1 .. SPDX-License-Identifier: GPL-2.0
3 ===================================================================
4 The Definitive KVM (Kernel-based Virtual Machine) API Documentation
5 ===================================================================
10 The kvm API is a set of ioctls that are issued to control various aspects
11 of a virtual machine. The ioctls belong to the following classes:
13 - System ioctls: These query and set global attributes which affect the
14 whole kvm subsystem. In addition a system ioctl is used to create
17 - VM ioctls: These query and set attributes that affect an entire virtual
18 machine, for example memory layout. In addition a VM ioctl is used to
19 create virtual cpus (vcpus) and devices.
21 VM ioctls must be issued from the same process (address space) that was
22 used to create the VM.
24 - vcpu ioctls: These query and set attributes that control the operation
25 of a single virtual cpu.
27 vcpu ioctls should be issued from the same thread that was used to create
28 the vcpu, except for asynchronous vcpu ioctl that are marked as such in
29 the documentation. Otherwise, the first ioctl after switching threads
30 could see a performance impact.
32 - device ioctls: These query and set attributes that control the operation
35 device ioctls must be issued from the same process (address space) that
36 was used to create the VM.
41 The kvm API is centered around file descriptors. An initial
42 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle
43 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this
44 handle will create a VM file descriptor which can be used to issue VM
45 ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVICE ioctl on a VM fd will
46 create a virtual cpu or device and return a file descriptor pointing to
47 the new resource. Finally, ioctls on a vcpu or device fd can be used
48 to control the vcpu or device. For vcpus, this includes the important
49 task of actually running guest code.
51 In general file descriptors can be migrated among processes by means
52 of fork() and the SCM_RIGHTS facility of unix domain socket. These
53 kinds of tricks are explicitly not supported by kvm. While they will
54 not cause harm to the host, their actual behavior is not guaranteed by
55 the API. See "General description" for details on the ioctl usage
56 model that is supported by KVM.
58 It is important to note that althought VM ioctls may only be issued from
59 the process that created the VM, a VM's lifecycle is associated with its
60 file descriptor, not its creator (process). In other words, the VM and
61 its resources, *including the associated address space*, are not freed
62 until the last reference to the VM's file descriptor has been released.
63 For example, if fork() is issued after ioctl(KVM_CREATE_VM), the VM will
64 not be freed until both the parent (original) process and its child have
65 put their references to the VM's file descriptor.
67 Because a VM's resources are not freed until the last reference to its
68 file descriptor is released, creating additional references to a VM
69 via fork(), dup(), etc... without careful consideration is strongly
70 discouraged and may have unwanted side effects, e.g. memory allocated
71 by and on behalf of the VM's process may not be freed/unaccounted when
78 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward
79 incompatible change are allowed. However, there is an extension
80 facility that allows backward-compatible extensions to the API to be
83 The extension mechanism is not based on the Linux version number.
84 Instead, kvm defines extension identifiers and a facility to query
85 whether a particular extension identifier is available. If it is, a
86 set of ioctls is available for application use.
92 This section describes ioctls that can be used to control kvm guests.
93 For each ioctl, the following information is provided along with a
97 which KVM extension provides this ioctl. Can be 'basic',
98 which means that is will be provided by any kernel that supports
99 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which
100 means availability needs to be checked with KVM_CHECK_EXTENSION
101 (see section 4.4), or 'none' which means that while not all kernels
102 support this ioctl, there's no capability bit to check its
103 availability: for kernels that don't support the ioctl,
104 the ioctl returns -ENOTTY.
107 which instruction set architectures provide this ioctl.
108 x86 includes both i386 and x86_64.
114 what parameters are accepted by the ioctl.
117 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
118 are not detailed, but errors with specific meanings are.
121 4.1 KVM_GET_API_VERSION
122 -----------------------
128 :Returns: the constant KVM_API_VERSION (=12)
130 This identifies the API version as the stable kvm API. It is not
131 expected that this number will change. However, Linux 2.6.20 and
132 2.6.21 report earlier versions; these are not documented and not
133 supported. Applications should refuse to run if KVM_GET_API_VERSION
134 returns a value other than 12. If this check passes, all ioctls
135 described as 'basic' will be available.
144 :Parameters: machine type identifier (KVM_VM_*)
145 :Returns: a VM fd that can be used to control the new virtual machine.
147 The new VM has no virtual cpus and no memory.
148 You probably want to use 0 as machine type.
150 In order to create user controlled virtual machines on S390, check
151 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
152 privileged user (CAP_SYS_ADMIN).
154 To use hardware assisted virtualization on MIPS (VZ ASE) rather than
155 the default trap & emulate implementation (which changes the virtual
156 memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
160 On arm64, the physical address size for a VM (IPA Size limit) is limited
161 to 40bits by default. The limit can be configured if the host supports the
162 extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use
163 KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type
164 identifier, where IPA_Bits is the maximum width of any physical
165 address used by the VM. The IPA_Bits is encoded in bits[7-0] of the
166 machine type identifier.
168 e.g, to configure a guest to use 48bit physical address size::
170 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48));
172 The requested size (IPA_Bits) must be:
174 == =========================================================
175 0 Implies default size, 40bits (for backward compatibility)
176 N Implies N bits, where N is a positive integer such that,
177 32 <= N <= Host_IPA_Limit
178 == =========================================================
180 Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and
181 is dependent on the CPU capability and the kernel configuration. The limit can
182 be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION
185 Creation of the VM will fail if the requested IPA size (whether it is
186 implicit or explicit) is unsupported on the host.
188 Please note that configuring the IPA size does not affect the capability
189 exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects
190 size of the address translated by the stage2 level (guest physical to
191 host physical address translations).
194 4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST
195 ----------------------------------------------------------
197 :Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST
200 :Parameters: struct kvm_msr_list (in/out)
201 :Returns: 0 on success; -1 on error
205 ====== ============================================================
206 EFAULT the msr index list cannot be read from or written to
207 E2BIG the msr index list is to be to fit in the array specified by
209 ====== ============================================================
213 struct kvm_msr_list {
214 __u32 nmsrs; /* number of msrs in entries */
218 The user fills in the size of the indices array in nmsrs, and in return
219 kvm adjusts nmsrs to reflect the actual number of msrs and fills in the
220 indices array with their numbers.
222 KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list
223 varies by kvm version and host processor, but does not change otherwise.
225 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
226 not returned in the MSR list, as different vcpus can have a different number
227 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
229 KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed
230 to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities
231 and processor features that are exposed via MSRs (e.g., VMX capabilities).
232 This list also varies by kvm version and host processor, but does not change
236 4.4 KVM_CHECK_EXTENSION
237 -----------------------
239 :Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
241 :Type: system ioctl, vm ioctl
242 :Parameters: extension identifier (KVM_CAP_*)
243 :Returns: 0 if unsupported; 1 (or some other positive integer) if supported
245 The API allows the application to query about extensions to the core
246 kvm API. Userspace passes an extension identifier (an integer) and
247 receives an integer that describes the extension availability.
248 Generally 0 means no and 1 means yes, but some extensions may report
249 additional information in the integer return value.
251 Based on their initialization different VMs may have different capabilities.
252 It is thus encouraged to use the vm ioctl to query for capabilities (available
253 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
255 4.5 KVM_GET_VCPU_MMAP_SIZE
256 --------------------------
262 :Returns: size of vcpu mmap area, in bytes
264 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
265 memory region. This ioctl returns the size of that region. See the
266 KVM_RUN documentation for details.
269 4.6 KVM_SET_MEMORY_REGION
270 -------------------------
275 :Parameters: struct kvm_memory_region (in)
276 :Returns: 0 on success, -1 on error
278 This ioctl is obsolete and has been removed.
287 :Parameters: vcpu id (apic id on x86)
288 :Returns: vcpu fd on success, -1 on error
290 This API adds a vcpu to a virtual machine. No more than max_vcpus may be added.
291 The vcpu id is an integer in the range [0, max_vcpu_id).
293 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
294 the KVM_CHECK_EXTENSION ioctl() at run-time.
295 The maximum possible value for max_vcpus can be retrieved using the
296 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
298 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
300 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
301 same as the value returned from KVM_CAP_NR_VCPUS.
303 The maximum possible value for max_vcpu_id can be retrieved using the
304 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time.
306 If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id
307 is the same as the value returned from KVM_CAP_MAX_VCPUS.
309 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
310 threads in one or more virtual CPU cores. (This is because the
311 hardware requires all the hardware threads in a CPU core to be in the
312 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
313 of vcpus per virtual core (vcore). The vcore id is obtained by
314 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
315 given vcore will always be in the same physical core as each other
316 (though that might be a different physical core from time to time).
317 Userspace can control the threading (SMT) mode of the guest by its
318 allocation of vcpu ids. For example, if userspace wants
319 single-threaded guest vcpus, it should make all vcpu ids be a multiple
320 of the number of vcpus per vcore.
322 For virtual cpus that have been created with S390 user controlled virtual
323 machines, the resulting vcpu fd can be memory mapped at page offset
324 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
325 cpu's hardware control block.
328 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
329 --------------------------------
334 :Parameters: struct kvm_dirty_log (in/out)
335 :Returns: 0 on success, -1 on error
339 /* for KVM_GET_DIRTY_LOG */
340 struct kvm_dirty_log {
344 void __user *dirty_bitmap; /* one bit per page */
349 Given a memory slot, return a bitmap containing any pages dirtied
350 since the last call to this ioctl. Bit 0 is the first page in the
351 memory slot. Ensure the entire structure is cleared to avoid padding
354 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies
355 the address space for which you want to return the dirty bitmap.
356 They must be less than the value that KVM_CHECK_EXTENSION returns for
357 the KVM_CAP_MULTI_ADDRESS_SPACE capability.
359 The bits in the dirty bitmap are cleared before the ioctl returns, unless
360 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information,
361 see the description of the capability.
363 4.9 KVM_SET_MEMORY_ALIAS
364 ------------------------
369 :Parameters: struct kvm_memory_alias (in)
370 :Returns: 0 (success), -1 (error)
372 This ioctl is obsolete and has been removed.
382 :Returns: 0 on success, -1 on error
386 ===== =============================
387 EINTR an unmasked signal is pending
388 ===== =============================
390 This ioctl is used to run a guest virtual cpu. While there are no
391 explicit parameters, there is an implicit parameter block that can be
392 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
393 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
394 kvm_run' (see below).
401 :Architectures: all except ARM, arm64
403 :Parameters: struct kvm_regs (out)
404 :Returns: 0 on success, -1 on error
406 Reads the general purpose registers from the vcpu.
412 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
413 __u64 rax, rbx, rcx, rdx;
414 __u64 rsi, rdi, rsp, rbp;
415 __u64 r8, r9, r10, r11;
416 __u64 r12, r13, r14, r15;
422 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
434 :Architectures: all except ARM, arm64
436 :Parameters: struct kvm_regs (in)
437 :Returns: 0 on success, -1 on error
439 Writes the general purpose registers into the vcpu.
441 See KVM_GET_REGS for the data structure.
448 :Architectures: x86, ppc
450 :Parameters: struct kvm_sregs (out)
451 :Returns: 0 on success, -1 on error
453 Reads special registers from the vcpu.
459 struct kvm_segment cs, ds, es, fs, gs, ss;
460 struct kvm_segment tr, ldt;
461 struct kvm_dtable gdt, idt;
462 __u64 cr0, cr2, cr3, cr4, cr8;
465 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
468 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
470 interrupt_bitmap is a bitmap of pending external interrupts. At most
471 one bit may be set. This interrupt has been acknowledged by the APIC
472 but not yet injected into the cpu core.
479 :Architectures: x86, ppc
481 :Parameters: struct kvm_sregs (in)
482 :Returns: 0 on success, -1 on error
484 Writes special registers into the vcpu. See KVM_GET_SREGS for the
494 :Parameters: struct kvm_translation (in/out)
495 :Returns: 0 on success, -1 on error
497 Translates a virtual address according to the vcpu's current address
502 struct kvm_translation {
504 __u64 linear_address;
507 __u64 physical_address;
519 :Architectures: x86, ppc, mips
521 :Parameters: struct kvm_interrupt (in)
522 :Returns: 0 on success, negative on failure.
524 Queues a hardware interrupt vector to be injected.
528 /* for KVM_INTERRUPT */
529 struct kvm_interrupt {
539 ========= ===================================
541 -EEXIST if an interrupt is already enqueued
542 -EINVAL the irq number is invalid
543 -ENXIO if the PIC is in the kernel
544 -EFAULT if the pointer is invalid
545 ========= ===================================
547 Note 'irq' is an interrupt vector, not an interrupt pin or line. This
548 ioctl is useful if the in-kernel PIC is not used.
553 Queues an external interrupt to be injected. This ioctl is overleaded
554 with 3 different irq values:
558 This injects an edge type external interrupt into the guest once it's ready
559 to receive interrupts. When injected, the interrupt is done.
561 b) KVM_INTERRUPT_UNSET
563 This unsets any pending interrupt.
565 Only available with KVM_CAP_PPC_UNSET_IRQ.
567 c) KVM_INTERRUPT_SET_LEVEL
569 This injects a level type external interrupt into the guest context. The
570 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
573 Only available with KVM_CAP_PPC_IRQ_LEVEL.
575 Note that any value for 'irq' other than the ones stated above is invalid
576 and incurs unexpected behavior.
578 This is an asynchronous vcpu ioctl and can be invoked from any thread.
583 Queues an external interrupt to be injected into the virtual CPU. A negative
584 interrupt number dequeues the interrupt.
586 This is an asynchronous vcpu ioctl and can be invoked from any thread.
596 :Returns: -1 on error
598 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
604 :Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system)
606 :Type: system ioctl, vcpu ioctl
607 :Parameters: struct kvm_msrs (in/out)
608 :Returns: number of msrs successfully returned;
611 When used as a system ioctl:
612 Reads the values of MSR-based features that are available for the VM. This
613 is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values.
614 The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST
617 When used as a vcpu ioctl:
618 Reads model-specific registers from the vcpu. Supported msr indices can
619 be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl.
624 __u32 nmsrs; /* number of msrs in entries */
627 struct kvm_msr_entry entries[0];
630 struct kvm_msr_entry {
636 Application code should set the 'nmsrs' member (which indicates the
637 size of the entries array) and the 'index' member of each array entry.
638 kvm will fill in the 'data' member.
647 :Parameters: struct kvm_msrs (in)
648 :Returns: number of msrs successfully set (see below), -1 on error
650 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
653 Application code should set the 'nmsrs' member (which indicates the
654 size of the entries array), and the 'index' and 'data' members of each
657 It tries to set the MSRs in array entries[] one by one. If setting an MSR
658 fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated
659 by KVM, etc..., it stops processing the MSR list and returns the number of
660 MSRs that have been set successfully.
669 :Parameters: struct kvm_cpuid (in)
670 :Returns: 0 on success, -1 on error
672 Defines the vcpu responses to the cpuid instruction. Applications
673 should use the KVM_SET_CPUID2 ioctl if available.
675 Note, when this IOCTL fails, KVM gives no guarantees that previous valid CPUID
676 configuration (if there is) is not corrupted. Userspace can get a copy of the
677 resulting CPUID configuration through KVM_GET_CPUID2 in case.
681 struct kvm_cpuid_entry {
690 /* for KVM_SET_CPUID */
694 struct kvm_cpuid_entry entries[0];
698 4.21 KVM_SET_SIGNAL_MASK
699 ------------------------
704 :Parameters: struct kvm_signal_mask (in)
705 :Returns: 0 on success, -1 on error
707 Defines which signals are blocked during execution of KVM_RUN. This
708 signal mask temporarily overrides the threads signal mask. Any
709 unblocked signal received (except SIGKILL and SIGSTOP, which retain
710 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
712 Note the signal will only be delivered if not blocked by the original
717 /* for KVM_SET_SIGNAL_MASK */
718 struct kvm_signal_mask {
730 :Parameters: struct kvm_fpu (out)
731 :Returns: 0 on success, -1 on error
733 Reads the floating point state from the vcpu.
737 /* for KVM_GET_FPU and KVM_SET_FPU */
742 __u8 ftwx; /* in fxsave format */
759 :Parameters: struct kvm_fpu (in)
760 :Returns: 0 on success, -1 on error
762 Writes the floating point state to the vcpu.
766 /* for KVM_GET_FPU and KVM_SET_FPU */
771 __u8 ftwx; /* in fxsave format */
782 4.24 KVM_CREATE_IRQCHIP
783 -----------------------
785 :Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
786 :Architectures: x86, ARM, arm64, s390
789 :Returns: 0 on success, -1 on error
791 Creates an interrupt controller model in the kernel.
792 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
793 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
794 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
795 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of
796 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
797 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
798 On s390, a dummy irq routing table is created.
800 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
801 before KVM_CREATE_IRQCHIP can be used.
807 :Capability: KVM_CAP_IRQCHIP
808 :Architectures: x86, arm, arm64
810 :Parameters: struct kvm_irq_level
811 :Returns: 0 on success, -1 on error
813 Sets the level of a GSI input to the interrupt controller model in the kernel.
814 On some architectures it is required that an interrupt controller model has
815 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
816 interrupts require the level to be set to 1 and then back to 0.
818 On real hardware, interrupt pins can be active-low or active-high. This
819 does not matter for the level field of struct kvm_irq_level: 1 always
820 means active (asserted), 0 means inactive (deasserted).
822 x86 allows the operating system to program the interrupt polarity
823 (active-low/active-high) for level-triggered interrupts, and KVM used
824 to consider the polarity. However, due to bitrot in the handling of
825 active-low interrupts, the above convention is now valid on x86 too.
826 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
827 should not present interrupts to the guest as active-low unless this
828 capability is present (or unless it is not using the in-kernel irqchip,
832 ARM/arm64 can signal an interrupt either at the CPU level, or at the
833 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
834 use PPIs designated for specific cpus. The irq field is interpreted
837 Â bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 |
838 field: | vcpu2_index | irq_type | vcpu_index | irq_id |
840 The irq_type field has the following values:
843 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
845 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
846 (the vcpu_index field is ignored)
848 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
850 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
852 In both cases, level is used to assert/deassert the line.
854 When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is
855 identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index
858 Note that on arm/arm64, the KVM_CAP_IRQCHIP capability only conditions
859 injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always
860 be used for a userspace interrupt controller.
864 struct kvm_irq_level {
867 __s32 status; /* not used for KVM_IRQ_LEVEL */
869 __u32 level; /* 0 or 1 */
876 :Capability: KVM_CAP_IRQCHIP
879 :Parameters: struct kvm_irqchip (in/out)
880 :Returns: 0 on success, -1 on error
882 Reads the state of a kernel interrupt controller created with
883 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
888 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
891 char dummy[512]; /* reserving space */
892 struct kvm_pic_state pic;
893 struct kvm_ioapic_state ioapic;
901 :Capability: KVM_CAP_IRQCHIP
904 :Parameters: struct kvm_irqchip (in)
905 :Returns: 0 on success, -1 on error
907 Sets the state of a kernel interrupt controller created with
908 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
913 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
916 char dummy[512]; /* reserving space */
917 struct kvm_pic_state pic;
918 struct kvm_ioapic_state ioapic;
923 4.28 KVM_XEN_HVM_CONFIG
924 -----------------------
926 :Capability: KVM_CAP_XEN_HVM
929 :Parameters: struct kvm_xen_hvm_config (in)
930 :Returns: 0 on success, -1 on error
932 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
933 page, and provides the starting address and size of the hypercall
934 blobs in userspace. When the guest writes the MSR, kvm copies one
935 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
940 struct kvm_xen_hvm_config {
954 :Capability: KVM_CAP_ADJUST_CLOCK
957 :Parameters: struct kvm_clock_data (out)
958 :Returns: 0 on success, -1 on error
960 Gets the current timestamp of kvmclock as seen by the current guest. In
961 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
964 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the
965 set of bits that KVM can return in struct kvm_clock_data's flag member.
967 The only flag defined now is KVM_CLOCK_TSC_STABLE. If set, the returned
968 value is the exact kvmclock value seen by all VCPUs at the instant
969 when KVM_GET_CLOCK was called. If clear, the returned value is simply
970 CLOCK_MONOTONIC plus a constant offset; the offset can be modified
971 with KVM_SET_CLOCK. KVM will try to make all VCPUs follow this clock,
972 but the exact value read by each VCPU could differ, because the host
977 struct kvm_clock_data {
978 __u64 clock; /* kvmclock current value */
987 :Capability: KVM_CAP_ADJUST_CLOCK
990 :Parameters: struct kvm_clock_data (in)
991 :Returns: 0 on success, -1 on error
993 Sets the current timestamp of kvmclock to the value specified in its parameter.
994 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
999 struct kvm_clock_data {
1000 __u64 clock; /* kvmclock current value */
1006 4.31 KVM_GET_VCPU_EVENTS
1007 ------------------------
1009 :Capability: KVM_CAP_VCPU_EVENTS
1010 :Extended by: KVM_CAP_INTR_SHADOW
1011 :Architectures: x86, arm, arm64
1013 :Parameters: struct kvm_vcpu_event (out)
1014 :Returns: 0 on success, -1 on error
1019 Gets currently pending exceptions, interrupts, and NMIs as well as related
1024 struct kvm_vcpu_events {
1028 __u8 has_error_code;
1049 __u8 smm_inside_nmi;
1053 __u8 exception_has_payload;
1054 __u64 exception_payload;
1057 The following bits are defined in the flags field:
1059 - KVM_VCPUEVENT_VALID_SHADOW may be set to signal that
1060 interrupt.shadow contains a valid state.
1062 - KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a
1065 - KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the
1066 exception_has_payload, exception_payload, and exception.pending
1067 fields contain a valid state. This bit will be set whenever
1068 KVM_CAP_EXCEPTION_PAYLOAD is enabled.
1073 If the guest accesses a device that is being emulated by the host kernel in
1074 such a way that a real device would generate a physical SError, KVM may make
1075 a virtual SError pending for that VCPU. This system error interrupt remains
1076 pending until the guest takes the exception by unmasking PSTATE.A.
1078 Running the VCPU may cause it to take a pending SError, or make an access that
1079 causes an SError to become pending. The event's description is only valid while
1080 the VPCU is not running.
1082 This API provides a way to read and write the pending 'event' state that is not
1083 visible to the guest. To save, restore or migrate a VCPU the struct representing
1084 the state can be read then written using this GET/SET API, along with the other
1085 guest-visible registers. It is not possible to 'cancel' an SError that has been
1088 A device being emulated in user-space may also wish to generate an SError. To do
1089 this the events structure can be populated by user-space. The current state
1090 should be read first, to ensure no existing SError is pending. If an existing
1091 SError is pending, the architecture's 'Multiple SError interrupts' rules should
1092 be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and
1093 Serviceability (RAS) Specification").
1095 SError exceptions always have an ESR value. Some CPUs have the ability to
1096 specify what the virtual SError's ESR value should be. These systems will
1097 advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will
1098 always have a non-zero value when read, and the agent making an SError pending
1099 should specify the ISS field in the lower 24 bits of exception.serror_esr. If
1100 the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events
1101 with exception.has_esr as zero, KVM will choose an ESR.
1103 Specifying exception.has_esr on a system that does not support it will return
1104 -EINVAL. Setting anything other than the lower 24bits of exception.serror_esr
1105 will return -EINVAL.
1107 It is not possible to read back a pending external abort (injected via
1108 KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered
1109 directly to the virtual CPU).
1113 struct kvm_vcpu_events {
1115 __u8 serror_pending;
1116 __u8 serror_has_esr;
1117 __u8 ext_dabt_pending;
1118 /* Align it to 8 bytes */
1125 4.32 KVM_SET_VCPU_EVENTS
1126 ------------------------
1128 :Capability: KVM_CAP_VCPU_EVENTS
1129 :Extended by: KVM_CAP_INTR_SHADOW
1130 :Architectures: x86, arm, arm64
1132 :Parameters: struct kvm_vcpu_event (in)
1133 :Returns: 0 on success, -1 on error
1138 Set pending exceptions, interrupts, and NMIs as well as related states of the
1141 See KVM_GET_VCPU_EVENTS for the data structure.
1143 Fields that may be modified asynchronously by running VCPUs can be excluded
1144 from the update. These fields are nmi.pending, sipi_vector, smi.smm,
1145 smi.pending. Keep the corresponding bits in the flags field cleared to
1146 suppress overwriting the current in-kernel state. The bits are:
1148 =============================== ==================================
1149 KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel
1150 KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector
1151 KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct.
1152 =============================== ==================================
1154 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
1155 the flags field to signal that interrupt.shadow contains a valid state and
1156 shall be written into the VCPU.
1158 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available.
1160 If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD
1161 can be set in the flags field to signal that the
1162 exception_has_payload, exception_payload, and exception.pending fields
1163 contain a valid state and shall be written into the VCPU.
1168 User space may need to inject several types of events to the guest.
1170 Set the pending SError exception state for this VCPU. It is not possible to
1171 'cancel' an Serror that has been made pending.
1173 If the guest performed an access to I/O memory which could not be handled by
1174 userspace, for example because of missing instruction syndrome decode
1175 information or because there is no device mapped at the accessed IPA, then
1176 userspace can ask the kernel to inject an external abort using the address
1177 from the exiting fault on the VCPU. It is a programming error to set
1178 ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or
1179 KVM_EXIT_ARM_NISV. This feature is only available if the system supports
1180 KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in
1181 how userspace reports accesses for the above cases to guests, across different
1182 userspace implementations. Nevertheless, userspace can still emulate all Arm
1183 exceptions by manipulating individual registers using the KVM_SET_ONE_REG API.
1185 See KVM_GET_VCPU_EVENTS for the data structure.
1188 4.33 KVM_GET_DEBUGREGS
1189 ----------------------
1191 :Capability: KVM_CAP_DEBUGREGS
1194 :Parameters: struct kvm_debugregs (out)
1195 :Returns: 0 on success, -1 on error
1197 Reads debug registers from the vcpu.
1201 struct kvm_debugregs {
1210 4.34 KVM_SET_DEBUGREGS
1211 ----------------------
1213 :Capability: KVM_CAP_DEBUGREGS
1216 :Parameters: struct kvm_debugregs (in)
1217 :Returns: 0 on success, -1 on error
1219 Writes debug registers into the vcpu.
1221 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
1222 yet and must be cleared on entry.
1225 4.35 KVM_SET_USER_MEMORY_REGION
1226 -------------------------------
1228 :Capability: KVM_CAP_USER_MEMORY
1231 :Parameters: struct kvm_userspace_memory_region (in)
1232 :Returns: 0 on success, -1 on error
1236 struct kvm_userspace_memory_region {
1239 __u64 guest_phys_addr;
1240 __u64 memory_size; /* bytes */
1241 __u64 userspace_addr; /* start of the userspace allocated memory */
1244 /* for kvm_memory_region::flags */
1245 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
1246 #define KVM_MEM_READONLY (1UL << 1)
1248 This ioctl allows the user to create, modify or delete a guest physical
1249 memory slot. Bits 0-15 of "slot" specify the slot id and this value
1250 should be less than the maximum number of user memory slots supported per
1251 VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS.
1252 Slots may not overlap in guest physical address space.
1254 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot"
1255 specifies the address space which is being modified. They must be
1256 less than the value that KVM_CHECK_EXTENSION returns for the
1257 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces
1258 are unrelated; the restriction on overlapping slots only applies within
1261 Deleting a slot is done by passing zero for memory_size. When changing
1262 an existing slot, it may be moved in the guest physical memory space,
1263 or its flags may be modified, but it may not be resized.
1265 Memory for the region is taken starting at the address denoted by the
1266 field userspace_addr, which must point at user addressable memory for
1267 the entire memory slot size. Any object may back this memory, including
1268 anonymous memory, ordinary files, and hugetlbfs.
1270 On architectures that support a form of address tagging, userspace_addr must
1271 be an untagged address.
1273 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
1274 be identical. This allows large pages in the guest to be backed by large
1277 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
1278 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
1279 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
1280 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
1281 to make a new slot read-only. In this case, writes to this memory will be
1282 posted to userspace as KVM_EXIT_MMIO exits.
1284 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
1285 the memory region are automatically reflected into the guest. For example, an
1286 mmap() that affects the region will be made visible immediately. Another
1287 example is madvise(MADV_DROP).
1289 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl.
1290 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory
1291 allocation and is deprecated.
1294 4.36 KVM_SET_TSS_ADDR
1295 ---------------------
1297 :Capability: KVM_CAP_SET_TSS_ADDR
1300 :Parameters: unsigned long tss_address (in)
1301 :Returns: 0 on success, -1 on error
1303 This ioctl defines the physical address of a three-page region in the guest
1304 physical address space. The region must be within the first 4GB of the
1305 guest physical address space and must not conflict with any memory slot
1306 or any mmio address. The guest may malfunction if it accesses this memory
1309 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1310 because of a quirk in the virtualization implementation (see the internals
1311 documentation when it pops into existence).
1317 :Capability: KVM_CAP_ENABLE_CAP
1318 :Architectures: mips, ppc, s390
1320 :Parameters: struct kvm_enable_cap (in)
1321 :Returns: 0 on success; -1 on error
1323 :Capability: KVM_CAP_ENABLE_CAP_VM
1326 :Parameters: struct kvm_enable_cap (in)
1327 :Returns: 0 on success; -1 on error
1331 Not all extensions are enabled by default. Using this ioctl the application
1332 can enable an extension, making it available to the guest.
1334 On systems that do not support this ioctl, it always fails. On systems that
1335 do support it, it only works for extensions that are supported for enablement.
1337 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
1342 struct kvm_enable_cap {
1346 The capability that is supposed to get enabled.
1352 A bitfield indicating future enhancements. Has to be 0 for now.
1358 Arguments for enabling a feature. If a feature needs initial values to
1359 function properly, this is the place to put them.
1366 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
1367 for vm-wide capabilities.
1369 4.38 KVM_GET_MP_STATE
1370 ---------------------
1372 :Capability: KVM_CAP_MP_STATE
1373 :Architectures: x86, s390, arm, arm64
1375 :Parameters: struct kvm_mp_state (out)
1376 :Returns: 0 on success; -1 on error
1380 struct kvm_mp_state {
1384 Returns the vcpu's current "multiprocessing state" (though also valid on
1385 uniprocessor guests).
1387 Possible values are:
1389 ========================== ===============================================
1390 KVM_MP_STATE_RUNNABLE the vcpu is currently running [x86,arm/arm64]
1391 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP)
1392 which has not yet received an INIT signal [x86]
1393 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is
1394 now ready for a SIPI [x86]
1395 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and
1396 is waiting for an interrupt [x86]
1397 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector
1398 accessible via KVM_GET_VCPU_EVENTS) [x86]
1399 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm/arm64]
1400 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390]
1401 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted)
1403 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state
1405 ========================== ===============================================
1407 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1408 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1409 these architectures.
1414 The only states that are valid are KVM_MP_STATE_STOPPED and
1415 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
1417 4.39 KVM_SET_MP_STATE
1418 ---------------------
1420 :Capability: KVM_CAP_MP_STATE
1421 :Architectures: x86, s390, arm, arm64
1423 :Parameters: struct kvm_mp_state (in)
1424 :Returns: 0 on success; -1 on error
1426 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1429 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1430 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1431 these architectures.
1436 The only states that are valid are KVM_MP_STATE_STOPPED and
1437 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
1439 4.40 KVM_SET_IDENTITY_MAP_ADDR
1440 ------------------------------
1442 :Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1445 :Parameters: unsigned long identity (in)
1446 :Returns: 0 on success, -1 on error
1448 This ioctl defines the physical address of a one-page region in the guest
1449 physical address space. The region must be within the first 4GB of the
1450 guest physical address space and must not conflict with any memory slot
1451 or any mmio address. The guest may malfunction if it accesses this memory
1454 Setting the address to 0 will result in resetting the address to its default
1457 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1458 because of a quirk in the virtualization implementation (see the internals
1459 documentation when it pops into existence).
1461 Fails if any VCPU has already been created.
1463 4.41 KVM_SET_BOOT_CPU_ID
1464 ------------------------
1466 :Capability: KVM_CAP_SET_BOOT_CPU_ID
1469 :Parameters: unsigned long vcpu_id
1470 :Returns: 0 on success, -1 on error
1472 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1473 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1480 :Capability: KVM_CAP_XSAVE
1483 :Parameters: struct kvm_xsave (out)
1484 :Returns: 0 on success, -1 on error
1493 This ioctl would copy current vcpu's xsave struct to the userspace.
1499 :Capability: KVM_CAP_XSAVE
1502 :Parameters: struct kvm_xsave (in)
1503 :Returns: 0 on success, -1 on error
1512 This ioctl would copy userspace's xsave struct to the kernel.
1518 :Capability: KVM_CAP_XCRS
1521 :Parameters: struct kvm_xcrs (out)
1522 :Returns: 0 on success, -1 on error
1535 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1539 This ioctl would copy current vcpu's xcrs to the userspace.
1545 :Capability: KVM_CAP_XCRS
1548 :Parameters: struct kvm_xcrs (in)
1549 :Returns: 0 on success, -1 on error
1562 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1566 This ioctl would set vcpu's xcr to the value userspace specified.
1569 4.46 KVM_GET_SUPPORTED_CPUID
1570 ----------------------------
1572 :Capability: KVM_CAP_EXT_CPUID
1575 :Parameters: struct kvm_cpuid2 (in/out)
1576 :Returns: 0 on success, -1 on error
1583 struct kvm_cpuid_entry2 entries[0];
1586 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1587 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */
1588 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */
1590 struct kvm_cpuid_entry2 {
1601 This ioctl returns x86 cpuid features which are supported by both the
1602 hardware and kvm in its default configuration. Userspace can use the
1603 information returned by this ioctl to construct cpuid information (for
1604 KVM_SET_CPUID2) that is consistent with hardware, kernel, and
1605 userspace capabilities, and with user requirements (for example, the
1606 user may wish to constrain cpuid to emulate older hardware, or for
1607 feature consistency across a cluster).
1609 Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may
1610 expose cpuid features (e.g. MONITOR) which are not supported by kvm in
1611 its default configuration. If userspace enables such capabilities, it
1612 is responsible for modifying the results of this ioctl appropriately.
1614 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1615 with the 'nent' field indicating the number of entries in the variable-size
1616 array 'entries'. If the number of entries is too low to describe the cpu
1617 capabilities, an error (E2BIG) is returned. If the number is too high,
1618 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1619 number is just right, the 'nent' field is adjusted to the number of valid
1620 entries in the 'entries' array, which is then filled.
1622 The entries returned are the host cpuid as returned by the cpuid instruction,
1623 with unknown or unsupported features masked out. Some features (for example,
1624 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1625 emulate them efficiently. The fields in each entry are defined as follows:
1628 the eax value used to obtain the entry
1631 the ecx value used to obtain the entry (for entries that are
1635 an OR of zero or more of the following:
1637 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1638 if the index field is valid
1641 the values returned by the cpuid instruction for
1642 this function/index combination
1644 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1645 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1646 support. Instead it is reported via::
1648 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1650 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1651 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1654 4.47 KVM_PPC_GET_PVINFO
1655 -----------------------
1657 :Capability: KVM_CAP_PPC_GET_PVINFO
1660 :Parameters: struct kvm_ppc_pvinfo (out)
1661 :Returns: 0 on success, !0 on error
1665 struct kvm_ppc_pvinfo {
1671 This ioctl fetches PV specific information that need to be passed to the guest
1672 using the device tree or other means from vm context.
1674 The hcall array defines 4 instructions that make up a hypercall.
1676 If any additional field gets added to this structure later on, a bit for that
1677 additional piece of information will be set in the flags bitmap.
1679 The flags bitmap is defined as::
1681 /* the host supports the ePAPR idle hcall
1682 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1684 4.52 KVM_SET_GSI_ROUTING
1685 ------------------------
1687 :Capability: KVM_CAP_IRQ_ROUTING
1688 :Architectures: x86 s390 arm arm64
1690 :Parameters: struct kvm_irq_routing (in)
1691 :Returns: 0 on success, -1 on error
1693 Sets the GSI routing table entries, overwriting any previously set entries.
1695 On arm/arm64, GSI routing has the following limitation:
1697 - GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD.
1701 struct kvm_irq_routing {
1704 struct kvm_irq_routing_entry entries[0];
1707 No flags are specified so far, the corresponding field must be set to zero.
1711 struct kvm_irq_routing_entry {
1717 struct kvm_irq_routing_irqchip irqchip;
1718 struct kvm_irq_routing_msi msi;
1719 struct kvm_irq_routing_s390_adapter adapter;
1720 struct kvm_irq_routing_hv_sint hv_sint;
1725 /* gsi routing entry types */
1726 #define KVM_IRQ_ROUTING_IRQCHIP 1
1727 #define KVM_IRQ_ROUTING_MSI 2
1728 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1729 #define KVM_IRQ_ROUTING_HV_SINT 4
1733 - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry
1734 type, specifies that the devid field contains a valid value. The per-VM
1735 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
1736 the device ID. If this capability is not available, userspace should
1737 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
1742 struct kvm_irq_routing_irqchip {
1747 struct kvm_irq_routing_msi {
1757 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
1758 for the device that wrote the MSI message. For PCI, this is usually a
1759 BFD identifier in the lower 16 bits.
1761 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
1762 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
1763 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
1764 address_hi must be zero.
1768 struct kvm_irq_routing_s390_adapter {
1772 __u32 summary_offset;
1776 struct kvm_irq_routing_hv_sint {
1782 4.55 KVM_SET_TSC_KHZ
1783 --------------------
1785 :Capability: KVM_CAP_TSC_CONTROL
1788 :Parameters: virtual tsc_khz
1789 :Returns: 0 on success, -1 on error
1791 Specifies the tsc frequency for the virtual machine. The unit of the
1795 4.56 KVM_GET_TSC_KHZ
1796 --------------------
1798 :Capability: KVM_CAP_GET_TSC_KHZ
1802 :Returns: virtual tsc-khz on success, negative value on error
1804 Returns the tsc frequency of the guest. The unit of the return value is
1805 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1812 :Capability: KVM_CAP_IRQCHIP
1815 :Parameters: struct kvm_lapic_state (out)
1816 :Returns: 0 on success, -1 on error
1820 #define KVM_APIC_REG_SIZE 0x400
1821 struct kvm_lapic_state {
1822 char regs[KVM_APIC_REG_SIZE];
1825 Reads the Local APIC registers and copies them into the input argument. The
1826 data format and layout are the same as documented in the architecture manual.
1828 If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is
1829 enabled, then the format of APIC_ID register depends on the APIC mode
1830 (reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in
1831 the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID
1832 which is stored in bits 31-24 of the APIC register, or equivalently in
1833 byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then
1834 be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR.
1836 If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state
1837 always uses xAPIC format.
1843 :Capability: KVM_CAP_IRQCHIP
1846 :Parameters: struct kvm_lapic_state (in)
1847 :Returns: 0 on success, -1 on error
1851 #define KVM_APIC_REG_SIZE 0x400
1852 struct kvm_lapic_state {
1853 char regs[KVM_APIC_REG_SIZE];
1856 Copies the input argument into the Local APIC registers. The data format
1857 and layout are the same as documented in the architecture manual.
1859 The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's
1860 regs field) depends on the state of the KVM_CAP_X2APIC_API capability.
1861 See the note in KVM_GET_LAPIC.
1867 :Capability: KVM_CAP_IOEVENTFD
1870 :Parameters: struct kvm_ioeventfd (in)
1871 :Returns: 0 on success, !0 on error
1873 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
1874 within the guest. A guest write in the registered address will signal the
1875 provided event instead of triggering an exit.
1879 struct kvm_ioeventfd {
1881 __u64 addr; /* legal pio/mmio address */
1882 __u32 len; /* 0, 1, 2, 4, or 8 bytes */
1888 For the special case of virtio-ccw devices on s390, the ioevent is matched
1889 to a subchannel/virtqueue tuple instead.
1891 The following flags are defined::
1893 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
1894 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
1895 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
1896 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
1897 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
1899 If datamatch flag is set, the event will be signaled only if the written value
1900 to the registered address is equal to datamatch in struct kvm_ioeventfd.
1902 For virtio-ccw devices, addr contains the subchannel id and datamatch the
1905 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
1906 the kernel will ignore the length of guest write and may get a faster vmexit.
1907 The speedup may only apply to specific architectures, but the ioeventfd will
1913 :Capability: KVM_CAP_SW_TLB
1916 :Parameters: struct kvm_dirty_tlb (in)
1917 :Returns: 0 on success, -1 on error
1921 struct kvm_dirty_tlb {
1926 This must be called whenever userspace has changed an entry in the shared
1927 TLB, prior to calling KVM_RUN on the associated vcpu.
1929 The "bitmap" field is the userspace address of an array. This array
1930 consists of a number of bits, equal to the total number of TLB entries as
1931 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
1932 nearest multiple of 64.
1934 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
1937 The array is little-endian: the bit 0 is the least significant bit of the
1938 first byte, bit 8 is the least significant bit of the second byte, etc.
1939 This avoids any complications with differing word sizes.
1941 The "num_dirty" field is a performance hint for KVM to determine whether it
1942 should skip processing the bitmap and just invalidate everything. It must
1943 be set to the number of set bits in the bitmap.
1946 4.62 KVM_CREATE_SPAPR_TCE
1947 -------------------------
1949 :Capability: KVM_CAP_SPAPR_TCE
1950 :Architectures: powerpc
1952 :Parameters: struct kvm_create_spapr_tce (in)
1953 :Returns: file descriptor for manipulating the created TCE table
1955 This creates a virtual TCE (translation control entry) table, which
1956 is an IOMMU for PAPR-style virtual I/O. It is used to translate
1957 logical addresses used in virtual I/O into guest physical addresses,
1958 and provides a scatter/gather capability for PAPR virtual I/O.
1962 /* for KVM_CAP_SPAPR_TCE */
1963 struct kvm_create_spapr_tce {
1968 The liobn field gives the logical IO bus number for which to create a
1969 TCE table. The window_size field specifies the size of the DMA window
1970 which this TCE table will translate - the table will contain one 64
1971 bit TCE entry for every 4kiB of the DMA window.
1973 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
1974 table has been created using this ioctl(), the kernel will handle it
1975 in real mode, updating the TCE table. H_PUT_TCE calls for other
1976 liobns will cause a vm exit and must be handled by userspace.
1978 The return value is a file descriptor which can be passed to mmap(2)
1979 to map the created TCE table into userspace. This lets userspace read
1980 the entries written by kernel-handled H_PUT_TCE calls, and also lets
1981 userspace update the TCE table directly which is useful in some
1985 4.63 KVM_ALLOCATE_RMA
1986 ---------------------
1988 :Capability: KVM_CAP_PPC_RMA
1989 :Architectures: powerpc
1991 :Parameters: struct kvm_allocate_rma (out)
1992 :Returns: file descriptor for mapping the allocated RMA
1994 This allocates a Real Mode Area (RMA) from the pool allocated at boot
1995 time by the kernel. An RMA is a physically-contiguous, aligned region
1996 of memory used on older POWER processors to provide the memory which
1997 will be accessed by real-mode (MMU off) accesses in a KVM guest.
1998 POWER processors support a set of sizes for the RMA that usually
1999 includes 64MB, 128MB, 256MB and some larger powers of two.
2003 /* for KVM_ALLOCATE_RMA */
2004 struct kvm_allocate_rma {
2008 The return value is a file descriptor which can be passed to mmap(2)
2009 to map the allocated RMA into userspace. The mapped area can then be
2010 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
2011 RMA for a virtual machine. The size of the RMA in bytes (which is
2012 fixed at host kernel boot time) is returned in the rma_size field of
2013 the argument structure.
2015 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
2016 is supported; 2 if the processor requires all virtual machines to have
2017 an RMA, or 1 if the processor can use an RMA but doesn't require it,
2018 because it supports the Virtual RMA (VRMA) facility.
2024 :Capability: KVM_CAP_USER_NMI
2028 :Returns: 0 on success, -1 on error
2030 Queues an NMI on the thread's vcpu. Note this is well defined only
2031 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
2032 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
2033 has been called, this interface is completely emulated within the kernel.
2035 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
2036 following algorithm:
2039 - read the local APIC's state (KVM_GET_LAPIC)
2040 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
2041 - if so, issue KVM_NMI
2044 Some guests configure the LINT1 NMI input to cause a panic, aiding in
2048 4.65 KVM_S390_UCAS_MAP
2049 ----------------------
2051 :Capability: KVM_CAP_S390_UCONTROL
2052 :Architectures: s390
2054 :Parameters: struct kvm_s390_ucas_mapping (in)
2055 :Returns: 0 in case of success
2057 The parameter is defined like this::
2059 struct kvm_s390_ucas_mapping {
2065 This ioctl maps the memory at "user_addr" with the length "length" to
2066 the vcpu's address space starting at "vcpu_addr". All parameters need to
2067 be aligned by 1 megabyte.
2070 4.66 KVM_S390_UCAS_UNMAP
2071 ------------------------
2073 :Capability: KVM_CAP_S390_UCONTROL
2074 :Architectures: s390
2076 :Parameters: struct kvm_s390_ucas_mapping (in)
2077 :Returns: 0 in case of success
2079 The parameter is defined like this::
2081 struct kvm_s390_ucas_mapping {
2087 This ioctl unmaps the memory in the vcpu's address space starting at
2088 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
2089 All parameters need to be aligned by 1 megabyte.
2092 4.67 KVM_S390_VCPU_FAULT
2093 ------------------------
2095 :Capability: KVM_CAP_S390_UCONTROL
2096 :Architectures: s390
2098 :Parameters: vcpu absolute address (in)
2099 :Returns: 0 in case of success
2101 This call creates a page table entry on the virtual cpu's address space
2102 (for user controlled virtual machines) or the virtual machine's address
2103 space (for regular virtual machines). This only works for minor faults,
2104 thus it's recommended to access subject memory page via the user page
2105 table upfront. This is useful to handle validity intercepts for user
2106 controlled virtual machines to fault in the virtual cpu's lowcore pages
2107 prior to calling the KVM_RUN ioctl.
2110 4.68 KVM_SET_ONE_REG
2111 --------------------
2113 :Capability: KVM_CAP_ONE_REG
2116 :Parameters: struct kvm_one_reg (in)
2117 :Returns: 0 on success, negative value on failure
2121 ====== ============================================================
2122 Â ENOENT Â Â no such register
2123 Â EINVAL Â Â invalid register ID, or no such register or used with VMs in
2124 protected virtualization mode on s390
2125 Â EPERM Â Â Â (arm64) register access not allowed before vcpu finalization
2126 ====== ============================================================
2128 (These error codes are indicative only: do not rely on a specific error
2129 code being returned in a specific situation.)
2133 struct kvm_one_reg {
2138 Using this ioctl, a single vcpu register can be set to a specific value
2139 defined by user space with the passed in struct kvm_one_reg, where id
2140 refers to the register identifier as described below and addr is a pointer
2141 to a variable with the respective size. There can be architecture agnostic
2142 and architecture specific registers. Each have their own range of operation
2143 and their own constants and width. To keep track of the implemented
2144 registers, find a list below:
2146 ======= =============================== ============
2147 Arch Register Width (bits)
2148 ======= =============================== ============
2149 PPC KVM_REG_PPC_HIOR 64
2150 PPC KVM_REG_PPC_IAC1 64
2151 PPC KVM_REG_PPC_IAC2 64
2152 PPC KVM_REG_PPC_IAC3 64
2153 PPC KVM_REG_PPC_IAC4 64
2154 PPC KVM_REG_PPC_DAC1 64
2155 PPC KVM_REG_PPC_DAC2 64
2156 PPC KVM_REG_PPC_DABR 64
2157 PPC KVM_REG_PPC_DSCR 64
2158 PPC KVM_REG_PPC_PURR 64
2159 PPC KVM_REG_PPC_SPURR 64
2160 PPC KVM_REG_PPC_DAR 64
2161 PPC KVM_REG_PPC_DSISR 32
2162 PPC KVM_REG_PPC_AMR 64
2163 PPC KVM_REG_PPC_UAMOR 64
2164 PPC KVM_REG_PPC_MMCR0 64
2165 PPC KVM_REG_PPC_MMCR1 64
2166 PPC KVM_REG_PPC_MMCRA 64
2167 PPC KVM_REG_PPC_MMCR2 64
2168 PPC KVM_REG_PPC_MMCRS 64
2169 PPC KVM_REG_PPC_MMCR3 64
2170 PPC KVM_REG_PPC_SIAR 64
2171 PPC KVM_REG_PPC_SDAR 64
2172 PPC KVM_REG_PPC_SIER 64
2173 PPC KVM_REG_PPC_SIER2 64
2174 PPC KVM_REG_PPC_SIER3 64
2175 PPC KVM_REG_PPC_PMC1 32
2176 PPC KVM_REG_PPC_PMC2 32
2177 PPC KVM_REG_PPC_PMC3 32
2178 PPC KVM_REG_PPC_PMC4 32
2179 PPC KVM_REG_PPC_PMC5 32
2180 PPC KVM_REG_PPC_PMC6 32
2181 PPC KVM_REG_PPC_PMC7 32
2182 PPC KVM_REG_PPC_PMC8 32
2183 PPC KVM_REG_PPC_FPR0 64
2185 PPC KVM_REG_PPC_FPR31 64
2186 PPC KVM_REG_PPC_VR0 128
2188 PPC KVM_REG_PPC_VR31 128
2189 PPC KVM_REG_PPC_VSR0 128
2191 PPC KVM_REG_PPC_VSR31 128
2192 PPC KVM_REG_PPC_FPSCR 64
2193 PPC KVM_REG_PPC_VSCR 32
2194 PPC KVM_REG_PPC_VPA_ADDR 64
2195 PPC KVM_REG_PPC_VPA_SLB 128
2196 PPC KVM_REG_PPC_VPA_DTL 128
2197 PPC KVM_REG_PPC_EPCR 32
2198 PPC KVM_REG_PPC_EPR 32
2199 PPC KVM_REG_PPC_TCR 32
2200 PPC KVM_REG_PPC_TSR 32
2201 PPC KVM_REG_PPC_OR_TSR 32
2202 PPC KVM_REG_PPC_CLEAR_TSR 32
2203 PPC KVM_REG_PPC_MAS0 32
2204 PPC KVM_REG_PPC_MAS1 32
2205 PPC KVM_REG_PPC_MAS2 64
2206 PPC KVM_REG_PPC_MAS7_3 64
2207 PPC KVM_REG_PPC_MAS4 32
2208 PPC KVM_REG_PPC_MAS6 32
2209 PPC KVM_REG_PPC_MMUCFG 32
2210 PPC KVM_REG_PPC_TLB0CFG 32
2211 PPC KVM_REG_PPC_TLB1CFG 32
2212 PPC KVM_REG_PPC_TLB2CFG 32
2213 PPC KVM_REG_PPC_TLB3CFG 32
2214 PPC KVM_REG_PPC_TLB0PS 32
2215 PPC KVM_REG_PPC_TLB1PS 32
2216 PPC KVM_REG_PPC_TLB2PS 32
2217 PPC KVM_REG_PPC_TLB3PS 32
2218 PPC KVM_REG_PPC_EPTCFG 32
2219 PPC KVM_REG_PPC_ICP_STATE 64
2220 PPC KVM_REG_PPC_VP_STATE 128
2221 PPC KVM_REG_PPC_TB_OFFSET 64
2222 PPC KVM_REG_PPC_SPMC1 32
2223 PPC KVM_REG_PPC_SPMC2 32
2224 PPC KVM_REG_PPC_IAMR 64
2225 PPC KVM_REG_PPC_TFHAR 64
2226 PPC KVM_REG_PPC_TFIAR 64
2227 PPC KVM_REG_PPC_TEXASR 64
2228 PPC KVM_REG_PPC_FSCR 64
2229 PPC KVM_REG_PPC_PSPB 32
2230 PPC KVM_REG_PPC_EBBHR 64
2231 PPC KVM_REG_PPC_EBBRR 64
2232 PPC KVM_REG_PPC_BESCR 64
2233 PPC KVM_REG_PPC_TAR 64
2234 PPC KVM_REG_PPC_DPDES 64
2235 PPC KVM_REG_PPC_DAWR 64
2236 PPC KVM_REG_PPC_DAWRX 64
2237 PPC KVM_REG_PPC_CIABR 64
2238 PPC KVM_REG_PPC_IC 64
2239 PPC KVM_REG_PPC_VTB 64
2240 PPC KVM_REG_PPC_CSIGR 64
2241 PPC KVM_REG_PPC_TACR 64
2242 PPC KVM_REG_PPC_TCSCR 64
2243 PPC KVM_REG_PPC_PID 64
2244 PPC KVM_REG_PPC_ACOP 64
2245 PPC KVM_REG_PPC_VRSAVE 32
2246 PPC KVM_REG_PPC_LPCR 32
2247 PPC KVM_REG_PPC_LPCR_64 64
2248 PPC KVM_REG_PPC_PPR 64
2249 PPC KVM_REG_PPC_ARCH_COMPAT 32
2250 PPC KVM_REG_PPC_DABRX 32
2251 PPC KVM_REG_PPC_WORT 64
2252 PPC KVM_REG_PPC_SPRG9 64
2253 PPC KVM_REG_PPC_DBSR 32
2254 PPC KVM_REG_PPC_TIDR 64
2255 PPC KVM_REG_PPC_PSSCR 64
2256 PPC KVM_REG_PPC_DEC_EXPIRY 64
2257 PPC KVM_REG_PPC_PTCR 64
2258 PPC KVM_REG_PPC_TM_GPR0 64
2260 PPC KVM_REG_PPC_TM_GPR31 64
2261 PPC KVM_REG_PPC_TM_VSR0 128
2263 PPC KVM_REG_PPC_TM_VSR63 128
2264 PPC KVM_REG_PPC_TM_CR 64
2265 PPC KVM_REG_PPC_TM_LR 64
2266 PPC KVM_REG_PPC_TM_CTR 64
2267 PPC KVM_REG_PPC_TM_FPSCR 64
2268 PPC KVM_REG_PPC_TM_AMR 64
2269 PPC KVM_REG_PPC_TM_PPR 64
2270 PPC KVM_REG_PPC_TM_VRSAVE 64
2271 PPC KVM_REG_PPC_TM_VSCR 32
2272 PPC KVM_REG_PPC_TM_DSCR 64
2273 PPC KVM_REG_PPC_TM_TAR 64
2274 PPC KVM_REG_PPC_TM_XER 64
2276 MIPS KVM_REG_MIPS_R0 64
2278 MIPS KVM_REG_MIPS_R31 64
2279 MIPS KVM_REG_MIPS_HI 64
2280 MIPS KVM_REG_MIPS_LO 64
2281 MIPS KVM_REG_MIPS_PC 64
2282 MIPS KVM_REG_MIPS_CP0_INDEX 32
2283 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64
2284 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64
2285 MIPS KVM_REG_MIPS_CP0_CONTEXT 64
2286 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32
2287 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64
2288 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64
2289 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32
2290 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32
2291 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64
2292 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64
2293 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64
2294 MIPS KVM_REG_MIPS_CP0_PWBASE 64
2295 MIPS KVM_REG_MIPS_CP0_PWFIELD 64
2296 MIPS KVM_REG_MIPS_CP0_PWSIZE 64
2297 MIPS KVM_REG_MIPS_CP0_WIRED 32
2298 MIPS KVM_REG_MIPS_CP0_PWCTL 32
2299 MIPS KVM_REG_MIPS_CP0_HWRENA 32
2300 MIPS KVM_REG_MIPS_CP0_BADVADDR 64
2301 MIPS KVM_REG_MIPS_CP0_BADINSTR 32
2302 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32
2303 MIPS KVM_REG_MIPS_CP0_COUNT 32
2304 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64
2305 MIPS KVM_REG_MIPS_CP0_COMPARE 32
2306 MIPS KVM_REG_MIPS_CP0_STATUS 32
2307 MIPS KVM_REG_MIPS_CP0_INTCTL 32
2308 MIPS KVM_REG_MIPS_CP0_CAUSE 32
2309 MIPS KVM_REG_MIPS_CP0_EPC 64
2310 MIPS KVM_REG_MIPS_CP0_PRID 32
2311 MIPS KVM_REG_MIPS_CP0_EBASE 64
2312 MIPS KVM_REG_MIPS_CP0_CONFIG 32
2313 MIPS KVM_REG_MIPS_CP0_CONFIG1 32
2314 MIPS KVM_REG_MIPS_CP0_CONFIG2 32
2315 MIPS KVM_REG_MIPS_CP0_CONFIG3 32
2316 MIPS KVM_REG_MIPS_CP0_CONFIG4 32
2317 MIPS KVM_REG_MIPS_CP0_CONFIG5 32
2318 MIPS KVM_REG_MIPS_CP0_CONFIG7 32
2319 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64
2320 MIPS KVM_REG_MIPS_CP0_ERROREPC 64
2321 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64
2322 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64
2323 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64
2324 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64
2325 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64
2326 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64
2327 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64
2328 MIPS KVM_REG_MIPS_COUNT_CTL 64
2329 MIPS KVM_REG_MIPS_COUNT_RESUME 64
2330 MIPS KVM_REG_MIPS_COUNT_HZ 64
2331 MIPS KVM_REG_MIPS_FPR_32(0..31) 32
2332 MIPS KVM_REG_MIPS_FPR_64(0..31) 64
2333 MIPS KVM_REG_MIPS_VEC_128(0..31) 128
2334 MIPS KVM_REG_MIPS_FCR_IR 32
2335 MIPS KVM_REG_MIPS_FCR_CSR 32
2336 MIPS KVM_REG_MIPS_MSA_IR 32
2337 MIPS KVM_REG_MIPS_MSA_CSR 32
2338 ======= =============================== ============
2340 ARM registers are mapped using the lower 32 bits. The upper 16 of that
2341 is the register group type, or coprocessor number:
2343 ARM core registers have the following id bit patterns::
2345 0x4020 0000 0010 <index into the kvm_regs struct:16>
2347 ARM 32-bit CP15 registers have the following id bit patterns::
2349 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
2351 ARM 64-bit CP15 registers have the following id bit patterns::
2353 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
2355 ARM CCSIDR registers are demultiplexed by CSSELR value::
2357 0x4020 0000 0011 00 <csselr:8>
2359 ARM 32-bit VFP control registers have the following id bit patterns::
2361 0x4020 0000 0012 1 <regno:12>
2363 ARM 64-bit FP registers have the following id bit patterns::
2365 0x4030 0000 0012 0 <regno:12>
2367 ARM firmware pseudo-registers have the following bit pattern::
2369 0x4030 0000 0014 <regno:16>
2372 arm64 registers are mapped using the lower 32 bits. The upper 16 of
2373 that is the register group type, or coprocessor number:
2375 arm64 core/FP-SIMD registers have the following id bit patterns. Note
2376 that the size of the access is variable, as the kvm_regs structure
2377 contains elements ranging from 32 to 128 bits. The index is a 32bit
2378 value in the kvm_regs structure seen as a 32bit array::
2380 0x60x0 0000 0010 <index into the kvm_regs struct:16>
2384 ======================= ========= ===== =======================================
2385 Encoding Register Bits kvm_regs member
2386 ======================= ========= ===== =======================================
2387 0x6030 0000 0010 0000 X0 64 regs.regs[0]
2388 0x6030 0000 0010 0002 X1 64 regs.regs[1]
2390 0x6030 0000 0010 003c X30 64 regs.regs[30]
2391 0x6030 0000 0010 003e SP 64 regs.sp
2392 0x6030 0000 0010 0040 PC 64 regs.pc
2393 0x6030 0000 0010 0042 PSTATE 64 regs.pstate
2394 0x6030 0000 0010 0044 SP_EL1 64 sp_el1
2395 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1
2396 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC)
2397 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT]
2398 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND]
2399 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ]
2400 0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ]
2401 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_
2402 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_
2404 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_
2405 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr
2406 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr
2407 ======================= ========= ===== =======================================
2409 .. [1] These encodings are not accepted for SVE-enabled vcpus. See
2412 The equivalent register content can be accessed via bits [127:0] of
2413 the corresponding SVE Zn registers instead for vcpus that have SVE
2414 enabled (see below).
2416 arm64 CCSIDR registers are demultiplexed by CSSELR value::
2418 0x6020 0000 0011 00 <csselr:8>
2420 arm64 system registers have the following id bit patterns::
2422 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
2426 Two system register IDs do not follow the specified pattern. These
2427 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to
2428 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These
2429 two had their values accidentally swapped, which means TIMER_CVAL is
2430 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is
2431 derived from the register encoding for CNTV_CVAL_EL0. As this is
2432 API, it must remain this way.
2434 arm64 firmware pseudo-registers have the following bit pattern::
2436 0x6030 0000 0014 <regno:16>
2438 arm64 SVE registers have the following bit patterns::
2440 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice]
2441 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice]
2442 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice]
2443 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register
2445 Access to register IDs where 2048 * slice >= 128 * max_vq will fail with
2446 ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit
2447 quadwords: see [2]_ below.
2449 These registers are only accessible on vcpus for which SVE is enabled.
2450 See KVM_ARM_VCPU_INIT for details.
2452 In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not
2453 accessible until the vcpu's SVE configuration has been finalized
2454 using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT
2455 and KVM_ARM_VCPU_FINALIZE for more information about this procedure.
2457 KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector
2458 lengths supported by the vcpu to be discovered and configured by
2459 userspace. When transferred to or from user memory via KVM_GET_ONE_REG
2460 or KVM_SET_ONE_REG, the value of this register is of type
2461 __u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as
2464 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS];
2466 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX &&
2467 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >>
2468 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1))
2469 /* Vector length vq * 16 bytes supported */
2471 /* Vector length vq * 16 bytes not supported */
2473 .. [2] The maximum value vq for which the above condition is true is
2474 max_vq. This is the maximum vector length available to the guest on
2475 this vcpu, and determines which register slices are visible through
2476 this ioctl interface.
2478 (See Documentation/arm64/sve.rst for an explanation of the "vq"
2481 KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT.
2482 KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that
2485 Userspace may subsequently modify it if desired until the vcpu's SVE
2486 configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE).
2488 Apart from simply removing all vector lengths from the host set that
2489 exceed some value, support for arbitrarily chosen sets of vector lengths
2490 is hardware-dependent and may not be available. Attempting to configure
2491 an invalid set of vector lengths via KVM_SET_ONE_REG will fail with
2494 After the vcpu's SVE configuration is finalized, further attempts to
2495 write this register will fail with EPERM.
2498 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
2499 the register group type:
2501 MIPS core registers (see above) have the following id bit patterns::
2503 0x7030 0000 0000 <reg:16>
2505 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
2506 patterns depending on whether they're 32-bit or 64-bit registers::
2508 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
2509 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2511 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64
2512 versions of the EntryLo registers regardless of the word size of the host
2513 hardware, host kernel, guest, and whether XPA is present in the guest, i.e.
2514 with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and
2515 the PFNX field starting at bit 30.
2517 MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit
2520 0x7030 0000 0001 01 <reg:8>
2522 MIPS KVM control registers (see above) have the following id bit patterns::
2524 0x7030 0000 0002 <reg:16>
2526 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
2527 id bit patterns depending on the size of the register being accessed. They are
2528 always accessed according to the current guest FPU mode (Status.FR and
2529 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
2530 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector
2531 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they
2532 overlap the FPU registers::
2534 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
2535 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
2536 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers)
2538 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
2539 following id bit patterns::
2541 0x7020 0000 0003 01 <0:3> <reg:5>
2543 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the
2544 following id bit patterns::
2546 0x7020 0000 0003 02 <0:3> <reg:5>
2549 4.69 KVM_GET_ONE_REG
2550 --------------------
2552 :Capability: KVM_CAP_ONE_REG
2555 :Parameters: struct kvm_one_reg (in and out)
2556 :Returns: 0 on success, negative value on failure
2560 ======== ============================================================
2561 Â ENOENT Â Â no such register
2562 Â EINVAL Â Â invalid register ID, or no such register or used with VMs in
2563 protected virtualization mode on s390
2564 Â EPERM Â Â Â (arm64) register access not allowed before vcpu finalization
2565 ======== ============================================================
2567 (These error codes are indicative only: do not rely on a specific error
2568 code being returned in a specific situation.)
2570 This ioctl allows to receive the value of a single register implemented
2571 in a vcpu. The register to read is indicated by the "id" field of the
2572 kvm_one_reg struct passed in. On success, the register value can be found
2573 at the memory location pointed to by "addr".
2575 The list of registers accessible using this interface is identical to the
2579 4.70 KVM_KVMCLOCK_CTRL
2580 ----------------------
2582 :Capability: KVM_CAP_KVMCLOCK_CTRL
2583 :Architectures: Any that implement pvclocks (currently x86 only)
2586 :Returns: 0 on success, -1 on error
2588 This ioctl sets a flag accessible to the guest indicating that the specified
2589 vCPU has been paused by the host userspace.
2591 The host will set a flag in the pvclock structure that is checked from the
2592 soft lockup watchdog. The flag is part of the pvclock structure that is
2593 shared between guest and host, specifically the second bit of the flags
2594 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2595 the host and read/cleared exclusively by the guest. The guest operation of
2596 checking and clearing the flag must be an atomic operation so
2597 load-link/store-conditional, or equivalent must be used. There are two cases
2598 where the guest will clear the flag: when the soft lockup watchdog timer resets
2599 itself or when a soft lockup is detected. This ioctl can be called any time
2600 after pausing the vcpu, but before it is resumed.
2606 :Capability: KVM_CAP_SIGNAL_MSI
2607 :Architectures: x86 arm arm64
2609 :Parameters: struct kvm_msi (in)
2610 :Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2612 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2627 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM
2628 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
2629 the device ID. If this capability is not available, userspace
2630 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
2632 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
2633 for the device that wrote the MSI message. For PCI, this is usually a
2634 BFD identifier in the lower 16 bits.
2636 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
2637 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
2638 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
2639 address_hi must be zero.
2642 4.71 KVM_CREATE_PIT2
2643 --------------------
2645 :Capability: KVM_CAP_PIT2
2648 :Parameters: struct kvm_pit_config (in)
2649 :Returns: 0 on success, -1 on error
2651 Creates an in-kernel device model for the i8254 PIT. This call is only valid
2652 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
2653 parameters have to be passed::
2655 struct kvm_pit_config {
2662 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
2664 PIT timer interrupts may use a per-VM kernel thread for injection. If it
2665 exists, this thread will have a name of the following pattern::
2667 kvm-pit/<owner-process-pid>
2669 When running a guest with elevated priorities, the scheduling parameters of
2670 this thread may have to be adjusted accordingly.
2672 This IOCTL replaces the obsolete KVM_CREATE_PIT.
2678 :Capability: KVM_CAP_PIT_STATE2
2681 :Parameters: struct kvm_pit_state2 (out)
2682 :Returns: 0 on success, -1 on error
2684 Retrieves the state of the in-kernel PIT model. Only valid after
2685 KVM_CREATE_PIT2. The state is returned in the following structure::
2687 struct kvm_pit_state2 {
2688 struct kvm_pit_channel_state channels[3];
2695 /* disable PIT in HPET legacy mode */
2696 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
2698 This IOCTL replaces the obsolete KVM_GET_PIT.
2704 :Capability: KVM_CAP_PIT_STATE2
2707 :Parameters: struct kvm_pit_state2 (in)
2708 :Returns: 0 on success, -1 on error
2710 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
2711 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
2713 This IOCTL replaces the obsolete KVM_SET_PIT.
2716 4.74 KVM_PPC_GET_SMMU_INFO
2717 --------------------------
2719 :Capability: KVM_CAP_PPC_GET_SMMU_INFO
2720 :Architectures: powerpc
2723 :Returns: 0 on success, -1 on error
2725 This populates and returns a structure describing the features of
2726 the "Server" class MMU emulation supported by KVM.
2727 This can in turn be used by userspace to generate the appropriate
2728 device-tree properties for the guest operating system.
2730 The structure contains some global information, followed by an
2731 array of supported segment page sizes::
2733 struct kvm_ppc_smmu_info {
2737 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
2740 The supported flags are:
2742 - KVM_PPC_PAGE_SIZES_REAL:
2743 When that flag is set, guest page sizes must "fit" the backing
2744 store page sizes. When not set, any page size in the list can
2745 be used regardless of how they are backed by userspace.
2747 - KVM_PPC_1T_SEGMENTS
2748 The emulated MMU supports 1T segments in addition to the
2752 This flag indicates that HPT guests are not supported by KVM,
2753 thus all guests must use radix MMU mode.
2755 The "slb_size" field indicates how many SLB entries are supported
2757 The "sps" array contains 8 entries indicating the supported base
2758 page sizes for a segment in increasing order. Each entry is defined
2761 struct kvm_ppc_one_seg_page_size {
2762 __u32 page_shift; /* Base page shift of segment (or 0) */
2763 __u32 slb_enc; /* SLB encoding for BookS */
2764 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
2767 An entry with a "page_shift" of 0 is unused. Because the array is
2768 organized in increasing order, a lookup can stop when encoutering
2771 The "slb_enc" field provides the encoding to use in the SLB for the
2772 page size. The bits are in positions such as the value can directly
2773 be OR'ed into the "vsid" argument of the slbmte instruction.
2775 The "enc" array is a list which for each of those segment base page
2776 size provides the list of supported actual page sizes (which can be
2777 only larger or equal to the base page size), along with the
2778 corresponding encoding in the hash PTE. Similarly, the array is
2779 8 entries sorted by increasing sizes and an entry with a "0" shift
2780 is an empty entry and a terminator::
2782 struct kvm_ppc_one_page_size {
2783 __u32 page_shift; /* Page shift (or 0) */
2784 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
2787 The "pte_enc" field provides a value that can OR'ed into the hash
2788 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
2789 into the hash PTE second double word).
2794 :Capability: KVM_CAP_IRQFD
2795 :Architectures: x86 s390 arm arm64
2797 :Parameters: struct kvm_irqfd (in)
2798 :Returns: 0 on success, -1 on error
2800 Allows setting an eventfd to directly trigger a guest interrupt.
2801 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
2802 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
2803 an event is triggered on the eventfd, an interrupt is injected into
2804 the guest using the specified gsi pin. The irqfd is removed using
2805 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
2808 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
2809 mechanism allowing emulation of level-triggered, irqfd-based
2810 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
2811 additional eventfd in the kvm_irqfd.resamplefd field. When operating
2812 in resample mode, posting of an interrupt through kvm_irq.fd asserts
2813 the specified gsi in the irqchip. When the irqchip is resampled, such
2814 as from an EOI, the gsi is de-asserted and the user is notified via
2815 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
2816 the interrupt if the device making use of it still requires service.
2817 Note that closing the resamplefd is not sufficient to disable the
2818 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
2819 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
2821 On arm/arm64, gsi routing being supported, the following can happen:
2823 - in case no routing entry is associated to this gsi, injection fails
2824 - in case the gsi is associated to an irqchip routing entry,
2825 irqchip.pin + 32 corresponds to the injected SPI ID.
2826 - in case the gsi is associated to an MSI routing entry, the MSI
2827 message and device ID are translated into an LPI (support restricted
2828 to GICv3 ITS in-kernel emulation).
2830 4.76 KVM_PPC_ALLOCATE_HTAB
2831 --------------------------
2833 :Capability: KVM_CAP_PPC_ALLOC_HTAB
2834 :Architectures: powerpc
2836 :Parameters: Pointer to u32 containing hash table order (in/out)
2837 :Returns: 0 on success, -1 on error
2839 This requests the host kernel to allocate an MMU hash table for a
2840 guest using the PAPR paravirtualization interface. This only does
2841 anything if the kernel is configured to use the Book 3S HV style of
2842 virtualization. Otherwise the capability doesn't exist and the ioctl
2843 returns an ENOTTY error. The rest of this description assumes Book 3S
2846 There must be no vcpus running when this ioctl is called; if there
2847 are, it will do nothing and return an EBUSY error.
2849 The parameter is a pointer to a 32-bit unsigned integer variable
2850 containing the order (log base 2) of the desired size of the hash
2851 table, which must be between 18 and 46. On successful return from the
2852 ioctl, the value will not be changed by the kernel.
2854 If no hash table has been allocated when any vcpu is asked to run
2855 (with the KVM_RUN ioctl), the host kernel will allocate a
2856 default-sized hash table (16 MB).
2858 If this ioctl is called when a hash table has already been allocated,
2859 with a different order from the existing hash table, the existing hash
2860 table will be freed and a new one allocated. If this is ioctl is
2861 called when a hash table has already been allocated of the same order
2862 as specified, the kernel will clear out the existing hash table (zero
2863 all HPTEs). In either case, if the guest is using the virtualized
2864 real-mode area (VRMA) facility, the kernel will re-create the VMRA
2865 HPTEs on the next KVM_RUN of any vcpu.
2867 4.77 KVM_S390_INTERRUPT
2868 -----------------------
2871 :Architectures: s390
2872 :Type: vm ioctl, vcpu ioctl
2873 :Parameters: struct kvm_s390_interrupt (in)
2874 :Returns: 0 on success, -1 on error
2876 Allows to inject an interrupt to the guest. Interrupts can be floating
2877 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
2879 Interrupt parameters are passed via kvm_s390_interrupt::
2881 struct kvm_s390_interrupt {
2887 type can be one of the following:
2889 KVM_S390_SIGP_STOP (vcpu)
2890 - sigp stop; optional flags in parm
2891 KVM_S390_PROGRAM_INT (vcpu)
2892 - program check; code in parm
2893 KVM_S390_SIGP_SET_PREFIX (vcpu)
2894 - sigp set prefix; prefix address in parm
2895 KVM_S390_RESTART (vcpu)
2897 KVM_S390_INT_CLOCK_COMP (vcpu)
2898 - clock comparator interrupt
2899 KVM_S390_INT_CPU_TIMER (vcpu)
2900 - CPU timer interrupt
2901 KVM_S390_INT_VIRTIO (vm)
2902 - virtio external interrupt; external interrupt
2903 parameters in parm and parm64
2904 KVM_S390_INT_SERVICE (vm)
2905 - sclp external interrupt; sclp parameter in parm
2906 KVM_S390_INT_EMERGENCY (vcpu)
2907 - sigp emergency; source cpu in parm
2908 KVM_S390_INT_EXTERNAL_CALL (vcpu)
2909 - sigp external call; source cpu in parm
2910 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm)
2911 - compound value to indicate an
2912 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
2913 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
2914 interruption subclass)
2915 KVM_S390_MCHK (vm, vcpu)
2916 - machine check interrupt; cr 14 bits in parm, machine check interrupt
2917 code in parm64 (note that machine checks needing further payload are not
2918 supported by this ioctl)
2920 This is an asynchronous vcpu ioctl and can be invoked from any thread.
2922 4.78 KVM_PPC_GET_HTAB_FD
2923 ------------------------
2925 :Capability: KVM_CAP_PPC_HTAB_FD
2926 :Architectures: powerpc
2928 :Parameters: Pointer to struct kvm_get_htab_fd (in)
2929 :Returns: file descriptor number (>= 0) on success, -1 on error
2931 This returns a file descriptor that can be used either to read out the
2932 entries in the guest's hashed page table (HPT), or to write entries to
2933 initialize the HPT. The returned fd can only be written to if the
2934 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
2935 can only be read if that bit is clear. The argument struct looks like
2938 /* For KVM_PPC_GET_HTAB_FD */
2939 struct kvm_get_htab_fd {
2945 /* Values for kvm_get_htab_fd.flags */
2946 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
2947 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
2949 The 'start_index' field gives the index in the HPT of the entry at
2950 which to start reading. It is ignored when writing.
2952 Reads on the fd will initially supply information about all
2953 "interesting" HPT entries. Interesting entries are those with the
2954 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
2955 all entries. When the end of the HPT is reached, the read() will
2956 return. If read() is called again on the fd, it will start again from
2957 the beginning of the HPT, but will only return HPT entries that have
2958 changed since they were last read.
2960 Data read or written is structured as a header (8 bytes) followed by a
2961 series of valid HPT entries (16 bytes) each. The header indicates how
2962 many valid HPT entries there are and how many invalid entries follow
2963 the valid entries. The invalid entries are not represented explicitly
2964 in the stream. The header format is::
2966 struct kvm_get_htab_header {
2972 Writes to the fd create HPT entries starting at the index given in the
2973 header; first 'n_valid' valid entries with contents from the data
2974 written, then 'n_invalid' invalid entries, invalidating any previously
2975 valid entries found.
2977 4.79 KVM_CREATE_DEVICE
2978 ----------------------
2980 :Capability: KVM_CAP_DEVICE_CTRL
2982 :Parameters: struct kvm_create_device (in/out)
2983 :Returns: 0 on success, -1 on error
2987 ====== =======================================================
2988 ENODEV The device type is unknown or unsupported
2989 EEXIST Device already created, and this type of device may not
2990 be instantiated multiple times
2991 ====== =======================================================
2993 Other error conditions may be defined by individual device types or
2994 have their standard meanings.
2996 Creates an emulated device in the kernel. The file descriptor returned
2997 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
2999 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
3000 device type is supported (not necessarily whether it can be created
3003 Individual devices should not define flags. Attributes should be used
3004 for specifying any behavior that is not implied by the device type
3009 struct kvm_create_device {
3010 __u32 type; /* in: KVM_DEV_TYPE_xxx */
3011 __u32 fd; /* out: device handle */
3012 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
3015 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
3016 --------------------------------------------
3018 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
3019 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
3020 :Type: device ioctl, vm ioctl, vcpu ioctl
3021 :Parameters: struct kvm_device_attr
3022 :Returns: 0 on success, -1 on error
3026 ===== =============================================================
3027 ENXIO The group or attribute is unknown/unsupported for this device
3028 or hardware support is missing.
3029 EPERM The attribute cannot (currently) be accessed this way
3030 (e.g. read-only attribute, or attribute that only makes
3031 sense when the device is in a different state)
3032 ===== =============================================================
3034 Other error conditions may be defined by individual device types.
3036 Gets/sets a specified piece of device configuration and/or state. The
3037 semantics are device-specific. See individual device documentation in
3038 the "devices" directory. As with ONE_REG, the size of the data
3039 transferred is defined by the particular attribute.
3043 struct kvm_device_attr {
3044 __u32 flags; /* no flags currently defined */
3045 __u32 group; /* device-defined */
3046 __u64 attr; /* group-defined */
3047 __u64 addr; /* userspace address of attr data */
3050 4.81 KVM_HAS_DEVICE_ATTR
3051 ------------------------
3053 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
3054 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
3055 :Type: device ioctl, vm ioctl, vcpu ioctl
3056 :Parameters: struct kvm_device_attr
3057 :Returns: 0 on success, -1 on error
3061 ===== =============================================================
3062 ENXIO The group or attribute is unknown/unsupported for this device
3063 or hardware support is missing.
3064 ===== =============================================================
3066 Tests whether a device supports a particular attribute. A successful
3067 return indicates the attribute is implemented. It does not necessarily
3068 indicate that the attribute can be read or written in the device's
3069 current state. "addr" is ignored.
3071 4.82 KVM_ARM_VCPU_INIT
3072 ----------------------
3075 :Architectures: arm, arm64
3077 :Parameters: struct kvm_vcpu_init (in)
3078 :Returns: 0 on success; -1 on error
3082 ====== =================================================================
3083 Â EINVAL Â Â Â the target is unknown, or the combination of features is invalid.
3084 Â ENOENT Â Â Â a features bit specified is unknown.
3085 ====== =================================================================
3087 This tells KVM what type of CPU to present to the guest, and what
3088 optional features it should have. Â This will cause a reset of the cpu
3089 registers to their initial values. Â If this is not called, KVM_RUN will
3090 return ENOEXEC for that vcpu.
3092 Note that because some registers reflect machine topology, all vcpus
3093 should be created before this ioctl is invoked.
3095 Userspace can call this function multiple times for a given vcpu, including
3096 after the vcpu has been run. This will reset the vcpu to its initial
3097 state. All calls to this function after the initial call must use the same
3098 target and same set of feature flags, otherwise EINVAL will be returned.
3102 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
3103 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
3104 and execute guest code when KVM_RUN is called.
3105 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
3106 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
3107 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision
3108 backward compatible with v0.2) for the CPU.
3109 Depends on KVM_CAP_ARM_PSCI_0_2.
3110 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU.
3111 Depends on KVM_CAP_ARM_PMU_V3.
3113 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication
3115 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS.
3116 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are
3117 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and
3118 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be
3121 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication
3123 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC.
3124 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are
3125 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and
3126 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be
3129 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only).
3130 Depends on KVM_CAP_ARM_SVE.
3131 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3133 * After KVM_ARM_VCPU_INIT:
3135 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the
3136 initial value of this pseudo-register indicates the best set of
3137 vector lengths possible for a vcpu on this host.
3139 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3141 - KVM_RUN and KVM_GET_REG_LIST are not available;
3143 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access
3144 the scalable archietctural SVE registers
3145 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or
3146 KVM_REG_ARM64_SVE_FFR;
3148 - KVM_REG_ARM64_SVE_VLS may optionally be written using
3149 KVM_SET_ONE_REG, to modify the set of vector lengths available
3152 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3154 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can
3155 no longer be written using KVM_SET_ONE_REG.
3157 4.83 KVM_ARM_PREFERRED_TARGET
3158 -----------------------------
3161 :Architectures: arm, arm64
3163 :Parameters: struct kvm_vcpu_init (out)
3164 :Returns: 0 on success; -1 on error
3168 ====== ==========================================
3169 ENODEV no preferred target available for the host
3170 ====== ==========================================
3172 This queries KVM for preferred CPU target type which can be emulated
3173 by KVM on underlying host.
3175 The ioctl returns struct kvm_vcpu_init instance containing information
3176 about preferred CPU target type and recommended features for it. The
3177 kvm_vcpu_init->features bitmap returned will have feature bits set if
3178 the preferred target recommends setting these features, but this is
3181 The information returned by this ioctl can be used to prepare an instance
3182 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
3183 VCPU matching underlying host.
3186 4.84 KVM_GET_REG_LIST
3187 ---------------------
3190 :Architectures: arm, arm64, mips
3192 :Parameters: struct kvm_reg_list (in/out)
3193 :Returns: 0 on success; -1 on error
3197 ===== ==============================================================
3198 Â E2BIG Â Â Â Â the reg index list is too big to fit in the array specified by
3199 Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n).
3200 ===== ==============================================================
3204 struct kvm_reg_list {
3205 __u64 n; /* number of registers in reg[] */
3209 This ioctl returns the guest registers that are supported for the
3210 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
3213 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
3214 -----------------------------------------
3216 :Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
3217 :Architectures: arm, arm64
3219 :Parameters: struct kvm_arm_device_address (in)
3220 :Returns: 0 on success, -1 on error
3224 ====== ============================================
3225 ENODEV The device id is unknown
3226 ENXIO Device not supported on current system
3227 EEXIST Address already set
3228 E2BIG Address outside guest physical address space
3229 EBUSY Address overlaps with other device range
3230 ====== ============================================
3234 struct kvm_arm_device_addr {
3239 Specify a device address in the guest's physical address space where guests
3240 can access emulated or directly exposed devices, which the host kernel needs
3241 to know about. The id field is an architecture specific identifier for a
3244 ARM/arm64 divides the id field into two parts, a device id and an
3245 address type id specific to the individual device::
3247 Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
3248 field: | 0x00000000 | device id | addr type id |
3250 ARM/arm64 currently only require this when using the in-kernel GIC
3251 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
3252 as the device id. When setting the base address for the guest's
3253 mapping of the VGIC virtual CPU and distributor interface, the ioctl
3254 must be called after calling KVM_CREATE_IRQCHIP, but before calling
3255 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
3256 base addresses will return -EEXIST.
3258 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
3259 should be used instead.
3262 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
3263 ------------------------------
3265 :Capability: KVM_CAP_PPC_RTAS
3268 :Parameters: struct kvm_rtas_token_args
3269 :Returns: 0 on success, -1 on error
3271 Defines a token value for a RTAS (Run Time Abstraction Services)
3272 service in order to allow it to be handled in the kernel. The
3273 argument struct gives the name of the service, which must be the name
3274 of a service that has a kernel-side implementation. If the token
3275 value is non-zero, it will be associated with that service, and
3276 subsequent RTAS calls by the guest specifying that token will be
3277 handled by the kernel. If the token value is 0, then any token
3278 associated with the service will be forgotten, and subsequent RTAS
3279 calls by the guest for that service will be passed to userspace to be
3282 4.87 KVM_SET_GUEST_DEBUG
3283 ------------------------
3285 :Capability: KVM_CAP_SET_GUEST_DEBUG
3286 :Architectures: x86, s390, ppc, arm64
3288 :Parameters: struct kvm_guest_debug (in)
3289 :Returns: 0 on success; -1 on error
3293 struct kvm_guest_debug {
3296 struct kvm_guest_debug_arch arch;
3299 Set up the processor specific debug registers and configure vcpu for
3300 handling guest debug events. There are two parts to the structure, the
3301 first a control bitfield indicates the type of debug events to handle
3302 when running. Common control bits are:
3304 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
3305 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
3307 The top 16 bits of the control field are architecture specific control
3308 flags which can include the following:
3310 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64]
3311 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64]
3312 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
3313 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
3314 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
3316 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
3317 are enabled in memory so we need to ensure breakpoint exceptions are
3318 correctly trapped and the KVM run loop exits at the breakpoint and not
3319 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
3320 we need to ensure the guest vCPUs architecture specific registers are
3321 updated to the correct (supplied) values.
3323 The second part of the structure is architecture specific and
3324 typically contains a set of debug registers.
3326 For arm64 the number of debug registers is implementation defined and
3327 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
3328 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
3329 indicating the number of supported registers.
3331 For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether
3332 the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported.
3334 When debug events exit the main run loop with the reason
3335 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
3336 structure containing architecture specific debug information.
3338 4.88 KVM_GET_EMULATED_CPUID
3339 ---------------------------
3341 :Capability: KVM_CAP_EXT_EMUL_CPUID
3344 :Parameters: struct kvm_cpuid2 (in/out)
3345 :Returns: 0 on success, -1 on error
3352 struct kvm_cpuid_entry2 entries[0];
3355 The member 'flags' is used for passing flags from userspace.
3359 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
3360 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */
3361 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */
3363 struct kvm_cpuid_entry2 {
3374 This ioctl returns x86 cpuid features which are emulated by
3375 kvm.Userspace can use the information returned by this ioctl to query
3376 which features are emulated by kvm instead of being present natively.
3378 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
3379 structure with the 'nent' field indicating the number of entries in
3380 the variable-size array 'entries'. If the number of entries is too low
3381 to describe the cpu capabilities, an error (E2BIG) is returned. If the
3382 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
3383 is returned. If the number is just right, the 'nent' field is adjusted
3384 to the number of valid entries in the 'entries' array, which is then
3387 The entries returned are the set CPUID bits of the respective features
3388 which kvm emulates, as returned by the CPUID instruction, with unknown
3389 or unsupported feature bits cleared.
3391 Features like x2apic, for example, may not be present in the host cpu
3392 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
3393 emulated efficiently and thus not included here.
3395 The fields in each entry are defined as follows:
3398 the eax value used to obtain the entry
3400 the ecx value used to obtain the entry (for entries that are
3403 an OR of zero or more of the following:
3405 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
3406 if the index field is valid
3410 the values returned by the cpuid instruction for
3411 this function/index combination
3413 4.89 KVM_S390_MEM_OP
3414 --------------------
3416 :Capability: KVM_CAP_S390_MEM_OP
3417 :Architectures: s390
3419 :Parameters: struct kvm_s390_mem_op (in)
3420 :Returns: = 0 on success,
3421 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
3422 > 0 if an exception occurred while walking the page tables
3424 Read or write data from/to the logical (virtual) memory of a VCPU.
3426 Parameters are specified via the following structure::
3428 struct kvm_s390_mem_op {
3429 __u64 gaddr; /* the guest address */
3430 __u64 flags; /* flags */
3431 __u32 size; /* amount of bytes */
3432 __u32 op; /* type of operation */
3433 __u64 buf; /* buffer in userspace */
3434 __u8 ar; /* the access register number */
3435 __u8 reserved[31]; /* should be set to 0 */
3438 The type of operation is specified in the "op" field. It is either
3439 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or
3440 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The
3441 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check
3442 whether the corresponding memory access would create an access exception
3443 (without touching the data in the memory at the destination). In case an
3444 access exception occurred while walking the MMU tables of the guest, the
3445 ioctl returns a positive error number to indicate the type of exception.
3446 This exception is also raised directly at the corresponding VCPU if the
3447 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field.
3449 The start address of the memory region has to be specified in the "gaddr"
3450 field, and the length of the region in the "size" field (which must not
3451 be 0). The maximum value for "size" can be obtained by checking the
3452 KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the
3453 userspace application where the read data should be written to for
3454 KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written is
3455 stored for a KVM_S390_MEMOP_LOGICAL_WRITE. When KVM_S390_MEMOP_F_CHECK_ONLY
3456 is specified, "buf" is unused and can be NULL. "ar" designates the access
3457 register number to be used; the valid range is 0..15.
3459 The "reserved" field is meant for future extensions. It is not used by
3460 KVM with the currently defined set of flags.
3462 4.90 KVM_S390_GET_SKEYS
3463 -----------------------
3465 :Capability: KVM_CAP_S390_SKEYS
3466 :Architectures: s390
3468 :Parameters: struct kvm_s390_skeys
3469 :Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage
3470 keys, negative value on error
3472 This ioctl is used to get guest storage key values on the s390
3473 architecture. The ioctl takes parameters via the kvm_s390_skeys struct::
3475 struct kvm_s390_skeys {
3478 __u64 skeydata_addr;
3483 The start_gfn field is the number of the first guest frame whose storage keys
3486 The count field is the number of consecutive frames (starting from start_gfn)
3487 whose storage keys to get. The count field must be at least 1 and the maximum
3488 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
3489 will cause the ioctl to return -EINVAL.
3491 The skeydata_addr field is the address to a buffer large enough to hold count
3492 bytes. This buffer will be filled with storage key data by the ioctl.
3494 4.91 KVM_S390_SET_SKEYS
3495 -----------------------
3497 :Capability: KVM_CAP_S390_SKEYS
3498 :Architectures: s390
3500 :Parameters: struct kvm_s390_skeys
3501 :Returns: 0 on success, negative value on error
3503 This ioctl is used to set guest storage key values on the s390
3504 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
3505 See section on KVM_S390_GET_SKEYS for struct definition.
3507 The start_gfn field is the number of the first guest frame whose storage keys
3510 The count field is the number of consecutive frames (starting from start_gfn)
3511 whose storage keys to get. The count field must be at least 1 and the maximum
3512 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
3513 will cause the ioctl to return -EINVAL.
3515 The skeydata_addr field is the address to a buffer containing count bytes of
3516 storage keys. Each byte in the buffer will be set as the storage key for a
3517 single frame starting at start_gfn for count frames.
3519 Note: If any architecturally invalid key value is found in the given data then
3520 the ioctl will return -EINVAL.
3525 :Capability: KVM_CAP_S390_INJECT_IRQ
3526 :Architectures: s390
3528 :Parameters: struct kvm_s390_irq (in)
3529 :Returns: 0 on success, -1 on error
3534 ====== =================================================================
3535 EINVAL interrupt type is invalid
3536 type is KVM_S390_SIGP_STOP and flag parameter is invalid value,
3537 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger
3538 than the maximum of VCPUs
3539 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped,
3540 type is KVM_S390_SIGP_STOP and a stop irq is already pending,
3541 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt
3543 ====== =================================================================
3545 Allows to inject an interrupt to the guest.
3547 Using struct kvm_s390_irq as a parameter allows
3548 to inject additional payload which is not
3549 possible via KVM_S390_INTERRUPT.
3551 Interrupt parameters are passed via kvm_s390_irq::
3553 struct kvm_s390_irq {
3556 struct kvm_s390_io_info io;
3557 struct kvm_s390_ext_info ext;
3558 struct kvm_s390_pgm_info pgm;
3559 struct kvm_s390_emerg_info emerg;
3560 struct kvm_s390_extcall_info extcall;
3561 struct kvm_s390_prefix_info prefix;
3562 struct kvm_s390_stop_info stop;
3563 struct kvm_s390_mchk_info mchk;
3568 type can be one of the following:
3570 - KVM_S390_SIGP_STOP - sigp stop; parameter in .stop
3571 - KVM_S390_PROGRAM_INT - program check; parameters in .pgm
3572 - KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix
3573 - KVM_S390_RESTART - restart; no parameters
3574 - KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters
3575 - KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters
3576 - KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg
3577 - KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall
3578 - KVM_S390_MCHK - machine check interrupt; parameters in .mchk
3580 This is an asynchronous vcpu ioctl and can be invoked from any thread.
3582 4.94 KVM_S390_GET_IRQ_STATE
3583 ---------------------------
3585 :Capability: KVM_CAP_S390_IRQ_STATE
3586 :Architectures: s390
3588 :Parameters: struct kvm_s390_irq_state (out)
3589 :Returns: >= number of bytes copied into buffer,
3590 -EINVAL if buffer size is 0,
3591 -ENOBUFS if buffer size is too small to fit all pending interrupts,
3592 -EFAULT if the buffer address was invalid
3594 This ioctl allows userspace to retrieve the complete state of all currently
3595 pending interrupts in a single buffer. Use cases include migration
3596 and introspection. The parameter structure contains the address of a
3597 userspace buffer and its length::
3599 struct kvm_s390_irq_state {
3601 __u32 flags; /* will stay unused for compatibility reasons */
3603 __u32 reserved[4]; /* will stay unused for compatibility reasons */
3606 Userspace passes in the above struct and for each pending interrupt a
3607 struct kvm_s390_irq is copied to the provided buffer.
3609 The structure contains a flags and a reserved field for future extensions. As
3610 the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and
3611 reserved, these fields can not be used in the future without breaking
3614 If -ENOBUFS is returned the buffer provided was too small and userspace
3615 may retry with a bigger buffer.
3617 4.95 KVM_S390_SET_IRQ_STATE
3618 ---------------------------
3620 :Capability: KVM_CAP_S390_IRQ_STATE
3621 :Architectures: s390
3623 :Parameters: struct kvm_s390_irq_state (in)
3624 :Returns: 0 on success,
3625 -EFAULT if the buffer address was invalid,
3626 -EINVAL for an invalid buffer length (see below),
3627 -EBUSY if there were already interrupts pending,
3628 errors occurring when actually injecting the
3629 interrupt. See KVM_S390_IRQ.
3631 This ioctl allows userspace to set the complete state of all cpu-local
3632 interrupts currently pending for the vcpu. It is intended for restoring
3633 interrupt state after a migration. The input parameter is a userspace buffer
3634 containing a struct kvm_s390_irq_state::
3636 struct kvm_s390_irq_state {
3638 __u32 flags; /* will stay unused for compatibility reasons */
3640 __u32 reserved[4]; /* will stay unused for compatibility reasons */
3643 The restrictions for flags and reserved apply as well.
3644 (see KVM_S390_GET_IRQ_STATE)
3646 The userspace memory referenced by buf contains a struct kvm_s390_irq
3647 for each interrupt to be injected into the guest.
3648 If one of the interrupts could not be injected for some reason the
3651 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
3652 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
3653 which is the maximum number of possibly pending cpu-local interrupts.
3658 :Capability: KVM_CAP_X86_SMM
3662 :Returns: 0 on success, -1 on error
3664 Queues an SMI on the thread's vcpu.
3666 4.97 KVM_CAP_PPC_MULTITCE
3667 -------------------------
3669 :Capability: KVM_CAP_PPC_MULTITCE
3673 This capability means the kernel is capable of handling hypercalls
3674 H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user
3675 space. This significantly accelerates DMA operations for PPC KVM guests.
3676 User space should expect that its handlers for these hypercalls
3677 are not going to be called if user space previously registered LIOBN
3678 in KVM (via KVM_CREATE_SPAPR_TCE or similar calls).
3680 In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest,
3681 user space might have to advertise it for the guest. For example,
3682 IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is
3683 present in the "ibm,hypertas-functions" device-tree property.
3685 The hypercalls mentioned above may or may not be processed successfully
3686 in the kernel based fast path. If they can not be handled by the kernel,
3687 they will get passed on to user space. So user space still has to have
3688 an implementation for these despite the in kernel acceleration.
3690 This capability is always enabled.
3692 4.98 KVM_CREATE_SPAPR_TCE_64
3693 ----------------------------
3695 :Capability: KVM_CAP_SPAPR_TCE_64
3696 :Architectures: powerpc
3698 :Parameters: struct kvm_create_spapr_tce_64 (in)
3699 :Returns: file descriptor for manipulating the created TCE table
3701 This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit
3702 windows, described in 4.62 KVM_CREATE_SPAPR_TCE
3704 This capability uses extended struct in ioctl interface::
3706 /* for KVM_CAP_SPAPR_TCE_64 */
3707 struct kvm_create_spapr_tce_64 {
3711 __u64 offset; /* in pages */
3712 __u64 size; /* in pages */
3715 The aim of extension is to support an additional bigger DMA window with
3716 a variable page size.
3717 KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and
3718 a bus offset of the corresponding DMA window, @size and @offset are numbers
3721 @flags are not used at the moment.
3723 The rest of functionality is identical to KVM_CREATE_SPAPR_TCE.
3725 4.99 KVM_REINJECT_CONTROL
3726 -------------------------
3728 :Capability: KVM_CAP_REINJECT_CONTROL
3731 :Parameters: struct kvm_reinject_control (in)
3732 :Returns: 0 on success,
3733 -EFAULT if struct kvm_reinject_control cannot be read,
3734 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier.
3736 i8254 (PIT) has two modes, reinject and !reinject. The default is reinject,
3737 where KVM queues elapsed i8254 ticks and monitors completion of interrupt from
3738 vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its
3739 interrupt whenever there isn't a pending interrupt from i8254.
3740 !reinject mode injects an interrupt as soon as a tick arrives.
3744 struct kvm_reinject_control {
3749 pit_reinject = 0 (!reinject mode) is recommended, unless running an old
3750 operating system that uses the PIT for timing (e.g. Linux 2.4.x).
3752 4.100 KVM_PPC_CONFIGURE_V3_MMU
3753 ------------------------------
3755 :Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3
3758 :Parameters: struct kvm_ppc_mmuv3_cfg (in)
3759 :Returns: 0 on success,
3760 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read,
3761 -EINVAL if the configuration is invalid
3763 This ioctl controls whether the guest will use radix or HPT (hashed
3764 page table) translation, and sets the pointer to the process table for
3769 struct kvm_ppc_mmuv3_cfg {
3771 __u64 process_table;
3774 There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and
3775 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest
3776 to use radix tree translation, and if clear, to use HPT translation.
3777 KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest
3778 to be able to use the global TLB and SLB invalidation instructions;
3779 if clear, the guest may not use these instructions.
3781 The process_table field specifies the address and size of the guest
3782 process table, which is in the guest's space. This field is formatted
3783 as the second doubleword of the partition table entry, as defined in
3784 the Power ISA V3.00, Book III section 5.7.6.1.
3786 4.101 KVM_PPC_GET_RMMU_INFO
3787 ---------------------------
3789 :Capability: KVM_CAP_PPC_RADIX_MMU
3792 :Parameters: struct kvm_ppc_rmmu_info (out)
3793 :Returns: 0 on success,
3794 -EFAULT if struct kvm_ppc_rmmu_info cannot be written,
3795 -EINVAL if no useful information can be returned
3797 This ioctl returns a structure containing two things: (a) a list
3798 containing supported radix tree geometries, and (b) a list that maps
3799 page sizes to put in the "AP" (actual page size) field for the tlbie
3800 (TLB invalidate entry) instruction.
3804 struct kvm_ppc_rmmu_info {
3805 struct kvm_ppc_radix_geom {
3810 __u32 ap_encodings[8];
3813 The geometries[] field gives up to 8 supported geometries for the
3814 radix page table, in terms of the log base 2 of the smallest page
3815 size, and the number of bits indexed at each level of the tree, from
3816 the PTE level up to the PGD level in that order. Any unused entries
3817 will have 0 in the page_shift field.
3819 The ap_encodings gives the supported page sizes and their AP field
3820 encodings, encoded with the AP value in the top 3 bits and the log
3821 base 2 of the page size in the bottom 6 bits.
3823 4.102 KVM_PPC_RESIZE_HPT_PREPARE
3824 --------------------------------
3826 :Capability: KVM_CAP_SPAPR_RESIZE_HPT
3827 :Architectures: powerpc
3829 :Parameters: struct kvm_ppc_resize_hpt (in)
3830 :Returns: 0 on successful completion,
3831 >0 if a new HPT is being prepared, the value is an estimated
3832 number of milliseconds until preparation is complete,
3833 -EFAULT if struct kvm_reinject_control cannot be read,
3834 -EINVAL if the supplied shift or flags are invalid,
3835 -ENOMEM if unable to allocate the new HPT,
3836 -ENOSPC if there was a hash collision
3840 struct kvm_ppc_rmmu_info {
3841 struct kvm_ppc_radix_geom {
3846 __u32 ap_encodings[8];
3849 The geometries[] field gives up to 8 supported geometries for the
3850 radix page table, in terms of the log base 2 of the smallest page
3851 size, and the number of bits indexed at each level of the tree, from
3852 the PTE level up to the PGD level in that order. Any unused entries
3853 will have 0 in the page_shift field.
3855 The ap_encodings gives the supported page sizes and their AP field
3856 encodings, encoded with the AP value in the top 3 bits and the log
3857 base 2 of the page size in the bottom 6 bits.
3859 4.102 KVM_PPC_RESIZE_HPT_PREPARE
3860 --------------------------------
3862 :Capability: KVM_CAP_SPAPR_RESIZE_HPT
3863 :Architectures: powerpc
3865 :Parameters: struct kvm_ppc_resize_hpt (in)
3866 :Returns: 0 on successful completion,
3867 >0 if a new HPT is being prepared, the value is an estimated
3868 number of milliseconds until preparation is complete,
3869 -EFAULT if struct kvm_reinject_control cannot be read,
3870 -EINVAL if the supplied shift or flags are invalid,when moving existing
3871 HPT entries to the new HPT,
3872 -EIO on other error conditions
3874 Used to implement the PAPR extension for runtime resizing of a guest's
3875 Hashed Page Table (HPT). Specifically this starts, stops or monitors
3876 the preparation of a new potential HPT for the guest, essentially
3877 implementing the H_RESIZE_HPT_PREPARE hypercall.
3879 If called with shift > 0 when there is no pending HPT for the guest,
3880 this begins preparation of a new pending HPT of size 2^(shift) bytes.
3881 It then returns a positive integer with the estimated number of
3882 milliseconds until preparation is complete.
3884 If called when there is a pending HPT whose size does not match that
3885 requested in the parameters, discards the existing pending HPT and
3886 creates a new one as above.
3888 If called when there is a pending HPT of the size requested, will:
3890 * If preparation of the pending HPT is already complete, return 0
3891 * If preparation of the pending HPT has failed, return an error
3892 code, then discard the pending HPT.
3893 * If preparation of the pending HPT is still in progress, return an
3894 estimated number of milliseconds until preparation is complete.
3896 If called with shift == 0, discards any currently pending HPT and
3897 returns 0 (i.e. cancels any in-progress preparation).
3899 flags is reserved for future expansion, currently setting any bits in
3900 flags will result in an -EINVAL.
3902 Normally this will be called repeatedly with the same parameters until
3903 it returns <= 0. The first call will initiate preparation, subsequent
3904 ones will monitor preparation until it completes or fails.
3908 struct kvm_ppc_resize_hpt {
3914 4.103 KVM_PPC_RESIZE_HPT_COMMIT
3915 -------------------------------
3917 :Capability: KVM_CAP_SPAPR_RESIZE_HPT
3918 :Architectures: powerpc
3920 :Parameters: struct kvm_ppc_resize_hpt (in)
3921 :Returns: 0 on successful completion,
3922 -EFAULT if struct kvm_reinject_control cannot be read,
3923 -EINVAL if the supplied shift or flags are invalid,
3924 -ENXIO is there is no pending HPT, or the pending HPT doesn't
3925 have the requested size,
3926 -EBUSY if the pending HPT is not fully prepared,
3927 -ENOSPC if there was a hash collision when moving existing
3928 HPT entries to the new HPT,
3929 -EIO on other error conditions
3931 Used to implement the PAPR extension for runtime resizing of a guest's
3932 Hashed Page Table (HPT). Specifically this requests that the guest be
3933 transferred to working with the new HPT, essentially implementing the
3934 H_RESIZE_HPT_COMMIT hypercall.
3936 This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has
3937 returned 0 with the same parameters. In other cases
3938 KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or
3939 -EBUSY, though others may be possible if the preparation was started,
3942 This will have undefined effects on the guest if it has not already
3943 placed itself in a quiescent state where no vcpu will make MMU enabled
3946 On succsful completion, the pending HPT will become the guest's active
3947 HPT and the previous HPT will be discarded.
3949 On failure, the guest will still be operating on its previous HPT.
3953 struct kvm_ppc_resize_hpt {
3959 4.104 KVM_X86_GET_MCE_CAP_SUPPORTED
3960 -----------------------------------
3962 :Capability: KVM_CAP_MCE
3965 :Parameters: u64 mce_cap (out)
3966 :Returns: 0 on success, -1 on error
3968 Returns supported MCE capabilities. The u64 mce_cap parameter
3969 has the same format as the MSR_IA32_MCG_CAP register. Supported
3970 capabilities will have the corresponding bits set.
3972 4.105 KVM_X86_SETUP_MCE
3973 -----------------------
3975 :Capability: KVM_CAP_MCE
3978 :Parameters: u64 mcg_cap (in)
3979 :Returns: 0 on success,
3980 -EFAULT if u64 mcg_cap cannot be read,
3981 -EINVAL if the requested number of banks is invalid,
3982 -EINVAL if requested MCE capability is not supported.
3984 Initializes MCE support for use. The u64 mcg_cap parameter
3985 has the same format as the MSR_IA32_MCG_CAP register and
3986 specifies which capabilities should be enabled. The maximum
3987 supported number of error-reporting banks can be retrieved when
3988 checking for KVM_CAP_MCE. The supported capabilities can be
3989 retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED.
3991 4.106 KVM_X86_SET_MCE
3992 ---------------------
3994 :Capability: KVM_CAP_MCE
3997 :Parameters: struct kvm_x86_mce (in)
3998 :Returns: 0 on success,
3999 -EFAULT if struct kvm_x86_mce cannot be read,
4000 -EINVAL if the bank number is invalid,
4001 -EINVAL if VAL bit is not set in status field.
4003 Inject a machine check error (MCE) into the guest. The input
4006 struct kvm_x86_mce {
4016 If the MCE being reported is an uncorrected error, KVM will
4017 inject it as an MCE exception into the guest. If the guest
4018 MCG_STATUS register reports that an MCE is in progress, KVM
4019 causes an KVM_EXIT_SHUTDOWN vmexit.
4021 Otherwise, if the MCE is a corrected error, KVM will just
4022 store it in the corresponding bank (provided this bank is
4023 not holding a previously reported uncorrected error).
4025 4.107 KVM_S390_GET_CMMA_BITS
4026 ----------------------------
4028 :Capability: KVM_CAP_S390_CMMA_MIGRATION
4029 :Architectures: s390
4031 :Parameters: struct kvm_s390_cmma_log (in, out)
4032 :Returns: 0 on success, a negative value on error
4036 ====== =============================================================
4037 ENOMEM not enough memory can be allocated to complete the task
4038 ENXIO if CMMA is not enabled
4039 EINVAL if KVM_S390_CMMA_PEEK is not set but migration mode was not enabled
4040 EINVAL if KVM_S390_CMMA_PEEK is not set but dirty tracking has been
4041 disabled (and thus migration mode was automatically disabled)
4042 EFAULT if the userspace address is invalid or if no page table is
4043 present for the addresses (e.g. when using hugepages).
4044 ====== =============================================================
4046 This ioctl is used to get the values of the CMMA bits on the s390
4047 architecture. It is meant to be used in two scenarios:
4049 - During live migration to save the CMMA values. Live migration needs
4050 to be enabled via the KVM_REQ_START_MIGRATION VM property.
4051 - To non-destructively peek at the CMMA values, with the flag
4052 KVM_S390_CMMA_PEEK set.
4054 The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired
4055 values are written to a buffer whose location is indicated via the "values"
4056 member in the kvm_s390_cmma_log struct. The values in the input struct are
4057 also updated as needed.
4059 Each CMMA value takes up one byte.
4063 struct kvm_s390_cmma_log {
4074 start_gfn is the number of the first guest frame whose CMMA values are
4077 count is the length of the buffer in bytes,
4079 values points to the buffer where the result will be written to.
4081 If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be
4082 KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with
4085 The result is written in the buffer pointed to by the field values, and
4086 the values of the input parameter are updated as follows.
4088 Depending on the flags, different actions are performed. The only
4089 supported flag so far is KVM_S390_CMMA_PEEK.
4091 The default behaviour if KVM_S390_CMMA_PEEK is not set is:
4092 start_gfn will indicate the first page frame whose CMMA bits were dirty.
4093 It is not necessarily the same as the one passed as input, as clean pages
4096 count will indicate the number of bytes actually written in the buffer.
4097 It can (and very often will) be smaller than the input value, since the
4098 buffer is only filled until 16 bytes of clean values are found (which
4099 are then not copied in the buffer). Since a CMMA migration block needs
4100 the base address and the length, for a total of 16 bytes, we will send
4101 back some clean data if there is some dirty data afterwards, as long as
4102 the size of the clean data does not exceed the size of the header. This
4103 allows to minimize the amount of data to be saved or transferred over
4104 the network at the expense of more roundtrips to userspace. The next
4105 invocation of the ioctl will skip over all the clean values, saving
4106 potentially more than just the 16 bytes we found.
4108 If KVM_S390_CMMA_PEEK is set:
4109 the existing storage attributes are read even when not in migration
4110 mode, and no other action is performed;
4112 the output start_gfn will be equal to the input start_gfn,
4114 the output count will be equal to the input count, except if the end of
4115 memory has been reached.
4118 the field "remaining" will indicate the total number of dirty CMMA values
4119 still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is
4124 values points to the userspace buffer where the result will be stored.
4126 4.108 KVM_S390_SET_CMMA_BITS
4127 ----------------------------
4129 :Capability: KVM_CAP_S390_CMMA_MIGRATION
4130 :Architectures: s390
4132 :Parameters: struct kvm_s390_cmma_log (in)
4133 :Returns: 0 on success, a negative value on error
4135 This ioctl is used to set the values of the CMMA bits on the s390
4136 architecture. It is meant to be used during live migration to restore
4137 the CMMA values, but there are no restrictions on its use.
4138 The ioctl takes parameters via the kvm_s390_cmma_values struct.
4139 Each CMMA value takes up one byte.
4143 struct kvm_s390_cmma_log {
4154 start_gfn indicates the starting guest frame number,
4156 count indicates how many values are to be considered in the buffer,
4158 flags is not used and must be 0.
4160 mask indicates which PGSTE bits are to be considered.
4162 remaining is not used.
4164 values points to the buffer in userspace where to store the values.
4166 This ioctl can fail with -ENOMEM if not enough memory can be allocated to
4167 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if
4168 the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or
4169 if the flags field was not 0, with -EFAULT if the userspace address is
4170 invalid, if invalid pages are written to (e.g. after the end of memory)
4171 or if no page table is present for the addresses (e.g. when using
4174 4.109 KVM_PPC_GET_CPU_CHAR
4175 --------------------------
4177 :Capability: KVM_CAP_PPC_GET_CPU_CHAR
4178 :Architectures: powerpc
4180 :Parameters: struct kvm_ppc_cpu_char (out)
4181 :Returns: 0 on successful completion,
4182 -EFAULT if struct kvm_ppc_cpu_char cannot be written
4184 This ioctl gives userspace information about certain characteristics
4185 of the CPU relating to speculative execution of instructions and
4186 possible information leakage resulting from speculative execution (see
4187 CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is
4188 returned in struct kvm_ppc_cpu_char, which looks like this::
4190 struct kvm_ppc_cpu_char {
4191 __u64 character; /* characteristics of the CPU */
4192 __u64 behaviour; /* recommended software behaviour */
4193 __u64 character_mask; /* valid bits in character */
4194 __u64 behaviour_mask; /* valid bits in behaviour */
4197 For extensibility, the character_mask and behaviour_mask fields
4198 indicate which bits of character and behaviour have been filled in by
4199 the kernel. If the set of defined bits is extended in future then
4200 userspace will be able to tell whether it is running on a kernel that
4201 knows about the new bits.
4203 The character field describes attributes of the CPU which can help
4204 with preventing inadvertent information disclosure - specifically,
4205 whether there is an instruction to flash-invalidate the L1 data cache
4206 (ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set
4207 to a mode where entries can only be used by the thread that created
4208 them, whether the bcctr[l] instruction prevents speculation, and
4209 whether a speculation barrier instruction (ori 31,31,0) is provided.
4211 The behaviour field describes actions that software should take to
4212 prevent inadvertent information disclosure, and thus describes which
4213 vulnerabilities the hardware is subject to; specifically whether the
4214 L1 data cache should be flushed when returning to user mode from the
4215 kernel, and whether a speculation barrier should be placed between an
4216 array bounds check and the array access.
4218 These fields use the same bit definitions as the new
4219 H_GET_CPU_CHARACTERISTICS hypercall.
4221 4.110 KVM_MEMORY_ENCRYPT_OP
4222 ---------------------------
4227 :Parameters: an opaque platform specific structure (in/out)
4228 :Returns: 0 on success; -1 on error
4230 If the platform supports creating encrypted VMs then this ioctl can be used
4231 for issuing platform-specific memory encryption commands to manage those
4234 Currently, this ioctl is used for issuing Secure Encrypted Virtualization
4235 (SEV) commands on AMD Processors. The SEV commands are defined in
4236 Documentation/virt/kvm/amd-memory-encryption.rst.
4238 4.111 KVM_MEMORY_ENCRYPT_REG_REGION
4239 -----------------------------------
4244 :Parameters: struct kvm_enc_region (in)
4245 :Returns: 0 on success; -1 on error
4247 This ioctl can be used to register a guest memory region which may
4248 contain encrypted data (e.g. guest RAM, SMRAM etc).
4250 It is used in the SEV-enabled guest. When encryption is enabled, a guest
4251 memory region may contain encrypted data. The SEV memory encryption
4252 engine uses a tweak such that two identical plaintext pages, each at
4253 different locations will have differing ciphertexts. So swapping or
4254 moving ciphertext of those pages will not result in plaintext being
4255 swapped. So relocating (or migrating) physical backing pages for the SEV
4256 guest will require some additional steps.
4258 Note: The current SEV key management spec does not provide commands to
4259 swap or migrate (move) ciphertext pages. Hence, for now we pin the guest
4260 memory region registered with the ioctl.
4262 4.112 KVM_MEMORY_ENCRYPT_UNREG_REGION
4263 -------------------------------------
4268 :Parameters: struct kvm_enc_region (in)
4269 :Returns: 0 on success; -1 on error
4271 This ioctl can be used to unregister the guest memory region registered
4272 with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above.
4274 4.113 KVM_HYPERV_EVENTFD
4275 ------------------------
4277 :Capability: KVM_CAP_HYPERV_EVENTFD
4280 :Parameters: struct kvm_hyperv_eventfd (in)
4282 This ioctl (un)registers an eventfd to receive notifications from the guest on
4283 the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without
4284 causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number
4285 (bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit.
4289 struct kvm_hyperv_eventfd {
4296 The conn_id field should fit within 24 bits::
4298 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff
4300 The acceptable values for the flags field are::
4302 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0)
4304 :Returns: 0 on success,
4305 -EINVAL if conn_id or flags is outside the allowed range,
4306 -ENOENT on deassign if the conn_id isn't registered,
4307 -EEXIST on assign if the conn_id is already registered
4309 4.114 KVM_GET_NESTED_STATE
4310 --------------------------
4312 :Capability: KVM_CAP_NESTED_STATE
4315 :Parameters: struct kvm_nested_state (in/out)
4316 :Returns: 0 on success, -1 on error
4320 ===== =============================================================
4321 E2BIG the total state size exceeds the value of 'size' specified by
4322 the user; the size required will be written into size.
4323 ===== =============================================================
4327 struct kvm_nested_state {
4333 struct kvm_vmx_nested_state_hdr vmx;
4334 struct kvm_svm_nested_state_hdr svm;
4336 /* Pad the header to 128 bytes. */
4341 struct kvm_vmx_nested_state_data vmx[0];
4342 struct kvm_svm_nested_state_data svm[0];
4346 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001
4347 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002
4348 #define KVM_STATE_NESTED_EVMCS 0x00000004
4350 #define KVM_STATE_NESTED_FORMAT_VMX 0
4351 #define KVM_STATE_NESTED_FORMAT_SVM 1
4353 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000
4355 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001
4356 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002
4358 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001
4360 struct kvm_vmx_nested_state_hdr {
4369 __u64 preemption_timer_deadline;
4372 struct kvm_vmx_nested_state_data {
4373 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE];
4374 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE];
4377 This ioctl copies the vcpu's nested virtualization state from the kernel to
4380 The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE
4381 to the KVM_CHECK_EXTENSION ioctl().
4383 4.115 KVM_SET_NESTED_STATE
4384 --------------------------
4386 :Capability: KVM_CAP_NESTED_STATE
4389 :Parameters: struct kvm_nested_state (in)
4390 :Returns: 0 on success, -1 on error
4392 This copies the vcpu's kvm_nested_state struct from userspace to the kernel.
4393 For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE.
4395 4.116 KVM_(UN)REGISTER_COALESCED_MMIO
4396 -------------------------------------
4398 :Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio)
4399 KVM_CAP_COALESCED_PIO (for coalesced pio)
4402 :Parameters: struct kvm_coalesced_mmio_zone
4403 :Returns: 0 on success, < 0 on error
4405 Coalesced I/O is a performance optimization that defers hardware
4406 register write emulation so that userspace exits are avoided. It is
4407 typically used to reduce the overhead of emulating frequently accessed
4410 When a hardware register is configured for coalesced I/O, write accesses
4411 do not exit to userspace and their value is recorded in a ring buffer
4412 that is shared between kernel and userspace.
4414 Coalesced I/O is used if one or more write accesses to a hardware
4415 register can be deferred until a read or a write to another hardware
4416 register on the same device. This last access will cause a vmexit and
4417 userspace will process accesses from the ring buffer before emulating
4418 it. That will avoid exiting to userspace on repeated writes.
4420 Coalesced pio is based on coalesced mmio. There is little difference
4421 between coalesced mmio and pio except that coalesced pio records accesses
4424 4.117 KVM_CLEAR_DIRTY_LOG (vm ioctl)
4425 ------------------------------------
4427 :Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
4428 :Architectures: x86, arm, arm64, mips
4430 :Parameters: struct kvm_dirty_log (in)
4431 :Returns: 0 on success, -1 on error
4435 /* for KVM_CLEAR_DIRTY_LOG */
4436 struct kvm_clear_dirty_log {
4441 void __user *dirty_bitmap; /* one bit per page */
4446 The ioctl clears the dirty status of pages in a memory slot, according to
4447 the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap
4448 field. Bit 0 of the bitmap corresponds to page "first_page" in the
4449 memory slot, and num_pages is the size in bits of the input bitmap.
4450 first_page must be a multiple of 64; num_pages must also be a multiple of
4451 64 unless first_page + num_pages is the size of the memory slot. For each
4452 bit that is set in the input bitmap, the corresponding page is marked "clean"
4453 in KVM's dirty bitmap, and dirty tracking is re-enabled for that page
4454 (for example via write-protection, or by clearing the dirty bit in
4455 a page table entry).
4457 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies
4458 the address space for which you want to return the dirty bitmap.
4459 They must be less than the value that KVM_CHECK_EXTENSION returns for
4460 the KVM_CAP_MULTI_ADDRESS_SPACE capability.
4462 This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
4463 is enabled; for more information, see the description of the capability.
4464 However, it can always be used as long as KVM_CHECK_EXTENSION confirms
4465 that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present.
4467 4.118 KVM_GET_SUPPORTED_HV_CPUID
4468 --------------------------------
4470 :Capability: KVM_CAP_HYPERV_CPUID
4473 :Parameters: struct kvm_cpuid2 (in/out)
4474 :Returns: 0 on success, -1 on error
4481 struct kvm_cpuid_entry2 entries[0];
4484 struct kvm_cpuid_entry2 {
4495 This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in
4496 KVM. Userspace can use the information returned by this ioctl to construct
4497 cpuid information presented to guests consuming Hyper-V enlightenments (e.g.
4498 Windows or Hyper-V guests).
4500 CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level
4501 Functional Specification (TLFS). These leaves can't be obtained with
4502 KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature
4503 leaves (0x40000000, 0x40000001).
4505 Currently, the following list of CPUID leaves are returned:
4506 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS
4507 - HYPERV_CPUID_INTERFACE
4508 - HYPERV_CPUID_VERSION
4509 - HYPERV_CPUID_FEATURES
4510 - HYPERV_CPUID_ENLIGHTMENT_INFO
4511 - HYPERV_CPUID_IMPLEMENT_LIMITS
4512 - HYPERV_CPUID_NESTED_FEATURES
4513 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS
4514 - HYPERV_CPUID_SYNDBG_INTERFACE
4515 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES
4517 HYPERV_CPUID_NESTED_FEATURES leaf is only exposed when Enlightened VMCS was
4518 enabled on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS).
4520 Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure
4521 with the 'nent' field indicating the number of entries in the variable-size
4522 array 'entries'. If the number of entries is too low to describe all Hyper-V
4523 feature leaves, an error (E2BIG) is returned. If the number is more or equal
4524 to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the
4525 number of valid entries in the 'entries' array, which is then filled.
4527 'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved,
4528 userspace should not expect to get any particular value there.
4530 4.119 KVM_ARM_VCPU_FINALIZE
4531 ---------------------------
4533 :Architectures: arm, arm64
4535 :Parameters: int feature (in)
4536 :Returns: 0 on success, -1 on error
4540 ====== ==============================================================
4541 EPERM feature not enabled, needs configuration, or already finalized
4542 EINVAL feature unknown or not present
4543 ====== ==============================================================
4545 Recognised values for feature:
4547 ===== ===========================================
4548 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE)
4549 ===== ===========================================
4551 Finalizes the configuration of the specified vcpu feature.
4553 The vcpu must already have been initialised, enabling the affected feature, by
4554 means of a successful KVM_ARM_VCPU_INIT call with the appropriate flag set in
4557 For affected vcpu features, this is a mandatory step that must be performed
4558 before the vcpu is fully usable.
4560 Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be
4561 configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration
4562 that should be performaned and how to do it are feature-dependent.
4564 Other calls that depend on a particular feature being finalized, such as
4565 KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with
4566 -EPERM unless the feature has already been finalized by means of a
4567 KVM_ARM_VCPU_FINALIZE call.
4569 See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization
4572 4.120 KVM_SET_PMU_EVENT_FILTER
4573 ------------------------------
4575 :Capability: KVM_CAP_PMU_EVENT_FILTER
4578 :Parameters: struct kvm_pmu_event_filter (in)
4579 :Returns: 0 on success, -1 on error
4583 struct kvm_pmu_event_filter {
4586 __u32 fixed_counter_bitmap;
4592 This ioctl restricts the set of PMU events that the guest can program.
4593 The argument holds a list of events which will be allowed or denied.
4594 The eventsel+umask of each event the guest attempts to program is compared
4595 against the events field to determine whether the guest should have access.
4596 The events field only controls general purpose counters; fixed purpose
4597 counters are controlled by the fixed_counter_bitmap.
4599 No flags are defined yet, the field must be zero.
4601 Valid values for 'action'::
4603 #define KVM_PMU_EVENT_ALLOW 0
4604 #define KVM_PMU_EVENT_DENY 1
4606 4.121 KVM_PPC_SVM_OFF
4607 ---------------------
4610 :Architectures: powerpc
4613 :Returns: 0 on successful completion,
4617 ====== ================================================================
4618 EINVAL if ultravisor failed to terminate the secure guest
4619 ENOMEM if hypervisor failed to allocate new radix page tables for guest
4620 ====== ================================================================
4622 This ioctl is used to turn off the secure mode of the guest or transition
4623 the guest from secure mode to normal mode. This is invoked when the guest
4624 is reset. This has no effect if called for a normal guest.
4626 This ioctl issues an ultravisor call to terminate the secure guest,
4627 unpins the VPA pages and releases all the device pages that are used to
4628 track the secure pages by hypervisor.
4630 4.122 KVM_S390_NORMAL_RESET
4631 ---------------------------
4633 :Capability: KVM_CAP_S390_VCPU_RESETS
4634 :Architectures: s390
4639 This ioctl resets VCPU registers and control structures according to
4640 the cpu reset definition in the POP (Principles Of Operation).
4642 4.123 KVM_S390_INITIAL_RESET
4643 ----------------------------
4646 :Architectures: s390
4651 This ioctl resets VCPU registers and control structures according to
4652 the initial cpu reset definition in the POP. However, the cpu is not
4653 put into ESA mode. This reset is a superset of the normal reset.
4655 4.124 KVM_S390_CLEAR_RESET
4656 --------------------------
4658 :Capability: KVM_CAP_S390_VCPU_RESETS
4659 :Architectures: s390
4664 This ioctl resets VCPU registers and control structures according to
4665 the clear cpu reset definition in the POP. However, the cpu is not put
4666 into ESA mode. This reset is a superset of the initial reset.
4669 4.125 KVM_S390_PV_COMMAND
4670 -------------------------
4672 :Capability: KVM_CAP_S390_PROTECTED
4673 :Architectures: s390
4675 :Parameters: struct kvm_pv_cmd
4676 :Returns: 0 on success, < 0 on error
4681 __u32 cmd; /* Command to be executed */
4682 __u16 rc; /* Ultravisor return code */
4683 __u16 rrc; /* Ultravisor return reason code */
4684 __u64 data; /* Data or address */
4685 __u32 flags; /* flags for future extensions. Must be 0 for now */
4692 Allocate memory and register the VM with the Ultravisor, thereby
4693 donating memory to the Ultravisor that will become inaccessible to
4694 KVM. All existing CPUs are converted to protected ones. After this
4695 command has succeeded, any CPU added via hotplug will become
4696 protected during its creation as well.
4700 ===== =============================
4701 EINTR an unmasked signal is pending
4702 ===== =============================
4706 Deregister the VM from the Ultravisor and reclaim the memory that
4707 had been donated to the Ultravisor, making it usable by the kernel
4708 again. All registered VCPUs are converted back to non-protected
4711 KVM_PV_VM_SET_SEC_PARMS
4712 Pass the image header from VM memory to the Ultravisor in
4713 preparation of image unpacking and verification.
4716 Unpack (protect and decrypt) a page of the encrypted boot image.
4719 Verify the integrity of the unpacked image. Only if this succeeds,
4720 KVM is allowed to start protected VCPUs.
4722 4.126 KVM_X86_SET_MSR_FILTER
4723 ----------------------------
4725 :Capability: KVM_X86_SET_MSR_FILTER
4728 :Parameters: struct kvm_msr_filter
4729 :Returns: 0 on success, < 0 on error
4733 struct kvm_msr_filter_range {
4734 #define KVM_MSR_FILTER_READ (1 << 0)
4735 #define KVM_MSR_FILTER_WRITE (1 << 1)
4737 __u32 nmsrs; /* number of msrs in bitmap */
4738 __u32 base; /* MSR index the bitmap starts at */
4739 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */
4742 #define KVM_MSR_FILTER_MAX_RANGES 16
4743 struct kvm_msr_filter {
4744 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0)
4745 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0)
4747 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES];
4750 flags values for ``struct kvm_msr_filter_range``:
4752 ``KVM_MSR_FILTER_READ``
4754 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap
4755 indicates that a read should immediately fail, while a 1 indicates that
4756 a read for a particular MSR should be handled regardless of the default
4759 ``KVM_MSR_FILTER_WRITE``
4761 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap
4762 indicates that a write should immediately fail, while a 1 indicates that
4763 a write for a particular MSR should be handled regardless of the default
4766 ``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE``
4768 Filter both read and write accesses to MSRs using the given bitmap. A 0
4769 in the bitmap indicates that both reads and writes should immediately fail,
4770 while a 1 indicates that reads and writes for a particular MSR are not
4771 filtered by this range.
4773 flags values for ``struct kvm_msr_filter``:
4775 ``KVM_MSR_FILTER_DEFAULT_ALLOW``
4777 If no filter range matches an MSR index that is getting accessed, KVM will
4778 fall back to allowing access to the MSR.
4780 ``KVM_MSR_FILTER_DEFAULT_DENY``
4782 If no filter range matches an MSR index that is getting accessed, KVM will
4783 fall back to rejecting access to the MSR. In this mode, all MSRs that should
4784 be processed by KVM need to explicitly be marked as allowed in the bitmaps.
4786 This ioctl allows user space to define up to 16 bitmaps of MSR ranges to
4787 specify whether a certain MSR access should be explicitly filtered for or not.
4789 If this ioctl has never been invoked, MSR accesses are not guarded and the
4790 default KVM in-kernel emulation behavior is fully preserved.
4792 Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR
4793 filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes
4796 As soon as the filtering is in place, every MSR access is processed through
4797 the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff);
4798 x2APIC MSRs are always allowed, independent of the ``default_allow`` setting,
4799 and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base
4802 If a bit is within one of the defined ranges, read and write accesses are
4803 guarded by the bitmap's value for the MSR index if the kind of access
4804 is included in the ``struct kvm_msr_filter_range`` flags. If no range
4805 cover this particular access, the behavior is determined by the flags
4806 field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW``
4807 and ``KVM_MSR_FILTER_DEFAULT_DENY``.
4809 Each bitmap range specifies a range of MSRs to potentially allow access on.
4810 The range goes from MSR index [base .. base+nmsrs]. The flags field
4811 indicates whether reads, writes or both reads and writes are filtered
4812 by setting a 1 bit in the bitmap for the corresponding MSR index.
4814 If an MSR access is not permitted through the filtering, it generates a
4815 #GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that
4816 allows user space to deflect and potentially handle various MSR accesses
4819 Note, invoking this ioctl with a vCPU is running is inherently racy. However,
4820 KVM does guarantee that vCPUs will see either the previous filter or the new
4821 filter, e.g. MSRs with identical settings in both the old and new filter will
4822 have deterministic behavior.
4825 5. The kvm_run structure
4826 ========================
4828 Application code obtains a pointer to the kvm_run structure by
4829 mmap()ing a vcpu fd. From that point, application code can control
4830 execution by changing fields in kvm_run prior to calling the KVM_RUN
4831 ioctl, and obtain information about the reason KVM_RUN returned by
4832 looking up structure members.
4838 __u8 request_interrupt_window;
4840 Request that KVM_RUN return when it becomes possible to inject external
4841 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
4845 __u8 immediate_exit;
4847 This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN
4848 exits immediately, returning -EINTR. In the common scenario where a
4849 signal is used to "kick" a VCPU out of KVM_RUN, this field can be used
4850 to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability.
4851 Rather than blocking the signal outside KVM_RUN, userspace can set up
4852 a signal handler that sets run->immediate_exit to a non-zero value.
4854 This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available.
4863 When KVM_RUN has returned successfully (return value 0), this informs
4864 application code why KVM_RUN has returned. Allowable values for this
4865 field are detailed below.
4869 __u8 ready_for_interrupt_injection;
4871 If request_interrupt_window has been specified, this field indicates
4872 an interrupt can be injected now with KVM_INTERRUPT.
4878 The value of the current interrupt flag. Only valid if in-kernel
4879 local APIC is not used.
4885 More architecture-specific flags detailing state of the VCPU that may
4886 affect the device's behavior. The only currently defined flag is
4887 KVM_RUN_X86_SMM, which is valid on x86 machines and is set if the
4888 VCPU is in system management mode.
4892 /* in (pre_kvm_run), out (post_kvm_run) */
4895 The value of the cr8 register. Only valid if in-kernel local APIC is
4896 not used. Both input and output.
4902 The value of the APIC BASE msr. Only valid if in-kernel local
4903 APIC is not used. Both input and output.
4908 /* KVM_EXIT_UNKNOWN */
4910 __u64 hardware_exit_reason;
4913 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
4914 reasons. Further architecture-specific information is available in
4915 hardware_exit_reason.
4919 /* KVM_EXIT_FAIL_ENTRY */
4921 __u64 hardware_entry_failure_reason;
4922 __u32 cpu; /* if KVM_LAST_CPU */
4925 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
4926 to unknown reasons. Further architecture-specific information is
4927 available in hardware_entry_failure_reason.
4931 /* KVM_EXIT_EXCEPTION */
4943 #define KVM_EXIT_IO_IN 0
4944 #define KVM_EXIT_IO_OUT 1
4946 __u8 size; /* bytes */
4949 __u64 data_offset; /* relative to kvm_run start */
4952 If exit_reason is KVM_EXIT_IO, then the vcpu has
4953 executed a port I/O instruction which could not be satisfied by kvm.
4954 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
4955 where kvm expects application code to place the data for the next
4956 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
4960 /* KVM_EXIT_DEBUG */
4962 struct kvm_debug_exit_arch arch;
4965 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event
4966 for which architecture specific information is returned.
4978 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
4979 executed a memory-mapped I/O instruction which could not be satisfied
4980 by kvm. The 'data' member contains the written data if 'is_write' is
4981 true, and should be filled by application code otherwise.
4983 The 'data' member contains, in its first 'len' bytes, the value as it would
4984 appear if the VCPU performed a load or store of the appropriate width directly
4989 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR,
4990 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding
4991 operations are complete (and guest state is consistent) only after userspace
4992 has re-entered the kernel with KVM_RUN. The kernel side will first finish
4993 incomplete operations and then check for pending signals. Userspace
4994 can re-enter the guest with an unmasked signal pending to complete
4999 /* KVM_EXIT_HYPERCALL */
5008 Unused. This was once used for 'hypercall to userspace'. To implement
5009 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390).
5011 .. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
5015 /* KVM_EXIT_TPR_ACCESS */
5022 To be documented (KVM_TPR_ACCESS_REPORTING).
5026 /* KVM_EXIT_S390_SIEIC */
5029 __u64 mask; /* psw upper half */
5030 __u64 addr; /* psw lower half */
5039 /* KVM_EXIT_S390_RESET */
5040 #define KVM_S390_RESET_POR 1
5041 #define KVM_S390_RESET_CLEAR 2
5042 #define KVM_S390_RESET_SUBSYSTEM 4
5043 #define KVM_S390_RESET_CPU_INIT 8
5044 #define KVM_S390_RESET_IPL 16
5045 __u64 s390_reset_flags;
5051 /* KVM_EXIT_S390_UCONTROL */
5053 __u64 trans_exc_code;
5057 s390 specific. A page fault has occurred for a user controlled virtual
5058 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be
5059 resolved by the kernel.
5060 The program code and the translation exception code that were placed
5061 in the cpu's lowcore are presented here as defined by the z Architecture
5062 Principles of Operation Book in the Chapter for Dynamic Address Translation
5074 Deprecated - was used for 440 KVM.
5083 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
5084 hypercalls and exit with this exit struct that contains all the guest gprs.
5086 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
5087 Userspace can now handle the hypercall and when it's done modify the gprs as
5088 necessary. Upon guest entry all guest GPRs will then be replaced by the values
5093 /* KVM_EXIT_PAPR_HCALL */
5100 This is used on 64-bit PowerPC when emulating a pSeries partition,
5101 e.g. with the 'pseries' machine type in qemu. It occurs when the
5102 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
5103 contains the hypercall number (from the guest R3), and 'args' contains
5104 the arguments (from the guest R4 - R12). Userspace should put the
5105 return code in 'ret' and any extra returned values in args[].
5106 The possible hypercalls are defined in the Power Architecture Platform
5107 Requirements (PAPR) document available from www.power.org (free
5108 developer registration required to access it).
5112 /* KVM_EXIT_S390_TSCH */
5114 __u16 subchannel_id;
5115 __u16 subchannel_nr;
5122 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
5123 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
5124 interrupt for the target subchannel has been dequeued and subchannel_id,
5125 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
5126 interrupt. ipb is needed for instruction parameter decoding.
5135 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
5136 interrupt acknowledge path to the core. When the core successfully
5137 delivers an interrupt, it automatically populates the EPR register with
5138 the interrupt vector number and acknowledges the interrupt inside
5139 the interrupt controller.
5141 In case the interrupt controller lives in user space, we need to do
5142 the interrupt acknowledge cycle through it to fetch the next to be
5143 delivered interrupt vector using this exit.
5145 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
5146 external interrupt has just been delivered into the guest. User space
5147 should put the acknowledged interrupt vector into the 'epr' field.
5151 /* KVM_EXIT_SYSTEM_EVENT */
5153 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
5154 #define KVM_SYSTEM_EVENT_RESET 2
5155 #define KVM_SYSTEM_EVENT_CRASH 3
5160 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
5161 a system-level event using some architecture specific mechanism (hypercall
5162 or some special instruction). In case of ARM/ARM64, this is triggered using
5163 HVC instruction based PSCI call from the vcpu. The 'type' field describes
5164 the system-level event type. The 'flags' field describes architecture
5165 specific flags for the system-level event.
5167 Valid values for 'type' are:
5169 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
5170 VM. Userspace is not obliged to honour this, and if it does honour
5171 this does not need to destroy the VM synchronously (ie it may call
5172 KVM_RUN again before shutdown finally occurs).
5173 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
5174 As with SHUTDOWN, userspace can choose to ignore the request, or
5175 to schedule the reset to occur in the future and may call KVM_RUN again.
5176 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest
5177 has requested a crash condition maintenance. Userspace can choose
5178 to ignore the request, or to gather VM memory core dump and/or
5179 reset/shutdown of the VM.
5183 /* KVM_EXIT_IOAPIC_EOI */
5188 Indicates that the VCPU's in-kernel local APIC received an EOI for a
5189 level-triggered IOAPIC interrupt. This exit only triggers when the
5190 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled);
5191 the userspace IOAPIC should process the EOI and retrigger the interrupt if
5192 it is still asserted. Vector is the LAPIC interrupt vector for which the
5197 struct kvm_hyperv_exit {
5198 #define KVM_EXIT_HYPERV_SYNIC 1
5199 #define KVM_EXIT_HYPERV_HCALL 2
5200 #define KVM_EXIT_HYPERV_SYNDBG 3
5227 /* KVM_EXIT_HYPERV */
5228 struct kvm_hyperv_exit hyperv;
5230 Indicates that the VCPU exits into userspace to process some tasks
5231 related to Hyper-V emulation.
5233 Valid values for 'type' are:
5235 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about
5237 Hyper-V SynIC state change. Notification is used to remap SynIC
5238 event/message pages and to enable/disable SynIC messages/events processing
5241 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about
5243 Hyper-V Synthetic debugger state change. Notification is used to either update
5244 the pending_page location or to send a control command (send the buffer located
5245 in send_page or recv a buffer to recv_page).
5249 /* KVM_EXIT_ARM_NISV */
5255 Used on arm and arm64 systems. If a guest accesses memory not in a memslot,
5256 KVM will typically return to userspace and ask it to do MMIO emulation on its
5257 behalf. However, for certain classes of instructions, no instruction decode
5258 (direction, length of memory access) is provided, and fetching and decoding
5259 the instruction from the VM is overly complicated to live in the kernel.
5261 Historically, when this situation occurred, KVM would print a warning and kill
5262 the VM. KVM assumed that if the guest accessed non-memslot memory, it was
5263 trying to do I/O, which just couldn't be emulated, and the warning message was
5264 phrased accordingly. However, what happened more often was that a guest bug
5265 caused access outside the guest memory areas which should lead to a more
5266 meaningful warning message and an external abort in the guest, if the access
5267 did not fall within an I/O window.
5269 Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable
5270 this capability at VM creation. Once this is done, these types of errors will
5271 instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from
5272 the HSR (arm) and ESR_EL2 (arm64) in the esr_iss field, and the faulting IPA
5273 in the fault_ipa field. Userspace can either fix up the access if it's
5274 actually an I/O access by decoding the instruction from guest memory (if it's
5275 very brave) and continue executing the guest, or it can decide to suspend,
5276 dump, or restart the guest.
5278 Note that KVM does not skip the faulting instruction as it does for
5279 KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state
5280 if it decides to decode and emulate the instruction.
5284 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */
5286 __u8 error; /* user -> kernel */
5288 __u32 reason; /* kernel -> user */
5289 __u32 index; /* kernel -> user */
5290 __u64 data; /* kernel <-> user */
5293 Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is
5294 enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code
5295 will instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR
5298 The "reason" field specifies why the MSR trap occurred. User space will only
5299 receive MSR exit traps when a particular reason was requested during through
5300 ENABLE_CAP. Currently valid exit reasons are:
5302 KVM_MSR_EXIT_REASON_UNKNOWN - access to MSR that is unknown to KVM
5303 KVM_MSR_EXIT_REASON_INVAL - access to invalid MSRs or reserved bits
5304 KVM_MSR_EXIT_REASON_FILTER - access blocked by KVM_X86_SET_MSR_FILTER
5306 For KVM_EXIT_X86_RDMSR, the "index" field tells user space which MSR the guest
5307 wants to read. To respond to this request with a successful read, user space
5308 writes the respective data into the "data" field and must continue guest
5309 execution to ensure the read data is transferred into guest register state.
5311 If the RDMSR request was unsuccessful, user space indicates that with a "1" in
5312 the "error" field. This will inject a #GP into the guest when the VCPU is
5315 For KVM_EXIT_X86_WRMSR, the "index" field tells user space which MSR the guest
5316 wants to write. Once finished processing the event, user space must continue
5317 vCPU execution. If the MSR write was unsuccessful, user space also sets the
5318 "error" field to "1".
5322 /* Fix the size of the union. */
5327 * shared registers between kvm and userspace.
5328 * kvm_valid_regs specifies the register classes set by the host
5329 * kvm_dirty_regs specified the register classes dirtied by userspace
5330 * struct kvm_sync_regs is architecture specific, as well as the
5331 * bits for kvm_valid_regs and kvm_dirty_regs
5333 __u64 kvm_valid_regs;
5334 __u64 kvm_dirty_regs;
5336 struct kvm_sync_regs regs;
5337 char padding[SYNC_REGS_SIZE_BYTES];
5340 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
5341 certain guest registers without having to call SET/GET_*REGS. Thus we can
5342 avoid some system call overhead if userspace has to handle the exit.
5343 Userspace can query the validity of the structure by checking
5344 kvm_valid_regs for specific bits. These bits are architecture specific
5345 and usually define the validity of a groups of registers. (e.g. one bit
5346 for general purpose registers)
5348 Please note that the kernel is allowed to use the kvm_run structure as the
5349 primary storage for certain register types. Therefore, the kernel may use the
5350 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
5358 6. Capabilities that can be enabled on vCPUs
5359 ============================================
5361 There are certain capabilities that change the behavior of the virtual CPU or
5362 the virtual machine when enabled. To enable them, please see section 4.37.
5363 Below you can find a list of capabilities and what their effect on the vCPU or
5364 the virtual machine is when enabling them.
5366 The following information is provided along with the description:
5369 which instruction set architectures provide this ioctl.
5370 x86 includes both i386 and x86_64.
5373 whether this is a per-vcpu or per-vm capability.
5376 what parameters are accepted by the capability.
5379 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
5380 are not detailed, but errors with specific meanings are.
5389 :Returns: 0 on success; -1 on error
5391 This capability enables interception of OSI hypercalls that otherwise would
5392 be treated as normal system calls to be injected into the guest. OSI hypercalls
5393 were invented by Mac-on-Linux to have a standardized communication mechanism
5394 between the guest and the host.
5396 When this capability is enabled, KVM_EXIT_OSI can occur.
5399 6.2 KVM_CAP_PPC_PAPR
5400 --------------------
5405 :Returns: 0 on success; -1 on error
5407 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
5408 done using the hypercall instruction "sc 1".
5410 It also sets the guest privilege level to "supervisor" mode. Usually the guest
5411 runs in "hypervisor" privilege mode with a few missing features.
5413 In addition to the above, it changes the semantics of SDR1. In this mode, the
5414 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
5415 HTAB invisible to the guest.
5417 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
5425 :Parameters: args[0] is the address of a struct kvm_config_tlb
5426 :Returns: 0 on success; -1 on error
5430 struct kvm_config_tlb {
5437 Configures the virtual CPU's TLB array, establishing a shared memory area
5438 between userspace and KVM. The "params" and "array" fields are userspace
5439 addresses of mmu-type-specific data structures. The "array_len" field is an
5440 safety mechanism, and should be set to the size in bytes of the memory that
5441 userspace has reserved for the array. It must be at least the size dictated
5442 by "mmu_type" and "params".
5444 While KVM_RUN is active, the shared region is under control of KVM. Its
5445 contents are undefined, and any modification by userspace results in
5446 boundedly undefined behavior.
5448 On return from KVM_RUN, the shared region will reflect the current state of
5449 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
5450 to tell KVM which entries have been changed, prior to calling KVM_RUN again
5453 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
5455 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
5456 - The "array" field points to an array of type "struct
5457 kvm_book3e_206_tlb_entry".
5458 - The array consists of all entries in the first TLB, followed by all
5459 entries in the second TLB.
5460 - Within a TLB, entries are ordered first by increasing set number. Within a
5461 set, entries are ordered by way (increasing ESEL).
5462 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
5463 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
5464 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
5465 hardware ignores this value for TLB0.
5467 6.4 KVM_CAP_S390_CSS_SUPPORT
5468 ----------------------------
5470 :Architectures: s390
5473 :Returns: 0 on success; -1 on error
5475 This capability enables support for handling of channel I/O instructions.
5477 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
5478 handled in-kernel, while the other I/O instructions are passed to userspace.
5480 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
5481 SUBCHANNEL intercepts.
5483 Note that even though this capability is enabled per-vcpu, the complete
5484 virtual machine is affected.
5491 :Parameters: args[0] defines whether the proxy facility is active
5492 :Returns: 0 on success; -1 on error
5494 This capability enables or disables the delivery of interrupts through the
5495 external proxy facility.
5497 When enabled (args[0] != 0), every time the guest gets an external interrupt
5498 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
5499 to receive the topmost interrupt vector.
5501 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
5503 When this capability is enabled, KVM_EXIT_EPR can occur.
5505 6.6 KVM_CAP_IRQ_MPIC
5506 --------------------
5509 :Parameters: args[0] is the MPIC device fd;
5510 args[1] is the MPIC CPU number for this vcpu
5512 This capability connects the vcpu to an in-kernel MPIC device.
5514 6.7 KVM_CAP_IRQ_XICS
5515 --------------------
5519 :Parameters: args[0] is the XICS device fd;
5520 args[1] is the XICS CPU number (server ID) for this vcpu
5522 This capability connects the vcpu to an in-kernel XICS device.
5524 6.8 KVM_CAP_S390_IRQCHIP
5525 ------------------------
5527 :Architectures: s390
5531 This capability enables the in-kernel irqchip for s390. Please refer to
5532 "4.24 KVM_CREATE_IRQCHIP" for details.
5534 6.9 KVM_CAP_MIPS_FPU
5535 --------------------
5537 :Architectures: mips
5539 :Parameters: args[0] is reserved for future use (should be 0).
5541 This capability allows the use of the host Floating Point Unit by the guest. It
5542 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is
5543 done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be
5544 accessed (depending on the current guest FPU register mode), and the Status.FR,
5545 Config5.FRE bits are accessible via the KVM API and also from the guest,
5546 depending on them being supported by the FPU.
5548 6.10 KVM_CAP_MIPS_MSA
5549 ---------------------
5551 :Architectures: mips
5553 :Parameters: args[0] is reserved for future use (should be 0).
5555 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest.
5556 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest.
5557 Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*``
5558 registers can be accessed, and the Config5.MSAEn bit is accessible via the
5559 KVM API and also from the guest.
5561 6.74 KVM_CAP_SYNC_REGS
5562 ----------------------
5564 :Architectures: s390, x86
5565 :Target: s390: always enabled, x86: vcpu
5567 :Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register
5569 (bitfields defined in arch/x86/include/uapi/asm/kvm.h).
5571 As described above in the kvm_sync_regs struct info in section 5 (kvm_run):
5572 KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers
5573 without having to call SET/GET_*REGS". This reduces overhead by eliminating
5574 repeated ioctl calls for setting and/or getting register values. This is
5575 particularly important when userspace is making synchronous guest state
5576 modifications, e.g. when emulating and/or intercepting instructions in
5579 For s390 specifics, please refer to the source code.
5583 - the register sets to be copied out to kvm_run are selectable
5584 by userspace (rather that all sets being copied out for every exit).
5585 - vcpu_events are available in addition to regs and sregs.
5587 For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to
5588 function as an input bit-array field set by userspace to indicate the
5589 specific register sets to be copied out on the next exit.
5591 To indicate when userspace has modified values that should be copied into
5592 the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set.
5593 This is done using the same bitflags as for the 'kvm_valid_regs' field.
5594 If the dirty bit is not set, then the register set values will not be copied
5595 into the vCPU even if they've been modified.
5597 Unused bitfields in the bitarrays must be set to zero.
5601 struct kvm_sync_regs {
5602 struct kvm_regs regs;
5603 struct kvm_sregs sregs;
5604 struct kvm_vcpu_events events;
5607 6.75 KVM_CAP_PPC_IRQ_XIVE
5608 -------------------------
5612 :Parameters: args[0] is the XIVE device fd;
5613 args[1] is the XIVE CPU number (server ID) for this vcpu
5615 This capability connects the vcpu to an in-kernel XIVE device.
5617 7. Capabilities that can be enabled on VMs
5618 ==========================================
5620 There are certain capabilities that change the behavior of the virtual
5621 machine when enabled. To enable them, please see section 4.37. Below
5622 you can find a list of capabilities and what their effect on the VM
5623 is when enabling them.
5625 The following information is provided along with the description:
5628 which instruction set architectures provide this ioctl.
5629 x86 includes both i386 and x86_64.
5632 what parameters are accepted by the capability.
5635 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
5636 are not detailed, but errors with specific meanings are.
5639 7.1 KVM_CAP_PPC_ENABLE_HCALL
5640 ----------------------------
5643 :Parameters: args[0] is the sPAPR hcall number;
5644 args[1] is 0 to disable, 1 to enable in-kernel handling
5646 This capability controls whether individual sPAPR hypercalls (hcalls)
5647 get handled by the kernel or not. Enabling or disabling in-kernel
5648 handling of an hcall is effective across the VM. On creation, an
5649 initial set of hcalls are enabled for in-kernel handling, which
5650 consists of those hcalls for which in-kernel handlers were implemented
5651 before this capability was implemented. If disabled, the kernel will
5652 not to attempt to handle the hcall, but will always exit to userspace
5653 to handle it. Note that it may not make sense to enable some and
5654 disable others of a group of related hcalls, but KVM does not prevent
5655 userspace from doing that.
5657 If the hcall number specified is not one that has an in-kernel
5658 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
5661 7.2 KVM_CAP_S390_USER_SIGP
5662 --------------------------
5664 :Architectures: s390
5667 This capability controls which SIGP orders will be handled completely in user
5668 space. With this capability enabled, all fast orders will be handled completely
5675 - CONDITIONAL EMERGENCY SIGNAL
5677 All other orders will be handled completely in user space.
5679 Only privileged operation exceptions will be checked for in the kernel (or even
5680 in the hardware prior to interception). If this capability is not enabled, the
5681 old way of handling SIGP orders is used (partially in kernel and user space).
5683 7.3 KVM_CAP_S390_VECTOR_REGISTERS
5684 ---------------------------------
5686 :Architectures: s390
5688 :Returns: 0 on success, negative value on error
5690 Allows use of the vector registers introduced with z13 processor, and
5691 provides for the synchronization between host and user space. Will
5692 return -EINVAL if the machine does not support vectors.
5694 7.4 KVM_CAP_S390_USER_STSI
5695 --------------------------
5697 :Architectures: s390
5700 This capability allows post-handlers for the STSI instruction. After
5701 initial handling in the kernel, KVM exits to user space with
5702 KVM_EXIT_S390_STSI to allow user space to insert further data.
5704 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
5716 @addr - guest address of STSI SYSIB
5720 @ar - access register number
5722 KVM handlers should exit to userspace with rc = -EREMOTE.
5724 7.5 KVM_CAP_SPLIT_IRQCHIP
5725 -------------------------
5728 :Parameters: args[0] - number of routes reserved for userspace IOAPICs
5729 :Returns: 0 on success, -1 on error
5731 Create a local apic for each processor in the kernel. This can be used
5732 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the
5733 IOAPIC and PIC (and also the PIT, even though this has to be enabled
5736 This capability also enables in kernel routing of interrupt requests;
5737 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are
5738 used in the IRQ routing table. The first args[0] MSI routes are reserved
5739 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes,
5740 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace.
5742 Fails if VCPU has already been created, or if the irqchip is already in the
5743 kernel (i.e. KVM_CREATE_IRQCHIP has already been called).
5748 :Architectures: s390
5751 Allows use of runtime-instrumentation introduced with zEC12 processor.
5752 Will return -EINVAL if the machine does not support runtime-instrumentation.
5753 Will return -EBUSY if a VCPU has already been created.
5755 7.7 KVM_CAP_X2APIC_API
5756 ----------------------
5759 :Parameters: args[0] - features that should be enabled
5760 :Returns: 0 on success, -EINVAL when args[0] contains invalid features
5762 Valid feature flags in args[0] are::
5764 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0)
5765 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1)
5767 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of
5768 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC,
5769 allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their
5770 respective sections.
5772 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work
5773 in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff
5774 as a broadcast even in x2APIC mode in order to support physical x2APIC
5775 without interrupt remapping. This is undesirable in logical mode,
5776 where 0xff represents CPUs 0-7 in cluster 0.
5778 7.8 KVM_CAP_S390_USER_INSTR0
5779 ----------------------------
5781 :Architectures: s390
5784 With this capability enabled, all illegal instructions 0x0000 (2 bytes) will
5785 be intercepted and forwarded to user space. User space can use this
5786 mechanism e.g. to realize 2-byte software breakpoints. The kernel will
5787 not inject an operating exception for these instructions, user space has
5788 to take care of that.
5790 This capability can be enabled dynamically even if VCPUs were already
5791 created and are running.
5796 :Architectures: s390
5798 :Returns: 0 on success; -EINVAL if the machine does not support
5799 guarded storage; -EBUSY if a VCPU has already been created.
5801 Allows use of guarded storage for the KVM guest.
5803 7.10 KVM_CAP_S390_AIS
5804 ---------------------
5806 :Architectures: s390
5809 Allow use of adapter-interruption suppression.
5810 :Returns: 0 on success; -EBUSY if a VCPU has already been created.
5812 7.11 KVM_CAP_PPC_SMT
5813 --------------------
5816 :Parameters: vsmt_mode, flags
5818 Enabling this capability on a VM provides userspace with a way to set
5819 the desired virtual SMT mode (i.e. the number of virtual CPUs per
5820 virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2
5821 between 1 and 8. On POWER8, vsmt_mode must also be no greater than
5822 the number of threads per subcore for the host. Currently flags must
5823 be 0. A successful call to enable this capability will result in
5824 vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is
5825 subsequently queried for the VM. This capability is only supported by
5826 HV KVM, and can only be set before any VCPUs have been created.
5827 The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT
5828 modes are available.
5830 7.12 KVM_CAP_PPC_FWNMI
5831 ----------------------
5836 With this capability a machine check exception in the guest address
5837 space will cause KVM to exit the guest with NMI exit reason. This
5838 enables QEMU to build error log and branch to guest kernel registered
5839 machine check handling routine. Without this capability KVM will
5840 branch to guests' 0x200 interrupt vector.
5842 7.13 KVM_CAP_X86_DISABLE_EXITS
5843 ------------------------------
5846 :Parameters: args[0] defines which exits are disabled
5847 :Returns: 0 on success, -EINVAL when args[0] contains invalid exits
5849 Valid bits in args[0] are::
5851 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0)
5852 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1)
5853 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2)
5854 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3)
5856 Enabling this capability on a VM provides userspace with a way to no
5857 longer intercept some instructions for improved latency in some
5858 workloads, and is suggested when vCPUs are associated to dedicated
5859 physical CPUs. More bits can be added in the future; userspace can
5860 just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable
5863 Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits.
5865 7.14 KVM_CAP_S390_HPAGE_1M
5866 --------------------------
5868 :Architectures: s390
5870 :Returns: 0 on success, -EINVAL if hpage module parameter was not set
5871 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL
5874 With this capability the KVM support for memory backing with 1m pages
5875 through hugetlbfs can be enabled for a VM. After the capability is
5876 enabled, cmma can't be enabled anymore and pfmfi and the storage key
5877 interpretation are disabled. If cmma has already been enabled or the
5878 hpage module parameter is not set to 1, -EINVAL is returned.
5880 While it is generally possible to create a huge page backed VM without
5881 this capability, the VM will not be able to run.
5883 7.15 KVM_CAP_MSR_PLATFORM_INFO
5884 ------------------------------
5887 :Parameters: args[0] whether feature should be enabled or not
5889 With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise,
5890 a #GP would be raised when the guest tries to access. Currently, this
5891 capability does not enable write permissions of this MSR for the guest.
5893 7.16 KVM_CAP_PPC_NESTED_HV
5894 --------------------------
5898 :Returns: 0 on success, -EINVAL when the implementation doesn't support
5899 nested-HV virtualization.
5901 HV-KVM on POWER9 and later systems allows for "nested-HV"
5902 virtualization, which provides a way for a guest VM to run guests that
5903 can run using the CPU's supervisor mode (privileged non-hypervisor
5904 state). Enabling this capability on a VM depends on the CPU having
5905 the necessary functionality and on the facility being enabled with a
5906 kvm-hv module parameter.
5908 7.17 KVM_CAP_EXCEPTION_PAYLOAD
5909 ------------------------------
5912 :Parameters: args[0] whether feature should be enabled or not
5914 With this capability enabled, CR2 will not be modified prior to the
5915 emulated VM-exit when L1 intercepts a #PF exception that occurs in
5916 L2. Similarly, for kvm-intel only, DR6 will not be modified prior to
5917 the emulated VM-exit when L1 intercepts a #DB exception that occurs in
5918 L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or
5919 #DB) exception for L2, exception.has_payload will be set and the
5920 faulting address (or the new DR6 bits*) will be reported in the
5921 exception_payload field. Similarly, when userspace injects a #PF (or
5922 #DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set
5923 exception.has_payload and to put the faulting address - or the new DR6
5924 bits\ [#]_ - in the exception_payload field.
5926 This capability also enables exception.pending in struct
5927 kvm_vcpu_events, which allows userspace to distinguish between pending
5928 and injected exceptions.
5931 .. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception
5934 7.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
5936 :Architectures: x86, arm, arm64, mips
5937 :Parameters: args[0] whether feature should be enabled or not
5941 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0)
5942 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1)
5944 With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not
5945 automatically clear and write-protect all pages that are returned as dirty.
5946 Rather, userspace will have to do this operation separately using
5947 KVM_CLEAR_DIRTY_LOG.
5949 At the cost of a slightly more complicated operation, this provides better
5950 scalability and responsiveness for two reasons. First,
5951 KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather
5952 than requiring to sync a full memslot; this ensures that KVM does not
5953 take spinlocks for an extended period of time. Second, in some cases a
5954 large amount of time can pass between a call to KVM_GET_DIRTY_LOG and
5955 userspace actually using the data in the page. Pages can be modified
5956 during this time, which is inefficient for both the guest and userspace:
5957 the guest will incur a higher penalty due to write protection faults,
5958 while userspace can see false reports of dirty pages. Manual reprotection
5959 helps reducing this time, improving guest performance and reducing the
5960 number of dirty log false positives.
5962 With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap
5963 will be initialized to 1 when created. This also improves performance because
5964 dirty logging can be enabled gradually in small chunks on the first call
5965 to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on
5966 KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on
5967 x86 and arm64 for now).
5969 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name
5970 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make
5971 it hard or impossible to use it correctly. The availability of
5972 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed.
5973 Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT.
5975 7.19 KVM_CAP_PPC_SECURE_GUEST
5976 ------------------------------
5980 This capability indicates that KVM is running on a host that has
5981 ultravisor firmware and thus can support a secure guest. On such a
5982 system, a guest can ask the ultravisor to make it a secure guest,
5983 one whose memory is inaccessible to the host except for pages which
5984 are explicitly requested to be shared with the host. The ultravisor
5985 notifies KVM when a guest requests to become a secure guest, and KVM
5986 has the opportunity to veto the transition.
5988 If present, this capability can be enabled for a VM, meaning that KVM
5989 will allow the transition to secure guest mode. Otherwise KVM will
5990 veto the transition.
5992 7.20 KVM_CAP_HALT_POLL
5993 ----------------------
5997 :Parameters: args[0] is the maximum poll time in nanoseconds
5998 :Returns: 0 on success; -1 on error
6000 This capability overrides the kvm module parameter halt_poll_ns for the
6003 VCPU polling allows a VCPU to poll for wakeup events instead of immediately
6004 scheduling during guest halts. The maximum time a VCPU can spend polling is
6005 controlled by the kvm module parameter halt_poll_ns. This capability allows
6006 the maximum halt time to specified on a per-VM basis, effectively overriding
6007 the module parameter for the target VM.
6009 7.21 KVM_CAP_X86_USER_SPACE_MSR
6010 -------------------------------
6014 :Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report
6015 :Returns: 0 on success; -1 on error
6017 This capability enables trapping of #GP invoking RDMSR and WRMSR instructions
6020 When a guest requests to read or write an MSR, KVM may not implement all MSRs
6021 that are relevant to a respective system. It also does not differentiate by
6024 To allow more fine grained control over MSR handling, user space may enable
6025 this capability. With it enabled, MSR accesses that match the mask specified in
6026 args[0] and trigger a #GP event inside the guest by KVM will instead trigger
6027 KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications which user space
6028 can then handle to implement model specific MSR handling and/or user notifications
6029 to inform a user that an MSR was not handled.
6031 8. Other capabilities.
6032 ======================
6034 This section lists capabilities that give information about other
6035 features of the KVM implementation.
6037 8.1 KVM_CAP_PPC_HWRNG
6038 ---------------------
6042 This capability, if KVM_CHECK_EXTENSION indicates that it is
6043 available, means that the kernel has an implementation of the
6044 H_RANDOM hypercall backed by a hardware random-number generator.
6045 If present, the kernel H_RANDOM handler can be enabled for guest use
6046 with the KVM_CAP_PPC_ENABLE_HCALL capability.
6048 8.2 KVM_CAP_HYPERV_SYNIC
6049 ------------------------
6053 This capability, if KVM_CHECK_EXTENSION indicates that it is
6054 available, means that the kernel has an implementation of the
6055 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
6056 used to support Windows Hyper-V based guest paravirt drivers(VMBus).
6058 In order to use SynIC, it has to be activated by setting this
6059 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
6060 will disable the use of APIC hardware virtualization even if supported
6061 by the CPU, as it's incompatible with SynIC auto-EOI behavior.
6063 8.3 KVM_CAP_PPC_RADIX_MMU
6064 -------------------------
6068 This capability, if KVM_CHECK_EXTENSION indicates that it is
6069 available, means that the kernel can support guests using the
6070 radix MMU defined in Power ISA V3.00 (as implemented in the POWER9
6073 8.4 KVM_CAP_PPC_HASH_MMU_V3
6074 ---------------------------
6078 This capability, if KVM_CHECK_EXTENSION indicates that it is
6079 available, means that the kernel can support guests using the
6080 hashed page table MMU defined in Power ISA V3.00 (as implemented in
6081 the POWER9 processor), including in-memory segment tables.
6086 :Architectures: mips
6088 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
6089 it is available, means that full hardware assisted virtualization capabilities
6090 of the hardware are available for use through KVM. An appropriate
6091 KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which
6094 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
6095 available, it means that the VM is using full hardware assisted virtualization
6096 capabilities of the hardware. This is useful to check after creating a VM with
6097 KVM_VM_MIPS_DEFAULT.
6099 The value returned by KVM_CHECK_EXTENSION should be compared against known
6100 values (see below). All other values are reserved. This is to allow for the
6101 possibility of other hardware assisted virtualization implementations which
6102 may be incompatible with the MIPS VZ ASE.
6104 == ==========================================================================
6105 0 The trap & emulate implementation is in use to run guest code in user
6106 mode. Guest virtual memory segments are rearranged to fit the guest in the
6107 user mode address space.
6109 1 The MIPS VZ ASE is in use, providing full hardware assisted
6110 virtualization, including standard guest virtual memory segments.
6111 == ==========================================================================
6116 :Architectures: mips
6118 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
6119 it is available, means that the trap & emulate implementation is available to
6120 run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware
6121 assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed
6122 to KVM_CREATE_VM to create a VM which utilises it.
6124 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
6125 available, it means that the VM is using trap & emulate.
6127 8.7 KVM_CAP_MIPS_64BIT
6128 ----------------------
6130 :Architectures: mips
6132 This capability indicates the supported architecture type of the guest, i.e. the
6133 supported register and address width.
6135 The values returned when this capability is checked by KVM_CHECK_EXTENSION on a
6136 kvm VM handle correspond roughly to the CP0_Config.AT register field, and should
6137 be checked specifically against known values (see below). All other values are
6140 == ========================================================================
6141 0 MIPS32 or microMIPS32.
6142 Both registers and addresses are 32-bits wide.
6143 It will only be possible to run 32-bit guest code.
6145 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments.
6146 Registers are 64-bits wide, but addresses are 32-bits wide.
6147 64-bit guest code may run but cannot access MIPS64 memory segments.
6148 It will also be possible to run 32-bit guest code.
6150 2 MIPS64 or microMIPS64 with access to all address segments.
6151 Both registers and addresses are 64-bits wide.
6152 It will be possible to run 64-bit or 32-bit guest code.
6153 == ========================================================================
6155 8.9 KVM_CAP_ARM_USER_IRQ
6156 ------------------------
6158 :Architectures: arm, arm64
6160 This capability, if KVM_CHECK_EXTENSION indicates that it is available, means
6161 that if userspace creates a VM without an in-kernel interrupt controller, it
6162 will be notified of changes to the output level of in-kernel emulated devices,
6163 which can generate virtual interrupts, presented to the VM.
6164 For such VMs, on every return to userspace, the kernel
6165 updates the vcpu's run->s.regs.device_irq_level field to represent the actual
6166 output level of the device.
6168 Whenever kvm detects a change in the device output level, kvm guarantees at
6169 least one return to userspace before running the VM. This exit could either
6170 be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way,
6171 userspace can always sample the device output level and re-compute the state of
6172 the userspace interrupt controller. Userspace should always check the state
6173 of run->s.regs.device_irq_level on every kvm exit.
6174 The value in run->s.regs.device_irq_level can represent both level and edge
6175 triggered interrupt signals, depending on the device. Edge triggered interrupt
6176 signals will exit to userspace with the bit in run->s.regs.device_irq_level
6177 set exactly once per edge signal.
6179 The field run->s.regs.device_irq_level is available independent of
6180 run->kvm_valid_regs or run->kvm_dirty_regs bits.
6182 If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a
6183 number larger than 0 indicating the version of this capability is implemented
6184 and thereby which bits in run->s.regs.device_irq_level can signal values.
6186 Currently the following bits are defined for the device_irq_level bitmap::
6188 KVM_CAP_ARM_USER_IRQ >= 1:
6190 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer
6191 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer
6192 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal
6194 Future versions of kvm may implement additional events. These will get
6195 indicated by returning a higher number from KVM_CHECK_EXTENSION and will be
6198 8.10 KVM_CAP_PPC_SMT_POSSIBLE
6199 -----------------------------
6203 Querying this capability returns a bitmap indicating the possible
6204 virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N
6205 (counting from the right) is set, then a virtual SMT mode of 2^N is
6208 8.11 KVM_CAP_HYPERV_SYNIC2
6209 --------------------------
6213 This capability enables a newer version of Hyper-V Synthetic interrupt
6214 controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM
6215 doesn't clear SynIC message and event flags pages when they are enabled by
6216 writing to the respective MSRs.
6218 8.12 KVM_CAP_HYPERV_VP_INDEX
6219 ----------------------------
6223 This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its
6224 value is used to denote the target vcpu for a SynIC interrupt. For
6225 compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this
6226 capability is absent, userspace can still query this msr's value.
6228 8.13 KVM_CAP_S390_AIS_MIGRATION
6229 -------------------------------
6231 :Architectures: s390
6234 This capability indicates if the flic device will be able to get/set the
6235 AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows
6236 to discover this without having to create a flic device.
6238 8.14 KVM_CAP_S390_PSW
6239 ---------------------
6241 :Architectures: s390
6243 This capability indicates that the PSW is exposed via the kvm_run structure.
6245 8.15 KVM_CAP_S390_GMAP
6246 ----------------------
6248 :Architectures: s390
6250 This capability indicates that the user space memory used as guest mapping can
6251 be anywhere in the user memory address space, as long as the memory slots are
6252 aligned and sized to a segment (1MB) boundary.
6254 8.16 KVM_CAP_S390_COW
6255 ---------------------
6257 :Architectures: s390
6259 This capability indicates that the user space memory used as guest mapping can
6260 use copy-on-write semantics as well as dirty pages tracking via read-only page
6263 8.17 KVM_CAP_S390_BPB
6264 ---------------------
6266 :Architectures: s390
6268 This capability indicates that kvm will implement the interfaces to handle
6269 reset, migration and nested KVM for branch prediction blocking. The stfle
6270 facility 82 should not be provided to the guest without this capability.
6272 8.18 KVM_CAP_HYPERV_TLBFLUSH
6273 ----------------------------
6277 This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush
6279 HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx,
6280 HvFlushVirtualAddressList, HvFlushVirtualAddressListEx.
6282 8.19 KVM_CAP_ARM_INJECT_SERROR_ESR
6283 ----------------------------------
6285 :Architectures: arm, arm64
6287 This capability indicates that userspace can specify (via the
6288 KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it
6289 takes a virtual SError interrupt exception.
6290 If KVM advertises this capability, userspace can only specify the ISS field for
6291 the ESR syndrome. Other parts of the ESR, such as the EC are generated by the
6292 CPU when the exception is taken. If this virtual SError is taken to EL1 using
6293 AArch64, this value will be reported in the ISS field of ESR_ELx.
6295 See KVM_CAP_VCPU_EVENTS for more details.
6297 8.20 KVM_CAP_HYPERV_SEND_IPI
6298 ----------------------------
6302 This capability indicates that KVM supports paravirtualized Hyper-V IPI send
6304 HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx.
6306 8.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH
6307 -----------------------------------
6311 This capability indicates that KVM running on top of Hyper-V hypervisor
6312 enables Direct TLB flush for its guests meaning that TLB flush
6313 hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM.
6314 Due to the different ABI for hypercall parameters between Hyper-V and
6315 KVM, enabling this capability effectively disables all hypercall
6316 handling by KVM (as some KVM hypercall may be mistakenly treated as TLB
6317 flush hypercalls by Hyper-V) so userspace should disable KVM identification
6318 in CPUID and only exposes Hyper-V identification. In this case, guest
6319 thinks it's running on Hyper-V and only use Hyper-V hypercalls.
6321 8.22 KVM_CAP_S390_VCPU_RESETS
6322 -----------------------------
6324 :Architectures: s390
6326 This capability indicates that the KVM_S390_NORMAL_RESET and
6327 KVM_S390_CLEAR_RESET ioctls are available.
6329 8.23 KVM_CAP_S390_PROTECTED
6330 ---------------------------
6332 :Architectures: s390
6334 This capability indicates that the Ultravisor has been initialized and
6335 KVM can therefore start protected VMs.
6336 This capability governs the KVM_S390_PV_COMMAND ioctl and the
6337 KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected
6338 guests when the state change is invalid.
6340 8.24 KVM_CAP_STEAL_TIME
6341 -----------------------
6343 :Architectures: arm64, x86
6345 This capability indicates that KVM supports steal time accounting.
6346 When steal time accounting is supported it may be enabled with
6347 architecture-specific interfaces. This capability and the architecture-
6348 specific interfaces must be consistent, i.e. if one says the feature
6349 is supported, than the other should as well and vice versa. For arm64
6350 see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL".
6351 For x86 see Documentation/virt/kvm/msr.rst "MSR_KVM_STEAL_TIME".
6353 8.25 KVM_CAP_S390_DIAG318
6354 -------------------------
6356 :Architectures: s390
6358 This capability enables a guest to set information about its control program
6359 (i.e. guest kernel type and version). The information is helpful during
6360 system/firmware service events, providing additional data about the guest
6361 environments running on the machine.
6363 The information is associated with the DIAGNOSE 0x318 instruction, which sets
6364 an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and
6365 a 7-byte Control Program Version Code (CPVC). The CPNC determines what
6366 environment the control program is running in (e.g. Linux, z/VM...), and the
6367 CPVC is used for information specific to OS (e.g. Linux version, Linux
6370 If this capability is available, then the CPNC and CPVC can be synchronized
6371 between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318).
6373 8.26 KVM_CAP_X86_USER_SPACE_MSR
6374 -------------------------------
6378 This capability indicates that KVM supports deflection of MSR reads and
6379 writes to user space. It can be enabled on a VM level. If enabled, MSR
6380 accesses that would usually trigger a #GP by KVM into the guest will
6381 instead get bounced to user space through the KVM_EXIT_X86_RDMSR and
6382 KVM_EXIT_X86_WRMSR exit notifications.
6384 8.27 KVM_X86_SET_MSR_FILTER
6385 ---------------------------
6389 This capability indicates that KVM supports that accesses to user defined MSRs
6390 may be rejected. With this capability exposed, KVM exports new VM ioctl
6391 KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR
6392 ranges that KVM should reject access to.
6394 In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to
6395 trap and emulate MSRs that are outside of the scope of KVM as well as
6396 limit the attack surface on KVM's MSR emulation code.
6398 8.28 KVM_CAP_ENFORCE_PV_CPUID
6399 -----------------------------
6403 When enabled, KVM will disable paravirtual features provided to the
6404 guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf
6405 (0x40000001). Otherwise, a guest may use the paravirtual features
6406 regardless of what has actually been exposed through the CPUID leaf.
6408 9. Known KVM API problems
6409 =========================
6411 In some cases, KVM's API has some inconsistencies or common pitfalls
6412 that userspace need to be aware of. This section details some of
6415 Most of them are architecture specific, so the section is split by
6421 ``KVM_GET_SUPPORTED_CPUID`` issues
6422 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6424 In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible
6425 to take its result and pass it directly to ``KVM_SET_CPUID2``. This section
6426 documents some cases in which that requires some care.
6431 CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``,
6432 but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or
6433 ``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of
6436 The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature.
6438 CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by ``KVM_GET_SUPPORTED_CPUID``.
6439 It can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` is present and the kernel
6440 has enabled in-kernel emulation of the local APIC.
6445 Several CPUID values include topology information for the host CPU:
6446 0x0b and 0x1f for Intel systems, 0x8000001e for AMD systems. Different
6447 versions of KVM return different values for this information and userspace
6448 should not rely on it. Currently they return all zeroes.
6450 If userspace wishes to set up a guest topology, it should be careful that
6451 the values of these three leaves differ for each CPU. In particular,
6452 the APIC ID is found in EDX for all subleaves of 0x0b and 0x1f, and in EAX
6453 for 0x8000001e; the latter also encodes the core id and node id in bits
6454 7:0 of EBX and ECX respectively.
6456 Obsolete ioctls and capabilities
6457 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6459 KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually
6460 available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if
6463 Ordering of KVM_GET_*/KVM_SET_* ioctls
6464 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^