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.
114 You probably want to use 0 as machine type.
116 In order to create user controlled virtual machines on S390, check
117 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
118 privileged user (CAP_SYS_ADMIN).
120 To use hardware assisted virtualization on MIPS (VZ ASE) rather than
121 the default trap & emulate implementation (which changes the virtual
122 memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
126 4.3 KVM_GET_MSR_INDEX_LIST
131 Parameters: struct kvm_msr_list (in/out)
132 Returns: 0 on success; -1 on error
134 E2BIG: the msr index list is to be to fit in the array specified by
137 struct kvm_msr_list {
138 __u32 nmsrs; /* number of msrs in entries */
142 This ioctl returns the guest msrs that are supported. The list varies
143 by kvm version and host processor, but does not change otherwise. The
144 user fills in the size of the indices array in nmsrs, and in return
145 kvm adjusts nmsrs to reflect the actual number of msrs and fills in
146 the indices array with their numbers.
148 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
149 not returned in the MSR list, as different vcpus can have a different number
150 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
153 4.4 KVM_CHECK_EXTENSION
155 Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
157 Type: system ioctl, vm ioctl
158 Parameters: extension identifier (KVM_CAP_*)
159 Returns: 0 if unsupported; 1 (or some other positive integer) if supported
161 The API allows the application to query about extensions to the core
162 kvm API. Userspace passes an extension identifier (an integer) and
163 receives an integer that describes the extension availability.
164 Generally 0 means no and 1 means yes, but some extensions may report
165 additional information in the integer return value.
167 Based on their initialization different VMs may have different capabilities.
168 It is thus encouraged to use the vm ioctl to query for capabilities (available
169 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
171 4.5 KVM_GET_VCPU_MMAP_SIZE
177 Returns: size of vcpu mmap area, in bytes
179 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
180 memory region. This ioctl returns the size of that region. See the
181 KVM_RUN documentation for details.
184 4.6 KVM_SET_MEMORY_REGION
189 Parameters: struct kvm_memory_region (in)
190 Returns: 0 on success, -1 on error
192 This ioctl is obsolete and has been removed.
200 Parameters: vcpu id (apic id on x86)
201 Returns: vcpu fd on success, -1 on error
203 This API adds a vcpu to a virtual machine. No more than max_vcpus may be added.
204 The vcpu id is an integer in the range [0, max_vcpu_id).
206 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
207 the KVM_CHECK_EXTENSION ioctl() at run-time.
208 The maximum possible value for max_vcpus can be retrieved using the
209 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
211 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
213 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
214 same as the value returned from KVM_CAP_NR_VCPUS.
216 The maximum possible value for max_vcpu_id can be retrieved using the
217 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time.
219 If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id
220 is the same as the value returned from KVM_CAP_MAX_VCPUS.
222 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
223 threads in one or more virtual CPU cores. (This is because the
224 hardware requires all the hardware threads in a CPU core to be in the
225 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
226 of vcpus per virtual core (vcore). The vcore id is obtained by
227 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
228 given vcore will always be in the same physical core as each other
229 (though that might be a different physical core from time to time).
230 Userspace can control the threading (SMT) mode of the guest by its
231 allocation of vcpu ids. For example, if userspace wants
232 single-threaded guest vcpus, it should make all vcpu ids be a multiple
233 of the number of vcpus per vcore.
235 For virtual cpus that have been created with S390 user controlled virtual
236 machines, the resulting vcpu fd can be memory mapped at page offset
237 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
238 cpu's hardware control block.
241 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
246 Parameters: struct kvm_dirty_log (in/out)
247 Returns: 0 on success, -1 on error
249 /* for KVM_GET_DIRTY_LOG */
250 struct kvm_dirty_log {
254 void __user *dirty_bitmap; /* one bit per page */
259 Given a memory slot, return a bitmap containing any pages dirtied
260 since the last call to this ioctl. Bit 0 is the first page in the
261 memory slot. Ensure the entire structure is cleared to avoid padding
264 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies
265 the address space for which you want to return the dirty bitmap.
266 They must be less than the value that KVM_CHECK_EXTENSION returns for
267 the KVM_CAP_MULTI_ADDRESS_SPACE capability.
270 4.9 KVM_SET_MEMORY_ALIAS
275 Parameters: struct kvm_memory_alias (in)
276 Returns: 0 (success), -1 (error)
278 This ioctl is obsolete and has been removed.
287 Returns: 0 on success, -1 on error
289 EINTR: an unmasked signal is pending
291 This ioctl is used to run a guest virtual cpu. While there are no
292 explicit parameters, there is an implicit parameter block that can be
293 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
294 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
295 kvm_run' (see below).
301 Architectures: all except ARM, arm64
303 Parameters: struct kvm_regs (out)
304 Returns: 0 on success, -1 on error
306 Reads the general purpose registers from the vcpu.
310 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
311 __u64 rax, rbx, rcx, rdx;
312 __u64 rsi, rdi, rsp, rbp;
313 __u64 r8, r9, r10, r11;
314 __u64 r12, r13, r14, r15;
320 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
331 Architectures: all except ARM, arm64
333 Parameters: struct kvm_regs (in)
334 Returns: 0 on success, -1 on error
336 Writes the general purpose registers into the vcpu.
338 See KVM_GET_REGS for the data structure.
344 Architectures: x86, ppc
346 Parameters: struct kvm_sregs (out)
347 Returns: 0 on success, -1 on error
349 Reads special registers from the vcpu.
353 struct kvm_segment cs, ds, es, fs, gs, ss;
354 struct kvm_segment tr, ldt;
355 struct kvm_dtable gdt, idt;
356 __u64 cr0, cr2, cr3, cr4, cr8;
359 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
362 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
364 interrupt_bitmap is a bitmap of pending external interrupts. At most
365 one bit may be set. This interrupt has been acknowledged by the APIC
366 but not yet injected into the cpu core.
372 Architectures: x86, ppc
374 Parameters: struct kvm_sregs (in)
375 Returns: 0 on success, -1 on error
377 Writes special registers into the vcpu. See KVM_GET_SREGS for the
386 Parameters: struct kvm_translation (in/out)
387 Returns: 0 on success, -1 on error
389 Translates a virtual address according to the vcpu's current address
392 struct kvm_translation {
394 __u64 linear_address;
397 __u64 physical_address;
408 Architectures: x86, ppc, mips
410 Parameters: struct kvm_interrupt (in)
411 Returns: 0 on success, negative on failure.
413 Queues a hardware interrupt vector to be injected.
415 /* for KVM_INTERRUPT */
416 struct kvm_interrupt {
423 Returns: 0 on success,
424 -EEXIST if an interrupt is already enqueued
425 -EINVAL the the irq number is invalid
426 -ENXIO if the PIC is in the kernel
427 -EFAULT if the pointer is invalid
429 Note 'irq' is an interrupt vector, not an interrupt pin or line. This
430 ioctl is useful if the in-kernel PIC is not used.
434 Queues an external interrupt to be injected. This ioctl is overleaded
435 with 3 different irq values:
439 This injects an edge type external interrupt into the guest once it's ready
440 to receive interrupts. When injected, the interrupt is done.
442 b) KVM_INTERRUPT_UNSET
444 This unsets any pending interrupt.
446 Only available with KVM_CAP_PPC_UNSET_IRQ.
448 c) KVM_INTERRUPT_SET_LEVEL
450 This injects a level type external interrupt into the guest context. The
451 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
454 Only available with KVM_CAP_PPC_IRQ_LEVEL.
456 Note that any value for 'irq' other than the ones stated above is invalid
457 and incurs unexpected behavior.
461 Queues an external interrupt to be injected into the virtual CPU. A negative
462 interrupt number dequeues the interrupt.
473 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
481 Parameters: struct kvm_msrs (in/out)
482 Returns: 0 on success, -1 on error
484 Reads model-specific registers from the vcpu. Supported msr indices can
485 be obtained using KVM_GET_MSR_INDEX_LIST.
488 __u32 nmsrs; /* number of msrs in entries */
491 struct kvm_msr_entry entries[0];
494 struct kvm_msr_entry {
500 Application code should set the 'nmsrs' member (which indicates the
501 size of the entries array) and the 'index' member of each array entry.
502 kvm will fill in the 'data' member.
510 Parameters: struct kvm_msrs (in)
511 Returns: 0 on success, -1 on error
513 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
516 Application code should set the 'nmsrs' member (which indicates the
517 size of the entries array), and the 'index' and 'data' members of each
526 Parameters: struct kvm_cpuid (in)
527 Returns: 0 on success, -1 on error
529 Defines the vcpu responses to the cpuid instruction. Applications
530 should use the KVM_SET_CPUID2 ioctl if available.
533 struct kvm_cpuid_entry {
542 /* for KVM_SET_CPUID */
546 struct kvm_cpuid_entry entries[0];
550 4.21 KVM_SET_SIGNAL_MASK
555 Parameters: struct kvm_signal_mask (in)
556 Returns: 0 on success, -1 on error
558 Defines which signals are blocked during execution of KVM_RUN. This
559 signal mask temporarily overrides the threads signal mask. Any
560 unblocked signal received (except SIGKILL and SIGSTOP, which retain
561 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
563 Note the signal will only be delivered if not blocked by the original
566 /* for KVM_SET_SIGNAL_MASK */
567 struct kvm_signal_mask {
578 Parameters: struct kvm_fpu (out)
579 Returns: 0 on success, -1 on error
581 Reads the floating point state from the vcpu.
583 /* for KVM_GET_FPU and KVM_SET_FPU */
588 __u8 ftwx; /* in fxsave format */
604 Parameters: struct kvm_fpu (in)
605 Returns: 0 on success, -1 on error
607 Writes the floating point state to the vcpu.
609 /* for KVM_GET_FPU and KVM_SET_FPU */
614 __u8 ftwx; /* in fxsave format */
625 4.24 KVM_CREATE_IRQCHIP
627 Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
628 Architectures: x86, ARM, arm64, s390
631 Returns: 0 on success, -1 on error
633 Creates an interrupt controller model in the kernel.
634 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
635 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
636 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
637 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of
638 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
639 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
640 On s390, a dummy irq routing table is created.
642 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
643 before KVM_CREATE_IRQCHIP can be used.
648 Capability: KVM_CAP_IRQCHIP
649 Architectures: x86, arm, arm64
651 Parameters: struct kvm_irq_level
652 Returns: 0 on success, -1 on error
654 Sets the level of a GSI input to the interrupt controller model in the kernel.
655 On some architectures it is required that an interrupt controller model has
656 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
657 interrupts require the level to be set to 1 and then back to 0.
659 On real hardware, interrupt pins can be active-low or active-high. This
660 does not matter for the level field of struct kvm_irq_level: 1 always
661 means active (asserted), 0 means inactive (deasserted).
663 x86 allows the operating system to program the interrupt polarity
664 (active-low/active-high) for level-triggered interrupts, and KVM used
665 to consider the polarity. However, due to bitrot in the handling of
666 active-low interrupts, the above convention is now valid on x86 too.
667 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
668 should not present interrupts to the guest as active-low unless this
669 capability is present (or unless it is not using the in-kernel irqchip,
673 ARM/arm64 can signal an interrupt either at the CPU level, or at the
674 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
675 use PPIs designated for specific cpus. The irq field is interpreted
678 Â bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 |
679 field: | irq_type | vcpu_index | irq_id |
681 The irq_type field has the following values:
682 - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
683 - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
684 (the vcpu_index field is ignored)
685 - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
687 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
689 In both cases, level is used to assert/deassert the line.
691 struct kvm_irq_level {
694 __s32 status; /* not used for KVM_IRQ_LEVEL */
696 __u32 level; /* 0 or 1 */
702 Capability: KVM_CAP_IRQCHIP
705 Parameters: struct kvm_irqchip (in/out)
706 Returns: 0 on success, -1 on error
708 Reads the state of a kernel interrupt controller created with
709 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
712 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
715 char dummy[512]; /* reserving space */
716 struct kvm_pic_state pic;
717 struct kvm_ioapic_state ioapic;
724 Capability: KVM_CAP_IRQCHIP
727 Parameters: struct kvm_irqchip (in)
728 Returns: 0 on success, -1 on error
730 Sets the state of a kernel interrupt controller created with
731 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
734 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
737 char dummy[512]; /* reserving space */
738 struct kvm_pic_state pic;
739 struct kvm_ioapic_state ioapic;
744 4.28 KVM_XEN_HVM_CONFIG
746 Capability: KVM_CAP_XEN_HVM
749 Parameters: struct kvm_xen_hvm_config (in)
750 Returns: 0 on success, -1 on error
752 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
753 page, and provides the starting address and size of the hypercall
754 blobs in userspace. When the guest writes the MSR, kvm copies one
755 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
758 struct kvm_xen_hvm_config {
771 Capability: KVM_CAP_ADJUST_CLOCK
774 Parameters: struct kvm_clock_data (out)
775 Returns: 0 on success, -1 on error
777 Gets the current timestamp of kvmclock as seen by the current guest. In
778 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
781 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the
782 set of bits that KVM can return in struct kvm_clock_data's flag member.
784 The only flag defined now is KVM_CLOCK_TSC_STABLE. If set, the returned
785 value is the exact kvmclock value seen by all VCPUs at the instant
786 when KVM_GET_CLOCK was called. If clear, the returned value is simply
787 CLOCK_MONOTONIC plus a constant offset; the offset can be modified
788 with KVM_SET_CLOCK. KVM will try to make all VCPUs follow this clock,
789 but the exact value read by each VCPU could differ, because the host
792 struct kvm_clock_data {
793 __u64 clock; /* kvmclock current value */
801 Capability: KVM_CAP_ADJUST_CLOCK
804 Parameters: struct kvm_clock_data (in)
805 Returns: 0 on success, -1 on error
807 Sets the current timestamp of kvmclock to the value specified in its parameter.
808 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
811 struct kvm_clock_data {
812 __u64 clock; /* kvmclock current value */
818 4.31 KVM_GET_VCPU_EVENTS
820 Capability: KVM_CAP_VCPU_EVENTS
821 Extended by: KVM_CAP_INTR_SHADOW
824 Parameters: struct kvm_vcpu_event (out)
825 Returns: 0 on success, -1 on error
827 Gets currently pending exceptions, interrupts, and NMIs as well as related
830 struct kvm_vcpu_events {
860 Only two fields are defined in the flags field:
862 - KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that
863 interrupt.shadow contains a valid state.
865 - KVM_VCPUEVENT_VALID_SMM may be set in the flags field to signal that
866 smi contains a valid state.
868 4.32 KVM_SET_VCPU_EVENTS
870 Capability: KVM_CAP_VCPU_EVENTS
871 Extended by: KVM_CAP_INTR_SHADOW
874 Parameters: struct kvm_vcpu_event (in)
875 Returns: 0 on success, -1 on error
877 Set pending exceptions, interrupts, and NMIs as well as related states of the
880 See KVM_GET_VCPU_EVENTS for the data structure.
882 Fields that may be modified asynchronously by running VCPUs can be excluded
883 from the update. These fields are nmi.pending, sipi_vector, smi.smm,
884 smi.pending. Keep the corresponding bits in the flags field cleared to
885 suppress overwriting the current in-kernel state. The bits are:
887 KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel
888 KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector
889 KVM_VCPUEVENT_VALID_SMM - transfer the smi sub-struct.
891 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
892 the flags field to signal that interrupt.shadow contains a valid state and
893 shall be written into the VCPU.
895 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available.
898 4.33 KVM_GET_DEBUGREGS
900 Capability: KVM_CAP_DEBUGREGS
903 Parameters: struct kvm_debugregs (out)
904 Returns: 0 on success, -1 on error
906 Reads debug registers from the vcpu.
908 struct kvm_debugregs {
917 4.34 KVM_SET_DEBUGREGS
919 Capability: KVM_CAP_DEBUGREGS
922 Parameters: struct kvm_debugregs (in)
923 Returns: 0 on success, -1 on error
925 Writes debug registers into the vcpu.
927 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
928 yet and must be cleared on entry.
931 4.35 KVM_SET_USER_MEMORY_REGION
933 Capability: KVM_CAP_USER_MEM
936 Parameters: struct kvm_userspace_memory_region (in)
937 Returns: 0 on success, -1 on error
939 struct kvm_userspace_memory_region {
942 __u64 guest_phys_addr;
943 __u64 memory_size; /* bytes */
944 __u64 userspace_addr; /* start of the userspace allocated memory */
947 /* for kvm_memory_region::flags */
948 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
949 #define KVM_MEM_READONLY (1UL << 1)
951 This ioctl allows the user to create or modify a guest physical memory
952 slot. When changing an existing slot, it may be moved in the guest
953 physical memory space, or its flags may be modified. It may not be
954 resized. Slots may not overlap in guest physical address space.
955 Bits 0-15 of "slot" specifies the slot id and this value should be
956 less than the maximum number of user memory slots supported per VM.
957 The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS,
958 if this capability is supported by the architecture.
960 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot"
961 specifies the address space which is being modified. They must be
962 less than the value that KVM_CHECK_EXTENSION returns for the
963 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces
964 are unrelated; the restriction on overlapping slots only applies within
967 Memory for the region is taken starting at the address denoted by the
968 field userspace_addr, which must point at user addressable memory for
969 the entire memory slot size. Any object may back this memory, including
970 anonymous memory, ordinary files, and hugetlbfs.
972 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
973 be identical. This allows large pages in the guest to be backed by large
976 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
977 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
978 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
979 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
980 to make a new slot read-only. In this case, writes to this memory will be
981 posted to userspace as KVM_EXIT_MMIO exits.
983 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
984 the memory region are automatically reflected into the guest. For example, an
985 mmap() that affects the region will be made visible immediately. Another
986 example is madvise(MADV_DROP).
988 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl.
989 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory
990 allocation and is deprecated.
993 4.36 KVM_SET_TSS_ADDR
995 Capability: KVM_CAP_SET_TSS_ADDR
998 Parameters: unsigned long tss_address (in)
999 Returns: 0 on success, -1 on error
1001 This ioctl defines the physical address of a three-page region in the guest
1002 physical address space. The region must be within the first 4GB of the
1003 guest physical address space and must not conflict with any memory slot
1004 or any mmio address. The guest may malfunction if it accesses this memory
1007 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1008 because of a quirk in the virtualization implementation (see the internals
1009 documentation when it pops into existence).
1014 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM
1015 Architectures: x86 (only KVM_CAP_ENABLE_CAP_VM),
1016 mips (only KVM_CAP_ENABLE_CAP), ppc, s390
1017 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM)
1018 Parameters: struct kvm_enable_cap (in)
1019 Returns: 0 on success; -1 on error
1021 +Not all extensions are enabled by default. Using this ioctl the application
1022 can enable an extension, making it available to the guest.
1024 On systems that do not support this ioctl, it always fails. On systems that
1025 do support it, it only works for extensions that are supported for enablement.
1027 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
1030 struct kvm_enable_cap {
1034 The capability that is supposed to get enabled.
1038 A bitfield indicating future enhancements. Has to be 0 for now.
1042 Arguments for enabling a feature. If a feature needs initial values to
1043 function properly, this is the place to put them.
1048 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
1049 for vm-wide capabilities.
1051 4.38 KVM_GET_MP_STATE
1053 Capability: KVM_CAP_MP_STATE
1054 Architectures: x86, s390, arm, arm64
1056 Parameters: struct kvm_mp_state (out)
1057 Returns: 0 on success; -1 on error
1059 struct kvm_mp_state {
1063 Returns the vcpu's current "multiprocessing state" (though also valid on
1064 uniprocessor guests).
1066 Possible values are:
1068 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86,arm/arm64]
1069 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP)
1070 which has not yet received an INIT signal [x86]
1071 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is
1072 now ready for a SIPI [x86]
1073 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and
1074 is waiting for an interrupt [x86]
1075 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector
1076 accessible via KVM_GET_VCPU_EVENTS) [x86]
1077 - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390,arm/arm64]
1078 - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390]
1079 - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted)
1081 - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state
1084 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1085 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1086 these architectures.
1090 The only states that are valid are KVM_MP_STATE_STOPPED and
1091 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
1093 4.39 KVM_SET_MP_STATE
1095 Capability: KVM_CAP_MP_STATE
1096 Architectures: x86, s390, arm, arm64
1098 Parameters: struct kvm_mp_state (in)
1099 Returns: 0 on success; -1 on error
1101 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1104 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1105 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1106 these architectures.
1110 The only states that are valid are KVM_MP_STATE_STOPPED and
1111 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
1113 4.40 KVM_SET_IDENTITY_MAP_ADDR
1115 Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1118 Parameters: unsigned long identity (in)
1119 Returns: 0 on success, -1 on error
1121 This ioctl defines the physical address of a one-page region in the guest
1122 physical address space. The region must be within the first 4GB of the
1123 guest physical address space and must not conflict with any memory slot
1124 or any mmio address. The guest may malfunction if it accesses this memory
1127 Setting the address to 0 will result in resetting the address to its default
1130 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1131 because of a quirk in the virtualization implementation (see the internals
1132 documentation when it pops into existence).
1134 Fails if any VCPU has already been created.
1136 4.41 KVM_SET_BOOT_CPU_ID
1138 Capability: KVM_CAP_SET_BOOT_CPU_ID
1141 Parameters: unsigned long vcpu_id
1142 Returns: 0 on success, -1 on error
1144 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1145 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1151 Capability: KVM_CAP_XSAVE
1154 Parameters: struct kvm_xsave (out)
1155 Returns: 0 on success, -1 on error
1161 This ioctl would copy current vcpu's xsave struct to the userspace.
1166 Capability: KVM_CAP_XSAVE
1169 Parameters: struct kvm_xsave (in)
1170 Returns: 0 on success, -1 on error
1176 This ioctl would copy userspace's xsave struct to the kernel.
1181 Capability: KVM_CAP_XCRS
1184 Parameters: struct kvm_xcrs (out)
1185 Returns: 0 on success, -1 on error
1196 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1200 This ioctl would copy current vcpu's xcrs to the userspace.
1205 Capability: KVM_CAP_XCRS
1208 Parameters: struct kvm_xcrs (in)
1209 Returns: 0 on success, -1 on error
1220 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1224 This ioctl would set vcpu's xcr to the value userspace specified.
1227 4.46 KVM_GET_SUPPORTED_CPUID
1229 Capability: KVM_CAP_EXT_CPUID
1232 Parameters: struct kvm_cpuid2 (in/out)
1233 Returns: 0 on success, -1 on error
1238 struct kvm_cpuid_entry2 entries[0];
1241 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1242 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
1243 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
1245 struct kvm_cpuid_entry2 {
1256 This ioctl returns x86 cpuid features which are supported by both the hardware
1257 and kvm. Userspace can use the information returned by this ioctl to
1258 construct cpuid information (for KVM_SET_CPUID2) that is consistent with
1259 hardware, kernel, and userspace capabilities, and with user requirements (for
1260 example, the user may wish to constrain cpuid to emulate older hardware,
1261 or for feature consistency across a cluster).
1263 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1264 with the 'nent' field indicating the number of entries in the variable-size
1265 array 'entries'. If the number of entries is too low to describe the cpu
1266 capabilities, an error (E2BIG) is returned. If the number is too high,
1267 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1268 number is just right, the 'nent' field is adjusted to the number of valid
1269 entries in the 'entries' array, which is then filled.
1271 The entries returned are the host cpuid as returned by the cpuid instruction,
1272 with unknown or unsupported features masked out. Some features (for example,
1273 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1274 emulate them efficiently. The fields in each entry are defined as follows:
1276 function: the eax value used to obtain the entry
1277 index: the ecx value used to obtain the entry (for entries that are
1279 flags: an OR of zero or more of the following:
1280 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1281 if the index field is valid
1282 KVM_CPUID_FLAG_STATEFUL_FUNC:
1283 if cpuid for this function returns different values for successive
1284 invocations; there will be several entries with the same function,
1285 all with this flag set
1286 KVM_CPUID_FLAG_STATE_READ_NEXT:
1287 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
1288 the first entry to be read by a cpu
1289 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
1290 this function/index combination
1292 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1293 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1294 support. Instead it is reported via
1296 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1298 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1299 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1302 4.47 KVM_PPC_GET_PVINFO
1304 Capability: KVM_CAP_PPC_GET_PVINFO
1307 Parameters: struct kvm_ppc_pvinfo (out)
1308 Returns: 0 on success, !0 on error
1310 struct kvm_ppc_pvinfo {
1316 This ioctl fetches PV specific information that need to be passed to the guest
1317 using the device tree or other means from vm context.
1319 The hcall array defines 4 instructions that make up a hypercall.
1321 If any additional field gets added to this structure later on, a bit for that
1322 additional piece of information will be set in the flags bitmap.
1324 The flags bitmap is defined as:
1326 /* the host supports the ePAPR idle hcall
1327 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1329 4.52 KVM_SET_GSI_ROUTING
1331 Capability: KVM_CAP_IRQ_ROUTING
1332 Architectures: x86 s390 arm arm64
1334 Parameters: struct kvm_irq_routing (in)
1335 Returns: 0 on success, -1 on error
1337 Sets the GSI routing table entries, overwriting any previously set entries.
1339 On arm/arm64, GSI routing has the following limitation:
1340 - GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD.
1342 struct kvm_irq_routing {
1345 struct kvm_irq_routing_entry entries[0];
1348 No flags are specified so far, the corresponding field must be set to zero.
1350 struct kvm_irq_routing_entry {
1356 struct kvm_irq_routing_irqchip irqchip;
1357 struct kvm_irq_routing_msi msi;
1358 struct kvm_irq_routing_s390_adapter adapter;
1359 struct kvm_irq_routing_hv_sint hv_sint;
1364 /* gsi routing entry types */
1365 #define KVM_IRQ_ROUTING_IRQCHIP 1
1366 #define KVM_IRQ_ROUTING_MSI 2
1367 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1368 #define KVM_IRQ_ROUTING_HV_SINT 4
1371 - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry
1372 type, specifies that the devid field contains a valid value. The per-VM
1373 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
1374 the device ID. If this capability is not available, userspace should
1375 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
1378 struct kvm_irq_routing_irqchip {
1383 struct kvm_irq_routing_msi {
1393 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
1394 for the device that wrote the MSI message. For PCI, this is usually a
1395 BFD identifier in the lower 16 bits.
1397 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
1398 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
1399 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
1400 address_hi must be zero.
1402 struct kvm_irq_routing_s390_adapter {
1406 __u32 summary_offset;
1410 struct kvm_irq_routing_hv_sint {
1416 4.55 KVM_SET_TSC_KHZ
1418 Capability: KVM_CAP_TSC_CONTROL
1421 Parameters: virtual tsc_khz
1422 Returns: 0 on success, -1 on error
1424 Specifies the tsc frequency for the virtual machine. The unit of the
1428 4.56 KVM_GET_TSC_KHZ
1430 Capability: KVM_CAP_GET_TSC_KHZ
1434 Returns: virtual tsc-khz on success, negative value on error
1436 Returns the tsc frequency of the guest. The unit of the return value is
1437 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1443 Capability: KVM_CAP_IRQCHIP
1446 Parameters: struct kvm_lapic_state (out)
1447 Returns: 0 on success, -1 on error
1449 #define KVM_APIC_REG_SIZE 0x400
1450 struct kvm_lapic_state {
1451 char regs[KVM_APIC_REG_SIZE];
1454 Reads the Local APIC registers and copies them into the input argument. The
1455 data format and layout are the same as documented in the architecture manual.
1457 If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is
1458 enabled, then the format of APIC_ID register depends on the APIC mode
1459 (reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in
1460 the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID
1461 which is stored in bits 31-24 of the APIC register, or equivalently in
1462 byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then
1463 be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR.
1465 If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state
1466 always uses xAPIC format.
1471 Capability: KVM_CAP_IRQCHIP
1474 Parameters: struct kvm_lapic_state (in)
1475 Returns: 0 on success, -1 on error
1477 #define KVM_APIC_REG_SIZE 0x400
1478 struct kvm_lapic_state {
1479 char regs[KVM_APIC_REG_SIZE];
1482 Copies the input argument into the Local APIC registers. The data format
1483 and layout are the same as documented in the architecture manual.
1485 The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's
1486 regs field) depends on the state of the KVM_CAP_X2APIC_API capability.
1487 See the note in KVM_GET_LAPIC.
1492 Capability: KVM_CAP_IOEVENTFD
1495 Parameters: struct kvm_ioeventfd (in)
1496 Returns: 0 on success, !0 on error
1498 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
1499 within the guest. A guest write in the registered address will signal the
1500 provided event instead of triggering an exit.
1502 struct kvm_ioeventfd {
1504 __u64 addr; /* legal pio/mmio address */
1505 __u32 len; /* 0, 1, 2, 4, or 8 bytes */
1511 For the special case of virtio-ccw devices on s390, the ioevent is matched
1512 to a subchannel/virtqueue tuple instead.
1514 The following flags are defined:
1516 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
1517 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
1518 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
1519 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
1520 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
1522 If datamatch flag is set, the event will be signaled only if the written value
1523 to the registered address is equal to datamatch in struct kvm_ioeventfd.
1525 For virtio-ccw devices, addr contains the subchannel id and datamatch the
1528 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
1529 the kernel will ignore the length of guest write and may get a faster vmexit.
1530 The speedup may only apply to specific architectures, but the ioeventfd will
1535 Capability: KVM_CAP_SW_TLB
1538 Parameters: struct kvm_dirty_tlb (in)
1539 Returns: 0 on success, -1 on error
1541 struct kvm_dirty_tlb {
1546 This must be called whenever userspace has changed an entry in the shared
1547 TLB, prior to calling KVM_RUN on the associated vcpu.
1549 The "bitmap" field is the userspace address of an array. This array
1550 consists of a number of bits, equal to the total number of TLB entries as
1551 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
1552 nearest multiple of 64.
1554 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
1557 The array is little-endian: the bit 0 is the least significant bit of the
1558 first byte, bit 8 is the least significant bit of the second byte, etc.
1559 This avoids any complications with differing word sizes.
1561 The "num_dirty" field is a performance hint for KVM to determine whether it
1562 should skip processing the bitmap and just invalidate everything. It must
1563 be set to the number of set bits in the bitmap.
1566 4.62 KVM_CREATE_SPAPR_TCE
1568 Capability: KVM_CAP_SPAPR_TCE
1569 Architectures: powerpc
1571 Parameters: struct kvm_create_spapr_tce (in)
1572 Returns: file descriptor for manipulating the created TCE table
1574 This creates a virtual TCE (translation control entry) table, which
1575 is an IOMMU for PAPR-style virtual I/O. It is used to translate
1576 logical addresses used in virtual I/O into guest physical addresses,
1577 and provides a scatter/gather capability for PAPR virtual I/O.
1579 /* for KVM_CAP_SPAPR_TCE */
1580 struct kvm_create_spapr_tce {
1585 The liobn field gives the logical IO bus number for which to create a
1586 TCE table. The window_size field specifies the size of the DMA window
1587 which this TCE table will translate - the table will contain one 64
1588 bit TCE entry for every 4kiB of the DMA window.
1590 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
1591 table has been created using this ioctl(), the kernel will handle it
1592 in real mode, updating the TCE table. H_PUT_TCE calls for other
1593 liobns will cause a vm exit and must be handled by userspace.
1595 The return value is a file descriptor which can be passed to mmap(2)
1596 to map the created TCE table into userspace. This lets userspace read
1597 the entries written by kernel-handled H_PUT_TCE calls, and also lets
1598 userspace update the TCE table directly which is useful in some
1602 4.63 KVM_ALLOCATE_RMA
1604 Capability: KVM_CAP_PPC_RMA
1605 Architectures: powerpc
1607 Parameters: struct kvm_allocate_rma (out)
1608 Returns: file descriptor for mapping the allocated RMA
1610 This allocates a Real Mode Area (RMA) from the pool allocated at boot
1611 time by the kernel. An RMA is a physically-contiguous, aligned region
1612 of memory used on older POWER processors to provide the memory which
1613 will be accessed by real-mode (MMU off) accesses in a KVM guest.
1614 POWER processors support a set of sizes for the RMA that usually
1615 includes 64MB, 128MB, 256MB and some larger powers of two.
1617 /* for KVM_ALLOCATE_RMA */
1618 struct kvm_allocate_rma {
1622 The return value is a file descriptor which can be passed to mmap(2)
1623 to map the allocated RMA into userspace. The mapped area can then be
1624 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
1625 RMA for a virtual machine. The size of the RMA in bytes (which is
1626 fixed at host kernel boot time) is returned in the rma_size field of
1627 the argument structure.
1629 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
1630 is supported; 2 if the processor requires all virtual machines to have
1631 an RMA, or 1 if the processor can use an RMA but doesn't require it,
1632 because it supports the Virtual RMA (VRMA) facility.
1637 Capability: KVM_CAP_USER_NMI
1641 Returns: 0 on success, -1 on error
1643 Queues an NMI on the thread's vcpu. Note this is well defined only
1644 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
1645 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
1646 has been called, this interface is completely emulated within the kernel.
1648 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
1649 following algorithm:
1652 - read the local APIC's state (KVM_GET_LAPIC)
1653 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
1654 - if so, issue KVM_NMI
1657 Some guests configure the LINT1 NMI input to cause a panic, aiding in
1661 4.65 KVM_S390_UCAS_MAP
1663 Capability: KVM_CAP_S390_UCONTROL
1666 Parameters: struct kvm_s390_ucas_mapping (in)
1667 Returns: 0 in case of success
1669 The parameter is defined like this:
1670 struct kvm_s390_ucas_mapping {
1676 This ioctl maps the memory at "user_addr" with the length "length" to
1677 the vcpu's address space starting at "vcpu_addr". All parameters need to
1678 be aligned by 1 megabyte.
1681 4.66 KVM_S390_UCAS_UNMAP
1683 Capability: KVM_CAP_S390_UCONTROL
1686 Parameters: struct kvm_s390_ucas_mapping (in)
1687 Returns: 0 in case of success
1689 The parameter is defined like this:
1690 struct kvm_s390_ucas_mapping {
1696 This ioctl unmaps the memory in the vcpu's address space starting at
1697 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
1698 All parameters need to be aligned by 1 megabyte.
1701 4.67 KVM_S390_VCPU_FAULT
1703 Capability: KVM_CAP_S390_UCONTROL
1706 Parameters: vcpu absolute address (in)
1707 Returns: 0 in case of success
1709 This call creates a page table entry on the virtual cpu's address space
1710 (for user controlled virtual machines) or the virtual machine's address
1711 space (for regular virtual machines). This only works for minor faults,
1712 thus it's recommended to access subject memory page via the user page
1713 table upfront. This is useful to handle validity intercepts for user
1714 controlled virtual machines to fault in the virtual cpu's lowcore pages
1715 prior to calling the KVM_RUN ioctl.
1718 4.68 KVM_SET_ONE_REG
1720 Capability: KVM_CAP_ONE_REG
1723 Parameters: struct kvm_one_reg (in)
1724 Returns: 0 on success, negative value on failure
1726 struct kvm_one_reg {
1731 Using this ioctl, a single vcpu register can be set to a specific value
1732 defined by user space with the passed in struct kvm_one_reg, where id
1733 refers to the register identifier as described below and addr is a pointer
1734 to a variable with the respective size. There can be architecture agnostic
1735 and architecture specific registers. Each have their own range of operation
1736 and their own constants and width. To keep track of the implemented
1737 registers, find a list below:
1739 Arch | Register | Width (bits)
1741 PPC | KVM_REG_PPC_HIOR | 64
1742 PPC | KVM_REG_PPC_IAC1 | 64
1743 PPC | KVM_REG_PPC_IAC2 | 64
1744 PPC | KVM_REG_PPC_IAC3 | 64
1745 PPC | KVM_REG_PPC_IAC4 | 64
1746 PPC | KVM_REG_PPC_DAC1 | 64
1747 PPC | KVM_REG_PPC_DAC2 | 64
1748 PPC | KVM_REG_PPC_DABR | 64
1749 PPC | KVM_REG_PPC_DSCR | 64
1750 PPC | KVM_REG_PPC_PURR | 64
1751 PPC | KVM_REG_PPC_SPURR | 64
1752 PPC | KVM_REG_PPC_DAR | 64
1753 PPC | KVM_REG_PPC_DSISR | 32
1754 PPC | KVM_REG_PPC_AMR | 64
1755 PPC | KVM_REG_PPC_UAMOR | 64
1756 PPC | KVM_REG_PPC_MMCR0 | 64
1757 PPC | KVM_REG_PPC_MMCR1 | 64
1758 PPC | KVM_REG_PPC_MMCRA | 64
1759 PPC | KVM_REG_PPC_MMCR2 | 64
1760 PPC | KVM_REG_PPC_MMCRS | 64
1761 PPC | KVM_REG_PPC_SIAR | 64
1762 PPC | KVM_REG_PPC_SDAR | 64
1763 PPC | KVM_REG_PPC_SIER | 64
1764 PPC | KVM_REG_PPC_PMC1 | 32
1765 PPC | KVM_REG_PPC_PMC2 | 32
1766 PPC | KVM_REG_PPC_PMC3 | 32
1767 PPC | KVM_REG_PPC_PMC4 | 32
1768 PPC | KVM_REG_PPC_PMC5 | 32
1769 PPC | KVM_REG_PPC_PMC6 | 32
1770 PPC | KVM_REG_PPC_PMC7 | 32
1771 PPC | KVM_REG_PPC_PMC8 | 32
1772 PPC | KVM_REG_PPC_FPR0 | 64
1774 PPC | KVM_REG_PPC_FPR31 | 64
1775 PPC | KVM_REG_PPC_VR0 | 128
1777 PPC | KVM_REG_PPC_VR31 | 128
1778 PPC | KVM_REG_PPC_VSR0 | 128
1780 PPC | KVM_REG_PPC_VSR31 | 128
1781 PPC | KVM_REG_PPC_FPSCR | 64
1782 PPC | KVM_REG_PPC_VSCR | 32
1783 PPC | KVM_REG_PPC_VPA_ADDR | 64
1784 PPC | KVM_REG_PPC_VPA_SLB | 128
1785 PPC | KVM_REG_PPC_VPA_DTL | 128
1786 PPC | KVM_REG_PPC_EPCR | 32
1787 PPC | KVM_REG_PPC_EPR | 32
1788 PPC | KVM_REG_PPC_TCR | 32
1789 PPC | KVM_REG_PPC_TSR | 32
1790 PPC | KVM_REG_PPC_OR_TSR | 32
1791 PPC | KVM_REG_PPC_CLEAR_TSR | 32
1792 PPC | KVM_REG_PPC_MAS0 | 32
1793 PPC | KVM_REG_PPC_MAS1 | 32
1794 PPC | KVM_REG_PPC_MAS2 | 64
1795 PPC | KVM_REG_PPC_MAS7_3 | 64
1796 PPC | KVM_REG_PPC_MAS4 | 32
1797 PPC | KVM_REG_PPC_MAS6 | 32
1798 PPC | KVM_REG_PPC_MMUCFG | 32
1799 PPC | KVM_REG_PPC_TLB0CFG | 32
1800 PPC | KVM_REG_PPC_TLB1CFG | 32
1801 PPC | KVM_REG_PPC_TLB2CFG | 32
1802 PPC | KVM_REG_PPC_TLB3CFG | 32
1803 PPC | KVM_REG_PPC_TLB0PS | 32
1804 PPC | KVM_REG_PPC_TLB1PS | 32
1805 PPC | KVM_REG_PPC_TLB2PS | 32
1806 PPC | KVM_REG_PPC_TLB3PS | 32
1807 PPC | KVM_REG_PPC_EPTCFG | 32
1808 PPC | KVM_REG_PPC_ICP_STATE | 64
1809 PPC | KVM_REG_PPC_TB_OFFSET | 64
1810 PPC | KVM_REG_PPC_SPMC1 | 32
1811 PPC | KVM_REG_PPC_SPMC2 | 32
1812 PPC | KVM_REG_PPC_IAMR | 64
1813 PPC | KVM_REG_PPC_TFHAR | 64
1814 PPC | KVM_REG_PPC_TFIAR | 64
1815 PPC | KVM_REG_PPC_TEXASR | 64
1816 PPC | KVM_REG_PPC_FSCR | 64
1817 PPC | KVM_REG_PPC_PSPB | 32
1818 PPC | KVM_REG_PPC_EBBHR | 64
1819 PPC | KVM_REG_PPC_EBBRR | 64
1820 PPC | KVM_REG_PPC_BESCR | 64
1821 PPC | KVM_REG_PPC_TAR | 64
1822 PPC | KVM_REG_PPC_DPDES | 64
1823 PPC | KVM_REG_PPC_DAWR | 64
1824 PPC | KVM_REG_PPC_DAWRX | 64
1825 PPC | KVM_REG_PPC_CIABR | 64
1826 PPC | KVM_REG_PPC_IC | 64
1827 PPC | KVM_REG_PPC_VTB | 64
1828 PPC | KVM_REG_PPC_CSIGR | 64
1829 PPC | KVM_REG_PPC_TACR | 64
1830 PPC | KVM_REG_PPC_TCSCR | 64
1831 PPC | KVM_REG_PPC_PID | 64
1832 PPC | KVM_REG_PPC_ACOP | 64
1833 PPC | KVM_REG_PPC_VRSAVE | 32
1834 PPC | KVM_REG_PPC_LPCR | 32
1835 PPC | KVM_REG_PPC_LPCR_64 | 64
1836 PPC | KVM_REG_PPC_PPR | 64
1837 PPC | KVM_REG_PPC_ARCH_COMPAT | 32
1838 PPC | KVM_REG_PPC_DABRX | 32
1839 PPC | KVM_REG_PPC_WORT | 64
1840 PPC | KVM_REG_PPC_SPRG9 | 64
1841 PPC | KVM_REG_PPC_DBSR | 32
1842 PPC | KVM_REG_PPC_TIDR | 64
1843 PPC | KVM_REG_PPC_PSSCR | 64
1844 PPC | KVM_REG_PPC_TM_GPR0 | 64
1846 PPC | KVM_REG_PPC_TM_GPR31 | 64
1847 PPC | KVM_REG_PPC_TM_VSR0 | 128
1849 PPC | KVM_REG_PPC_TM_VSR63 | 128
1850 PPC | KVM_REG_PPC_TM_CR | 64
1851 PPC | KVM_REG_PPC_TM_LR | 64
1852 PPC | KVM_REG_PPC_TM_CTR | 64
1853 PPC | KVM_REG_PPC_TM_FPSCR | 64
1854 PPC | KVM_REG_PPC_TM_AMR | 64
1855 PPC | KVM_REG_PPC_TM_PPR | 64
1856 PPC | KVM_REG_PPC_TM_VRSAVE | 64
1857 PPC | KVM_REG_PPC_TM_VSCR | 32
1858 PPC | KVM_REG_PPC_TM_DSCR | 64
1859 PPC | KVM_REG_PPC_TM_TAR | 64
1860 PPC | KVM_REG_PPC_TM_XER | 64
1862 MIPS | KVM_REG_MIPS_R0 | 64
1864 MIPS | KVM_REG_MIPS_R31 | 64
1865 MIPS | KVM_REG_MIPS_HI | 64
1866 MIPS | KVM_REG_MIPS_LO | 64
1867 MIPS | KVM_REG_MIPS_PC | 64
1868 MIPS | KVM_REG_MIPS_CP0_INDEX | 32
1869 MIPS | KVM_REG_MIPS_CP0_ENTRYLO0 | 64
1870 MIPS | KVM_REG_MIPS_CP0_ENTRYLO1 | 64
1871 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64
1872 MIPS | KVM_REG_MIPS_CP0_CONTEXTCONFIG| 32
1873 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64
1874 MIPS | KVM_REG_MIPS_CP0_XCONTEXTCONFIG| 64
1875 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32
1876 MIPS | KVM_REG_MIPS_CP0_PAGEGRAIN | 32
1877 MIPS | KVM_REG_MIPS_CP0_SEGCTL0 | 64
1878 MIPS | KVM_REG_MIPS_CP0_SEGCTL1 | 64
1879 MIPS | KVM_REG_MIPS_CP0_SEGCTL2 | 64
1880 MIPS | KVM_REG_MIPS_CP0_PWBASE | 64
1881 MIPS | KVM_REG_MIPS_CP0_PWFIELD | 64
1882 MIPS | KVM_REG_MIPS_CP0_PWSIZE | 64
1883 MIPS | KVM_REG_MIPS_CP0_WIRED | 32
1884 MIPS | KVM_REG_MIPS_CP0_PWCTL | 32
1885 MIPS | KVM_REG_MIPS_CP0_HWRENA | 32
1886 MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64
1887 MIPS | KVM_REG_MIPS_CP0_BADINSTR | 32
1888 MIPS | KVM_REG_MIPS_CP0_BADINSTRP | 32
1889 MIPS | KVM_REG_MIPS_CP0_COUNT | 32
1890 MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64
1891 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32
1892 MIPS | KVM_REG_MIPS_CP0_STATUS | 32
1893 MIPS | KVM_REG_MIPS_CP0_INTCTL | 32
1894 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32
1895 MIPS | KVM_REG_MIPS_CP0_EPC | 64
1896 MIPS | KVM_REG_MIPS_CP0_PRID | 32
1897 MIPS | KVM_REG_MIPS_CP0_EBASE | 64
1898 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32
1899 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32
1900 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32
1901 MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32
1902 MIPS | KVM_REG_MIPS_CP0_CONFIG4 | 32
1903 MIPS | KVM_REG_MIPS_CP0_CONFIG5 | 32
1904 MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32
1905 MIPS | KVM_REG_MIPS_CP0_XCONTEXT | 64
1906 MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64
1907 MIPS | KVM_REG_MIPS_CP0_KSCRATCH1 | 64
1908 MIPS | KVM_REG_MIPS_CP0_KSCRATCH2 | 64
1909 MIPS | KVM_REG_MIPS_CP0_KSCRATCH3 | 64
1910 MIPS | KVM_REG_MIPS_CP0_KSCRATCH4 | 64
1911 MIPS | KVM_REG_MIPS_CP0_KSCRATCH5 | 64
1912 MIPS | KVM_REG_MIPS_CP0_KSCRATCH6 | 64
1913 MIPS | KVM_REG_MIPS_CP0_MAAR(0..63) | 64
1914 MIPS | KVM_REG_MIPS_COUNT_CTL | 64
1915 MIPS | KVM_REG_MIPS_COUNT_RESUME | 64
1916 MIPS | KVM_REG_MIPS_COUNT_HZ | 64
1917 MIPS | KVM_REG_MIPS_FPR_32(0..31) | 32
1918 MIPS | KVM_REG_MIPS_FPR_64(0..31) | 64
1919 MIPS | KVM_REG_MIPS_VEC_128(0..31) | 128
1920 MIPS | KVM_REG_MIPS_FCR_IR | 32
1921 MIPS | KVM_REG_MIPS_FCR_CSR | 32
1922 MIPS | KVM_REG_MIPS_MSA_IR | 32
1923 MIPS | KVM_REG_MIPS_MSA_CSR | 32
1925 ARM registers are mapped using the lower 32 bits. The upper 16 of that
1926 is the register group type, or coprocessor number:
1928 ARM core registers have the following id bit patterns:
1929 0x4020 0000 0010 <index into the kvm_regs struct:16>
1931 ARM 32-bit CP15 registers have the following id bit patterns:
1932 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
1934 ARM 64-bit CP15 registers have the following id bit patterns:
1935 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
1937 ARM CCSIDR registers are demultiplexed by CSSELR value:
1938 0x4020 0000 0011 00 <csselr:8>
1940 ARM 32-bit VFP control registers have the following id bit patterns:
1941 0x4020 0000 0012 1 <regno:12>
1943 ARM 64-bit FP registers have the following id bit patterns:
1944 0x4030 0000 0012 0 <regno:12>
1947 arm64 registers are mapped using the lower 32 bits. The upper 16 of
1948 that is the register group type, or coprocessor number:
1950 arm64 core/FP-SIMD registers have the following id bit patterns. Note
1951 that the size of the access is variable, as the kvm_regs structure
1952 contains elements ranging from 32 to 128 bits. The index is a 32bit
1953 value in the kvm_regs structure seen as a 32bit array.
1954 0x60x0 0000 0010 <index into the kvm_regs struct:16>
1956 arm64 CCSIDR registers are demultiplexed by CSSELR value:
1957 0x6020 0000 0011 00 <csselr:8>
1959 arm64 system registers have the following id bit patterns:
1960 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
1963 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
1964 the register group type:
1966 MIPS core registers (see above) have the following id bit patterns:
1967 0x7030 0000 0000 <reg:16>
1969 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
1970 patterns depending on whether they're 32-bit or 64-bit registers:
1971 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
1972 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
1974 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64
1975 versions of the EntryLo registers regardless of the word size of the host
1976 hardware, host kernel, guest, and whether XPA is present in the guest, i.e.
1977 with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and
1978 the PFNX field starting at bit 30.
1980 MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit
1982 0x7030 0000 0001 01 <reg:8>
1984 MIPS KVM control registers (see above) have the following id bit patterns:
1985 0x7030 0000 0002 <reg:16>
1987 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
1988 id bit patterns depending on the size of the register being accessed. They are
1989 always accessed according to the current guest FPU mode (Status.FR and
1990 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
1991 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector
1992 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they
1993 overlap the FPU registers:
1994 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
1995 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
1996 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers)
1998 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
1999 following id bit patterns:
2000 0x7020 0000 0003 01 <0:3> <reg:5>
2002 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the
2003 following id bit patterns:
2004 0x7020 0000 0003 02 <0:3> <reg:5>
2007 4.69 KVM_GET_ONE_REG
2009 Capability: KVM_CAP_ONE_REG
2012 Parameters: struct kvm_one_reg (in and out)
2013 Returns: 0 on success, negative value on failure
2015 This ioctl allows to receive the value of a single register implemented
2016 in a vcpu. The register to read is indicated by the "id" field of the
2017 kvm_one_reg struct passed in. On success, the register value can be found
2018 at the memory location pointed to by "addr".
2020 The list of registers accessible using this interface is identical to the
2024 4.70 KVM_KVMCLOCK_CTRL
2026 Capability: KVM_CAP_KVMCLOCK_CTRL
2027 Architectures: Any that implement pvclocks (currently x86 only)
2030 Returns: 0 on success, -1 on error
2032 This signals to the host kernel that the specified guest is being paused by
2033 userspace. The host will set a flag in the pvclock structure that is checked
2034 from the soft lockup watchdog. The flag is part of the pvclock structure that
2035 is shared between guest and host, specifically the second bit of the flags
2036 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2037 the host and read/cleared exclusively by the guest. The guest operation of
2038 checking and clearing the flag must an atomic operation so
2039 load-link/store-conditional, or equivalent must be used. There are two cases
2040 where the guest will clear the flag: when the soft lockup watchdog timer resets
2041 itself or when a soft lockup is detected. This ioctl can be called any time
2042 after pausing the vcpu, but before it is resumed.
2047 Capability: KVM_CAP_SIGNAL_MSI
2048 Architectures: x86 arm arm64
2050 Parameters: struct kvm_msi (in)
2051 Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2053 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2065 flags: KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM
2066 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
2067 the device ID. If this capability is not available, userspace
2068 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
2070 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
2071 for the device that wrote the MSI message. For PCI, this is usually a
2072 BFD identifier in the lower 16 bits.
2074 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
2075 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
2076 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
2077 address_hi must be zero.
2080 4.71 KVM_CREATE_PIT2
2082 Capability: KVM_CAP_PIT2
2085 Parameters: struct kvm_pit_config (in)
2086 Returns: 0 on success, -1 on error
2088 Creates an in-kernel device model for the i8254 PIT. This call is only valid
2089 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
2090 parameters have to be passed:
2092 struct kvm_pit_config {
2099 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
2101 PIT timer interrupts may use a per-VM kernel thread for injection. If it
2102 exists, this thread will have a name of the following pattern:
2104 kvm-pit/<owner-process-pid>
2106 When running a guest with elevated priorities, the scheduling parameters of
2107 this thread may have to be adjusted accordingly.
2109 This IOCTL replaces the obsolete KVM_CREATE_PIT.
2114 Capability: KVM_CAP_PIT_STATE2
2117 Parameters: struct kvm_pit_state2 (out)
2118 Returns: 0 on success, -1 on error
2120 Retrieves the state of the in-kernel PIT model. Only valid after
2121 KVM_CREATE_PIT2. The state is returned in the following structure:
2123 struct kvm_pit_state2 {
2124 struct kvm_pit_channel_state channels[3];
2131 /* disable PIT in HPET legacy mode */
2132 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
2134 This IOCTL replaces the obsolete KVM_GET_PIT.
2139 Capability: KVM_CAP_PIT_STATE2
2142 Parameters: struct kvm_pit_state2 (in)
2143 Returns: 0 on success, -1 on error
2145 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
2146 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
2148 This IOCTL replaces the obsolete KVM_SET_PIT.
2151 4.74 KVM_PPC_GET_SMMU_INFO
2153 Capability: KVM_CAP_PPC_GET_SMMU_INFO
2154 Architectures: powerpc
2157 Returns: 0 on success, -1 on error
2159 This populates and returns a structure describing the features of
2160 the "Server" class MMU emulation supported by KVM.
2161 This can in turn be used by userspace to generate the appropriate
2162 device-tree properties for the guest operating system.
2164 The structure contains some global information, followed by an
2165 array of supported segment page sizes:
2167 struct kvm_ppc_smmu_info {
2171 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
2174 The supported flags are:
2176 - KVM_PPC_PAGE_SIZES_REAL:
2177 When that flag is set, guest page sizes must "fit" the backing
2178 store page sizes. When not set, any page size in the list can
2179 be used regardless of how they are backed by userspace.
2181 - KVM_PPC_1T_SEGMENTS
2182 The emulated MMU supports 1T segments in addition to the
2185 The "slb_size" field indicates how many SLB entries are supported
2187 The "sps" array contains 8 entries indicating the supported base
2188 page sizes for a segment in increasing order. Each entry is defined
2191 struct kvm_ppc_one_seg_page_size {
2192 __u32 page_shift; /* Base page shift of segment (or 0) */
2193 __u32 slb_enc; /* SLB encoding for BookS */
2194 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
2197 An entry with a "page_shift" of 0 is unused. Because the array is
2198 organized in increasing order, a lookup can stop when encoutering
2201 The "slb_enc" field provides the encoding to use in the SLB for the
2202 page size. The bits are in positions such as the value can directly
2203 be OR'ed into the "vsid" argument of the slbmte instruction.
2205 The "enc" array is a list which for each of those segment base page
2206 size provides the list of supported actual page sizes (which can be
2207 only larger or equal to the base page size), along with the
2208 corresponding encoding in the hash PTE. Similarly, the array is
2209 8 entries sorted by increasing sizes and an entry with a "0" shift
2210 is an empty entry and a terminator:
2212 struct kvm_ppc_one_page_size {
2213 __u32 page_shift; /* Page shift (or 0) */
2214 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
2217 The "pte_enc" field provides a value that can OR'ed into the hash
2218 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
2219 into the hash PTE second double word).
2223 Capability: KVM_CAP_IRQFD
2224 Architectures: x86 s390 arm arm64
2226 Parameters: struct kvm_irqfd (in)
2227 Returns: 0 on success, -1 on error
2229 Allows setting an eventfd to directly trigger a guest interrupt.
2230 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
2231 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
2232 an event is triggered on the eventfd, an interrupt is injected into
2233 the guest using the specified gsi pin. The irqfd is removed using
2234 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
2237 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
2238 mechanism allowing emulation of level-triggered, irqfd-based
2239 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
2240 additional eventfd in the kvm_irqfd.resamplefd field. When operating
2241 in resample mode, posting of an interrupt through kvm_irq.fd asserts
2242 the specified gsi in the irqchip. When the irqchip is resampled, such
2243 as from an EOI, the gsi is de-asserted and the user is notified via
2244 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
2245 the interrupt if the device making use of it still requires service.
2246 Note that closing the resamplefd is not sufficient to disable the
2247 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
2248 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
2250 On arm/arm64, gsi routing being supported, the following can happen:
2251 - in case no routing entry is associated to this gsi, injection fails
2252 - in case the gsi is associated to an irqchip routing entry,
2253 irqchip.pin + 32 corresponds to the injected SPI ID.
2254 - in case the gsi is associated to an MSI routing entry, the MSI
2255 message and device ID are translated into an LPI (support restricted
2256 to GICv3 ITS in-kernel emulation).
2258 4.76 KVM_PPC_ALLOCATE_HTAB
2260 Capability: KVM_CAP_PPC_ALLOC_HTAB
2261 Architectures: powerpc
2263 Parameters: Pointer to u32 containing hash table order (in/out)
2264 Returns: 0 on success, -1 on error
2266 This requests the host kernel to allocate an MMU hash table for a
2267 guest using the PAPR paravirtualization interface. This only does
2268 anything if the kernel is configured to use the Book 3S HV style of
2269 virtualization. Otherwise the capability doesn't exist and the ioctl
2270 returns an ENOTTY error. The rest of this description assumes Book 3S
2273 There must be no vcpus running when this ioctl is called; if there
2274 are, it will do nothing and return an EBUSY error.
2276 The parameter is a pointer to a 32-bit unsigned integer variable
2277 containing the order (log base 2) of the desired size of the hash
2278 table, which must be between 18 and 46. On successful return from the
2279 ioctl, the value will not be changed by the kernel.
2281 If no hash table has been allocated when any vcpu is asked to run
2282 (with the KVM_RUN ioctl), the host kernel will allocate a
2283 default-sized hash table (16 MB).
2285 If this ioctl is called when a hash table has already been allocated,
2286 with a different order from the existing hash table, the existing hash
2287 table will be freed and a new one allocated. If this is ioctl is
2288 called when a hash table has already been allocated of the same order
2289 as specified, the kernel will clear out the existing hash table (zero
2290 all HPTEs). In either case, if the guest is using the virtualized
2291 real-mode area (VRMA) facility, the kernel will re-create the VMRA
2292 HPTEs on the next KVM_RUN of any vcpu.
2294 4.77 KVM_S390_INTERRUPT
2298 Type: vm ioctl, vcpu ioctl
2299 Parameters: struct kvm_s390_interrupt (in)
2300 Returns: 0 on success, -1 on error
2302 Allows to inject an interrupt to the guest. Interrupts can be floating
2303 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
2305 Interrupt parameters are passed via kvm_s390_interrupt:
2307 struct kvm_s390_interrupt {
2313 type can be one of the following:
2315 KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm
2316 KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm
2317 KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm
2318 KVM_S390_RESTART (vcpu) - restart
2319 KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt
2320 KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt
2321 KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt
2322 parameters in parm and parm64
2323 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm
2324 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm
2325 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm
2326 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
2327 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
2328 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
2329 interruption subclass)
2330 KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm,
2331 machine check interrupt code in parm64 (note that
2332 machine checks needing further payload are not
2333 supported by this ioctl)
2335 Note that the vcpu ioctl is asynchronous to vcpu execution.
2337 4.78 KVM_PPC_GET_HTAB_FD
2339 Capability: KVM_CAP_PPC_HTAB_FD
2340 Architectures: powerpc
2342 Parameters: Pointer to struct kvm_get_htab_fd (in)
2343 Returns: file descriptor number (>= 0) on success, -1 on error
2345 This returns a file descriptor that can be used either to read out the
2346 entries in the guest's hashed page table (HPT), or to write entries to
2347 initialize the HPT. The returned fd can only be written to if the
2348 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
2349 can only be read if that bit is clear. The argument struct looks like
2352 /* For KVM_PPC_GET_HTAB_FD */
2353 struct kvm_get_htab_fd {
2359 /* Values for kvm_get_htab_fd.flags */
2360 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
2361 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
2363 The `start_index' field gives the index in the HPT of the entry at
2364 which to start reading. It is ignored when writing.
2366 Reads on the fd will initially supply information about all
2367 "interesting" HPT entries. Interesting entries are those with the
2368 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
2369 all entries. When the end of the HPT is reached, the read() will
2370 return. If read() is called again on the fd, it will start again from
2371 the beginning of the HPT, but will only return HPT entries that have
2372 changed since they were last read.
2374 Data read or written is structured as a header (8 bytes) followed by a
2375 series of valid HPT entries (16 bytes) each. The header indicates how
2376 many valid HPT entries there are and how many invalid entries follow
2377 the valid entries. The invalid entries are not represented explicitly
2378 in the stream. The header format is:
2380 struct kvm_get_htab_header {
2386 Writes to the fd create HPT entries starting at the index given in the
2387 header; first `n_valid' valid entries with contents from the data
2388 written, then `n_invalid' invalid entries, invalidating any previously
2389 valid entries found.
2391 4.79 KVM_CREATE_DEVICE
2393 Capability: KVM_CAP_DEVICE_CTRL
2395 Parameters: struct kvm_create_device (in/out)
2396 Returns: 0 on success, -1 on error
2398 ENODEV: The device type is unknown or unsupported
2399 EEXIST: Device already created, and this type of device may not
2400 be instantiated multiple times
2402 Other error conditions may be defined by individual device types or
2403 have their standard meanings.
2405 Creates an emulated device in the kernel. The file descriptor returned
2406 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
2408 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
2409 device type is supported (not necessarily whether it can be created
2412 Individual devices should not define flags. Attributes should be used
2413 for specifying any behavior that is not implied by the device type
2416 struct kvm_create_device {
2417 __u32 type; /* in: KVM_DEV_TYPE_xxx */
2418 __u32 fd; /* out: device handle */
2419 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
2422 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
2424 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
2425 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
2426 Type: device ioctl, vm ioctl, vcpu ioctl
2427 Parameters: struct kvm_device_attr
2428 Returns: 0 on success, -1 on error
2430 ENXIO: The group or attribute is unknown/unsupported for this device
2431 or hardware support is missing.
2432 EPERM: The attribute cannot (currently) be accessed this way
2433 (e.g. read-only attribute, or attribute that only makes
2434 sense when the device is in a different state)
2436 Other error conditions may be defined by individual device types.
2438 Gets/sets a specified piece of device configuration and/or state. The
2439 semantics are device-specific. See individual device documentation in
2440 the "devices" directory. As with ONE_REG, the size of the data
2441 transferred is defined by the particular attribute.
2443 struct kvm_device_attr {
2444 __u32 flags; /* no flags currently defined */
2445 __u32 group; /* device-defined */
2446 __u64 attr; /* group-defined */
2447 __u64 addr; /* userspace address of attr data */
2450 4.81 KVM_HAS_DEVICE_ATTR
2452 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
2453 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
2454 Type: device ioctl, vm ioctl, vcpu ioctl
2455 Parameters: struct kvm_device_attr
2456 Returns: 0 on success, -1 on error
2458 ENXIO: The group or attribute is unknown/unsupported for this device
2459 or hardware support is missing.
2461 Tests whether a device supports a particular attribute. A successful
2462 return indicates the attribute is implemented. It does not necessarily
2463 indicate that the attribute can be read or written in the device's
2464 current state. "addr" is ignored.
2466 4.82 KVM_ARM_VCPU_INIT
2469 Architectures: arm, arm64
2471 Parameters: struct kvm_vcpu_init (in)
2472 Returns: 0 on success; -1 on error
2474 Â EINVAL: Â Â Â the target is unknown, or the combination of features is invalid.
2475 Â ENOENT: Â Â Â a features bit specified is unknown.
2477 This tells KVM what type of CPU to present to the guest, and what
2478 optional features it should have. Â This will cause a reset of the cpu
2479 registers to their initial values. Â If this is not called, KVM_RUN will
2480 return ENOEXEC for that vcpu.
2482 Note that because some registers reflect machine topology, all vcpus
2483 should be created before this ioctl is invoked.
2485 Userspace can call this function multiple times for a given vcpu, including
2486 after the vcpu has been run. This will reset the vcpu to its initial
2487 state. All calls to this function after the initial call must use the same
2488 target and same set of feature flags, otherwise EINVAL will be returned.
2491 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
2492 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
2493 and execute guest code when KVM_RUN is called.
2494 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
2495 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
2496 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU.
2497 Depends on KVM_CAP_ARM_PSCI_0_2.
2498 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU.
2499 Depends on KVM_CAP_ARM_PMU_V3.
2502 4.83 KVM_ARM_PREFERRED_TARGET
2505 Architectures: arm, arm64
2507 Parameters: struct struct kvm_vcpu_init (out)
2508 Returns: 0 on success; -1 on error
2510 ENODEV: no preferred target available for the host
2512 This queries KVM for preferred CPU target type which can be emulated
2513 by KVM on underlying host.
2515 The ioctl returns struct kvm_vcpu_init instance containing information
2516 about preferred CPU target type and recommended features for it. The
2517 kvm_vcpu_init->features bitmap returned will have feature bits set if
2518 the preferred target recommends setting these features, but this is
2521 The information returned by this ioctl can be used to prepare an instance
2522 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
2523 in VCPU matching underlying host.
2526 4.84 KVM_GET_REG_LIST
2529 Architectures: arm, arm64, mips
2531 Parameters: struct kvm_reg_list (in/out)
2532 Returns: 0 on success; -1 on error
2534 Â E2BIG: Â Â Â Â the reg index list is too big to fit in the array specified by
2535 Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n).
2537 struct kvm_reg_list {
2538 __u64 n; /* number of registers in reg[] */
2542 This ioctl returns the guest registers that are supported for the
2543 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
2546 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
2548 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
2549 Architectures: arm, arm64
2551 Parameters: struct kvm_arm_device_address (in)
2552 Returns: 0 on success, -1 on error
2554 ENODEV: The device id is unknown
2555 ENXIO: Device not supported on current system
2556 EEXIST: Address already set
2557 E2BIG: Address outside guest physical address space
2558 EBUSY: Address overlaps with other device range
2560 struct kvm_arm_device_addr {
2565 Specify a device address in the guest's physical address space where guests
2566 can access emulated or directly exposed devices, which the host kernel needs
2567 to know about. The id field is an architecture specific identifier for a
2570 ARM/arm64 divides the id field into two parts, a device id and an
2571 address type id specific to the individual device.
2573 Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
2574 field: | 0x00000000 | device id | addr type id |
2576 ARM/arm64 currently only require this when using the in-kernel GIC
2577 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
2578 as the device id. When setting the base address for the guest's
2579 mapping of the VGIC virtual CPU and distributor interface, the ioctl
2580 must be called after calling KVM_CREATE_IRQCHIP, but before calling
2581 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
2582 base addresses will return -EEXIST.
2584 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
2585 should be used instead.
2588 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
2590 Capability: KVM_CAP_PPC_RTAS
2593 Parameters: struct kvm_rtas_token_args
2594 Returns: 0 on success, -1 on error
2596 Defines a token value for a RTAS (Run Time Abstraction Services)
2597 service in order to allow it to be handled in the kernel. The
2598 argument struct gives the name of the service, which must be the name
2599 of a service that has a kernel-side implementation. If the token
2600 value is non-zero, it will be associated with that service, and
2601 subsequent RTAS calls by the guest specifying that token will be
2602 handled by the kernel. If the token value is 0, then any token
2603 associated with the service will be forgotten, and subsequent RTAS
2604 calls by the guest for that service will be passed to userspace to be
2607 4.87 KVM_SET_GUEST_DEBUG
2609 Capability: KVM_CAP_SET_GUEST_DEBUG
2610 Architectures: x86, s390, ppc, arm64
2612 Parameters: struct kvm_guest_debug (in)
2613 Returns: 0 on success; -1 on error
2615 struct kvm_guest_debug {
2618 struct kvm_guest_debug_arch arch;
2621 Set up the processor specific debug registers and configure vcpu for
2622 handling guest debug events. There are two parts to the structure, the
2623 first a control bitfield indicates the type of debug events to handle
2624 when running. Common control bits are:
2626 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
2627 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
2629 The top 16 bits of the control field are architecture specific control
2630 flags which can include the following:
2632 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64]
2633 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64]
2634 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
2635 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
2636 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
2638 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
2639 are enabled in memory so we need to ensure breakpoint exceptions are
2640 correctly trapped and the KVM run loop exits at the breakpoint and not
2641 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
2642 we need to ensure the guest vCPUs architecture specific registers are
2643 updated to the correct (supplied) values.
2645 The second part of the structure is architecture specific and
2646 typically contains a set of debug registers.
2648 For arm64 the number of debug registers is implementation defined and
2649 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
2650 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
2651 indicating the number of supported registers.
2653 When debug events exit the main run loop with the reason
2654 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
2655 structure containing architecture specific debug information.
2657 4.88 KVM_GET_EMULATED_CPUID
2659 Capability: KVM_CAP_EXT_EMUL_CPUID
2662 Parameters: struct kvm_cpuid2 (in/out)
2663 Returns: 0 on success, -1 on error
2668 struct kvm_cpuid_entry2 entries[0];
2671 The member 'flags' is used for passing flags from userspace.
2673 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
2674 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
2675 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
2677 struct kvm_cpuid_entry2 {
2688 This ioctl returns x86 cpuid features which are emulated by
2689 kvm.Userspace can use the information returned by this ioctl to query
2690 which features are emulated by kvm instead of being present natively.
2692 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
2693 structure with the 'nent' field indicating the number of entries in
2694 the variable-size array 'entries'. If the number of entries is too low
2695 to describe the cpu capabilities, an error (E2BIG) is returned. If the
2696 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
2697 is returned. If the number is just right, the 'nent' field is adjusted
2698 to the number of valid entries in the 'entries' array, which is then
2701 The entries returned are the set CPUID bits of the respective features
2702 which kvm emulates, as returned by the CPUID instruction, with unknown
2703 or unsupported feature bits cleared.
2705 Features like x2apic, for example, may not be present in the host cpu
2706 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
2707 emulated efficiently and thus not included here.
2709 The fields in each entry are defined as follows:
2711 function: the eax value used to obtain the entry
2712 index: the ecx value used to obtain the entry (for entries that are
2714 flags: an OR of zero or more of the following:
2715 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
2716 if the index field is valid
2717 KVM_CPUID_FLAG_STATEFUL_FUNC:
2718 if cpuid for this function returns different values for successive
2719 invocations; there will be several entries with the same function,
2720 all with this flag set
2721 KVM_CPUID_FLAG_STATE_READ_NEXT:
2722 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
2723 the first entry to be read by a cpu
2724 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
2725 this function/index combination
2727 4.89 KVM_S390_MEM_OP
2729 Capability: KVM_CAP_S390_MEM_OP
2732 Parameters: struct kvm_s390_mem_op (in)
2733 Returns: = 0 on success,
2734 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
2735 > 0 if an exception occurred while walking the page tables
2737 Read or write data from/to the logical (virtual) memory of a VCPU.
2739 Parameters are specified via the following structure:
2741 struct kvm_s390_mem_op {
2742 __u64 gaddr; /* the guest address */
2743 __u64 flags; /* flags */
2744 __u32 size; /* amount of bytes */
2745 __u32 op; /* type of operation */
2746 __u64 buf; /* buffer in userspace */
2747 __u8 ar; /* the access register number */
2748 __u8 reserved[31]; /* should be set to 0 */
2751 The type of operation is specified in the "op" field. It is either
2752 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or
2753 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The
2754 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check
2755 whether the corresponding memory access would create an access exception
2756 (without touching the data in the memory at the destination). In case an
2757 access exception occurred while walking the MMU tables of the guest, the
2758 ioctl returns a positive error number to indicate the type of exception.
2759 This exception is also raised directly at the corresponding VCPU if the
2760 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field.
2762 The start address of the memory region has to be specified in the "gaddr"
2763 field, and the length of the region in the "size" field. "buf" is the buffer
2764 supplied by the userspace application where the read data should be written
2765 to for KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written
2766 is stored for a KVM_S390_MEMOP_LOGICAL_WRITE. "buf" is unused and can be NULL
2767 when KVM_S390_MEMOP_F_CHECK_ONLY is specified. "ar" designates the access
2768 register number to be used.
2770 The "reserved" field is meant for future extensions. It is not used by
2771 KVM with the currently defined set of flags.
2773 4.90 KVM_S390_GET_SKEYS
2775 Capability: KVM_CAP_S390_SKEYS
2778 Parameters: struct kvm_s390_skeys
2779 Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage
2780 keys, negative value on error
2782 This ioctl is used to get guest storage key values on the s390
2783 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2785 struct kvm_s390_skeys {
2788 __u64 skeydata_addr;
2793 The start_gfn field is the number of the first guest frame whose storage keys
2796 The count field is the number of consecutive frames (starting from start_gfn)
2797 whose storage keys to get. The count field must be at least 1 and the maximum
2798 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2799 will cause the ioctl to return -EINVAL.
2801 The skeydata_addr field is the address to a buffer large enough to hold count
2802 bytes. This buffer will be filled with storage key data by the ioctl.
2804 4.91 KVM_S390_SET_SKEYS
2806 Capability: KVM_CAP_S390_SKEYS
2809 Parameters: struct kvm_s390_skeys
2810 Returns: 0 on success, negative value on error
2812 This ioctl is used to set guest storage key values on the s390
2813 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2814 See section on KVM_S390_GET_SKEYS for struct definition.
2816 The start_gfn field is the number of the first guest frame whose storage keys
2819 The count field is the number of consecutive frames (starting from start_gfn)
2820 whose storage keys to get. The count field must be at least 1 and the maximum
2821 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2822 will cause the ioctl to return -EINVAL.
2824 The skeydata_addr field is the address to a buffer containing count bytes of
2825 storage keys. Each byte in the buffer will be set as the storage key for a
2826 single frame starting at start_gfn for count frames.
2828 Note: If any architecturally invalid key value is found in the given data then
2829 the ioctl will return -EINVAL.
2833 Capability: KVM_CAP_S390_INJECT_IRQ
2836 Parameters: struct kvm_s390_irq (in)
2837 Returns: 0 on success, -1 on error
2839 EINVAL: interrupt type is invalid
2840 type is KVM_S390_SIGP_STOP and flag parameter is invalid value
2841 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger
2842 than the maximum of VCPUs
2843 EBUSY: type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped
2844 type is KVM_S390_SIGP_STOP and a stop irq is already pending
2845 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt
2848 Allows to inject an interrupt to the guest.
2850 Using struct kvm_s390_irq as a parameter allows
2851 to inject additional payload which is not
2852 possible via KVM_S390_INTERRUPT.
2854 Interrupt parameters are passed via kvm_s390_irq:
2856 struct kvm_s390_irq {
2859 struct kvm_s390_io_info io;
2860 struct kvm_s390_ext_info ext;
2861 struct kvm_s390_pgm_info pgm;
2862 struct kvm_s390_emerg_info emerg;
2863 struct kvm_s390_extcall_info extcall;
2864 struct kvm_s390_prefix_info prefix;
2865 struct kvm_s390_stop_info stop;
2866 struct kvm_s390_mchk_info mchk;
2871 type can be one of the following:
2873 KVM_S390_SIGP_STOP - sigp stop; parameter in .stop
2874 KVM_S390_PROGRAM_INT - program check; parameters in .pgm
2875 KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix
2876 KVM_S390_RESTART - restart; no parameters
2877 KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters
2878 KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters
2879 KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg
2880 KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall
2881 KVM_S390_MCHK - machine check interrupt; parameters in .mchk
2884 Note that the vcpu ioctl is asynchronous to vcpu execution.
2886 4.94 KVM_S390_GET_IRQ_STATE
2888 Capability: KVM_CAP_S390_IRQ_STATE
2891 Parameters: struct kvm_s390_irq_state (out)
2892 Returns: >= number of bytes copied into buffer,
2893 -EINVAL if buffer size is 0,
2894 -ENOBUFS if buffer size is too small to fit all pending interrupts,
2895 -EFAULT if the buffer address was invalid
2897 This ioctl allows userspace to retrieve the complete state of all currently
2898 pending interrupts in a single buffer. Use cases include migration
2899 and introspection. The parameter structure contains the address of a
2900 userspace buffer and its length:
2902 struct kvm_s390_irq_state {
2904 __u32 flags; /* will stay unused for compatibility reasons */
2906 __u32 reserved[4]; /* will stay unused for compatibility reasons */
2909 Userspace passes in the above struct and for each pending interrupt a
2910 struct kvm_s390_irq is copied to the provided buffer.
2912 The structure contains a flags and a reserved field for future extensions. As
2913 the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and
2914 reserved, these fields can not be used in the future without breaking
2917 If -ENOBUFS is returned the buffer provided was too small and userspace
2918 may retry with a bigger buffer.
2920 4.95 KVM_S390_SET_IRQ_STATE
2922 Capability: KVM_CAP_S390_IRQ_STATE
2925 Parameters: struct kvm_s390_irq_state (in)
2926 Returns: 0 on success,
2927 -EFAULT if the buffer address was invalid,
2928 -EINVAL for an invalid buffer length (see below),
2929 -EBUSY if there were already interrupts pending,
2930 errors occurring when actually injecting the
2931 interrupt. See KVM_S390_IRQ.
2933 This ioctl allows userspace to set the complete state of all cpu-local
2934 interrupts currently pending for the vcpu. It is intended for restoring
2935 interrupt state after a migration. The input parameter is a userspace buffer
2936 containing a struct kvm_s390_irq_state:
2938 struct kvm_s390_irq_state {
2940 __u32 flags; /* will stay unused for compatibility reasons */
2942 __u32 reserved[4]; /* will stay unused for compatibility reasons */
2945 The restrictions for flags and reserved apply as well.
2946 (see KVM_S390_GET_IRQ_STATE)
2948 The userspace memory referenced by buf contains a struct kvm_s390_irq
2949 for each interrupt to be injected into the guest.
2950 If one of the interrupts could not be injected for some reason the
2953 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
2954 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
2955 which is the maximum number of possibly pending cpu-local interrupts.
2959 Capability: KVM_CAP_X86_SMM
2963 Returns: 0 on success, -1 on error
2965 Queues an SMI on the thread's vcpu.
2967 4.97 KVM_CAP_PPC_MULTITCE
2969 Capability: KVM_CAP_PPC_MULTITCE
2973 This capability means the kernel is capable of handling hypercalls
2974 H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user
2975 space. This significantly accelerates DMA operations for PPC KVM guests.
2976 User space should expect that its handlers for these hypercalls
2977 are not going to be called if user space previously registered LIOBN
2978 in KVM (via KVM_CREATE_SPAPR_TCE or similar calls).
2980 In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest,
2981 user space might have to advertise it for the guest. For example,
2982 IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is
2983 present in the "ibm,hypertas-functions" device-tree property.
2985 The hypercalls mentioned above may or may not be processed successfully
2986 in the kernel based fast path. If they can not be handled by the kernel,
2987 they will get passed on to user space. So user space still has to have
2988 an implementation for these despite the in kernel acceleration.
2990 This capability is always enabled.
2992 4.98 KVM_CREATE_SPAPR_TCE_64
2994 Capability: KVM_CAP_SPAPR_TCE_64
2995 Architectures: powerpc
2997 Parameters: struct kvm_create_spapr_tce_64 (in)
2998 Returns: file descriptor for manipulating the created TCE table
3000 This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit
3001 windows, described in 4.62 KVM_CREATE_SPAPR_TCE
3003 This capability uses extended struct in ioctl interface:
3005 /* for KVM_CAP_SPAPR_TCE_64 */
3006 struct kvm_create_spapr_tce_64 {
3010 __u64 offset; /* in pages */
3011 __u64 size; /* in pages */
3014 The aim of extension is to support an additional bigger DMA window with
3015 a variable page size.
3016 KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and
3017 a bus offset of the corresponding DMA window, @size and @offset are numbers
3020 @flags are not used at the moment.
3022 The rest of functionality is identical to KVM_CREATE_SPAPR_TCE.
3024 4.99 KVM_REINJECT_CONTROL
3026 Capability: KVM_CAP_REINJECT_CONTROL
3029 Parameters: struct kvm_reinject_control (in)
3030 Returns: 0 on success,
3031 -EFAULT if struct kvm_reinject_control cannot be read,
3032 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier.
3034 i8254 (PIT) has two modes, reinject and !reinject. The default is reinject,
3035 where KVM queues elapsed i8254 ticks and monitors completion of interrupt from
3036 vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its
3037 interrupt whenever there isn't a pending interrupt from i8254.
3038 !reinject mode injects an interrupt as soon as a tick arrives.
3040 struct kvm_reinject_control {
3045 pit_reinject = 0 (!reinject mode) is recommended, unless running an old
3046 operating system that uses the PIT for timing (e.g. Linux 2.4.x).
3048 4.100 KVM_PPC_CONFIGURE_V3_MMU
3050 Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3
3053 Parameters: struct kvm_ppc_mmuv3_cfg (in)
3054 Returns: 0 on success,
3055 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read,
3056 -EINVAL if the configuration is invalid
3058 This ioctl controls whether the guest will use radix or HPT (hashed
3059 page table) translation, and sets the pointer to the process table for
3062 struct kvm_ppc_mmuv3_cfg {
3064 __u64 process_table;
3067 There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and
3068 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest
3069 to use radix tree translation, and if clear, to use HPT translation.
3070 KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest
3071 to be able to use the global TLB and SLB invalidation instructions;
3072 if clear, the guest may not use these instructions.
3074 The process_table field specifies the address and size of the guest
3075 process table, which is in the guest's space. This field is formatted
3076 as the second doubleword of the partition table entry, as defined in
3077 the Power ISA V3.00, Book III section 5.7.6.1.
3079 4.101 KVM_PPC_GET_RMMU_INFO
3081 Capability: KVM_CAP_PPC_RADIX_MMU
3084 Parameters: struct kvm_ppc_rmmu_info (out)
3085 Returns: 0 on success,
3086 -EFAULT if struct kvm_ppc_rmmu_info cannot be written,
3087 -EINVAL if no useful information can be returned
3089 This ioctl returns a structure containing two things: (a) a list
3090 containing supported radix tree geometries, and (b) a list that maps
3091 page sizes to put in the "AP" (actual page size) field for the tlbie
3092 (TLB invalidate entry) instruction.
3094 struct kvm_ppc_rmmu_info {
3095 struct kvm_ppc_radix_geom {
3100 __u32 ap_encodings[8];
3103 The geometries[] field gives up to 8 supported geometries for the
3104 radix page table, in terms of the log base 2 of the smallest page
3105 size, and the number of bits indexed at each level of the tree, from
3106 the PTE level up to the PGD level in that order. Any unused entries
3107 will have 0 in the page_shift field.
3109 The ap_encodings gives the supported page sizes and their AP field
3110 encodings, encoded with the AP value in the top 3 bits and the log
3111 base 2 of the page size in the bottom 6 bits.
3113 4.102 KVM_PPC_RESIZE_HPT_PREPARE
3115 Capability: KVM_CAP_SPAPR_RESIZE_HPT
3116 Architectures: powerpc
3118 Parameters: struct kvm_ppc_resize_hpt (in)
3119 Returns: 0 on successful completion,
3120 >0 if a new HPT is being prepared, the value is an estimated
3121 number of milliseconds until preparation is complete
3122 -EFAULT if struct kvm_reinject_control cannot be read,
3123 -EINVAL if the supplied shift or flags are invalid
3124 -ENOMEM if unable to allocate the new HPT
3125 -ENOSPC if there was a hash collision when moving existing
3126 HPT entries to the new HPT
3127 -EIO on other error conditions
3129 Used to implement the PAPR extension for runtime resizing of a guest's
3130 Hashed Page Table (HPT). Specifically this starts, stops or monitors
3131 the preparation of a new potential HPT for the guest, essentially
3132 implementing the H_RESIZE_HPT_PREPARE hypercall.
3134 If called with shift > 0 when there is no pending HPT for the guest,
3135 this begins preparation of a new pending HPT of size 2^(shift) bytes.
3136 It then returns a positive integer with the estimated number of
3137 milliseconds until preparation is complete.
3139 If called when there is a pending HPT whose size does not match that
3140 requested in the parameters, discards the existing pending HPT and
3141 creates a new one as above.
3143 If called when there is a pending HPT of the size requested, will:
3144 * If preparation of the pending HPT is already complete, return 0
3145 * If preparation of the pending HPT has failed, return an error
3146 code, then discard the pending HPT.
3147 * If preparation of the pending HPT is still in progress, return an
3148 estimated number of milliseconds until preparation is complete.
3150 If called with shift == 0, discards any currently pending HPT and
3151 returns 0 (i.e. cancels any in-progress preparation).
3153 flags is reserved for future expansion, currently setting any bits in
3154 flags will result in an -EINVAL.
3156 Normally this will be called repeatedly with the same parameters until
3157 it returns <= 0. The first call will initiate preparation, subsequent
3158 ones will monitor preparation until it completes or fails.
3160 struct kvm_ppc_resize_hpt {
3166 4.103 KVM_PPC_RESIZE_HPT_COMMIT
3168 Capability: KVM_CAP_SPAPR_RESIZE_HPT
3169 Architectures: powerpc
3171 Parameters: struct kvm_ppc_resize_hpt (in)
3172 Returns: 0 on successful completion,
3173 -EFAULT if struct kvm_reinject_control cannot be read,
3174 -EINVAL if the supplied shift or flags are invalid
3175 -ENXIO is there is no pending HPT, or the pending HPT doesn't
3176 have the requested size
3177 -EBUSY if the pending HPT is not fully prepared
3178 -ENOSPC if there was a hash collision when moving existing
3179 HPT entries to the new HPT
3180 -EIO on other error conditions
3182 Used to implement the PAPR extension for runtime resizing of a guest's
3183 Hashed Page Table (HPT). Specifically this requests that the guest be
3184 transferred to working with the new HPT, essentially implementing the
3185 H_RESIZE_HPT_COMMIT hypercall.
3187 This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has
3188 returned 0 with the same parameters. In other cases
3189 KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or
3190 -EBUSY, though others may be possible if the preparation was started,
3193 This will have undefined effects on the guest if it has not already
3194 placed itself in a quiescent state where no vcpu will make MMU enabled
3197 On succsful completion, the pending HPT will become the guest's active
3198 HPT and the previous HPT will be discarded.
3200 On failure, the guest will still be operating on its previous HPT.
3202 struct kvm_ppc_resize_hpt {
3208 4.104 KVM_X86_GET_MCE_CAP_SUPPORTED
3210 Capability: KVM_CAP_MCE
3213 Parameters: u64 mce_cap (out)
3214 Returns: 0 on success, -1 on error
3216 Returns supported MCE capabilities. The u64 mce_cap parameter
3217 has the same format as the MSR_IA32_MCG_CAP register. Supported
3218 capabilities will have the corresponding bits set.
3220 4.105 KVM_X86_SETUP_MCE
3222 Capability: KVM_CAP_MCE
3225 Parameters: u64 mcg_cap (in)
3226 Returns: 0 on success,
3227 -EFAULT if u64 mcg_cap cannot be read,
3228 -EINVAL if the requested number of banks is invalid,
3229 -EINVAL if requested MCE capability is not supported.
3231 Initializes MCE support for use. The u64 mcg_cap parameter
3232 has the same format as the MSR_IA32_MCG_CAP register and
3233 specifies which capabilities should be enabled. The maximum
3234 supported number of error-reporting banks can be retrieved when
3235 checking for KVM_CAP_MCE. The supported capabilities can be
3236 retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED.
3238 4.106 KVM_X86_SET_MCE
3240 Capability: KVM_CAP_MCE
3243 Parameters: struct kvm_x86_mce (in)
3244 Returns: 0 on success,
3245 -EFAULT if struct kvm_x86_mce cannot be read,
3246 -EINVAL if the bank number is invalid,
3247 -EINVAL if VAL bit is not set in status field.
3249 Inject a machine check error (MCE) into the guest. The input
3252 struct kvm_x86_mce {
3262 If the MCE being reported is an uncorrected error, KVM will
3263 inject it as an MCE exception into the guest. If the guest
3264 MCG_STATUS register reports that an MCE is in progress, KVM
3265 causes an KVM_EXIT_SHUTDOWN vmexit.
3267 Otherwise, if the MCE is a corrected error, KVM will just
3268 store it in the corresponding bank (provided this bank is
3269 not holding a previously reported uncorrected error).
3271 4.107 KVM_S390_GET_CMMA_BITS
3273 Capability: KVM_CAP_S390_CMMA_MIGRATION
3276 Parameters: struct kvm_s390_cmma_log (in, out)
3277 Returns: 0 on success, a negative value on error
3279 This ioctl is used to get the values of the CMMA bits on the s390
3280 architecture. It is meant to be used in two scenarios:
3281 - During live migration to save the CMMA values. Live migration needs
3282 to be enabled via the KVM_REQ_START_MIGRATION VM property.
3283 - To non-destructively peek at the CMMA values, with the flag
3284 KVM_S390_CMMA_PEEK set.
3286 The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired
3287 values are written to a buffer whose location is indicated via the "values"
3288 member in the kvm_s390_cmma_log struct. The values in the input struct are
3289 also updated as needed.
3290 Each CMMA value takes up one byte.
3292 struct kvm_s390_cmma_log {
3303 start_gfn is the number of the first guest frame whose CMMA values are
3306 count is the length of the buffer in bytes,
3308 values points to the buffer where the result will be written to.
3310 If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be
3311 KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with
3314 The result is written in the buffer pointed to by the field values, and
3315 the values of the input parameter are updated as follows.
3317 Depending on the flags, different actions are performed. The only
3318 supported flag so far is KVM_S390_CMMA_PEEK.
3320 The default behaviour if KVM_S390_CMMA_PEEK is not set is:
3321 start_gfn will indicate the first page frame whose CMMA bits were dirty.
3322 It is not necessarily the same as the one passed as input, as clean pages
3325 count will indicate the number of bytes actually written in the buffer.
3326 It can (and very often will) be smaller than the input value, since the
3327 buffer is only filled until 16 bytes of clean values are found (which
3328 are then not copied in the buffer). Since a CMMA migration block needs
3329 the base address and the length, for a total of 16 bytes, we will send
3330 back some clean data if there is some dirty data afterwards, as long as
3331 the size of the clean data does not exceed the size of the header. This
3332 allows to minimize the amount of data to be saved or transferred over
3333 the network at the expense of more roundtrips to userspace. The next
3334 invocation of the ioctl will skip over all the clean values, saving
3335 potentially more than just the 16 bytes we found.
3337 If KVM_S390_CMMA_PEEK is set:
3338 the existing storage attributes are read even when not in migration
3339 mode, and no other action is performed;
3341 the output start_gfn will be equal to the input start_gfn,
3343 the output count will be equal to the input count, except if the end of
3344 memory has been reached.
3347 the field "remaining" will indicate the total number of dirty CMMA values
3348 still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is
3353 values points to the userspace buffer where the result will be stored.
3355 This ioctl can fail with -ENOMEM if not enough memory can be allocated to
3356 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if
3357 KVM_S390_CMMA_PEEK is not set but migration mode was not enabled, with
3358 -EFAULT if the userspace address is invalid or if no page table is
3359 present for the addresses (e.g. when using hugepages).
3361 4.108 KVM_S390_SET_CMMA_BITS
3363 Capability: KVM_CAP_S390_CMMA_MIGRATION
3366 Parameters: struct kvm_s390_cmma_log (in)
3367 Returns: 0 on success, a negative value on error
3369 This ioctl is used to set the values of the CMMA bits on the s390
3370 architecture. It is meant to be used during live migration to restore
3371 the CMMA values, but there are no restrictions on its use.
3372 The ioctl takes parameters via the kvm_s390_cmma_values struct.
3373 Each CMMA value takes up one byte.
3375 struct kvm_s390_cmma_log {
3386 start_gfn indicates the starting guest frame number,
3388 count indicates how many values are to be considered in the buffer,
3390 flags is not used and must be 0.
3392 mask indicates which PGSTE bits are to be considered.
3394 remaining is not used.
3396 values points to the buffer in userspace where to store the values.
3398 This ioctl can fail with -ENOMEM if not enough memory can be allocated to
3399 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if
3400 the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or
3401 if the flags field was not 0, with -EFAULT if the userspace address is
3402 invalid, if invalid pages are written to (e.g. after the end of memory)
3403 or if no page table is present for the addresses (e.g. when using
3406 4.108 KVM_PPC_GET_CPU_CHAR
3408 Capability: KVM_CAP_PPC_GET_CPU_CHAR
3409 Architectures: powerpc
3411 Parameters: struct kvm_ppc_cpu_char (out)
3412 Returns: 0 on successful completion
3413 -EFAULT if struct kvm_ppc_cpu_char cannot be written
3415 This ioctl gives userspace information about certain characteristics
3416 of the CPU relating to speculative execution of instructions and
3417 possible information leakage resulting from speculative execution (see
3418 CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is
3419 returned in struct kvm_ppc_cpu_char, which looks like this:
3421 struct kvm_ppc_cpu_char {
3422 __u64 character; /* characteristics of the CPU */
3423 __u64 behaviour; /* recommended software behaviour */
3424 __u64 character_mask; /* valid bits in character */
3425 __u64 behaviour_mask; /* valid bits in behaviour */
3428 For extensibility, the character_mask and behaviour_mask fields
3429 indicate which bits of character and behaviour have been filled in by
3430 the kernel. If the set of defined bits is extended in future then
3431 userspace will be able to tell whether it is running on a kernel that
3432 knows about the new bits.
3434 The character field describes attributes of the CPU which can help
3435 with preventing inadvertent information disclosure - specifically,
3436 whether there is an instruction to flash-invalidate the L1 data cache
3437 (ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set
3438 to a mode where entries can only be used by the thread that created
3439 them, whether the bcctr[l] instruction prevents speculation, and
3440 whether a speculation barrier instruction (ori 31,31,0) is provided.
3442 The behaviour field describes actions that software should take to
3443 prevent inadvertent information disclosure, and thus describes which
3444 vulnerabilities the hardware is subject to; specifically whether the
3445 L1 data cache should be flushed when returning to user mode from the
3446 kernel, and whether a speculation barrier should be placed between an
3447 array bounds check and the array access.
3449 These fields use the same bit definitions as the new
3450 H_GET_CPU_CHARACTERISTICS hypercall.
3452 5. The kvm_run structure
3453 ------------------------
3455 Application code obtains a pointer to the kvm_run structure by
3456 mmap()ing a vcpu fd. From that point, application code can control
3457 execution by changing fields in kvm_run prior to calling the KVM_RUN
3458 ioctl, and obtain information about the reason KVM_RUN returned by
3459 looking up structure members.
3463 __u8 request_interrupt_window;
3465 Request that KVM_RUN return when it becomes possible to inject external
3466 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
3468 __u8 immediate_exit;
3470 This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN
3471 exits immediately, returning -EINTR. In the common scenario where a
3472 signal is used to "kick" a VCPU out of KVM_RUN, this field can be used
3473 to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability.
3474 Rather than blocking the signal outside KVM_RUN, userspace can set up
3475 a signal handler that sets run->immediate_exit to a non-zero value.
3477 This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available.
3484 When KVM_RUN has returned successfully (return value 0), this informs
3485 application code why KVM_RUN has returned. Allowable values for this
3486 field are detailed below.
3488 __u8 ready_for_interrupt_injection;
3490 If request_interrupt_window has been specified, this field indicates
3491 an interrupt can be injected now with KVM_INTERRUPT.
3495 The value of the current interrupt flag. Only valid if in-kernel
3496 local APIC is not used.
3500 More architecture-specific flags detailing state of the VCPU that may
3501 affect the device's behavior. The only currently defined flag is
3502 KVM_RUN_X86_SMM, which is valid on x86 machines and is set if the
3503 VCPU is in system management mode.
3505 /* in (pre_kvm_run), out (post_kvm_run) */
3508 The value of the cr8 register. Only valid if in-kernel local APIC is
3509 not used. Both input and output.
3513 The value of the APIC BASE msr. Only valid if in-kernel local
3514 APIC is not used. Both input and output.
3517 /* KVM_EXIT_UNKNOWN */
3519 __u64 hardware_exit_reason;
3522 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
3523 reasons. Further architecture-specific information is available in
3524 hardware_exit_reason.
3526 /* KVM_EXIT_FAIL_ENTRY */
3528 __u64 hardware_entry_failure_reason;
3531 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
3532 to unknown reasons. Further architecture-specific information is
3533 available in hardware_entry_failure_reason.
3535 /* KVM_EXIT_EXCEPTION */
3545 #define KVM_EXIT_IO_IN 0
3546 #define KVM_EXIT_IO_OUT 1
3548 __u8 size; /* bytes */
3551 __u64 data_offset; /* relative to kvm_run start */
3554 If exit_reason is KVM_EXIT_IO, then the vcpu has
3555 executed a port I/O instruction which could not be satisfied by kvm.
3556 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
3557 where kvm expects application code to place the data for the next
3558 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
3560 /* KVM_EXIT_DEBUG */
3562 struct kvm_debug_exit_arch arch;
3565 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event
3566 for which architecture specific information is returned.
3576 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
3577 executed a memory-mapped I/O instruction which could not be satisfied
3578 by kvm. The 'data' member contains the written data if 'is_write' is
3579 true, and should be filled by application code otherwise.
3581 The 'data' member contains, in its first 'len' bytes, the value as it would
3582 appear if the VCPU performed a load or store of the appropriate width directly
3585 NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and
3586 KVM_EXIT_EPR the corresponding
3587 operations are complete (and guest state is consistent) only after userspace
3588 has re-entered the kernel with KVM_RUN. The kernel side will first finish
3589 incomplete operations and then check for pending signals. Userspace
3590 can re-enter the guest with an unmasked signal pending to complete
3593 /* KVM_EXIT_HYPERCALL */
3602 Unused. This was once used for 'hypercall to userspace'. To implement
3603 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390).
3604 Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
3606 /* KVM_EXIT_TPR_ACCESS */
3613 To be documented (KVM_TPR_ACCESS_REPORTING).
3615 /* KVM_EXIT_S390_SIEIC */
3618 __u64 mask; /* psw upper half */
3619 __u64 addr; /* psw lower half */
3626 /* KVM_EXIT_S390_RESET */
3627 #define KVM_S390_RESET_POR 1
3628 #define KVM_S390_RESET_CLEAR 2
3629 #define KVM_S390_RESET_SUBSYSTEM 4
3630 #define KVM_S390_RESET_CPU_INIT 8
3631 #define KVM_S390_RESET_IPL 16
3632 __u64 s390_reset_flags;
3636 /* KVM_EXIT_S390_UCONTROL */
3638 __u64 trans_exc_code;
3642 s390 specific. A page fault has occurred for a user controlled virtual
3643 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be
3644 resolved by the kernel.
3645 The program code and the translation exception code that were placed
3646 in the cpu's lowcore are presented here as defined by the z Architecture
3647 Principles of Operation Book in the Chapter for Dynamic Address Translation
3657 Deprecated - was used for 440 KVM.
3664 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
3665 hypercalls and exit with this exit struct that contains all the guest gprs.
3667 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
3668 Userspace can now handle the hypercall and when it's done modify the gprs as
3669 necessary. Upon guest entry all guest GPRs will then be replaced by the values
3672 /* KVM_EXIT_PAPR_HCALL */
3679 This is used on 64-bit PowerPC when emulating a pSeries partition,
3680 e.g. with the 'pseries' machine type in qemu. It occurs when the
3681 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
3682 contains the hypercall number (from the guest R3), and 'args' contains
3683 the arguments (from the guest R4 - R12). Userspace should put the
3684 return code in 'ret' and any extra returned values in args[].
3685 The possible hypercalls are defined in the Power Architecture Platform
3686 Requirements (PAPR) document available from www.power.org (free
3687 developer registration required to access it).
3689 /* KVM_EXIT_S390_TSCH */
3691 __u16 subchannel_id;
3692 __u16 subchannel_nr;
3699 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
3700 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
3701 interrupt for the target subchannel has been dequeued and subchannel_id,
3702 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
3703 interrupt. ipb is needed for instruction parameter decoding.
3710 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
3711 interrupt acknowledge path to the core. When the core successfully
3712 delivers an interrupt, it automatically populates the EPR register with
3713 the interrupt vector number and acknowledges the interrupt inside
3714 the interrupt controller.
3716 In case the interrupt controller lives in user space, we need to do
3717 the interrupt acknowledge cycle through it to fetch the next to be
3718 delivered interrupt vector using this exit.
3720 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
3721 external interrupt has just been delivered into the guest. User space
3722 should put the acknowledged interrupt vector into the 'epr' field.
3724 /* KVM_EXIT_SYSTEM_EVENT */
3726 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
3727 #define KVM_SYSTEM_EVENT_RESET 2
3728 #define KVM_SYSTEM_EVENT_CRASH 3
3733 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
3734 a system-level event using some architecture specific mechanism (hypercall
3735 or some special instruction). In case of ARM/ARM64, this is triggered using
3736 HVC instruction based PSCI call from the vcpu. The 'type' field describes
3737 the system-level event type. The 'flags' field describes architecture
3738 specific flags for the system-level event.
3740 Valid values for 'type' are:
3741 KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
3742 VM. Userspace is not obliged to honour this, and if it does honour
3743 this does not need to destroy the VM synchronously (ie it may call
3744 KVM_RUN again before shutdown finally occurs).
3745 KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
3746 As with SHUTDOWN, userspace can choose to ignore the request, or
3747 to schedule the reset to occur in the future and may call KVM_RUN again.
3748 KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest
3749 has requested a crash condition maintenance. Userspace can choose
3750 to ignore the request, or to gather VM memory core dump and/or
3751 reset/shutdown of the VM.
3753 /* KVM_EXIT_IOAPIC_EOI */
3758 Indicates that the VCPU's in-kernel local APIC received an EOI for a
3759 level-triggered IOAPIC interrupt. This exit only triggers when the
3760 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled);
3761 the userspace IOAPIC should process the EOI and retrigger the interrupt if
3762 it is still asserted. Vector is the LAPIC interrupt vector for which the
3765 struct kvm_hyperv_exit {
3766 #define KVM_EXIT_HYPERV_SYNIC 1
3767 #define KVM_EXIT_HYPERV_HCALL 2
3783 /* KVM_EXIT_HYPERV */
3784 struct kvm_hyperv_exit hyperv;
3785 Indicates that the VCPU exits into userspace to process some tasks
3786 related to Hyper-V emulation.
3787 Valid values for 'type' are:
3788 KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about
3789 Hyper-V SynIC state change. Notification is used to remap SynIC
3790 event/message pages and to enable/disable SynIC messages/events processing
3793 /* Fix the size of the union. */
3798 * shared registers between kvm and userspace.
3799 * kvm_valid_regs specifies the register classes set by the host
3800 * kvm_dirty_regs specified the register classes dirtied by userspace
3801 * struct kvm_sync_regs is architecture specific, as well as the
3802 * bits for kvm_valid_regs and kvm_dirty_regs
3804 __u64 kvm_valid_regs;
3805 __u64 kvm_dirty_regs;
3807 struct kvm_sync_regs regs;
3811 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
3812 certain guest registers without having to call SET/GET_*REGS. Thus we can
3813 avoid some system call overhead if userspace has to handle the exit.
3814 Userspace can query the validity of the structure by checking
3815 kvm_valid_regs for specific bits. These bits are architecture specific
3816 and usually define the validity of a groups of registers. (e.g. one bit
3817 for general purpose registers)
3819 Please note that the kernel is allowed to use the kvm_run structure as the
3820 primary storage for certain register types. Therefore, the kernel may use the
3821 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
3827 6. Capabilities that can be enabled on vCPUs
3828 --------------------------------------------
3830 There are certain capabilities that change the behavior of the virtual CPU or
3831 the virtual machine when enabled. To enable them, please see section 4.37.
3832 Below you can find a list of capabilities and what their effect on the vCPU or
3833 the virtual machine is when enabling them.
3835 The following information is provided along with the description:
3837 Architectures: which instruction set architectures provide this ioctl.
3838 x86 includes both i386 and x86_64.
3840 Target: whether this is a per-vcpu or per-vm capability.
3842 Parameters: what parameters are accepted by the capability.
3844 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3845 are not detailed, but errors with specific meanings are.
3853 Returns: 0 on success; -1 on error
3855 This capability enables interception of OSI hypercalls that otherwise would
3856 be treated as normal system calls to be injected into the guest. OSI hypercalls
3857 were invented by Mac-on-Linux to have a standardized communication mechanism
3858 between the guest and the host.
3860 When this capability is enabled, KVM_EXIT_OSI can occur.
3863 6.2 KVM_CAP_PPC_PAPR
3868 Returns: 0 on success; -1 on error
3870 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
3871 done using the hypercall instruction "sc 1".
3873 It also sets the guest privilege level to "supervisor" mode. Usually the guest
3874 runs in "hypervisor" privilege mode with a few missing features.
3876 In addition to the above, it changes the semantics of SDR1. In this mode, the
3877 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
3878 HTAB invisible to the guest.
3880 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
3887 Parameters: args[0] is the address of a struct kvm_config_tlb
3888 Returns: 0 on success; -1 on error
3890 struct kvm_config_tlb {
3897 Configures the virtual CPU's TLB array, establishing a shared memory area
3898 between userspace and KVM. The "params" and "array" fields are userspace
3899 addresses of mmu-type-specific data structures. The "array_len" field is an
3900 safety mechanism, and should be set to the size in bytes of the memory that
3901 userspace has reserved for the array. It must be at least the size dictated
3902 by "mmu_type" and "params".
3904 While KVM_RUN is active, the shared region is under control of KVM. Its
3905 contents are undefined, and any modification by userspace results in
3906 boundedly undefined behavior.
3908 On return from KVM_RUN, the shared region will reflect the current state of
3909 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
3910 to tell KVM which entries have been changed, prior to calling KVM_RUN again
3913 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
3914 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
3915 - The "array" field points to an array of type "struct
3916 kvm_book3e_206_tlb_entry".
3917 - The array consists of all entries in the first TLB, followed by all
3918 entries in the second TLB.
3919 - Within a TLB, entries are ordered first by increasing set number. Within a
3920 set, entries are ordered by way (increasing ESEL).
3921 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
3922 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
3923 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
3924 hardware ignores this value for TLB0.
3926 6.4 KVM_CAP_S390_CSS_SUPPORT
3931 Returns: 0 on success; -1 on error
3933 This capability enables support for handling of channel I/O instructions.
3935 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
3936 handled in-kernel, while the other I/O instructions are passed to userspace.
3938 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
3939 SUBCHANNEL intercepts.
3941 Note that even though this capability is enabled per-vcpu, the complete
3942 virtual machine is affected.
3948 Parameters: args[0] defines whether the proxy facility is active
3949 Returns: 0 on success; -1 on error
3951 This capability enables or disables the delivery of interrupts through the
3952 external proxy facility.
3954 When enabled (args[0] != 0), every time the guest gets an external interrupt
3955 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
3956 to receive the topmost interrupt vector.
3958 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
3960 When this capability is enabled, KVM_EXIT_EPR can occur.
3962 6.6 KVM_CAP_IRQ_MPIC
3965 Parameters: args[0] is the MPIC device fd
3966 args[1] is the MPIC CPU number for this vcpu
3968 This capability connects the vcpu to an in-kernel MPIC device.
3970 6.7 KVM_CAP_IRQ_XICS
3974 Parameters: args[0] is the XICS device fd
3975 args[1] is the XICS CPU number (server ID) for this vcpu
3977 This capability connects the vcpu to an in-kernel XICS device.
3979 6.8 KVM_CAP_S390_IRQCHIP
3985 This capability enables the in-kernel irqchip for s390. Please refer to
3986 "4.24 KVM_CREATE_IRQCHIP" for details.
3988 6.9 KVM_CAP_MIPS_FPU
3992 Parameters: args[0] is reserved for future use (should be 0).
3994 This capability allows the use of the host Floating Point Unit by the guest. It
3995 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is
3996 done the KVM_REG_MIPS_FPR_* and KVM_REG_MIPS_FCR_* registers can be accessed
3997 (depending on the current guest FPU register mode), and the Status.FR,
3998 Config5.FRE bits are accessible via the KVM API and also from the guest,
3999 depending on them being supported by the FPU.
4001 6.10 KVM_CAP_MIPS_MSA
4005 Parameters: args[0] is reserved for future use (should be 0).
4007 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest.
4008 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest.
4009 Once this is done the KVM_REG_MIPS_VEC_* and KVM_REG_MIPS_MSA_* registers can be
4010 accessed, and the Config5.MSAEn bit is accessible via the KVM API and also from
4013 7. Capabilities that can be enabled on VMs
4014 ------------------------------------------
4016 There are certain capabilities that change the behavior of the virtual
4017 machine when enabled. To enable them, please see section 4.37. Below
4018 you can find a list of capabilities and what their effect on the VM
4019 is when enabling them.
4021 The following information is provided along with the description:
4023 Architectures: which instruction set architectures provide this ioctl.
4024 x86 includes both i386 and x86_64.
4026 Parameters: what parameters are accepted by the capability.
4028 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
4029 are not detailed, but errors with specific meanings are.
4032 7.1 KVM_CAP_PPC_ENABLE_HCALL
4035 Parameters: args[0] is the sPAPR hcall number
4036 args[1] is 0 to disable, 1 to enable in-kernel handling
4038 This capability controls whether individual sPAPR hypercalls (hcalls)
4039 get handled by the kernel or not. Enabling or disabling in-kernel
4040 handling of an hcall is effective across the VM. On creation, an
4041 initial set of hcalls are enabled for in-kernel handling, which
4042 consists of those hcalls for which in-kernel handlers were implemented
4043 before this capability was implemented. If disabled, the kernel will
4044 not to attempt to handle the hcall, but will always exit to userspace
4045 to handle it. Note that it may not make sense to enable some and
4046 disable others of a group of related hcalls, but KVM does not prevent
4047 userspace from doing that.
4049 If the hcall number specified is not one that has an in-kernel
4050 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
4053 7.2 KVM_CAP_S390_USER_SIGP
4058 This capability controls which SIGP orders will be handled completely in user
4059 space. With this capability enabled, all fast orders will be handled completely
4065 - CONDITIONAL EMERGENCY SIGNAL
4067 All other orders will be handled completely in user space.
4069 Only privileged operation exceptions will be checked for in the kernel (or even
4070 in the hardware prior to interception). If this capability is not enabled, the
4071 old way of handling SIGP orders is used (partially in kernel and user space).
4073 7.3 KVM_CAP_S390_VECTOR_REGISTERS
4077 Returns: 0 on success, negative value on error
4079 Allows use of the vector registers introduced with z13 processor, and
4080 provides for the synchronization between host and user space. Will
4081 return -EINVAL if the machine does not support vectors.
4083 7.4 KVM_CAP_S390_USER_STSI
4088 This capability allows post-handlers for the STSI instruction. After
4089 initial handling in the kernel, KVM exits to user space with
4090 KVM_EXIT_S390_STSI to allow user space to insert further data.
4092 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
4103 @addr - guest address of STSI SYSIB
4107 @ar - access register number
4109 KVM handlers should exit to userspace with rc = -EREMOTE.
4111 7.5 KVM_CAP_SPLIT_IRQCHIP
4114 Parameters: args[0] - number of routes reserved for userspace IOAPICs
4115 Returns: 0 on success, -1 on error
4117 Create a local apic for each processor in the kernel. This can be used
4118 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the
4119 IOAPIC and PIC (and also the PIT, even though this has to be enabled
4122 This capability also enables in kernel routing of interrupt requests;
4123 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are
4124 used in the IRQ routing table. The first args[0] MSI routes are reserved
4125 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes,
4126 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace.
4128 Fails if VCPU has already been created, or if the irqchip is already in the
4129 kernel (i.e. KVM_CREATE_IRQCHIP has already been called).
4136 Allows use of runtime-instrumentation introduced with zEC12 processor.
4137 Will return -EINVAL if the machine does not support runtime-instrumentation.
4138 Will return -EBUSY if a VCPU has already been created.
4140 7.7 KVM_CAP_X2APIC_API
4143 Parameters: args[0] - features that should be enabled
4144 Returns: 0 on success, -EINVAL when args[0] contains invalid features
4146 Valid feature flags in args[0] are
4148 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0)
4149 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1)
4151 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of
4152 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC,
4153 allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their
4154 respective sections.
4156 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work
4157 in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff
4158 as a broadcast even in x2APIC mode in order to support physical x2APIC
4159 without interrupt remapping. This is undesirable in logical mode,
4160 where 0xff represents CPUs 0-7 in cluster 0.
4162 7.8 KVM_CAP_S390_USER_INSTR0
4167 With this capability enabled, all illegal instructions 0x0000 (2 bytes) will
4168 be intercepted and forwarded to user space. User space can use this
4169 mechanism e.g. to realize 2-byte software breakpoints. The kernel will
4170 not inject an operating exception for these instructions, user space has
4171 to take care of that.
4173 This capability can be enabled dynamically even if VCPUs were already
4174 created and are running.
4180 Returns: 0 on success; -EINVAL if the machine does not support
4181 guarded storage; -EBUSY if a VCPU has already been created.
4183 Allows use of guarded storage for the KVM guest.
4185 7.10 KVM_CAP_S390_AIS
4190 Allow use of adapter-interruption suppression.
4191 Returns: 0 on success; -EBUSY if a VCPU has already been created.
4193 7.11 KVM_CAP_PPC_SMT
4196 Parameters: vsmt_mode, flags
4198 Enabling this capability on a VM provides userspace with a way to set
4199 the desired virtual SMT mode (i.e. the number of virtual CPUs per
4200 virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2
4201 between 1 and 8. On POWER8, vsmt_mode must also be no greater than
4202 the number of threads per subcore for the host. Currently flags must
4203 be 0. A successful call to enable this capability will result in
4204 vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is
4205 subsequently queried for the VM. This capability is only supported by
4206 HV KVM, and can only be set before any VCPUs have been created.
4207 The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT
4208 modes are available.
4210 7.12 KVM_CAP_PPC_FWNMI
4215 With this capability a machine check exception in the guest address
4216 space will cause KVM to exit the guest with NMI exit reason. This
4217 enables QEMU to build error log and branch to guest kernel registered
4218 machine check handling routine. Without this capability KVM will
4219 branch to guests' 0x200 interrupt vector.
4221 8. Other capabilities.
4222 ----------------------
4224 This section lists capabilities that give information about other
4225 features of the KVM implementation.
4227 8.1 KVM_CAP_PPC_HWRNG
4231 This capability, if KVM_CHECK_EXTENSION indicates that it is
4232 available, means that that the kernel has an implementation of the
4233 H_RANDOM hypercall backed by a hardware random-number generator.
4234 If present, the kernel H_RANDOM handler can be enabled for guest use
4235 with the KVM_CAP_PPC_ENABLE_HCALL capability.
4237 8.2 KVM_CAP_HYPERV_SYNIC
4240 This capability, if KVM_CHECK_EXTENSION indicates that it is
4241 available, means that that the kernel has an implementation of the
4242 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
4243 used to support Windows Hyper-V based guest paravirt drivers(VMBus).
4245 In order to use SynIC, it has to be activated by setting this
4246 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
4247 will disable the use of APIC hardware virtualization even if supported
4248 by the CPU, as it's incompatible with SynIC auto-EOI behavior.
4250 8.3 KVM_CAP_PPC_RADIX_MMU
4254 This capability, if KVM_CHECK_EXTENSION indicates that it is
4255 available, means that that the kernel can support guests using the
4256 radix MMU defined in Power ISA V3.00 (as implemented in the POWER9
4259 8.4 KVM_CAP_PPC_HASH_MMU_V3
4263 This capability, if KVM_CHECK_EXTENSION indicates that it is
4264 available, means that that the kernel can support guests using the
4265 hashed page table MMU defined in Power ISA V3.00 (as implemented in
4266 the POWER9 processor), including in-memory segment tables.
4272 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
4273 it is available, means that full hardware assisted virtualization capabilities
4274 of the hardware are available for use through KVM. An appropriate
4275 KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which
4278 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
4279 available, it means that the VM is using full hardware assisted virtualization
4280 capabilities of the hardware. This is useful to check after creating a VM with
4281 KVM_VM_MIPS_DEFAULT.
4283 The value returned by KVM_CHECK_EXTENSION should be compared against known
4284 values (see below). All other values are reserved. This is to allow for the
4285 possibility of other hardware assisted virtualization implementations which
4286 may be incompatible with the MIPS VZ ASE.
4288 0: The trap & emulate implementation is in use to run guest code in user
4289 mode. Guest virtual memory segments are rearranged to fit the guest in the
4290 user mode address space.
4292 1: The MIPS VZ ASE is in use, providing full hardware assisted
4293 virtualization, including standard guest virtual memory segments.
4299 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
4300 it is available, means that the trap & emulate implementation is available to
4301 run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware
4302 assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed
4303 to KVM_CREATE_VM to create a VM which utilises it.
4305 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
4306 available, it means that the VM is using trap & emulate.
4308 8.7 KVM_CAP_MIPS_64BIT
4312 This capability indicates the supported architecture type of the guest, i.e. the
4313 supported register and address width.
4315 The values returned when this capability is checked by KVM_CHECK_EXTENSION on a
4316 kvm VM handle correspond roughly to the CP0_Config.AT register field, and should
4317 be checked specifically against known values (see below). All other values are
4320 0: MIPS32 or microMIPS32.
4321 Both registers and addresses are 32-bits wide.
4322 It will only be possible to run 32-bit guest code.
4324 1: MIPS64 or microMIPS64 with access only to 32-bit compatibility segments.
4325 Registers are 64-bits wide, but addresses are 32-bits wide.
4326 64-bit guest code may run but cannot access MIPS64 memory segments.
4327 It will also be possible to run 32-bit guest code.
4329 2: MIPS64 or microMIPS64 with access to all address segments.
4330 Both registers and addresses are 64-bits wide.
4331 It will be possible to run 64-bit or 32-bit guest code.
4333 8.8 KVM_CAP_X86_GUEST_MWAIT
4337 This capability indicates that guest using memory monotoring instructions
4338 (MWAIT/MWAITX) to stop the virtual CPU will not cause a VM exit. As such time
4339 spent while virtual CPU is halted in this way will then be accounted for as
4340 guest running time on the host (as opposed to e.g. HLT).
4342 8.9 KVM_CAP_ARM_USER_IRQ
4344 Architectures: arm, arm64
4345 This capability, if KVM_CHECK_EXTENSION indicates that it is available, means
4346 that if userspace creates a VM without an in-kernel interrupt controller, it
4347 will be notified of changes to the output level of in-kernel emulated devices,
4348 which can generate virtual interrupts, presented to the VM.
4349 For such VMs, on every return to userspace, the kernel
4350 updates the vcpu's run->s.regs.device_irq_level field to represent the actual
4351 output level of the device.
4353 Whenever kvm detects a change in the device output level, kvm guarantees at
4354 least one return to userspace before running the VM. This exit could either
4355 be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way,
4356 userspace can always sample the device output level and re-compute the state of
4357 the userspace interrupt controller. Userspace should always check the state
4358 of run->s.regs.device_irq_level on every kvm exit.
4359 The value in run->s.regs.device_irq_level can represent both level and edge
4360 triggered interrupt signals, depending on the device. Edge triggered interrupt
4361 signals will exit to userspace with the bit in run->s.regs.device_irq_level
4362 set exactly once per edge signal.
4364 The field run->s.regs.device_irq_level is available independent of
4365 run->kvm_valid_regs or run->kvm_dirty_regs bits.
4367 If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a
4368 number larger than 0 indicating the version of this capability is implemented
4369 and thereby which bits in in run->s.regs.device_irq_level can signal values.
4371 Currently the following bits are defined for the device_irq_level bitmap:
4373 KVM_CAP_ARM_USER_IRQ >= 1:
4375 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer
4376 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer
4377 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal
4379 Future versions of kvm may implement additional events. These will get
4380 indicated by returning a higher number from KVM_CHECK_EXTENSION and will be
4383 8.10 KVM_CAP_PPC_SMT_POSSIBLE
4387 Querying this capability returns a bitmap indicating the possible
4388 virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N
4389 (counting from the right) is set, then a virtual SMT mode of 2^N is
4392 8.11 KVM_CAP_HYPERV_SYNIC2
4396 This capability enables a newer version of Hyper-V Synthetic interrupt
4397 controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM
4398 doesn't clear SynIC message and event flags pages when they are enabled by
4399 writing to the respective MSRs.
4401 8.12 KVM_CAP_HYPERV_VP_INDEX
4405 This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its
4406 value is used to denote the target vcpu for a SynIC interrupt. For
4407 compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this
4408 capability is absent, userspace can still query this msr's value.
4410 8.13 KVM_CAP_S390_AIS_MIGRATION
4415 This capability indicates if the flic device will be able to get/set the
4416 AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows
4417 to discover this without having to create a flic device.