1 The Definitive KVM (Kernel-based Virtual Machine) API Documentation
2 ===================================================================
7 The kvm API is a set of ioctls that are issued to control various aspects
8 of a virtual machine. The ioctls belong to three classes
10 - System ioctls: These query and set global attributes which affect the
11 whole kvm subsystem. In addition a system ioctl is used to create
14 - VM ioctls: These query and set attributes that affect an entire virtual
15 machine, for example memory layout. In addition a VM ioctl is used to
16 create virtual cpus (vcpus).
18 Only run VM ioctls from the same process (address space) that was used
21 - vcpu ioctls: These query and set attributes that control the operation
22 of a single virtual cpu.
24 Only run vcpu ioctls from the same thread that was used to create the
31 The kvm API is centered around file descriptors. An initial
32 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle
33 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this
34 handle will create a VM file descriptor which can be used to issue VM
35 ioctls. A KVM_CREATE_VCPU ioctl on a VM fd will create a virtual cpu
36 and return a file descriptor pointing to it. Finally, ioctls on a vcpu
37 fd can be used to control the vcpu, including the important task of
38 actually running guest code.
40 In general file descriptors can be migrated among processes by means
41 of fork() and the SCM_RIGHTS facility of unix domain socket. These
42 kinds of tricks are explicitly not supported by kvm. While they will
43 not cause harm to the host, their actual behavior is not guaranteed by
44 the API. The only supported use is one virtual machine per process,
45 and one vcpu per thread.
51 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward
52 incompatible change are allowed. However, there is an extension
53 facility that allows backward-compatible extensions to the API to be
56 The extension mechanism is not based on the Linux version number.
57 Instead, kvm defines extension identifiers and a facility to query
58 whether a particular extension identifier is available. If it is, a
59 set of ioctls is available for application use.
65 This section describes ioctls that can be used to control kvm guests.
66 For each ioctl, the following information is provided along with a
69 Capability: which KVM extension provides this ioctl. Can be 'basic',
70 which means that is will be provided by any kernel that supports
71 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which
72 means availability needs to be checked with KVM_CHECK_EXTENSION
73 (see section 4.4), or 'none' which means that while not all kernels
74 support this ioctl, there's no capability bit to check its
75 availability: for kernels that don't support the ioctl,
76 the ioctl returns -ENOTTY.
78 Architectures: which instruction set architectures provide this ioctl.
79 x86 includes both i386 and x86_64.
81 Type: system, vm, or vcpu.
83 Parameters: what parameters are accepted by the ioctl.
85 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
86 are not detailed, but errors with specific meanings are.
89 4.1 KVM_GET_API_VERSION
95 Returns: the constant KVM_API_VERSION (=12)
97 This identifies the API version as the stable kvm API. It is not
98 expected that this number will change. However, Linux 2.6.20 and
99 2.6.21 report earlier versions; these are not documented and not
100 supported. Applications should refuse to run if KVM_GET_API_VERSION
101 returns a value other than 12. If this check passes, all ioctls
102 described as 'basic' will be available.
110 Parameters: machine type identifier (KVM_VM_*)
111 Returns: a VM fd that can be used to control the new virtual machine.
113 The new VM has no virtual cpus and no memory. An mmap() of a VM fd
114 will access the virtual machine's physical address space; offset zero
115 corresponds to guest physical address zero. Use of mmap() on a VM fd
116 is discouraged if userspace memory allocation (KVM_CAP_USER_MEMORY) is
118 You most certainly want to use 0 as machine type.
120 In order to create user controlled virtual machines on S390, check
121 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
122 privileged user (CAP_SYS_ADMIN).
125 4.3 KVM_GET_MSR_INDEX_LIST
130 Parameters: struct kvm_msr_list (in/out)
131 Returns: 0 on success; -1 on error
133 E2BIG: the msr index list is to be to fit in the array specified by
136 struct kvm_msr_list {
137 __u32 nmsrs; /* number of msrs in entries */
141 This ioctl returns the guest msrs that are supported. The list varies
142 by kvm version and host processor, but does not change otherwise. The
143 user fills in the size of the indices array in nmsrs, and in return
144 kvm adjusts nmsrs to reflect the actual number of msrs and fills in
145 the indices array with their numbers.
147 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
148 not returned in the MSR list, as different vcpus can have a different number
149 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
152 4.4 KVM_CHECK_EXTENSION
154 Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
156 Type: system ioctl, vm ioctl
157 Parameters: extension identifier (KVM_CAP_*)
158 Returns: 0 if unsupported; 1 (or some other positive integer) if supported
160 The API allows the application to query about extensions to the core
161 kvm API. Userspace passes an extension identifier (an integer) and
162 receives an integer that describes the extension availability.
163 Generally 0 means no and 1 means yes, but some extensions may report
164 additional information in the integer return value.
166 Based on their initialization different VMs may have different capabilities.
167 It is thus encouraged to use the vm ioctl to query for capabilities (available
168 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
170 4.5 KVM_GET_VCPU_MMAP_SIZE
176 Returns: size of vcpu mmap area, in bytes
178 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
179 memory region. This ioctl returns the size of that region. See the
180 KVM_RUN documentation for details.
183 4.6 KVM_SET_MEMORY_REGION
188 Parameters: struct kvm_memory_region (in)
189 Returns: 0 on success, -1 on error
191 This ioctl is obsolete and has been removed.
199 Parameters: vcpu id (apic id on x86)
200 Returns: vcpu fd on success, -1 on error
202 This API adds a vcpu to a virtual machine. No more than max_vcpus may be added.
203 The vcpu id is an integer in the range [0, max_vcpu_id).
205 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
206 the KVM_CHECK_EXTENSION ioctl() at run-time.
207 The maximum possible value for max_vcpus can be retrieved using the
208 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
210 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
212 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
213 same as the value returned from KVM_CAP_NR_VCPUS.
215 The maximum possible value for max_vcpu_id can be retrieved using the
216 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time.
218 If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id
219 is the same as the value returned from KVM_CAP_MAX_VCPUS.
221 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
222 threads in one or more virtual CPU cores. (This is because the
223 hardware requires all the hardware threads in a CPU core to be in the
224 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
225 of vcpus per virtual core (vcore). The vcore id is obtained by
226 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
227 given vcore will always be in the same physical core as each other
228 (though that might be a different physical core from time to time).
229 Userspace can control the threading (SMT) mode of the guest by its
230 allocation of vcpu ids. For example, if userspace wants
231 single-threaded guest vcpus, it should make all vcpu ids be a multiple
232 of the number of vcpus per vcore.
234 For virtual cpus that have been created with S390 user controlled virtual
235 machines, the resulting vcpu fd can be memory mapped at page offset
236 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
237 cpu's hardware control block.
240 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
245 Parameters: struct kvm_dirty_log (in/out)
246 Returns: 0 on success, -1 on error
248 /* for KVM_GET_DIRTY_LOG */
249 struct kvm_dirty_log {
253 void __user *dirty_bitmap; /* one bit per page */
258 Given a memory slot, return a bitmap containing any pages dirtied
259 since the last call to this ioctl. Bit 0 is the first page in the
260 memory slot. Ensure the entire structure is cleared to avoid padding
263 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies
264 the address space for which you want to return the dirty bitmap.
265 They must be less than the value that KVM_CHECK_EXTENSION returns for
266 the KVM_CAP_MULTI_ADDRESS_SPACE capability.
269 4.9 KVM_SET_MEMORY_ALIAS
274 Parameters: struct kvm_memory_alias (in)
275 Returns: 0 (success), -1 (error)
277 This ioctl is obsolete and has been removed.
286 Returns: 0 on success, -1 on error
288 EINTR: an unmasked signal is pending
290 This ioctl is used to run a guest virtual cpu. While there are no
291 explicit parameters, there is an implicit parameter block that can be
292 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
293 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
294 kvm_run' (see below).
300 Architectures: all except ARM, arm64
302 Parameters: struct kvm_regs (out)
303 Returns: 0 on success, -1 on error
305 Reads the general purpose registers from the vcpu.
309 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
310 __u64 rax, rbx, rcx, rdx;
311 __u64 rsi, rdi, rsp, rbp;
312 __u64 r8, r9, r10, r11;
313 __u64 r12, r13, r14, r15;
319 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
330 Architectures: all except ARM, arm64
332 Parameters: struct kvm_regs (in)
333 Returns: 0 on success, -1 on error
335 Writes the general purpose registers into the vcpu.
337 See KVM_GET_REGS for the data structure.
343 Architectures: x86, ppc
345 Parameters: struct kvm_sregs (out)
346 Returns: 0 on success, -1 on error
348 Reads special registers from the vcpu.
352 struct kvm_segment cs, ds, es, fs, gs, ss;
353 struct kvm_segment tr, ldt;
354 struct kvm_dtable gdt, idt;
355 __u64 cr0, cr2, cr3, cr4, cr8;
358 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
361 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
363 interrupt_bitmap is a bitmap of pending external interrupts. At most
364 one bit may be set. This interrupt has been acknowledged by the APIC
365 but not yet injected into the cpu core.
371 Architectures: x86, ppc
373 Parameters: struct kvm_sregs (in)
374 Returns: 0 on success, -1 on error
376 Writes special registers into the vcpu. See KVM_GET_SREGS for the
385 Parameters: struct kvm_translation (in/out)
386 Returns: 0 on success, -1 on error
388 Translates a virtual address according to the vcpu's current address
391 struct kvm_translation {
393 __u64 linear_address;
396 __u64 physical_address;
407 Architectures: x86, ppc, mips
409 Parameters: struct kvm_interrupt (in)
410 Returns: 0 on success, negative on failure.
412 Queues a hardware interrupt vector to be injected.
414 /* for KVM_INTERRUPT */
415 struct kvm_interrupt {
422 Returns: 0 on success,
423 -EEXIST if an interrupt is already enqueued
424 -EINVAL the the irq number is invalid
425 -ENXIO if the PIC is in the kernel
426 -EFAULT if the pointer is invalid
428 Note 'irq' is an interrupt vector, not an interrupt pin or line. This
429 ioctl is useful if the in-kernel PIC is not used.
433 Queues an external interrupt to be injected. This ioctl is overleaded
434 with 3 different irq values:
438 This injects an edge type external interrupt into the guest once it's ready
439 to receive interrupts. When injected, the interrupt is done.
441 b) KVM_INTERRUPT_UNSET
443 This unsets any pending interrupt.
445 Only available with KVM_CAP_PPC_UNSET_IRQ.
447 c) KVM_INTERRUPT_SET_LEVEL
449 This injects a level type external interrupt into the guest context. The
450 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
453 Only available with KVM_CAP_PPC_IRQ_LEVEL.
455 Note that any value for 'irq' other than the ones stated above is invalid
456 and incurs unexpected behavior.
460 Queues an external interrupt to be injected into the virtual CPU. A negative
461 interrupt number dequeues the interrupt.
472 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
480 Parameters: struct kvm_msrs (in/out)
481 Returns: 0 on success, -1 on error
483 Reads model-specific registers from the vcpu. Supported msr indices can
484 be obtained using KVM_GET_MSR_INDEX_LIST.
487 __u32 nmsrs; /* number of msrs in entries */
490 struct kvm_msr_entry entries[0];
493 struct kvm_msr_entry {
499 Application code should set the 'nmsrs' member (which indicates the
500 size of the entries array) and the 'index' member of each array entry.
501 kvm will fill in the 'data' member.
509 Parameters: struct kvm_msrs (in)
510 Returns: 0 on success, -1 on error
512 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
515 Application code should set the 'nmsrs' member (which indicates the
516 size of the entries array), and the 'index' and 'data' members of each
525 Parameters: struct kvm_cpuid (in)
526 Returns: 0 on success, -1 on error
528 Defines the vcpu responses to the cpuid instruction. Applications
529 should use the KVM_SET_CPUID2 ioctl if available.
532 struct kvm_cpuid_entry {
541 /* for KVM_SET_CPUID */
545 struct kvm_cpuid_entry entries[0];
549 4.21 KVM_SET_SIGNAL_MASK
554 Parameters: struct kvm_signal_mask (in)
555 Returns: 0 on success, -1 on error
557 Defines which signals are blocked during execution of KVM_RUN. This
558 signal mask temporarily overrides the threads signal mask. Any
559 unblocked signal received (except SIGKILL and SIGSTOP, which retain
560 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
562 Note the signal will only be delivered if not blocked by the original
565 /* for KVM_SET_SIGNAL_MASK */
566 struct kvm_signal_mask {
577 Parameters: struct kvm_fpu (out)
578 Returns: 0 on success, -1 on error
580 Reads the floating point state from the vcpu.
582 /* for KVM_GET_FPU and KVM_SET_FPU */
587 __u8 ftwx; /* in fxsave format */
603 Parameters: struct kvm_fpu (in)
604 Returns: 0 on success, -1 on error
606 Writes the floating point state to the vcpu.
608 /* for KVM_GET_FPU and KVM_SET_FPU */
613 __u8 ftwx; /* in fxsave format */
624 4.24 KVM_CREATE_IRQCHIP
626 Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
627 Architectures: x86, ARM, arm64, s390
630 Returns: 0 on success, -1 on error
632 Creates an interrupt controller model in the kernel.
633 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
634 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
635 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
636 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of
637 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
638 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
639 On s390, a dummy irq routing table is created.
641 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
642 before KVM_CREATE_IRQCHIP can be used.
647 Capability: KVM_CAP_IRQCHIP
648 Architectures: x86, arm, arm64
650 Parameters: struct kvm_irq_level
651 Returns: 0 on success, -1 on error
653 Sets the level of a GSI input to the interrupt controller model in the kernel.
654 On some architectures it is required that an interrupt controller model has
655 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
656 interrupts require the level to be set to 1 and then back to 0.
658 On real hardware, interrupt pins can be active-low or active-high. This
659 does not matter for the level field of struct kvm_irq_level: 1 always
660 means active (asserted), 0 means inactive (deasserted).
662 x86 allows the operating system to program the interrupt polarity
663 (active-low/active-high) for level-triggered interrupts, and KVM used
664 to consider the polarity. However, due to bitrot in the handling of
665 active-low interrupts, the above convention is now valid on x86 too.
666 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
667 should not present interrupts to the guest as active-low unless this
668 capability is present (or unless it is not using the in-kernel irqchip,
672 ARM/arm64 can signal an interrupt either at the CPU level, or at the
673 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
674 use PPIs designated for specific cpus. The irq field is interpreted
677 Â bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 |
678 field: | irq_type | vcpu_index | irq_id |
680 The irq_type field has the following values:
681 - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
682 - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
683 (the vcpu_index field is ignored)
684 - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
686 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
688 In both cases, level is used to assert/deassert the line.
690 struct kvm_irq_level {
693 __s32 status; /* not used for KVM_IRQ_LEVEL */
695 __u32 level; /* 0 or 1 */
701 Capability: KVM_CAP_IRQCHIP
704 Parameters: struct kvm_irqchip (in/out)
705 Returns: 0 on success, -1 on error
707 Reads the state of a kernel interrupt controller created with
708 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
711 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
714 char dummy[512]; /* reserving space */
715 struct kvm_pic_state pic;
716 struct kvm_ioapic_state ioapic;
723 Capability: KVM_CAP_IRQCHIP
726 Parameters: struct kvm_irqchip (in)
727 Returns: 0 on success, -1 on error
729 Sets the state of a kernel interrupt controller created with
730 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
733 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
736 char dummy[512]; /* reserving space */
737 struct kvm_pic_state pic;
738 struct kvm_ioapic_state ioapic;
743 4.28 KVM_XEN_HVM_CONFIG
745 Capability: KVM_CAP_XEN_HVM
748 Parameters: struct kvm_xen_hvm_config (in)
749 Returns: 0 on success, -1 on error
751 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
752 page, and provides the starting address and size of the hypercall
753 blobs in userspace. When the guest writes the MSR, kvm copies one
754 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
757 struct kvm_xen_hvm_config {
770 Capability: KVM_CAP_ADJUST_CLOCK
773 Parameters: struct kvm_clock_data (out)
774 Returns: 0 on success, -1 on error
776 Gets the current timestamp of kvmclock as seen by the current guest. In
777 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
780 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the
781 set of bits that KVM can return in struct kvm_clock_data's flag member.
783 The only flag defined now is KVM_CLOCK_TSC_STABLE. If set, the returned
784 value is the exact kvmclock value seen by all VCPUs at the instant
785 when KVM_GET_CLOCK was called. If clear, the returned value is simply
786 CLOCK_MONOTONIC plus a constant offset; the offset can be modified
787 with KVM_SET_CLOCK. KVM will try to make all VCPUs follow this clock,
788 but the exact value read by each VCPU could differ, because the host
791 struct kvm_clock_data {
792 __u64 clock; /* kvmclock current value */
800 Capability: KVM_CAP_ADJUST_CLOCK
803 Parameters: struct kvm_clock_data (in)
804 Returns: 0 on success, -1 on error
806 Sets the current timestamp of kvmclock to the value specified in its parameter.
807 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
810 struct kvm_clock_data {
811 __u64 clock; /* kvmclock current value */
817 4.31 KVM_GET_VCPU_EVENTS
819 Capability: KVM_CAP_VCPU_EVENTS
820 Extended by: KVM_CAP_INTR_SHADOW
823 Parameters: struct kvm_vcpu_event (out)
824 Returns: 0 on success, -1 on error
826 Gets currently pending exceptions, interrupts, and NMIs as well as related
829 struct kvm_vcpu_events {
859 Only two fields are defined in the flags field:
861 - KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that
862 interrupt.shadow contains a valid state.
864 - KVM_VCPUEVENT_VALID_SMM may be set in the flags field to signal that
865 smi contains a valid state.
867 4.32 KVM_SET_VCPU_EVENTS
869 Capability: KVM_CAP_VCPU_EVENTS
870 Extended by: KVM_CAP_INTR_SHADOW
873 Parameters: struct kvm_vcpu_event (in)
874 Returns: 0 on success, -1 on error
876 Set pending exceptions, interrupts, and NMIs as well as related states of the
879 See KVM_GET_VCPU_EVENTS for the data structure.
881 Fields that may be modified asynchronously by running VCPUs can be excluded
882 from the update. These fields are nmi.pending, sipi_vector, smi.smm,
883 smi.pending. Keep the corresponding bits in the flags field cleared to
884 suppress overwriting the current in-kernel state. The bits are:
886 KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel
887 KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector
888 KVM_VCPUEVENT_VALID_SMM - transfer the smi sub-struct.
890 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
891 the flags field to signal that interrupt.shadow contains a valid state and
892 shall be written into the VCPU.
894 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available.
897 4.33 KVM_GET_DEBUGREGS
899 Capability: KVM_CAP_DEBUGREGS
902 Parameters: struct kvm_debugregs (out)
903 Returns: 0 on success, -1 on error
905 Reads debug registers from the vcpu.
907 struct kvm_debugregs {
916 4.34 KVM_SET_DEBUGREGS
918 Capability: KVM_CAP_DEBUGREGS
921 Parameters: struct kvm_debugregs (in)
922 Returns: 0 on success, -1 on error
924 Writes debug registers into the vcpu.
926 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
927 yet and must be cleared on entry.
930 4.35 KVM_SET_USER_MEMORY_REGION
932 Capability: KVM_CAP_USER_MEM
935 Parameters: struct kvm_userspace_memory_region (in)
936 Returns: 0 on success, -1 on error
938 struct kvm_userspace_memory_region {
941 __u64 guest_phys_addr;
942 __u64 memory_size; /* bytes */
943 __u64 userspace_addr; /* start of the userspace allocated memory */
946 /* for kvm_memory_region::flags */
947 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
948 #define KVM_MEM_READONLY (1UL << 1)
950 This ioctl allows the user to create or modify a guest physical memory
951 slot. When changing an existing slot, it may be moved in the guest
952 physical memory space, or its flags may be modified. It may not be
953 resized. Slots may not overlap in guest physical address space.
955 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot"
956 specifies the address space which is being modified. They must be
957 less than the value that KVM_CHECK_EXTENSION returns for the
958 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces
959 are unrelated; the restriction on overlapping slots only applies within
962 Memory for the region is taken starting at the address denoted by the
963 field userspace_addr, which must point at user addressable memory for
964 the entire memory slot size. Any object may back this memory, including
965 anonymous memory, ordinary files, and hugetlbfs.
967 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
968 be identical. This allows large pages in the guest to be backed by large
971 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
972 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
973 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
974 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
975 to make a new slot read-only. In this case, writes to this memory will be
976 posted to userspace as KVM_EXIT_MMIO exits.
978 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
979 the memory region are automatically reflected into the guest. For example, an
980 mmap() that affects the region will be made visible immediately. Another
981 example is madvise(MADV_DROP).
983 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl.
984 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory
985 allocation and is deprecated.
988 4.36 KVM_SET_TSS_ADDR
990 Capability: KVM_CAP_SET_TSS_ADDR
993 Parameters: unsigned long tss_address (in)
994 Returns: 0 on success, -1 on error
996 This ioctl defines the physical address of a three-page region in the guest
997 physical address space. The region must be within the first 4GB of the
998 guest physical address space and must not conflict with any memory slot
999 or any mmio address. The guest may malfunction if it accesses this memory
1002 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1003 because of a quirk in the virtualization implementation (see the internals
1004 documentation when it pops into existence).
1009 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM
1010 Architectures: x86 (only KVM_CAP_ENABLE_CAP_VM),
1011 mips (only KVM_CAP_ENABLE_CAP), ppc, s390
1012 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM)
1013 Parameters: struct kvm_enable_cap (in)
1014 Returns: 0 on success; -1 on error
1016 +Not all extensions are enabled by default. Using this ioctl the application
1017 can enable an extension, making it available to the guest.
1019 On systems that do not support this ioctl, it always fails. On systems that
1020 do support it, it only works for extensions that are supported for enablement.
1022 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
1025 struct kvm_enable_cap {
1029 The capability that is supposed to get enabled.
1033 A bitfield indicating future enhancements. Has to be 0 for now.
1037 Arguments for enabling a feature. If a feature needs initial values to
1038 function properly, this is the place to put them.
1043 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
1044 for vm-wide capabilities.
1046 4.38 KVM_GET_MP_STATE
1048 Capability: KVM_CAP_MP_STATE
1049 Architectures: x86, s390, arm, arm64
1051 Parameters: struct kvm_mp_state (out)
1052 Returns: 0 on success; -1 on error
1054 struct kvm_mp_state {
1058 Returns the vcpu's current "multiprocessing state" (though also valid on
1059 uniprocessor guests).
1061 Possible values are:
1063 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86,arm/arm64]
1064 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP)
1065 which has not yet received an INIT signal [x86]
1066 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is
1067 now ready for a SIPI [x86]
1068 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and
1069 is waiting for an interrupt [x86]
1070 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector
1071 accessible via KVM_GET_VCPU_EVENTS) [x86]
1072 - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390,arm/arm64]
1073 - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390]
1074 - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted)
1076 - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state
1079 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1080 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1081 these architectures.
1085 The only states that are valid are KVM_MP_STATE_STOPPED and
1086 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
1088 4.39 KVM_SET_MP_STATE
1090 Capability: KVM_CAP_MP_STATE
1091 Architectures: x86, s390, arm, arm64
1093 Parameters: struct kvm_mp_state (in)
1094 Returns: 0 on success; -1 on error
1096 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1099 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1100 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1101 these architectures.
1105 The only states that are valid are KVM_MP_STATE_STOPPED and
1106 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
1108 4.40 KVM_SET_IDENTITY_MAP_ADDR
1110 Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1113 Parameters: unsigned long identity (in)
1114 Returns: 0 on success, -1 on error
1116 This ioctl defines the physical address of a one-page region in the guest
1117 physical address space. The region must be within the first 4GB of the
1118 guest physical address space and must not conflict with any memory slot
1119 or any mmio address. The guest may malfunction if it accesses this memory
1122 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1123 because of a quirk in the virtualization implementation (see the internals
1124 documentation when it pops into existence).
1127 4.41 KVM_SET_BOOT_CPU_ID
1129 Capability: KVM_CAP_SET_BOOT_CPU_ID
1132 Parameters: unsigned long vcpu_id
1133 Returns: 0 on success, -1 on error
1135 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1136 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1142 Capability: KVM_CAP_XSAVE
1145 Parameters: struct kvm_xsave (out)
1146 Returns: 0 on success, -1 on error
1152 This ioctl would copy current vcpu's xsave struct to the userspace.
1157 Capability: KVM_CAP_XSAVE
1160 Parameters: struct kvm_xsave (in)
1161 Returns: 0 on success, -1 on error
1167 This ioctl would copy userspace's xsave struct to the kernel.
1172 Capability: KVM_CAP_XCRS
1175 Parameters: struct kvm_xcrs (out)
1176 Returns: 0 on success, -1 on error
1187 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1191 This ioctl would copy current vcpu's xcrs to the userspace.
1196 Capability: KVM_CAP_XCRS
1199 Parameters: struct kvm_xcrs (in)
1200 Returns: 0 on success, -1 on error
1211 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1215 This ioctl would set vcpu's xcr to the value userspace specified.
1218 4.46 KVM_GET_SUPPORTED_CPUID
1220 Capability: KVM_CAP_EXT_CPUID
1223 Parameters: struct kvm_cpuid2 (in/out)
1224 Returns: 0 on success, -1 on error
1229 struct kvm_cpuid_entry2 entries[0];
1232 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1233 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
1234 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
1236 struct kvm_cpuid_entry2 {
1247 This ioctl returns x86 cpuid features which are supported by both the hardware
1248 and kvm. Userspace can use the information returned by this ioctl to
1249 construct cpuid information (for KVM_SET_CPUID2) that is consistent with
1250 hardware, kernel, and userspace capabilities, and with user requirements (for
1251 example, the user may wish to constrain cpuid to emulate older hardware,
1252 or for feature consistency across a cluster).
1254 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1255 with the 'nent' field indicating the number of entries in the variable-size
1256 array 'entries'. If the number of entries is too low to describe the cpu
1257 capabilities, an error (E2BIG) is returned. If the number is too high,
1258 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1259 number is just right, the 'nent' field is adjusted to the number of valid
1260 entries in the 'entries' array, which is then filled.
1262 The entries returned are the host cpuid as returned by the cpuid instruction,
1263 with unknown or unsupported features masked out. Some features (for example,
1264 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1265 emulate them efficiently. The fields in each entry are defined as follows:
1267 function: the eax value used to obtain the entry
1268 index: the ecx value used to obtain the entry (for entries that are
1270 flags: an OR of zero or more of the following:
1271 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1272 if the index field is valid
1273 KVM_CPUID_FLAG_STATEFUL_FUNC:
1274 if cpuid for this function returns different values for successive
1275 invocations; there will be several entries with the same function,
1276 all with this flag set
1277 KVM_CPUID_FLAG_STATE_READ_NEXT:
1278 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
1279 the first entry to be read by a cpu
1280 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
1281 this function/index combination
1283 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1284 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1285 support. Instead it is reported via
1287 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1289 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1290 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1293 4.47 KVM_PPC_GET_PVINFO
1295 Capability: KVM_CAP_PPC_GET_PVINFO
1298 Parameters: struct kvm_ppc_pvinfo (out)
1299 Returns: 0 on success, !0 on error
1301 struct kvm_ppc_pvinfo {
1307 This ioctl fetches PV specific information that need to be passed to the guest
1308 using the device tree or other means from vm context.
1310 The hcall array defines 4 instructions that make up a hypercall.
1312 If any additional field gets added to this structure later on, a bit for that
1313 additional piece of information will be set in the flags bitmap.
1315 The flags bitmap is defined as:
1317 /* the host supports the ePAPR idle hcall
1318 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1320 4.48 KVM_ASSIGN_PCI_DEVICE (deprecated)
1325 Parameters: struct kvm_assigned_pci_dev (in)
1326 Returns: 0 on success, -1 on error
1328 Assigns a host PCI device to the VM.
1330 struct kvm_assigned_pci_dev {
1331 __u32 assigned_dev_id;
1341 The PCI device is specified by the triple segnr, busnr, and devfn.
1342 Identification in succeeding service requests is done via assigned_dev_id. The
1343 following flags are specified:
1345 /* Depends on KVM_CAP_IOMMU */
1346 #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0)
1347 /* The following two depend on KVM_CAP_PCI_2_3 */
1348 #define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1)
1349 #define KVM_DEV_ASSIGN_MASK_INTX (1 << 2)
1351 If KVM_DEV_ASSIGN_PCI_2_3 is set, the kernel will manage legacy INTx interrupts
1352 via the PCI-2.3-compliant device-level mask, thus enable IRQ sharing with other
1353 assigned devices or host devices. KVM_DEV_ASSIGN_MASK_INTX specifies the
1354 guest's view on the INTx mask, see KVM_ASSIGN_SET_INTX_MASK for details.
1356 The KVM_DEV_ASSIGN_ENABLE_IOMMU flag is a mandatory option to ensure
1357 isolation of the device. Usages not specifying this flag are deprecated.
1359 Only PCI header type 0 devices with PCI BAR resources are supported by
1360 device assignment. The user requesting this ioctl must have read/write
1361 access to the PCI sysfs resource files associated with the device.
1364 ENOTTY: kernel does not support this ioctl
1366 Other error conditions may be defined by individual device types or
1367 have their standard meanings.
1370 4.49 KVM_DEASSIGN_PCI_DEVICE (deprecated)
1375 Parameters: struct kvm_assigned_pci_dev (in)
1376 Returns: 0 on success, -1 on error
1378 Ends PCI device assignment, releasing all associated resources.
1380 See KVM_ASSIGN_PCI_DEVICE for the data structure. Only assigned_dev_id is
1381 used in kvm_assigned_pci_dev to identify the device.
1384 ENOTTY: kernel does not support this ioctl
1386 Other error conditions may be defined by individual device types or
1387 have their standard meanings.
1389 4.50 KVM_ASSIGN_DEV_IRQ (deprecated)
1391 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1394 Parameters: struct kvm_assigned_irq (in)
1395 Returns: 0 on success, -1 on error
1397 Assigns an IRQ to a passed-through device.
1399 struct kvm_assigned_irq {
1400 __u32 assigned_dev_id;
1401 __u32 host_irq; /* ignored (legacy field) */
1409 The following flags are defined:
1411 #define KVM_DEV_IRQ_HOST_INTX (1 << 0)
1412 #define KVM_DEV_IRQ_HOST_MSI (1 << 1)
1413 #define KVM_DEV_IRQ_HOST_MSIX (1 << 2)
1415 #define KVM_DEV_IRQ_GUEST_INTX (1 << 8)
1416 #define KVM_DEV_IRQ_GUEST_MSI (1 << 9)
1417 #define KVM_DEV_IRQ_GUEST_MSIX (1 << 10)
1419 It is not valid to specify multiple types per host or guest IRQ. However, the
1420 IRQ type of host and guest can differ or can even be null.
1423 ENOTTY: kernel does not support this ioctl
1425 Other error conditions may be defined by individual device types or
1426 have their standard meanings.
1429 4.51 KVM_DEASSIGN_DEV_IRQ (deprecated)
1431 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1434 Parameters: struct kvm_assigned_irq (in)
1435 Returns: 0 on success, -1 on error
1437 Ends an IRQ assignment to a passed-through device.
1439 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1440 by assigned_dev_id, flags must correspond to the IRQ type specified on
1441 KVM_ASSIGN_DEV_IRQ. Partial deassignment of host or guest IRQ is allowed.
1444 4.52 KVM_SET_GSI_ROUTING
1446 Capability: KVM_CAP_IRQ_ROUTING
1447 Architectures: x86 s390 arm arm64
1449 Parameters: struct kvm_irq_routing (in)
1450 Returns: 0 on success, -1 on error
1452 Sets the GSI routing table entries, overwriting any previously set entries.
1454 On arm/arm64, GSI routing has the following limitation:
1455 - GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD.
1457 struct kvm_irq_routing {
1460 struct kvm_irq_routing_entry entries[0];
1463 No flags are specified so far, the corresponding field must be set to zero.
1465 struct kvm_irq_routing_entry {
1471 struct kvm_irq_routing_irqchip irqchip;
1472 struct kvm_irq_routing_msi msi;
1473 struct kvm_irq_routing_s390_adapter adapter;
1474 struct kvm_irq_routing_hv_sint hv_sint;
1479 /* gsi routing entry types */
1480 #define KVM_IRQ_ROUTING_IRQCHIP 1
1481 #define KVM_IRQ_ROUTING_MSI 2
1482 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1483 #define KVM_IRQ_ROUTING_HV_SINT 4
1486 - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry
1487 type, specifies that the devid field contains a valid value. The per-VM
1488 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
1489 the device ID. If this capability is not available, userspace should
1490 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
1493 struct kvm_irq_routing_irqchip {
1498 struct kvm_irq_routing_msi {
1508 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
1509 for the device that wrote the MSI message. For PCI, this is usually a
1510 BFD identifier in the lower 16 bits.
1512 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
1513 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
1514 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
1515 address_hi must be zero.
1517 struct kvm_irq_routing_s390_adapter {
1521 __u32 summary_offset;
1525 struct kvm_irq_routing_hv_sint {
1530 4.53 KVM_ASSIGN_SET_MSIX_NR (deprecated)
1535 Parameters: struct kvm_assigned_msix_nr (in)
1536 Returns: 0 on success, -1 on error
1538 Set the number of MSI-X interrupts for an assigned device. The number is
1539 reset again by terminating the MSI-X assignment of the device via
1540 KVM_DEASSIGN_DEV_IRQ. Calling this service more than once at any earlier
1543 struct kvm_assigned_msix_nr {
1544 __u32 assigned_dev_id;
1549 #define KVM_MAX_MSIX_PER_DEV 256
1552 4.54 KVM_ASSIGN_SET_MSIX_ENTRY (deprecated)
1557 Parameters: struct kvm_assigned_msix_entry (in)
1558 Returns: 0 on success, -1 on error
1560 Specifies the routing of an MSI-X assigned device interrupt to a GSI. Setting
1561 the GSI vector to zero means disabling the interrupt.
1563 struct kvm_assigned_msix_entry {
1564 __u32 assigned_dev_id;
1566 __u16 entry; /* The index of entry in the MSI-X table */
1571 ENOTTY: kernel does not support this ioctl
1573 Other error conditions may be defined by individual device types or
1574 have their standard meanings.
1577 4.55 KVM_SET_TSC_KHZ
1579 Capability: KVM_CAP_TSC_CONTROL
1582 Parameters: virtual tsc_khz
1583 Returns: 0 on success, -1 on error
1585 Specifies the tsc frequency for the virtual machine. The unit of the
1589 4.56 KVM_GET_TSC_KHZ
1591 Capability: KVM_CAP_GET_TSC_KHZ
1595 Returns: virtual tsc-khz on success, negative value on error
1597 Returns the tsc frequency of the guest. The unit of the return value is
1598 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1604 Capability: KVM_CAP_IRQCHIP
1607 Parameters: struct kvm_lapic_state (out)
1608 Returns: 0 on success, -1 on error
1610 #define KVM_APIC_REG_SIZE 0x400
1611 struct kvm_lapic_state {
1612 char regs[KVM_APIC_REG_SIZE];
1615 Reads the Local APIC registers and copies them into the input argument. The
1616 data format and layout are the same as documented in the architecture manual.
1618 If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is
1619 enabled, then the format of APIC_ID register depends on the APIC mode
1620 (reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in
1621 the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID
1622 which is stored in bits 31-24 of the APIC register, or equivalently in
1623 byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then
1624 be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR.
1626 If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state
1627 always uses xAPIC format.
1632 Capability: KVM_CAP_IRQCHIP
1635 Parameters: struct kvm_lapic_state (in)
1636 Returns: 0 on success, -1 on error
1638 #define KVM_APIC_REG_SIZE 0x400
1639 struct kvm_lapic_state {
1640 char regs[KVM_APIC_REG_SIZE];
1643 Copies the input argument into the Local APIC registers. The data format
1644 and layout are the same as documented in the architecture manual.
1646 The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's
1647 regs field) depends on the state of the KVM_CAP_X2APIC_API capability.
1648 See the note in KVM_GET_LAPIC.
1653 Capability: KVM_CAP_IOEVENTFD
1656 Parameters: struct kvm_ioeventfd (in)
1657 Returns: 0 on success, !0 on error
1659 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
1660 within the guest. A guest write in the registered address will signal the
1661 provided event instead of triggering an exit.
1663 struct kvm_ioeventfd {
1665 __u64 addr; /* legal pio/mmio address */
1666 __u32 len; /* 0, 1, 2, 4, or 8 bytes */
1672 For the special case of virtio-ccw devices on s390, the ioevent is matched
1673 to a subchannel/virtqueue tuple instead.
1675 The following flags are defined:
1677 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
1678 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
1679 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
1680 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
1681 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
1683 If datamatch flag is set, the event will be signaled only if the written value
1684 to the registered address is equal to datamatch in struct kvm_ioeventfd.
1686 For virtio-ccw devices, addr contains the subchannel id and datamatch the
1689 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
1690 the kernel will ignore the length of guest write and may get a faster vmexit.
1691 The speedup may only apply to specific architectures, but the ioeventfd will
1696 Capability: KVM_CAP_SW_TLB
1699 Parameters: struct kvm_dirty_tlb (in)
1700 Returns: 0 on success, -1 on error
1702 struct kvm_dirty_tlb {
1707 This must be called whenever userspace has changed an entry in the shared
1708 TLB, prior to calling KVM_RUN on the associated vcpu.
1710 The "bitmap" field is the userspace address of an array. This array
1711 consists of a number of bits, equal to the total number of TLB entries as
1712 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
1713 nearest multiple of 64.
1715 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
1718 The array is little-endian: the bit 0 is the least significant bit of the
1719 first byte, bit 8 is the least significant bit of the second byte, etc.
1720 This avoids any complications with differing word sizes.
1722 The "num_dirty" field is a performance hint for KVM to determine whether it
1723 should skip processing the bitmap and just invalidate everything. It must
1724 be set to the number of set bits in the bitmap.
1727 4.61 KVM_ASSIGN_SET_INTX_MASK (deprecated)
1729 Capability: KVM_CAP_PCI_2_3
1732 Parameters: struct kvm_assigned_pci_dev (in)
1733 Returns: 0 on success, -1 on error
1735 Allows userspace to mask PCI INTx interrupts from the assigned device. The
1736 kernel will not deliver INTx interrupts to the guest between setting and
1737 clearing of KVM_ASSIGN_SET_INTX_MASK via this interface. This enables use of
1738 and emulation of PCI 2.3 INTx disable command register behavior.
1740 This may be used for both PCI 2.3 devices supporting INTx disable natively and
1741 older devices lacking this support. Userspace is responsible for emulating the
1742 read value of the INTx disable bit in the guest visible PCI command register.
1743 When modifying the INTx disable state, userspace should precede updating the
1744 physical device command register by calling this ioctl to inform the kernel of
1745 the new intended INTx mask state.
1747 Note that the kernel uses the device INTx disable bit to internally manage the
1748 device interrupt state for PCI 2.3 devices. Reads of this register may
1749 therefore not match the expected value. Writes should always use the guest
1750 intended INTx disable value rather than attempting to read-copy-update the
1751 current physical device state. Races between user and kernel updates to the
1752 INTx disable bit are handled lazily in the kernel. It's possible the device
1753 may generate unintended interrupts, but they will not be injected into the
1756 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1757 by assigned_dev_id. In the flags field, only KVM_DEV_ASSIGN_MASK_INTX is
1761 4.62 KVM_CREATE_SPAPR_TCE
1763 Capability: KVM_CAP_SPAPR_TCE
1764 Architectures: powerpc
1766 Parameters: struct kvm_create_spapr_tce (in)
1767 Returns: file descriptor for manipulating the created TCE table
1769 This creates a virtual TCE (translation control entry) table, which
1770 is an IOMMU for PAPR-style virtual I/O. It is used to translate
1771 logical addresses used in virtual I/O into guest physical addresses,
1772 and provides a scatter/gather capability for PAPR virtual I/O.
1774 /* for KVM_CAP_SPAPR_TCE */
1775 struct kvm_create_spapr_tce {
1780 The liobn field gives the logical IO bus number for which to create a
1781 TCE table. The window_size field specifies the size of the DMA window
1782 which this TCE table will translate - the table will contain one 64
1783 bit TCE entry for every 4kiB of the DMA window.
1785 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
1786 table has been created using this ioctl(), the kernel will handle it
1787 in real mode, updating the TCE table. H_PUT_TCE calls for other
1788 liobns will cause a vm exit and must be handled by userspace.
1790 The return value is a file descriptor which can be passed to mmap(2)
1791 to map the created TCE table into userspace. This lets userspace read
1792 the entries written by kernel-handled H_PUT_TCE calls, and also lets
1793 userspace update the TCE table directly which is useful in some
1797 4.63 KVM_ALLOCATE_RMA
1799 Capability: KVM_CAP_PPC_RMA
1800 Architectures: powerpc
1802 Parameters: struct kvm_allocate_rma (out)
1803 Returns: file descriptor for mapping the allocated RMA
1805 This allocates a Real Mode Area (RMA) from the pool allocated at boot
1806 time by the kernel. An RMA is a physically-contiguous, aligned region
1807 of memory used on older POWER processors to provide the memory which
1808 will be accessed by real-mode (MMU off) accesses in a KVM guest.
1809 POWER processors support a set of sizes for the RMA that usually
1810 includes 64MB, 128MB, 256MB and some larger powers of two.
1812 /* for KVM_ALLOCATE_RMA */
1813 struct kvm_allocate_rma {
1817 The return value is a file descriptor which can be passed to mmap(2)
1818 to map the allocated RMA into userspace. The mapped area can then be
1819 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
1820 RMA for a virtual machine. The size of the RMA in bytes (which is
1821 fixed at host kernel boot time) is returned in the rma_size field of
1822 the argument structure.
1824 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
1825 is supported; 2 if the processor requires all virtual machines to have
1826 an RMA, or 1 if the processor can use an RMA but doesn't require it,
1827 because it supports the Virtual RMA (VRMA) facility.
1832 Capability: KVM_CAP_USER_NMI
1836 Returns: 0 on success, -1 on error
1838 Queues an NMI on the thread's vcpu. Note this is well defined only
1839 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
1840 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
1841 has been called, this interface is completely emulated within the kernel.
1843 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
1844 following algorithm:
1847 - read the local APIC's state (KVM_GET_LAPIC)
1848 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
1849 - if so, issue KVM_NMI
1852 Some guests configure the LINT1 NMI input to cause a panic, aiding in
1856 4.65 KVM_S390_UCAS_MAP
1858 Capability: KVM_CAP_S390_UCONTROL
1861 Parameters: struct kvm_s390_ucas_mapping (in)
1862 Returns: 0 in case of success
1864 The parameter is defined like this:
1865 struct kvm_s390_ucas_mapping {
1871 This ioctl maps the memory at "user_addr" with the length "length" to
1872 the vcpu's address space starting at "vcpu_addr". All parameters need to
1873 be aligned by 1 megabyte.
1876 4.66 KVM_S390_UCAS_UNMAP
1878 Capability: KVM_CAP_S390_UCONTROL
1881 Parameters: struct kvm_s390_ucas_mapping (in)
1882 Returns: 0 in case of success
1884 The parameter is defined like this:
1885 struct kvm_s390_ucas_mapping {
1891 This ioctl unmaps the memory in the vcpu's address space starting at
1892 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
1893 All parameters need to be aligned by 1 megabyte.
1896 4.67 KVM_S390_VCPU_FAULT
1898 Capability: KVM_CAP_S390_UCONTROL
1901 Parameters: vcpu absolute address (in)
1902 Returns: 0 in case of success
1904 This call creates a page table entry on the virtual cpu's address space
1905 (for user controlled virtual machines) or the virtual machine's address
1906 space (for regular virtual machines). This only works for minor faults,
1907 thus it's recommended to access subject memory page via the user page
1908 table upfront. This is useful to handle validity intercepts for user
1909 controlled virtual machines to fault in the virtual cpu's lowcore pages
1910 prior to calling the KVM_RUN ioctl.
1913 4.68 KVM_SET_ONE_REG
1915 Capability: KVM_CAP_ONE_REG
1918 Parameters: struct kvm_one_reg (in)
1919 Returns: 0 on success, negative value on failure
1921 struct kvm_one_reg {
1926 Using this ioctl, a single vcpu register can be set to a specific value
1927 defined by user space with the passed in struct kvm_one_reg, where id
1928 refers to the register identifier as described below and addr is a pointer
1929 to a variable with the respective size. There can be architecture agnostic
1930 and architecture specific registers. Each have their own range of operation
1931 and their own constants and width. To keep track of the implemented
1932 registers, find a list below:
1934 Arch | Register | Width (bits)
1936 PPC | KVM_REG_PPC_HIOR | 64
1937 PPC | KVM_REG_PPC_IAC1 | 64
1938 PPC | KVM_REG_PPC_IAC2 | 64
1939 PPC | KVM_REG_PPC_IAC3 | 64
1940 PPC | KVM_REG_PPC_IAC4 | 64
1941 PPC | KVM_REG_PPC_DAC1 | 64
1942 PPC | KVM_REG_PPC_DAC2 | 64
1943 PPC | KVM_REG_PPC_DABR | 64
1944 PPC | KVM_REG_PPC_DSCR | 64
1945 PPC | KVM_REG_PPC_PURR | 64
1946 PPC | KVM_REG_PPC_SPURR | 64
1947 PPC | KVM_REG_PPC_DAR | 64
1948 PPC | KVM_REG_PPC_DSISR | 32
1949 PPC | KVM_REG_PPC_AMR | 64
1950 PPC | KVM_REG_PPC_UAMOR | 64
1951 PPC | KVM_REG_PPC_MMCR0 | 64
1952 PPC | KVM_REG_PPC_MMCR1 | 64
1953 PPC | KVM_REG_PPC_MMCRA | 64
1954 PPC | KVM_REG_PPC_MMCR2 | 64
1955 PPC | KVM_REG_PPC_MMCRS | 64
1956 PPC | KVM_REG_PPC_SIAR | 64
1957 PPC | KVM_REG_PPC_SDAR | 64
1958 PPC | KVM_REG_PPC_SIER | 64
1959 PPC | KVM_REG_PPC_PMC1 | 32
1960 PPC | KVM_REG_PPC_PMC2 | 32
1961 PPC | KVM_REG_PPC_PMC3 | 32
1962 PPC | KVM_REG_PPC_PMC4 | 32
1963 PPC | KVM_REG_PPC_PMC5 | 32
1964 PPC | KVM_REG_PPC_PMC6 | 32
1965 PPC | KVM_REG_PPC_PMC7 | 32
1966 PPC | KVM_REG_PPC_PMC8 | 32
1967 PPC | KVM_REG_PPC_FPR0 | 64
1969 PPC | KVM_REG_PPC_FPR31 | 64
1970 PPC | KVM_REG_PPC_VR0 | 128
1972 PPC | KVM_REG_PPC_VR31 | 128
1973 PPC | KVM_REG_PPC_VSR0 | 128
1975 PPC | KVM_REG_PPC_VSR31 | 128
1976 PPC | KVM_REG_PPC_FPSCR | 64
1977 PPC | KVM_REG_PPC_VSCR | 32
1978 PPC | KVM_REG_PPC_VPA_ADDR | 64
1979 PPC | KVM_REG_PPC_VPA_SLB | 128
1980 PPC | KVM_REG_PPC_VPA_DTL | 128
1981 PPC | KVM_REG_PPC_EPCR | 32
1982 PPC | KVM_REG_PPC_EPR | 32
1983 PPC | KVM_REG_PPC_TCR | 32
1984 PPC | KVM_REG_PPC_TSR | 32
1985 PPC | KVM_REG_PPC_OR_TSR | 32
1986 PPC | KVM_REG_PPC_CLEAR_TSR | 32
1987 PPC | KVM_REG_PPC_MAS0 | 32
1988 PPC | KVM_REG_PPC_MAS1 | 32
1989 PPC | KVM_REG_PPC_MAS2 | 64
1990 PPC | KVM_REG_PPC_MAS7_3 | 64
1991 PPC | KVM_REG_PPC_MAS4 | 32
1992 PPC | KVM_REG_PPC_MAS6 | 32
1993 PPC | KVM_REG_PPC_MMUCFG | 32
1994 PPC | KVM_REG_PPC_TLB0CFG | 32
1995 PPC | KVM_REG_PPC_TLB1CFG | 32
1996 PPC | KVM_REG_PPC_TLB2CFG | 32
1997 PPC | KVM_REG_PPC_TLB3CFG | 32
1998 PPC | KVM_REG_PPC_TLB0PS | 32
1999 PPC | KVM_REG_PPC_TLB1PS | 32
2000 PPC | KVM_REG_PPC_TLB2PS | 32
2001 PPC | KVM_REG_PPC_TLB3PS | 32
2002 PPC | KVM_REG_PPC_EPTCFG | 32
2003 PPC | KVM_REG_PPC_ICP_STATE | 64
2004 PPC | KVM_REG_PPC_TB_OFFSET | 64
2005 PPC | KVM_REG_PPC_SPMC1 | 32
2006 PPC | KVM_REG_PPC_SPMC2 | 32
2007 PPC | KVM_REG_PPC_IAMR | 64
2008 PPC | KVM_REG_PPC_TFHAR | 64
2009 PPC | KVM_REG_PPC_TFIAR | 64
2010 PPC | KVM_REG_PPC_TEXASR | 64
2011 PPC | KVM_REG_PPC_FSCR | 64
2012 PPC | KVM_REG_PPC_PSPB | 32
2013 PPC | KVM_REG_PPC_EBBHR | 64
2014 PPC | KVM_REG_PPC_EBBRR | 64
2015 PPC | KVM_REG_PPC_BESCR | 64
2016 PPC | KVM_REG_PPC_TAR | 64
2017 PPC | KVM_REG_PPC_DPDES | 64
2018 PPC | KVM_REG_PPC_DAWR | 64
2019 PPC | KVM_REG_PPC_DAWRX | 64
2020 PPC | KVM_REG_PPC_CIABR | 64
2021 PPC | KVM_REG_PPC_IC | 64
2022 PPC | KVM_REG_PPC_VTB | 64
2023 PPC | KVM_REG_PPC_CSIGR | 64
2024 PPC | KVM_REG_PPC_TACR | 64
2025 PPC | KVM_REG_PPC_TCSCR | 64
2026 PPC | KVM_REG_PPC_PID | 64
2027 PPC | KVM_REG_PPC_ACOP | 64
2028 PPC | KVM_REG_PPC_VRSAVE | 32
2029 PPC | KVM_REG_PPC_LPCR | 32
2030 PPC | KVM_REG_PPC_LPCR_64 | 64
2031 PPC | KVM_REG_PPC_PPR | 64
2032 PPC | KVM_REG_PPC_ARCH_COMPAT | 32
2033 PPC | KVM_REG_PPC_DABRX | 32
2034 PPC | KVM_REG_PPC_WORT | 64
2035 PPC | KVM_REG_PPC_SPRG9 | 64
2036 PPC | KVM_REG_PPC_DBSR | 32
2037 PPC | KVM_REG_PPC_TIDR | 64
2038 PPC | KVM_REG_PPC_PSSCR | 64
2039 PPC | KVM_REG_PPC_TM_GPR0 | 64
2041 PPC | KVM_REG_PPC_TM_GPR31 | 64
2042 PPC | KVM_REG_PPC_TM_VSR0 | 128
2044 PPC | KVM_REG_PPC_TM_VSR63 | 128
2045 PPC | KVM_REG_PPC_TM_CR | 64
2046 PPC | KVM_REG_PPC_TM_LR | 64
2047 PPC | KVM_REG_PPC_TM_CTR | 64
2048 PPC | KVM_REG_PPC_TM_FPSCR | 64
2049 PPC | KVM_REG_PPC_TM_AMR | 64
2050 PPC | KVM_REG_PPC_TM_PPR | 64
2051 PPC | KVM_REG_PPC_TM_VRSAVE | 64
2052 PPC | KVM_REG_PPC_TM_VSCR | 32
2053 PPC | KVM_REG_PPC_TM_DSCR | 64
2054 PPC | KVM_REG_PPC_TM_TAR | 64
2055 PPC | KVM_REG_PPC_TM_XER | 64
2057 MIPS | KVM_REG_MIPS_R0 | 64
2059 MIPS | KVM_REG_MIPS_R31 | 64
2060 MIPS | KVM_REG_MIPS_HI | 64
2061 MIPS | KVM_REG_MIPS_LO | 64
2062 MIPS | KVM_REG_MIPS_PC | 64
2063 MIPS | KVM_REG_MIPS_CP0_INDEX | 32
2064 MIPS | KVM_REG_MIPS_CP0_ENTRYLO0 | 64
2065 MIPS | KVM_REG_MIPS_CP0_ENTRYLO1 | 64
2066 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64
2067 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64
2068 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32
2069 MIPS | KVM_REG_MIPS_CP0_WIRED | 32
2070 MIPS | KVM_REG_MIPS_CP0_HWRENA | 32
2071 MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64
2072 MIPS | KVM_REG_MIPS_CP0_COUNT | 32
2073 MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64
2074 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32
2075 MIPS | KVM_REG_MIPS_CP0_STATUS | 32
2076 MIPS | KVM_REG_MIPS_CP0_INTCTL | 32
2077 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32
2078 MIPS | KVM_REG_MIPS_CP0_EPC | 64
2079 MIPS | KVM_REG_MIPS_CP0_PRID | 32
2080 MIPS | KVM_REG_MIPS_CP0_EBASE | 64
2081 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32
2082 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32
2083 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32
2084 MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32
2085 MIPS | KVM_REG_MIPS_CP0_CONFIG4 | 32
2086 MIPS | KVM_REG_MIPS_CP0_CONFIG5 | 32
2087 MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32
2088 MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64
2089 MIPS | KVM_REG_MIPS_CP0_KSCRATCH1 | 64
2090 MIPS | KVM_REG_MIPS_CP0_KSCRATCH2 | 64
2091 MIPS | KVM_REG_MIPS_CP0_KSCRATCH3 | 64
2092 MIPS | KVM_REG_MIPS_CP0_KSCRATCH4 | 64
2093 MIPS | KVM_REG_MIPS_CP0_KSCRATCH5 | 64
2094 MIPS | KVM_REG_MIPS_CP0_KSCRATCH6 | 64
2095 MIPS | KVM_REG_MIPS_COUNT_CTL | 64
2096 MIPS | KVM_REG_MIPS_COUNT_RESUME | 64
2097 MIPS | KVM_REG_MIPS_COUNT_HZ | 64
2098 MIPS | KVM_REG_MIPS_FPR_32(0..31) | 32
2099 MIPS | KVM_REG_MIPS_FPR_64(0..31) | 64
2100 MIPS | KVM_REG_MIPS_VEC_128(0..31) | 128
2101 MIPS | KVM_REG_MIPS_FCR_IR | 32
2102 MIPS | KVM_REG_MIPS_FCR_CSR | 32
2103 MIPS | KVM_REG_MIPS_MSA_IR | 32
2104 MIPS | KVM_REG_MIPS_MSA_CSR | 32
2106 ARM registers are mapped using the lower 32 bits. The upper 16 of that
2107 is the register group type, or coprocessor number:
2109 ARM core registers have the following id bit patterns:
2110 0x4020 0000 0010 <index into the kvm_regs struct:16>
2112 ARM 32-bit CP15 registers have the following id bit patterns:
2113 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
2115 ARM 64-bit CP15 registers have the following id bit patterns:
2116 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
2118 ARM CCSIDR registers are demultiplexed by CSSELR value:
2119 0x4020 0000 0011 00 <csselr:8>
2121 ARM 32-bit VFP control registers have the following id bit patterns:
2122 0x4020 0000 0012 1 <regno:12>
2124 ARM 64-bit FP registers have the following id bit patterns:
2125 0x4030 0000 0012 0 <regno:12>
2128 arm64 registers are mapped using the lower 32 bits. The upper 16 of
2129 that is the register group type, or coprocessor number:
2131 arm64 core/FP-SIMD registers have the following id bit patterns. Note
2132 that the size of the access is variable, as the kvm_regs structure
2133 contains elements ranging from 32 to 128 bits. The index is a 32bit
2134 value in the kvm_regs structure seen as a 32bit array.
2135 0x60x0 0000 0010 <index into the kvm_regs struct:16>
2137 arm64 CCSIDR registers are demultiplexed by CSSELR value:
2138 0x6020 0000 0011 00 <csselr:8>
2140 arm64 system registers have the following id bit patterns:
2141 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
2144 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
2145 the register group type:
2147 MIPS core registers (see above) have the following id bit patterns:
2148 0x7030 0000 0000 <reg:16>
2150 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
2151 patterns depending on whether they're 32-bit or 64-bit registers:
2152 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
2153 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2155 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64
2156 versions of the EntryLo registers regardless of the word size of the host
2157 hardware, host kernel, guest, and whether XPA is present in the guest, i.e.
2158 with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and
2159 the PFNX field starting at bit 30.
2161 MIPS KVM control registers (see above) have the following id bit patterns:
2162 0x7030 0000 0002 <reg:16>
2164 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
2165 id bit patterns depending on the size of the register being accessed. They are
2166 always accessed according to the current guest FPU mode (Status.FR and
2167 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
2168 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector
2169 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they
2170 overlap the FPU registers:
2171 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
2172 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
2173 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers)
2175 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
2176 following id bit patterns:
2177 0x7020 0000 0003 01 <0:3> <reg:5>
2179 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the
2180 following id bit patterns:
2181 0x7020 0000 0003 02 <0:3> <reg:5>
2184 4.69 KVM_GET_ONE_REG
2186 Capability: KVM_CAP_ONE_REG
2189 Parameters: struct kvm_one_reg (in and out)
2190 Returns: 0 on success, negative value on failure
2192 This ioctl allows to receive the value of a single register implemented
2193 in a vcpu. The register to read is indicated by the "id" field of the
2194 kvm_one_reg struct passed in. On success, the register value can be found
2195 at the memory location pointed to by "addr".
2197 The list of registers accessible using this interface is identical to the
2201 4.70 KVM_KVMCLOCK_CTRL
2203 Capability: KVM_CAP_KVMCLOCK_CTRL
2204 Architectures: Any that implement pvclocks (currently x86 only)
2207 Returns: 0 on success, -1 on error
2209 This signals to the host kernel that the specified guest is being paused by
2210 userspace. The host will set a flag in the pvclock structure that is checked
2211 from the soft lockup watchdog. The flag is part of the pvclock structure that
2212 is shared between guest and host, specifically the second bit of the flags
2213 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2214 the host and read/cleared exclusively by the guest. The guest operation of
2215 checking and clearing the flag must an atomic operation so
2216 load-link/store-conditional, or equivalent must be used. There are two cases
2217 where the guest will clear the flag: when the soft lockup watchdog timer resets
2218 itself or when a soft lockup is detected. This ioctl can be called any time
2219 after pausing the vcpu, but before it is resumed.
2224 Capability: KVM_CAP_SIGNAL_MSI
2225 Architectures: x86 arm arm64
2227 Parameters: struct kvm_msi (in)
2228 Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2230 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2242 flags: KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM
2243 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
2244 the device ID. If this capability is not available, userspace
2245 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
2247 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
2248 for the device that wrote the MSI message. For PCI, this is usually a
2249 BFD identifier in the lower 16 bits.
2251 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
2252 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
2253 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
2254 address_hi must be zero.
2257 4.71 KVM_CREATE_PIT2
2259 Capability: KVM_CAP_PIT2
2262 Parameters: struct kvm_pit_config (in)
2263 Returns: 0 on success, -1 on error
2265 Creates an in-kernel device model for the i8254 PIT. This call is only valid
2266 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
2267 parameters have to be passed:
2269 struct kvm_pit_config {
2276 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
2278 PIT timer interrupts may use a per-VM kernel thread for injection. If it
2279 exists, this thread will have a name of the following pattern:
2281 kvm-pit/<owner-process-pid>
2283 When running a guest with elevated priorities, the scheduling parameters of
2284 this thread may have to be adjusted accordingly.
2286 This IOCTL replaces the obsolete KVM_CREATE_PIT.
2291 Capability: KVM_CAP_PIT_STATE2
2294 Parameters: struct kvm_pit_state2 (out)
2295 Returns: 0 on success, -1 on error
2297 Retrieves the state of the in-kernel PIT model. Only valid after
2298 KVM_CREATE_PIT2. The state is returned in the following structure:
2300 struct kvm_pit_state2 {
2301 struct kvm_pit_channel_state channels[3];
2308 /* disable PIT in HPET legacy mode */
2309 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
2311 This IOCTL replaces the obsolete KVM_GET_PIT.
2316 Capability: KVM_CAP_PIT_STATE2
2319 Parameters: struct kvm_pit_state2 (in)
2320 Returns: 0 on success, -1 on error
2322 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
2323 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
2325 This IOCTL replaces the obsolete KVM_SET_PIT.
2328 4.74 KVM_PPC_GET_SMMU_INFO
2330 Capability: KVM_CAP_PPC_GET_SMMU_INFO
2331 Architectures: powerpc
2334 Returns: 0 on success, -1 on error
2336 This populates and returns a structure describing the features of
2337 the "Server" class MMU emulation supported by KVM.
2338 This can in turn be used by userspace to generate the appropriate
2339 device-tree properties for the guest operating system.
2341 The structure contains some global information, followed by an
2342 array of supported segment page sizes:
2344 struct kvm_ppc_smmu_info {
2348 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
2351 The supported flags are:
2353 - KVM_PPC_PAGE_SIZES_REAL:
2354 When that flag is set, guest page sizes must "fit" the backing
2355 store page sizes. When not set, any page size in the list can
2356 be used regardless of how they are backed by userspace.
2358 - KVM_PPC_1T_SEGMENTS
2359 The emulated MMU supports 1T segments in addition to the
2362 The "slb_size" field indicates how many SLB entries are supported
2364 The "sps" array contains 8 entries indicating the supported base
2365 page sizes for a segment in increasing order. Each entry is defined
2368 struct kvm_ppc_one_seg_page_size {
2369 __u32 page_shift; /* Base page shift of segment (or 0) */
2370 __u32 slb_enc; /* SLB encoding for BookS */
2371 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
2374 An entry with a "page_shift" of 0 is unused. Because the array is
2375 organized in increasing order, a lookup can stop when encoutering
2378 The "slb_enc" field provides the encoding to use in the SLB for the
2379 page size. The bits are in positions such as the value can directly
2380 be OR'ed into the "vsid" argument of the slbmte instruction.
2382 The "enc" array is a list which for each of those segment base page
2383 size provides the list of supported actual page sizes (which can be
2384 only larger or equal to the base page size), along with the
2385 corresponding encoding in the hash PTE. Similarly, the array is
2386 8 entries sorted by increasing sizes and an entry with a "0" shift
2387 is an empty entry and a terminator:
2389 struct kvm_ppc_one_page_size {
2390 __u32 page_shift; /* Page shift (or 0) */
2391 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
2394 The "pte_enc" field provides a value that can OR'ed into the hash
2395 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
2396 into the hash PTE second double word).
2400 Capability: KVM_CAP_IRQFD
2401 Architectures: x86 s390 arm arm64
2403 Parameters: struct kvm_irqfd (in)
2404 Returns: 0 on success, -1 on error
2406 Allows setting an eventfd to directly trigger a guest interrupt.
2407 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
2408 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
2409 an event is triggered on the eventfd, an interrupt is injected into
2410 the guest using the specified gsi pin. The irqfd is removed using
2411 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
2414 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
2415 mechanism allowing emulation of level-triggered, irqfd-based
2416 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
2417 additional eventfd in the kvm_irqfd.resamplefd field. When operating
2418 in resample mode, posting of an interrupt through kvm_irq.fd asserts
2419 the specified gsi in the irqchip. When the irqchip is resampled, such
2420 as from an EOI, the gsi is de-asserted and the user is notified via
2421 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
2422 the interrupt if the device making use of it still requires service.
2423 Note that closing the resamplefd is not sufficient to disable the
2424 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
2425 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
2427 On arm/arm64, gsi routing being supported, the following can happen:
2428 - in case no routing entry is associated to this gsi, injection fails
2429 - in case the gsi is associated to an irqchip routing entry,
2430 irqchip.pin + 32 corresponds to the injected SPI ID.
2431 - in case the gsi is associated to an MSI routing entry, the MSI
2432 message and device ID are translated into an LPI (support restricted
2433 to GICv3 ITS in-kernel emulation).
2435 4.76 KVM_PPC_ALLOCATE_HTAB
2437 Capability: KVM_CAP_PPC_ALLOC_HTAB
2438 Architectures: powerpc
2440 Parameters: Pointer to u32 containing hash table order (in/out)
2441 Returns: 0 on success, -1 on error
2443 This requests the host kernel to allocate an MMU hash table for a
2444 guest using the PAPR paravirtualization interface. This only does
2445 anything if the kernel is configured to use the Book 3S HV style of
2446 virtualization. Otherwise the capability doesn't exist and the ioctl
2447 returns an ENOTTY error. The rest of this description assumes Book 3S
2450 There must be no vcpus running when this ioctl is called; if there
2451 are, it will do nothing and return an EBUSY error.
2453 The parameter is a pointer to a 32-bit unsigned integer variable
2454 containing the order (log base 2) of the desired size of the hash
2455 table, which must be between 18 and 46. On successful return from the
2456 ioctl, the value will not be changed by the kernel.
2458 If no hash table has been allocated when any vcpu is asked to run
2459 (with the KVM_RUN ioctl), the host kernel will allocate a
2460 default-sized hash table (16 MB).
2462 If this ioctl is called when a hash table has already been allocated,
2463 with a different order from the existing hash table, the existing hash
2464 table will be freed and a new one allocated. If this is ioctl is
2465 called when a hash table has already been allocated of the same order
2466 as specified, the kernel will clear out the existing hash table (zero
2467 all HPTEs). In either case, if the guest is using the virtualized
2468 real-mode area (VRMA) facility, the kernel will re-create the VMRA
2469 HPTEs on the next KVM_RUN of any vcpu.
2471 4.77 KVM_S390_INTERRUPT
2475 Type: vm ioctl, vcpu ioctl
2476 Parameters: struct kvm_s390_interrupt (in)
2477 Returns: 0 on success, -1 on error
2479 Allows to inject an interrupt to the guest. Interrupts can be floating
2480 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
2482 Interrupt parameters are passed via kvm_s390_interrupt:
2484 struct kvm_s390_interrupt {
2490 type can be one of the following:
2492 KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm
2493 KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm
2494 KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm
2495 KVM_S390_RESTART (vcpu) - restart
2496 KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt
2497 KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt
2498 KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt
2499 parameters in parm and parm64
2500 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm
2501 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm
2502 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm
2503 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
2504 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
2505 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
2506 interruption subclass)
2507 KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm,
2508 machine check interrupt code in parm64 (note that
2509 machine checks needing further payload are not
2510 supported by this ioctl)
2512 Note that the vcpu ioctl is asynchronous to vcpu execution.
2514 4.78 KVM_PPC_GET_HTAB_FD
2516 Capability: KVM_CAP_PPC_HTAB_FD
2517 Architectures: powerpc
2519 Parameters: Pointer to struct kvm_get_htab_fd (in)
2520 Returns: file descriptor number (>= 0) on success, -1 on error
2522 This returns a file descriptor that can be used either to read out the
2523 entries in the guest's hashed page table (HPT), or to write entries to
2524 initialize the HPT. The returned fd can only be written to if the
2525 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
2526 can only be read if that bit is clear. The argument struct looks like
2529 /* For KVM_PPC_GET_HTAB_FD */
2530 struct kvm_get_htab_fd {
2536 /* Values for kvm_get_htab_fd.flags */
2537 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
2538 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
2540 The `start_index' field gives the index in the HPT of the entry at
2541 which to start reading. It is ignored when writing.
2543 Reads on the fd will initially supply information about all
2544 "interesting" HPT entries. Interesting entries are those with the
2545 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
2546 all entries. When the end of the HPT is reached, the read() will
2547 return. If read() is called again on the fd, it will start again from
2548 the beginning of the HPT, but will only return HPT entries that have
2549 changed since they were last read.
2551 Data read or written is structured as a header (8 bytes) followed by a
2552 series of valid HPT entries (16 bytes) each. The header indicates how
2553 many valid HPT entries there are and how many invalid entries follow
2554 the valid entries. The invalid entries are not represented explicitly
2555 in the stream. The header format is:
2557 struct kvm_get_htab_header {
2563 Writes to the fd create HPT entries starting at the index given in the
2564 header; first `n_valid' valid entries with contents from the data
2565 written, then `n_invalid' invalid entries, invalidating any previously
2566 valid entries found.
2568 4.79 KVM_CREATE_DEVICE
2570 Capability: KVM_CAP_DEVICE_CTRL
2572 Parameters: struct kvm_create_device (in/out)
2573 Returns: 0 on success, -1 on error
2575 ENODEV: The device type is unknown or unsupported
2576 EEXIST: Device already created, and this type of device may not
2577 be instantiated multiple times
2579 Other error conditions may be defined by individual device types or
2580 have their standard meanings.
2582 Creates an emulated device in the kernel. The file descriptor returned
2583 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
2585 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
2586 device type is supported (not necessarily whether it can be created
2589 Individual devices should not define flags. Attributes should be used
2590 for specifying any behavior that is not implied by the device type
2593 struct kvm_create_device {
2594 __u32 type; /* in: KVM_DEV_TYPE_xxx */
2595 __u32 fd; /* out: device handle */
2596 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
2599 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
2601 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
2602 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
2603 Type: device ioctl, vm ioctl, vcpu ioctl
2604 Parameters: struct kvm_device_attr
2605 Returns: 0 on success, -1 on error
2607 ENXIO: The group or attribute is unknown/unsupported for this device
2608 or hardware support is missing.
2609 EPERM: The attribute cannot (currently) be accessed this way
2610 (e.g. read-only attribute, or attribute that only makes
2611 sense when the device is in a different state)
2613 Other error conditions may be defined by individual device types.
2615 Gets/sets a specified piece of device configuration and/or state. The
2616 semantics are device-specific. See individual device documentation in
2617 the "devices" directory. As with ONE_REG, the size of the data
2618 transferred is defined by the particular attribute.
2620 struct kvm_device_attr {
2621 __u32 flags; /* no flags currently defined */
2622 __u32 group; /* device-defined */
2623 __u64 attr; /* group-defined */
2624 __u64 addr; /* userspace address of attr data */
2627 4.81 KVM_HAS_DEVICE_ATTR
2629 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
2630 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
2631 Type: device ioctl, vm ioctl, vcpu ioctl
2632 Parameters: struct kvm_device_attr
2633 Returns: 0 on success, -1 on error
2635 ENXIO: The group or attribute is unknown/unsupported for this device
2636 or hardware support is missing.
2638 Tests whether a device supports a particular attribute. A successful
2639 return indicates the attribute is implemented. It does not necessarily
2640 indicate that the attribute can be read or written in the device's
2641 current state. "addr" is ignored.
2643 4.82 KVM_ARM_VCPU_INIT
2646 Architectures: arm, arm64
2648 Parameters: struct kvm_vcpu_init (in)
2649 Returns: 0 on success; -1 on error
2651 Â EINVAL: Â Â Â the target is unknown, or the combination of features is invalid.
2652 Â ENOENT: Â Â Â a features bit specified is unknown.
2654 This tells KVM what type of CPU to present to the guest, and what
2655 optional features it should have. Â This will cause a reset of the cpu
2656 registers to their initial values. Â If this is not called, KVM_RUN will
2657 return ENOEXEC for that vcpu.
2659 Note that because some registers reflect machine topology, all vcpus
2660 should be created before this ioctl is invoked.
2662 Userspace can call this function multiple times for a given vcpu, including
2663 after the vcpu has been run. This will reset the vcpu to its initial
2664 state. All calls to this function after the initial call must use the same
2665 target and same set of feature flags, otherwise EINVAL will be returned.
2668 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
2669 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
2670 and execute guest code when KVM_RUN is called.
2671 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
2672 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
2673 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU.
2674 Depends on KVM_CAP_ARM_PSCI_0_2.
2675 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU.
2676 Depends on KVM_CAP_ARM_PMU_V3.
2679 4.83 KVM_ARM_PREFERRED_TARGET
2682 Architectures: arm, arm64
2684 Parameters: struct struct kvm_vcpu_init (out)
2685 Returns: 0 on success; -1 on error
2687 ENODEV: no preferred target available for the host
2689 This queries KVM for preferred CPU target type which can be emulated
2690 by KVM on underlying host.
2692 The ioctl returns struct kvm_vcpu_init instance containing information
2693 about preferred CPU target type and recommended features for it. The
2694 kvm_vcpu_init->features bitmap returned will have feature bits set if
2695 the preferred target recommends setting these features, but this is
2698 The information returned by this ioctl can be used to prepare an instance
2699 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
2700 in VCPU matching underlying host.
2703 4.84 KVM_GET_REG_LIST
2706 Architectures: arm, arm64, mips
2708 Parameters: struct kvm_reg_list (in/out)
2709 Returns: 0 on success; -1 on error
2711 Â E2BIG: Â Â Â Â the reg index list is too big to fit in the array specified by
2712 Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n).
2714 struct kvm_reg_list {
2715 __u64 n; /* number of registers in reg[] */
2719 This ioctl returns the guest registers that are supported for the
2720 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
2723 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
2725 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
2726 Architectures: arm, arm64
2728 Parameters: struct kvm_arm_device_address (in)
2729 Returns: 0 on success, -1 on error
2731 ENODEV: The device id is unknown
2732 ENXIO: Device not supported on current system
2733 EEXIST: Address already set
2734 E2BIG: Address outside guest physical address space
2735 EBUSY: Address overlaps with other device range
2737 struct kvm_arm_device_addr {
2742 Specify a device address in the guest's physical address space where guests
2743 can access emulated or directly exposed devices, which the host kernel needs
2744 to know about. The id field is an architecture specific identifier for a
2747 ARM/arm64 divides the id field into two parts, a device id and an
2748 address type id specific to the individual device.
2750 Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
2751 field: | 0x00000000 | device id | addr type id |
2753 ARM/arm64 currently only require this when using the in-kernel GIC
2754 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
2755 as the device id. When setting the base address for the guest's
2756 mapping of the VGIC virtual CPU and distributor interface, the ioctl
2757 must be called after calling KVM_CREATE_IRQCHIP, but before calling
2758 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
2759 base addresses will return -EEXIST.
2761 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
2762 should be used instead.
2765 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
2767 Capability: KVM_CAP_PPC_RTAS
2770 Parameters: struct kvm_rtas_token_args
2771 Returns: 0 on success, -1 on error
2773 Defines a token value for a RTAS (Run Time Abstraction Services)
2774 service in order to allow it to be handled in the kernel. The
2775 argument struct gives the name of the service, which must be the name
2776 of a service that has a kernel-side implementation. If the token
2777 value is non-zero, it will be associated with that service, and
2778 subsequent RTAS calls by the guest specifying that token will be
2779 handled by the kernel. If the token value is 0, then any token
2780 associated with the service will be forgotten, and subsequent RTAS
2781 calls by the guest for that service will be passed to userspace to be
2784 4.87 KVM_SET_GUEST_DEBUG
2786 Capability: KVM_CAP_SET_GUEST_DEBUG
2787 Architectures: x86, s390, ppc, arm64
2789 Parameters: struct kvm_guest_debug (in)
2790 Returns: 0 on success; -1 on error
2792 struct kvm_guest_debug {
2795 struct kvm_guest_debug_arch arch;
2798 Set up the processor specific debug registers and configure vcpu for
2799 handling guest debug events. There are two parts to the structure, the
2800 first a control bitfield indicates the type of debug events to handle
2801 when running. Common control bits are:
2803 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
2804 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
2806 The top 16 bits of the control field are architecture specific control
2807 flags which can include the following:
2809 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64]
2810 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64]
2811 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
2812 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
2813 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
2815 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
2816 are enabled in memory so we need to ensure breakpoint exceptions are
2817 correctly trapped and the KVM run loop exits at the breakpoint and not
2818 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
2819 we need to ensure the guest vCPUs architecture specific registers are
2820 updated to the correct (supplied) values.
2822 The second part of the structure is architecture specific and
2823 typically contains a set of debug registers.
2825 For arm64 the number of debug registers is implementation defined and
2826 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
2827 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
2828 indicating the number of supported registers.
2830 When debug events exit the main run loop with the reason
2831 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
2832 structure containing architecture specific debug information.
2834 4.88 KVM_GET_EMULATED_CPUID
2836 Capability: KVM_CAP_EXT_EMUL_CPUID
2839 Parameters: struct kvm_cpuid2 (in/out)
2840 Returns: 0 on success, -1 on error
2845 struct kvm_cpuid_entry2 entries[0];
2848 The member 'flags' is used for passing flags from userspace.
2850 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
2851 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
2852 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
2854 struct kvm_cpuid_entry2 {
2865 This ioctl returns x86 cpuid features which are emulated by
2866 kvm.Userspace can use the information returned by this ioctl to query
2867 which features are emulated by kvm instead of being present natively.
2869 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
2870 structure with the 'nent' field indicating the number of entries in
2871 the variable-size array 'entries'. If the number of entries is too low
2872 to describe the cpu capabilities, an error (E2BIG) is returned. If the
2873 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
2874 is returned. If the number is just right, the 'nent' field is adjusted
2875 to the number of valid entries in the 'entries' array, which is then
2878 The entries returned are the set CPUID bits of the respective features
2879 which kvm emulates, as returned by the CPUID instruction, with unknown
2880 or unsupported feature bits cleared.
2882 Features like x2apic, for example, may not be present in the host cpu
2883 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
2884 emulated efficiently and thus not included here.
2886 The fields in each entry are defined as follows:
2888 function: the eax value used to obtain the entry
2889 index: the ecx value used to obtain the entry (for entries that are
2891 flags: an OR of zero or more of the following:
2892 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
2893 if the index field is valid
2894 KVM_CPUID_FLAG_STATEFUL_FUNC:
2895 if cpuid for this function returns different values for successive
2896 invocations; there will be several entries with the same function,
2897 all with this flag set
2898 KVM_CPUID_FLAG_STATE_READ_NEXT:
2899 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
2900 the first entry to be read by a cpu
2901 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
2902 this function/index combination
2904 4.89 KVM_S390_MEM_OP
2906 Capability: KVM_CAP_S390_MEM_OP
2909 Parameters: struct kvm_s390_mem_op (in)
2910 Returns: = 0 on success,
2911 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
2912 > 0 if an exception occurred while walking the page tables
2914 Read or write data from/to the logical (virtual) memory of a VCPU.
2916 Parameters are specified via the following structure:
2918 struct kvm_s390_mem_op {
2919 __u64 gaddr; /* the guest address */
2920 __u64 flags; /* flags */
2921 __u32 size; /* amount of bytes */
2922 __u32 op; /* type of operation */
2923 __u64 buf; /* buffer in userspace */
2924 __u8 ar; /* the access register number */
2925 __u8 reserved[31]; /* should be set to 0 */
2928 The type of operation is specified in the "op" field. It is either
2929 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or
2930 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The
2931 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check
2932 whether the corresponding memory access would create an access exception
2933 (without touching the data in the memory at the destination). In case an
2934 access exception occurred while walking the MMU tables of the guest, the
2935 ioctl returns a positive error number to indicate the type of exception.
2936 This exception is also raised directly at the corresponding VCPU if the
2937 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field.
2939 The start address of the memory region has to be specified in the "gaddr"
2940 field, and the length of the region in the "size" field. "buf" is the buffer
2941 supplied by the userspace application where the read data should be written
2942 to for KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written
2943 is stored for a KVM_S390_MEMOP_LOGICAL_WRITE. "buf" is unused and can be NULL
2944 when KVM_S390_MEMOP_F_CHECK_ONLY is specified. "ar" designates the access
2945 register number to be used.
2947 The "reserved" field is meant for future extensions. It is not used by
2948 KVM with the currently defined set of flags.
2950 4.90 KVM_S390_GET_SKEYS
2952 Capability: KVM_CAP_S390_SKEYS
2955 Parameters: struct kvm_s390_skeys
2956 Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage
2957 keys, negative value on error
2959 This ioctl is used to get guest storage key values on the s390
2960 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2962 struct kvm_s390_skeys {
2965 __u64 skeydata_addr;
2970 The start_gfn field is the number of the first guest frame whose storage keys
2973 The count field is the number of consecutive frames (starting from start_gfn)
2974 whose storage keys to get. The count field must be at least 1 and the maximum
2975 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2976 will cause the ioctl to return -EINVAL.
2978 The skeydata_addr field is the address to a buffer large enough to hold count
2979 bytes. This buffer will be filled with storage key data by the ioctl.
2981 4.91 KVM_S390_SET_SKEYS
2983 Capability: KVM_CAP_S390_SKEYS
2986 Parameters: struct kvm_s390_skeys
2987 Returns: 0 on success, negative value on error
2989 This ioctl is used to set guest storage key values on the s390
2990 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2991 See section on KVM_S390_GET_SKEYS for struct definition.
2993 The start_gfn field is the number of the first guest frame whose storage keys
2996 The count field is the number of consecutive frames (starting from start_gfn)
2997 whose storage keys to get. The count field must be at least 1 and the maximum
2998 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2999 will cause the ioctl to return -EINVAL.
3001 The skeydata_addr field is the address to a buffer containing count bytes of
3002 storage keys. Each byte in the buffer will be set as the storage key for a
3003 single frame starting at start_gfn for count frames.
3005 Note: If any architecturally invalid key value is found in the given data then
3006 the ioctl will return -EINVAL.
3010 Capability: KVM_CAP_S390_INJECT_IRQ
3013 Parameters: struct kvm_s390_irq (in)
3014 Returns: 0 on success, -1 on error
3016 EINVAL: interrupt type is invalid
3017 type is KVM_S390_SIGP_STOP and flag parameter is invalid value
3018 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger
3019 than the maximum of VCPUs
3020 EBUSY: type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped
3021 type is KVM_S390_SIGP_STOP and a stop irq is already pending
3022 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt
3025 Allows to inject an interrupt to the guest.
3027 Using struct kvm_s390_irq as a parameter allows
3028 to inject additional payload which is not
3029 possible via KVM_S390_INTERRUPT.
3031 Interrupt parameters are passed via kvm_s390_irq:
3033 struct kvm_s390_irq {
3036 struct kvm_s390_io_info io;
3037 struct kvm_s390_ext_info ext;
3038 struct kvm_s390_pgm_info pgm;
3039 struct kvm_s390_emerg_info emerg;
3040 struct kvm_s390_extcall_info extcall;
3041 struct kvm_s390_prefix_info prefix;
3042 struct kvm_s390_stop_info stop;
3043 struct kvm_s390_mchk_info mchk;
3048 type can be one of the following:
3050 KVM_S390_SIGP_STOP - sigp stop; parameter in .stop
3051 KVM_S390_PROGRAM_INT - program check; parameters in .pgm
3052 KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix
3053 KVM_S390_RESTART - restart; no parameters
3054 KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters
3055 KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters
3056 KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg
3057 KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall
3058 KVM_S390_MCHK - machine check interrupt; parameters in .mchk
3061 Note that the vcpu ioctl is asynchronous to vcpu execution.
3063 4.94 KVM_S390_GET_IRQ_STATE
3065 Capability: KVM_CAP_S390_IRQ_STATE
3068 Parameters: struct kvm_s390_irq_state (out)
3069 Returns: >= number of bytes copied into buffer,
3070 -EINVAL if buffer size is 0,
3071 -ENOBUFS if buffer size is too small to fit all pending interrupts,
3072 -EFAULT if the buffer address was invalid
3074 This ioctl allows userspace to retrieve the complete state of all currently
3075 pending interrupts in a single buffer. Use cases include migration
3076 and introspection. The parameter structure contains the address of a
3077 userspace buffer and its length:
3079 struct kvm_s390_irq_state {
3086 Userspace passes in the above struct and for each pending interrupt a
3087 struct kvm_s390_irq is copied to the provided buffer.
3089 If -ENOBUFS is returned the buffer provided was too small and userspace
3090 may retry with a bigger buffer.
3092 4.95 KVM_S390_SET_IRQ_STATE
3094 Capability: KVM_CAP_S390_IRQ_STATE
3097 Parameters: struct kvm_s390_irq_state (in)
3098 Returns: 0 on success,
3099 -EFAULT if the buffer address was invalid,
3100 -EINVAL for an invalid buffer length (see below),
3101 -EBUSY if there were already interrupts pending,
3102 errors occurring when actually injecting the
3103 interrupt. See KVM_S390_IRQ.
3105 This ioctl allows userspace to set the complete state of all cpu-local
3106 interrupts currently pending for the vcpu. It is intended for restoring
3107 interrupt state after a migration. The input parameter is a userspace buffer
3108 containing a struct kvm_s390_irq_state:
3110 struct kvm_s390_irq_state {
3116 The userspace memory referenced by buf contains a struct kvm_s390_irq
3117 for each interrupt to be injected into the guest.
3118 If one of the interrupts could not be injected for some reason the
3121 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
3122 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
3123 which is the maximum number of possibly pending cpu-local interrupts.
3127 Capability: KVM_CAP_X86_SMM
3131 Returns: 0 on success, -1 on error
3133 Queues an SMI on the thread's vcpu.
3135 4.97 KVM_CAP_PPC_MULTITCE
3137 Capability: KVM_CAP_PPC_MULTITCE
3141 This capability means the kernel is capable of handling hypercalls
3142 H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user
3143 space. This significantly accelerates DMA operations for PPC KVM guests.
3144 User space should expect that its handlers for these hypercalls
3145 are not going to be called if user space previously registered LIOBN
3146 in KVM (via KVM_CREATE_SPAPR_TCE or similar calls).
3148 In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest,
3149 user space might have to advertise it for the guest. For example,
3150 IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is
3151 present in the "ibm,hypertas-functions" device-tree property.
3153 The hypercalls mentioned above may or may not be processed successfully
3154 in the kernel based fast path. If they can not be handled by the kernel,
3155 they will get passed on to user space. So user space still has to have
3156 an implementation for these despite the in kernel acceleration.
3158 This capability is always enabled.
3160 4.98 KVM_CREATE_SPAPR_TCE_64
3162 Capability: KVM_CAP_SPAPR_TCE_64
3163 Architectures: powerpc
3165 Parameters: struct kvm_create_spapr_tce_64 (in)
3166 Returns: file descriptor for manipulating the created TCE table
3168 This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit
3169 windows, described in 4.62 KVM_CREATE_SPAPR_TCE
3171 This capability uses extended struct in ioctl interface:
3173 /* for KVM_CAP_SPAPR_TCE_64 */
3174 struct kvm_create_spapr_tce_64 {
3178 __u64 offset; /* in pages */
3179 __u64 size; /* in pages */
3182 The aim of extension is to support an additional bigger DMA window with
3183 a variable page size.
3184 KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and
3185 a bus offset of the corresponding DMA window, @size and @offset are numbers
3188 @flags are not used at the moment.
3190 The rest of functionality is identical to KVM_CREATE_SPAPR_TCE.
3192 4.99 KVM_REINJECT_CONTROL
3194 Capability: KVM_CAP_REINJECT_CONTROL
3197 Parameters: struct kvm_reinject_control (in)
3198 Returns: 0 on success,
3199 -EFAULT if struct kvm_reinject_control cannot be read,
3200 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier.
3202 i8254 (PIT) has two modes, reinject and !reinject. The default is reinject,
3203 where KVM queues elapsed i8254 ticks and monitors completion of interrupt from
3204 vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its
3205 interrupt whenever there isn't a pending interrupt from i8254.
3206 !reinject mode injects an interrupt as soon as a tick arrives.
3208 struct kvm_reinject_control {
3213 pit_reinject = 0 (!reinject mode) is recommended, unless running an old
3214 operating system that uses the PIT for timing (e.g. Linux 2.4.x).
3216 4.100 KVM_PPC_CONFIGURE_V3_MMU
3218 Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3
3221 Parameters: struct kvm_ppc_mmuv3_cfg (in)
3222 Returns: 0 on success,
3223 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read,
3224 -EINVAL if the configuration is invalid
3226 This ioctl controls whether the guest will use radix or HPT (hashed
3227 page table) translation, and sets the pointer to the process table for
3230 struct kvm_ppc_mmuv3_cfg {
3232 __u64 process_table;
3235 There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and
3236 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest
3237 to use radix tree translation, and if clear, to use HPT translation.
3238 KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest
3239 to be able to use the global TLB and SLB invalidation instructions;
3240 if clear, the guest may not use these instructions.
3242 The process_table field specifies the address and size of the guest
3243 process table, which is in the guest's space. This field is formatted
3244 as the second doubleword of the partition table entry, as defined in
3245 the Power ISA V3.00, Book III section 5.7.6.1.
3247 4.101 KVM_PPC_GET_RMMU_INFO
3249 Capability: KVM_CAP_PPC_RADIX_MMU
3252 Parameters: struct kvm_ppc_rmmu_info (out)
3253 Returns: 0 on success,
3254 -EFAULT if struct kvm_ppc_rmmu_info cannot be written,
3255 -EINVAL if no useful information can be returned
3257 This ioctl returns a structure containing two things: (a) a list
3258 containing supported radix tree geometries, and (b) a list that maps
3259 page sizes to put in the "AP" (actual page size) field for the tlbie
3260 (TLB invalidate entry) instruction.
3262 struct kvm_ppc_rmmu_info {
3263 struct kvm_ppc_radix_geom {
3268 __u32 ap_encodings[8];
3271 The geometries[] field gives up to 8 supported geometries for the
3272 radix page table, in terms of the log base 2 of the smallest page
3273 size, and the number of bits indexed at each level of the tree, from
3274 the PTE level up to the PGD level in that order. Any unused entries
3275 will have 0 in the page_shift field.
3277 The ap_encodings gives the supported page sizes and their AP field
3278 encodings, encoded with the AP value in the top 3 bits and the log
3279 base 2 of the page size in the bottom 6 bits.
3281 4.102 KVM_PPC_RESIZE_HPT_PREPARE
3283 Capability: KVM_CAP_SPAPR_RESIZE_HPT
3284 Architectures: powerpc
3286 Parameters: struct kvm_ppc_resize_hpt (in)
3287 Returns: 0 on successful completion,
3288 >0 if a new HPT is being prepared, the value is an estimated
3289 number of milliseconds until preparation is complete
3290 -EFAULT if struct kvm_reinject_control cannot be read,
3291 -EINVAL if the supplied shift or flags are invalid
3292 -ENOMEM if unable to allocate the new HPT
3293 -ENOSPC if there was a hash collision when moving existing
3294 HPT entries to the new HPT
3295 -EIO on other error conditions
3297 Used to implement the PAPR extension for runtime resizing of a guest's
3298 Hashed Page Table (HPT). Specifically this starts, stops or monitors
3299 the preparation of a new potential HPT for the guest, essentially
3300 implementing the H_RESIZE_HPT_PREPARE hypercall.
3302 If called with shift > 0 when there is no pending HPT for the guest,
3303 this begins preparation of a new pending HPT of size 2^(shift) bytes.
3304 It then returns a positive integer with the estimated number of
3305 milliseconds until preparation is complete.
3307 If called when there is a pending HPT whose size does not match that
3308 requested in the parameters, discards the existing pending HPT and
3309 creates a new one as above.
3311 If called when there is a pending HPT of the size requested, will:
3312 * If preparation of the pending HPT is already complete, return 0
3313 * If preparation of the pending HPT has failed, return an error
3314 code, then discard the pending HPT.
3315 * If preparation of the pending HPT is still in progress, return an
3316 estimated number of milliseconds until preparation is complete.
3318 If called with shift == 0, discards any currently pending HPT and
3319 returns 0 (i.e. cancels any in-progress preparation).
3321 flags is reserved for future expansion, currently setting any bits in
3322 flags will result in an -EINVAL.
3324 Normally this will be called repeatedly with the same parameters until
3325 it returns <= 0. The first call will initiate preparation, subsequent
3326 ones will monitor preparation until it completes or fails.
3328 struct kvm_ppc_resize_hpt {
3334 4.103 KVM_PPC_RESIZE_HPT_COMMIT
3336 Capability: KVM_CAP_SPAPR_RESIZE_HPT
3337 Architectures: powerpc
3339 Parameters: struct kvm_ppc_resize_hpt (in)
3340 Returns: 0 on successful completion,
3341 -EFAULT if struct kvm_reinject_control cannot be read,
3342 -EINVAL if the supplied shift or flags are invalid
3343 -ENXIO is there is no pending HPT, or the pending HPT doesn't
3344 have the requested size
3345 -EBUSY if the pending HPT is not fully prepared
3346 -ENOSPC if there was a hash collision when moving existing
3347 HPT entries to the new HPT
3348 -EIO on other error conditions
3350 Used to implement the PAPR extension for runtime resizing of a guest's
3351 Hashed Page Table (HPT). Specifically this requests that the guest be
3352 transferred to working with the new HPT, essentially implementing the
3353 H_RESIZE_HPT_COMMIT hypercall.
3355 This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has
3356 returned 0 with the same parameters. In other cases
3357 KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or
3358 -EBUSY, though others may be possible if the preparation was started,
3361 This will have undefined effects on the guest if it has not already
3362 placed itself in a quiescent state where no vcpu will make MMU enabled
3365 On succsful completion, the pending HPT will become the guest's active
3366 HPT and the previous HPT will be discarded.
3368 On failure, the guest will still be operating on its previous HPT.
3370 struct kvm_ppc_resize_hpt {
3376 5. The kvm_run structure
3377 ------------------------
3379 Application code obtains a pointer to the kvm_run structure by
3380 mmap()ing a vcpu fd. From that point, application code can control
3381 execution by changing fields in kvm_run prior to calling the KVM_RUN
3382 ioctl, and obtain information about the reason KVM_RUN returned by
3383 looking up structure members.
3387 __u8 request_interrupt_window;
3389 Request that KVM_RUN return when it becomes possible to inject external
3390 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
3392 __u8 immediate_exit;
3394 This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN
3395 exits immediately, returning -EINTR. In the common scenario where a
3396 signal is used to "kick" a VCPU out of KVM_RUN, this field can be used
3397 to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability.
3398 Rather than blocking the signal outside KVM_RUN, userspace can set up
3399 a signal handler that sets run->immediate_exit to a non-zero value.
3401 This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available.
3408 When KVM_RUN has returned successfully (return value 0), this informs
3409 application code why KVM_RUN has returned. Allowable values for this
3410 field are detailed below.
3412 __u8 ready_for_interrupt_injection;
3414 If request_interrupt_window has been specified, this field indicates
3415 an interrupt can be injected now with KVM_INTERRUPT.
3419 The value of the current interrupt flag. Only valid if in-kernel
3420 local APIC is not used.
3424 More architecture-specific flags detailing state of the VCPU that may
3425 affect the device's behavior. The only currently defined flag is
3426 KVM_RUN_X86_SMM, which is valid on x86 machines and is set if the
3427 VCPU is in system management mode.
3429 /* in (pre_kvm_run), out (post_kvm_run) */
3432 The value of the cr8 register. Only valid if in-kernel local APIC is
3433 not used. Both input and output.
3437 The value of the APIC BASE msr. Only valid if in-kernel local
3438 APIC is not used. Both input and output.
3441 /* KVM_EXIT_UNKNOWN */
3443 __u64 hardware_exit_reason;
3446 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
3447 reasons. Further architecture-specific information is available in
3448 hardware_exit_reason.
3450 /* KVM_EXIT_FAIL_ENTRY */
3452 __u64 hardware_entry_failure_reason;
3455 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
3456 to unknown reasons. Further architecture-specific information is
3457 available in hardware_entry_failure_reason.
3459 /* KVM_EXIT_EXCEPTION */
3469 #define KVM_EXIT_IO_IN 0
3470 #define KVM_EXIT_IO_OUT 1
3472 __u8 size; /* bytes */
3475 __u64 data_offset; /* relative to kvm_run start */
3478 If exit_reason is KVM_EXIT_IO, then the vcpu has
3479 executed a port I/O instruction which could not be satisfied by kvm.
3480 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
3481 where kvm expects application code to place the data for the next
3482 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
3484 /* KVM_EXIT_DEBUG */
3486 struct kvm_debug_exit_arch arch;
3489 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event
3490 for which architecture specific information is returned.
3500 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
3501 executed a memory-mapped I/O instruction which could not be satisfied
3502 by kvm. The 'data' member contains the written data if 'is_write' is
3503 true, and should be filled by application code otherwise.
3505 The 'data' member contains, in its first 'len' bytes, the value as it would
3506 appear if the VCPU performed a load or store of the appropriate width directly
3509 NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and
3510 KVM_EXIT_EPR the corresponding
3511 operations are complete (and guest state is consistent) only after userspace
3512 has re-entered the kernel with KVM_RUN. The kernel side will first finish
3513 incomplete operations and then check for pending signals. Userspace
3514 can re-enter the guest with an unmasked signal pending to complete
3517 /* KVM_EXIT_HYPERCALL */
3526 Unused. This was once used for 'hypercall to userspace'. To implement
3527 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390).
3528 Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
3530 /* KVM_EXIT_TPR_ACCESS */
3537 To be documented (KVM_TPR_ACCESS_REPORTING).
3539 /* KVM_EXIT_S390_SIEIC */
3542 __u64 mask; /* psw upper half */
3543 __u64 addr; /* psw lower half */
3550 /* KVM_EXIT_S390_RESET */
3551 #define KVM_S390_RESET_POR 1
3552 #define KVM_S390_RESET_CLEAR 2
3553 #define KVM_S390_RESET_SUBSYSTEM 4
3554 #define KVM_S390_RESET_CPU_INIT 8
3555 #define KVM_S390_RESET_IPL 16
3556 __u64 s390_reset_flags;
3560 /* KVM_EXIT_S390_UCONTROL */
3562 __u64 trans_exc_code;
3566 s390 specific. A page fault has occurred for a user controlled virtual
3567 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be
3568 resolved by the kernel.
3569 The program code and the translation exception code that were placed
3570 in the cpu's lowcore are presented here as defined by the z Architecture
3571 Principles of Operation Book in the Chapter for Dynamic Address Translation
3581 Deprecated - was used for 440 KVM.
3588 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
3589 hypercalls and exit with this exit struct that contains all the guest gprs.
3591 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
3592 Userspace can now handle the hypercall and when it's done modify the gprs as
3593 necessary. Upon guest entry all guest GPRs will then be replaced by the values
3596 /* KVM_EXIT_PAPR_HCALL */
3603 This is used on 64-bit PowerPC when emulating a pSeries partition,
3604 e.g. with the 'pseries' machine type in qemu. It occurs when the
3605 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
3606 contains the hypercall number (from the guest R3), and 'args' contains
3607 the arguments (from the guest R4 - R12). Userspace should put the
3608 return code in 'ret' and any extra returned values in args[].
3609 The possible hypercalls are defined in the Power Architecture Platform
3610 Requirements (PAPR) document available from www.power.org (free
3611 developer registration required to access it).
3613 /* KVM_EXIT_S390_TSCH */
3615 __u16 subchannel_id;
3616 __u16 subchannel_nr;
3623 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
3624 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
3625 interrupt for the target subchannel has been dequeued and subchannel_id,
3626 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
3627 interrupt. ipb is needed for instruction parameter decoding.
3634 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
3635 interrupt acknowledge path to the core. When the core successfully
3636 delivers an interrupt, it automatically populates the EPR register with
3637 the interrupt vector number and acknowledges the interrupt inside
3638 the interrupt controller.
3640 In case the interrupt controller lives in user space, we need to do
3641 the interrupt acknowledge cycle through it to fetch the next to be
3642 delivered interrupt vector using this exit.
3644 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
3645 external interrupt has just been delivered into the guest. User space
3646 should put the acknowledged interrupt vector into the 'epr' field.
3648 /* KVM_EXIT_SYSTEM_EVENT */
3650 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
3651 #define KVM_SYSTEM_EVENT_RESET 2
3652 #define KVM_SYSTEM_EVENT_CRASH 3
3657 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
3658 a system-level event using some architecture specific mechanism (hypercall
3659 or some special instruction). In case of ARM/ARM64, this is triggered using
3660 HVC instruction based PSCI call from the vcpu. The 'type' field describes
3661 the system-level event type. The 'flags' field describes architecture
3662 specific flags for the system-level event.
3664 Valid values for 'type' are:
3665 KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
3666 VM. Userspace is not obliged to honour this, and if it does honour
3667 this does not need to destroy the VM synchronously (ie it may call
3668 KVM_RUN again before shutdown finally occurs).
3669 KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
3670 As with SHUTDOWN, userspace can choose to ignore the request, or
3671 to schedule the reset to occur in the future and may call KVM_RUN again.
3672 KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest
3673 has requested a crash condition maintenance. Userspace can choose
3674 to ignore the request, or to gather VM memory core dump and/or
3675 reset/shutdown of the VM.
3677 /* KVM_EXIT_IOAPIC_EOI */
3682 Indicates that the VCPU's in-kernel local APIC received an EOI for a
3683 level-triggered IOAPIC interrupt. This exit only triggers when the
3684 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled);
3685 the userspace IOAPIC should process the EOI and retrigger the interrupt if
3686 it is still asserted. Vector is the LAPIC interrupt vector for which the
3689 struct kvm_hyperv_exit {
3690 #define KVM_EXIT_HYPERV_SYNIC 1
3691 #define KVM_EXIT_HYPERV_HCALL 2
3707 /* KVM_EXIT_HYPERV */
3708 struct kvm_hyperv_exit hyperv;
3709 Indicates that the VCPU exits into userspace to process some tasks
3710 related to Hyper-V emulation.
3711 Valid values for 'type' are:
3712 KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about
3713 Hyper-V SynIC state change. Notification is used to remap SynIC
3714 event/message pages and to enable/disable SynIC messages/events processing
3717 /* Fix the size of the union. */
3722 * shared registers between kvm and userspace.
3723 * kvm_valid_regs specifies the register classes set by the host
3724 * kvm_dirty_regs specified the register classes dirtied by userspace
3725 * struct kvm_sync_regs is architecture specific, as well as the
3726 * bits for kvm_valid_regs and kvm_dirty_regs
3728 __u64 kvm_valid_regs;
3729 __u64 kvm_dirty_regs;
3731 struct kvm_sync_regs regs;
3735 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
3736 certain guest registers without having to call SET/GET_*REGS. Thus we can
3737 avoid some system call overhead if userspace has to handle the exit.
3738 Userspace can query the validity of the structure by checking
3739 kvm_valid_regs for specific bits. These bits are architecture specific
3740 and usually define the validity of a groups of registers. (e.g. one bit
3741 for general purpose registers)
3743 Please note that the kernel is allowed to use the kvm_run structure as the
3744 primary storage for certain register types. Therefore, the kernel may use the
3745 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
3751 6. Capabilities that can be enabled on vCPUs
3752 --------------------------------------------
3754 There are certain capabilities that change the behavior of the virtual CPU or
3755 the virtual machine when enabled. To enable them, please see section 4.37.
3756 Below you can find a list of capabilities and what their effect on the vCPU or
3757 the virtual machine is when enabling them.
3759 The following information is provided along with the description:
3761 Architectures: which instruction set architectures provide this ioctl.
3762 x86 includes both i386 and x86_64.
3764 Target: whether this is a per-vcpu or per-vm capability.
3766 Parameters: what parameters are accepted by the capability.
3768 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3769 are not detailed, but errors with specific meanings are.
3777 Returns: 0 on success; -1 on error
3779 This capability enables interception of OSI hypercalls that otherwise would
3780 be treated as normal system calls to be injected into the guest. OSI hypercalls
3781 were invented by Mac-on-Linux to have a standardized communication mechanism
3782 between the guest and the host.
3784 When this capability is enabled, KVM_EXIT_OSI can occur.
3787 6.2 KVM_CAP_PPC_PAPR
3792 Returns: 0 on success; -1 on error
3794 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
3795 done using the hypercall instruction "sc 1".
3797 It also sets the guest privilege level to "supervisor" mode. Usually the guest
3798 runs in "hypervisor" privilege mode with a few missing features.
3800 In addition to the above, it changes the semantics of SDR1. In this mode, the
3801 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
3802 HTAB invisible to the guest.
3804 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
3811 Parameters: args[0] is the address of a struct kvm_config_tlb
3812 Returns: 0 on success; -1 on error
3814 struct kvm_config_tlb {
3821 Configures the virtual CPU's TLB array, establishing a shared memory area
3822 between userspace and KVM. The "params" and "array" fields are userspace
3823 addresses of mmu-type-specific data structures. The "array_len" field is an
3824 safety mechanism, and should be set to the size in bytes of the memory that
3825 userspace has reserved for the array. It must be at least the size dictated
3826 by "mmu_type" and "params".
3828 While KVM_RUN is active, the shared region is under control of KVM. Its
3829 contents are undefined, and any modification by userspace results in
3830 boundedly undefined behavior.
3832 On return from KVM_RUN, the shared region will reflect the current state of
3833 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
3834 to tell KVM which entries have been changed, prior to calling KVM_RUN again
3837 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
3838 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
3839 - The "array" field points to an array of type "struct
3840 kvm_book3e_206_tlb_entry".
3841 - The array consists of all entries in the first TLB, followed by all
3842 entries in the second TLB.
3843 - Within a TLB, entries are ordered first by increasing set number. Within a
3844 set, entries are ordered by way (increasing ESEL).
3845 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
3846 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
3847 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
3848 hardware ignores this value for TLB0.
3850 6.4 KVM_CAP_S390_CSS_SUPPORT
3855 Returns: 0 on success; -1 on error
3857 This capability enables support for handling of channel I/O instructions.
3859 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
3860 handled in-kernel, while the other I/O instructions are passed to userspace.
3862 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
3863 SUBCHANNEL intercepts.
3865 Note that even though this capability is enabled per-vcpu, the complete
3866 virtual machine is affected.
3872 Parameters: args[0] defines whether the proxy facility is active
3873 Returns: 0 on success; -1 on error
3875 This capability enables or disables the delivery of interrupts through the
3876 external proxy facility.
3878 When enabled (args[0] != 0), every time the guest gets an external interrupt
3879 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
3880 to receive the topmost interrupt vector.
3882 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
3884 When this capability is enabled, KVM_EXIT_EPR can occur.
3886 6.6 KVM_CAP_IRQ_MPIC
3889 Parameters: args[0] is the MPIC device fd
3890 args[1] is the MPIC CPU number for this vcpu
3892 This capability connects the vcpu to an in-kernel MPIC device.
3894 6.7 KVM_CAP_IRQ_XICS
3898 Parameters: args[0] is the XICS device fd
3899 args[1] is the XICS CPU number (server ID) for this vcpu
3901 This capability connects the vcpu to an in-kernel XICS device.
3903 6.8 KVM_CAP_S390_IRQCHIP
3909 This capability enables the in-kernel irqchip for s390. Please refer to
3910 "4.24 KVM_CREATE_IRQCHIP" for details.
3912 6.9 KVM_CAP_MIPS_FPU
3916 Parameters: args[0] is reserved for future use (should be 0).
3918 This capability allows the use of the host Floating Point Unit by the guest. It
3919 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is
3920 done the KVM_REG_MIPS_FPR_* and KVM_REG_MIPS_FCR_* registers can be accessed
3921 (depending on the current guest FPU register mode), and the Status.FR,
3922 Config5.FRE bits are accessible via the KVM API and also from the guest,
3923 depending on them being supported by the FPU.
3925 6.10 KVM_CAP_MIPS_MSA
3929 Parameters: args[0] is reserved for future use (should be 0).
3931 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest.
3932 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest.
3933 Once this is done the KVM_REG_MIPS_VEC_* and KVM_REG_MIPS_MSA_* registers can be
3934 accessed, and the Config5.MSAEn bit is accessible via the KVM API and also from
3937 7. Capabilities that can be enabled on VMs
3938 ------------------------------------------
3940 There are certain capabilities that change the behavior of the virtual
3941 machine when enabled. To enable them, please see section 4.37. Below
3942 you can find a list of capabilities and what their effect on the VM
3943 is when enabling them.
3945 The following information is provided along with the description:
3947 Architectures: which instruction set architectures provide this ioctl.
3948 x86 includes both i386 and x86_64.
3950 Parameters: what parameters are accepted by the capability.
3952 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3953 are not detailed, but errors with specific meanings are.
3956 7.1 KVM_CAP_PPC_ENABLE_HCALL
3959 Parameters: args[0] is the sPAPR hcall number
3960 args[1] is 0 to disable, 1 to enable in-kernel handling
3962 This capability controls whether individual sPAPR hypercalls (hcalls)
3963 get handled by the kernel or not. Enabling or disabling in-kernel
3964 handling of an hcall is effective across the VM. On creation, an
3965 initial set of hcalls are enabled for in-kernel handling, which
3966 consists of those hcalls for which in-kernel handlers were implemented
3967 before this capability was implemented. If disabled, the kernel will
3968 not to attempt to handle the hcall, but will always exit to userspace
3969 to handle it. Note that it may not make sense to enable some and
3970 disable others of a group of related hcalls, but KVM does not prevent
3971 userspace from doing that.
3973 If the hcall number specified is not one that has an in-kernel
3974 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
3977 7.2 KVM_CAP_S390_USER_SIGP
3982 This capability controls which SIGP orders will be handled completely in user
3983 space. With this capability enabled, all fast orders will be handled completely
3989 - CONDITIONAL EMERGENCY SIGNAL
3991 All other orders will be handled completely in user space.
3993 Only privileged operation exceptions will be checked for in the kernel (or even
3994 in the hardware prior to interception). If this capability is not enabled, the
3995 old way of handling SIGP orders is used (partially in kernel and user space).
3997 7.3 KVM_CAP_S390_VECTOR_REGISTERS
4001 Returns: 0 on success, negative value on error
4003 Allows use of the vector registers introduced with z13 processor, and
4004 provides for the synchronization between host and user space. Will
4005 return -EINVAL if the machine does not support vectors.
4007 7.4 KVM_CAP_S390_USER_STSI
4012 This capability allows post-handlers for the STSI instruction. After
4013 initial handling in the kernel, KVM exits to user space with
4014 KVM_EXIT_S390_STSI to allow user space to insert further data.
4016 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
4027 @addr - guest address of STSI SYSIB
4031 @ar - access register number
4033 KVM handlers should exit to userspace with rc = -EREMOTE.
4035 7.5 KVM_CAP_SPLIT_IRQCHIP
4038 Parameters: args[0] - number of routes reserved for userspace IOAPICs
4039 Returns: 0 on success, -1 on error
4041 Create a local apic for each processor in the kernel. This can be used
4042 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the
4043 IOAPIC and PIC (and also the PIT, even though this has to be enabled
4046 This capability also enables in kernel routing of interrupt requests;
4047 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are
4048 used in the IRQ routing table. The first args[0] MSI routes are reserved
4049 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes,
4050 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace.
4052 Fails if VCPU has already been created, or if the irqchip is already in the
4053 kernel (i.e. KVM_CREATE_IRQCHIP has already been called).
4060 Allows use of runtime-instrumentation introduced with zEC12 processor.
4061 Will return -EINVAL if the machine does not support runtime-instrumentation.
4062 Will return -EBUSY if a VCPU has already been created.
4064 7.7 KVM_CAP_X2APIC_API
4067 Parameters: args[0] - features that should be enabled
4068 Returns: 0 on success, -EINVAL when args[0] contains invalid features
4070 Valid feature flags in args[0] are
4072 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0)
4073 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1)
4075 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of
4076 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC,
4077 allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their
4078 respective sections.
4080 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work
4081 in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff
4082 as a broadcast even in x2APIC mode in order to support physical x2APIC
4083 without interrupt remapping. This is undesirable in logical mode,
4084 where 0xff represents CPUs 0-7 in cluster 0.
4086 7.8 KVM_CAP_S390_USER_INSTR0
4091 With this capability enabled, all illegal instructions 0x0000 (2 bytes) will
4092 be intercepted and forwarded to user space. User space can use this
4093 mechanism e.g. to realize 2-byte software breakpoints. The kernel will
4094 not inject an operating exception for these instructions, user space has
4095 to take care of that.
4097 This capability can be enabled dynamically even if VCPUs were already
4098 created and are running.
4100 8. Other capabilities.
4101 ----------------------
4103 This section lists capabilities that give information about other
4104 features of the KVM implementation.
4106 8.1 KVM_CAP_PPC_HWRNG
4110 This capability, if KVM_CHECK_EXTENSION indicates that it is
4111 available, means that that the kernel has an implementation of the
4112 H_RANDOM hypercall backed by a hardware random-number generator.
4113 If present, the kernel H_RANDOM handler can be enabled for guest use
4114 with the KVM_CAP_PPC_ENABLE_HCALL capability.
4116 8.2 KVM_CAP_HYPERV_SYNIC
4119 This capability, if KVM_CHECK_EXTENSION indicates that it is
4120 available, means that that the kernel has an implementation of the
4121 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
4122 used to support Windows Hyper-V based guest paravirt drivers(VMBus).
4124 In order to use SynIC, it has to be activated by setting this
4125 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
4126 will disable the use of APIC hardware virtualization even if supported
4127 by the CPU, as it's incompatible with SynIC auto-EOI behavior.
4129 8.3 KVM_CAP_PPC_RADIX_MMU
4133 This capability, if KVM_CHECK_EXTENSION indicates that it is
4134 available, means that that the kernel can support guests using the
4135 radix MMU defined in Power ISA V3.00 (as implemented in the POWER9
4138 8.4 KVM_CAP_PPC_HASH_MMU_V3
4142 This capability, if KVM_CHECK_EXTENSION indicates that it is
4143 available, means that that the kernel can support guests using the
4144 hashed page table MMU defined in Power ISA V3.00 (as implemented in
4145 the POWER9 processor), including in-memory segment tables.