1 The Definitive KVM (Kernel-based Virtual Machine) API Documentation
2 ===================================================================
7 The kvm API is a set of ioctls that are issued to control various aspects
8 of a virtual machine. The ioctls belong to three classes
10 - System ioctls: These query and set global attributes which affect the
11 whole kvm subsystem. In addition a system ioctl is used to create
14 - VM ioctls: These query and set attributes that affect an entire virtual
15 machine, for example memory layout. In addition a VM ioctl is used to
16 create virtual cpus (vcpus).
18 Only run VM ioctls from the same process (address space) that was used
21 - vcpu ioctls: These query and set attributes that control the operation
22 of a single virtual cpu.
24 Only run vcpu ioctls from the same thread that was used to create the
31 The kvm API is centered around file descriptors. An initial
32 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle
33 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this
34 handle will create a VM file descriptor which can be used to issue VM
35 ioctls. A KVM_CREATE_VCPU ioctl on a VM fd will create a virtual cpu
36 and return a file descriptor pointing to it. Finally, ioctls on a vcpu
37 fd can be used to control the vcpu, including the important task of
38 actually running guest code.
40 In general file descriptors can be migrated among processes by means
41 of fork() and the SCM_RIGHTS facility of unix domain socket. These
42 kinds of tricks are explicitly not supported by kvm. While they will
43 not cause harm to the host, their actual behavior is not guaranteed by
44 the API. The only supported use is one virtual machine per process,
45 and one vcpu per thread.
51 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward
52 incompatible change are allowed. However, there is an extension
53 facility that allows backward-compatible extensions to the API to be
56 The extension mechanism is not based on the Linux version number.
57 Instead, kvm defines extension identifiers and a facility to query
58 whether a particular extension identifier is available. If it is, a
59 set of ioctls is available for application use.
65 This section describes ioctls that can be used to control kvm guests.
66 For each ioctl, the following information is provided along with a
69 Capability: which KVM extension provides this ioctl. Can be 'basic',
70 which means that is will be provided by any kernel that supports
71 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which
72 means availability needs to be checked with KVM_CHECK_EXTENSION
73 (see section 4.4), or 'none' which means that while not all kernels
74 support this ioctl, there's no capability bit to check its
75 availability: for kernels that don't support the ioctl,
76 the ioctl returns -ENOTTY.
78 Architectures: which instruction set architectures provide this ioctl.
79 x86 includes both i386 and x86_64.
81 Type: system, vm, or vcpu.
83 Parameters: what parameters are accepted by the ioctl.
85 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
86 are not detailed, but errors with specific meanings are.
89 4.1 KVM_GET_API_VERSION
95 Returns: the constant KVM_API_VERSION (=12)
97 This identifies the API version as the stable kvm API. It is not
98 expected that this number will change. However, Linux 2.6.20 and
99 2.6.21 report earlier versions; these are not documented and not
100 supported. Applications should refuse to run if KVM_GET_API_VERSION
101 returns a value other than 12. If this check passes, all ioctls
102 described as 'basic' will be available.
110 Parameters: machine type identifier (KVM_VM_*)
111 Returns: a VM fd that can be used to control the new virtual machine.
113 The new VM has no virtual cpus and no memory. An mmap() of a VM fd
114 will access the virtual machine's physical address space; offset zero
115 corresponds to guest physical address zero. Use of mmap() on a VM fd
116 is discouraged if userspace memory allocation (KVM_CAP_USER_MEMORY) is
118 You most certainly want to use 0 as machine type.
120 In order to create user controlled virtual machines on S390, check
121 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
122 privileged user (CAP_SYS_ADMIN).
125 4.3 KVM_GET_MSR_INDEX_LIST
130 Parameters: struct kvm_msr_list (in/out)
131 Returns: 0 on success; -1 on error
133 E2BIG: the msr index list is to be to fit in the array specified by
136 struct kvm_msr_list {
137 __u32 nmsrs; /* number of msrs in entries */
141 This ioctl returns the guest msrs that are supported. The list varies
142 by kvm version and host processor, but does not change otherwise. The
143 user fills in the size of the indices array in nmsrs, and in return
144 kvm adjusts nmsrs to reflect the actual number of msrs and fills in
145 the indices array with their numbers.
147 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
148 not returned in the MSR list, as different vcpus can have a different number
149 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
152 4.4 KVM_CHECK_EXTENSION
154 Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
156 Type: system ioctl, vm ioctl
157 Parameters: extension identifier (KVM_CAP_*)
158 Returns: 0 if unsupported; 1 (or some other positive integer) if supported
160 The API allows the application to query about extensions to the core
161 kvm API. Userspace passes an extension identifier (an integer) and
162 receives an integer that describes the extension availability.
163 Generally 0 means no and 1 means yes, but some extensions may report
164 additional information in the integer return value.
166 Based on their initialization different VMs may have different capabilities.
167 It is thus encouraged to use the vm ioctl to query for capabilities (available
168 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
170 4.5 KVM_GET_VCPU_MMAP_SIZE
176 Returns: size of vcpu mmap area, in bytes
178 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
179 memory region. This ioctl returns the size of that region. See the
180 KVM_RUN documentation for details.
183 4.6 KVM_SET_MEMORY_REGION
188 Parameters: struct kvm_memory_region (in)
189 Returns: 0 on success, -1 on error
191 This ioctl is obsolete and has been removed.
199 Parameters: vcpu id (apic id on x86)
200 Returns: vcpu fd on success, -1 on error
202 This API adds a vcpu to a virtual machine. The vcpu id is a small integer
203 in the range [0, max_vcpus).
205 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
206 the KVM_CHECK_EXTENSION ioctl() at run-time.
207 The maximum possible value for max_vcpus can be retrieved using the
208 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
210 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
212 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
213 same as the value returned from KVM_CAP_NR_VCPUS.
215 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
216 threads in one or more virtual CPU cores. (This is because the
217 hardware requires all the hardware threads in a CPU core to be in the
218 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
219 of vcpus per virtual core (vcore). The vcore id is obtained by
220 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
221 given vcore will always be in the same physical core as each other
222 (though that might be a different physical core from time to time).
223 Userspace can control the threading (SMT) mode of the guest by its
224 allocation of vcpu ids. For example, if userspace wants
225 single-threaded guest vcpus, it should make all vcpu ids be a multiple
226 of the number of vcpus per vcore.
228 For virtual cpus that have been created with S390 user controlled virtual
229 machines, the resulting vcpu fd can be memory mapped at page offset
230 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
231 cpu's hardware control block.
234 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
239 Parameters: struct kvm_dirty_log (in/out)
240 Returns: 0 on success, -1 on error
242 /* for KVM_GET_DIRTY_LOG */
243 struct kvm_dirty_log {
247 void __user *dirty_bitmap; /* one bit per page */
252 Given a memory slot, return a bitmap containing any pages dirtied
253 since the last call to this ioctl. Bit 0 is the first page in the
254 memory slot. Ensure the entire structure is cleared to avoid padding
258 4.9 KVM_SET_MEMORY_ALIAS
263 Parameters: struct kvm_memory_alias (in)
264 Returns: 0 (success), -1 (error)
266 This ioctl is obsolete and has been removed.
275 Returns: 0 on success, -1 on error
277 EINTR: an unmasked signal is pending
279 This ioctl is used to run a guest virtual cpu. While there are no
280 explicit parameters, there is an implicit parameter block that can be
281 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
282 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
283 kvm_run' (see below).
289 Architectures: all except ARM, arm64
291 Parameters: struct kvm_regs (out)
292 Returns: 0 on success, -1 on error
294 Reads the general purpose registers from the vcpu.
298 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
299 __u64 rax, rbx, rcx, rdx;
300 __u64 rsi, rdi, rsp, rbp;
301 __u64 r8, r9, r10, r11;
302 __u64 r12, r13, r14, r15;
308 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
319 Architectures: all except ARM, arm64
321 Parameters: struct kvm_regs (in)
322 Returns: 0 on success, -1 on error
324 Writes the general purpose registers into the vcpu.
326 See KVM_GET_REGS for the data structure.
332 Architectures: x86, ppc
334 Parameters: struct kvm_sregs (out)
335 Returns: 0 on success, -1 on error
337 Reads special registers from the vcpu.
341 struct kvm_segment cs, ds, es, fs, gs, ss;
342 struct kvm_segment tr, ldt;
343 struct kvm_dtable gdt, idt;
344 __u64 cr0, cr2, cr3, cr4, cr8;
347 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
350 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
352 interrupt_bitmap is a bitmap of pending external interrupts. At most
353 one bit may be set. This interrupt has been acknowledged by the APIC
354 but not yet injected into the cpu core.
360 Architectures: x86, ppc
362 Parameters: struct kvm_sregs (in)
363 Returns: 0 on success, -1 on error
365 Writes special registers into the vcpu. See KVM_GET_SREGS for the
374 Parameters: struct kvm_translation (in/out)
375 Returns: 0 on success, -1 on error
377 Translates a virtual address according to the vcpu's current address
380 struct kvm_translation {
382 __u64 linear_address;
385 __u64 physical_address;
396 Architectures: x86, ppc, mips
398 Parameters: struct kvm_interrupt (in)
399 Returns: 0 on success, -1 on error
401 Queues a hardware interrupt vector to be injected. This is only
402 useful if in-kernel local APIC or equivalent is not used.
404 /* for KVM_INTERRUPT */
405 struct kvm_interrupt {
412 Note 'irq' is an interrupt vector, not an interrupt pin or line.
416 Queues an external interrupt to be injected. This ioctl is overleaded
417 with 3 different irq values:
421 This injects an edge type external interrupt into the guest once it's ready
422 to receive interrupts. When injected, the interrupt is done.
424 b) KVM_INTERRUPT_UNSET
426 This unsets any pending interrupt.
428 Only available with KVM_CAP_PPC_UNSET_IRQ.
430 c) KVM_INTERRUPT_SET_LEVEL
432 This injects a level type external interrupt into the guest context. The
433 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
436 Only available with KVM_CAP_PPC_IRQ_LEVEL.
438 Note that any value for 'irq' other than the ones stated above is invalid
439 and incurs unexpected behavior.
443 Queues an external interrupt to be injected into the virtual CPU. A negative
444 interrupt number dequeues the interrupt.
455 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
463 Parameters: struct kvm_msrs (in/out)
464 Returns: 0 on success, -1 on error
466 Reads model-specific registers from the vcpu. Supported msr indices can
467 be obtained using KVM_GET_MSR_INDEX_LIST.
470 __u32 nmsrs; /* number of msrs in entries */
473 struct kvm_msr_entry entries[0];
476 struct kvm_msr_entry {
482 Application code should set the 'nmsrs' member (which indicates the
483 size of the entries array) and the 'index' member of each array entry.
484 kvm will fill in the 'data' member.
492 Parameters: struct kvm_msrs (in)
493 Returns: 0 on success, -1 on error
495 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
498 Application code should set the 'nmsrs' member (which indicates the
499 size of the entries array), and the 'index' and 'data' members of each
508 Parameters: struct kvm_cpuid (in)
509 Returns: 0 on success, -1 on error
511 Defines the vcpu responses to the cpuid instruction. Applications
512 should use the KVM_SET_CPUID2 ioctl if available.
515 struct kvm_cpuid_entry {
524 /* for KVM_SET_CPUID */
528 struct kvm_cpuid_entry entries[0];
532 4.21 KVM_SET_SIGNAL_MASK
537 Parameters: struct kvm_signal_mask (in)
538 Returns: 0 on success, -1 on error
540 Defines which signals are blocked during execution of KVM_RUN. This
541 signal mask temporarily overrides the threads signal mask. Any
542 unblocked signal received (except SIGKILL and SIGSTOP, which retain
543 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
545 Note the signal will only be delivered if not blocked by the original
548 /* for KVM_SET_SIGNAL_MASK */
549 struct kvm_signal_mask {
560 Parameters: struct kvm_fpu (out)
561 Returns: 0 on success, -1 on error
563 Reads the floating point state from the vcpu.
565 /* for KVM_GET_FPU and KVM_SET_FPU */
570 __u8 ftwx; /* in fxsave format */
586 Parameters: struct kvm_fpu (in)
587 Returns: 0 on success, -1 on error
589 Writes the floating point state to the vcpu.
591 /* for KVM_GET_FPU and KVM_SET_FPU */
596 __u8 ftwx; /* in fxsave format */
607 4.24 KVM_CREATE_IRQCHIP
609 Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
610 Architectures: x86, ARM, arm64, s390
613 Returns: 0 on success, -1 on error
615 Creates an interrupt controller model in the kernel.
616 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
617 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
618 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
619 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of
620 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
621 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
622 On s390, a dummy irq routing table is created.
624 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
625 before KVM_CREATE_IRQCHIP can be used.
630 Capability: KVM_CAP_IRQCHIP
631 Architectures: x86, arm, arm64
633 Parameters: struct kvm_irq_level
634 Returns: 0 on success, -1 on error
636 Sets the level of a GSI input to the interrupt controller model in the kernel.
637 On some architectures it is required that an interrupt controller model has
638 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
639 interrupts require the level to be set to 1 and then back to 0.
641 On real hardware, interrupt pins can be active-low or active-high. This
642 does not matter for the level field of struct kvm_irq_level: 1 always
643 means active (asserted), 0 means inactive (deasserted).
645 x86 allows the operating system to program the interrupt polarity
646 (active-low/active-high) for level-triggered interrupts, and KVM used
647 to consider the polarity. However, due to bitrot in the handling of
648 active-low interrupts, the above convention is now valid on x86 too.
649 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
650 should not present interrupts to the guest as active-low unless this
651 capability is present (or unless it is not using the in-kernel irqchip,
655 ARM/arm64 can signal an interrupt either at the CPU level, or at the
656 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
657 use PPIs designated for specific cpus. The irq field is interpreted
660 Â bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 |
661 field: | irq_type | vcpu_index | irq_id |
663 The irq_type field has the following values:
664 - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
665 - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
666 (the vcpu_index field is ignored)
667 - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
669 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
671 In both cases, level is used to assert/deassert the line.
673 struct kvm_irq_level {
676 __s32 status; /* not used for KVM_IRQ_LEVEL */
678 __u32 level; /* 0 or 1 */
684 Capability: KVM_CAP_IRQCHIP
687 Parameters: struct kvm_irqchip (in/out)
688 Returns: 0 on success, -1 on error
690 Reads the state of a kernel interrupt controller created with
691 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
694 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
697 char dummy[512]; /* reserving space */
698 struct kvm_pic_state pic;
699 struct kvm_ioapic_state ioapic;
706 Capability: KVM_CAP_IRQCHIP
709 Parameters: struct kvm_irqchip (in)
710 Returns: 0 on success, -1 on error
712 Sets the state of a kernel interrupt controller created with
713 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
716 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
719 char dummy[512]; /* reserving space */
720 struct kvm_pic_state pic;
721 struct kvm_ioapic_state ioapic;
726 4.28 KVM_XEN_HVM_CONFIG
728 Capability: KVM_CAP_XEN_HVM
731 Parameters: struct kvm_xen_hvm_config (in)
732 Returns: 0 on success, -1 on error
734 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
735 page, and provides the starting address and size of the hypercall
736 blobs in userspace. When the guest writes the MSR, kvm copies one
737 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
740 struct kvm_xen_hvm_config {
753 Capability: KVM_CAP_ADJUST_CLOCK
756 Parameters: struct kvm_clock_data (out)
757 Returns: 0 on success, -1 on error
759 Gets the current timestamp of kvmclock as seen by the current guest. In
760 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
763 struct kvm_clock_data {
764 __u64 clock; /* kvmclock current value */
772 Capability: KVM_CAP_ADJUST_CLOCK
775 Parameters: struct kvm_clock_data (in)
776 Returns: 0 on success, -1 on error
778 Sets the current timestamp of kvmclock to the value specified in its parameter.
779 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
782 struct kvm_clock_data {
783 __u64 clock; /* kvmclock current value */
789 4.31 KVM_GET_VCPU_EVENTS
791 Capability: KVM_CAP_VCPU_EVENTS
792 Extended by: KVM_CAP_INTR_SHADOW
795 Parameters: struct kvm_vcpu_event (out)
796 Returns: 0 on success, -1 on error
798 Gets currently pending exceptions, interrupts, and NMIs as well as related
801 struct kvm_vcpu_events {
825 KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that
826 interrupt.shadow contains a valid state. Otherwise, this field is undefined.
829 4.32 KVM_SET_VCPU_EVENTS
831 Capability: KVM_CAP_VCPU_EVENTS
832 Extended by: KVM_CAP_INTR_SHADOW
835 Parameters: struct kvm_vcpu_event (in)
836 Returns: 0 on success, -1 on error
838 Set pending exceptions, interrupts, and NMIs as well as related states of the
841 See KVM_GET_VCPU_EVENTS for the data structure.
843 Fields that may be modified asynchronously by running VCPUs can be excluded
844 from the update. These fields are nmi.pending and sipi_vector. Keep the
845 corresponding bits in the flags field cleared to suppress overwriting the
846 current in-kernel state. The bits are:
848 KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel
849 KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector
851 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
852 the flags field to signal that interrupt.shadow contains a valid state and
853 shall be written into the VCPU.
856 4.33 KVM_GET_DEBUGREGS
858 Capability: KVM_CAP_DEBUGREGS
861 Parameters: struct kvm_debugregs (out)
862 Returns: 0 on success, -1 on error
864 Reads debug registers from the vcpu.
866 struct kvm_debugregs {
875 4.34 KVM_SET_DEBUGREGS
877 Capability: KVM_CAP_DEBUGREGS
880 Parameters: struct kvm_debugregs (in)
881 Returns: 0 on success, -1 on error
883 Writes debug registers into the vcpu.
885 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
886 yet and must be cleared on entry.
889 4.35 KVM_SET_USER_MEMORY_REGION
891 Capability: KVM_CAP_USER_MEM
894 Parameters: struct kvm_userspace_memory_region (in)
895 Returns: 0 on success, -1 on error
897 struct kvm_userspace_memory_region {
900 __u64 guest_phys_addr;
901 __u64 memory_size; /* bytes */
902 __u64 userspace_addr; /* start of the userspace allocated memory */
905 /* for kvm_memory_region::flags */
906 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
907 #define KVM_MEM_READONLY (1UL << 1)
909 This ioctl allows the user to create or modify a guest physical memory
910 slot. When changing an existing slot, it may be moved in the guest
911 physical memory space, or its flags may be modified. It may not be
912 resized. Slots may not overlap in guest physical address space.
914 Memory for the region is taken starting at the address denoted by the
915 field userspace_addr, which must point at user addressable memory for
916 the entire memory slot size. Any object may back this memory, including
917 anonymous memory, ordinary files, and hugetlbfs.
919 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
920 be identical. This allows large pages in the guest to be backed by large
923 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
924 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
925 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
926 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
927 to make a new slot read-only. In this case, writes to this memory will be
928 posted to userspace as KVM_EXIT_MMIO exits.
930 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
931 the memory region are automatically reflected into the guest. For example, an
932 mmap() that affects the region will be made visible immediately. Another
933 example is madvise(MADV_DROP).
935 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl.
936 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory
937 allocation and is deprecated.
940 4.36 KVM_SET_TSS_ADDR
942 Capability: KVM_CAP_SET_TSS_ADDR
945 Parameters: unsigned long tss_address (in)
946 Returns: 0 on success, -1 on error
948 This ioctl defines the physical address of a three-page region in the guest
949 physical address space. The region must be within the first 4GB of the
950 guest physical address space and must not conflict with any memory slot
951 or any mmio address. The guest may malfunction if it accesses this memory
954 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
955 because of a quirk in the virtualization implementation (see the internals
956 documentation when it pops into existence).
961 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM
962 Architectures: ppc, s390
963 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM)
964 Parameters: struct kvm_enable_cap (in)
965 Returns: 0 on success; -1 on error
967 +Not all extensions are enabled by default. Using this ioctl the application
968 can enable an extension, making it available to the guest.
970 On systems that do not support this ioctl, it always fails. On systems that
971 do support it, it only works for extensions that are supported for enablement.
973 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
976 struct kvm_enable_cap {
980 The capability that is supposed to get enabled.
984 A bitfield indicating future enhancements. Has to be 0 for now.
988 Arguments for enabling a feature. If a feature needs initial values to
989 function properly, this is the place to put them.
994 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
995 for vm-wide capabilities.
997 4.38 KVM_GET_MP_STATE
999 Capability: KVM_CAP_MP_STATE
1000 Architectures: x86, s390
1002 Parameters: struct kvm_mp_state (out)
1003 Returns: 0 on success; -1 on error
1005 struct kvm_mp_state {
1009 Returns the vcpu's current "multiprocessing state" (though also valid on
1010 uniprocessor guests).
1012 Possible values are:
1014 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86]
1015 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP)
1016 which has not yet received an INIT signal [x86]
1017 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is
1018 now ready for a SIPI [x86]
1019 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and
1020 is waiting for an interrupt [x86]
1021 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector
1022 accessible via KVM_GET_VCPU_EVENTS) [x86]
1023 - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390]
1024 - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390]
1025 - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted)
1027 - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state
1030 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1031 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1032 these architectures.
1035 4.39 KVM_SET_MP_STATE
1037 Capability: KVM_CAP_MP_STATE
1038 Architectures: x86, s390
1040 Parameters: struct kvm_mp_state (in)
1041 Returns: 0 on success; -1 on error
1043 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1046 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1047 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1048 these architectures.
1051 4.40 KVM_SET_IDENTITY_MAP_ADDR
1053 Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1056 Parameters: unsigned long identity (in)
1057 Returns: 0 on success, -1 on error
1059 This ioctl defines the physical address of a one-page region in the guest
1060 physical address space. The region must be within the first 4GB of the
1061 guest physical address space and must not conflict with any memory slot
1062 or any mmio address. The guest may malfunction if it accesses this memory
1065 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1066 because of a quirk in the virtualization implementation (see the internals
1067 documentation when it pops into existence).
1070 4.41 KVM_SET_BOOT_CPU_ID
1072 Capability: KVM_CAP_SET_BOOT_CPU_ID
1075 Parameters: unsigned long vcpu_id
1076 Returns: 0 on success, -1 on error
1078 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1079 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1085 Capability: KVM_CAP_XSAVE
1088 Parameters: struct kvm_xsave (out)
1089 Returns: 0 on success, -1 on error
1095 This ioctl would copy current vcpu's xsave struct to the userspace.
1100 Capability: KVM_CAP_XSAVE
1103 Parameters: struct kvm_xsave (in)
1104 Returns: 0 on success, -1 on error
1110 This ioctl would copy userspace's xsave struct to the kernel.
1115 Capability: KVM_CAP_XCRS
1118 Parameters: struct kvm_xcrs (out)
1119 Returns: 0 on success, -1 on error
1130 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1134 This ioctl would copy current vcpu's xcrs to the userspace.
1139 Capability: KVM_CAP_XCRS
1142 Parameters: struct kvm_xcrs (in)
1143 Returns: 0 on success, -1 on error
1154 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1158 This ioctl would set vcpu's xcr to the value userspace specified.
1161 4.46 KVM_GET_SUPPORTED_CPUID
1163 Capability: KVM_CAP_EXT_CPUID
1166 Parameters: struct kvm_cpuid2 (in/out)
1167 Returns: 0 on success, -1 on error
1172 struct kvm_cpuid_entry2 entries[0];
1175 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1176 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
1177 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
1179 struct kvm_cpuid_entry2 {
1190 This ioctl returns x86 cpuid features which are supported by both the hardware
1191 and kvm. Userspace can use the information returned by this ioctl to
1192 construct cpuid information (for KVM_SET_CPUID2) that is consistent with
1193 hardware, kernel, and userspace capabilities, and with user requirements (for
1194 example, the user may wish to constrain cpuid to emulate older hardware,
1195 or for feature consistency across a cluster).
1197 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1198 with the 'nent' field indicating the number of entries in the variable-size
1199 array 'entries'. If the number of entries is too low to describe the cpu
1200 capabilities, an error (E2BIG) is returned. If the number is too high,
1201 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1202 number is just right, the 'nent' field is adjusted to the number of valid
1203 entries in the 'entries' array, which is then filled.
1205 The entries returned are the host cpuid as returned by the cpuid instruction,
1206 with unknown or unsupported features masked out. Some features (for example,
1207 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1208 emulate them efficiently. The fields in each entry are defined as follows:
1210 function: the eax value used to obtain the entry
1211 index: the ecx value used to obtain the entry (for entries that are
1213 flags: an OR of zero or more of the following:
1214 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1215 if the index field is valid
1216 KVM_CPUID_FLAG_STATEFUL_FUNC:
1217 if cpuid for this function returns different values for successive
1218 invocations; there will be several entries with the same function,
1219 all with this flag set
1220 KVM_CPUID_FLAG_STATE_READ_NEXT:
1221 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
1222 the first entry to be read by a cpu
1223 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
1224 this function/index combination
1226 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1227 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1228 support. Instead it is reported via
1230 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1232 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1233 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1236 4.47 KVM_PPC_GET_PVINFO
1238 Capability: KVM_CAP_PPC_GET_PVINFO
1241 Parameters: struct kvm_ppc_pvinfo (out)
1242 Returns: 0 on success, !0 on error
1244 struct kvm_ppc_pvinfo {
1250 This ioctl fetches PV specific information that need to be passed to the guest
1251 using the device tree or other means from vm context.
1253 The hcall array defines 4 instructions that make up a hypercall.
1255 If any additional field gets added to this structure later on, a bit for that
1256 additional piece of information will be set in the flags bitmap.
1258 The flags bitmap is defined as:
1260 /* the host supports the ePAPR idle hcall
1261 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1263 4.48 KVM_ASSIGN_PCI_DEVICE
1268 Parameters: struct kvm_assigned_pci_dev (in)
1269 Returns: 0 on success, -1 on error
1271 Assigns a host PCI device to the VM.
1273 struct kvm_assigned_pci_dev {
1274 __u32 assigned_dev_id;
1284 The PCI device is specified by the triple segnr, busnr, and devfn.
1285 Identification in succeeding service requests is done via assigned_dev_id. The
1286 following flags are specified:
1288 /* Depends on KVM_CAP_IOMMU */
1289 #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0)
1290 /* The following two depend on KVM_CAP_PCI_2_3 */
1291 #define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1)
1292 #define KVM_DEV_ASSIGN_MASK_INTX (1 << 2)
1294 If KVM_DEV_ASSIGN_PCI_2_3 is set, the kernel will manage legacy INTx interrupts
1295 via the PCI-2.3-compliant device-level mask, thus enable IRQ sharing with other
1296 assigned devices or host devices. KVM_DEV_ASSIGN_MASK_INTX specifies the
1297 guest's view on the INTx mask, see KVM_ASSIGN_SET_INTX_MASK for details.
1299 The KVM_DEV_ASSIGN_ENABLE_IOMMU flag is a mandatory option to ensure
1300 isolation of the device. Usages not specifying this flag are deprecated.
1302 Only PCI header type 0 devices with PCI BAR resources are supported by
1303 device assignment. The user requesting this ioctl must have read/write
1304 access to the PCI sysfs resource files associated with the device.
1307 ENOTTY: kernel does not support this ioctl
1309 Other error conditions may be defined by individual device types or
1310 have their standard meanings.
1313 4.49 KVM_DEASSIGN_PCI_DEVICE
1318 Parameters: struct kvm_assigned_pci_dev (in)
1319 Returns: 0 on success, -1 on error
1321 Ends PCI device assignment, releasing all associated resources.
1323 See KVM_ASSIGN_PCI_DEVICE for the data structure. Only assigned_dev_id is
1324 used in kvm_assigned_pci_dev to identify the device.
1327 ENOTTY: kernel does not support this ioctl
1329 Other error conditions may be defined by individual device types or
1330 have their standard meanings.
1332 4.50 KVM_ASSIGN_DEV_IRQ
1334 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1337 Parameters: struct kvm_assigned_irq (in)
1338 Returns: 0 on success, -1 on error
1340 Assigns an IRQ to a passed-through device.
1342 struct kvm_assigned_irq {
1343 __u32 assigned_dev_id;
1344 __u32 host_irq; /* ignored (legacy field) */
1352 The following flags are defined:
1354 #define KVM_DEV_IRQ_HOST_INTX (1 << 0)
1355 #define KVM_DEV_IRQ_HOST_MSI (1 << 1)
1356 #define KVM_DEV_IRQ_HOST_MSIX (1 << 2)
1358 #define KVM_DEV_IRQ_GUEST_INTX (1 << 8)
1359 #define KVM_DEV_IRQ_GUEST_MSI (1 << 9)
1360 #define KVM_DEV_IRQ_GUEST_MSIX (1 << 10)
1362 It is not valid to specify multiple types per host or guest IRQ. However, the
1363 IRQ type of host and guest can differ or can even be null.
1366 ENOTTY: kernel does not support this ioctl
1368 Other error conditions may be defined by individual device types or
1369 have their standard meanings.
1372 4.51 KVM_DEASSIGN_DEV_IRQ
1374 Capability: KVM_CAP_ASSIGN_DEV_IRQ
1377 Parameters: struct kvm_assigned_irq (in)
1378 Returns: 0 on success, -1 on error
1380 Ends an IRQ assignment to a passed-through device.
1382 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1383 by assigned_dev_id, flags must correspond to the IRQ type specified on
1384 KVM_ASSIGN_DEV_IRQ. Partial deassignment of host or guest IRQ is allowed.
1387 4.52 KVM_SET_GSI_ROUTING
1389 Capability: KVM_CAP_IRQ_ROUTING
1390 Architectures: x86 s390
1392 Parameters: struct kvm_irq_routing (in)
1393 Returns: 0 on success, -1 on error
1395 Sets the GSI routing table entries, overwriting any previously set entries.
1397 struct kvm_irq_routing {
1400 struct kvm_irq_routing_entry entries[0];
1403 No flags are specified so far, the corresponding field must be set to zero.
1405 struct kvm_irq_routing_entry {
1411 struct kvm_irq_routing_irqchip irqchip;
1412 struct kvm_irq_routing_msi msi;
1413 struct kvm_irq_routing_s390_adapter adapter;
1418 /* gsi routing entry types */
1419 #define KVM_IRQ_ROUTING_IRQCHIP 1
1420 #define KVM_IRQ_ROUTING_MSI 2
1421 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1423 No flags are specified so far, the corresponding field must be set to zero.
1425 struct kvm_irq_routing_irqchip {
1430 struct kvm_irq_routing_msi {
1437 struct kvm_irq_routing_s390_adapter {
1441 __u32 summary_offset;
1446 4.53 KVM_ASSIGN_SET_MSIX_NR
1451 Parameters: struct kvm_assigned_msix_nr (in)
1452 Returns: 0 on success, -1 on error
1454 Set the number of MSI-X interrupts for an assigned device. The number is
1455 reset again by terminating the MSI-X assignment of the device via
1456 KVM_DEASSIGN_DEV_IRQ. Calling this service more than once at any earlier
1459 struct kvm_assigned_msix_nr {
1460 __u32 assigned_dev_id;
1465 #define KVM_MAX_MSIX_PER_DEV 256
1468 4.54 KVM_ASSIGN_SET_MSIX_ENTRY
1473 Parameters: struct kvm_assigned_msix_entry (in)
1474 Returns: 0 on success, -1 on error
1476 Specifies the routing of an MSI-X assigned device interrupt to a GSI. Setting
1477 the GSI vector to zero means disabling the interrupt.
1479 struct kvm_assigned_msix_entry {
1480 __u32 assigned_dev_id;
1482 __u16 entry; /* The index of entry in the MSI-X table */
1487 ENOTTY: kernel does not support this ioctl
1489 Other error conditions may be defined by individual device types or
1490 have their standard meanings.
1493 4.55 KVM_SET_TSC_KHZ
1495 Capability: KVM_CAP_TSC_CONTROL
1498 Parameters: virtual tsc_khz
1499 Returns: 0 on success, -1 on error
1501 Specifies the tsc frequency for the virtual machine. The unit of the
1505 4.56 KVM_GET_TSC_KHZ
1507 Capability: KVM_CAP_GET_TSC_KHZ
1511 Returns: virtual tsc-khz on success, negative value on error
1513 Returns the tsc frequency of the guest. The unit of the return value is
1514 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1520 Capability: KVM_CAP_IRQCHIP
1523 Parameters: struct kvm_lapic_state (out)
1524 Returns: 0 on success, -1 on error
1526 #define KVM_APIC_REG_SIZE 0x400
1527 struct kvm_lapic_state {
1528 char regs[KVM_APIC_REG_SIZE];
1531 Reads the Local APIC registers and copies them into the input argument. The
1532 data format and layout are the same as documented in the architecture manual.
1537 Capability: KVM_CAP_IRQCHIP
1540 Parameters: struct kvm_lapic_state (in)
1541 Returns: 0 on success, -1 on error
1543 #define KVM_APIC_REG_SIZE 0x400
1544 struct kvm_lapic_state {
1545 char regs[KVM_APIC_REG_SIZE];
1548 Copies the input argument into the Local APIC registers. The data format
1549 and layout are the same as documented in the architecture manual.
1554 Capability: KVM_CAP_IOEVENTFD
1557 Parameters: struct kvm_ioeventfd (in)
1558 Returns: 0 on success, !0 on error
1560 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
1561 within the guest. A guest write in the registered address will signal the
1562 provided event instead of triggering an exit.
1564 struct kvm_ioeventfd {
1566 __u64 addr; /* legal pio/mmio address */
1567 __u32 len; /* 1, 2, 4, or 8 bytes */
1573 For the special case of virtio-ccw devices on s390, the ioevent is matched
1574 to a subchannel/virtqueue tuple instead.
1576 The following flags are defined:
1578 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
1579 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
1580 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
1581 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
1582 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
1584 If datamatch flag is set, the event will be signaled only if the written value
1585 to the registered address is equal to datamatch in struct kvm_ioeventfd.
1587 For virtio-ccw devices, addr contains the subchannel id and datamatch the
1593 Capability: KVM_CAP_SW_TLB
1596 Parameters: struct kvm_dirty_tlb (in)
1597 Returns: 0 on success, -1 on error
1599 struct kvm_dirty_tlb {
1604 This must be called whenever userspace has changed an entry in the shared
1605 TLB, prior to calling KVM_RUN on the associated vcpu.
1607 The "bitmap" field is the userspace address of an array. This array
1608 consists of a number of bits, equal to the total number of TLB entries as
1609 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
1610 nearest multiple of 64.
1612 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
1615 The array is little-endian: the bit 0 is the least significant bit of the
1616 first byte, bit 8 is the least significant bit of the second byte, etc.
1617 This avoids any complications with differing word sizes.
1619 The "num_dirty" field is a performance hint for KVM to determine whether it
1620 should skip processing the bitmap and just invalidate everything. It must
1621 be set to the number of set bits in the bitmap.
1624 4.61 KVM_ASSIGN_SET_INTX_MASK
1626 Capability: KVM_CAP_PCI_2_3
1629 Parameters: struct kvm_assigned_pci_dev (in)
1630 Returns: 0 on success, -1 on error
1632 Allows userspace to mask PCI INTx interrupts from the assigned device. The
1633 kernel will not deliver INTx interrupts to the guest between setting and
1634 clearing of KVM_ASSIGN_SET_INTX_MASK via this interface. This enables use of
1635 and emulation of PCI 2.3 INTx disable command register behavior.
1637 This may be used for both PCI 2.3 devices supporting INTx disable natively and
1638 older devices lacking this support. Userspace is responsible for emulating the
1639 read value of the INTx disable bit in the guest visible PCI command register.
1640 When modifying the INTx disable state, userspace should precede updating the
1641 physical device command register by calling this ioctl to inform the kernel of
1642 the new intended INTx mask state.
1644 Note that the kernel uses the device INTx disable bit to internally manage the
1645 device interrupt state for PCI 2.3 devices. Reads of this register may
1646 therefore not match the expected value. Writes should always use the guest
1647 intended INTx disable value rather than attempting to read-copy-update the
1648 current physical device state. Races between user and kernel updates to the
1649 INTx disable bit are handled lazily in the kernel. It's possible the device
1650 may generate unintended interrupts, but they will not be injected into the
1653 See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified
1654 by assigned_dev_id. In the flags field, only KVM_DEV_ASSIGN_MASK_INTX is
1658 4.62 KVM_CREATE_SPAPR_TCE
1660 Capability: KVM_CAP_SPAPR_TCE
1661 Architectures: powerpc
1663 Parameters: struct kvm_create_spapr_tce (in)
1664 Returns: file descriptor for manipulating the created TCE table
1666 This creates a virtual TCE (translation control entry) table, which
1667 is an IOMMU for PAPR-style virtual I/O. It is used to translate
1668 logical addresses used in virtual I/O into guest physical addresses,
1669 and provides a scatter/gather capability for PAPR virtual I/O.
1671 /* for KVM_CAP_SPAPR_TCE */
1672 struct kvm_create_spapr_tce {
1677 The liobn field gives the logical IO bus number for which to create a
1678 TCE table. The window_size field specifies the size of the DMA window
1679 which this TCE table will translate - the table will contain one 64
1680 bit TCE entry for every 4kiB of the DMA window.
1682 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
1683 table has been created using this ioctl(), the kernel will handle it
1684 in real mode, updating the TCE table. H_PUT_TCE calls for other
1685 liobns will cause a vm exit and must be handled by userspace.
1687 The return value is a file descriptor which can be passed to mmap(2)
1688 to map the created TCE table into userspace. This lets userspace read
1689 the entries written by kernel-handled H_PUT_TCE calls, and also lets
1690 userspace update the TCE table directly which is useful in some
1694 4.63 KVM_ALLOCATE_RMA
1696 Capability: KVM_CAP_PPC_RMA
1697 Architectures: powerpc
1699 Parameters: struct kvm_allocate_rma (out)
1700 Returns: file descriptor for mapping the allocated RMA
1702 This allocates a Real Mode Area (RMA) from the pool allocated at boot
1703 time by the kernel. An RMA is a physically-contiguous, aligned region
1704 of memory used on older POWER processors to provide the memory which
1705 will be accessed by real-mode (MMU off) accesses in a KVM guest.
1706 POWER processors support a set of sizes for the RMA that usually
1707 includes 64MB, 128MB, 256MB and some larger powers of two.
1709 /* for KVM_ALLOCATE_RMA */
1710 struct kvm_allocate_rma {
1714 The return value is a file descriptor which can be passed to mmap(2)
1715 to map the allocated RMA into userspace. The mapped area can then be
1716 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
1717 RMA for a virtual machine. The size of the RMA in bytes (which is
1718 fixed at host kernel boot time) is returned in the rma_size field of
1719 the argument structure.
1721 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
1722 is supported; 2 if the processor requires all virtual machines to have
1723 an RMA, or 1 if the processor can use an RMA but doesn't require it,
1724 because it supports the Virtual RMA (VRMA) facility.
1729 Capability: KVM_CAP_USER_NMI
1733 Returns: 0 on success, -1 on error
1735 Queues an NMI on the thread's vcpu. Note this is well defined only
1736 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
1737 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
1738 has been called, this interface is completely emulated within the kernel.
1740 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
1741 following algorithm:
1744 - read the local APIC's state (KVM_GET_LAPIC)
1745 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
1746 - if so, issue KVM_NMI
1749 Some guests configure the LINT1 NMI input to cause a panic, aiding in
1753 4.65 KVM_S390_UCAS_MAP
1755 Capability: KVM_CAP_S390_UCONTROL
1758 Parameters: struct kvm_s390_ucas_mapping (in)
1759 Returns: 0 in case of success
1761 The parameter is defined like this:
1762 struct kvm_s390_ucas_mapping {
1768 This ioctl maps the memory at "user_addr" with the length "length" to
1769 the vcpu's address space starting at "vcpu_addr". All parameters need to
1770 be aligned by 1 megabyte.
1773 4.66 KVM_S390_UCAS_UNMAP
1775 Capability: KVM_CAP_S390_UCONTROL
1778 Parameters: struct kvm_s390_ucas_mapping (in)
1779 Returns: 0 in case of success
1781 The parameter is defined like this:
1782 struct kvm_s390_ucas_mapping {
1788 This ioctl unmaps the memory in the vcpu's address space starting at
1789 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
1790 All parameters need to be aligned by 1 megabyte.
1793 4.67 KVM_S390_VCPU_FAULT
1795 Capability: KVM_CAP_S390_UCONTROL
1798 Parameters: vcpu absolute address (in)
1799 Returns: 0 in case of success
1801 This call creates a page table entry on the virtual cpu's address space
1802 (for user controlled virtual machines) or the virtual machine's address
1803 space (for regular virtual machines). This only works for minor faults,
1804 thus it's recommended to access subject memory page via the user page
1805 table upfront. This is useful to handle validity intercepts for user
1806 controlled virtual machines to fault in the virtual cpu's lowcore pages
1807 prior to calling the KVM_RUN ioctl.
1810 4.68 KVM_SET_ONE_REG
1812 Capability: KVM_CAP_ONE_REG
1815 Parameters: struct kvm_one_reg (in)
1816 Returns: 0 on success, negative value on failure
1818 struct kvm_one_reg {
1823 Using this ioctl, a single vcpu register can be set to a specific value
1824 defined by user space with the passed in struct kvm_one_reg, where id
1825 refers to the register identifier as described below and addr is a pointer
1826 to a variable with the respective size. There can be architecture agnostic
1827 and architecture specific registers. Each have their own range of operation
1828 and their own constants and width. To keep track of the implemented
1829 registers, find a list below:
1831 Arch | Register | Width (bits)
1833 PPC | KVM_REG_PPC_HIOR | 64
1834 PPC | KVM_REG_PPC_IAC1 | 64
1835 PPC | KVM_REG_PPC_IAC2 | 64
1836 PPC | KVM_REG_PPC_IAC3 | 64
1837 PPC | KVM_REG_PPC_IAC4 | 64
1838 PPC | KVM_REG_PPC_DAC1 | 64
1839 PPC | KVM_REG_PPC_DAC2 | 64
1840 PPC | KVM_REG_PPC_DABR | 64
1841 PPC | KVM_REG_PPC_DSCR | 64
1842 PPC | KVM_REG_PPC_PURR | 64
1843 PPC | KVM_REG_PPC_SPURR | 64
1844 PPC | KVM_REG_PPC_DAR | 64
1845 PPC | KVM_REG_PPC_DSISR | 32
1846 PPC | KVM_REG_PPC_AMR | 64
1847 PPC | KVM_REG_PPC_UAMOR | 64
1848 PPC | KVM_REG_PPC_MMCR0 | 64
1849 PPC | KVM_REG_PPC_MMCR1 | 64
1850 PPC | KVM_REG_PPC_MMCRA | 64
1851 PPC | KVM_REG_PPC_MMCR2 | 64
1852 PPC | KVM_REG_PPC_MMCRS | 64
1853 PPC | KVM_REG_PPC_SIAR | 64
1854 PPC | KVM_REG_PPC_SDAR | 64
1855 PPC | KVM_REG_PPC_SIER | 64
1856 PPC | KVM_REG_PPC_PMC1 | 32
1857 PPC | KVM_REG_PPC_PMC2 | 32
1858 PPC | KVM_REG_PPC_PMC3 | 32
1859 PPC | KVM_REG_PPC_PMC4 | 32
1860 PPC | KVM_REG_PPC_PMC5 | 32
1861 PPC | KVM_REG_PPC_PMC6 | 32
1862 PPC | KVM_REG_PPC_PMC7 | 32
1863 PPC | KVM_REG_PPC_PMC8 | 32
1864 PPC | KVM_REG_PPC_FPR0 | 64
1866 PPC | KVM_REG_PPC_FPR31 | 64
1867 PPC | KVM_REG_PPC_VR0 | 128
1869 PPC | KVM_REG_PPC_VR31 | 128
1870 PPC | KVM_REG_PPC_VSR0 | 128
1872 PPC | KVM_REG_PPC_VSR31 | 128
1873 PPC | KVM_REG_PPC_FPSCR | 64
1874 PPC | KVM_REG_PPC_VSCR | 32
1875 PPC | KVM_REG_PPC_VPA_ADDR | 64
1876 PPC | KVM_REG_PPC_VPA_SLB | 128
1877 PPC | KVM_REG_PPC_VPA_DTL | 128
1878 PPC | KVM_REG_PPC_EPCR | 32
1879 PPC | KVM_REG_PPC_EPR | 32
1880 PPC | KVM_REG_PPC_TCR | 32
1881 PPC | KVM_REG_PPC_TSR | 32
1882 PPC | KVM_REG_PPC_OR_TSR | 32
1883 PPC | KVM_REG_PPC_CLEAR_TSR | 32
1884 PPC | KVM_REG_PPC_MAS0 | 32
1885 PPC | KVM_REG_PPC_MAS1 | 32
1886 PPC | KVM_REG_PPC_MAS2 | 64
1887 PPC | KVM_REG_PPC_MAS7_3 | 64
1888 PPC | KVM_REG_PPC_MAS4 | 32
1889 PPC | KVM_REG_PPC_MAS6 | 32
1890 PPC | KVM_REG_PPC_MMUCFG | 32
1891 PPC | KVM_REG_PPC_TLB0CFG | 32
1892 PPC | KVM_REG_PPC_TLB1CFG | 32
1893 PPC | KVM_REG_PPC_TLB2CFG | 32
1894 PPC | KVM_REG_PPC_TLB3CFG | 32
1895 PPC | KVM_REG_PPC_TLB0PS | 32
1896 PPC | KVM_REG_PPC_TLB1PS | 32
1897 PPC | KVM_REG_PPC_TLB2PS | 32
1898 PPC | KVM_REG_PPC_TLB3PS | 32
1899 PPC | KVM_REG_PPC_EPTCFG | 32
1900 PPC | KVM_REG_PPC_ICP_STATE | 64
1901 PPC | KVM_REG_PPC_TB_OFFSET | 64
1902 PPC | KVM_REG_PPC_SPMC1 | 32
1903 PPC | KVM_REG_PPC_SPMC2 | 32
1904 PPC | KVM_REG_PPC_IAMR | 64
1905 PPC | KVM_REG_PPC_TFHAR | 64
1906 PPC | KVM_REG_PPC_TFIAR | 64
1907 PPC | KVM_REG_PPC_TEXASR | 64
1908 PPC | KVM_REG_PPC_FSCR | 64
1909 PPC | KVM_REG_PPC_PSPB | 32
1910 PPC | KVM_REG_PPC_EBBHR | 64
1911 PPC | KVM_REG_PPC_EBBRR | 64
1912 PPC | KVM_REG_PPC_BESCR | 64
1913 PPC | KVM_REG_PPC_TAR | 64
1914 PPC | KVM_REG_PPC_DPDES | 64
1915 PPC | KVM_REG_PPC_DAWR | 64
1916 PPC | KVM_REG_PPC_DAWRX | 64
1917 PPC | KVM_REG_PPC_CIABR | 64
1918 PPC | KVM_REG_PPC_IC | 64
1919 PPC | KVM_REG_PPC_VTB | 64
1920 PPC | KVM_REG_PPC_CSIGR | 64
1921 PPC | KVM_REG_PPC_TACR | 64
1922 PPC | KVM_REG_PPC_TCSCR | 64
1923 PPC | KVM_REG_PPC_PID | 64
1924 PPC | KVM_REG_PPC_ACOP | 64
1925 PPC | KVM_REG_PPC_VRSAVE | 32
1926 PPC | KVM_REG_PPC_LPCR | 32
1927 PPC | KVM_REG_PPC_LPCR_64 | 64
1928 PPC | KVM_REG_PPC_PPR | 64
1929 PPC | KVM_REG_PPC_ARCH_COMPAT | 32
1930 PPC | KVM_REG_PPC_DABRX | 32
1931 PPC | KVM_REG_PPC_WORT | 64
1932 PPC | KVM_REG_PPC_SPRG9 | 64
1933 PPC | KVM_REG_PPC_DBSR | 32
1934 PPC | KVM_REG_PPC_TM_GPR0 | 64
1936 PPC | KVM_REG_PPC_TM_GPR31 | 64
1937 PPC | KVM_REG_PPC_TM_VSR0 | 128
1939 PPC | KVM_REG_PPC_TM_VSR63 | 128
1940 PPC | KVM_REG_PPC_TM_CR | 64
1941 PPC | KVM_REG_PPC_TM_LR | 64
1942 PPC | KVM_REG_PPC_TM_CTR | 64
1943 PPC | KVM_REG_PPC_TM_FPSCR | 64
1944 PPC | KVM_REG_PPC_TM_AMR | 64
1945 PPC | KVM_REG_PPC_TM_PPR | 64
1946 PPC | KVM_REG_PPC_TM_VRSAVE | 64
1947 PPC | KVM_REG_PPC_TM_VSCR | 32
1948 PPC | KVM_REG_PPC_TM_DSCR | 64
1949 PPC | KVM_REG_PPC_TM_TAR | 64
1951 MIPS | KVM_REG_MIPS_R0 | 64
1953 MIPS | KVM_REG_MIPS_R31 | 64
1954 MIPS | KVM_REG_MIPS_HI | 64
1955 MIPS | KVM_REG_MIPS_LO | 64
1956 MIPS | KVM_REG_MIPS_PC | 64
1957 MIPS | KVM_REG_MIPS_CP0_INDEX | 32
1958 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64
1959 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64
1960 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32
1961 MIPS | KVM_REG_MIPS_CP0_WIRED | 32
1962 MIPS | KVM_REG_MIPS_CP0_HWRENA | 32
1963 MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64
1964 MIPS | KVM_REG_MIPS_CP0_COUNT | 32
1965 MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64
1966 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32
1967 MIPS | KVM_REG_MIPS_CP0_STATUS | 32
1968 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32
1969 MIPS | KVM_REG_MIPS_CP0_EPC | 64
1970 MIPS | KVM_REG_MIPS_CP0_PRID | 32
1971 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32
1972 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32
1973 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32
1974 MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32
1975 MIPS | KVM_REG_MIPS_CP0_CONFIG4 | 32
1976 MIPS | KVM_REG_MIPS_CP0_CONFIG5 | 32
1977 MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32
1978 MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64
1979 MIPS | KVM_REG_MIPS_COUNT_CTL | 64
1980 MIPS | KVM_REG_MIPS_COUNT_RESUME | 64
1981 MIPS | KVM_REG_MIPS_COUNT_HZ | 64
1982 MIPS | KVM_REG_MIPS_FPR_32(0..31) | 32
1983 MIPS | KVM_REG_MIPS_FPR_64(0..31) | 64
1984 MIPS | KVM_REG_MIPS_FCR_IR | 32
1985 MIPS | KVM_REG_MIPS_FCR_CSR | 32
1987 ARM registers are mapped using the lower 32 bits. The upper 16 of that
1988 is the register group type, or coprocessor number:
1990 ARM core registers have the following id bit patterns:
1991 0x4020 0000 0010 <index into the kvm_regs struct:16>
1993 ARM 32-bit CP15 registers have the following id bit patterns:
1994 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
1996 ARM 64-bit CP15 registers have the following id bit patterns:
1997 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
1999 ARM CCSIDR registers are demultiplexed by CSSELR value:
2000 0x4020 0000 0011 00 <csselr:8>
2002 ARM 32-bit VFP control registers have the following id bit patterns:
2003 0x4020 0000 0012 1 <regno:12>
2005 ARM 64-bit FP registers have the following id bit patterns:
2006 0x4030 0000 0012 0 <regno:12>
2009 arm64 registers are mapped using the lower 32 bits. The upper 16 of
2010 that is the register group type, or coprocessor number:
2012 arm64 core/FP-SIMD registers have the following id bit patterns. Note
2013 that the size of the access is variable, as the kvm_regs structure
2014 contains elements ranging from 32 to 128 bits. The index is a 32bit
2015 value in the kvm_regs structure seen as a 32bit array.
2016 0x60x0 0000 0010 <index into the kvm_regs struct:16>
2018 arm64 CCSIDR registers are demultiplexed by CSSELR value:
2019 0x6020 0000 0011 00 <csselr:8>
2021 arm64 system registers have the following id bit patterns:
2022 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
2025 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
2026 the register group type:
2028 MIPS core registers (see above) have the following id bit patterns:
2029 0x7030 0000 0000 <reg:16>
2031 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
2032 patterns depending on whether they're 32-bit or 64-bit registers:
2033 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
2034 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2036 MIPS KVM control registers (see above) have the following id bit patterns:
2037 0x7030 0000 0002 <reg:16>
2039 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
2040 id bit patterns depending on the size of the register being accessed. They are
2041 always accessed according to the current guest FPU mode (Status.FR and
2042 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
2043 if the guest FPU mode is changed:
2044 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
2045 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
2047 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
2048 following id bit patterns:
2049 0x7020 0000 0003 01 <0:3> <reg:5>
2052 4.69 KVM_GET_ONE_REG
2054 Capability: KVM_CAP_ONE_REG
2057 Parameters: struct kvm_one_reg (in and out)
2058 Returns: 0 on success, negative value on failure
2060 This ioctl allows to receive the value of a single register implemented
2061 in a vcpu. The register to read is indicated by the "id" field of the
2062 kvm_one_reg struct passed in. On success, the register value can be found
2063 at the memory location pointed to by "addr".
2065 The list of registers accessible using this interface is identical to the
2069 4.70 KVM_KVMCLOCK_CTRL
2071 Capability: KVM_CAP_KVMCLOCK_CTRL
2072 Architectures: Any that implement pvclocks (currently x86 only)
2075 Returns: 0 on success, -1 on error
2077 This signals to the host kernel that the specified guest is being paused by
2078 userspace. The host will set a flag in the pvclock structure that is checked
2079 from the soft lockup watchdog. The flag is part of the pvclock structure that
2080 is shared between guest and host, specifically the second bit of the flags
2081 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2082 the host and read/cleared exclusively by the guest. The guest operation of
2083 checking and clearing the flag must an atomic operation so
2084 load-link/store-conditional, or equivalent must be used. There are two cases
2085 where the guest will clear the flag: when the soft lockup watchdog timer resets
2086 itself or when a soft lockup is detected. This ioctl can be called any time
2087 after pausing the vcpu, but before it is resumed.
2092 Capability: KVM_CAP_SIGNAL_MSI
2095 Parameters: struct kvm_msi (in)
2096 Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2098 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2109 No flags are defined so far. The corresponding field must be 0.
2112 4.71 KVM_CREATE_PIT2
2114 Capability: KVM_CAP_PIT2
2117 Parameters: struct kvm_pit_config (in)
2118 Returns: 0 on success, -1 on error
2120 Creates an in-kernel device model for the i8254 PIT. This call is only valid
2121 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
2122 parameters have to be passed:
2124 struct kvm_pit_config {
2131 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
2133 PIT timer interrupts may use a per-VM kernel thread for injection. If it
2134 exists, this thread will have a name of the following pattern:
2136 kvm-pit/<owner-process-pid>
2138 When running a guest with elevated priorities, the scheduling parameters of
2139 this thread may have to be adjusted accordingly.
2141 This IOCTL replaces the obsolete KVM_CREATE_PIT.
2146 Capability: KVM_CAP_PIT_STATE2
2149 Parameters: struct kvm_pit_state2 (out)
2150 Returns: 0 on success, -1 on error
2152 Retrieves the state of the in-kernel PIT model. Only valid after
2153 KVM_CREATE_PIT2. The state is returned in the following structure:
2155 struct kvm_pit_state2 {
2156 struct kvm_pit_channel_state channels[3];
2163 /* disable PIT in HPET legacy mode */
2164 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
2166 This IOCTL replaces the obsolete KVM_GET_PIT.
2171 Capability: KVM_CAP_PIT_STATE2
2174 Parameters: struct kvm_pit_state2 (in)
2175 Returns: 0 on success, -1 on error
2177 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
2178 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
2180 This IOCTL replaces the obsolete KVM_SET_PIT.
2183 4.74 KVM_PPC_GET_SMMU_INFO
2185 Capability: KVM_CAP_PPC_GET_SMMU_INFO
2186 Architectures: powerpc
2189 Returns: 0 on success, -1 on error
2191 This populates and returns a structure describing the features of
2192 the "Server" class MMU emulation supported by KVM.
2193 This can in turn be used by userspace to generate the appropriate
2194 device-tree properties for the guest operating system.
2196 The structure contains some global information, followed by an
2197 array of supported segment page sizes:
2199 struct kvm_ppc_smmu_info {
2203 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
2206 The supported flags are:
2208 - KVM_PPC_PAGE_SIZES_REAL:
2209 When that flag is set, guest page sizes must "fit" the backing
2210 store page sizes. When not set, any page size in the list can
2211 be used regardless of how they are backed by userspace.
2213 - KVM_PPC_1T_SEGMENTS
2214 The emulated MMU supports 1T segments in addition to the
2217 The "slb_size" field indicates how many SLB entries are supported
2219 The "sps" array contains 8 entries indicating the supported base
2220 page sizes for a segment in increasing order. Each entry is defined
2223 struct kvm_ppc_one_seg_page_size {
2224 __u32 page_shift; /* Base page shift of segment (or 0) */
2225 __u32 slb_enc; /* SLB encoding for BookS */
2226 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
2229 An entry with a "page_shift" of 0 is unused. Because the array is
2230 organized in increasing order, a lookup can stop when encoutering
2233 The "slb_enc" field provides the encoding to use in the SLB for the
2234 page size. The bits are in positions such as the value can directly
2235 be OR'ed into the "vsid" argument of the slbmte instruction.
2237 The "enc" array is a list which for each of those segment base page
2238 size provides the list of supported actual page sizes (which can be
2239 only larger or equal to the base page size), along with the
2240 corresponding encoding in the hash PTE. Similarly, the array is
2241 8 entries sorted by increasing sizes and an entry with a "0" shift
2242 is an empty entry and a terminator:
2244 struct kvm_ppc_one_page_size {
2245 __u32 page_shift; /* Page shift (or 0) */
2246 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
2249 The "pte_enc" field provides a value that can OR'ed into the hash
2250 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
2251 into the hash PTE second double word).
2255 Capability: KVM_CAP_IRQFD
2256 Architectures: x86 s390
2258 Parameters: struct kvm_irqfd (in)
2259 Returns: 0 on success, -1 on error
2261 Allows setting an eventfd to directly trigger a guest interrupt.
2262 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
2263 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
2264 an event is triggered on the eventfd, an interrupt is injected into
2265 the guest using the specified gsi pin. The irqfd is removed using
2266 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
2269 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
2270 mechanism allowing emulation of level-triggered, irqfd-based
2271 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
2272 additional eventfd in the kvm_irqfd.resamplefd field. When operating
2273 in resample mode, posting of an interrupt through kvm_irq.fd asserts
2274 the specified gsi in the irqchip. When the irqchip is resampled, such
2275 as from an EOI, the gsi is de-asserted and the user is notified via
2276 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
2277 the interrupt if the device making use of it still requires service.
2278 Note that closing the resamplefd is not sufficient to disable the
2279 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
2280 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
2282 4.76 KVM_PPC_ALLOCATE_HTAB
2284 Capability: KVM_CAP_PPC_ALLOC_HTAB
2285 Architectures: powerpc
2287 Parameters: Pointer to u32 containing hash table order (in/out)
2288 Returns: 0 on success, -1 on error
2290 This requests the host kernel to allocate an MMU hash table for a
2291 guest using the PAPR paravirtualization interface. This only does
2292 anything if the kernel is configured to use the Book 3S HV style of
2293 virtualization. Otherwise the capability doesn't exist and the ioctl
2294 returns an ENOTTY error. The rest of this description assumes Book 3S
2297 There must be no vcpus running when this ioctl is called; if there
2298 are, it will do nothing and return an EBUSY error.
2300 The parameter is a pointer to a 32-bit unsigned integer variable
2301 containing the order (log base 2) of the desired size of the hash
2302 table, which must be between 18 and 46. On successful return from the
2303 ioctl, it will have been updated with the order of the hash table that
2306 If no hash table has been allocated when any vcpu is asked to run
2307 (with the KVM_RUN ioctl), the host kernel will allocate a
2308 default-sized hash table (16 MB).
2310 If this ioctl is called when a hash table has already been allocated,
2311 the kernel will clear out the existing hash table (zero all HPTEs) and
2312 return the hash table order in the parameter. (If the guest is using
2313 the virtualized real-mode area (VRMA) facility, the kernel will
2314 re-create the VMRA HPTEs on the next KVM_RUN of any vcpu.)
2316 4.77 KVM_S390_INTERRUPT
2320 Type: vm ioctl, vcpu ioctl
2321 Parameters: struct kvm_s390_interrupt (in)
2322 Returns: 0 on success, -1 on error
2324 Allows to inject an interrupt to the guest. Interrupts can be floating
2325 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
2327 Interrupt parameters are passed via kvm_s390_interrupt:
2329 struct kvm_s390_interrupt {
2335 type can be one of the following:
2337 KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm
2338 KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm
2339 KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm
2340 KVM_S390_RESTART (vcpu) - restart
2341 KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt
2342 KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt
2343 KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt
2344 parameters in parm and parm64
2345 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm
2346 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm
2347 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm
2348 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
2349 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
2350 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
2351 interruption subclass)
2352 KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm,
2353 machine check interrupt code in parm64 (note that
2354 machine checks needing further payload are not
2355 supported by this ioctl)
2357 Note that the vcpu ioctl is asynchronous to vcpu execution.
2359 4.78 KVM_PPC_GET_HTAB_FD
2361 Capability: KVM_CAP_PPC_HTAB_FD
2362 Architectures: powerpc
2364 Parameters: Pointer to struct kvm_get_htab_fd (in)
2365 Returns: file descriptor number (>= 0) on success, -1 on error
2367 This returns a file descriptor that can be used either to read out the
2368 entries in the guest's hashed page table (HPT), or to write entries to
2369 initialize the HPT. The returned fd can only be written to if the
2370 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
2371 can only be read if that bit is clear. The argument struct looks like
2374 /* For KVM_PPC_GET_HTAB_FD */
2375 struct kvm_get_htab_fd {
2381 /* Values for kvm_get_htab_fd.flags */
2382 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
2383 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
2385 The `start_index' field gives the index in the HPT of the entry at
2386 which to start reading. It is ignored when writing.
2388 Reads on the fd will initially supply information about all
2389 "interesting" HPT entries. Interesting entries are those with the
2390 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
2391 all entries. When the end of the HPT is reached, the read() will
2392 return. If read() is called again on the fd, it will start again from
2393 the beginning of the HPT, but will only return HPT entries that have
2394 changed since they were last read.
2396 Data read or written is structured as a header (8 bytes) followed by a
2397 series of valid HPT entries (16 bytes) each. The header indicates how
2398 many valid HPT entries there are and how many invalid entries follow
2399 the valid entries. The invalid entries are not represented explicitly
2400 in the stream. The header format is:
2402 struct kvm_get_htab_header {
2408 Writes to the fd create HPT entries starting at the index given in the
2409 header; first `n_valid' valid entries with contents from the data
2410 written, then `n_invalid' invalid entries, invalidating any previously
2411 valid entries found.
2413 4.79 KVM_CREATE_DEVICE
2415 Capability: KVM_CAP_DEVICE_CTRL
2417 Parameters: struct kvm_create_device (in/out)
2418 Returns: 0 on success, -1 on error
2420 ENODEV: The device type is unknown or unsupported
2421 EEXIST: Device already created, and this type of device may not
2422 be instantiated multiple times
2424 Other error conditions may be defined by individual device types or
2425 have their standard meanings.
2427 Creates an emulated device in the kernel. The file descriptor returned
2428 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
2430 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
2431 device type is supported (not necessarily whether it can be created
2434 Individual devices should not define flags. Attributes should be used
2435 for specifying any behavior that is not implied by the device type
2438 struct kvm_create_device {
2439 __u32 type; /* in: KVM_DEV_TYPE_xxx */
2440 __u32 fd; /* out: device handle */
2441 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
2444 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
2446 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device
2447 Type: device ioctl, vm ioctl
2448 Parameters: struct kvm_device_attr
2449 Returns: 0 on success, -1 on error
2451 ENXIO: The group or attribute is unknown/unsupported for this device
2452 EPERM: The attribute cannot (currently) be accessed this way
2453 (e.g. read-only attribute, or attribute that only makes
2454 sense when the device is in a different state)
2456 Other error conditions may be defined by individual device types.
2458 Gets/sets a specified piece of device configuration and/or state. The
2459 semantics are device-specific. See individual device documentation in
2460 the "devices" directory. As with ONE_REG, the size of the data
2461 transferred is defined by the particular attribute.
2463 struct kvm_device_attr {
2464 __u32 flags; /* no flags currently defined */
2465 __u32 group; /* device-defined */
2466 __u64 attr; /* group-defined */
2467 __u64 addr; /* userspace address of attr data */
2470 4.81 KVM_HAS_DEVICE_ATTR
2472 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device
2473 Type: device ioctl, vm ioctl
2474 Parameters: struct kvm_device_attr
2475 Returns: 0 on success, -1 on error
2477 ENXIO: The group or attribute is unknown/unsupported for this device
2479 Tests whether a device supports a particular attribute. A successful
2480 return indicates the attribute is implemented. It does not necessarily
2481 indicate that the attribute can be read or written in the device's
2482 current state. "addr" is ignored.
2484 4.82 KVM_ARM_VCPU_INIT
2487 Architectures: arm, arm64
2489 Parameters: struct kvm_vcpu_init (in)
2490 Returns: 0 on success; -1 on error
2492 Â EINVAL: Â Â Â the target is unknown, or the combination of features is invalid.
2493 Â ENOENT: Â Â Â a features bit specified is unknown.
2495 This tells KVM what type of CPU to present to the guest, and what
2496 optional features it should have. Â This will cause a reset of the cpu
2497 registers to their initial values. Â If this is not called, KVM_RUN will
2498 return ENOEXEC for that vcpu.
2500 Note that because some registers reflect machine topology, all vcpus
2501 should be created before this ioctl is invoked.
2503 Userspace can call this function multiple times for a given vcpu, including
2504 after the vcpu has been run. This will reset the vcpu to its initial
2505 state. All calls to this function after the initial call must use the same
2506 target and same set of feature flags, otherwise EINVAL will be returned.
2509 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
2510 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
2511 and execute guest code when KVM_RUN is called.
2512 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
2513 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
2514 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU.
2515 Depends on KVM_CAP_ARM_PSCI_0_2.
2518 4.83 KVM_ARM_PREFERRED_TARGET
2521 Architectures: arm, arm64
2523 Parameters: struct struct kvm_vcpu_init (out)
2524 Returns: 0 on success; -1 on error
2526 ENODEV: no preferred target available for the host
2528 This queries KVM for preferred CPU target type which can be emulated
2529 by KVM on underlying host.
2531 The ioctl returns struct kvm_vcpu_init instance containing information
2532 about preferred CPU target type and recommended features for it. The
2533 kvm_vcpu_init->features bitmap returned will have feature bits set if
2534 the preferred target recommends setting these features, but this is
2537 The information returned by this ioctl can be used to prepare an instance
2538 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
2539 in VCPU matching underlying host.
2542 4.84 KVM_GET_REG_LIST
2545 Architectures: arm, arm64, mips
2547 Parameters: struct kvm_reg_list (in/out)
2548 Returns: 0 on success; -1 on error
2550 Â E2BIG: Â Â Â Â the reg index list is too big to fit in the array specified by
2551 Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n).
2553 struct kvm_reg_list {
2554 __u64 n; /* number of registers in reg[] */
2558 This ioctl returns the guest registers that are supported for the
2559 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
2562 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
2564 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
2565 Architectures: arm, arm64
2567 Parameters: struct kvm_arm_device_address (in)
2568 Returns: 0 on success, -1 on error
2570 ENODEV: The device id is unknown
2571 ENXIO: Device not supported on current system
2572 EEXIST: Address already set
2573 E2BIG: Address outside guest physical address space
2574 EBUSY: Address overlaps with other device range
2576 struct kvm_arm_device_addr {
2581 Specify a device address in the guest's physical address space where guests
2582 can access emulated or directly exposed devices, which the host kernel needs
2583 to know about. The id field is an architecture specific identifier for a
2586 ARM/arm64 divides the id field into two parts, a device id and an
2587 address type id specific to the individual device.
2589 Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
2590 field: | 0x00000000 | device id | addr type id |
2592 ARM/arm64 currently only require this when using the in-kernel GIC
2593 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
2594 as the device id. When setting the base address for the guest's
2595 mapping of the VGIC virtual CPU and distributor interface, the ioctl
2596 must be called after calling KVM_CREATE_IRQCHIP, but before calling
2597 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
2598 base addresses will return -EEXIST.
2600 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
2601 should be used instead.
2604 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
2606 Capability: KVM_CAP_PPC_RTAS
2609 Parameters: struct kvm_rtas_token_args
2610 Returns: 0 on success, -1 on error
2612 Defines a token value for a RTAS (Run Time Abstraction Services)
2613 service in order to allow it to be handled in the kernel. The
2614 argument struct gives the name of the service, which must be the name
2615 of a service that has a kernel-side implementation. If the token
2616 value is non-zero, it will be associated with that service, and
2617 subsequent RTAS calls by the guest specifying that token will be
2618 handled by the kernel. If the token value is 0, then any token
2619 associated with the service will be forgotten, and subsequent RTAS
2620 calls by the guest for that service will be passed to userspace to be
2623 4.87 KVM_SET_GUEST_DEBUG
2625 Capability: KVM_CAP_SET_GUEST_DEBUG
2626 Architectures: x86, s390, ppc
2628 Parameters: struct kvm_guest_debug (in)
2629 Returns: 0 on success; -1 on error
2631 struct kvm_guest_debug {
2634 struct kvm_guest_debug_arch arch;
2637 Set up the processor specific debug registers and configure vcpu for
2638 handling guest debug events. There are two parts to the structure, the
2639 first a control bitfield indicates the type of debug events to handle
2640 when running. Common control bits are:
2642 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
2643 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
2645 The top 16 bits of the control field are architecture specific control
2646 flags which can include the following:
2648 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86]
2649 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390]
2650 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
2651 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
2652 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
2654 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
2655 are enabled in memory so we need to ensure breakpoint exceptions are
2656 correctly trapped and the KVM run loop exits at the breakpoint and not
2657 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
2658 we need to ensure the guest vCPUs architecture specific registers are
2659 updated to the correct (supplied) values.
2661 The second part of the structure is architecture specific and
2662 typically contains a set of debug registers.
2664 When debug events exit the main run loop with the reason
2665 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
2666 structure containing architecture specific debug information.
2668 4.88 KVM_GET_EMULATED_CPUID
2670 Capability: KVM_CAP_EXT_EMUL_CPUID
2673 Parameters: struct kvm_cpuid2 (in/out)
2674 Returns: 0 on success, -1 on error
2679 struct kvm_cpuid_entry2 entries[0];
2682 The member 'flags' is used for passing flags from userspace.
2684 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
2685 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1)
2686 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2)
2688 struct kvm_cpuid_entry2 {
2699 This ioctl returns x86 cpuid features which are emulated by
2700 kvm.Userspace can use the information returned by this ioctl to query
2701 which features are emulated by kvm instead of being present natively.
2703 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
2704 structure with the 'nent' field indicating the number of entries in
2705 the variable-size array 'entries'. If the number of entries is too low
2706 to describe the cpu capabilities, an error (E2BIG) is returned. If the
2707 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
2708 is returned. If the number is just right, the 'nent' field is adjusted
2709 to the number of valid entries in the 'entries' array, which is then
2712 The entries returned are the set CPUID bits of the respective features
2713 which kvm emulates, as returned by the CPUID instruction, with unknown
2714 or unsupported feature bits cleared.
2716 Features like x2apic, for example, may not be present in the host cpu
2717 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
2718 emulated efficiently and thus not included here.
2720 The fields in each entry are defined as follows:
2722 function: the eax value used to obtain the entry
2723 index: the ecx value used to obtain the entry (for entries that are
2725 flags: an OR of zero or more of the following:
2726 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
2727 if the index field is valid
2728 KVM_CPUID_FLAG_STATEFUL_FUNC:
2729 if cpuid for this function returns different values for successive
2730 invocations; there will be several entries with the same function,
2731 all with this flag set
2732 KVM_CPUID_FLAG_STATE_READ_NEXT:
2733 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
2734 the first entry to be read by a cpu
2735 eax, ebx, ecx, edx: the values returned by the cpuid instruction for
2736 this function/index combination
2738 4.89 KVM_S390_MEM_OP
2740 Capability: KVM_CAP_S390_MEM_OP
2743 Parameters: struct kvm_s390_mem_op (in)
2744 Returns: = 0 on success,
2745 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
2746 > 0 if an exception occurred while walking the page tables
2748 Read or write data from/to the logical (virtual) memory of a VPCU.
2750 Parameters are specified via the following structure:
2752 struct kvm_s390_mem_op {
2753 __u64 gaddr; /* the guest address */
2754 __u64 flags; /* flags */
2755 __u32 size; /* amount of bytes */
2756 __u32 op; /* type of operation */
2757 __u64 buf; /* buffer in userspace */
2758 __u8 ar; /* the access register number */
2759 __u8 reserved[31]; /* should be set to 0 */
2762 The type of operation is specified in the "op" field. It is either
2763 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or
2764 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The
2765 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check
2766 whether the corresponding memory access would create an access exception
2767 (without touching the data in the memory at the destination). In case an
2768 access exception occurred while walking the MMU tables of the guest, the
2769 ioctl returns a positive error number to indicate the type of exception.
2770 This exception is also raised directly at the corresponding VCPU if the
2771 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field.
2773 The start address of the memory region has to be specified in the "gaddr"
2774 field, and the length of the region in the "size" field. "buf" is the buffer
2775 supplied by the userspace application where the read data should be written
2776 to for KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written
2777 is stored for a KVM_S390_MEMOP_LOGICAL_WRITE. "buf" is unused and can be NULL
2778 when KVM_S390_MEMOP_F_CHECK_ONLY is specified. "ar" designates the access
2779 register number to be used.
2781 The "reserved" field is meant for future extensions. It is not used by
2782 KVM with the currently defined set of flags.
2784 4.90 KVM_S390_GET_SKEYS
2786 Capability: KVM_CAP_S390_SKEYS
2789 Parameters: struct kvm_s390_skeys
2790 Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage
2791 keys, negative value on error
2793 This ioctl is used to get guest storage key values on the s390
2794 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2796 struct kvm_s390_skeys {
2799 __u64 skeydata_addr;
2804 The start_gfn field is the number of the first guest frame whose storage keys
2807 The count field is the number of consecutive frames (starting from start_gfn)
2808 whose storage keys to get. The count field must be at least 1 and the maximum
2809 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2810 will cause the ioctl to return -EINVAL.
2812 The skeydata_addr field is the address to a buffer large enough to hold count
2813 bytes. This buffer will be filled with storage key data by the ioctl.
2815 4.91 KVM_S390_SET_SKEYS
2817 Capability: KVM_CAP_S390_SKEYS
2820 Parameters: struct kvm_s390_skeys
2821 Returns: 0 on success, negative value on error
2823 This ioctl is used to set guest storage key values on the s390
2824 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
2825 See section on KVM_S390_GET_SKEYS for struct definition.
2827 The start_gfn field is the number of the first guest frame whose storage keys
2830 The count field is the number of consecutive frames (starting from start_gfn)
2831 whose storage keys to get. The count field must be at least 1 and the maximum
2832 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range
2833 will cause the ioctl to return -EINVAL.
2835 The skeydata_addr field is the address to a buffer containing count bytes of
2836 storage keys. Each byte in the buffer will be set as the storage key for a
2837 single frame starting at start_gfn for count frames.
2839 Note: If any architecturally invalid key value is found in the given data then
2840 the ioctl will return -EINVAL.
2842 5. The kvm_run structure
2843 ------------------------
2845 Application code obtains a pointer to the kvm_run structure by
2846 mmap()ing a vcpu fd. From that point, application code can control
2847 execution by changing fields in kvm_run prior to calling the KVM_RUN
2848 ioctl, and obtain information about the reason KVM_RUN returned by
2849 looking up structure members.
2853 __u8 request_interrupt_window;
2855 Request that KVM_RUN return when it becomes possible to inject external
2856 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
2863 When KVM_RUN has returned successfully (return value 0), this informs
2864 application code why KVM_RUN has returned. Allowable values for this
2865 field are detailed below.
2867 __u8 ready_for_interrupt_injection;
2869 If request_interrupt_window has been specified, this field indicates
2870 an interrupt can be injected now with KVM_INTERRUPT.
2874 The value of the current interrupt flag. Only valid if in-kernel
2875 local APIC is not used.
2879 /* in (pre_kvm_run), out (post_kvm_run) */
2882 The value of the cr8 register. Only valid if in-kernel local APIC is
2883 not used. Both input and output.
2887 The value of the APIC BASE msr. Only valid if in-kernel local
2888 APIC is not used. Both input and output.
2891 /* KVM_EXIT_UNKNOWN */
2893 __u64 hardware_exit_reason;
2896 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
2897 reasons. Further architecture-specific information is available in
2898 hardware_exit_reason.
2900 /* KVM_EXIT_FAIL_ENTRY */
2902 __u64 hardware_entry_failure_reason;
2905 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
2906 to unknown reasons. Further architecture-specific information is
2907 available in hardware_entry_failure_reason.
2909 /* KVM_EXIT_EXCEPTION */
2919 #define KVM_EXIT_IO_IN 0
2920 #define KVM_EXIT_IO_OUT 1
2922 __u8 size; /* bytes */
2925 __u64 data_offset; /* relative to kvm_run start */
2928 If exit_reason is KVM_EXIT_IO, then the vcpu has
2929 executed a port I/O instruction which could not be satisfied by kvm.
2930 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
2931 where kvm expects application code to place the data for the next
2932 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
2935 struct kvm_debug_exit_arch arch;
2948 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
2949 executed a memory-mapped I/O instruction which could not be satisfied
2950 by kvm. The 'data' member contains the written data if 'is_write' is
2951 true, and should be filled by application code otherwise.
2953 The 'data' member contains, in its first 'len' bytes, the value as it would
2954 appear if the VCPU performed a load or store of the appropriate width directly
2957 NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and
2958 KVM_EXIT_EPR the corresponding
2959 operations are complete (and guest state is consistent) only after userspace
2960 has re-entered the kernel with KVM_RUN. The kernel side will first finish
2961 incomplete operations and then check for pending signals. Userspace
2962 can re-enter the guest with an unmasked signal pending to complete
2965 /* KVM_EXIT_HYPERCALL */
2974 Unused. This was once used for 'hypercall to userspace'. To implement
2975 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390).
2976 Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
2978 /* KVM_EXIT_TPR_ACCESS */
2985 To be documented (KVM_TPR_ACCESS_REPORTING).
2987 /* KVM_EXIT_S390_SIEIC */
2990 __u64 mask; /* psw upper half */
2991 __u64 addr; /* psw lower half */
2998 /* KVM_EXIT_S390_RESET */
2999 #define KVM_S390_RESET_POR 1
3000 #define KVM_S390_RESET_CLEAR 2
3001 #define KVM_S390_RESET_SUBSYSTEM 4
3002 #define KVM_S390_RESET_CPU_INIT 8
3003 #define KVM_S390_RESET_IPL 16
3004 __u64 s390_reset_flags;
3008 /* KVM_EXIT_S390_UCONTROL */
3010 __u64 trans_exc_code;
3014 s390 specific. A page fault has occurred for a user controlled virtual
3015 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be
3016 resolved by the kernel.
3017 The program code and the translation exception code that were placed
3018 in the cpu's lowcore are presented here as defined by the z Architecture
3019 Principles of Operation Book in the Chapter for Dynamic Address Translation
3029 Deprecated - was used for 440 KVM.
3036 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
3037 hypercalls and exit with this exit struct that contains all the guest gprs.
3039 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
3040 Userspace can now handle the hypercall and when it's done modify the gprs as
3041 necessary. Upon guest entry all guest GPRs will then be replaced by the values
3044 /* KVM_EXIT_PAPR_HCALL */
3051 This is used on 64-bit PowerPC when emulating a pSeries partition,
3052 e.g. with the 'pseries' machine type in qemu. It occurs when the
3053 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
3054 contains the hypercall number (from the guest R3), and 'args' contains
3055 the arguments (from the guest R4 - R12). Userspace should put the
3056 return code in 'ret' and any extra returned values in args[].
3057 The possible hypercalls are defined in the Power Architecture Platform
3058 Requirements (PAPR) document available from www.power.org (free
3059 developer registration required to access it).
3061 /* KVM_EXIT_S390_TSCH */
3063 __u16 subchannel_id;
3064 __u16 subchannel_nr;
3071 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
3072 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
3073 interrupt for the target subchannel has been dequeued and subchannel_id,
3074 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
3075 interrupt. ipb is needed for instruction parameter decoding.
3082 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
3083 interrupt acknowledge path to the core. When the core successfully
3084 delivers an interrupt, it automatically populates the EPR register with
3085 the interrupt vector number and acknowledges the interrupt inside
3086 the interrupt controller.
3088 In case the interrupt controller lives in user space, we need to do
3089 the interrupt acknowledge cycle through it to fetch the next to be
3090 delivered interrupt vector using this exit.
3092 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
3093 external interrupt has just been delivered into the guest. User space
3094 should put the acknowledged interrupt vector into the 'epr' field.
3096 /* KVM_EXIT_SYSTEM_EVENT */
3098 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
3099 #define KVM_SYSTEM_EVENT_RESET 2
3104 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
3105 a system-level event using some architecture specific mechanism (hypercall
3106 or some special instruction). In case of ARM/ARM64, this is triggered using
3107 HVC instruction based PSCI call from the vcpu. The 'type' field describes
3108 the system-level event type. The 'flags' field describes architecture
3109 specific flags for the system-level event.
3111 Valid values for 'type' are:
3112 KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
3113 VM. Userspace is not obliged to honour this, and if it does honour
3114 this does not need to destroy the VM synchronously (ie it may call
3115 KVM_RUN again before shutdown finally occurs).
3116 KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
3117 As with SHUTDOWN, userspace can choose to ignore the request, or
3118 to schedule the reset to occur in the future and may call KVM_RUN again.
3120 /* Fix the size of the union. */
3125 * shared registers between kvm and userspace.
3126 * kvm_valid_regs specifies the register classes set by the host
3127 * kvm_dirty_regs specified the register classes dirtied by userspace
3128 * struct kvm_sync_regs is architecture specific, as well as the
3129 * bits for kvm_valid_regs and kvm_dirty_regs
3131 __u64 kvm_valid_regs;
3132 __u64 kvm_dirty_regs;
3134 struct kvm_sync_regs regs;
3138 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
3139 certain guest registers without having to call SET/GET_*REGS. Thus we can
3140 avoid some system call overhead if userspace has to handle the exit.
3141 Userspace can query the validity of the structure by checking
3142 kvm_valid_regs for specific bits. These bits are architecture specific
3143 and usually define the validity of a groups of registers. (e.g. one bit
3144 for general purpose registers)
3146 Please note that the kernel is allowed to use the kvm_run structure as the
3147 primary storage for certain register types. Therefore, the kernel may use the
3148 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
3154 6. Capabilities that can be enabled on vCPUs
3155 --------------------------------------------
3157 There are certain capabilities that change the behavior of the virtual CPU or
3158 the virtual machine when enabled. To enable them, please see section 4.37.
3159 Below you can find a list of capabilities and what their effect on the vCPU or
3160 the virtual machine is when enabling them.
3162 The following information is provided along with the description:
3164 Architectures: which instruction set architectures provide this ioctl.
3165 x86 includes both i386 and x86_64.
3167 Target: whether this is a per-vcpu or per-vm capability.
3169 Parameters: what parameters are accepted by the capability.
3171 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3172 are not detailed, but errors with specific meanings are.
3180 Returns: 0 on success; -1 on error
3182 This capability enables interception of OSI hypercalls that otherwise would
3183 be treated as normal system calls to be injected into the guest. OSI hypercalls
3184 were invented by Mac-on-Linux to have a standardized communication mechanism
3185 between the guest and the host.
3187 When this capability is enabled, KVM_EXIT_OSI can occur.
3190 6.2 KVM_CAP_PPC_PAPR
3195 Returns: 0 on success; -1 on error
3197 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
3198 done using the hypercall instruction "sc 1".
3200 It also sets the guest privilege level to "supervisor" mode. Usually the guest
3201 runs in "hypervisor" privilege mode with a few missing features.
3203 In addition to the above, it changes the semantics of SDR1. In this mode, the
3204 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
3205 HTAB invisible to the guest.
3207 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
3214 Parameters: args[0] is the address of a struct kvm_config_tlb
3215 Returns: 0 on success; -1 on error
3217 struct kvm_config_tlb {
3224 Configures the virtual CPU's TLB array, establishing a shared memory area
3225 between userspace and KVM. The "params" and "array" fields are userspace
3226 addresses of mmu-type-specific data structures. The "array_len" field is an
3227 safety mechanism, and should be set to the size in bytes of the memory that
3228 userspace has reserved for the array. It must be at least the size dictated
3229 by "mmu_type" and "params".
3231 While KVM_RUN is active, the shared region is under control of KVM. Its
3232 contents are undefined, and any modification by userspace results in
3233 boundedly undefined behavior.
3235 On return from KVM_RUN, the shared region will reflect the current state of
3236 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
3237 to tell KVM which entries have been changed, prior to calling KVM_RUN again
3240 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
3241 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
3242 - The "array" field points to an array of type "struct
3243 kvm_book3e_206_tlb_entry".
3244 - The array consists of all entries in the first TLB, followed by all
3245 entries in the second TLB.
3246 - Within a TLB, entries are ordered first by increasing set number. Within a
3247 set, entries are ordered by way (increasing ESEL).
3248 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
3249 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
3250 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
3251 hardware ignores this value for TLB0.
3253 6.4 KVM_CAP_S390_CSS_SUPPORT
3258 Returns: 0 on success; -1 on error
3260 This capability enables support for handling of channel I/O instructions.
3262 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
3263 handled in-kernel, while the other I/O instructions are passed to userspace.
3265 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
3266 SUBCHANNEL intercepts.
3268 Note that even though this capability is enabled per-vcpu, the complete
3269 virtual machine is affected.
3275 Parameters: args[0] defines whether the proxy facility is active
3276 Returns: 0 on success; -1 on error
3278 This capability enables or disables the delivery of interrupts through the
3279 external proxy facility.
3281 When enabled (args[0] != 0), every time the guest gets an external interrupt
3282 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
3283 to receive the topmost interrupt vector.
3285 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
3287 When this capability is enabled, KVM_EXIT_EPR can occur.
3289 6.6 KVM_CAP_IRQ_MPIC
3292 Parameters: args[0] is the MPIC device fd
3293 args[1] is the MPIC CPU number for this vcpu
3295 This capability connects the vcpu to an in-kernel MPIC device.
3297 6.7 KVM_CAP_IRQ_XICS
3301 Parameters: args[0] is the XICS device fd
3302 args[1] is the XICS CPU number (server ID) for this vcpu
3304 This capability connects the vcpu to an in-kernel XICS device.
3306 6.8 KVM_CAP_S390_IRQCHIP
3312 This capability enables the in-kernel irqchip for s390. Please refer to
3313 "4.24 KVM_CREATE_IRQCHIP" for details.
3315 7. Capabilities that can be enabled on VMs
3316 ------------------------------------------
3318 There are certain capabilities that change the behavior of the virtual
3319 machine when enabled. To enable them, please see section 4.37. Below
3320 you can find a list of capabilities and what their effect on the VM
3321 is when enabling them.
3323 The following information is provided along with the description:
3325 Architectures: which instruction set architectures provide this ioctl.
3326 x86 includes both i386 and x86_64.
3328 Parameters: what parameters are accepted by the capability.
3330 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL)
3331 are not detailed, but errors with specific meanings are.
3334 7.1 KVM_CAP_PPC_ENABLE_HCALL
3337 Parameters: args[0] is the sPAPR hcall number
3338 args[1] is 0 to disable, 1 to enable in-kernel handling
3340 This capability controls whether individual sPAPR hypercalls (hcalls)
3341 get handled by the kernel or not. Enabling or disabling in-kernel
3342 handling of an hcall is effective across the VM. On creation, an
3343 initial set of hcalls are enabled for in-kernel handling, which
3344 consists of those hcalls for which in-kernel handlers were implemented
3345 before this capability was implemented. If disabled, the kernel will
3346 not to attempt to handle the hcall, but will always exit to userspace
3347 to handle it. Note that it may not make sense to enable some and
3348 disable others of a group of related hcalls, but KVM does not prevent
3349 userspace from doing that.
3351 If the hcall number specified is not one that has an in-kernel
3352 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
3355 7.2 KVM_CAP_S390_USER_SIGP
3360 This capability controls which SIGP orders will be handled completely in user
3361 space. With this capability enabled, all fast orders will be handled completely
3367 - CONDITIONAL EMERGENCY SIGNAL
3369 All other orders will be handled completely in user space.
3371 Only privileged operation exceptions will be checked for in the kernel (or even
3372 in the hardware prior to interception). If this capability is not enabled, the
3373 old way of handling SIGP orders is used (partially in kernel and user space).
3375 7.3 KVM_CAP_S390_VECTOR_REGISTERS
3379 Returns: 0 on success, negative value on error
3381 Allows use of the vector registers introduced with z13 processor, and
3382 provides for the synchronization between host and user space. Will
3383 return -EINVAL if the machine does not support vectors.
3385 7.4 KVM_CAP_S390_USER_STSI
3390 This capability allows post-handlers for the STSI instruction. After
3391 initial handling in the kernel, KVM exits to user space with
3392 KVM_EXIT_S390_STSI to allow user space to insert further data.
3394 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
3405 @addr - guest address of STSI SYSIB
3409 @ar - access register number
3411 KVM handlers should exit to userspace with rc = -EREMOTE.