optarg);
+			if (test_args.timer_period_ms <= 0) {
+				pr_info("Positive value needed for -p\n");
+				goto err;
+			}
+			break;
+		case 'm':
+			test_args.migration_freq_ms = atoi(optarg);
+			if (test_args.migration_freq_ms < 0) {
+				pr_info("0 or positive value needed for -m\n");
+				goto err;
+			}
+			break;
+		case 'h':
+		default:
+			goto err;
+		}
+	}
+
+	return true;
+
+err:
+	test_print_help(argv[0]);
+	return false;
+}
+
+int main(int argc, char *argv[])
+{
+	struct kvm_vm *vm;
+
+	/* Tell stdout not to buffer its content */
+	setbuf(stdout, NULL);
+
+	if (!parse_args(argc, argv))
+		exit(KSFT_SKIP);
+
+	if (test_args.migration_freq_ms && get_nprocs() < 2) {
+		print_skip("At least two physical CPUs needed for vCPU migration");
+		exit(KSFT_SKIP);
+	}
+
+	vm = test_vm_create();
+	test_run(vm);
+	kvm_vm_free(vm);
+
+	return 0;
+}

tools/testing/selftests/kvm/aarch64/debug-exceptions.c

···
 {
 	asm volatile("msr daifset, #8");
 
-	write_sysreg(osdlr_el1, 0);
-	write_sysreg(oslar_el1, 0);
+	write_sysreg(0, osdlr_el1);
+	write_sysreg(0, oslar_el1);
 	isb();
 
-	write_sysreg(mdscr_el1, 0);
+	write_sysreg(0, mdscr_el1);
 	/* This test only uses the first bp and wp slot. */
-	write_sysreg(dbgbvr0_el1, 0);
-	write_sysreg(dbgbcr0_el1, 0);
-	write_sysreg(dbgwcr0_el1, 0);
-	write_sysreg(dbgwvr0_el1, 0);
+	write_sysreg(0, dbgbvr0_el1);
+	write_sysreg(0, dbgbcr0_el1);
+	write_sysreg(0, dbgwcr0_el1);
+	write_sysreg(0, dbgwvr0_el1);
 	isb();
 }
 
···
 	uint32_t mdscr;
 
 	wcr = DBGWCR_LEN8 | DBGWCR_RD | DBGWCR_WR | DBGWCR_EL1 | DBGWCR_E;
-	write_sysreg(dbgwcr0_el1, wcr);
-	write_sysreg(dbgwvr0_el1, addr);
+	write_sysreg(wcr, dbgwcr0_el1);
+	write_sysreg(addr, dbgwvr0_el1);
 	isb();
 
 	asm volatile("msr daifclr, #8");
 
 	mdscr = read_sysreg(mdscr_el1) | MDSCR_KDE | MDSCR_MDE;
-	write_sysreg(mdscr_el1, mdscr);
+	write_sysreg(mdscr, mdscr_el1);
 	isb();
 }
 
···
 	uint32_t mdscr;
 
 	bcr = DBGBCR_LEN8 | DBGBCR_EXEC | DBGBCR_EL1 | DBGBCR_E;
-	write_sysreg(dbgbcr0_el1, bcr);
-	write_sysreg(dbgbvr0_el1, addr);
+	write_sysreg(bcr, dbgbcr0_el1);
+	write_sysreg(addr, dbgbvr0_el1);
 	isb();
 
 	asm volatile("msr daifclr, #8");
 
 	mdscr = read_sysreg(mdscr_el1) | MDSCR_KDE | MDSCR_MDE;
-	write_sysreg(mdscr_el1, mdscr);
+	write_sysreg(mdscr, mdscr_el1);
 	isb();
 }
 
···
 	asm volatile("msr daifclr, #8");
 
 	mdscr = read_sysreg(mdscr_el1) | MDSCR_KDE | MDSCR_SS;
-	write_sysreg(mdscr_el1, mdscr);
+	write_sysreg(mdscr, mdscr_el1);
 	isb();
 }
 
···
 {
 	uint64_t id_aa64dfr0;
 
-	get_reg(vm, VCPU_ID, ARM64_SYS_REG(ID_AA64DFR0_EL1), &id_aa64dfr0);
+	get_reg(vm, VCPU_ID, KVM_ARM64_SYS_REG(SYS_ID_AA64DFR0_EL1), &id_aa64dfr0);
 	return id_aa64dfr0 & 0xf;
 }
 

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

···
 	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, ARM64_SYS_REG(MPIDR_EL1), &target_mpidr);
+	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);
 

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

···
 #include "test_util.h"
 #include "kvm_util.h"
 #include "processor.h"
+#include "vgic.h"
 
 #define NR_VCPUS		4
 
-#define REDIST_REGION_ATTR_ADDR(count, base, flags, index) (((uint64_t)(count) << 52) | \
-	((uint64_t)((base) >> 16) << 16) | ((uint64_t)(flags) << 12) | index)
 #define REG_OFFSET(vcpu, offset) (((uint64_t)vcpu << 32) | offset)
 
 #define GICR_TYPER 0x8
 
+#define VGIC_DEV_IS_V2(_d) ((_d) == KVM_DEV_TYPE_ARM_VGIC_V2)
+#define VGIC_DEV_IS_V3(_d) ((_d) == KVM_DEV_TYPE_ARM_VGIC_V3)
+
 struct vm_gic {
 	struct kvm_vm *vm;
 	int gic_fd;
+	uint32_t gic_dev_type;
 };
 
-static int max_ipa_bits;
+static uint64_t max_phys_size;
 
 /* helper to access a redistributor register */
-static int access_redist_reg(int gicv3_fd, int vcpu, int offset,
-			     uint32_t *val, bool write)
+static int access_v3_redist_reg(int gicv3_fd, int vcpu, int offset,
+				uint32_t *val, bool write)
 {
 	uint64_t attr = REG_OFFSET(vcpu, offset);
 
···
 	return 0;
 }
 
-static struct vm_gic vm_gic_create(void)
+static struct vm_gic vm_gic_create_with_vcpus(uint32_t gic_dev_type, uint32_t nr_vcpus)
 {
 	struct vm_gic v;
 
-	v.vm = vm_create_default_with_vcpus(NR_VCPUS, 0, 0, guest_code, NULL);
-	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+	v.gic_dev_type = gic_dev_type;
+	v.vm = vm_create_default_with_vcpus(nr_vcpus, 0, 0, guest_code, NULL);
+	v.gic_fd = kvm_create_device(v.vm, gic_dev_type, false);
 
 	return v;
 }
···
 	kvm_vm_free(v->vm);
 }
 
+struct vgic_region_attr {
+	uint64_t attr;
+	uint64_t size;
+	uint64_t alignment;
+};
+
+struct vgic_region_attr gic_v3_dist_region = {
+	.attr = KVM_VGIC_V3_ADDR_TYPE_DIST,
+	.size = 0x10000,
+	.alignment = 0x10000,
+};
+
+struct vgic_region_attr gic_v3_redist_region = {
+	.attr = KVM_VGIC_V3_ADDR_TYPE_REDIST,
+	.size = NR_VCPUS * 0x20000,
+	.alignment = 0x10000,
+};
+
+struct vgic_region_attr gic_v2_dist_region = {
+	.attr = KVM_VGIC_V2_ADDR_TYPE_DIST,
+	.size = 0x1000,
+	.alignment = 0x1000,
+};
+
+struct vgic_region_attr gic_v2_cpu_region = {
+	.attr = KVM_VGIC_V2_ADDR_TYPE_CPU,
+	.size = 0x2000,
+	.alignment = 0x1000,
+};
+
 /**
- * Helper routine that performs KVM device tests in general and
- * especially ARM_VGIC_V3 ones. Eventually the ARM_VGIC_V3
- * device gets created, a legacy RDIST region is set at @0x0
- * and a DIST region is set @0x60000
+ * Helper routine that performs KVM device tests in general. Eventually the
+ * ARM_VGIC (GICv2 or GICv3) device gets created with an overlapping
+ * DIST/REDIST (or DIST/CPUIF for GICv2). Assumption is 4 vcpus are going to be
+ * used hence the overlap. In the case of GICv3, A RDIST region is set at @0x0
+ * and a DIST region is set @0x70000. The GICv2 case sets a CPUIF @0x0 and a
+ * DIST region @0x1000.
  */
 static void subtest_dist_rdist(struct vm_gic *v)
 {
 	int ret;
 	uint64_t addr;
+	struct vgic_region_attr rdist; /* CPU interface in GICv2*/
+	struct vgic_region_attr dist;
+
+	rdist = VGIC_DEV_IS_V3(v->gic_dev_type) ? gic_v3_redist_region
+						: gic_v2_cpu_region;
+	dist = VGIC_DEV_IS_V3(v->gic_dev_type) ? gic_v3_dist_region
+						: gic_v2_dist_region;
 
 	/* Check existing group/attributes */
 	kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-			      KVM_VGIC_V3_ADDR_TYPE_DIST);
+			      dist.attr);
 
 	kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-			      KVM_VGIC_V3_ADDR_TYPE_REDIST);
+			      rdist.attr);
 
 	/* check non existing attribute */
-	ret = _kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR, 0);
+	ret = _kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR, -1);
 	TEST_ASSERT(ret && errno == ENXIO, "attribute not supported");
 
 	/* misaligned DIST and REDIST address settings */
-	addr = 0x1000;
+	addr = dist.alignment / 0x10;
 	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-				 KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
-	TEST_ASSERT(ret && errno == EINVAL, "GICv3 dist base not 64kB aligned");
+				 dist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "GIC dist base not aligned");
 
+	addr = rdist.alignment / 0x10;
 	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
-	TEST_ASSERT(ret && errno == EINVAL, "GICv3 redist base not 64kB aligned");
+				 rdist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "GIC redist/cpu base not aligned");
 
 	/* out of range address */
-	if (max_ipa_bits) {
-		addr = 1ULL << max_ipa_bits;
-		ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-					 KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
-		TEST_ASSERT(ret && errno == E2BIG, "dist address beyond IPA limit");
+	addr = max_phys_size;
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 dist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG, "dist address beyond IPA limit");
 
-		ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-					 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
-		TEST_ASSERT(ret && errno == E2BIG, "redist address beyond IPA limit");
-	}
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 rdist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG, "redist address beyond IPA limit");
+
+	/* Space for half a rdist (a rdist is: 2 * rdist.alignment). */
+	addr = max_phys_size - dist.alignment;
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 rdist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG,
+			"half of the redist is beyond IPA limit");
 
 	/* set REDIST base address @0x0*/
 	addr = 0x00000;
 	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-			  KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+			  rdist.attr, &addr, true);
 
 	/* Attempt to create a second legacy redistributor region */
 	addr = 0xE0000;
 	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
-	TEST_ASSERT(ret && errno == EEXIST, "GICv3 redist base set again");
+				 rdist.attr, &addr, true);
+	TEST_ASSERT(ret && errno == EEXIST, "GIC redist base set again");
 
-	/* Attempt to mix legacy and new redistributor regions */
-	addr = REDIST_REGION_ATTR_ADDR(NR_VCPUS, 0x100000, 0, 0);
-	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
-				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
-	TEST_ASSERT(ret && errno == EINVAL, "attempt to mix GICv3 REDIST and REDIST_REGION");
+	ret = _kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				     KVM_VGIC_V3_ADDR_TYPE_REDIST);
+	if (!ret) {
+		/* Attempt to mix legacy and new redistributor regions */
+		addr = REDIST_REGION_ATTR_ADDR(NR_VCPUS, 0x100000, 0, 0);
+		ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+					 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION,
+					 &addr, true);
+		TEST_ASSERT(ret && errno == EINVAL,
+			    "attempt to mix GICv3 REDIST and REDIST_REGION");
+	}
 
 	/*
 	 * Set overlapping DIST / REDIST, cannot be detected here. Will be detected
 	 * on first vcpu run instead.
 	 */
-	addr = 3 * 2 * 0x10000;
-	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR, KVM_VGIC_V3_ADDR_TYPE_DIST,
-			  &addr, true);
+	addr = rdist.size - rdist.alignment;
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  dist.attr, &addr, true);
 }
 
 /* Test the new REDIST region API */
-static void subtest_redist_regions(struct vm_gic *v)
+static void subtest_v3_redist_regions(struct vm_gic *v)
 {
 	uint64_t addr, expected_addr;
 	int ret;
···
 	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 
-	addr = REDIST_REGION_ATTR_ADDR(1, 1ULL << max_ipa_bits, 0, 2);
+	addr = REDIST_REGION_ATTR_ADDR(1, max_phys_size, 0, 2);
 	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 	TEST_ASSERT(ret && errno == E2BIG,
 		    "register redist region with base address beyond IPA range");
+
+	/* The last redist is above the pa range. */
+	addr = REDIST_REGION_ATTR_ADDR(2, max_phys_size - 0x30000, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG,
+		    "register redist region with top address beyond IPA range");
 
 	addr = 0x260000;
 	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
···
  * VGIC KVM device is created and initialized before the secondary CPUs
  * get created
  */
-static void test_vgic_then_vcpus(void)
+static void test_vgic_then_vcpus(uint32_t gic_dev_type)
 {
 	struct vm_gic v;
 	int ret, i;
 
-	v.vm = vm_create_default(0, 0, guest_code);
-	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+	v = vm_gic_create_with_vcpus(gic_dev_type, 1);
 
 	subtest_dist_rdist(&v);
 
···
 }
 
 /* All the VCPUs are created before the VGIC KVM device gets initialized */
-static void test_vcpus_then_vgic(void)
+static void test_vcpus_then_vgic(uint32_t gic_dev_type)
 {
 	struct vm_gic v;
 	int ret;
 
-	v = vm_gic_create();
+	v = vm_gic_create_with_vcpus(gic_dev_type, NR_VCPUS);
 
 	subtest_dist_rdist(&v);
 
···
 	vm_gic_destroy(&v);
 }
 
-static void test_new_redist_regions(void)
+static void test_v3_new_redist_regions(void)
 {
 	void *dummy = NULL;
 	struct vm_gic v;
 	uint64_t addr;
 	int ret;
 
-	v = vm_gic_create();
-	subtest_redist_regions(&v);
+	v = vm_gic_create_with_vcpus(KVM_DEV_TYPE_ARM_VGIC_V3, NR_VCPUS);
+	subtest_v3_redist_regions(&v);
 	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
 			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
 
···
 
 	/* step2 */
 
-	v = vm_gic_create();
-	subtest_redist_regions(&v);
+	v = vm_gic_create_with_vcpus(KVM_DEV_TYPE_ARM_VGIC_V3, NR_VCPUS);
+	subtest_v3_redist_regions(&v);
 
 	addr = REDIST_REGION_ATTR_ADDR(1, 0x280000, 0, 2);
 	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
···
 
 	/* step 3 */
 
-	v = vm_gic_create();
-	subtest_redist_regions(&v);
+	v = vm_gic_create_with_vcpus(KVM_DEV_TYPE_ARM_VGIC_V3, NR_VCPUS);
+	subtest_v3_redist_regions(&v);
 
 	_kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, dummy, true);
···
 	vm_gic_destroy(&v);
 }
 
-static void test_typer_accesses(void)
+static void test_v3_typer_accesses(void)
 {
 	struct vm_gic v;
 	uint64_t addr;
···
 
 	vm_vcpu_add_default(v.vm, 3, guest_code);
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(ret && errno == EINVAL, "attempting to read GICR_TYPER of non created vcpu");
 
 	vm_vcpu_add_default(v.vm, 1, guest_code);
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(ret && errno == EBUSY, "read GICR_TYPER before GIC initialized");
 
 	vm_vcpu_add_default(v.vm, 2, guest_code);
···
 			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
 
 	for (i = 0; i < NR_VCPUS ; i++) {
-		ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+		ret = access_v3_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
 		TEST_ASSERT(!ret && !val, "read GICR_TYPER before rdist region setting");
 	}
 
···
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 
 	/* The 2 first rdists should be put there (vcpu 0 and 3) */
-	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && !val, "read typer of rdist #0");
 
-	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x310, "read typer of rdist #1");
 
 	addr = REDIST_REGION_ATTR_ADDR(10, 0x100000, 0, 1);
···
 				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 	TEST_ASSERT(ret && errno == EINVAL, "collision with previous rdist region");
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x100,
 		    "no redist region attached to vcpu #1 yet, last cannot be returned");
 
-	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x200,
 		    "no redist region attached to vcpu #2, last cannot be returned");
 
···
 	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #1");
 
-	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x210,
 		    "read typer of rdist #1, last properly returned");
 
···
  * rdist region #2 @0x200000 2 rdist capacity
  *     rdists: 1, 2
  */
-static void test_last_bit_redist_regions(void)
+static void test_v3_last_bit_redist_regions(void)
 {
 	uint32_t vcpuids[] = { 0, 3, 5, 4, 1, 2 };
 	struct vm_gic v;
···
 	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
 
-	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x000, "read typer of rdist #0");
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #1");
 
-	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x200, "read typer of rdist #2");
 
-	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x310, "read typer of rdist #3");
 
-	ret = access_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x500, "read typer of rdist #5");
 
-	ret = access_redist_reg(v.gic_fd, 4, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 4, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x410, "read typer of rdist #4");
 
 	vm_gic_destroy(&v);
 }
 
 /* Test last bit with legacy region */
-static void test_last_bit_single_rdist(void)
+static void test_v3_last_bit_single_rdist(void)
 {
 	uint32_t vcpuids[] = { 0, 3, 5, 4, 1, 2 };
 	struct vm_gic v;
···
 	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
 			  KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
 
-	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x000, "read typer of rdist #0");
 
-	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x300, "read typer of rdist #1");
 
-	ret = access_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x500, "read typer of rdist #2");
 
-	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #3");
 
-	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	ret = access_v3_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
 	TEST_ASSERT(!ret && val == 0x210, "read typer of rdist #3");
 
 	vm_gic_destroy(&v);
 }
 
-void test_kvm_device(void)
+/* Uses the legacy REDIST region API. */
+static void test_v3_redist_ipa_range_check_at_vcpu_run(void)
+{
+	struct vm_gic v;
+	int ret, i;
+	uint64_t addr;
+
+	v = vm_gic_create_with_vcpus(KVM_DEV_TYPE_ARM_VGIC_V3, 1);
+
+	/* Set space for 3 redists, we have 1 vcpu, so this succeeds. */
+	addr = max_phys_size - (3 * 2 * 0x10000);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+
+	addr = 0x00000;
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
+
+	/* Add the rest of the VCPUs */
+	for (i = 1; i < NR_VCPUS; ++i)
+		vm_vcpu_add_default(v.vm, i, guest_code);
+
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	/* Attempt to run a vcpu without enough redist space. */
+	ret = run_vcpu(v.vm, 2);
+	TEST_ASSERT(ret && errno == EINVAL,
+		"redist base+size above PA range detected on 1st vcpu run");
+
+	vm_gic_destroy(&v);
+}
+
+static void test_v3_its_region(void)
+{
+	struct vm_gic v;
+	uint64_t addr;
+	int its_fd, ret;
+
+	v = vm_gic_create_with_vcpus(KVM_DEV_TYPE_ARM_VGIC_V3, NR_VCPUS);
+	its_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_ITS, false);
+
+	addr = 0x401000;
+	ret = _kvm_device_access(its_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_ITS_ADDR_TYPE, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL,
+		"ITS region with misaligned address");
+
+	addr = max_phys_size;
+	ret = _kvm_device_access(its_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_ITS_ADDR_TYPE, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG,
+		"register ITS region with base address beyond IPA range");
+
+	addr = max_phys_size - 0x10000;
+	ret = _kvm_device_access(its_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_ITS_ADDR_TYPE, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG,
+		"Half of ITS region is beyond IPA range");
+
+	/* This one succeeds setting the ITS base */
+	addr = 0x400000;
+	kvm_device_access(its_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_ITS_ADDR_TYPE, &addr, true);
+
+	addr = 0x300000;
+	ret = _kvm_device_access(its_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_ITS_ADDR_TYPE, &addr, true);
+	TEST_ASSERT(ret && errno == EEXIST, "ITS base set again");
+
+	close(its_fd);
+	vm_gic_destroy(&v);
+}
+
+/*
+ * Returns 0 if it's possible to create GIC device of a given type (V2 or V3).
+ */
+int test_kvm_device(uint32_t gic_dev_type)
 {
 	struct vm_gic v;
 	int ret, fd;
+	uint32_t other;
 
 	v.vm = vm_create_default_with_vcpus(NR_VCPUS, 0, 0, guest_code, NULL);
 
···
 	ret = _kvm_create_device(v.vm, 0, true, &fd);
 	TEST_ASSERT(ret && errno == ENODEV, "unsupported device");
 
-	/* trial mode with VGIC_V3 device */
-	ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, true, &fd);
-	if (ret) {
-		print_skip("GICv3 not supported");
-		exit(KSFT_SKIP);
-	}
-	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+	/* trial mode */
+	ret = _kvm_create_device(v.vm, gic_dev_type, true, &fd);
+	if (ret)
+		return ret;
+	v.gic_fd = kvm_create_device(v.vm, gic_dev_type, false);
 
-	ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false, &fd);
-	TEST_ASSERT(ret && errno == EEXIST, "create GICv3 device twice");
+	ret = _kvm_create_device(v.vm, gic_dev_type, false, &fd);
+	TEST_ASSERT(ret && errno == EEXIST, "create GIC device twice");
 
-	kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, true);
+	kvm_create_device(v.vm, gic_dev_type, true);
 
-	if (!_kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V2, true, &fd)) {
-		ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V2, false, &fd);
-		TEST_ASSERT(ret && errno == EINVAL, "create GICv2 while v3 exists");
+	/* try to create the other gic_dev_type */
+	other = VGIC_DEV_IS_V2(gic_dev_type) ? KVM_DEV_TYPE_ARM_VGIC_V3
+					     : KVM_DEV_TYPE_ARM_VGIC_V2;
+
+	if (!_kvm_create_device(v.vm, other, true, &fd)) {
+		ret = _kvm_create_device(v.vm, other, false, &fd);
+		TEST_ASSERT(ret && errno == EINVAL,
+				"create GIC device while other version exists");
 	}
 
 	vm_gic_destroy(&v);
+
+	return 0;
+}
+
+void run_tests(uint32_t gic_dev_type)
+{
+	test_vcpus_then_vgic(gic_dev_type);
+	test_vgic_then_vcpus(gic_dev_type);
+
+	if (VGIC_DEV_IS_V3(gic_dev_type)) {
+		test_v3_new_redist_regions();
+		test_v3_typer_accesses();
+		test_v3_last_bit_redist_regions();
+		test_v3_last_bit_single_rdist();
+		test_v3_redist_ipa_range_check_at_vcpu_run();
+		test_v3_its_region();
+	}
 }
 
 int main(int ac, char **av)
 {
-	max_ipa_bits = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
+	int ret;
+	int pa_bits;
 
-	test_kvm_device();
-	test_vcpus_then_vgic();
-	test_vgic_then_vcpus();
-	test_new_redist_regions();
-	test_typer_accesses();
-	test_last_bit_redist_regions();
-	test_last_bit_single_rdist();
+	pa_bits = vm_guest_mode_params[VM_MODE_DEFAULT].pa_bits;
+	max_phys_size = 1ULL << pa_bits;
 
+	ret = test_kvm_device(KVM_DEV_TYPE_ARM_VGIC_V3);
+	if (!ret) {
+		pr_info("Running GIC_v3 tests.\n");
+		run_tests(KVM_DEV_TYPE_ARM_VGIC_V3);
+		return 0;
+	}
+
+	ret = test_kvm_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+	if (!ret) {
+		pr_info("Running GIC_v2 tests.\n");
+		run_tests(KVM_DEV_TYPE_ARM_VGIC_V2);
+		return 0;
+	}
+
+	print_skip("No GICv2 nor GICv3 support");
+	exit(KSFT_SKIP);
 	return 0;
 }

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

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM Generic Timer specific interface
+ */
+
+#ifndef SELFTEST_KVM_ARCH_TIMER_H
+#define SELFTEST_KVM_ARCH_TIMER_H
+
+#include "processor.h"
+
+enum arch_timer {
+	VIRTUAL,
+	PHYSICAL,
+};
+
+#define CTL_ENABLE	(1 << 0)
+#define CTL_IMASK	(1 << 1)
+#define CTL_ISTATUS	(1 << 2)
+
+#define msec_to_cycles(msec)	\
+	(timer_get_cntfrq() * (uint64_t)(msec) / 1000)
+
+#define usec_to_cycles(usec)	\
+	(timer_get_cntfrq() * (uint64_t)(usec) / 1000000)
+
+#define cycles_to_usec(cycles) \
+	((uint64_t)(cycles) * 1000000 / timer_get_cntfrq())
+
+static inline uint32_t timer_get_cntfrq(void)
+{
+	return read_sysreg(cntfrq_el0);
+}
+
+static inline uint64_t timer_get_cntct(enum arch_timer timer)
+{
+	isb();
+
+	switch (timer) {
+	case VIRTUAL:
+		return read_sysreg(cntvct_el0);
+	case PHYSICAL:
+		return read_sysreg(cntpct_el0);
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	/* We should not reach here */
+	return 0;
+}
+
+static inline void timer_set_cval(enum arch_timer timer, uint64_t cval)
+{
+	switch (timer) {
+	case VIRTUAL:
+		write_sysreg(cval, cntv_cval_el0);
+		break;
+	case PHYSICAL:
+		write_sysreg(cval, cntp_cval_el0);
+		break;
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	isb();
+}
+
+static inline uint64_t timer_get_cval(enum arch_timer timer)
+{
+	switch (timer) {
+	case VIRTUAL:
+		return read_sysreg(cntv_cval_el0);
+	case PHYSICAL:
+		return read_sysreg(cntp_cval_el0);
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	/* We should not reach here */
+	return 0;
+}
+
+static inline void timer_set_tval(enum arch_timer timer, uint32_t tval)
+{
+	switch (timer) {
+	case VIRTUAL:
+		write_sysreg(tval, cntv_tval_el0);
+		break;
+	case PHYSICAL:
+		write_sysreg(tval, cntp_tval_el0);
+		break;
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	isb();
+}
+
+static inline void timer_set_ctl(enum arch_timer timer, uint32_t ctl)
+{
+	switch (timer) {
+	case VIRTUAL:
+		write_sysreg(ctl, cntv_ctl_el0);
+		break;
+	case PHYSICAL:
+		write_sysreg(ctl, cntp_ctl_el0);
+		break;
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	isb();
+}
+
+static inline uint32_t timer_get_ctl(enum arch_timer timer)
+{
+	switch (timer) {
+	case VIRTUAL:
+		return read_sysreg(cntv_ctl_el0);
+	case PHYSICAL:
+		return read_sysreg(cntp_ctl_el0);
+	default:
+		GUEST_ASSERT_1(0, timer);
+	}
+
+	/* We should not reach here */
+	return 0;
+}
+
+static inline void timer_set_next_cval_ms(enum arch_timer timer, uint32_t msec)
+{
+	uint64_t now_ct = timer_get_cntct(timer);
+	uint64_t next_ct = now_ct + msec_to_cycles(msec);
+
+	timer_set_cval(timer, next_ct);
+}
+
+static inline void timer_set_next_tval_ms(enum arch_timer timer, uint32_t msec)
+{
+	timer_set_tval(timer, msec_to_cycles(msec));
+}
+
+#endif /* SELFTEST_KVM_ARCH_TIMER_H */

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

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM simple delay routines
+ */
+
+#ifndef SELFTEST_KVM_ARM_DELAY_H
+#define SELFTEST_KVM_ARM_DELAY_H
+
+#include "arch_timer.h"
+
+static inline void __delay(uint64_t cycles)
+{
+	enum arch_timer timer = VIRTUAL;
+	uint64_t start = timer_get_cntct(timer);
+
+	while ((timer_get_cntct(timer) - start) < cycles)
+		cpu_relax();
+}
+
+static inline void udelay(unsigned long usec)
+{
+	__delay(usec_to_cycles(usec));
+}
+
+#endif /* SELFTEST_KVM_ARM_DELAY_H */

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

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM Generic Interrupt Controller (GIC) specific defines
+ */
+
+#ifndef SELFTEST_KVM_GIC_H
+#define SELFTEST_KVM_GIC_H
+
+enum gic_type {
+	GIC_V3,
+	GIC_TYPE_MAX,
+};
+
+void gic_init(enum gic_type type, unsigned int nr_cpus,
+		void *dist_base, void *redist_base);
+void gic_irq_enable(unsigned int intid);
+void gic_irq_disable(unsigned int intid);
+unsigned int gic_get_and_ack_irq(void);
+void gic_set_eoi(unsigned int intid);
+
+#endif /* SELFTEST_KVM_GIC_H */

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

···
 
 #include "kvm_util.h"
 #include <linux/stringify.h>
+#include <linux/types.h>
+#include <asm/sysreg.h>
 
 
 #define ARM64_CORE_REG(x) (KVM_REG_ARM64 | KVM_REG_SIZE_U64 | \
 			   KVM_REG_ARM_CORE | KVM_REG_ARM_CORE_REG(x))
 
-#define CPACR_EL1               3, 0,  1, 0, 2
-#define TCR_EL1                 3, 0,  2, 0, 2
-#define MAIR_EL1                3, 0, 10, 2, 0
-#define MPIDR_EL1               3, 0,  0, 0, 5
-#define TTBR0_EL1               3, 0,  2, 0, 0
-#define SCTLR_EL1               3, 0,  1, 0, 0
-#define VBAR_EL1                3, 0, 12, 0, 0
-
-#define ID_AA64DFR0_EL1         3, 0,  0, 5, 0
+/*
+ * KVM_ARM64_SYS_REG(sys_reg_id): Helper macro to convert
+ * SYS_* register definitions in asm/sysreg.h to use in KVM
+ * calls such as get_reg() and set_reg().
+ */
+#define KVM_ARM64_SYS_REG(sys_reg_id)			\
+	ARM64_SYS_REG(sys_reg_Op0(sys_reg_id),		\
+			sys_reg_Op1(sys_reg_id),	\
+			sys_reg_CRn(sys_reg_id),	\
+			sys_reg_CRm(sys_reg_id),	\
+			sys_reg_Op2(sys_reg_id))
 
 /*
  * Default MAIR
···
 	vcpu_ioctl(vm, vcpuid, KVM_SET_ONE_REG, &reg);
 }
 
-void aarch64_vcpu_setup(struct kvm_vm *vm, int vcpuid, struct kvm_vcpu_init *init);
+void aarch64_vcpu_setup(struct kvm_vm *vm, uint32_t vcpuid, struct kvm_vcpu_init *init);
 void aarch64_vcpu_add_default(struct kvm_vm *vm, uint32_t vcpuid,
 			      struct kvm_vcpu_init *init, void *guest_code);
 
···
 void vm_install_sync_handler(struct kvm_vm *vm,
 		int vector, int ec, handler_fn handler);
 
-#define write_sysreg(reg, val)						  \
-({									  \
-	u64 __val = (u64)(val);						  \
-	asm volatile("msr " __stringify(reg) ", %x0" : : "rZ" (__val));	  \
+static inline void cpu_relax(void)
+{
+	asm volatile("yield" ::: "memory");
+}
+
+#define isb()		asm volatile("isb" : : : "memory")
+#define dsb(opt)	asm volatile("dsb " #opt : : : "memory")
+#define dmb(opt)	asm volatile("dmb " #opt : : : "memory")
+
+#define dma_wmb()	dmb(oshst)
+#define __iowmb()	dma_wmb()
+
+#define dma_rmb()	dmb(oshld)
+
+#define __iormb(v)							\
+({									\
+	unsigned long tmp;						\
+									\
+	dma_rmb();							\
+									\
+	/*								\
+	 * Courtesy of arch/arm64/include/asm/io.h:			\
+	 * Create a dummy control dependency from the IO read to any	\
+	 * later instructions. This ensures that a subsequent call	\
+	 * to udelay() will be ordered due to the ISB in __delay().	\
+	 */								\
+	asm volatile("eor	%0, %1, %1\n"				\
+		     "cbnz	%0, ."					\
+		     : "=r" (tmp) : "r" ((unsigned long)(v))		\
+		     : "memory");					\
 })
 
-#define read_sysreg(reg)						  \
-({	u64 val;							  \
-	asm volatile("mrs %0, "__stringify(reg) : "=r"(val) : : "memory");\
-	val;								  \
-})
+static __always_inline void __raw_writel(u32 val, volatile void *addr)
+{
+	asm volatile("str %w0, [%1]" : : "rZ" (val), "r" (addr));
+}
 
-#define isb()	asm volatile("isb" : : : "memory")
+static __always_inline u32 __raw_readl(const volatile void *addr)
+{
+	u32 val;
+	asm volatile("ldr %w0, [%1]" : "=r" (val) : "r" (addr));
+	return val;
+}
+
+#define writel_relaxed(v,c)	((void)__raw_writel((__force u32)cpu_to_le32(v),(c)))
+#define readl_relaxed(c)	({ u32 __r = le32_to_cpu((__force __le32)__raw_readl(c)); __r; })
+
+#define writel(v,c)		({ __iowmb(); writel_relaxed((v),(c));})
+#define readl(c)		({ u32 __v = readl_relaxed(c); __iormb(__v); __v; })
+
+static inline void local_irq_enable(void)
+{
+	asm volatile("msr daifclr, #3" : : : "memory");
+}
+
+static inline void local_irq_disable(void)
+{
+	asm volatile("msr daifset, #3" : : : "memory");
+}
 
 #endif /* SELFTEST_KVM_PROCESSOR_H */

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

···
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef SELFTEST_KVM_ARM64_SPINLOCK_H
+#define SELFTEST_KVM_ARM64_SPINLOCK_H
+
+struct spinlock {
+	int v;
+};
+
+extern void spin_lock(struct spinlock *lock);
+extern void spin_unlock(struct spinlock *lock);
+
+#endif /* SELFTEST_KVM_ARM64_SPINLOCK_H */

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

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM Generic Interrupt Controller (GIC) host specific defines
+ */
+
+#ifndef SELFTEST_KVM_VGIC_H
+#define SELFTEST_KVM_VGIC_H
+
+#include <linux/kvm.h>
+
+#define REDIST_REGION_ATTR_ADDR(count, base, flags, index) \
+	(((uint64_t)(count) << 52) | \
+	((uint64_t)((base) >> 16) << 16) | \
+	((uint64_t)(flags) << 12) | \
+	index)
+
+int vgic_v3_setup(struct kvm_vm *vm, unsigned int nr_vcpus,
+		uint64_t gicd_base_gpa, uint64_t gicr_base_gpa);
+
+#endif /* SELFTEST_KVM_VGIC_H */

tools/testing/selftests/kvm/include/kvm_util.h

···
 #define KVM_DEV_PATH "/dev/kvm"
 #define KVM_MAX_VCPUS 512
 
+#define NSEC_PER_SEC 1000000000L
+
 /*
  * Callers of kvm_util only have an incomplete/opaque description of the
  * structure kvm_util is using to maintain the state of a VM.
···
 int kvm_device_access(int dev_fd, uint32_t group, uint64_t attr,
 		      void *val, bool write);
 
+int _vcpu_has_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			  uint64_t attr);
+int vcpu_has_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			 uint64_t attr);
+int _vcpu_access_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			  uint64_t attr, void *val, bool write);
+int vcpu_access_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			 uint64_t attr, void *val, bool write);
+
 const char *exit_reason_str(unsigned int exit_reason);
 
 void virt_pgd_alloc(struct kvm_vm *vm);
···
 
 int vm_get_stats_fd(struct kvm_vm *vm);
 int vcpu_get_stats_fd(struct kvm_vm *vm, uint32_t vcpuid);
+
+uint32_t guest_get_vcpuid(void);
 
 #endif /* SELFTEST_KVM_UTIL_H */

tools/testing/selftests/kvm/kvm_create_max_vcpus.c

···
 		kvm_max_vcpu_id = kvm_max_vcpus;
 
 	TEST_ASSERT(kvm_max_vcpu_id >= kvm_max_vcpus,
-		    "KVM_MAX_VCPU_ID (%d) must be at least as large as KVM_MAX_VCPUS (%d).",
+		    "KVM_MAX_VCPU_IDS (%d) must be at least as large as KVM_MAX_VCPUS (%d).",
 		    kvm_max_vcpu_id, kvm_max_vcpus);
 
 	test_vcpu_creation(0, kvm_max_vcpus);

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

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ARM Generic Interrupt Controller (GIC) support
+ */
+
+#include <errno.h>
+#include <linux/bits.h>
+#include <linux/sizes.h>
+
+#include "kvm_util.h"
+
+#include <gic.h>
+#include "gic_private.h"
+#include "processor.h"
+#include "spinlock.h"
+
+static const struct gic_common_ops *gic_common_ops;
+static struct spinlock gic_lock;
+
+static void gic_cpu_init(unsigned int cpu, void *redist_base)
+{
+	gic_common_ops->gic_cpu_init(cpu, redist_base);
+}
+
+static void
+gic_dist_init(enum gic_type type, unsigned int nr_cpus, void *dist_base)
+{
+	const struct gic_common_ops *gic_ops = NULL;
+
+	spin_lock(&gic_lock);
+
+	/* Distributor initialization is needed only once per VM */
+	if (gic_common_ops) {
+		spin_unlock(&gic_lock);
+		return;
+	}
+
+	if (type == GIC_V3)
+		gic_ops = &gicv3_ops;
+
+	GUEST_ASSERT(gic_ops);
+
+	gic_ops->gic_init(nr_cpus, dist_base);
+	gic_common_ops = gic_ops;
+
+	/* Make sure that the initialized data is visible to all the vCPUs */
+	dsb(sy);
+
+	spin_unlock(&gic_lock);
+}
+
+void gic_init(enum gic_type type, unsigned int nr_cpus,
+		void *dist_base, void *redist_base)
+{
+	uint32_t cpu = guest_get_vcpuid();
+
+	GUEST_ASSERT(type < GIC_TYPE_MAX);
+	GUEST_ASSERT(dist_base);
+	GUEST_ASSERT(redist_base);
+	GUEST_ASSERT(nr_cpus);
+
+	gic_dist_init(type, nr_cpus, dist_base);
+	gic_cpu_init(cpu, redist_base);
+}
+
+void gic_irq_enable(unsigned int intid)
+{
+	GUEST_ASSERT(gic_common_ops);
+	gic_common_ops->gic_irq_enable(intid);
+}
+
+void gic_irq_disable(unsigned int intid)
+{
+	GUEST_ASSERT(gic_common_ops);
+	gic_common_ops->gic_irq_disable(intid);
+}
+
+unsigned int gic_get_and_ack_irq(void)
+{
+	uint64_t irqstat;
+	unsigned int intid;
+
+	GUEST_ASSERT(gic_common_ops);
+
+	irqstat = gic_common_ops->gic_read_iar();
+	intid = irqstat & GENMASK(23, 0);
+
+	return intid;
+}
+
+void gic_set_eoi(unsigned int intid)
+{
+	GUEST_ASSERT(gic_common_ops);
+	gic_common_ops->gic_write_eoir(intid);
+}

tools/testing/selftests/kvm/lib/aarch64/gic_private.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM Generic Interrupt Controller (GIC) private defines that's only
+ * shared among the GIC library code.
+ */
+
+#ifndef SELFTEST_KVM_GIC_PRIVATE_H
+#define SELFTEST_KVM_GIC_PRIVATE_H
+
+struct gic_common_ops {
+	void (*gic_init)(unsigned int nr_cpus, void *dist_base);
+	void (*gic_cpu_init)(unsigned int cpu, void *redist_base);
+	void (*gic_irq_enable)(unsigned int intid);
+	void (*gic_irq_disable)(unsigned int intid);
+	uint64_t (*gic_read_iar)(void);
+	void (*gic_write_eoir)(uint32_t irq);
+};
+
+extern const struct gic_common_ops gicv3_ops;
+
+#endif /* SELFTEST_KVM_GIC_PRIVATE_H */

tools/testing/selftests/kvm/lib/aarch64/gic_v3.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ARM Generic Interrupt Controller (GIC) v3 support
+ */
+
+#include <linux/sizes.h>
+
+#include "kvm_util.h"
+#include "processor.h"
+#include "delay.h"
+
+#include "gic_v3.h"
+#include "gic_private.h"
+
+struct gicv3_data {
+	void *dist_base;
+	void *redist_base[GICV3_MAX_CPUS];
+	unsigned int nr_cpus;
+	unsigned int nr_spis;
+};
+
+#define sgi_base_from_redist(redist_base) (redist_base + SZ_64K)
+
+enum gicv3_intid_range {
+	SGI_RANGE,
+	PPI_RANGE,
+	SPI_RANGE,
+	INVALID_RANGE,
+};
+
+static struct gicv3_data gicv3_data;
+
+static void gicv3_gicd_wait_for_rwp(void)
+{
+	unsigned int count = 100000; /* 1s */
+
+	while (readl(gicv3_data.dist_base + GICD_CTLR) & GICD_CTLR_RWP) {
+		GUEST_ASSERT(count--);
+		udelay(10);
+	}
+}
+
+static void gicv3_gicr_wait_for_rwp(void *redist_base)
+{
+	unsigned int count = 100000; /* 1s */
+
+	while (readl(redist_base + GICR_CTLR) & GICR_CTLR_RWP) {
+		GUEST_ASSERT(count--);
+		udelay(10);
+	}
+}
+
+static enum gicv3_intid_range get_intid_range(unsigned int intid)
+{
+	switch (intid) {
+	case 0 ... 15:
+		return SGI_RANGE;
+	case 16 ... 31:
+		return PPI_RANGE;
+	case 32 ... 1019:
+		return SPI_RANGE;
+	}
+
+	/* We should not be reaching here */
+	GUEST_ASSERT(0);
+
+	return INVALID_RANGE;
+}
+
+static uint64_t gicv3_read_iar(void)
+{
+	uint64_t irqstat = read_sysreg_s(SYS_ICC_IAR1_EL1);
+
+	dsb(sy);
+	return irqstat;
+}
+
+static void gicv3_write_eoir(uint32_t irq)
+{
+	write_sysreg_s(irq, SYS_ICC_EOIR1_EL1);
+	isb();
+}
+
+static void
+gicv3_config_irq(unsigned int intid, unsigned int offset)
+{
+	uint32_t cpu = guest_get_vcpuid();
+	uint32_t mask = 1 << (intid % 32);
+	enum gicv3_intid_range intid_range = get_intid_range(intid);
+	void *reg;
+
+	/* We care about 'cpu' only for SGIs or PPIs */
+	if (intid_range == SGI_RANGE || intid_range == PPI_RANGE) {
+		GUEST_ASSERT(cpu < gicv3_data.nr_cpus);
+
+		reg = sgi_base_from_redist(gicv3_data.redist_base[cpu]) +
+			offset;
+		writel(mask, reg);
+		gicv3_gicr_wait_for_rwp(gicv3_data.redist_base[cpu]);
+	} else if (intid_range == SPI_RANGE) {
+		reg = gicv3_data.dist_base + offset + (intid / 32) * 4;
+		writel(mask, reg);
+		gicv3_gicd_wait_for_rwp();
+	} else {
+		GUEST_ASSERT(0);
+	}
+}
+
+static void gicv3_irq_enable(unsigned int intid)
+{
+	gicv3_config_irq(intid, GICD_ISENABLER);
+}
+
+static void gicv3_irq_disable(unsigned int intid)
+{
+	gicv3_config_irq(intid, GICD_ICENABLER);
+}
+
+static void gicv3_enable_redist(void *redist_base)
+{
+	uint32_t val = readl(redist_base + GICR_WAKER);
+	unsigned int count = 100000; /* 1s */
+
+	val &= ~GICR_WAKER_ProcessorSleep;
+	writel(val, redist_base + GICR_WAKER);
+
+	/* Wait until the processor is 'active' */
+	while (readl(redist_base + GICR_WAKER) & GICR_WAKER_ChildrenAsleep) {
+		GUEST_ASSERT(count--);
+		udelay(10);
+	}
+}
+
+static inline void *gicr_base_cpu(void *redist_base, uint32_t cpu)
+{
+	/* Align all the redistributors sequentially */
+	return redist_base + cpu * SZ_64K * 2;
+}
+
+static void gicv3_cpu_init(unsigned int cpu, void *redist_base)
+{
+	void *sgi_base;
+	unsigned int i;
+	void *redist_base_cpu;
+
+	GUEST_ASSERT(cpu < gicv3_data.nr_cpus);
+
+	redist_base_cpu = gicr_base_cpu(redist_base, cpu);
+	sgi_base = sgi_base_from_redist(redist_base_cpu);
+
+	gicv3_enable_redist(redist_base_cpu);
+
+	/*
+	 * Mark all the SGI and PPI interrupts as non-secure Group-1.
+	 * Also, deactivate and disable them.
+	 */
+	writel(~0, sgi_base + GICR_IGROUPR0);
+	writel(~0, sgi_base + GICR_ICACTIVER0);
+	writel(~0, sgi_base + GICR_ICENABLER0);
+
+	/* Set a default priority for all the SGIs and PPIs */
+	for (i = 0; i < 32; i += 4)
+		writel(GICD_INT_DEF_PRI_X4,
+				sgi_base + GICR_IPRIORITYR0 + i);
+
+	gicv3_gicr_wait_for_rwp(redist_base_cpu);
+
+	/* Enable the GIC system register (ICC_*) access */
+	write_sysreg_s(read_sysreg_s(SYS_ICC_SRE_EL1) | ICC_SRE_EL1_SRE,
+			SYS_ICC_SRE_EL1);
+
+	/* Set a default priority threshold */
+	write_sysreg_s(ICC_PMR_DEF_PRIO, SYS_ICC_PMR_EL1);
+
+	/* Enable non-secure Group-1 interrupts */
+	write_sysreg_s(ICC_IGRPEN1_EL1_ENABLE, SYS_ICC_GRPEN1_EL1);
+
+	gicv3_data.redist_base[cpu] = redist_base_cpu;
+}
+
+static void gicv3_dist_init(void)
+{
+	void *dist_base = gicv3_data.dist_base;
+	unsigned int i;
+
+	/* Disable the distributor until we set things up */
+	writel(0, dist_base + GICD_CTLR);
+	gicv3_gicd_wait_for_rwp();
+
+	/*
+	 * Mark all the SPI interrupts as non-secure Group-1.
+	 * Also, deactivate and disable them.
+	 */
+	for (i = 32; i < gicv3_data.nr_spis; i += 32) {
+		writel(~0, dist_base + GICD_IGROUPR + i / 8);
+		writel(~0, dist_base + GICD_ICACTIVER + i / 8);
+		writel(~0, dist_base + GICD_ICENABLER + i / 8);
+	}
+
+	/* Set a default priority for all the SPIs */
+	for (i = 32; i < gicv3_data.nr_spis; i += 4)
+		writel(GICD_INT_DEF_PRI_X4,
+				dist_base + GICD_IPRIORITYR + i);
+
+	/* Wait for the settings to sync-in */
+	gicv3_gicd_wait_for_rwp();
+
+	/* Finally, enable the distributor globally with ARE */
+	writel(GICD_CTLR_ARE_NS | GICD_CTLR_ENABLE_G1A |
+			GICD_CTLR_ENABLE_G1, dist_base + GICD_CTLR);
+	gicv3_gicd_wait_for_rwp();
+}
+
+static void gicv3_init(unsigned int nr_cpus, void *dist_base)
+{
+	GUEST_ASSERT(nr_cpus <= GICV3_MAX_CPUS);
+
+	gicv3_data.nr_cpus = nr_cpus;
+	gicv3_data.dist_base = dist_base;
+	gicv3_data.nr_spis = GICD_TYPER_SPIS(
+				readl(gicv3_data.dist_base + GICD_TYPER));
+	if (gicv3_data.nr_spis > 1020)
+		gicv3_data.nr_spis = 1020;
+
+	/*
+	 * Initialize only the distributor for now.
+	 * The redistributor and CPU interfaces are initialized
+	 * later for every PE.
+	 */
+	gicv3_dist_init();
+}
+
+const struct gic_common_ops gicv3_ops = {
+	.gic_init = gicv3_init,
+	.gic_cpu_init = gicv3_cpu_init,
+	.gic_irq_enable = gicv3_irq_enable,
+	.gic_irq_disable = gicv3_irq_disable,
+	.gic_read_iar = gicv3_read_iar,
+	.gic_write_eoir = gicv3_write_eoir,
+};

tools/testing/selftests/kvm/lib/aarch64/gic_v3.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * ARM Generic Interrupt Controller (GIC) v3 specific defines
+ */
+
+#ifndef SELFTEST_KVM_GICV3_H
+#define SELFTEST_KVM_GICV3_H
+
+#include <asm/sysreg.h>
+
+/*
+ * Distributor registers
+ */
+#define GICD_CTLR			0x0000
+#define GICD_TYPER			0x0004
+#define GICD_IGROUPR			0x0080
+#define GICD_ISENABLER			0x0100
+#define GICD_ICENABLER			0x0180
+#define GICD_ICACTIVER			0x0380
+#define GICD_IPRIORITYR			0x0400
+
+/*
+ * The assumption is that the guest runs in a non-secure mode.
+ * The following bits of GICD_CTLR are defined accordingly.
+ */
+#define GICD_CTLR_RWP			(1U << 31)
+#define GICD_CTLR_nASSGIreq		(1U << 8)
+#define GICD_CTLR_ARE_NS		(1U << 4)
+#define GICD_CTLR_ENABLE_G1A		(1U << 1)
+#define GICD_CTLR_ENABLE_G1		(1U << 0)
+
+#define GICD_TYPER_SPIS(typer)		((((typer) & 0x1f) + 1) * 32)
+#define GICD_INT_DEF_PRI_X4		0xa0a0a0a0
+
+/*
+ * Redistributor registers
+ */
+#define GICR_CTLR			0x000
+#define GICR_WAKER			0x014
+
+#define GICR_CTLR_RWP			(1U << 3)
+
+#define GICR_WAKER_ProcessorSleep	(1U << 1)
+#define GICR_WAKER_ChildrenAsleep	(1U << 2)
+
+/*
+ * Redistributor registers, offsets from SGI base
+ */
+#define GICR_IGROUPR0			GICD_IGROUPR
+#define GICR_ISENABLER0			GICD_ISENABLER
+#define GICR_ICENABLER0			GICD_ICENABLER
+#define GICR_ICACTIVER0			GICD_ICACTIVER
+#define GICR_IPRIORITYR0		GICD_IPRIORITYR
+
+/* CPU interface registers */
+#define SYS_ICC_PMR_EL1			sys_reg(3, 0, 4, 6, 0)
+#define SYS_ICC_IAR1_EL1		sys_reg(3, 0, 12, 12, 0)
+#define SYS_ICC_EOIR1_EL1		sys_reg(3, 0, 12, 12, 1)
+#define SYS_ICC_SRE_EL1			sys_reg(3, 0, 12, 12, 5)
+#define SYS_ICC_GRPEN1_EL1		sys_reg(3, 0, 12, 12, 7)
+
+#define ICC_PMR_DEF_PRIO		0xf0
+
+#define ICC_SRE_EL1_SRE			(1U << 0)
+
+#define ICC_IGRPEN1_EL1_ENABLE		(1U << 0)
+
+#define GICV3_MAX_CPUS			512
+
+#endif /* SELFTEST_KVM_GICV3_H */

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

···
 	}
 }
 
-void aarch64_vcpu_setup(struct kvm_vm *vm, int vcpuid, struct kvm_vcpu_init *init)
+void aarch64_vcpu_setup(struct kvm_vm *vm, uint32_t vcpuid, struct kvm_vcpu_init *init)
 {
 	struct kvm_vcpu_init default_init = { .target = -1, };
 	uint64_t sctlr_el1, tcr_el1;
···
 	 * Enable FP/ASIMD to avoid trapping when accessing Q0-Q15
 	 * registers, which the variable argument list macros do.
 	 */
-	set_reg(vm, vcpuid, ARM64_SYS_REG(CPACR_EL1), 3 << 20);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_CPACR_EL1), 3 << 20);
 
-	get_reg(vm, vcpuid, ARM64_SYS_REG(SCTLR_EL1), &sctlr_el1);
-	get_reg(vm, vcpuid, ARM64_SYS_REG(TCR_EL1), &tcr_el1);
+	get_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_SCTLR_EL1), &sctlr_el1);
+	get_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_TCR_EL1), &tcr_el1);
 
 	switch (vm->mode) {
 	case VM_MODE_P52V48_4K:
···
 	tcr_el1 |= (1 << 8) | (1 << 10) | (3 << 12);
 	tcr_el1 |= (64 - vm->va_bits) /* T0SZ */;
 
-	set_reg(vm, vcpuid, ARM64_SYS_REG(SCTLR_EL1), sctlr_el1);
-	set_reg(vm, vcpuid, ARM64_SYS_REG(TCR_EL1), tcr_el1);
-	set_reg(vm, vcpuid, ARM64_SYS_REG(MAIR_EL1), DEFAULT_MAIR_EL1);
-	set_reg(vm, vcpuid, ARM64_SYS_REG(TTBR0_EL1), vm->pgd);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_SCTLR_EL1), sctlr_el1);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_TCR_EL1), tcr_el1);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_MAIR_EL1), DEFAULT_MAIR_EL1);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_TTBR0_EL1), vm->pgd);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_TPIDR_EL1), vcpuid);
 }
 
 void vcpu_dump(FILE *stream, struct kvm_vm *vm, uint32_t vcpuid, uint8_t indent)
···
 {
 	extern char vectors;
 
-	set_reg(vm, vcpuid, ARM64_SYS_REG(VBAR_EL1), (uint64_t)&vectors);
+	set_reg(vm, vcpuid, KVM_ARM64_SYS_REG(SYS_VBAR_EL1), (uint64_t)&vectors);
 }
 
 void route_exception(struct ex_regs *regs, int vector)
···
 	assert(!VECTOR_IS_SYNC(vector));
 	assert(vector < VECTOR_NUM);
 	handlers->exception_handlers[vector][0] = handler;
+}
+
+uint32_t guest_get_vcpuid(void)
+{
+	return read_sysreg(tpidr_el1);
 }

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

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ARM64 Spinlock support
+ */
+#include <stdint.h>
+
+#include "spinlock.h"
+
+void spin_lock(struct spinlock *lock)
+{
+	int val, res;
+
+	asm volatile(
+	"1:	ldaxr	%w0, [%2]\n"
+	"	cbnz	%w0, 1b\n"
+	"	mov	%w0, #1\n"
+	"	stxr	%w1, %w0, [%2]\n"
+	"	cbnz	%w1, 1b\n"
+	: "=&r" (val), "=&r" (res)
+	: "r" (&lock->v)
+	: "memory");
+}
+
+void spin_unlock(struct spinlock *lock)
+{
+	asm volatile("stlr wzr, [%0]\n"	: : "r" (&lock->v) : "memory");
+}

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

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ARM Generic Interrupt Controller (GIC) v3 host support
+ */
+
+#include <linux/kvm.h>
+#include <linux/sizes.h>
+#include <asm/kvm.h>
+
+#include "kvm_util.h"
+#include "../kvm_util_internal.h"
+#include "vgic.h"
+
+/*
+ * vGIC-v3 default host setup
+ *
+ * Input args:
+ *	vm - KVM VM
+ *	nr_vcpus - Number of vCPUs supported by this VM
+ *	gicd_base_gpa - Guest Physical Address of the Distributor region
+ *	gicr_base_gpa - Guest Physical Address of the Redistributor region
+ *
+ * Output args: None
+ *
+ * Return: GIC file-descriptor or negative error code upon failure
+ *
+ * The function creates a vGIC-v3 device and maps the distributor and
+ * redistributor regions of the guest. Since it depends on the number of
+ * vCPUs for the VM, it must be called after all the vCPUs have been created.
+ */
+int vgic_v3_setup(struct kvm_vm *vm, unsigned int nr_vcpus,
+		uint64_t gicd_base_gpa, uint64_t gicr_base_gpa)
+{
+	int gic_fd;
+	uint64_t redist_attr;
+	struct list_head *iter;
+	unsigned int nr_gic_pages, nr_vcpus_created = 0;
+
+	TEST_ASSERT(nr_vcpus, "Number of vCPUs cannot be empty\n");
+
+	/*
+	 * Make sure that the caller is infact calling this
+	 * function after all the vCPUs are added.
+	 */
+	list_for_each(iter, &vm->vcpus)
+		nr_vcpus_created++;
+	TEST_ASSERT(nr_vcpus == nr_vcpus_created,
+			"Number of vCPUs requested (%u) doesn't match with the ones created for the VM (%u)\n",
+			nr_vcpus, nr_vcpus_created);
+
+	/* Distributor setup */
+	gic_fd = kvm_create_device(vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+	kvm_device_access(gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			KVM_VGIC_V3_ADDR_TYPE_DIST, &gicd_base_gpa, true);
+	nr_gic_pages = vm_calc_num_guest_pages(vm->mode, KVM_VGIC_V3_DIST_SIZE);
+	virt_map(vm, gicd_base_gpa, gicd_base_gpa,  nr_gic_pages);
+
+	/* Redistributor setup */
+	redist_attr = REDIST_REGION_ATTR_ADDR(nr_vcpus, gicr_base_gpa, 0, 0);
+	kvm_device_access(gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &redist_attr, true);
+	nr_gic_pages = vm_calc_num_guest_pages(vm->mode,
+						KVM_VGIC_V3_REDIST_SIZE * nr_vcpus);
+	virt_map(vm, gicr_base_gpa, gicr_base_gpa,  nr_gic_pages);
+
+	kvm_device_access(gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+				KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	return gic_fd;
+}

tools/testing/selftests/kvm/lib/kvm_util.c

···
 void vcpu_sregs_set(struct kvm_vm *vm, uint32_t vcpuid, struct kvm_sregs *sregs)
 {
 	int ret = _vcpu_sregs_set(vm, vcpuid, sregs);
-	TEST_ASSERT(ret == 0, "KVM_RUN IOCTL failed, "
+	TEST_ASSERT(ret == 0, "KVM_SET_SREGS IOCTL failed, "
 		"rc: %i errno: %i", ret, errno);
 }
 
···
 {
 	int ret = _kvm_device_check_attr(dev_fd, group, attr);
 
-	TEST_ASSERT(ret >= 0, "KVM_HAS_DEVICE_ATTR failed, rc: %i errno: %i", ret, errno);
+	TEST_ASSERT(!ret, "KVM_HAS_DEVICE_ATTR failed, rc: %i errno: %i", ret, errno);
 	return ret;
 }
 
···
 	ret = _kvm_create_device(vm, type, test, &fd);
 
 	if (!test) {
-		TEST_ASSERT(ret >= 0,
+		TEST_ASSERT(!ret,
 			    "KVM_CREATE_DEVICE IOCTL failed, rc: %i errno: %i", ret, errno);
 		return fd;
 	}
···
 {
 	int ret = _kvm_device_access(dev_fd, group, attr, val, write);
 
-	TEST_ASSERT(ret >= 0, "KVM_SET|GET_DEVICE_ATTR IOCTL failed, rc: %i errno: %i", ret, errno);
+	TEST_ASSERT(!ret, "KVM_SET|GET_DEVICE_ATTR IOCTL failed, rc: %i errno: %i", ret, errno);
+	return ret;
+}
+
+int _vcpu_has_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			  uint64_t attr)
+{
+	struct vcpu *vcpu = vcpu_find(vm, vcpuid);
+
+	TEST_ASSERT(vcpu, "nonexistent vcpu id: %d", vcpuid);
+
+	return _kvm_device_check_attr(vcpu->fd, group, attr);
+}
+
+int vcpu_has_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+				 uint64_t attr)
+{
+	int ret = _vcpu_has_device_attr(vm, vcpuid, group, attr);
+
+	TEST_ASSERT(!ret, "KVM_HAS_DEVICE_ATTR IOCTL failed, rc: %i errno: %i", ret, errno);
+	return ret;
+}
+
+int _vcpu_access_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			     uint64_t attr, void *val, bool write)
+{
+	struct vcpu *vcpu = vcpu_find(vm, vcpuid);
+
+	TEST_ASSERT(vcpu, "nonexistent vcpu id: %d", vcpuid);
+
+	return _kvm_device_access(vcpu->fd, group, attr, val, write);
+}
+
+int vcpu_access_device_attr(struct kvm_vm *vm, uint32_t vcpuid, uint32_t group,
+			    uint64_t attr, void *val, bool write)
+{
+	int ret = _vcpu_access_device_attr(vm, vcpuid, group, attr, val, write);
+
+	TEST_ASSERT(!ret, "KVM_SET|GET_DEVICE_ATTR IOCTL failed, rc: %i errno: %i", ret, errno);
 	return ret;
 }
 

tools/testing/selftests/kvm/lib/sparsebit.c

···
 		 * of total bits set.
 		 */
 		if (s->num_set != total_bits_set) {
-			fprintf(stderr, "Number of bits set missmatch,\n"
+			fprintf(stderr, "Number of bits set mismatch,\n"
 				"  s->num_set: 0x%lx total_bits_set: 0x%lx",
 				s->num_set, total_bits_set);
 

tools/testing/selftests/kvm/lib/x86_64/processor.c

···
 
 	/* Create VCPU */
 	vm_vcpu_add(vm, vcpuid);
+	vcpu_set_cpuid(vm, vcpuid, kvm_get_supported_cpuid());
 	vcpu_setup(vm, vcpuid);
 
 	/* Setup guest general purpose registers */
···
 	/* Setup the MP state */
 	mp_state.mp_state = 0;
 	vcpu_set_mp_state(vm, vcpuid, &mp_state);
-
-	/* Setup supported CPUIDs */
-	vcpu_set_cpuid(vm, vcpuid, kvm_get_supported_cpuid());
 }
 
 /*

tools/testing/selftests/kvm/lib/x86_64/svm.c

···
 	seg->base = base;
 }
 
+/*
+ * Avoid using memset to clear the vmcb, since libc may not be
+ * available in L1 (and, even if it is, features that libc memset may
+ * want to use, like AVX, may not be enabled).
+ */
+static void clear_vmcb(struct vmcb *vmcb)
+{
+	int n = sizeof(*vmcb) / sizeof(u32);
+
+	asm volatile ("rep stosl" : "+c"(n), "+D"(vmcb) : "a"(0) : "memory");
+}
+
 void generic_svm_setup(struct svm_test_data *svm, void *guest_rip, void *guest_rsp)
 {
 	struct vmcb *vmcb = svm->vmcb;
···
 	wrmsr(MSR_EFER, efer | EFER_SVME);
 	wrmsr(MSR_VM_HSAVE_PA, svm->save_area_gpa);
 
-	memset(vmcb, 0, sizeof(*vmcb));
+	clear_vmcb(vmcb);
 	asm volatile ("vmsave %0\n\t" : : "a" (vmcb_gpa) : "memory");
 	vmcb_set_seg(&save->es, get_es(), 0, -1U, data_seg_attr);
 	vmcb_set_seg(&save->cs, get_cs(), 0, -1U, code_seg_attr);

tools/testing/selftests/kvm/memslot_perf_test.c

···
 			pr_info(__VA_ARGS__);	\
 	} while (0)
 
+static void check_mmio_access(struct vm_data *vm, struct kvm_run *run)
+{
+	TEST_ASSERT(vm->mmio_ok, "Unexpected mmio exit");
+	TEST_ASSERT(run->mmio.is_write, "Unexpected mmio read");
+	TEST_ASSERT(run->mmio.len == 8,
+		    "Unexpected exit mmio size = %u", run->mmio.len);
+	TEST_ASSERT(run->mmio.phys_addr >= vm->mmio_gpa_min &&
+		    run->mmio.phys_addr <= vm->mmio_gpa_max,
+		    "Unexpected exit mmio address = 0x%llx",
+		    run->mmio.phys_addr);
+}
+
 static void *vcpu_worker(void *data)
 {
 	struct vm_data *vm = data;
 	struct kvm_run *run;
 	struct ucall uc;
-	uint64_t cmd;
 
 	run = vcpu_state(vm->vm, VCPU_ID);
 	while (1) {
 		vcpu_run(vm->vm, VCPU_ID);
 
-		if (run->exit_reason == KVM_EXIT_IO) {
-			cmd = get_ucall(vm->vm, VCPU_ID, &uc);
-			if (cmd != UCALL_SYNC)
-				break;
-
+		switch (get_ucall(vm->vm, VCPU_ID, &uc)) {
+		case UCALL_SYNC:
+			TEST_ASSERT(uc.args[1] == 0,
+				"Unexpected sync ucall, got %lx",
+				(ulong)uc.args[1]);
 			sem_post(&vcpu_ready);
 			continue;
-		}
-
-		if (run->exit_reason != KVM_EXIT_MMIO)
+		case UCALL_NONE:
+			if (run->exit_reason == KVM_EXIT_MMIO)
+				check_mmio_access(vm, run);
+			else
+				goto done;
 			break;
-
-		TEST_ASSERT(vm->mmio_ok, "Unexpected mmio exit");
-		TEST_ASSERT(run->mmio.is_write, "Unexpected mmio read");
-		TEST_ASSERT(run->mmio.len == 8,
-			    "Unexpected exit mmio size = %u", run->mmio.len);
-		TEST_ASSERT(run->mmio.phys_addr >= vm->mmio_gpa_min &&
-			    run->mmio.phys_addr <= vm->mmio_gpa_max,
-			    "Unexpected exit mmio address = 0x%llx",
-			    run->mmio.phys_addr);
+		case UCALL_ABORT:
+			TEST_FAIL("%s at %s:%ld, val = %lu",
+					(const char *)uc.args[0],
+					__FILE__, uc.args[1], uc.args[2]);
+			break;
+		case UCALL_DONE:
+			goto done;
+		default:
+			TEST_FAIL("Unknown ucall %lu", uc.cmd);
+		}
 	}
 
-	if (run->exit_reason == KVM_EXIT_IO && cmd == UCALL_ABORT)
-		TEST_FAIL("%s at %s:%ld, val = %lu", (const char *)uc.args[0],
-			  __FILE__, uc.args[1], uc.args[2]);
-
+done:
 	return NULL;
 }
 
···
 	TEST_ASSERT(data->hva_slots, "malloc() fail");
 
 	data->vm = vm_create_default(VCPU_ID, mempages, guest_code);
+	ucall_init(data->vm, NULL);
 
 	pr_info_v("Adding slots 1..%i, each slot with %"PRIu64" pages + %"PRIu64" extra pages last\n",
 		max_mem_slots - 1, data->pages_per_slot, rempages);

tools/testing/selftests/kvm/system_counter_offset_test.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021, Google LLC.
+ *
+ * Tests for adjusting the system counter from userspace
+ */
+#include <asm/kvm_para.h>
+#include <stdint.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <time.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#define VCPU_ID 0
+
+#ifdef __x86_64__
+
+struct test_case {
+	uint64_t tsc_offset;
+};
+
+static struct test_case test_cases[] = {
+	{ 0 },
+	{ 180 * NSEC_PER_SEC },
+	{ -180 * NSEC_PER_SEC },
+};
+
+static void check_preconditions(struct kvm_vm *vm)
+{
+	if (!_vcpu_has_device_attr(vm, VCPU_ID, KVM_VCPU_TSC_CTRL, KVM_VCPU_TSC_OFFSET))
+		return;
+
+	print_skip("KVM_VCPU_TSC_OFFSET not supported; skipping test");
+	exit(KSFT_SKIP);
+}
+
+static void setup_system_counter(struct kvm_vm *vm, struct test_case *test)
+{
+	vcpu_access_device_attr(vm, VCPU_ID, KVM_VCPU_TSC_CTRL,
+				KVM_VCPU_TSC_OFFSET, &test->tsc_offset, true);
+}
+
+static uint64_t guest_read_system_counter(struct test_case *test)
+{
+	return rdtsc();
+}
+
+static uint64_t host_read_guest_system_counter(struct test_case *test)
+{
+	return rdtsc() + test->tsc_offset;
+}
+
+#else /* __x86_64__ */
+
+#error test not implemented for this architecture!
+
+#endif
+
+#define GUEST_SYNC_CLOCK(__stage, __val)			\
+		GUEST_SYNC_ARGS(__stage, __val, 0, 0, 0)
+
+static void guest_main(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		struct test_case *test = &test_cases[i];
+
+		GUEST_SYNC_CLOCK(i, guest_read_system_counter(test));
+	}
+}
+
+static void handle_sync(struct ucall *uc, uint64_t start, uint64_t end)
+{
+	uint64_t obs = uc->args[2];
+
+	TEST_ASSERT(start <= obs && obs <= end,
+		    "unexpected system counter value: %"PRIu64" expected range: [%"PRIu64", %"PRIu64"]",
+		    obs, start, end);
+
+	pr_info("system counter value: %"PRIu64" expected range [%"PRIu64", %"PRIu64"]\n",
+		obs, start, end);
+}
+
+static void handle_abort(struct ucall *uc)
+{
+	TEST_FAIL("%s at %s:%ld", (const char *)uc->args[0],
+		  __FILE__, uc->args[1]);
+}
+
+static void enter_guest(struct kvm_vm *vm)
+{
+	uint64_t start, end;
+	struct ucall uc;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		struct test_case *test = &test_cases[i];
+
+		setup_system_counter(vm, test);
+		start = host_read_guest_system_counter(test);
+		vcpu_run(vm, VCPU_ID);
+		end = host_read_guest_system_counter(test);
+
+		switch (get_ucall(vm, VCPU_ID, &uc)) {
+		case UCALL_SYNC:
+			handle_sync(&uc, start, end);
+			break;
+		case UCALL_ABORT:
+			handle_abort(&uc);
+			return;
+		default:
+			TEST_ASSERT(0, "unhandled ucall %ld\n",
+				    get_ucall(vm, VCPU_ID, &uc));
+		}
+	}
+}
+
+int main(void)
+{
+	struct kvm_vm *vm;
+
+	vm = vm_create_default(VCPU_ID, 0, guest_main);
+	check_preconditions(vm);
+	ucall_init(vm, NULL);
+
+	enter_guest(vm);
+	kvm_vm_free(vm);
+}

tools/testing/selftests/kvm/x86_64/cr4_cpuid_sync_test.c

···
 		}
 	}
 
-	kvm_vm_free(vm);
-
 done:
+	kvm_vm_free(vm);
 	return 0;
 }

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

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021, Google LLC.
+ *
+ * Tests for adjusting the KVM clock from userspace
+ */
+#include <asm/kvm_para.h>
+#include <asm/pvclock.h>
+#include <asm/pvclock-abi.h>
+#include <stdint.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <time.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#define VCPU_ID 0
+
+struct test_case {
+	uint64_t kvmclock_base;
+	int64_t realtime_offset;
+};
+
+static struct test_case test_cases[] = {
+	{ .kvmclock_base = 0 },
+	{ .kvmclock_base = 180 * NSEC_PER_SEC },
+	{ .kvmclock_base = 0, .realtime_offset = -180 * NSEC_PER_SEC },
+	{ .kvmclock_base = 0, .realtime_offset = 180 * NSEC_PER_SEC },
+};
+
+#define GUEST_SYNC_CLOCK(__stage, __val)			\
+		GUEST_SYNC_ARGS(__stage, __val, 0, 0, 0)
+
+static void guest_main(vm_paddr_t pvti_pa, struct pvclock_vcpu_time_info *pvti)
+{
+	int i;
+
+	wrmsr(MSR_KVM_SYSTEM_TIME_NEW, pvti_pa | KVM_MSR_ENABLED);
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++)
+		GUEST_SYNC_CLOCK(i, __pvclock_read_cycles(pvti, rdtsc()));
+}
+
+#define EXPECTED_FLAGS (KVM_CLOCK_REALTIME | KVM_CLOCK_HOST_TSC)
+
+static inline void assert_flags(struct kvm_clock_data *data)
+{
+	TEST_ASSERT((data->flags & EXPECTED_FLAGS) == EXPECTED_FLAGS,
+		    "unexpected clock data flags: %x (want set: %x)",
+		    data->flags, EXPECTED_FLAGS);
+}
+
+static void handle_sync(struct ucall *uc, struct kvm_clock_data *start,
+			struct kvm_clock_data *end)
+{
+	uint64_t obs, exp_lo, exp_hi;
+
+	obs = uc->args[2];
+	exp_lo = start->clock;
+	exp_hi = end->clock;
+
+	assert_flags(start);
+	assert_flags(end);
+
+	TEST_ASSERT(exp_lo <= obs && obs <= exp_hi,
+		    "unexpected kvm-clock value: %"PRIu64" expected range: [%"PRIu64", %"PRIu64"]",
+		    obs, exp_lo, exp_hi);
+
+	pr_info("kvm-clock value: %"PRIu64" expected range [%"PRIu64", %"PRIu64"]\n",
+		obs, exp_lo, exp_hi);
+}
+
+static void handle_abort(struct ucall *uc)
+{
+	TEST_FAIL("%s at %s:%ld", (const char *)uc->args[0],
+		  __FILE__, uc->args[1]);
+}
+
+static void setup_clock(struct kvm_vm *vm, struct test_case *test_case)
+{
+	struct kvm_clock_data data;
+
+	memset(&data, 0, sizeof(data));
+
+	data.clock = test_case->kvmclock_base;
+	if (test_case->realtime_offset) {
+		struct timespec ts;
+		int r;
+
+		data.flags |= KVM_CLOCK_REALTIME;
+		do {
+			r = clock_gettime(CLOCK_REALTIME, &ts);
+			if (!r)
+				break;
+		} while (errno == EINTR);
+
+		TEST_ASSERT(!r, "clock_gettime() failed: %d\n", r);
+
+		data.realtime = ts.tv_sec * NSEC_PER_SEC;
+		data.realtime += ts.tv_nsec;
+		data.realtime += test_case->realtime_offset;
+	}
+
+	vm_ioctl(vm, KVM_SET_CLOCK, &data);
+}
+
+static void enter_guest(struct kvm_vm *vm)
+{
+	struct kvm_clock_data start, end;
+	struct kvm_run *run;
+	struct ucall uc;
+	int i, r;
+
+	run = vcpu_state(vm, VCPU_ID);
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		setup_clock(vm, &test_cases[i]);
+
+		vm_ioctl(vm, KVM_GET_CLOCK, &start);
+
+		r = _vcpu_run(vm, VCPU_ID);
+		vm_ioctl(vm, KVM_GET_CLOCK, &end);
+
+		TEST_ASSERT(!r, "vcpu_run failed: %d\n", r);
+		TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
+			    "unexpected exit reason: %u (%s)",
+			    run->exit_reason, exit_reason_str(run->exit_reason));
+
+		switch (get_ucall(vm, VCPU_ID, &uc)) {
+		case UCALL_SYNC:
+			handle_sync(&uc, &start, &end);
+			break;
+		case UCALL_ABORT:
+			handle_abort(&uc);
+			return;
+		default:
+			TEST_ASSERT(0, "unhandled ucall: %ld\n", uc.cmd);
+		}
+	}
+}
+
+#define CLOCKSOURCE_PATH "/sys/devices/system/clocksource/clocksource0/current_clocksource"
+
+static void check_clocksource(void)
+{
+	char *clk_name;
+	struct stat st;
+	FILE *fp;
+
+	fp = fopen(CLOCKSOURCE_PATH, "r");
+	if (!fp) {
+		pr_info("failed to open clocksource file: %d; assuming TSC.\n",
+			errno);
+		return;
+	}
+
+	if (fstat(fileno(fp), &st)) {
+		pr_info("failed to stat clocksource file: %d; assuming TSC.\n",
+			errno);
+		goto out;
+	}
+
+	clk_name = malloc(st.st_size);
+	TEST_ASSERT(clk_name, "failed to allocate buffer to read file\n");
+
+	if (!fgets(clk_name, st.st_size, fp)) {
+		pr_info("failed to read clocksource file: %d; assuming TSC.\n",
+			ferror(fp));
+		goto out;
+	}
+
+	TEST_ASSERT(!strncmp(clk_name, "tsc\n", st.st_size),
+		    "clocksource not supported: %s", clk_name);
+out:
+	fclose(fp);
+}
+
+int main(void)
+{
+	vm_vaddr_t pvti_gva;
+	vm_paddr_t pvti_gpa;
+	struct kvm_vm *vm;
+	int flags;
+
+	flags = kvm_check_cap(KVM_CAP_ADJUST_CLOCK);
+	if (!(flags & KVM_CLOCK_REALTIME)) {
+		print_skip("KVM_CLOCK_REALTIME not supported; flags: %x",
+			   flags);
+		exit(KSFT_SKIP);
+	}
+
+	check_clocksource();
+
+	vm = vm_create_default(VCPU_ID, 0, guest_main);
+
+	pvti_gva = vm_vaddr_alloc(vm, getpagesize(), 0x10000);
+	pvti_gpa = addr_gva2gpa(vm, pvti_gva);
+	vcpu_args_set(vm, VCPU_ID, 2, pvti_gpa, pvti_gva);
+
+	enter_guest(vm);
+	kvm_vm_free(vm);
+}

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

···
 		}
 	}
 
-	kvm_vm_free(vm);
 done:
+	kvm_vm_free(vm);
 	return 0;
 }

virt/kvm/eventfd.c

···
 {
 	return 0;
 }
+
+bool __attribute__((weak)) kvm_arch_irqfd_route_changed(
+				struct kvm_kernel_irq_routing_entry *old,
+				struct kvm_kernel_irq_routing_entry *new)
+{
+	return true;
+}
 #endif
 
 static int
···
 	spin_lock_irq(&kvm->irqfds.lock);
 
 	list_for_each_entry(irqfd, &kvm->irqfds.items, list) {
+#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+		/* Under irqfds.lock, so can read irq_entry safely */
+		struct kvm_kernel_irq_routing_entry old = irqfd->irq_entry;
+#endif
+
 		irqfd_update(kvm, irqfd);
 
 #ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
-		if (irqfd->producer) {
+		if (irqfd->producer &&
+		    kvm_arch_irqfd_route_changed(&old, &irqfd->irq_entry)) {
 			int ret = kvm_arch_update_irqfd_routing(
 					irqfd->kvm, irqfd->producer->irq,
 					irqfd->gsi, 1);

virt/kvm/kvm_main.c

···
 static unsigned long long kvm_createvm_count;
 static unsigned long long kvm_active_vms;
 
+static DEFINE_PER_CPU(cpumask_var_t, cpu_kick_mask);
+
 __weak void kvm_arch_mmu_notifier_invalidate_range(struct kvm *kvm,
 						   unsigned long start, unsigned long end)
 {
···
 {
 }
 
-static inline bool kvm_kick_many_cpus(cpumask_var_t tmp, bool wait)
+static inline bool kvm_kick_many_cpus(struct cpumask *cpus, bool wait)
 {
-	const struct cpumask *cpus;
-
-	if (likely(cpumask_available(tmp)))
-		cpus = tmp;
-	else
-		cpus = cpu_online_mask;
-
 	if (cpumask_empty(cpus))
 		return false;
 
···
 	return true;
 }
 
-bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
-				 struct kvm_vcpu *except,
-				 unsigned long *vcpu_bitmap, cpumask_var_t tmp)
+static void kvm_make_vcpu_request(struct kvm *kvm, struct kvm_vcpu *vcpu,
+				  unsigned int req, struct cpumask *tmp,
+				  int current_cpu)
 {
-	int i, cpu, me;
+	int cpu;
+
+	kvm_make_request(req, vcpu);
+
+	if (!(req & KVM_REQUEST_NO_WAKEUP) && kvm_vcpu_wake_up(vcpu))
+		return;
+
+	/*
+	 * Note, the vCPU could get migrated to a different pCPU at any point
+	 * after kvm_request_needs_ipi(), which could result in sending an IPI
+	 * to the previous pCPU.  But, that's OK because the purpose of the IPI
+	 * is to ensure the vCPU returns to OUTSIDE_GUEST_MODE, which is
+	 * satisfied if the vCPU migrates. Entering READING_SHADOW_PAGE_TABLES
+	 * after this point is also OK, as the requirement is only that KVM wait
+	 * for vCPUs that were reading SPTEs _before_ any changes were
+	 * finalized. See kvm_vcpu_kick() for more details on handling requests.
+	 */
+	if (kvm_request_needs_ipi(vcpu, req)) {
+		cpu = READ_ONCE(vcpu->cpu);
+		if (cpu != -1 && cpu != current_cpu)
+			__cpumask_set_cpu(cpu, tmp);
+	}
+}
+
+bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
+				 unsigned long *vcpu_bitmap)
+{
 	struct kvm_vcpu *vcpu;
+	struct cpumask *cpus;
+	int i, me;
 	bool called;
 
 	me = get_cpu();
 
-	kvm_for_each_vcpu(i, vcpu, kvm) {
-		if ((vcpu_bitmap && !test_bit(i, vcpu_bitmap)) ||
-		    vcpu == except)
+	cpus = this_cpu_cpumask_var_ptr(cpu_kick_mask);
+	cpumask_clear(cpus);
+
+	for_each_set_bit(i, vcpu_bitmap, KVM_MAX_VCPUS) {
+		vcpu = kvm_get_vcpu(kvm, i);
+		if (!vcpu)
 			continue;
-
-		kvm_make_request(req, vcpu);
-
-		if (!(req & KVM_REQUEST_NO_WAKEUP) && kvm_vcpu_wake_up(vcpu))
-			continue;
-
-		/*
-		 * tmp can be "unavailable" if cpumasks are allocated off stack
-		 * as allocation of the mask is deliberately not fatal and is
-		 * handled by falling back to kicking all online CPUs.
-		 */
-		if (!cpumask_available(tmp))
-			continue;
-
-		/*
-		 * Note, the vCPU could get migrated to a different pCPU at any
-		 * point after kvm_request_needs_ipi(), which could result in
-		 * sending an IPI to the previous pCPU.  But, that's ok because
-		 * the purpose of the IPI is to ensure the vCPU returns to
-		 * OUTSIDE_GUEST_MODE, which is satisfied if the vCPU migrates.
-		 * Entering READING_SHADOW_PAGE_TABLES after this point is also
-		 * ok, as the requirement is only that KVM wait for vCPUs that
-		 * were reading SPTEs _before_ any changes were finalized.  See
-		 * kvm_vcpu_kick() for more details on handling requests.
-		 */
-		if (kvm_request_needs_ipi(vcpu, req)) {
-			cpu = READ_ONCE(vcpu->cpu);
-			if (cpu != -1 && cpu != me)
-				__cpumask_set_cpu(cpu, tmp);
-		}
+		kvm_make_vcpu_request(kvm, vcpu, req, cpus, me);
 	}
 
-	called = kvm_kick_many_cpus(tmp, !!(req & KVM_REQUEST_WAIT));
+	called = kvm_kick_many_cpus(cpus, !!(req & KVM_REQUEST_WAIT));
 	put_cpu();
 
 	return called;
···
 bool kvm_make_all_cpus_request_except(struct kvm *kvm, unsigned int req,
 				      struct kvm_vcpu *except)
 {
-	cpumask_var_t cpus;
+	struct kvm_vcpu *vcpu;
+	struct cpumask *cpus;
 	bool called;
+	int i, me;
 
-	zalloc_cpumask_var(&cpus, GFP_ATOMIC);
+	me = get_cpu();
 
-	called = kvm_make_vcpus_request_mask(kvm, req, except, NULL, cpus);
+	cpus = this_cpu_cpumask_var_ptr(cpu_kick_mask);
+	cpumask_clear(cpus);
 
-	free_cpumask_var(cpus);
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		if (vcpu == except)
+			continue;
+		kvm_make_vcpu_request(kvm, vcpu, req, cpus, me);
+	}
+
+	called = kvm_kick_many_cpus(cpus, !!(req & KVM_REQUEST_WAIT));
+	put_cpu();
+
 	return called;
 }
 
···
 static int kvm_vcpu_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct kvm_vcpu *vcpu = file->private_data;
-	unsigned long pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+	unsigned long pages = vma_pages(vma);
 
 	if ((kvm_page_in_dirty_ring(vcpu->kvm, vma->vm_pgoff) ||
 	     kvm_page_in_dirty_ring(vcpu->kvm, vma->vm_pgoff + pages - 1)) &&
···
 	struct kvm_vcpu *vcpu;
 	struct page *page;
 
-	if (id >= KVM_MAX_VCPU_ID)
+	if (id >= KVM_MAX_VCPU_IDS)
 		return -EINVAL;
 
 	mutex_lock(&kvm->lock);
···
 		goto out_free_3;
 	}
 
+	for_each_possible_cpu(cpu) {
+		if (!alloc_cpumask_var_node(&per_cpu(cpu_kick_mask, cpu),
+					    GFP_KERNEL, cpu_to_node(cpu))) {
+			r = -ENOMEM;
+			goto out_free_4;
+		}
+	}
+
 	r = kvm_async_pf_init();
 	if (r)
-		goto out_free;
+		goto out_free_5;
 
 	kvm_chardev_ops.owner = module;
 	kvm_vm_fops.owner = module;
···
 
 out_unreg:
 	kvm_async_pf_deinit();
-out_free:
+out_free_5:
+	for_each_possible_cpu(cpu)
+		free_cpumask_var(per_cpu(cpu_kick_mask, cpu));
+out_free_4:
 	kmem_cache_destroy(kvm_vcpu_cache);
 out_free_3:
 	unregister_reboot_notifier(&kvm_reboot_notifier);
···
 
 void kvm_exit(void)
 {
+	int cpu;
+
 	debugfs_remove_recursive(kvm_debugfs_dir);
 	misc_deregister(&kvm_dev);
+	for_each_possible_cpu(cpu)
+		free_cpumask_var(per_cpu(cpu_kick_mask, cpu));
 	kmem_cache_destroy(kvm_vcpu_cache);
 	kvm_async_pf_deinit();
 	unregister_syscore_ops(&kvm_syscore_ops);