1 The Definitive KVM (Kernel-based Virtual Machine) API Documentation
2 ===================================================================
7 The kvm API is a set of ioctls that are issued to control various aspects
8 of a virtual machine. The ioctls belong to three classes
10 - System ioctls: These query and set global attributes which affect the
11 whole kvm subsystem. In addition a system ioctl is used to create
14 - VM ioctls: These query and set attributes that affect an entire virtual
15 machine, for example memory layout. In addition a VM ioctl is used to
16 create virtual cpus (vcpus).
18 Only run VM ioctls from the same process (address space) that was used
21 - vcpu ioctls: These query and set attributes that control the operation
22 of a single virtual cpu.
24 Only run vcpu ioctls from the same thread that was used to create the
31 The kvm API is centered around file descriptors. An initial
32 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle
33 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this
34 handle will create a VM file descriptor which can be used to issue VM
35 ioctls. A KVM_CREATE_VCPU ioctl on a VM fd will create a virtual cpu
36 and return a file descriptor pointing to it. Finally, ioctls on a vcpu
37 fd can be used to control the vcpu, including the important task of
38 actually running guest code.
40 In general file descriptors can be migrated among processes by means
41 of fork() and the SCM_RIGHTS facility of unix domain socket. These
42 kinds of tricks are explicitly not supported by kvm. While they will
43 not cause harm to the host, their actual behavior is not guaranteed by
44 the API. The only supported use is one virtual machine per process,
45 and one vcpu per thread.
51 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward
52 incompatible change are allowed. However, there is an extension
53 facility that allows backward-compatible extensions to the API to be
56 The extension mechanism is not based on the Linux version number.
57 Instead, kvm defines extension identifiers and a facility to query
58 whether a particular extension identifier is available. If it is, a
59 set of ioctls is available for application use.
65 This section describes ioctls that can be used to control kvm guests.
66 For each ioctl, the following information is provided along with a
69 Capability: which KVM extension provides this ioctl. Can be 'basic',
70 which means that is will be provided by any kernel that supports
71 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which
72 means availability needs to be checked with KVM_CHECK_EXTENSION
73 (see section 4.4), or 'none' which means that while not all kernels
74 support this ioctl, there's no capability bit to check its
75 availability: for kernels that don't support the ioctl,
76 the ioctl returns -ENOTTY.
78 Architectures: which instruction set architectures provide this ioctl.
79 x86 includes both i386 and x86_64.
81 Type: system, vm, or vcpu.
83 Parameters: what parameters are accepted by the ioctl.
85 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
86 are not detailed, but errors with specific meanings are.
89 4.1 KVM_GET_API_VERSION
95 Returns: the constant KVM_API_VERSION (=12)
97 This identifies the API version as the stable kvm API. It is not
98 expected that this number will change. However, Linux 2.6.20 and
99 2.6.21 report earlier versions; these are not documented and not
100 supported. Applications should refuse to run if KVM_GET_API_VERSION
101 returns a value other than 12. If this check passes, all ioctls
102 described as 'basic' will be available.
110 Parameters: machine type identifier (KVM_VM_*)
111 Returns: a VM fd that can be used to control the new virtual machine.
113 The new VM has no virtual cpus and no memory. An mmap() of a VM fd
114 will access the virtual machine's physical address space; offset zero
115 corresponds to guest physical address zero. Use of mmap() on a VM fd
116 is discouraged if userspace memory allocation (KVM_CAP_USER_MEMORY) is
118 You most certainly want to use 0 as machine type.
120 In order to create user controlled virtual machines on S390, check
121 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
122 privileged user (CAP_SYS_ADMIN).
125 4.3 KVM_GET_MSR_INDEX_LIST
130 Parameters: struct kvm_msr_list (in/out)
131 Returns: 0 on success; -1 on error
133 E2BIG: the msr index list is to be to fit in the array specified by
136 struct kvm_msr_list {
137 __u32 nmsrs; /* number of msrs in entries */
141 This ioctl returns the guest msrs that are supported. The list varies
142 by kvm version and host processor, but does not change otherwise. The
143 user fills in the size of the indices array in nmsrs, and in return
144 kvm adjusts nmsrs to reflect the actual number of msrs and fills in
145 the indices array with their numbers.
147 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
148 not returned in the MSR list, as different vcpus can have a different number
149 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
152 4.4 KVM_CHECK_EXTENSION
154 Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
156 Type: system ioctl, vm ioctl
157 Parameters: extension identifier (KVM_CAP_*)
158 Returns: 0 if unsupported; 1 (or some other positive integer) if supported
160 The API allows the application to query about extensions to the core
161 kvm API. Userspace passes an extension identifier (an integer) and
162 receives an integer that describes the extension availability.
163 Generally 0 means no and 1 means yes, but some extensions may report
164 additional information in the integer return value.
166 Based on their initialization different VMs may have different capabilities.
167 It is thus encouraged to use the vm ioctl to query for capabilities (available
168 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
170 4.5 KVM_GET_VCPU_MMAP_SIZE
176 Returns: size of vcpu mmap area, in bytes
178 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
179 memory region. This ioctl returns the size of that region. See the
180 KVM_RUN documentation for details.
183 4.6 KVM_SET_MEMORY_REGION
188 Parameters: struct kvm_memory_region (in)
189 Returns: 0 on success, -1 on error
191 This ioctl is obsolete and has been removed.
199 Parameters: vcpu id (apic id on x86)
200 Returns: vcpu fd on success, -1 on error
202 This API adds a vcpu to a virtual machine. The vcpu id is a small integer
203 in the range [0, max_vcpus).
205 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
206 the KVM_CHECK_EXTENSION ioctl() at run-time.
207 The maximum possible value for max_vcpus can be retrieved using the
208 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
210 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
212 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
213 same as the value returned from KVM_CAP_NR_VCPUS.
215 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
216 threads in one or more virtual CPU cores. (This is because the
217 hardware requires all the hardware threads in a CPU core to be in the
218 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
219 of vcpus per virtual core (vcore). The vcore id is obtained by
220 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
221 given vcore will always be in the same physical core as each other
222 (though that might be a different physical core from time to time).
223 Userspace can control the threading (SMT) mode of the guest by its
224 allocation of vcpu ids. For example, if userspace wants
225 single-threaded guest vcpus, it should make all vcpu ids be a multiple
226 of the number of vcpus per vcore.
228 For virtual cpus that have been created with S390 user controlled virtual
229 machines, the resulting vcpu fd can be memory mapped at page offset
230 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
231 cpu's hardware control block.
234 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
239 Parameters: struct kvm_dirty_log (in/out)
240 Returns: 0 on success, -1 on error
242 /* for KVM_GET_DIRTY_LOG */
243 struct kvm_dirty_log {
247 void __user *dirty_bitmap; /* one bit per page */
252 Given a memory slot, return a bitmap containing any pages dirtied
253 since the last call to this ioctl. Bit 0 is the first page in the
254 memory slot. Ensure the entire structure is cleared to avoid padding
257 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies
258 the address space for which you want to return the dirty bitmap.
259 They must be less than the value that KVM_CHECK_EXTENSION returns for
260 the KVM_CAP_MULTI_ADDRESS_SPACE capability.
263 4.9 KVM_SET_MEMORY_ALIAS
268 Parameters: struct kvm_memory_alias (in)
269 Returns: 0 (success), -1 (error)
271 This ioctl is obsolete and has been removed.
280 Returns: 0 on success, -1 on error
282 EINTR: an unmasked signal is pending
284 This ioctl is used to run a guest virtual cpu. While there are no
285 explicit parameters, there is an implicit parameter block that can be
286 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
287 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
288 kvm_run' (see below).
294 Architectures: all except ARM, arm64
296 Parameters: struct kvm_regs (out)
297 Returns: 0 on success, -1 on error
299 Reads the general purpose registers from the vcpu.
303 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
304 __u64 rax, rbx, rcx, rdx;
305 __u64 rsi, rdi, rsp, rbp;
306 __u64 r8, r9, r10, r11;
307 __u64 r12, r13, r14, r15;
313 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
324 Architectures: all except ARM, arm64
326 Parameters: struct kvm_regs (in)
327 Returns: 0 on success, -1 on error
329 Writes the general purpose registers into the vcpu.
331 See KVM_GET_REGS for the data structure.
337 Architectures: x86, ppc
339 Parameters: struct kvm_sregs (out)
340 Returns: 0 on success, -1 on error
342 Reads special registers from the vcpu.
346 struct kvm_segment cs, ds, es, fs, gs, ss;
347 struct kvm_segment tr, ldt;
348 struct kvm_dtable gdt, idt;
349 __u64 cr0, cr2, cr3, cr4, cr8;
352 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
355 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
357 interrupt_bitmap is a bitmap of pending external interrupts. At most
358 one bit may be set. This interrupt has been acknowledged by the APIC
359 but not yet injected into the cpu core.
365 Architectures: x86, ppc
367 Parameters: struct kvm_sregs (in)
368 Returns: 0 on success, -1 on error
370 Writes special registers into the vcpu. See KVM_GET_SREGS for the
379 Parameters: struct kvm_translation (in/out)
380 Returns: 0 on success, -1 on error
382 Translates a virtual address according to the vcpu's current address
385 struct kvm_translation {
387 __u64 linear_address;
390 __u64 physical_address;
401 Architectures: x86, ppc, mips
403 Parameters: struct kvm_interrupt (in)
404 Returns: 0 on success, negative on failure.
406 Queues a hardware interrupt vector to be injected.
408 /* for KVM_INTERRUPT */
409 struct kvm_interrupt {
416 Returns: 0 on success,
417 -EEXIST if an interrupt is already enqueued
418 -EINVAL the the irq number is invalid
419 -ENXIO if the PIC is in the kernel
420 -EFAULT if the pointer is invalid
422 Note 'irq' is an interrupt vector, not an interrupt pin or line. This
423 ioctl is useful if the in-kernel PIC is not used.
427 Queues an external interrupt to be injected. This ioctl is overleaded
428 with 3 different irq values:
432 This injects an edge type external interrupt into the guest once it's ready
433 to receive interrupts. When injected, the interrupt is done.
435 b) KVM_INTERRUPT_UNSET
437 This unsets any pending interrupt.
439 Only available with KVM_CAP_PPC_UNSET_IRQ.
441 c) KVM_INTERRUPT_SET_LEVEL
443 This injects a level type external interrupt into the guest context. The
444 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
447 Only available with KVM_CAP_PPC_IRQ_LEVEL.
449 Note that any value for 'irq' other than the ones stated above is invalid
450 and incurs unexpected behavior.
454 Queues an external interrupt to be injected into the virtual CPU. A negative
455 interrupt number dequeues the interrupt.
466 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
474 Parameters: struct kvm_msrs (in/out)
475 Returns: 0 on success, -1 on error
477 Reads model-specific registers from the vcpu. Supported msr indices can
478 be obtained using KVM_GET_MSR_INDEX_LIST.
481 __u32 nmsrs; /* number of msrs in entries */
484 struct kvm_msr_entry entries[0];
487 struct kvm_msr_entry {
493 Application code should set the 'nmsrs' member (which indicates the
494 size of the entries array) and the 'index' member of each array entry.
495 kvm will fill in the 'data' member.
503 Parameters: struct kvm_msrs (in)
504 Returns: 0 on success, -1 on error
506 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
509 Application code should set the 'nmsrs' member (which indicates the
510 size of the entries array), and the 'index' and 'data' members of each
519 Parameters: struct kvm_cpuid (in)
520 Returns: 0 on success, -1 on error
522 Defines the vcpu responses to the cpuid instruction. Applications
523 should use the KVM_SET_CPUID2 ioctl if available.
526 struct kvm_cpuid_entry {
535 /* for KVM_SET_CPUID */
539 struct kvm_cpuid_entry entries[0];
543 4.21 KVM_SET_SIGNAL_MASK
548 Parameters: struct kvm_signal_mask (in)
549 Returns: 0 on success, -1 on error
551 Defines which signals are blocked during execution of KVM_RUN. This
552 signal mask temporarily overrides the threads signal mask. Any
553 unblocked signal received (except SIGKILL and SIGSTOP, which retain
554 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
556 Note the signal will only be delivered if not blocked by the original
559 /* for KVM_SET_SIGNAL_MASK */
560 struct kvm_signal_mask {
571 Parameters: struct kvm_fpu (out)
572 Returns: 0 on success, -1 on error
574 Reads the floating point state from the vcpu.
576 /* for KVM_GET_FPU and KVM_SET_FPU */
581 __u8 ftwx; /* in fxsave format */
597 Parameters: struct kvm_fpu (in)
598 Returns: 0 on success, -1 on error
600 Writes the floating point state to the vcpu.
602 /* for KVM_GET_FPU and KVM_SET_FPU */
607 __u8 ftwx; /* in fxsave format */
618 4.24 KVM_CREATE_IRQCHIP
620 Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
621 Architectures: x86, ARM, arm64, s390
624 Returns: 0 on success, -1 on error
626 Creates an interrupt controller model in the kernel.
627 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
628 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
629 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
630 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of
631 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
632 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
633 On s390, a dummy irq routing table is created.
635 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
636 before KVM_CREATE_IRQCHIP can be used.
641 Capability: KVM_CAP_IRQCHIP
642 Architectures: x86, arm, arm64
644 Parameters: struct kvm_irq_level
645 Returns: 0 on success, -1 on error
647 Sets the level of a GSI input to the interrupt controller model in the kernel.
648 On some architectures it is required that an interrupt controller model has
649 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
650 interrupts require the level to be set to 1 and then back to 0.
652 On real hardware, interrupt pins can be active-low or active-high. This
653 does not matter for the level field of struct kvm_irq_level: 1 always
654 means active (asserted), 0 means inactive (deasserted).
656 x86 allows the operating system to program the interrupt polarity
657 (active-low/active-high) for level-triggered interrupts, and KVM used
658 to consider the polarity. However, due to bitrot in the handling of
659 active-low interrupts, the above convention is now valid on x86 too.
660 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
661 should not present interrupts to the guest as active-low unless this
662 capability is present (or unless it is not using the in-kernel irqchip,
666 ARM/arm64 can signal an interrupt either at the CPU level, or at the
667 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
668 use PPIs designated for specific cpus. The irq field is interpreted
671 Â bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 |
672 field: | irq_type | vcpu_index | irq_id |
674 The irq_type field has the following values:
675 - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
676 - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
677 (the vcpu_index field is ignored)
678 - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
680 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
682 In both cases, level is used to assert/deassert the line.
684 struct kvm_irq_level {
687 __s32 status; /* not used for KVM_IRQ_LEVEL */
689 __u32 level; /* 0 or 1 */
695 Capability: KVM_CAP_IRQCHIP
698 Parameters: struct kvm_irqchip (in/out)
699 Returns: 0 on success, -1 on error
701 Reads the state of a kernel interrupt controller created with
702 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
705 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
708 char dummy[512]; /* reserving space */
709 struct kvm_pic_state pic;
710 struct kvm_ioapic_state ioapic;
717 Capability: KVM_CAP_IRQCHIP
720 Parameters: struct kvm_irqchip (in)
721 Returns: 0 on success, -1 on error
723 Sets the state of a kernel interrupt controller created with
724 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
727 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
730 char dummy[512]; /* reserving space */
731 struct kvm_pic_state pic;
732 struct kvm_ioapic_state ioapic;
737 4.28 KVM_XEN_HVM_CONFIG
739 Capability: KVM_CAP_XEN_HVM
742 Parameters: struct kvm_xen_hvm_config (in)
743 Returns: 0 on success, -1 on error
745 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
746 page, and provides the starting address and size of the hypercall
747 blobs in userspace. When the guest writes the MSR, kvm copies one
748 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
751 struct kvm_xen_hvm_config {
764 Capability: KVM_CAP_ADJUST_CLOCK
767 Parameters: struct kvm_clock_data (out)
768 Returns: 0 on success, -1 on error
770 Gets the current timestamp of kvmclock as seen by the current guest. In
771 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
774 struct kvm_clock_data {
775 __u64 clock; /* kvmclock current value */
783 Capability: KVM_CAP_ADJUST_CLOCK
786 Parameters: struct kvm_clock_data (in)
787 Returns: 0 on success, -1 on error
789 Sets the current timestamp of kvmclock to the value specified in its parameter.
790 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
793 struct kvm_clock_data {
794 __u64 clock; /* kvmclock current value */
800 4.31 KVM_GET_VCPU_EVENTS
802 Capability: KVM_CAP_VCPU_EVENTS
803 Extended by: KVM_CAP_INTR_SHADOW
806 Parameters: struct kvm_vcpu_event (out)
807 Returns: 0 on success, -1 on error
809 Gets currently pending exceptions, interrupts, and NMIs as well as related
812 struct kvm_vcpu_events {
842 Only two fields are defined in the flags field:
844 - KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that
845 interrupt.shadow contains a valid state.
847 - KVM_VCPUEVENT_VALID_SMM may be set in the flags field to signal that
848 smi contains a valid state.
850 4.32 KVM_SET_VCPU_EVENTS
852 Capability: KVM_CAP_VCPU_EVENTS
853 Extended by: KVM_CAP_INTR_SHADOW
856 Parameters: struct kvm_vcpu_event (in)
857 Returns: 0 on success, -1 on error
859 Set pending exceptions, interrupts, and NMIs as well as related states of the
862 See KVM_GET_VCPU_EVENTS for the data structure.
864 Fields that may be modified asynchronously by running VCPUs can be excluded
865 from the update. These fields are nmi.pending, sipi_vector, smi.smm,
866 smi.pending. Keep the corresponding bits in the flags field cleared to
867 suppress overwriting the current in-kernel state. The bits are:
869 KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel
870 KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector
871 KVM_VCPUEVENT_VALID_SMM - transfer the smi sub-struct.
873 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
874 the flags field to signal that interrupt.shadow contains a valid state and
875 shall be written into the VCPU.
877 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available.
880 4.33 KVM_GET_DEBUGREGS
882 Capability: KVM_CAP_DEBUGREGS
885 Parameters: struct kvm_debugregs (out)
886 Returns: 0 on success, -1 on error
888 Reads debug registers from the vcpu.
890 struct kvm_debugregs {
899 4.34 KVM_SET_DEBUGREGS
901 Capability: KVM_CAP_DEBUGREGS
904 Parameters: struct kvm_debugregs (in)
905 Returns: 0 on success, -1 on error
907 Writes debug registers into the vcpu.
909 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
910 yet and must be cleared on entry.
913 4.35 KVM_SET_USER_MEMORY_REGION
915 Capability: KVM_CAP_USER_MEM
918 Parameters: struct kvm_userspace_memory_region (in)
919 Returns: 0 on success, -1 on error
921 struct kvm_userspace_memory_region {
924 __u64 guest_phys_addr;
925 __u64 memory_size; /* bytes */
926 __u64 userspace_addr; /* start of the userspace allocated memory */
929 /* for kvm_memory_region::flags */
930 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
931 #define KVM_MEM_READONLY (1UL << 1)
933 This ioctl allows the user to create or modify a guest physical memory
934 slot. When changing an existing slot, it may be moved in the guest
935 physical memory space, or its flags may be modified. It may not be
936 resized. Slots may not overlap in guest physical address space.
938 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot"
939 specifies the address space which is being modified. They must be
940 less than the value that KVM_CHECK_EXTENSION returns for the
941 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces
942 are unrelated; the restriction on overlapping slots only applies within
945 Memory for the region is taken starting at the address denoted by the
946 field userspace_addr, which must point at user addressable memory for
947 the entire memory slot size. Any object may back this memory, including
948 anonymous memory, ordinary files, and hugetlbfs.
950 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
951 be identical. This allows large pages in the guest to be backed by large
954 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
955 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
956 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
957 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
958 to make a new slot read-only. In this case, writes to this memory will be
959 posted to userspace as KVM_EXIT_MMIO exits.
961 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
962 the memory region are automatically reflected into the guest. For example, an
963 mmap() that affects the region will be made visible immediately. Another
964 example is madvise(MADV_DROP).
966 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl.
967 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory
968 allocation and is deprecated.
971 4.36 KVM_SET_TSS_ADDR
973 Capability: KVM_CAP_SET_TSS_ADDR
976 Parameters: unsigned long tss_address (in)
977 Returns: 0 on success, -1 on error
979 This ioctl defines the physical address of a three-page region in the guest
980 physical address space. The region must be within the first 4GB of the
981 guest physical address space and must not conflict with any memory slot
982 or any mmio address. The guest may malfunction if it accesses this memory
985 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
986 because of a quirk in the virtualization implementation (see the internals
987 documentation when it pops into existence).
992 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM
993 Architectures: x86 (only KVM_CAP_ENABLE_CAP_VM),
994 mips (only KVM_CAP_ENABLE_CAP), ppc, s390
995 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM)
996 Parameters: struct kvm_enable_cap (in)
997 Returns: 0 on success; -1 on error
999 +Not all extensions are enabled by default. Using this ioctl the application
1000 can enable an extension, making it available to the guest.
1002 On systems that do not support this ioctl, it always fails. On systems that
1003 do support it, it only works for extensions that are supported for enablement.
1005 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
1008 struct kvm_enable_cap {
1012 The capability that is supposed to get enabled.
1016 A bitfield indicating future enhancements. Has to be 0 for now.
1020 Arguments for enabling a feature. If a feature needs initial values to
1021 function properly, this is the place to put them.
1026 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
1027 for vm-wide capabilities.
1029 4.38 KVM_GET_MP_STATE
1031 Capability: KVM_CAP_MP_STATE
1032 Architectures: x86, s390, arm, arm64
1034 Parameters: struct kvm_mp_state (out)
1035 Returns: 0 on success; -1 on error
1037 struct kvm_mp_state {
1041 Returns the vcpu's current "multiprocessing state" (though also valid on
1042 uniprocessor guests).
1044 Possible values are:
1046 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86,arm/arm64]
1047 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP)
1048 which has not yet received an INIT signal [x86]
1049 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is
1050 now ready for a SIPI [x86]
1051 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and
1052 is waiting for an interrupt [x86]
1053 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector
1054 accessible via KVM_GET_VCPU_EVENTS) [x86]
1055 - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390,arm/arm64]
1056 - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390]
1057 - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted)
1059 - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state
1062 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1063 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1064 these architectures.
1068 The only states that are valid are KVM_MP_STATE_STOPPED and
1069 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
1071 4.39 KVM_SET_MP_STATE
1073 Capability: KVM_CAP_MP_STATE
1074 Architectures: x86, s390, arm, arm64
1076 Parameters: struct kvm_mp_state (in)
1077 Returns: 0 on success; -1 on error
1079 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1082 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1083 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1084 these architectures.
1088 The only states that are valid are KVM_MP_STATE_STOPPED and
1089 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
1091 4.40 KVM_SET_IDENTITY_MAP_ADDR
1093 Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1096 Parameters: unsigned long identity (in)
1097 Returns: 0 on success, -1 on error
1099 This ioctl defines the physical address of a one-page region in the guest
1100 physical address space. The region must be within the first 4GB of the
1101 guest physical address space and must not conflict with any memory slot
1102 or any mmio address. The guest may malfunction if it accesses this memory
1105 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1106 because of a quirk in the virtualization implementation (see the internals
1107 documentation when it pops into existence).
1110 4.41 KVM_SET_BOOT_CPU_ID
1112 Capability: KVM_CAP_SET_BOOT_CPU_ID
1115 Parameters: unsigned long vcpu_id
1116 Returns: 0 on success, -1 on error
1118 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1119 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1125 Capability: KVM_CAP_XSAVE
1128 Parameters: struct kvm_xsave (out)
1129 Returns: 0 on success, -1 on error
1135 This ioctl would copy current vcpu's xsave struct to the userspace.
1140 Capability: KVM_CAP_XSAVE
1143 Parameters: struct kvm_xsave (in)
1144 Returns: 0 on success, -1 on error
1150 This ioctl would copy userspace's xsave struct to the kernel.
1155 Capability: KVM_CAP_XCRS
1158 Parameters: struct kvm_xcrs (out)
1159 Returns: 0 on success, -1 on error
1170 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1174 This ioctl would copy current vcpu's xcrs to the userspace.
1179 Capability: KVM_CAP_XCRS
1182 Parameters: struct kvm_xcrs (in)
1183 Returns: 0 on success, -1 on error
1194 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1198 This ioctl would set vcpu's xcr to the value userspace specified.
1201 4.46 KVM_GET_SUPPORTED_CPUID
1203 Capability: KVM_CAP_EXT_CPUID
1206 Parameters: struct kvm_cpuid2 (in/out)
1207 Returns: 0 on success, -1 on error
1212 struct kvm_cpuid_entry2 entries[0];
1215 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1216 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
1217 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
1219 struct kvm_cpuid_entry2 {
1230 This ioctl returns x86 cpuid features which are supported by both the hardware
1231 and kvm. Userspace can use the information returned by this ioctl to
1232 construct cpuid information (for KVM_SET_CPUID2) that is consistent with
1233 hardware, kernel, and userspace capabilities, and with user requirements (for
1234 example, the user may wish to constrain cpuid to emulate older hardware,
1235 or for feature consistency across a cluster).
1237 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1238 with the 'nent' field indicating the number of entries in the variable-size
1239 array 'entries'. If the number of entries is too low to describe the cpu
1240 capabilities, an error (E2BIG) is returned. If the number is too high,
1241 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1242 number is just right, the 'nent' field is adjusted to the number of valid
1243 entries in the 'entries' array, which is then filled.
1245 The entries returned are the host cpuid as returned by the cpuid instruction,
1246 with unknown or unsupported features masked out. Some features (for example,
1247 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1248 emulate them efficiently. The fields in each entry are defined as follows:
1250 function: the eax value used to obtain the entry
1251 index: the ecx value used to obtain the entry (for entries that are
1253 flags: an OR of zero or more of the following:
1254 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1255 if the index field is valid
1256 KVM_CPUID_FLAG_STATEFUL_FUNC:
1257 if cpuid for this function returns different values for successive
1258 invocations; there will be several entries with the same function,
1259 all with this flag set
1260 KVM_CPUID_FLAG_STATE_READ_NEXT:
1261 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
1262 the first entry to be read by a cpu
1263 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
1264 this function/index combination
1266 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1267 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1268 support. Instead it is reported via
1270 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1272 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1273 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1276 4.47 KVM_PPC_GET_PVINFO
1278 Capability: KVM_CAP_PPC_GET_PVINFO
1281 Parameters: struct kvm_ppc_pvinfo (out)
1282 Returns: 0 on success, !0 on error
1284 struct kvm_ppc_pvinfo {
1290 This ioctl fetches PV specific information that need to be passed to the guest
1291 using the device tree or other means from vm context.
1293 The hcall array defines 4 instructions that make up a hypercall.
1295 If any additional field gets added to this structure later on, a bit for that
1296 additional piece of information will be set in the flags bitmap.
1298 The flags bitmap is defined as:
1300 /* the host supports the ePAPR idle hcall
1301 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1303 4.48 KVM_ASSIGN_PCI_DEVICE (deprecated)
1308 Parameters: struct kvm_assigned_pci_dev (in)
1309 Returns: 0 on success, -1 on error
1311 Assigns a host PCI device to the VM.
1313 struct kvm_assigned_pci_dev {
1314 __u32 assigned_dev_id;
1324 The PCI device is specified by the triple segnr, busnr, and devfn.
1325 Identification in succeeding service requests is done via assigned_dev_id. The
1326 following flags are specified:
1328 /* Depends on KVM_CAP_IOMMU */
1329 #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0)
1330 /* The following two depend on KVM_CAP_PCI_2_3 */
1331 #define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1)
1332 #define KVM_DEV_ASSIGN_MASK_INTX (1 << 2)
1334 If KVM_DEV_ASSIGN_PCI_2_3 is set, the kernel will manage legacy INTx interrupts
1335 via the PCI-2.3-compliant device-level mask, thus enable IRQ sharing with other
1336 assigned devices or host devices. KVM_DEV_ASSIGN_MASK_INTX specifies the
1337 guest's view on the INTx mask, see KVM_ASSIGN_SET_INTX_MASK for details.
1339 The KVM_DEV_ASSIGN_ENABLE_IOMMU flag is a mandatory option to ensure
1340 isolation of the device. Usages not specifying this flag are deprecated.
1342 Only PCI header type 0 devices with PCI BAR resources are supported by
1343 device assignment. The user requesting this ioctl must have read/write
1344 access to the PCI sysfs resource files associated with the device.
1347 ENOTTY: kernel does not support this ioctl
1349 Other error conditions may be defined by individual device types or
1350 have their standard meanings.
1353 4.49 KVM_DEASSIGN_PCI_DEVICE (deprecated)
1358 Parameters: struct kvm_assigned_pci_dev (in)
1359 Returns: 0 on success, -1 on error
1361 Ends PCI device assignment, releasing all associated resources.
1363 See KVM_ASSIGN_PCI_DEVICE for the data structure. Only assigned_dev_id is
1364 used in kvm_assigned_pci_dev to identify the device.
1367 ENOTTY: kernel does not support this ioctl
1369 Other error conditions may be defined by individual device types or
1370 have their standard meanings.
1372 4.50 KVM_ASSIGN_DEV_IRQ (deprecated)
1374 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1377 Parameters: struct kvm_assigned_irq (in)
1378 Returns: 0 on success, -1 on error
1380 Assigns an IRQ to a passed-through device.
1382 struct kvm_assigned_irq {
1383 __u32 assigned_dev_id;
1384 __u32 host_irq; /* ignored (legacy field) */
1392 The following flags are defined:
1394 #define KVM_DEV_IRQ_HOST_INTX (1 << 0)
1395 #define KVM_DEV_IRQ_HOST_MSI (1 << 1)
1396 #define KVM_DEV_IRQ_HOST_MSIX (1 << 2)
1398 #define KVM_DEV_IRQ_GUEST_INTX (1 << 8)
1399 #define KVM_DEV_IRQ_GUEST_MSI (1 << 9)
1400 #define KVM_DEV_IRQ_GUEST_MSIX (1 << 10)
1402 It is not valid to specify multiple types per host or guest IRQ. However, the
1403 IRQ type of host and guest can differ or can even be null.
1406 ENOTTY: kernel does not support this ioctl
1408 Other error conditions may be defined by individual device types or
1409 have their standard meanings.
1412 4.51 KVM_DEASSIGN_DEV_IRQ (deprecated)
1414 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1417 Parameters: struct kvm_assigned_irq (in)
1418 Returns: 0 on success, -1 on error
1420 Ends an IRQ assignment to a passed-through device.
1422 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1423 by assigned_dev_id, flags must correspond to the IRQ type specified on
1424 KVM_ASSIGN_DEV_IRQ. Partial deassignment of host or guest IRQ is allowed.
1427 4.52 KVM_SET_GSI_ROUTING
1429 Capability: KVM_CAP_IRQ_ROUTING
1430 Architectures: x86 s390
1432 Parameters: struct kvm_irq_routing (in)
1433 Returns: 0 on success, -1 on error
1435 Sets the GSI routing table entries, overwriting any previously set entries.
1437 struct kvm_irq_routing {
1440 struct kvm_irq_routing_entry entries[0];
1443 No flags are specified so far, the corresponding field must be set to zero.
1445 struct kvm_irq_routing_entry {
1451 struct kvm_irq_routing_irqchip irqchip;
1452 struct kvm_irq_routing_msi msi;
1453 struct kvm_irq_routing_s390_adapter adapter;
1454 struct kvm_irq_routing_hv_sint hv_sint;
1459 /* gsi routing entry types */
1460 #define KVM_IRQ_ROUTING_IRQCHIP 1
1461 #define KVM_IRQ_ROUTING_MSI 2
1462 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1463 #define KVM_IRQ_ROUTING_HV_SINT 4
1465 No flags are specified so far, the corresponding field must be set to zero.
1467 struct kvm_irq_routing_irqchip {
1472 struct kvm_irq_routing_msi {
1479 struct kvm_irq_routing_s390_adapter {
1483 __u32 summary_offset;
1487 struct kvm_irq_routing_hv_sint {
1492 4.53 KVM_ASSIGN_SET_MSIX_NR (deprecated)
1497 Parameters: struct kvm_assigned_msix_nr (in)
1498 Returns: 0 on success, -1 on error
1500 Set the number of MSI-X interrupts for an assigned device. The number is
1501 reset again by terminating the MSI-X assignment of the device via
1502 KVM_DEASSIGN_DEV_IRQ. Calling this service more than once at any earlier
1505 struct kvm_assigned_msix_nr {
1506 __u32 assigned_dev_id;
1511 #define KVM_MAX_MSIX_PER_DEV 256
1514 4.54 KVM_ASSIGN_SET_MSIX_ENTRY (deprecated)
1519 Parameters: struct kvm_assigned_msix_entry (in)
1520 Returns: 0 on success, -1 on error
1522 Specifies the routing of an MSI-X assigned device interrupt to a GSI. Setting
1523 the GSI vector to zero means disabling the interrupt.
1525 struct kvm_assigned_msix_entry {
1526 __u32 assigned_dev_id;
1528 __u16 entry; /* The index of entry in the MSI-X table */
1533 ENOTTY: kernel does not support this ioctl
1535 Other error conditions may be defined by individual device types or
1536 have their standard meanings.
1539 4.55 KVM_SET_TSC_KHZ
1541 Capability: KVM_CAP_TSC_CONTROL
1544 Parameters: virtual tsc_khz
1545 Returns: 0 on success, -1 on error
1547 Specifies the tsc frequency for the virtual machine. The unit of the
1551 4.56 KVM_GET_TSC_KHZ
1553 Capability: KVM_CAP_GET_TSC_KHZ
1557 Returns: virtual tsc-khz on success, negative value on error
1559 Returns the tsc frequency of the guest. The unit of the return value is
1560 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1566 Capability: KVM_CAP_IRQCHIP
1569 Parameters: struct kvm_lapic_state (out)
1570 Returns: 0 on success, -1 on error
1572 #define KVM_APIC_REG_SIZE 0x400
1573 struct kvm_lapic_state {
1574 char regs[KVM_APIC_REG_SIZE];
1577 Reads the Local APIC registers and copies them into the input argument. The
1578 data format and layout are the same as documented in the architecture manual.
1583 Capability: KVM_CAP_IRQCHIP
1586 Parameters: struct kvm_lapic_state (in)
1587 Returns: 0 on success, -1 on error
1589 #define KVM_APIC_REG_SIZE 0x400
1590 struct kvm_lapic_state {
1591 char regs[KVM_APIC_REG_SIZE];
1594 Copies the input argument into the Local APIC registers. The data format
1595 and layout are the same as documented in the architecture manual.
1600 Capability: KVM_CAP_IOEVENTFD
1603 Parameters: struct kvm_ioeventfd (in)
1604 Returns: 0 on success, !0 on error
1606 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
1607 within the guest. A guest write in the registered address will signal the
1608 provided event instead of triggering an exit.
1610 struct kvm_ioeventfd {
1612 __u64 addr; /* legal pio/mmio address */
1613 __u32 len; /* 0, 1, 2, 4, or 8 bytes */
1619 For the special case of virtio-ccw devices on s390, the ioevent is matched
1620 to a subchannel/virtqueue tuple instead.
1622 The following flags are defined:
1624 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
1625 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
1626 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
1627 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
1628 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
1630 If datamatch flag is set, the event will be signaled only if the written value
1631 to the registered address is equal to datamatch in struct kvm_ioeventfd.
1633 For virtio-ccw devices, addr contains the subchannel id and datamatch the
1636 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
1637 the kernel will ignore the length of guest write and may get a faster vmexit.
1638 The speedup may only apply to specific architectures, but the ioeventfd will
1643 Capability: KVM_CAP_SW_TLB
1646 Parameters: struct kvm_dirty_tlb (in)
1647 Returns: 0 on success, -1 on error
1649 struct kvm_dirty_tlb {
1654 This must be called whenever userspace has changed an entry in the shared
1655 TLB, prior to calling KVM_RUN on the associated vcpu.
1657 The "bitmap" field is the userspace address of an array. This array
1658 consists of a number of bits, equal to the total number of TLB entries as
1659 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
1660 nearest multiple of 64.
1662 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
1665 The array is little-endian: the bit 0 is the least significant bit of the
1666 first byte, bit 8 is the least significant bit of the second byte, etc.
1667 This avoids any complications with differing word sizes.
1669 The "num_dirty" field is a performance hint for KVM to determine whether it
1670 should skip processing the bitmap and just invalidate everything. It must
1671 be set to the number of set bits in the bitmap.
1674 4.61 KVM_ASSIGN_SET_INTX_MASK (deprecated)
1676 Capability: KVM_CAP_PCI_2_3
1679 Parameters: struct kvm_assigned_pci_dev (in)
1680 Returns: 0 on success, -1 on error
1682 Allows userspace to mask PCI INTx interrupts from the assigned device. The
1683 kernel will not deliver INTx interrupts to the guest between setting and
1684 clearing of KVM_ASSIGN_SET_INTX_MASK via this interface. This enables use of
1685 and emulation of PCI 2.3 INTx disable command register behavior.
1687 This may be used for both PCI 2.3 devices supporting INTx disable natively and
1688 older devices lacking this support. Userspace is responsible for emulating the
1689 read value of the INTx disable bit in the guest visible PCI command register.
1690 When modifying the INTx disable state, userspace should precede updating the
1691 physical device command register by calling this ioctl to inform the kernel of
1692 the new intended INTx mask state.
1694 Note that the kernel uses the device INTx disable bit to internally manage the
1695 device interrupt state for PCI 2.3 devices. Reads of this register may
1696 therefore not match the expected value. Writes should always use the guest
1697 intended INTx disable value rather than attempting to read-copy-update the
1698 current physical device state. Races between user and kernel updates to the
1699 INTx disable bit are handled lazily in the kernel. It's possible the device
1700 may generate unintended interrupts, but they will not be injected into the
1703 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1704 by assigned_dev_id. In the flags field, only KVM_DEV_ASSIGN_MASK_INTX is
1708 4.62 KVM_CREATE_SPAPR_TCE
1710 Capability: KVM_CAP_SPAPR_TCE
1711 Architectures: powerpc
1713 Parameters: struct kvm_create_spapr_tce (in)
1714 Returns: file descriptor for manipulating the created TCE table
1716 This creates a virtual TCE (translation control entry) table, which
1717 is an IOMMU for PAPR-style virtual I/O. It is used to translate
1718 logical addresses used in virtual I/O into guest physical addresses,
1719 and provides a scatter/gather capability for PAPR virtual I/O.
1721 /* for KVM_CAP_SPAPR_TCE */
1722 struct kvm_create_spapr_tce {
1727 The liobn field gives the logical IO bus number for which to create a
1728 TCE table. The window_size field specifies the size of the DMA window
1729 which this TCE table will translate - the table will contain one 64
1730 bit TCE entry for every 4kiB of the DMA window.
1732 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
1733 table has been created using this ioctl(), the kernel will handle it
1734 in real mode, updating the TCE table. H_PUT_TCE calls for other
1735 liobns will cause a vm exit and must be handled by userspace.
1737 The return value is a file descriptor which can be passed to mmap(2)
1738 to map the created TCE table into userspace. This lets userspace read
1739 the entries written by kernel-handled H_PUT_TCE calls, and also lets
1740 userspace update the TCE table directly which is useful in some
1744 4.63 KVM_ALLOCATE_RMA
1746 Capability: KVM_CAP_PPC_RMA
1747 Architectures: powerpc
1749 Parameters: struct kvm_allocate_rma (out)
1750 Returns: file descriptor for mapping the allocated RMA
1752 This allocates a Real Mode Area (RMA) from the pool allocated at boot
1753 time by the kernel. An RMA is a physically-contiguous, aligned region
1754 of memory used on older POWER processors to provide the memory which
1755 will be accessed by real-mode (MMU off) accesses in a KVM guest.
1756 POWER processors support a set of sizes for the RMA that usually
1757 includes 64MB, 128MB, 256MB and some larger powers of two.
1759 /* for KVM_ALLOCATE_RMA */
1760 struct kvm_allocate_rma {
1764 The return value is a file descriptor which can be passed to mmap(2)
1765 to map the allocated RMA into userspace. The mapped area can then be
1766 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
1767 RMA for a virtual machine. The size of the RMA in bytes (which is
1768 fixed at host kernel boot time) is returned in the rma_size field of
1769 the argument structure.
1771 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
1772 is supported; 2 if the processor requires all virtual machines to have
1773 an RMA, or 1 if the processor can use an RMA but doesn't require it,
1774 because it supports the Virtual RMA (VRMA) facility.
1779 Capability: KVM_CAP_USER_NMI
1783 Returns: 0 on success, -1 on error
1785 Queues an NMI on the thread's vcpu. Note this is well defined only
1786 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
1787 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
1788 has been called, this interface is completely emulated within the kernel.
1790 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
1791 following algorithm:
1794 - read the local APIC's state (KVM_GET_LAPIC)
1795 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
1796 - if so, issue KVM_NMI
1799 Some guests configure the LINT1 NMI input to cause a panic, aiding in
1803 4.65 KVM_S390_UCAS_MAP
1805 Capability: KVM_CAP_S390_UCONTROL
1808 Parameters: struct kvm_s390_ucas_mapping (in)
1809 Returns: 0 in case of success
1811 The parameter is defined like this:
1812 struct kvm_s390_ucas_mapping {
1818 This ioctl maps the memory at "user_addr" with the length "length" to
1819 the vcpu's address space starting at "vcpu_addr". All parameters need to
1820 be aligned by 1 megabyte.
1823 4.66 KVM_S390_UCAS_UNMAP
1825 Capability: KVM_CAP_S390_UCONTROL
1828 Parameters: struct kvm_s390_ucas_mapping (in)
1829 Returns: 0 in case of success
1831 The parameter is defined like this:
1832 struct kvm_s390_ucas_mapping {
1838 This ioctl unmaps the memory in the vcpu's address space starting at
1839 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
1840 All parameters need to be aligned by 1 megabyte.
1843 4.67 KVM_S390_VCPU_FAULT
1845 Capability: KVM_CAP_S390_UCONTROL
1848 Parameters: vcpu absolute address (in)
1849 Returns: 0 in case of success
1851 This call creates a page table entry on the virtual cpu's address space
1852 (for user controlled virtual machines) or the virtual machine's address
1853 space (for regular virtual machines). This only works for minor faults,
1854 thus it's recommended to access subject memory page via the user page
1855 table upfront. This is useful to handle validity intercepts for user
1856 controlled virtual machines to fault in the virtual cpu's lowcore pages
1857 prior to calling the KVM_RUN ioctl.
1860 4.68 KVM_SET_ONE_REG
1862 Capability: KVM_CAP_ONE_REG
1865 Parameters: struct kvm_one_reg (in)
1866 Returns: 0 on success, negative value on failure
1868 struct kvm_one_reg {
1873 Using this ioctl, a single vcpu register can be set to a specific value
1874 defined by user space with the passed in struct kvm_one_reg, where id
1875 refers to the register identifier as described below and addr is a pointer
1876 to a variable with the respective size. There can be architecture agnostic
1877 and architecture specific registers. Each have their own range of operation
1878 and their own constants and width. To keep track of the implemented
1879 registers, find a list below:
1881 Arch | Register | Width (bits)
1883 PPC | KVM_REG_PPC_HIOR | 64
1884 PPC | KVM_REG_PPC_IAC1 | 64
1885 PPC | KVM_REG_PPC_IAC2 | 64
1886 PPC | KVM_REG_PPC_IAC3 | 64
1887 PPC | KVM_REG_PPC_IAC4 | 64
1888 PPC | KVM_REG_PPC_DAC1 | 64
1889 PPC | KVM_REG_PPC_DAC2 | 64
1890 PPC | KVM_REG_PPC_DABR | 64
1891 PPC | KVM_REG_PPC_DSCR | 64
1892 PPC | KVM_REG_PPC_PURR | 64
1893 PPC | KVM_REG_PPC_SPURR | 64
1894 PPC | KVM_REG_PPC_DAR | 64
1895 PPC | KVM_REG_PPC_DSISR | 32
1896 PPC | KVM_REG_PPC_AMR | 64
1897 PPC | KVM_REG_PPC_UAMOR | 64
1898 PPC | KVM_REG_PPC_MMCR0 | 64
1899 PPC | KVM_REG_PPC_MMCR1 | 64
1900 PPC | KVM_REG_PPC_MMCRA | 64
1901 PPC | KVM_REG_PPC_MMCR2 | 64
1902 PPC | KVM_REG_PPC_MMCRS | 64
1903 PPC | KVM_REG_PPC_SIAR | 64
1904 PPC | KVM_REG_PPC_SDAR | 64
1905 PPC | KVM_REG_PPC_SIER | 64
1906 PPC | KVM_REG_PPC_PMC1 | 32
1907 PPC | KVM_REG_PPC_PMC2 | 32
1908 PPC | KVM_REG_PPC_PMC3 | 32
1909 PPC | KVM_REG_PPC_PMC4 | 32
1910 PPC | KVM_REG_PPC_PMC5 | 32
1911 PPC | KVM_REG_PPC_PMC6 | 32
1912 PPC | KVM_REG_PPC_PMC7 | 32
1913 PPC | KVM_REG_PPC_PMC8 | 32
1914 PPC | KVM_REG_PPC_FPR0 | 64
1916 PPC | KVM_REG_PPC_FPR31 | 64
1917 PPC | KVM_REG_PPC_VR0 | 128
1919 PPC | KVM_REG_PPC_VR31 | 128
1920 PPC | KVM_REG_PPC_VSR0 | 128
1922 PPC | KVM_REG_PPC_VSR31 | 128
1923 PPC | KVM_REG_PPC_FPSCR | 64
1924 PPC | KVM_REG_PPC_VSCR | 32
1925 PPC | KVM_REG_PPC_VPA_ADDR | 64
1926 PPC | KVM_REG_PPC_VPA_SLB | 128
1927 PPC | KVM_REG_PPC_VPA_DTL | 128
1928 PPC | KVM_REG_PPC_EPCR | 32
1929 PPC | KVM_REG_PPC_EPR | 32
1930 PPC | KVM_REG_PPC_TCR | 32
1931 PPC | KVM_REG_PPC_TSR | 32
1932 PPC | KVM_REG_PPC_OR_TSR | 32
1933 PPC | KVM_REG_PPC_CLEAR_TSR | 32
1934 PPC | KVM_REG_PPC_MAS0 | 32
1935 PPC | KVM_REG_PPC_MAS1 | 32
1936 PPC | KVM_REG_PPC_MAS2 | 64
1937 PPC | KVM_REG_PPC_MAS7_3 | 64
1938 PPC | KVM_REG_PPC_MAS4 | 32
1939 PPC | KVM_REG_PPC_MAS6 | 32
1940 PPC | KVM_REG_PPC_MMUCFG | 32
1941 PPC | KVM_REG_PPC_TLB0CFG | 32
1942 PPC | KVM_REG_PPC_TLB1CFG | 32
1943 PPC | KVM_REG_PPC_TLB2CFG | 32
1944 PPC | KVM_REG_PPC_TLB3CFG | 32
1945 PPC | KVM_REG_PPC_TLB0PS | 32
1946 PPC | KVM_REG_PPC_TLB1PS | 32
1947 PPC | KVM_REG_PPC_TLB2PS | 32
1948 PPC | KVM_REG_PPC_TLB3PS | 32
1949 PPC | KVM_REG_PPC_EPTCFG | 32
1950 PPC | KVM_REG_PPC_ICP_STATE | 64
1951 PPC | KVM_REG_PPC_TB_OFFSET | 64
1952 PPC | KVM_REG_PPC_SPMC1 | 32
1953 PPC | KVM_REG_PPC_SPMC2 | 32
1954 PPC | KVM_REG_PPC_IAMR | 64
1955 PPC | KVM_REG_PPC_TFHAR | 64
1956 PPC | KVM_REG_PPC_TFIAR | 64
1957 PPC | KVM_REG_PPC_TEXASR | 64
1958 PPC | KVM_REG_PPC_FSCR | 64
1959 PPC | KVM_REG_PPC_PSPB | 32
1960 PPC | KVM_REG_PPC_EBBHR | 64
1961 PPC | KVM_REG_PPC_EBBRR | 64
1962 PPC | KVM_REG_PPC_BESCR | 64
1963 PPC | KVM_REG_PPC_TAR | 64
1964 PPC | KVM_REG_PPC_DPDES | 64
1965 PPC | KVM_REG_PPC_DAWR | 64
1966 PPC | KVM_REG_PPC_DAWRX | 64
1967 PPC | KVM_REG_PPC_CIABR | 64
1968 PPC | KVM_REG_PPC_IC | 64
1969 PPC | KVM_REG_PPC_VTB | 64
1970 PPC | KVM_REG_PPC_CSIGR | 64
1971 PPC | KVM_REG_PPC_TACR | 64
1972 PPC | KVM_REG_PPC_TCSCR | 64
1973 PPC | KVM_REG_PPC_PID | 64
1974 PPC | KVM_REG_PPC_ACOP | 64
1975 PPC | KVM_REG_PPC_VRSAVE | 32
1976 PPC | KVM_REG_PPC_LPCR | 32
1977 PPC | KVM_REG_PPC_LPCR_64 | 64
1978 PPC | KVM_REG_PPC_PPR | 64
1979 PPC | KVM_REG_PPC_ARCH_COMPAT | 32
1980 PPC | KVM_REG_PPC_DABRX | 32
1981 PPC | KVM_REG_PPC_WORT | 64
1982 PPC | KVM_REG_PPC_SPRG9 | 64
1983 PPC | KVM_REG_PPC_DBSR | 32
1984 PPC | KVM_REG_PPC_TM_GPR0 | 64
1986 PPC | KVM_REG_PPC_TM_GPR31 | 64
1987 PPC | KVM_REG_PPC_TM_VSR0 | 128
1989 PPC | KVM_REG_PPC_TM_VSR63 | 128
1990 PPC | KVM_REG_PPC_TM_CR | 64
1991 PPC | KVM_REG_PPC_TM_LR | 64
1992 PPC | KVM_REG_PPC_TM_CTR | 64
1993 PPC | KVM_REG_PPC_TM_FPSCR | 64
1994 PPC | KVM_REG_PPC_TM_AMR | 64
1995 PPC | KVM_REG_PPC_TM_PPR | 64
1996 PPC | KVM_REG_PPC_TM_VRSAVE | 64
1997 PPC | KVM_REG_PPC_TM_VSCR | 32
1998 PPC | KVM_REG_PPC_TM_DSCR | 64
1999 PPC | KVM_REG_PPC_TM_TAR | 64
2001 MIPS | KVM_REG_MIPS_R0 | 64
2003 MIPS | KVM_REG_MIPS_R31 | 64
2004 MIPS | KVM_REG_MIPS_HI | 64
2005 MIPS | KVM_REG_MIPS_LO | 64
2006 MIPS | KVM_REG_MIPS_PC | 64
2007 MIPS | KVM_REG_MIPS_CP0_INDEX | 32
2008 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64
2009 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64
2010 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32
2011 MIPS | KVM_REG_MIPS_CP0_WIRED | 32
2012 MIPS | KVM_REG_MIPS_CP0_HWRENA | 32
2013 MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64
2014 MIPS | KVM_REG_MIPS_CP0_COUNT | 32
2015 MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64
2016 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32
2017 MIPS | KVM_REG_MIPS_CP0_STATUS | 32
2018 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32
2019 MIPS | KVM_REG_MIPS_CP0_EPC | 64
2020 MIPS | KVM_REG_MIPS_CP0_PRID | 32
2021 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32
2022 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32
2023 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32
2024 MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32
2025 MIPS | KVM_REG_MIPS_CP0_CONFIG4 | 32
2026 MIPS | KVM_REG_MIPS_CP0_CONFIG5 | 32
2027 MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32
2028 MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64
2029 MIPS | KVM_REG_MIPS_COUNT_CTL | 64
2030 MIPS | KVM_REG_MIPS_COUNT_RESUME | 64
2031 MIPS | KVM_REG_MIPS_COUNT_HZ | 64
2032 MIPS | KVM_REG_MIPS_FPR_32(0..31) | 32
2033 MIPS | KVM_REG_MIPS_FPR_64(0..31) | 64
2034 MIPS | KVM_REG_MIPS_VEC_128(0..31) | 128
2035 MIPS | KVM_REG_MIPS_FCR_IR | 32
2036 MIPS | KVM_REG_MIPS_FCR_CSR | 32
2037 MIPS | KVM_REG_MIPS_MSA_IR | 32
2038 MIPS | KVM_REG_MIPS_MSA_CSR | 32
2040 ARM registers are mapped using the lower 32 bits. The upper 16 of that
2041 is the register group type, or coprocessor number:
2043 ARM core registers have the following id bit patterns:
2044 0x4020 0000 0010 <index into the kvm_regs struct:16>
2046 ARM 32-bit CP15 registers have the following id bit patterns:
2047 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
2049 ARM 64-bit CP15 registers have the following id bit patterns:
2050 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
2052 ARM CCSIDR registers are demultiplexed by CSSELR value:
2053 0x4020 0000 0011 00 <csselr:8>
2055 ARM 32-bit VFP control registers have the following id bit patterns:
2056 0x4020 0000 0012 1 <regno:12>
2058 ARM 64-bit FP registers have the following id bit patterns:
2059 0x4030 0000 0012 0 <regno:12>
2062 arm64 registers are mapped using the lower 32 bits. The upper 16 of
2063 that is the register group type, or coprocessor number:
2065 arm64 core/FP-SIMD registers have the following id bit patterns. Note
2066 that the size of the access is variable, as the kvm_regs structure
2067 contains elements ranging from 32 to 128 bits. The index is a 32bit
2068 value in the kvm_regs structure seen as a 32bit array.
2069 0x60x0 0000 0010 <index into the kvm_regs struct:16>
2071 arm64 CCSIDR registers are demultiplexed by CSSELR value:
2072 0x6020 0000 0011 00 <csselr:8>
2074 arm64 system registers have the following id bit patterns:
2075 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
2078 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
2079 the register group type:
2081 MIPS core registers (see above) have the following id bit patterns:
2082 0x7030 0000 0000 <reg:16>
2084 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
2085 patterns depending on whether they're 32-bit or 64-bit registers:
2086 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
2087 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2089 MIPS KVM control registers (see above) have the following id bit patterns:
2090 0x7030 0000 0002 <reg:16>
2092 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
2093 id bit patterns depending on the size of the register being accessed. They are
2094 always accessed according to the current guest FPU mode (Status.FR and
2095 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
2096 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector
2097 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they
2098 overlap the FPU registers:
2099 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
2100 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
2101 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers)
2103 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
2104 following id bit patterns:
2105 0x7020 0000 0003 01 <0:3> <reg:5>
2107 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the
2108 following id bit patterns:
2109 0x7020 0000 0003 02 <0:3> <reg:5>
2112 4.69 KVM_GET_ONE_REG
2114 Capability: KVM_CAP_ONE_REG
2117 Parameters: struct kvm_one_reg (in and out)
2118 Returns: 0 on success, negative value on failure
2120 This ioctl allows to receive the value of a single register implemented
2121 in a vcpu. The register to read is indicated by the "id" field of the
2122 kvm_one_reg struct passed in. On success, the register value can be found
2123 at the memory location pointed to by "addr".
2125 The list of registers accessible using this interface is identical to the
2129 4.70 KVM_KVMCLOCK_CTRL
2131 Capability: KVM_CAP_KVMCLOCK_CTRL
2132 Architectures: Any that implement pvclocks (currently x86 only)
2135 Returns: 0 on success, -1 on error
2137 This signals to the host kernel that the specified guest is being paused by
2138 userspace. The host will set a flag in the pvclock structure that is checked
2139 from the soft lockup watchdog. The flag is part of the pvclock structure that
2140 is shared between guest and host, specifically the second bit of the flags
2141 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2142 the host and read/cleared exclusively by the guest. The guest operation of
2143 checking and clearing the flag must an atomic operation so
2144 load-link/store-conditional, or equivalent must be used. There are two cases
2145 where the guest will clear the flag: when the soft lockup watchdog timer resets
2146 itself or when a soft lockup is detected. This ioctl can be called any time
2147 after pausing the vcpu, but before it is resumed.
2152 Capability: KVM_CAP_SIGNAL_MSI
2155 Parameters: struct kvm_msi (in)
2156 Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2158 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2169 No flags are defined so far. The corresponding field must be 0.
2172 4.71 KVM_CREATE_PIT2
2174 Capability: KVM_CAP_PIT2
2177 Parameters: struct kvm_pit_config (in)
2178 Returns: 0 on success, -1 on error
2180 Creates an in-kernel device model for the i8254 PIT. This call is only valid
2181 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
2182 parameters have to be passed:
2184 struct kvm_pit_config {
2191 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
2193 PIT timer interrupts may use a per-VM kernel thread for injection. If it
2194 exists, this thread will have a name of the following pattern:
2196 kvm-pit/<owner-process-pid>
2198 When running a guest with elevated priorities, the scheduling parameters of
2199 this thread may have to be adjusted accordingly.
2201 This IOCTL replaces the obsolete KVM_CREATE_PIT.
2206 Capability: KVM_CAP_PIT_STATE2
2209 Parameters: struct kvm_pit_state2 (out)
2210 Returns: 0 on success, -1 on error
2212 Retrieves the state of the in-kernel PIT model. Only valid after
2213 KVM_CREATE_PIT2. The state is returned in the following structure:
2215 struct kvm_pit_state2 {
2216 struct kvm_pit_channel_state channels[3];
2223 /* disable PIT in HPET legacy mode */
2224 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
2226 This IOCTL replaces the obsolete KVM_GET_PIT.
2231 Capability: KVM_CAP_PIT_STATE2
2234 Parameters: struct kvm_pit_state2 (in)
2235 Returns: 0 on success, -1 on error
2237 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
2238 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
2240 This IOCTL replaces the obsolete KVM_SET_PIT.
2243 4.74 KVM_PPC_GET_SMMU_INFO
2245 Capability: KVM_CAP_PPC_GET_SMMU_INFO
2246 Architectures: powerpc
2249 Returns: 0 on success, -1 on error
2251 This populates and returns a structure describing the features of
2252 the "Server" class MMU emulation supported by KVM.
2253 This can in turn be used by userspace to generate the appropriate
2254 device-tree properties for the guest operating system.
2256 The structure contains some global information, followed by an
2257 array of supported segment page sizes:
2259 struct kvm_ppc_smmu_info {
2263 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
2266 The supported flags are:
2268 - KVM_PPC_PAGE_SIZES_REAL:
2269 When that flag is set, guest page sizes must "fit" the backing
2270 store page sizes. When not set, any page size in the list can
2271 be used regardless of how they are backed by userspace.
2273 - KVM_PPC_1T_SEGMENTS
2274 The emulated MMU supports 1T segments in addition to the
2277 The "slb_size" field indicates how many SLB entries are supported
2279 The "sps" array contains 8 entries indicating the supported base
2280 page sizes for a segment in increasing order. Each entry is defined
2283 struct kvm_ppc_one_seg_page_size {
2284 __u32 page_shift; /* Base page shift of segment (or 0) */
2285 __u32 slb_enc; /* SLB encoding for BookS */
2286 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
2289 An entry with a "page_shift" of 0 is unused. Because the array is
2290 organized in increasing order, a lookup can stop when encoutering
2293 The "slb_enc" field provides the encoding to use in the SLB for the
2294 page size. The bits are in positions such as the value can directly
2295 be OR'ed into the "vsid" argument of the slbmte instruction.
2297 The "enc" array is a list which for each of those segment base page
2298 size provides the list of supported actual page sizes (which can be
2299 only larger or equal to the base page size), along with the
2300 corresponding encoding in the hash PTE. Similarly, the array is
2301 8 entries sorted by increasing sizes and an entry with a "0" shift
2302 is an empty entry and a terminator:
2304 struct kvm_ppc_one_page_size {
2305 __u32 page_shift; /* Page shift (or 0) */
2306 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
2309 The "pte_enc" field provides a value that can OR'ed into the hash
2310 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
2311 into the hash PTE second double word).
2315 Capability: KVM_CAP_IRQFD
2316 Architectures: x86 s390 arm arm64
2318 Parameters: struct kvm_irqfd (in)
2319 Returns: 0 on success, -1 on error
2321 Allows setting an eventfd to directly trigger a guest interrupt.
2322 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
2323 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
2324 an event is triggered on the eventfd, an interrupt is injected into
2325 the guest using the specified gsi pin. The irqfd is removed using
2326 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
2329 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
2330 mechanism allowing emulation of level-triggered, irqfd-based
2331 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
2332 additional eventfd in the kvm_irqfd.resamplefd field. When operating
2333 in resample mode, posting of an interrupt through kvm_irq.fd asserts
2334 the specified gsi in the irqchip. When the irqchip is resampled, such
2335 as from an EOI, the gsi is de-asserted and the user is notified via
2336 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
2337 the interrupt if the device making use of it still requires service.
2338 Note that closing the resamplefd is not sufficient to disable the
2339 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
2340 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
2342 On ARM/ARM64, the gsi field in the kvm_irqfd struct specifies the Shared
2343 Peripheral Interrupt (SPI) index, such that the GIC interrupt ID is
2346 4.76 KVM_PPC_ALLOCATE_HTAB
2348 Capability: KVM_CAP_PPC_ALLOC_HTAB
2349 Architectures: powerpc
2351 Parameters: Pointer to u32 containing hash table order (in/out)
2352 Returns: 0 on success, -1 on error
2354 This requests the host kernel to allocate an MMU hash table for a
2355 guest using the PAPR paravirtualization interface. This only does
2356 anything if the kernel is configured to use the Book 3S HV style of
2357 virtualization. Otherwise the capability doesn't exist and the ioctl
2358 returns an ENOTTY error. The rest of this description assumes Book 3S
2361 There must be no vcpus running when this ioctl is called; if there
2362 are, it will do nothing and return an EBUSY error.
2364 The parameter is a pointer to a 32-bit unsigned integer variable
2365 containing the order (log base 2) of the desired size of the hash
2366 table, which must be between 18 and 46. On successful return from the
2367 ioctl, it will have been updated with the order of the hash table that
2370 If no hash table has been allocated when any vcpu is asked to run
2371 (with the KVM_RUN ioctl), the host kernel will allocate a
2372 default-sized hash table (16 MB).
2374 If this ioctl is called when a hash table has already been allocated,
2375 the kernel will clear out the existing hash table (zero all HPTEs) and
2376 return the hash table order in the parameter. (If the guest is using
2377 the virtualized real-mode area (VRMA) facility, the kernel will
2378 re-create the VMRA HPTEs on the next KVM_RUN of any vcpu.)
2380 4.77 KVM_S390_INTERRUPT
2384 Type: vm ioctl, vcpu ioctl
2385 Parameters: struct kvm_s390_interrupt (in)
2386 Returns: 0 on success, -1 on error
2388 Allows to inject an interrupt to the guest. Interrupts can be floating
2389 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
2391 Interrupt parameters are passed via kvm_s390_interrupt:
2393 struct kvm_s390_interrupt {
2399 type can be one of the following:
2401 KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm
2402 KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm
2403 KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm
2404 KVM_S390_RESTART (vcpu) - restart
2405 KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt
2406 KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt
2407 KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt
2408 parameters in parm and parm64
2409 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm
2410 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm
2411 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm
2412 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
2413 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
2414 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
2415 interruption subclass)
2416 KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm,
2417 machine check interrupt code in parm64 (note that
2418 machine checks needing further payload are not
2419 supported by this ioctl)
2421 Note that the vcpu ioctl is asynchronous to vcpu execution.
2423 4.78 KVM_PPC_GET_HTAB_FD
2425 Capability: KVM_CAP_PPC_HTAB_FD
2426 Architectures: powerpc
2428 Parameters: Pointer to struct kvm_get_htab_fd (in)
2429 Returns: file descriptor number (>= 0) on success, -1 on error
2431 This returns a file descriptor that can be used either to read out the
2432 entries in the guest's hashed page table (HPT), or to write entries to
2433 initialize the HPT. The returned fd can only be written to if the
2434 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
2435 can only be read if that bit is clear. The argument struct looks like
2438 /* For KVM_PPC_GET_HTAB_FD */
2439 struct kvm_get_htab_fd {
2445 /* Values for kvm_get_htab_fd.flags */
2446 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
2447 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
2449 The `start_index' field gives the index in the HPT of the entry at
2450 which to start reading. It is ignored when writing.
2452 Reads on the fd will initially supply information about all
2453 "interesting" HPT entries. Interesting entries are those with the
2454 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
2455 all entries. When the end of the HPT is reached, the read() will
2456 return. If read() is called again on the fd, it will start again from
2457 the beginning of the HPT, but will only return HPT entries that have
2458 changed since they were last read.
2460 Data read or written is structured as a header (8 bytes) followed by a
2461 series of valid HPT entries (16 bytes) each. The header indicates how
2462 many valid HPT entries there are and how many invalid entries follow
2463 the valid entries. The invalid entries are not represented explicitly
2464 in the stream. The header format is:
2466 struct kvm_get_htab_header {
2472 Writes to the fd create HPT entries starting at the index given in the
2473 header; first `n_valid' valid entries with contents from the data
2474 written, then `n_invalid' invalid entries, invalidating any previously
2475 valid entries found.
2477 4.79 KVM_CREATE_DEVICE
2479 Capability: KVM_CAP_DEVICE_CTRL
2481 Parameters: struct kvm_create_device (in/out)
2482 Returns: 0 on success, -1 on error
2484 ENODEV: The device type is unknown or unsupported
2485 EEXIST: Device already created, and this type of device may not
2486 be instantiated multiple times
2488 Other error conditions may be defined by individual device types or
2489 have their standard meanings.
2491 Creates an emulated device in the kernel. The file descriptor returned
2492 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
2494 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
2495 device type is supported (not necessarily whether it can be created
2498 Individual devices should not define flags. Attributes should be used
2499 for specifying any behavior that is not implied by the device type
2502 struct kvm_create_device {
2503 __u32 type; /* in: KVM_DEV_TYPE_xxx */
2504 __u32 fd; /* out: device handle */
2505 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
2508 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
2510 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device
2511 Type: device ioctl, vm ioctl
2512 Parameters: struct kvm_device_attr
2513 Returns: 0 on success, -1 on error
2515 ENXIO: The group or attribute is unknown/unsupported for this device
2516 EPERM: The attribute cannot (currently) be accessed this way
2517 (e.g. read-only attribute, or attribute that only makes
2518 sense when the device is in a different state)
2520 Other error conditions may be defined by individual device types.
2522 Gets/sets a specified piece of device configuration and/or state. The
2523 semantics are device-specific. See individual device documentation in
2524 the "devices" directory. As with ONE_REG, the size of the data
2525 transferred is defined by the particular attribute.
2527 struct kvm_device_attr {
2528 __u32 flags; /* no flags currently defined */
2529 __u32 group; /* device-defined */
2530 __u64 attr; /* group-defined */
2531 __u64 addr; /* userspace address of attr data */
2534 4.81 KVM_HAS_DEVICE_ATTR
2536 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device
2537 Type: device ioctl, vm ioctl
2538 Parameters: struct kvm_device_attr
2539 Returns: 0 on success, -1 on error
2541 ENXIO: The group or attribute is unknown/unsupported for this device
2543 Tests whether a device supports a particular attribute. A successful
2544 return indicates the attribute is implemented. It does not necessarily
2545 indicate that the attribute can be read or written in the device's
2546 current state. "addr" is ignored.
2548 4.82 KVM_ARM_VCPU_INIT
2551 Architectures: arm, arm64
2553 Parameters: struct kvm_vcpu_init (in)
2554 Returns: 0 on success; -1 on error
2556 Â EINVAL: Â Â Â the target is unknown, or the combination of features is invalid.
2557 Â ENOENT: Â Â Â a features bit specified is unknown.
2559 This tells KVM what type of CPU to present to the guest, and what
2560 optional features it should have. Â This will cause a reset of the cpu
2561 registers to their initial values. Â If this is not called, KVM_RUN will
2562 return ENOEXEC for that vcpu.
2564 Note that because some registers reflect machine topology, all vcpus
2565 should be created before this ioctl is invoked.
2567 Userspace can call this function multiple times for a given vcpu, including
2568 after the vcpu has been run. This will reset the vcpu to its initial
2569 state. All calls to this function after the initial call must use the same
2570 target and same set of feature flags, otherwise EINVAL will be returned.
2573 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
2574 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
2575 and execute guest code when KVM_RUN is called.
2576 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
2577 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
2578 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU.
2579 Depends on KVM_CAP_ARM_PSCI_0_2.
2582 4.83 KVM_ARM_PREFERRED_TARGET
2585 Architectures: arm, arm64
2587 Parameters: struct struct kvm_vcpu_init (out)
2588 Returns: 0 on success; -1 on error
2590 ENODEV: no preferred target available for the host
2592 This queries KVM for preferred CPU target type which can be emulated
2593 by KVM on underlying host.
2595 The ioctl returns struct kvm_vcpu_init instance containing information
2596 about preferred CPU target type and recommended features for it. The
2597 kvm_vcpu_init->features bitmap returned will have feature bits set if
2598 the preferred target recommends setting these features, but this is
2601 The information returned by this ioctl can be used to prepare an instance
2602 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
2603 in VCPU matching underlying host.
2606 4.84 KVM_GET_REG_LIST
2609 Architectures: arm, arm64, mips
2611 Parameters: struct kvm_reg_list (in/out)
2612 Returns: 0 on success; -1 on error
2614 Â E2BIG: Â Â Â Â the reg index list is too big to fit in the array specified by
2615 Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n).
2617 struct kvm_reg_list {
2618 __u64 n; /* number of registers in reg[] */
2622 This ioctl returns the guest registers that are supported for the
2623 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
2626 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
2628 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
2629 Architectures: arm, arm64
2631 Parameters: struct kvm_arm_device_address (in)
2632 Returns: 0 on success, -1 on error
2634 ENODEV: The device id is unknown
2635 ENXIO: Device not supported on current system
2636 EEXIST: Address already set
2637 E2BIG: Address outside guest physical address space
2638 EBUSY: Address overlaps with other device range
2640 struct kvm_arm_device_addr {
2645 Specify a device address in the guest's physical address space where guests
2646 can access emulated or directly exposed devices, which the host kernel needs
2647 to know about. The id field is an architecture specific identifier for a
2650 ARM/arm64 divides the id field into two parts, a device id and an
2651 address type id specific to the individual device.
2653 Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
2654 field: | 0x00000000 | device id | addr type id |
2656 ARM/arm64 currently only require this when using the in-kernel GIC
2657 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
2658 as the device id. When setting the base address for the guest's
2659 mapping of the VGIC virtual CPU and distributor interface, the ioctl
2660 must be called after calling KVM_CREATE_IRQCHIP, but before calling
2661 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
2662 base addresses will return -EEXIST.
2664 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
2665 should be used instead.
2668 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
2670 Capability: KVM_CAP_PPC_RTAS
2673 Parameters: struct kvm_rtas_token_args
2674 Returns: 0 on success, -1 on error
2676 Defines a token value for a RTAS (Run Time Abstraction Services)
2677 service in order to allow it to be handled in the kernel. The
2678 argument struct gives the name of the service, which must be the name
2679 of a service that has a kernel-side implementation. If the token
2680 value is non-zero, it will be associated with that service, and
2681 subsequent RTAS calls by the guest specifying that token will be
2682 handled by the kernel. If the token value is 0, then any token
2683 associated with the service will be forgotten, and subsequent RTAS
2684 calls by the guest for that service will be passed to userspace to be
2687 4.87 KVM_SET_GUEST_DEBUG
2689 Capability: KVM_CAP_SET_GUEST_DEBUG
2690 Architectures: x86, s390, ppc, arm64
2692 Parameters: struct kvm_guest_debug (in)
2693 Returns: 0 on success; -1 on error
2695 struct kvm_guest_debug {
2698 struct kvm_guest_debug_arch arch;
2701 Set up the processor specific debug registers and configure vcpu for
2702 handling guest debug events. There are two parts to the structure, the
2703 first a control bitfield indicates the type of debug events to handle
2704 when running. Common control bits are:
2706 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
2707 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
2709 The top 16 bits of the control field are architecture specific control
2710 flags which can include the following:
2712 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64]
2713 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64]
2714 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
2715 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
2716 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
2718 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
2719 are enabled in memory so we need to ensure breakpoint exceptions are
2720 correctly trapped and the KVM run loop exits at the breakpoint and not
2721 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
2722 we need to ensure the guest vCPUs architecture specific registers are
2723 updated to the correct (supplied) values.
2725 The second part of the structure is architecture specific and
2726 typically contains a set of debug registers.
2728 For arm64 the number of debug registers is implementation defined and
2729 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
2730 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
2731 indicating the number of supported registers.
2733 When debug events exit the main run loop with the reason
2734 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
2735 structure containing architecture specific debug information.
2737 4.88 KVM_GET_EMULATED_CPUID
2739 Capability: KVM_CAP_EXT_EMUL_CPUID
2742 Parameters: struct kvm_cpuid2 (in/out)
2743 Returns: 0 on success, -1 on error
2748 struct kvm_cpuid_entry2 entries[0];
2751 The member 'flags' is used for passing flags from userspace.
2753 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
2754 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
2755 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
2757 struct kvm_cpuid_entry2 {
2768 This ioctl returns x86 cpuid features which are emulated by
2769 kvm.Userspace can use the information returned by this ioctl to query
2770 which features are emulated by kvm instead of being present natively.
2772 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
2773 structure with the 'nent' field indicating the number of entries in
2774 the variable-size array 'entries'. If the number of entries is too low
2775 to describe the cpu capabilities, an error (E2BIG) is returned. If the
2776 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
2777 is returned. If the number is just right, the 'nent' field is adjusted
2778 to the number of valid entries in the 'entries' array, which is then
2781 The entries returned are the set CPUID bits of the respective features
2782 which kvm emulates, as returned by the CPUID instruction, with unknown
2783 or unsupported feature bits cleared.
2785 Features like x2apic, for example, may not be present in the host cpu
2786 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
2787 emulated efficiently and thus not included here.
2789 The fields in each entry are defined as follows:
2791 function: the eax value used to obtain the entry
2792 index: the ecx value used to obtain the entry (for entries that are
2794 flags: an OR of zero or more of the following:
2795 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
2796 if the index field is valid
2797 KVM_CPUID_FLAG_STATEFUL_FUNC:
2798 if cpuid for this function returns different values for successive
2799 invocations; there will be several entries with the same function,
2800 all with this flag set
2801 KVM_CPUID_FLAG_STATE_READ_NEXT:
2802 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
2803 the first entry to be read by a cpu
2804 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
2805 this function/index combination
2807 4.89 KVM_S390_MEM_OP
2809 Capability: KVM_CAP_S390_MEM_OP
2812 Parameters: struct kvm_s390_mem_op (in)
2813 Returns: = 0 on success,
2814 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
2815 > 0 if an exception occurred while walking the page tables
2817 Read or write data from/to the logical (virtual) memory of a VCPU.
2819 Parameters are specified via the following structure:
2821 struct kvm_s390_mem_op {
2822 __u64 gaddr; /* the guest address */
2823 __u64 flags; /* flags */
2824 __u32 size; /* amount of bytes */
2825 __u32 op; /* type of operation */
2826 __u64 buf; /* buffer in userspace */
2827 __u8 ar; /* the access register number */
2828 __u8 reserved[31]; /* should be set to 0 */
2831 The type of operation is specified in the "op" field. It is either
2832 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or
2833 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The
2834 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check
2835 whether the corresponding memory access would create an access exception
2836 (without touching the data in the memory at the destination). In case an
2837 access exception occurred while walking the MMU tables of the guest, the
2838 ioctl returns a positive error number to indicate the type of exception.
2839 This exception is also raised directly at the corresponding VCPU if the
2840 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field.
2842 The start address of the memory region has to be specified in the "gaddr"
2843 field, and the length of the region in the "size" field. "buf" is the buffer
2844 supplied by the userspace application where the read data should be written
2845 to for KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written
2846 is stored for a KVM_S390_MEMOP_LOGICAL_WRITE. "buf" is unused and can be NULL
2847 when KVM_S390_MEMOP_F_CHECK_ONLY is specified. "ar" designates the access
2848 register number to be used.
2850 The "reserved" field is meant for future extensions. It is not used by
2851 KVM with the currently defined set of flags.
2853 4.90 KVM_S390_GET_SKEYS
2855 Capability: KVM_CAP_S390_SKEYS
2858 Parameters: struct kvm_s390_skeys
2859 Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage
2860 keys, negative value on error
2862 This ioctl is used to get guest storage key values on the s390
2863 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2865 struct kvm_s390_skeys {
2868 __u64 skeydata_addr;
2873 The start_gfn field is the number of the first guest frame whose storage keys
2876 The count field is the number of consecutive frames (starting from start_gfn)
2877 whose storage keys to get. The count field must be at least 1 and the maximum
2878 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2879 will cause the ioctl to return -EINVAL.
2881 The skeydata_addr field is the address to a buffer large enough to hold count
2882 bytes. This buffer will be filled with storage key data by the ioctl.
2884 4.91 KVM_S390_SET_SKEYS
2886 Capability: KVM_CAP_S390_SKEYS
2889 Parameters: struct kvm_s390_skeys
2890 Returns: 0 on success, negative value on error
2892 This ioctl is used to set guest storage key values on the s390
2893 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2894 See section on KVM_S390_GET_SKEYS for struct definition.
2896 The start_gfn field is the number of the first guest frame whose storage keys
2899 The count field is the number of consecutive frames (starting from start_gfn)
2900 whose storage keys to get. The count field must be at least 1 and the maximum
2901 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2902 will cause the ioctl to return -EINVAL.
2904 The skeydata_addr field is the address to a buffer containing count bytes of
2905 storage keys. Each byte in the buffer will be set as the storage key for a
2906 single frame starting at start_gfn for count frames.
2908 Note: If any architecturally invalid key value is found in the given data then
2909 the ioctl will return -EINVAL.
2913 Capability: KVM_CAP_S390_INJECT_IRQ
2916 Parameters: struct kvm_s390_irq (in)
2917 Returns: 0 on success, -1 on error
2919 EINVAL: interrupt type is invalid
2920 type is KVM_S390_SIGP_STOP and flag parameter is invalid value
2921 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger
2922 than the maximum of VCPUs
2923 EBUSY: type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped
2924 type is KVM_S390_SIGP_STOP and a stop irq is already pending
2925 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt
2928 Allows to inject an interrupt to the guest.
2930 Using struct kvm_s390_irq as a parameter allows
2931 to inject additional payload which is not
2932 possible via KVM_S390_INTERRUPT.
2934 Interrupt parameters are passed via kvm_s390_irq:
2936 struct kvm_s390_irq {
2939 struct kvm_s390_io_info io;
2940 struct kvm_s390_ext_info ext;
2941 struct kvm_s390_pgm_info pgm;
2942 struct kvm_s390_emerg_info emerg;
2943 struct kvm_s390_extcall_info extcall;
2944 struct kvm_s390_prefix_info prefix;
2945 struct kvm_s390_stop_info stop;
2946 struct kvm_s390_mchk_info mchk;
2951 type can be one of the following:
2953 KVM_S390_SIGP_STOP - sigp stop; parameter in .stop
2954 KVM_S390_PROGRAM_INT - program check; parameters in .pgm
2955 KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix
2956 KVM_S390_RESTART - restart; no parameters
2957 KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters
2958 KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters
2959 KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg
2960 KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall
2961 KVM_S390_MCHK - machine check interrupt; parameters in .mchk
2964 Note that the vcpu ioctl is asynchronous to vcpu execution.
2966 4.94 KVM_S390_GET_IRQ_STATE
2968 Capability: KVM_CAP_S390_IRQ_STATE
2971 Parameters: struct kvm_s390_irq_state (out)
2972 Returns: >= number of bytes copied into buffer,
2973 -EINVAL if buffer size is 0,
2974 -ENOBUFS if buffer size is too small to fit all pending interrupts,
2975 -EFAULT if the buffer address was invalid
2977 This ioctl allows userspace to retrieve the complete state of all currently
2978 pending interrupts in a single buffer. Use cases include migration
2979 and introspection. The parameter structure contains the address of a
2980 userspace buffer and its length:
2982 struct kvm_s390_irq_state {
2989 Userspace passes in the above struct and for each pending interrupt a
2990 struct kvm_s390_irq is copied to the provided buffer.
2992 If -ENOBUFS is returned the buffer provided was too small and userspace
2993 may retry with a bigger buffer.
2995 4.95 KVM_S390_SET_IRQ_STATE
2997 Capability: KVM_CAP_S390_IRQ_STATE
3000 Parameters: struct kvm_s390_irq_state (in)
3001 Returns: 0 on success,
3002 -EFAULT if the buffer address was invalid,
3003 -EINVAL for an invalid buffer length (see below),
3004 -EBUSY if there were already interrupts pending,
3005 errors occurring when actually injecting the
3006 interrupt. See KVM_S390_IRQ.
3008 This ioctl allows userspace to set the complete state of all cpu-local
3009 interrupts currently pending for the vcpu. It is intended for restoring
3010 interrupt state after a migration. The input parameter is a userspace buffer
3011 containing a struct kvm_s390_irq_state:
3013 struct kvm_s390_irq_state {
3019 The userspace memory referenced by buf contains a struct kvm_s390_irq
3020 for each interrupt to be injected into the guest.
3021 If one of the interrupts could not be injected for some reason the
3024 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
3025 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
3026 which is the maximum number of possibly pending cpu-local interrupts.
3030 Capability: KVM_CAP_X86_SMM
3034 Returns: 0 on success, -1 on error
3036 Queues an SMI on the thread's vcpu.
3038 5. The kvm_run structure
3039 ------------------------
3041 Application code obtains a pointer to the kvm_run structure by
3042 mmap()ing a vcpu fd. From that point, application code can control
3043 execution by changing fields in kvm_run prior to calling the KVM_RUN
3044 ioctl, and obtain information about the reason KVM_RUN returned by
3045 looking up structure members.
3049 __u8 request_interrupt_window;
3051 Request that KVM_RUN return when it becomes possible to inject external
3052 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
3059 When KVM_RUN has returned successfully (return value 0), this informs
3060 application code why KVM_RUN has returned. Allowable values for this
3061 field are detailed below.
3063 __u8 ready_for_interrupt_injection;
3065 If request_interrupt_window has been specified, this field indicates
3066 an interrupt can be injected now with KVM_INTERRUPT.
3070 The value of the current interrupt flag. Only valid if in-kernel
3071 local APIC is not used.
3075 More architecture-specific flags detailing state of the VCPU that may
3076 affect the device's behavior. The only currently defined flag is
3077 KVM_RUN_X86_SMM, which is valid on x86 machines and is set if the
3078 VCPU is in system management mode.
3080 /* in (pre_kvm_run), out (post_kvm_run) */
3083 The value of the cr8 register. Only valid if in-kernel local APIC is
3084 not used. Both input and output.
3088 The value of the APIC BASE msr. Only valid if in-kernel local
3089 APIC is not used. Both input and output.
3092 /* KVM_EXIT_UNKNOWN */
3094 __u64 hardware_exit_reason;
3097 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
3098 reasons. Further architecture-specific information is available in
3099 hardware_exit_reason.
3101 /* KVM_EXIT_FAIL_ENTRY */
3103 __u64 hardware_entry_failure_reason;
3106 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
3107 to unknown reasons. Further architecture-specific information is
3108 available in hardware_entry_failure_reason.
3110 /* KVM_EXIT_EXCEPTION */
3120 #define KVM_EXIT_IO_IN 0
3121 #define KVM_EXIT_IO_OUT 1
3123 __u8 size; /* bytes */
3126 __u64 data_offset; /* relative to kvm_run start */
3129 If exit_reason is KVM_EXIT_IO, then the vcpu has
3130 executed a port I/O instruction which could not be satisfied by kvm.
3131 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
3132 where kvm expects application code to place the data for the next
3133 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
3135 /* KVM_EXIT_DEBUG */
3137 struct kvm_debug_exit_arch arch;
3140 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event
3141 for which architecture specific information is returned.
3151 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
3152 executed a memory-mapped I/O instruction which could not be satisfied
3153 by kvm. The 'data' member contains the written data if 'is_write' is
3154 true, and should be filled by application code otherwise.
3156 The 'data' member contains, in its first 'len' bytes, the value as it would
3157 appear if the VCPU performed a load or store of the appropriate width directly
3160 NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and
3161 KVM_EXIT_EPR the corresponding
3162 operations are complete (and guest state is consistent) only after userspace
3163 has re-entered the kernel with KVM_RUN. The kernel side will first finish
3164 incomplete operations and then check for pending signals. Userspace
3165 can re-enter the guest with an unmasked signal pending to complete
3168 /* KVM_EXIT_HYPERCALL */
3177 Unused. This was once used for 'hypercall to userspace'. To implement
3178 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390).
3179 Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
3181 /* KVM_EXIT_TPR_ACCESS */
3188 To be documented (KVM_TPR_ACCESS_REPORTING).
3190 /* KVM_EXIT_S390_SIEIC */
3193 __u64 mask; /* psw upper half */
3194 __u64 addr; /* psw lower half */
3201 /* KVM_EXIT_S390_RESET */
3202 #define KVM_S390_RESET_POR 1
3203 #define KVM_S390_RESET_CLEAR 2
3204 #define KVM_S390_RESET_SUBSYSTEM 4
3205 #define KVM_S390_RESET_CPU_INIT 8
3206 #define KVM_S390_RESET_IPL 16
3207 __u64 s390_reset_flags;
3211 /* KVM_EXIT_S390_UCONTROL */
3213 __u64 trans_exc_code;
3217 s390 specific. A page fault has occurred for a user controlled virtual
3218 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be
3219 resolved by the kernel.
3220 The program code and the translation exception code that were placed
3221 in the cpu's lowcore are presented here as defined by the z Architecture
3222 Principles of Operation Book in the Chapter for Dynamic Address Translation
3232 Deprecated - was used for 440 KVM.
3239 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
3240 hypercalls and exit with this exit struct that contains all the guest gprs.
3242 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
3243 Userspace can now handle the hypercall and when it's done modify the gprs as
3244 necessary. Upon guest entry all guest GPRs will then be replaced by the values
3247 /* KVM_EXIT_PAPR_HCALL */
3254 This is used on 64-bit PowerPC when emulating a pSeries partition,
3255 e.g. with the 'pseries' machine type in qemu. It occurs when the
3256 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
3257 contains the hypercall number (from the guest R3), and 'args' contains
3258 the arguments (from the guest R4 - R12). Userspace should put the
3259 return code in 'ret' and any extra returned values in args[].
3260 The possible hypercalls are defined in the Power Architecture Platform
3261 Requirements (PAPR) document available from www.power.org (free
3262 developer registration required to access it).
3264 /* KVM_EXIT_S390_TSCH */
3266 __u16 subchannel_id;
3267 __u16 subchannel_nr;
3274 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
3275 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
3276 interrupt for the target subchannel has been dequeued and subchannel_id,
3277 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
3278 interrupt. ipb is needed for instruction parameter decoding.
3285 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
3286 interrupt acknowledge path to the core. When the core successfully
3287 delivers an interrupt, it automatically populates the EPR register with
3288 the interrupt vector number and acknowledges the interrupt inside
3289 the interrupt controller.
3291 In case the interrupt controller lives in user space, we need to do
3292 the interrupt acknowledge cycle through it to fetch the next to be
3293 delivered interrupt vector using this exit.
3295 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
3296 external interrupt has just been delivered into the guest. User space
3297 should put the acknowledged interrupt vector into the 'epr' field.
3299 /* KVM_EXIT_SYSTEM_EVENT */
3301 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
3302 #define KVM_SYSTEM_EVENT_RESET 2
3303 #define KVM_SYSTEM_EVENT_CRASH 3
3308 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
3309 a system-level event using some architecture specific mechanism (hypercall
3310 or some special instruction). In case of ARM/ARM64, this is triggered using
3311 HVC instruction based PSCI call from the vcpu. The 'type' field describes
3312 the system-level event type. The 'flags' field describes architecture
3313 specific flags for the system-level event.
3315 Valid values for 'type' are:
3316 KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
3317 VM. Userspace is not obliged to honour this, and if it does honour
3318 this does not need to destroy the VM synchronously (ie it may call
3319 KVM_RUN again before shutdown finally occurs).
3320 KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
3321 As with SHUTDOWN, userspace can choose to ignore the request, or
3322 to schedule the reset to occur in the future and may call KVM_RUN again.
3323 KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest
3324 has requested a crash condition maintenance. Userspace can choose
3325 to ignore the request, or to gather VM memory core dump and/or
3326 reset/shutdown of the VM.
3328 /* KVM_EXIT_IOAPIC_EOI */
3333 Indicates that the VCPU's in-kernel local APIC received an EOI for a
3334 level-triggered IOAPIC interrupt. This exit only triggers when the
3335 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled);
3336 the userspace IOAPIC should process the EOI and retrigger the interrupt if
3337 it is still asserted. Vector is the LAPIC interrupt vector for which the
3340 struct kvm_hyperv_exit {
3341 #define KVM_EXIT_HYPERV_SYNIC 1
3352 /* KVM_EXIT_HYPERV */
3353 struct kvm_hyperv_exit hyperv;
3354 Indicates that the VCPU exits into userspace to process some tasks
3355 related to Hyper-V emulation.
3356 Valid values for 'type' are:
3357 KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about
3358 Hyper-V SynIC state change. Notification is used to remap SynIC
3359 event/message pages and to enable/disable SynIC messages/events processing
3362 /* Fix the size of the union. */
3367 * shared registers between kvm and userspace.
3368 * kvm_valid_regs specifies the register classes set by the host
3369 * kvm_dirty_regs specified the register classes dirtied by userspace
3370 * struct kvm_sync_regs is architecture specific, as well as the
3371 * bits for kvm_valid_regs and kvm_dirty_regs
3373 __u64 kvm_valid_regs;
3374 __u64 kvm_dirty_regs;
3376 struct kvm_sync_regs regs;
3380 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
3381 certain guest registers without having to call SET/GET_*REGS. Thus we can
3382 avoid some system call overhead if userspace has to handle the exit.
3383 Userspace can query the validity of the structure by checking
3384 kvm_valid_regs for specific bits. These bits are architecture specific
3385 and usually define the validity of a groups of registers. (e.g. one bit
3386 for general purpose registers)
3388 Please note that the kernel is allowed to use the kvm_run structure as the
3389 primary storage for certain register types. Therefore, the kernel may use the
3390 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
3396 6. Capabilities that can be enabled on vCPUs
3397 --------------------------------------------
3399 There are certain capabilities that change the behavior of the virtual CPU or
3400 the virtual machine when enabled. To enable them, please see section 4.37.
3401 Below you can find a list of capabilities and what their effect on the vCPU or
3402 the virtual machine is when enabling them.
3404 The following information is provided along with the description:
3406 Architectures: which instruction set architectures provide this ioctl.
3407 x86 includes both i386 and x86_64.
3409 Target: whether this is a per-vcpu or per-vm capability.
3411 Parameters: what parameters are accepted by the capability.
3413 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3414 are not detailed, but errors with specific meanings are.
3422 Returns: 0 on success; -1 on error
3424 This capability enables interception of OSI hypercalls that otherwise would
3425 be treated as normal system calls to be injected into the guest. OSI hypercalls
3426 were invented by Mac-on-Linux to have a standardized communication mechanism
3427 between the guest and the host.
3429 When this capability is enabled, KVM_EXIT_OSI can occur.
3432 6.2 KVM_CAP_PPC_PAPR
3437 Returns: 0 on success; -1 on error
3439 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
3440 done using the hypercall instruction "sc 1".
3442 It also sets the guest privilege level to "supervisor" mode. Usually the guest
3443 runs in "hypervisor" privilege mode with a few missing features.
3445 In addition to the above, it changes the semantics of SDR1. In this mode, the
3446 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
3447 HTAB invisible to the guest.
3449 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
3456 Parameters: args[0] is the address of a struct kvm_config_tlb
3457 Returns: 0 on success; -1 on error
3459 struct kvm_config_tlb {
3466 Configures the virtual CPU's TLB array, establishing a shared memory area
3467 between userspace and KVM. The "params" and "array" fields are userspace
3468 addresses of mmu-type-specific data structures. The "array_len" field is an
3469 safety mechanism, and should be set to the size in bytes of the memory that
3470 userspace has reserved for the array. It must be at least the size dictated
3471 by "mmu_type" and "params".
3473 While KVM_RUN is active, the shared region is under control of KVM. Its
3474 contents are undefined, and any modification by userspace results in
3475 boundedly undefined behavior.
3477 On return from KVM_RUN, the shared region will reflect the current state of
3478 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
3479 to tell KVM which entries have been changed, prior to calling KVM_RUN again
3482 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
3483 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
3484 - The "array" field points to an array of type "struct
3485 kvm_book3e_206_tlb_entry".
3486 - The array consists of all entries in the first TLB, followed by all
3487 entries in the second TLB.
3488 - Within a TLB, entries are ordered first by increasing set number. Within a
3489 set, entries are ordered by way (increasing ESEL).
3490 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
3491 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
3492 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
3493 hardware ignores this value for TLB0.
3495 6.4 KVM_CAP_S390_CSS_SUPPORT
3500 Returns: 0 on success; -1 on error
3502 This capability enables support for handling of channel I/O instructions.
3504 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
3505 handled in-kernel, while the other I/O instructions are passed to userspace.
3507 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
3508 SUBCHANNEL intercepts.
3510 Note that even though this capability is enabled per-vcpu, the complete
3511 virtual machine is affected.
3517 Parameters: args[0] defines whether the proxy facility is active
3518 Returns: 0 on success; -1 on error
3520 This capability enables or disables the delivery of interrupts through the
3521 external proxy facility.
3523 When enabled (args[0] != 0), every time the guest gets an external interrupt
3524 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
3525 to receive the topmost interrupt vector.
3527 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
3529 When this capability is enabled, KVM_EXIT_EPR can occur.
3531 6.6 KVM_CAP_IRQ_MPIC
3534 Parameters: args[0] is the MPIC device fd
3535 args[1] is the MPIC CPU number for this vcpu
3537 This capability connects the vcpu to an in-kernel MPIC device.
3539 6.7 KVM_CAP_IRQ_XICS
3543 Parameters: args[0] is the XICS device fd
3544 args[1] is the XICS CPU number (server ID) for this vcpu
3546 This capability connects the vcpu to an in-kernel XICS device.
3548 6.8 KVM_CAP_S390_IRQCHIP
3554 This capability enables the in-kernel irqchip for s390. Please refer to
3555 "4.24 KVM_CREATE_IRQCHIP" for details.
3557 6.9 KVM_CAP_MIPS_FPU
3561 Parameters: args[0] is reserved for future use (should be 0).
3563 This capability allows the use of the host Floating Point Unit by the guest. It
3564 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is
3565 done the KVM_REG_MIPS_FPR_* and KVM_REG_MIPS_FCR_* registers can be accessed
3566 (depending on the current guest FPU register mode), and the Status.FR,
3567 Config5.FRE bits are accessible via the KVM API and also from the guest,
3568 depending on them being supported by the FPU.
3570 6.10 KVM_CAP_MIPS_MSA
3574 Parameters: args[0] is reserved for future use (should be 0).
3576 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest.
3577 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest.
3578 Once this is done the KVM_REG_MIPS_VEC_* and KVM_REG_MIPS_MSA_* registers can be
3579 accessed, and the Config5.MSAEn bit is accessible via the KVM API and also from
3582 7. Capabilities that can be enabled on VMs
3583 ------------------------------------------
3585 There are certain capabilities that change the behavior of the virtual
3586 machine when enabled. To enable them, please see section 4.37. Below
3587 you can find a list of capabilities and what their effect on the VM
3588 is when enabling them.
3590 The following information is provided along with the description:
3592 Architectures: which instruction set architectures provide this ioctl.
3593 x86 includes both i386 and x86_64.
3595 Parameters: what parameters are accepted by the capability.
3597 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3598 are not detailed, but errors with specific meanings are.
3601 7.1 KVM_CAP_PPC_ENABLE_HCALL
3604 Parameters: args[0] is the sPAPR hcall number
3605 args[1] is 0 to disable, 1 to enable in-kernel handling
3607 This capability controls whether individual sPAPR hypercalls (hcalls)
3608 get handled by the kernel or not. Enabling or disabling in-kernel
3609 handling of an hcall is effective across the VM. On creation, an
3610 initial set of hcalls are enabled for in-kernel handling, which
3611 consists of those hcalls for which in-kernel handlers were implemented
3612 before this capability was implemented. If disabled, the kernel will
3613 not to attempt to handle the hcall, but will always exit to userspace
3614 to handle it. Note that it may not make sense to enable some and
3615 disable others of a group of related hcalls, but KVM does not prevent
3616 userspace from doing that.
3618 If the hcall number specified is not one that has an in-kernel
3619 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
3622 7.2 KVM_CAP_S390_USER_SIGP
3627 This capability controls which SIGP orders will be handled completely in user
3628 space. With this capability enabled, all fast orders will be handled completely
3634 - CONDITIONAL EMERGENCY SIGNAL
3636 All other orders will be handled completely in user space.
3638 Only privileged operation exceptions will be checked for in the kernel (or even
3639 in the hardware prior to interception). If this capability is not enabled, the
3640 old way of handling SIGP orders is used (partially in kernel and user space).
3642 7.3 KVM_CAP_S390_VECTOR_REGISTERS
3646 Returns: 0 on success, negative value on error
3648 Allows use of the vector registers introduced with z13 processor, and
3649 provides for the synchronization between host and user space. Will
3650 return -EINVAL if the machine does not support vectors.
3652 7.4 KVM_CAP_S390_USER_STSI
3657 This capability allows post-handlers for the STSI instruction. After
3658 initial handling in the kernel, KVM exits to user space with
3659 KVM_EXIT_S390_STSI to allow user space to insert further data.
3661 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
3672 @addr - guest address of STSI SYSIB
3676 @ar - access register number
3678 KVM handlers should exit to userspace with rc = -EREMOTE.
3680 7.5 KVM_CAP_SPLIT_IRQCHIP
3683 Parameters: args[0] - number of routes reserved for userspace IOAPICs
3684 Returns: 0 on success, -1 on error
3686 Create a local apic for each processor in the kernel. This can be used
3687 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the
3688 IOAPIC and PIC (and also the PIT, even though this has to be enabled
3691 This capability also enables in kernel routing of interrupt requests;
3692 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are
3693 used in the IRQ routing table. The first args[0] MSI routes are reserved
3694 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes,
3695 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace.
3697 Fails if VCPU has already been created, or if the irqchip is already in the
3698 kernel (i.e. KVM_CREATE_IRQCHIP has already been called).
3701 8. Other capabilities.
3702 ----------------------
3704 This section lists capabilities that give information about other
3705 features of the KVM implementation.
3707 8.1 KVM_CAP_PPC_HWRNG
3711 This capability, if KVM_CHECK_EXTENSION indicates that it is
3712 available, means that that the kernel has an implementation of the
3713 H_RANDOM hypercall backed by a hardware random-number generator.
3714 If present, the kernel H_RANDOM handler can be enabled for guest use
3715 with the KVM_CAP_PPC_ENABLE_HCALL capability.
3717 8.2 KVM_CAP_HYPERV_SYNIC
3720 This capability, if KVM_CHECK_EXTENSION indicates that it is
3721 available, means that that the kernel has an implementation of the
3722 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
3723 used to support Windows Hyper-V based guest paravirt drivers(VMBus).
3725 In order to use SynIC, it has to be activated by setting this
3726 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
3727 will disable the use of APIC hardware virtualization even if supported
3728 by the CPU, as it's incompatible with SynIC auto-EOI behavior.