1The Definitive KVM (Kernel-based Virtual Machine) API Documentation 2=================================================================== 3 41. General description 5---------------------- 6 7The kvm API is a set of ioctls that are issued to control various aspects 8of a virtual machine. The ioctls belong to three classes 9 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 12 virtual machines 13 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). 17 18 Only run VM ioctls from the same process (address space) that was used 19 to create the VM. 20 21 - vcpu ioctls: These query and set attributes that control the operation 22 of a single virtual cpu. 23 24 Only run vcpu ioctls from the same thread that was used to create the 25 vcpu. 26 27 282. File descriptors 29------------------- 30 31The kvm API is centered around file descriptors. An initial 32open("/dev/kvm") obtains a handle to the kvm subsystem; this handle 33can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this 34handle will create a VM file descriptor which can be used to issue VM 35ioctls. A KVM_CREATE_VCPU ioctl on a VM fd will create a virtual cpu 36and return a file descriptor pointing to it. Finally, ioctls on a vcpu 37fd can be used to control the vcpu, including the important task of 38actually running guest code. 39 40In general file descriptors can be migrated among processes by means 41of fork() and the SCM_RIGHTS facility of unix domain socket. These 42kinds of tricks are explicitly not supported by kvm. While they will 43not cause harm to the host, their actual behavior is not guaranteed by 44the API. The only supported use is one virtual machine per process, 45and one vcpu per thread. 46 47 483. Extensions 49------------- 50 51As of Linux 2.6.22, the KVM ABI has been stabilized: no backward 52incompatible change are allowed. However, there is an extension 53facility that allows backward-compatible extensions to the API to be 54queried and used. 55 56The extension mechanism is not based on on the Linux version number. 57Instead, kvm defines extension identifiers and a facility to query 58whether a particular extension identifier is available. If it is, a 59set of ioctls is available for application use. 60 61 624. API description 63------------------ 64 65This section describes ioctls that can be used to control kvm guests. 66For each ioctl, the following information is provided along with a 67description: 68 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), or a KVM_CAP_xyz constant, which 72 means availability needs to be checked with KVM_CHECK_EXTENSION 73 (see section 4.4). 74 75 Architectures: which instruction set architectures provide this ioctl. 76 x86 includes both i386 and x86_64. 77 78 Type: system, vm, or vcpu. 79 80 Parameters: what parameters are accepted by the ioctl. 81 82 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) 83 are not detailed, but errors with specific meanings are. 84 85 864.1 KVM_GET_API_VERSION 87 88Capability: basic 89Architectures: all 90Type: system ioctl 91Parameters: none 92Returns: the constant KVM_API_VERSION (=12) 93 94This identifies the API version as the stable kvm API. It is not 95expected that this number will change. However, Linux 2.6.20 and 962.6.21 report earlier versions; these are not documented and not 97supported. Applications should refuse to run if KVM_GET_API_VERSION 98returns a value other than 12. If this check passes, all ioctls 99described as 'basic' will be available. 100 101 1024.2 KVM_CREATE_VM 103 104Capability: basic 105Architectures: all 106Type: system ioctl 107Parameters: machine type identifier (KVM_VM_*) 108Returns: a VM fd that can be used to control the new virtual machine. 109 110The new VM has no virtual cpus and no memory. An mmap() of a VM fd 111will access the virtual machine's physical address space; offset zero 112corresponds to guest physical address zero. Use of mmap() on a VM fd 113is discouraged if userspace memory allocation (KVM_CAP_USER_MEMORY) is 114available. 115You most certainly want to use 0 as machine type. 116 117In order to create user controlled virtual machines on S390, check 118KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as 119privileged user (CAP_SYS_ADMIN). 120 121 1224.3 KVM_GET_MSR_INDEX_LIST 123 124Capability: basic 125Architectures: x86 126Type: system 127Parameters: struct kvm_msr_list (in/out) 128Returns: 0 on success; -1 on error 129Errors: 130 E2BIG: the msr index list is to be to fit in the array specified by 131 the user. 132 133struct kvm_msr_list { 134 __u32 nmsrs; /* number of msrs in entries */ 135 __u32 indices[0]; 136}; 137 138This ioctl returns the guest msrs that are supported. The list varies 139by kvm version and host processor, but does not change otherwise. The 140user fills in the size of the indices array in nmsrs, and in return 141kvm adjusts nmsrs to reflect the actual number of msrs and fills in 142the indices array with their numbers. 143 144Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 145not returned in the MSR list, as different vcpus can have a different number 146of banks, as set via the KVM_X86_SETUP_MCE ioctl. 147 148 1494.4 KVM_CHECK_EXTENSION 150 151Capability: basic 152Architectures: all 153Type: system ioctl 154Parameters: extension identifier (KVM_CAP_*) 155Returns: 0 if unsupported; 1 (or some other positive integer) if supported 156 157The API allows the application to query about extensions to the core 158kvm API. Userspace passes an extension identifier (an integer) and 159receives an integer that describes the extension availability. 160Generally 0 means no and 1 means yes, but some extensions may report 161additional information in the integer return value. 162 163 1644.5 KVM_GET_VCPU_MMAP_SIZE 165 166Capability: basic 167Architectures: all 168Type: system ioctl 169Parameters: none 170Returns: size of vcpu mmap area, in bytes 171 172The KVM_RUN ioctl (cf.) communicates with userspace via a shared 173memory region. This ioctl returns the size of that region. See the 174KVM_RUN documentation for details. 175 176 1774.6 KVM_SET_MEMORY_REGION 178 179Capability: basic 180Architectures: all 181Type: vm ioctl 182Parameters: struct kvm_memory_region (in) 183Returns: 0 on success, -1 on error 184 185This ioctl is obsolete and has been removed. 186 187 1884.7 KVM_CREATE_VCPU 189 190Capability: basic 191Architectures: all 192Type: vm ioctl 193Parameters: vcpu id (apic id on x86) 194Returns: vcpu fd on success, -1 on error 195 196This API adds a vcpu to a virtual machine. The vcpu id is a small integer 197in the range [0, max_vcpus). 198 199The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 200the KVM_CHECK_EXTENSION ioctl() at run-time. 201The maximum possible value for max_vcpus can be retrieved using the 202KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 203 204If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 205cpus max. 206If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 207same as the value returned from KVM_CAP_NR_VCPUS. 208 209On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 210threads in one or more virtual CPU cores. (This is because the 211hardware requires all the hardware threads in a CPU core to be in the 212same partition.) The KVM_CAP_PPC_SMT capability indicates the number 213of vcpus per virtual core (vcore). The vcore id is obtained by 214dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 215given vcore will always be in the same physical core as each other 216(though that might be a different physical core from time to time). 217Userspace can control the threading (SMT) mode of the guest by its 218allocation of vcpu ids. For example, if userspace wants 219single-threaded guest vcpus, it should make all vcpu ids be a multiple 220of the number of vcpus per vcore. 221 222For virtual cpus that have been created with S390 user controlled virtual 223machines, the resulting vcpu fd can be memory mapped at page offset 224KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 225cpu's hardware control block. 226 227 2284.8 KVM_GET_DIRTY_LOG (vm ioctl) 229 230Capability: basic 231Architectures: x86 232Type: vm ioctl 233Parameters: struct kvm_dirty_log (in/out) 234Returns: 0 on success, -1 on error 235 236/* for KVM_GET_DIRTY_LOG */ 237struct kvm_dirty_log { 238 __u32 slot; 239 __u32 padding; 240 union { 241 void __user *dirty_bitmap; /* one bit per page */ 242 __u64 padding; 243 }; 244}; 245 246Given a memory slot, return a bitmap containing any pages dirtied 247since the last call to this ioctl. Bit 0 is the first page in the 248memory slot. Ensure the entire structure is cleared to avoid padding 249issues. 250 251 2524.9 KVM_SET_MEMORY_ALIAS 253 254Capability: basic 255Architectures: x86 256Type: vm ioctl 257Parameters: struct kvm_memory_alias (in) 258Returns: 0 (success), -1 (error) 259 260This ioctl is obsolete and has been removed. 261 262 2634.10 KVM_RUN 264 265Capability: basic 266Architectures: all 267Type: vcpu ioctl 268Parameters: none 269Returns: 0 on success, -1 on error 270Errors: 271 EINTR: an unmasked signal is pending 272 273This ioctl is used to run a guest virtual cpu. While there are no 274explicit parameters, there is an implicit parameter block that can be 275obtained by mmap()ing the vcpu fd at offset 0, with the size given by 276KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct 277kvm_run' (see below). 278 279 2804.11 KVM_GET_REGS 281 282Capability: basic 283Architectures: all except ARM 284Type: vcpu ioctl 285Parameters: struct kvm_regs (out) 286Returns: 0 on success, -1 on error 287 288Reads the general purpose registers from the vcpu. 289 290/* x86 */ 291struct kvm_regs { 292 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 293 __u64 rax, rbx, rcx, rdx; 294 __u64 rsi, rdi, rsp, rbp; 295 __u64 r8, r9, r10, r11; 296 __u64 r12, r13, r14, r15; 297 __u64 rip, rflags; 298}; 299 300 3014.12 KVM_SET_REGS 302 303Capability: basic 304Architectures: all except ARM 305Type: vcpu ioctl 306Parameters: struct kvm_regs (in) 307Returns: 0 on success, -1 on error 308 309Writes the general purpose registers into the vcpu. 310 311See KVM_GET_REGS for the data structure. 312 313 3144.13 KVM_GET_SREGS 315 316Capability: basic 317Architectures: x86, ppc 318Type: vcpu ioctl 319Parameters: struct kvm_sregs (out) 320Returns: 0 on success, -1 on error 321 322Reads special registers from the vcpu. 323 324/* x86 */ 325struct kvm_sregs { 326 struct kvm_segment cs, ds, es, fs, gs, ss; 327 struct kvm_segment tr, ldt; 328 struct kvm_dtable gdt, idt; 329 __u64 cr0, cr2, cr3, cr4, cr8; 330 __u64 efer; 331 __u64 apic_base; 332 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; 333}; 334 335/* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ 336 337interrupt_bitmap is a bitmap of pending external interrupts. At most 338one bit may be set. This interrupt has been acknowledged by the APIC 339but not yet injected into the cpu core. 340 341 3424.14 KVM_SET_SREGS 343 344Capability: basic 345Architectures: x86, ppc 346Type: vcpu ioctl 347Parameters: struct kvm_sregs (in) 348Returns: 0 on success, -1 on error 349 350Writes special registers into the vcpu. See KVM_GET_SREGS for the 351data structures. 352 353 3544.15 KVM_TRANSLATE 355 356Capability: basic 357Architectures: x86 358Type: vcpu ioctl 359Parameters: struct kvm_translation (in/out) 360Returns: 0 on success, -1 on error 361 362Translates a virtual address according to the vcpu's current address 363translation mode. 364 365struct kvm_translation { 366 /* in */ 367 __u64 linear_address; 368 369 /* out */ 370 __u64 physical_address; 371 __u8 valid; 372 __u8 writeable; 373 __u8 usermode; 374 __u8 pad[5]; 375}; 376 377 3784.16 KVM_INTERRUPT 379 380Capability: basic 381Architectures: x86, ppc 382Type: vcpu ioctl 383Parameters: struct kvm_interrupt (in) 384Returns: 0 on success, -1 on error 385 386Queues a hardware interrupt vector to be injected. This is only 387useful if in-kernel local APIC or equivalent is not used. 388 389/* for KVM_INTERRUPT */ 390struct kvm_interrupt { 391 /* in */ 392 __u32 irq; 393}; 394 395X86: 396 397Note 'irq' is an interrupt vector, not an interrupt pin or line. 398 399PPC: 400 401Queues an external interrupt to be injected. This ioctl is overleaded 402with 3 different irq values: 403 404a) KVM_INTERRUPT_SET 405 406 This injects an edge type external interrupt into the guest once it's ready 407 to receive interrupts. When injected, the interrupt is done. 408 409b) KVM_INTERRUPT_UNSET 410 411 This unsets any pending interrupt. 412 413 Only available with KVM_CAP_PPC_UNSET_IRQ. 414 415c) KVM_INTERRUPT_SET_LEVEL 416 417 This injects a level type external interrupt into the guest context. The 418 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET 419 is triggered. 420 421 Only available with KVM_CAP_PPC_IRQ_LEVEL. 422 423Note that any value for 'irq' other than the ones stated above is invalid 424and incurs unexpected behavior. 425 426 4274.17 KVM_DEBUG_GUEST 428 429Capability: basic 430Architectures: none 431Type: vcpu ioctl 432Parameters: none) 433Returns: -1 on error 434 435Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. 436 437 4384.18 KVM_GET_MSRS 439 440Capability: basic 441Architectures: x86 442Type: vcpu ioctl 443Parameters: struct kvm_msrs (in/out) 444Returns: 0 on success, -1 on error 445 446Reads model-specific registers from the vcpu. Supported msr indices can 447be obtained using KVM_GET_MSR_INDEX_LIST. 448 449struct kvm_msrs { 450 __u32 nmsrs; /* number of msrs in entries */ 451 __u32 pad; 452 453 struct kvm_msr_entry entries[0]; 454}; 455 456struct kvm_msr_entry { 457 __u32 index; 458 __u32 reserved; 459 __u64 data; 460}; 461 462Application code should set the 'nmsrs' member (which indicates the 463size of the entries array) and the 'index' member of each array entry. 464kvm will fill in the 'data' member. 465 466 4674.19 KVM_SET_MSRS 468 469Capability: basic 470Architectures: x86 471Type: vcpu ioctl 472Parameters: struct kvm_msrs (in) 473Returns: 0 on success, -1 on error 474 475Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 476data structures. 477 478Application code should set the 'nmsrs' member (which indicates the 479size of the entries array), and the 'index' and 'data' members of each 480array entry. 481 482 4834.20 KVM_SET_CPUID 484 485Capability: basic 486Architectures: x86 487Type: vcpu ioctl 488Parameters: struct kvm_cpuid (in) 489Returns: 0 on success, -1 on error 490 491Defines the vcpu responses to the cpuid instruction. Applications 492should use the KVM_SET_CPUID2 ioctl if available. 493 494 495struct kvm_cpuid_entry { 496 __u32 function; 497 __u32 eax; 498 __u32 ebx; 499 __u32 ecx; 500 __u32 edx; 501 __u32 padding; 502}; 503 504/* for KVM_SET_CPUID */ 505struct kvm_cpuid { 506 __u32 nent; 507 __u32 padding; 508 struct kvm_cpuid_entry entries[0]; 509}; 510 511 5124.21 KVM_SET_SIGNAL_MASK 513 514Capability: basic 515Architectures: x86 516Type: vcpu ioctl 517Parameters: struct kvm_signal_mask (in) 518Returns: 0 on success, -1 on error 519 520Defines which signals are blocked during execution of KVM_RUN. This 521signal mask temporarily overrides the threads signal mask. Any 522unblocked signal received (except SIGKILL and SIGSTOP, which retain 523their traditional behaviour) will cause KVM_RUN to return with -EINTR. 524 525Note the signal will only be delivered if not blocked by the original 526signal mask. 527 528/* for KVM_SET_SIGNAL_MASK */ 529struct kvm_signal_mask { 530 __u32 len; 531 __u8 sigset[0]; 532}; 533 534 5354.22 KVM_GET_FPU 536 537Capability: basic 538Architectures: x86 539Type: vcpu ioctl 540Parameters: struct kvm_fpu (out) 541Returns: 0 on success, -1 on error 542 543Reads the floating point state from the vcpu. 544 545/* for KVM_GET_FPU and KVM_SET_FPU */ 546struct kvm_fpu { 547 __u8 fpr[8][16]; 548 __u16 fcw; 549 __u16 fsw; 550 __u8 ftwx; /* in fxsave format */ 551 __u8 pad1; 552 __u16 last_opcode; 553 __u64 last_ip; 554 __u64 last_dp; 555 __u8 xmm[16][16]; 556 __u32 mxcsr; 557 __u32 pad2; 558}; 559 560 5614.23 KVM_SET_FPU 562 563Capability: basic 564Architectures: x86 565Type: vcpu ioctl 566Parameters: struct kvm_fpu (in) 567Returns: 0 on success, -1 on error 568 569Writes the floating point state to the vcpu. 570 571/* for KVM_GET_FPU and KVM_SET_FPU */ 572struct kvm_fpu { 573 __u8 fpr[8][16]; 574 __u16 fcw; 575 __u16 fsw; 576 __u8 ftwx; /* in fxsave format */ 577 __u8 pad1; 578 __u16 last_opcode; 579 __u64 last_ip; 580 __u64 last_dp; 581 __u8 xmm[16][16]; 582 __u32 mxcsr; 583 __u32 pad2; 584}; 585 586 5874.24 KVM_CREATE_IRQCHIP 588 589Capability: KVM_CAP_IRQCHIP 590Architectures: x86, ia64, ARM 591Type: vm ioctl 592Parameters: none 593Returns: 0 on success, -1 on error 594 595Creates an interrupt controller model in the kernel. On x86, creates a virtual 596ioapic, a virtual PIC (two PICs, nested), and sets up future vcpus to have a 597local APIC. IRQ routing for GSIs 0-15 is set to both PIC and IOAPIC; GSI 16-23 598only go to the IOAPIC. On ia64, a IOSAPIC is created. On ARM, a GIC is 599created. 600 601 6024.25 KVM_IRQ_LINE 603 604Capability: KVM_CAP_IRQCHIP 605Architectures: x86, ia64, arm 606Type: vm ioctl 607Parameters: struct kvm_irq_level 608Returns: 0 on success, -1 on error 609 610Sets the level of a GSI input to the interrupt controller model in the kernel. 611On some architectures it is required that an interrupt controller model has 612been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 613interrupts require the level to be set to 1 and then back to 0. 614 615ARM can signal an interrupt either at the CPU level, or at the in-kernel irqchip 616(GIC), and for in-kernel irqchip can tell the GIC to use PPIs designated for 617specific cpus. The irq field is interpreted like this: 618 619 bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 | 620 field: | irq_type | vcpu_index | irq_id | 621 622The irq_type field has the following values: 623- irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 624- irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 625 (the vcpu_index field is ignored) 626- irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 627 628(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 629 630In both cases, level is used to raise/lower the line. 631 632struct kvm_irq_level { 633 union { 634 __u32 irq; /* GSI */ 635 __s32 status; /* not used for KVM_IRQ_LEVEL */ 636 }; 637 __u32 level; /* 0 or 1 */ 638}; 639 640 6414.26 KVM_GET_IRQCHIP 642 643Capability: KVM_CAP_IRQCHIP 644Architectures: x86, ia64 645Type: vm ioctl 646Parameters: struct kvm_irqchip (in/out) 647Returns: 0 on success, -1 on error 648 649Reads the state of a kernel interrupt controller created with 650KVM_CREATE_IRQCHIP into a buffer provided by the caller. 651 652struct kvm_irqchip { 653 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 654 __u32 pad; 655 union { 656 char dummy[512]; /* reserving space */ 657 struct kvm_pic_state pic; 658 struct kvm_ioapic_state ioapic; 659 } chip; 660}; 661 662 6634.27 KVM_SET_IRQCHIP 664 665Capability: KVM_CAP_IRQCHIP 666Architectures: x86, ia64 667Type: vm ioctl 668Parameters: struct kvm_irqchip (in) 669Returns: 0 on success, -1 on error 670 671Sets the state of a kernel interrupt controller created with 672KVM_CREATE_IRQCHIP from a buffer provided by the caller. 673 674struct kvm_irqchip { 675 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 676 __u32 pad; 677 union { 678 char dummy[512]; /* reserving space */ 679 struct kvm_pic_state pic; 680 struct kvm_ioapic_state ioapic; 681 } chip; 682}; 683 684 6854.28 KVM_XEN_HVM_CONFIG 686 687Capability: KVM_CAP_XEN_HVM 688Architectures: x86 689Type: vm ioctl 690Parameters: struct kvm_xen_hvm_config (in) 691Returns: 0 on success, -1 on error 692 693Sets the MSR that the Xen HVM guest uses to initialize its hypercall 694page, and provides the starting address and size of the hypercall 695blobs in userspace. When the guest writes the MSR, kvm copies one 696page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 697memory. 698 699struct kvm_xen_hvm_config { 700 __u32 flags; 701 __u32 msr; 702 __u64 blob_addr_32; 703 __u64 blob_addr_64; 704 __u8 blob_size_32; 705 __u8 blob_size_64; 706 __u8 pad2[30]; 707}; 708 709 7104.29 KVM_GET_CLOCK 711 712Capability: KVM_CAP_ADJUST_CLOCK 713Architectures: x86 714Type: vm ioctl 715Parameters: struct kvm_clock_data (out) 716Returns: 0 on success, -1 on error 717 718Gets the current timestamp of kvmclock as seen by the current guest. In 719conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 720such as migration. 721 722struct kvm_clock_data { 723 __u64 clock; /* kvmclock current value */ 724 __u32 flags; 725 __u32 pad[9]; 726}; 727 728 7294.30 KVM_SET_CLOCK 730 731Capability: KVM_CAP_ADJUST_CLOCK 732Architectures: x86 733Type: vm ioctl 734Parameters: struct kvm_clock_data (in) 735Returns: 0 on success, -1 on error 736 737Sets the current timestamp of kvmclock to the value specified in its parameter. 738In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 739such as migration. 740 741struct kvm_clock_data { 742 __u64 clock; /* kvmclock current value */ 743 __u32 flags; 744 __u32 pad[9]; 745}; 746 747 7484.31 KVM_GET_VCPU_EVENTS 749 750Capability: KVM_CAP_VCPU_EVENTS 751Extended by: KVM_CAP_INTR_SHADOW 752Architectures: x86 753Type: vm ioctl 754Parameters: struct kvm_vcpu_event (out) 755Returns: 0 on success, -1 on error 756 757Gets currently pending exceptions, interrupts, and NMIs as well as related 758states of the vcpu. 759 760struct kvm_vcpu_events { 761 struct { 762 __u8 injected; 763 __u8 nr; 764 __u8 has_error_code; 765 __u8 pad; 766 __u32 error_code; 767 } exception; 768 struct { 769 __u8 injected; 770 __u8 nr; 771 __u8 soft; 772 __u8 shadow; 773 } interrupt; 774 struct { 775 __u8 injected; 776 __u8 pending; 777 __u8 masked; 778 __u8 pad; 779 } nmi; 780 __u32 sipi_vector; 781 __u32 flags; 782}; 783 784KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that 785interrupt.shadow contains a valid state. Otherwise, this field is undefined. 786 787 7884.32 KVM_SET_VCPU_EVENTS 789 790Capability: KVM_CAP_VCPU_EVENTS 791Extended by: KVM_CAP_INTR_SHADOW 792Architectures: x86 793Type: vm ioctl 794Parameters: struct kvm_vcpu_event (in) 795Returns: 0 on success, -1 on error 796 797Set pending exceptions, interrupts, and NMIs as well as related states of the 798vcpu. 799 800See KVM_GET_VCPU_EVENTS for the data structure. 801 802Fields that may be modified asynchronously by running VCPUs can be excluded 803from the update. These fields are nmi.pending and sipi_vector. Keep the 804corresponding bits in the flags field cleared to suppress overwriting the 805current in-kernel state. The bits are: 806 807KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel 808KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector 809 810If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 811the flags field to signal that interrupt.shadow contains a valid state and 812shall be written into the VCPU. 813 814 8154.33 KVM_GET_DEBUGREGS 816 817Capability: KVM_CAP_DEBUGREGS 818Architectures: x86 819Type: vm ioctl 820Parameters: struct kvm_debugregs (out) 821Returns: 0 on success, -1 on error 822 823Reads debug registers from the vcpu. 824 825struct kvm_debugregs { 826 __u64 db[4]; 827 __u64 dr6; 828 __u64 dr7; 829 __u64 flags; 830 __u64 reserved[9]; 831}; 832 833 8344.34 KVM_SET_DEBUGREGS 835 836Capability: KVM_CAP_DEBUGREGS 837Architectures: x86 838Type: vm ioctl 839Parameters: struct kvm_debugregs (in) 840Returns: 0 on success, -1 on error 841 842Writes debug registers into the vcpu. 843 844See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 845yet and must be cleared on entry. 846 847 8484.35 KVM_SET_USER_MEMORY_REGION 849 850Capability: KVM_CAP_USER_MEM 851Architectures: all 852Type: vm ioctl 853Parameters: struct kvm_userspace_memory_region (in) 854Returns: 0 on success, -1 on error 855 856struct kvm_userspace_memory_region { 857 __u32 slot; 858 __u32 flags; 859 __u64 guest_phys_addr; 860 __u64 memory_size; /* bytes */ 861 __u64 userspace_addr; /* start of the userspace allocated memory */ 862}; 863 864/* for kvm_memory_region::flags */ 865#define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 866#define KVM_MEM_READONLY (1UL << 1) 867 868This ioctl allows the user to create or modify a guest physical memory 869slot. When changing an existing slot, it may be moved in the guest 870physical memory space, or its flags may be modified. It may not be 871resized. Slots may not overlap in guest physical address space. 872 873Memory for the region is taken starting at the address denoted by the 874field userspace_addr, which must point at user addressable memory for 875the entire memory slot size. Any object may back this memory, including 876anonymous memory, ordinary files, and hugetlbfs. 877 878It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 879be identical. This allows large pages in the guest to be backed by large 880pages in the host. 881 882The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 883KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 884writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 885use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 886to make a new slot read-only. In this case, writes to this memory will be 887posted to userspace as KVM_EXIT_MMIO exits. 888 889When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 890the memory region are automatically reflected into the guest. For example, an 891mmap() that affects the region will be made visible immediately. Another 892example is madvise(MADV_DROP). 893 894It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. 895The KVM_SET_MEMORY_REGION does not allow fine grained control over memory 896allocation and is deprecated. 897 898 8994.36 KVM_SET_TSS_ADDR 900 901Capability: KVM_CAP_SET_TSS_ADDR 902Architectures: x86 903Type: vm ioctl 904Parameters: unsigned long tss_address (in) 905Returns: 0 on success, -1 on error 906 907This ioctl defines the physical address of a three-page region in the guest 908physical address space. The region must be within the first 4GB of the 909guest physical address space and must not conflict with any memory slot 910or any mmio address. The guest may malfunction if it accesses this memory 911region. 912 913This ioctl is required on Intel-based hosts. This is needed on Intel hardware 914because of a quirk in the virtualization implementation (see the internals 915documentation when it pops into existence). 916 917 9184.37 KVM_ENABLE_CAP 919 920Capability: KVM_CAP_ENABLE_CAP 921Architectures: ppc, s390 922Type: vcpu ioctl 923Parameters: struct kvm_enable_cap (in) 924Returns: 0 on success; -1 on error 925 926+Not all extensions are enabled by default. Using this ioctl the application 927can enable an extension, making it available to the guest. 928 929On systems that do not support this ioctl, it always fails. On systems that 930do support it, it only works for extensions that are supported for enablement. 931 932To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 933be used. 934 935struct kvm_enable_cap { 936 /* in */ 937 __u32 cap; 938 939The capability that is supposed to get enabled. 940 941 __u32 flags; 942 943A bitfield indicating future enhancements. Has to be 0 for now. 944 945 __u64 args[4]; 946 947Arguments for enabling a feature. If a feature needs initial values to 948function properly, this is the place to put them. 949 950 __u8 pad[64]; 951}; 952 953 9544.38 KVM_GET_MP_STATE 955 956Capability: KVM_CAP_MP_STATE 957Architectures: x86, ia64 958Type: vcpu ioctl 959Parameters: struct kvm_mp_state (out) 960Returns: 0 on success; -1 on error 961 962struct kvm_mp_state { 963 __u32 mp_state; 964}; 965 966Returns the vcpu's current "multiprocessing state" (though also valid on 967uniprocessor guests). 968 969Possible values are: 970 971 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running 972 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP) 973 which has not yet received an INIT signal 974 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is 975 now ready for a SIPI 976 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and 977 is waiting for an interrupt 978 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector 979 accessible via KVM_GET_VCPU_EVENTS) 980 981This ioctl is only useful after KVM_CREATE_IRQCHIP. Without an in-kernel 982irqchip, the multiprocessing state must be maintained by userspace. 983 984 9854.39 KVM_SET_MP_STATE 986 987Capability: KVM_CAP_MP_STATE 988Architectures: x86, ia64 989Type: vcpu ioctl 990Parameters: struct kvm_mp_state (in) 991Returns: 0 on success; -1 on error 992 993Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 994arguments. 995 996This ioctl is only useful after KVM_CREATE_IRQCHIP. Without an in-kernel 997irqchip, the multiprocessing state must be maintained by userspace. 998 999 10004.40 KVM_SET_IDENTITY_MAP_ADDR 1001 1002Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1003Architectures: x86 1004Type: vm ioctl 1005Parameters: unsigned long identity (in) 1006Returns: 0 on success, -1 on error 1007 1008This ioctl defines the physical address of a one-page region in the guest 1009physical address space. The region must be within the first 4GB of the 1010guest physical address space and must not conflict with any memory slot 1011or any mmio address. The guest may malfunction if it accesses this memory 1012region. 1013 1014This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1015because of a quirk in the virtualization implementation (see the internals 1016documentation when it pops into existence). 1017 1018 10194.41 KVM_SET_BOOT_CPU_ID 1020 1021Capability: KVM_CAP_SET_BOOT_CPU_ID 1022Architectures: x86, ia64 1023Type: vm ioctl 1024Parameters: unsigned long vcpu_id 1025Returns: 0 on success, -1 on error 1026 1027Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1028as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1029is vcpu 0. 1030 1031 10324.42 KVM_GET_XSAVE 1033 1034Capability: KVM_CAP_XSAVE 1035Architectures: x86 1036Type: vcpu ioctl 1037Parameters: struct kvm_xsave (out) 1038Returns: 0 on success, -1 on error 1039 1040struct kvm_xsave { 1041 __u32 region[1024]; 1042}; 1043 1044This ioctl would copy current vcpu's xsave struct to the userspace. 1045 1046 10474.43 KVM_SET_XSAVE 1048 1049Capability: KVM_CAP_XSAVE 1050Architectures: x86 1051Type: vcpu ioctl 1052Parameters: struct kvm_xsave (in) 1053Returns: 0 on success, -1 on error 1054 1055struct kvm_xsave { 1056 __u32 region[1024]; 1057}; 1058 1059This ioctl would copy userspace's xsave struct to the kernel. 1060 1061 10624.44 KVM_GET_XCRS 1063 1064Capability: KVM_CAP_XCRS 1065Architectures: x86 1066Type: vcpu ioctl 1067Parameters: struct kvm_xcrs (out) 1068Returns: 0 on success, -1 on error 1069 1070struct kvm_xcr { 1071 __u32 xcr; 1072 __u32 reserved; 1073 __u64 value; 1074}; 1075 1076struct kvm_xcrs { 1077 __u32 nr_xcrs; 1078 __u32 flags; 1079 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1080 __u64 padding[16]; 1081}; 1082 1083This ioctl would copy current vcpu's xcrs to the userspace. 1084 1085 10864.45 KVM_SET_XCRS 1087 1088Capability: KVM_CAP_XCRS 1089Architectures: x86 1090Type: vcpu ioctl 1091Parameters: struct kvm_xcrs (in) 1092Returns: 0 on success, -1 on error 1093 1094struct kvm_xcr { 1095 __u32 xcr; 1096 __u32 reserved; 1097 __u64 value; 1098}; 1099 1100struct kvm_xcrs { 1101 __u32 nr_xcrs; 1102 __u32 flags; 1103 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1104 __u64 padding[16]; 1105}; 1106 1107This ioctl would set vcpu's xcr to the value userspace specified. 1108 1109 11104.46 KVM_GET_SUPPORTED_CPUID 1111 1112Capability: KVM_CAP_EXT_CPUID 1113Architectures: x86 1114Type: system ioctl 1115Parameters: struct kvm_cpuid2 (in/out) 1116Returns: 0 on success, -1 on error 1117 1118struct kvm_cpuid2 { 1119 __u32 nent; 1120 __u32 padding; 1121 struct kvm_cpuid_entry2 entries[0]; 1122}; 1123 1124#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 1 1125#define KVM_CPUID_FLAG_STATEFUL_FUNC 2 1126#define KVM_CPUID_FLAG_STATE_READ_NEXT 4 1127 1128struct kvm_cpuid_entry2 { 1129 __u32 function; 1130 __u32 index; 1131 __u32 flags; 1132 __u32 eax; 1133 __u32 ebx; 1134 __u32 ecx; 1135 __u32 edx; 1136 __u32 padding[3]; 1137}; 1138 1139This ioctl returns x86 cpuid features which are supported by both the hardware 1140and kvm. Userspace can use the information returned by this ioctl to 1141construct cpuid information (for KVM_SET_CPUID2) that is consistent with 1142hardware, kernel, and userspace capabilities, and with user requirements (for 1143example, the user may wish to constrain cpuid to emulate older hardware, 1144or for feature consistency across a cluster). 1145 1146Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1147with the 'nent' field indicating the number of entries in the variable-size 1148array 'entries'. If the number of entries is too low to describe the cpu 1149capabilities, an error (E2BIG) is returned. If the number is too high, 1150the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1151number is just right, the 'nent' field is adjusted to the number of valid 1152entries in the 'entries' array, which is then filled. 1153 1154The entries returned are the host cpuid as returned by the cpuid instruction, 1155with unknown or unsupported features masked out. Some features (for example, 1156x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1157emulate them efficiently. The fields in each entry are defined as follows: 1158 1159 function: the eax value used to obtain the entry 1160 index: the ecx value used to obtain the entry (for entries that are 1161 affected by ecx) 1162 flags: an OR of zero or more of the following: 1163 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1164 if the index field is valid 1165 KVM_CPUID_FLAG_STATEFUL_FUNC: 1166 if cpuid for this function returns different values for successive 1167 invocations; there will be several entries with the same function, 1168 all with this flag set 1169 KVM_CPUID_FLAG_STATE_READ_NEXT: 1170 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is 1171 the first entry to be read by a cpu 1172 eax, ebx, ecx, edx: the values returned by the cpuid instruction for 1173 this function/index combination 1174 1175The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned 1176as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC 1177support. Instead it is reported via 1178 1179 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1180 1181if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1182feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1183 1184 11854.47 KVM_PPC_GET_PVINFO 1186 1187Capability: KVM_CAP_PPC_GET_PVINFO 1188Architectures: ppc 1189Type: vm ioctl 1190Parameters: struct kvm_ppc_pvinfo (out) 1191Returns: 0 on success, !0 on error 1192 1193struct kvm_ppc_pvinfo { 1194 __u32 flags; 1195 __u32 hcall[4]; 1196 __u8 pad[108]; 1197}; 1198 1199This ioctl fetches PV specific information that need to be passed to the guest 1200using the device tree or other means from vm context. 1201 1202The hcall array defines 4 instructions that make up a hypercall. 1203 1204If any additional field gets added to this structure later on, a bit for that 1205additional piece of information will be set in the flags bitmap. 1206 1207The flags bitmap is defined as: 1208 1209 /* the host supports the ePAPR idle hcall 1210 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1211 12124.48 KVM_ASSIGN_PCI_DEVICE 1213 1214Capability: KVM_CAP_DEVICE_ASSIGNMENT 1215Architectures: x86 ia64 1216Type: vm ioctl 1217Parameters: struct kvm_assigned_pci_dev (in) 1218Returns: 0 on success, -1 on error 1219 1220Assigns a host PCI device to the VM. 1221 1222struct kvm_assigned_pci_dev { 1223 __u32 assigned_dev_id; 1224 __u32 busnr; 1225 __u32 devfn; 1226 __u32 flags; 1227 __u32 segnr; 1228 union { 1229 __u32 reserved[11]; 1230 }; 1231}; 1232 1233The PCI device is specified by the triple segnr, busnr, and devfn. 1234Identification in succeeding service requests is done via assigned_dev_id. The 1235following flags are specified: 1236 1237/* Depends on KVM_CAP_IOMMU */ 1238#define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0) 1239/* The following two depend on KVM_CAP_PCI_2_3 */ 1240#define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1) 1241#define KVM_DEV_ASSIGN_MASK_INTX (1 << 2) 1242 1243If KVM_DEV_ASSIGN_PCI_2_3 is set, the kernel will manage legacy INTx interrupts 1244via the PCI-2.3-compliant device-level mask, thus enable IRQ sharing with other 1245assigned devices or host devices. KVM_DEV_ASSIGN_MASK_INTX specifies the 1246guest's view on the INTx mask, see KVM_ASSIGN_SET_INTX_MASK for details. 1247 1248The KVM_DEV_ASSIGN_ENABLE_IOMMU flag is a mandatory option to ensure 1249isolation of the device. Usages not specifying this flag are deprecated. 1250 1251Only PCI header type 0 devices with PCI BAR resources are supported by 1252device assignment. The user requesting this ioctl must have read/write 1253access to the PCI sysfs resource files associated with the device. 1254 1255 12564.49 KVM_DEASSIGN_PCI_DEVICE 1257 1258Capability: KVM_CAP_DEVICE_DEASSIGNMENT 1259Architectures: x86 ia64 1260Type: vm ioctl 1261Parameters: struct kvm_assigned_pci_dev (in) 1262Returns: 0 on success, -1 on error 1263 1264Ends PCI device assignment, releasing all associated resources. 1265 1266See KVM_CAP_DEVICE_ASSIGNMENT for the data structure. Only assigned_dev_id is 1267used in kvm_assigned_pci_dev to identify the device. 1268 1269 12704.50 KVM_ASSIGN_DEV_IRQ 1271 1272Capability: KVM_CAP_ASSIGN_DEV_IRQ 1273Architectures: x86 ia64 1274Type: vm ioctl 1275Parameters: struct kvm_assigned_irq (in) 1276Returns: 0 on success, -1 on error 1277 1278Assigns an IRQ to a passed-through device. 1279 1280struct kvm_assigned_irq { 1281 __u32 assigned_dev_id; 1282 __u32 host_irq; /* ignored (legacy field) */ 1283 __u32 guest_irq; 1284 __u32 flags; 1285 union { 1286 __u32 reserved[12]; 1287 }; 1288}; 1289 1290The following flags are defined: 1291 1292#define KVM_DEV_IRQ_HOST_INTX (1 << 0) 1293#define KVM_DEV_IRQ_HOST_MSI (1 << 1) 1294#define KVM_DEV_IRQ_HOST_MSIX (1 << 2) 1295 1296#define KVM_DEV_IRQ_GUEST_INTX (1 << 8) 1297#define KVM_DEV_IRQ_GUEST_MSI (1 << 9) 1298#define KVM_DEV_IRQ_GUEST_MSIX (1 << 10) 1299 1300It is not valid to specify multiple types per host or guest IRQ. However, the 1301IRQ type of host and guest can differ or can even be null. 1302 1303 13044.51 KVM_DEASSIGN_DEV_IRQ 1305 1306Capability: KVM_CAP_ASSIGN_DEV_IRQ 1307Architectures: x86 ia64 1308Type: vm ioctl 1309Parameters: struct kvm_assigned_irq (in) 1310Returns: 0 on success, -1 on error 1311 1312Ends an IRQ assignment to a passed-through device. 1313 1314See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified 1315by assigned_dev_id, flags must correspond to the IRQ type specified on 1316KVM_ASSIGN_DEV_IRQ. Partial deassignment of host or guest IRQ is allowed. 1317 1318 13194.52 KVM_SET_GSI_ROUTING 1320 1321Capability: KVM_CAP_IRQ_ROUTING 1322Architectures: x86 ia64 1323Type: vm ioctl 1324Parameters: struct kvm_irq_routing (in) 1325Returns: 0 on success, -1 on error 1326 1327Sets the GSI routing table entries, overwriting any previously set entries. 1328 1329struct kvm_irq_routing { 1330 __u32 nr; 1331 __u32 flags; 1332 struct kvm_irq_routing_entry entries[0]; 1333}; 1334 1335No flags are specified so far, the corresponding field must be set to zero. 1336 1337struct kvm_irq_routing_entry { 1338 __u32 gsi; 1339 __u32 type; 1340 __u32 flags; 1341 __u32 pad; 1342 union { 1343 struct kvm_irq_routing_irqchip irqchip; 1344 struct kvm_irq_routing_msi msi; 1345 __u32 pad[8]; 1346 } u; 1347}; 1348 1349/* gsi routing entry types */ 1350#define KVM_IRQ_ROUTING_IRQCHIP 1 1351#define KVM_IRQ_ROUTING_MSI 2 1352 1353No flags are specified so far, the corresponding field must be set to zero. 1354 1355struct kvm_irq_routing_irqchip { 1356 __u32 irqchip; 1357 __u32 pin; 1358}; 1359 1360struct kvm_irq_routing_msi { 1361 __u32 address_lo; 1362 __u32 address_hi; 1363 __u32 data; 1364 __u32 pad; 1365}; 1366 1367 13684.53 KVM_ASSIGN_SET_MSIX_NR 1369 1370Capability: KVM_CAP_DEVICE_MSIX 1371Architectures: x86 ia64 1372Type: vm ioctl 1373Parameters: struct kvm_assigned_msix_nr (in) 1374Returns: 0 on success, -1 on error 1375 1376Set the number of MSI-X interrupts for an assigned device. The number is 1377reset again by terminating the MSI-X assignment of the device via 1378KVM_DEASSIGN_DEV_IRQ. Calling this service more than once at any earlier 1379point will fail. 1380 1381struct kvm_assigned_msix_nr { 1382 __u32 assigned_dev_id; 1383 __u16 entry_nr; 1384 __u16 padding; 1385}; 1386 1387#define KVM_MAX_MSIX_PER_DEV 256 1388 1389 13904.54 KVM_ASSIGN_SET_MSIX_ENTRY 1391 1392Capability: KVM_CAP_DEVICE_MSIX 1393Architectures: x86 ia64 1394Type: vm ioctl 1395Parameters: struct kvm_assigned_msix_entry (in) 1396Returns: 0 on success, -1 on error 1397 1398Specifies the routing of an MSI-X assigned device interrupt to a GSI. Setting 1399the GSI vector to zero means disabling the interrupt. 1400 1401struct kvm_assigned_msix_entry { 1402 __u32 assigned_dev_id; 1403 __u32 gsi; 1404 __u16 entry; /* The index of entry in the MSI-X table */ 1405 __u16 padding[3]; 1406}; 1407 1408 14094.55 KVM_SET_TSC_KHZ 1410 1411Capability: KVM_CAP_TSC_CONTROL 1412Architectures: x86 1413Type: vcpu ioctl 1414Parameters: virtual tsc_khz 1415Returns: 0 on success, -1 on error 1416 1417Specifies the tsc frequency for the virtual machine. The unit of the 1418frequency is KHz. 1419 1420 14214.56 KVM_GET_TSC_KHZ 1422 1423Capability: KVM_CAP_GET_TSC_KHZ 1424Architectures: x86 1425Type: vcpu ioctl 1426Parameters: none 1427Returns: virtual tsc-khz on success, negative value on error 1428 1429Returns the tsc frequency of the guest. The unit of the return value is 1430KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 1431error. 1432 1433 14344.57 KVM_GET_LAPIC 1435 1436Capability: KVM_CAP_IRQCHIP 1437Architectures: x86 1438Type: vcpu ioctl 1439Parameters: struct kvm_lapic_state (out) 1440Returns: 0 on success, -1 on error 1441 1442#define KVM_APIC_REG_SIZE 0x400 1443struct kvm_lapic_state { 1444 char regs[KVM_APIC_REG_SIZE]; 1445}; 1446 1447Reads the Local APIC registers and copies them into the input argument. The 1448data format and layout are the same as documented in the architecture manual. 1449 1450 14514.58 KVM_SET_LAPIC 1452 1453Capability: KVM_CAP_IRQCHIP 1454Architectures: x86 1455Type: vcpu ioctl 1456Parameters: struct kvm_lapic_state (in) 1457Returns: 0 on success, -1 on error 1458 1459#define KVM_APIC_REG_SIZE 0x400 1460struct kvm_lapic_state { 1461 char regs[KVM_APIC_REG_SIZE]; 1462}; 1463 1464Copies the input argument into the the Local APIC registers. The data format 1465and layout are the same as documented in the architecture manual. 1466 1467 14684.59 KVM_IOEVENTFD 1469 1470Capability: KVM_CAP_IOEVENTFD 1471Architectures: all 1472Type: vm ioctl 1473Parameters: struct kvm_ioeventfd (in) 1474Returns: 0 on success, !0 on error 1475 1476This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 1477within the guest. A guest write in the registered address will signal the 1478provided event instead of triggering an exit. 1479 1480struct kvm_ioeventfd { 1481 __u64 datamatch; 1482 __u64 addr; /* legal pio/mmio address */ 1483 __u32 len; /* 1, 2, 4, or 8 bytes */ 1484 __s32 fd; 1485 __u32 flags; 1486 __u8 pad[36]; 1487}; 1488 1489For the special case of virtio-ccw devices on s390, the ioevent is matched 1490to a subchannel/virtqueue tuple instead. 1491 1492The following flags are defined: 1493 1494#define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 1495#define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 1496#define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 1497#define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 1498 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 1499 1500If datamatch flag is set, the event will be signaled only if the written value 1501to the registered address is equal to datamatch in struct kvm_ioeventfd. 1502 1503For virtio-ccw devices, addr contains the subchannel id and datamatch the 1504virtqueue index. 1505 1506 15074.60 KVM_DIRTY_TLB 1508 1509Capability: KVM_CAP_SW_TLB 1510Architectures: ppc 1511Type: vcpu ioctl 1512Parameters: struct kvm_dirty_tlb (in) 1513Returns: 0 on success, -1 on error 1514 1515struct kvm_dirty_tlb { 1516 __u64 bitmap; 1517 __u32 num_dirty; 1518}; 1519 1520This must be called whenever userspace has changed an entry in the shared 1521TLB, prior to calling KVM_RUN on the associated vcpu. 1522 1523The "bitmap" field is the userspace address of an array. This array 1524consists of a number of bits, equal to the total number of TLB entries as 1525determined by the last successful call to KVM_CONFIG_TLB, rounded up to the 1526nearest multiple of 64. 1527 1528Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 1529array. 1530 1531The array is little-endian: the bit 0 is the least significant bit of the 1532first byte, bit 8 is the least significant bit of the second byte, etc. 1533This avoids any complications with differing word sizes. 1534 1535The "num_dirty" field is a performance hint for KVM to determine whether it 1536should skip processing the bitmap and just invalidate everything. It must 1537be set to the number of set bits in the bitmap. 1538 1539 15404.61 KVM_ASSIGN_SET_INTX_MASK 1541 1542Capability: KVM_CAP_PCI_2_3 1543Architectures: x86 1544Type: vm ioctl 1545Parameters: struct kvm_assigned_pci_dev (in) 1546Returns: 0 on success, -1 on error 1547 1548Allows userspace to mask PCI INTx interrupts from the assigned device. The 1549kernel will not deliver INTx interrupts to the guest between setting and 1550clearing of KVM_ASSIGN_SET_INTX_MASK via this interface. This enables use of 1551and emulation of PCI 2.3 INTx disable command register behavior. 1552 1553This may be used for both PCI 2.3 devices supporting INTx disable natively and 1554older devices lacking this support. Userspace is responsible for emulating the 1555read value of the INTx disable bit in the guest visible PCI command register. 1556When modifying the INTx disable state, userspace should precede updating the 1557physical device command register by calling this ioctl to inform the kernel of 1558the new intended INTx mask state. 1559 1560Note that the kernel uses the device INTx disable bit to internally manage the 1561device interrupt state for PCI 2.3 devices. Reads of this register may 1562therefore not match the expected value. Writes should always use the guest 1563intended INTx disable value rather than attempting to read-copy-update the 1564current physical device state. Races between user and kernel updates to the 1565INTx disable bit are handled lazily in the kernel. It's possible the device 1566may generate unintended interrupts, but they will not be injected into the 1567guest. 1568 1569See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified 1570by assigned_dev_id. In the flags field, only KVM_DEV_ASSIGN_MASK_INTX is 1571evaluated. 1572 1573 15744.62 KVM_CREATE_SPAPR_TCE 1575 1576Capability: KVM_CAP_SPAPR_TCE 1577Architectures: powerpc 1578Type: vm ioctl 1579Parameters: struct kvm_create_spapr_tce (in) 1580Returns: file descriptor for manipulating the created TCE table 1581 1582This creates a virtual TCE (translation control entry) table, which 1583is an IOMMU for PAPR-style virtual I/O. It is used to translate 1584logical addresses used in virtual I/O into guest physical addresses, 1585and provides a scatter/gather capability for PAPR virtual I/O. 1586 1587/* for KVM_CAP_SPAPR_TCE */ 1588struct kvm_create_spapr_tce { 1589 __u64 liobn; 1590 __u32 window_size; 1591}; 1592 1593The liobn field gives the logical IO bus number for which to create a 1594TCE table. The window_size field specifies the size of the DMA window 1595which this TCE table will translate - the table will contain one 64 1596bit TCE entry for every 4kiB of the DMA window. 1597 1598When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 1599table has been created using this ioctl(), the kernel will handle it 1600in real mode, updating the TCE table. H_PUT_TCE calls for other 1601liobns will cause a vm exit and must be handled by userspace. 1602 1603The return value is a file descriptor which can be passed to mmap(2) 1604to map the created TCE table into userspace. This lets userspace read 1605the entries written by kernel-handled H_PUT_TCE calls, and also lets 1606userspace update the TCE table directly which is useful in some 1607circumstances. 1608 1609 16104.63 KVM_ALLOCATE_RMA 1611 1612Capability: KVM_CAP_PPC_RMA 1613Architectures: powerpc 1614Type: vm ioctl 1615Parameters: struct kvm_allocate_rma (out) 1616Returns: file descriptor for mapping the allocated RMA 1617 1618This allocates a Real Mode Area (RMA) from the pool allocated at boot 1619time by the kernel. An RMA is a physically-contiguous, aligned region 1620of memory used on older POWER processors to provide the memory which 1621will be accessed by real-mode (MMU off) accesses in a KVM guest. 1622POWER processors support a set of sizes for the RMA that usually 1623includes 64MB, 128MB, 256MB and some larger powers of two. 1624 1625/* for KVM_ALLOCATE_RMA */ 1626struct kvm_allocate_rma { 1627 __u64 rma_size; 1628}; 1629 1630The return value is a file descriptor which can be passed to mmap(2) 1631to map the allocated RMA into userspace. The mapped area can then be 1632passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the 1633RMA for a virtual machine. The size of the RMA in bytes (which is 1634fixed at host kernel boot time) is returned in the rma_size field of 1635the argument structure. 1636 1637The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl 1638is supported; 2 if the processor requires all virtual machines to have 1639an RMA, or 1 if the processor can use an RMA but doesn't require it, 1640because it supports the Virtual RMA (VRMA) facility. 1641 1642 16434.64 KVM_NMI 1644 1645Capability: KVM_CAP_USER_NMI 1646Architectures: x86 1647Type: vcpu ioctl 1648Parameters: none 1649Returns: 0 on success, -1 on error 1650 1651Queues an NMI on the thread's vcpu. Note this is well defined only 1652when KVM_CREATE_IRQCHIP has not been called, since this is an interface 1653between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 1654has been called, this interface is completely emulated within the kernel. 1655 1656To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 1657following algorithm: 1658 1659 - pause the vpcu 1660 - read the local APIC's state (KVM_GET_LAPIC) 1661 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 1662 - if so, issue KVM_NMI 1663 - resume the vcpu 1664 1665Some guests configure the LINT1 NMI input to cause a panic, aiding in 1666debugging. 1667 1668 16694.65 KVM_S390_UCAS_MAP 1670 1671Capability: KVM_CAP_S390_UCONTROL 1672Architectures: s390 1673Type: vcpu ioctl 1674Parameters: struct kvm_s390_ucas_mapping (in) 1675Returns: 0 in case of success 1676 1677The parameter is defined like this: 1678 struct kvm_s390_ucas_mapping { 1679 __u64 user_addr; 1680 __u64 vcpu_addr; 1681 __u64 length; 1682 }; 1683 1684This ioctl maps the memory at "user_addr" with the length "length" to 1685the vcpu's address space starting at "vcpu_addr". All parameters need to 1686be alligned by 1 megabyte. 1687 1688 16894.66 KVM_S390_UCAS_UNMAP 1690 1691Capability: KVM_CAP_S390_UCONTROL 1692Architectures: s390 1693Type: vcpu ioctl 1694Parameters: struct kvm_s390_ucas_mapping (in) 1695Returns: 0 in case of success 1696 1697The parameter is defined like this: 1698 struct kvm_s390_ucas_mapping { 1699 __u64 user_addr; 1700 __u64 vcpu_addr; 1701 __u64 length; 1702 }; 1703 1704This ioctl unmaps the memory in the vcpu's address space starting at 1705"vcpu_addr" with the length "length". The field "user_addr" is ignored. 1706All parameters need to be alligned by 1 megabyte. 1707 1708 17094.67 KVM_S390_VCPU_FAULT 1710 1711Capability: KVM_CAP_S390_UCONTROL 1712Architectures: s390 1713Type: vcpu ioctl 1714Parameters: vcpu absolute address (in) 1715Returns: 0 in case of success 1716 1717This call creates a page table entry on the virtual cpu's address space 1718(for user controlled virtual machines) or the virtual machine's address 1719space (for regular virtual machines). This only works for minor faults, 1720thus it's recommended to access subject memory page via the user page 1721table upfront. This is useful to handle validity intercepts for user 1722controlled virtual machines to fault in the virtual cpu's lowcore pages 1723prior to calling the KVM_RUN ioctl. 1724 1725 17264.68 KVM_SET_ONE_REG 1727 1728Capability: KVM_CAP_ONE_REG 1729Architectures: all 1730Type: vcpu ioctl 1731Parameters: struct kvm_one_reg (in) 1732Returns: 0 on success, negative value on failure 1733 1734struct kvm_one_reg { 1735 __u64 id; 1736 __u64 addr; 1737}; 1738 1739Using this ioctl, a single vcpu register can be set to a specific value 1740defined by user space with the passed in struct kvm_one_reg, where id 1741refers to the register identifier as described below and addr is a pointer 1742to a variable with the respective size. There can be architecture agnostic 1743and architecture specific registers. Each have their own range of operation 1744and their own constants and width. To keep track of the implemented 1745registers, find a list below: 1746 1747 Arch | Register | Width (bits) 1748 | | 1749 PPC | KVM_REG_PPC_HIOR | 64 1750 PPC | KVM_REG_PPC_IAC1 | 64 1751 PPC | KVM_REG_PPC_IAC2 | 64 1752 PPC | KVM_REG_PPC_IAC3 | 64 1753 PPC | KVM_REG_PPC_IAC4 | 64 1754 PPC | KVM_REG_PPC_DAC1 | 64 1755 PPC | KVM_REG_PPC_DAC2 | 64 1756 PPC | KVM_REG_PPC_DABR | 64 1757 PPC | KVM_REG_PPC_DSCR | 64 1758 PPC | KVM_REG_PPC_PURR | 64 1759 PPC | KVM_REG_PPC_SPURR | 64 1760 PPC | KVM_REG_PPC_DAR | 64 1761 PPC | KVM_REG_PPC_DSISR | 32 1762 PPC | KVM_REG_PPC_AMR | 64 1763 PPC | KVM_REG_PPC_UAMOR | 64 1764 PPC | KVM_REG_PPC_MMCR0 | 64 1765 PPC | KVM_REG_PPC_MMCR1 | 64 1766 PPC | KVM_REG_PPC_MMCRA | 64 1767 PPC | KVM_REG_PPC_PMC1 | 32 1768 PPC | KVM_REG_PPC_PMC2 | 32 1769 PPC | KVM_REG_PPC_PMC3 | 32 1770 PPC | KVM_REG_PPC_PMC4 | 32 1771 PPC | KVM_REG_PPC_PMC5 | 32 1772 PPC | KVM_REG_PPC_PMC6 | 32 1773 PPC | KVM_REG_PPC_PMC7 | 32 1774 PPC | KVM_REG_PPC_PMC8 | 32 1775 PPC | KVM_REG_PPC_FPR0 | 64 1776 ... 1777 PPC | KVM_REG_PPC_FPR31 | 64 1778 PPC | KVM_REG_PPC_VR0 | 128 1779 ... 1780 PPC | KVM_REG_PPC_VR31 | 128 1781 PPC | KVM_REG_PPC_VSR0 | 128 1782 ... 1783 PPC | KVM_REG_PPC_VSR31 | 128 1784 PPC | KVM_REG_PPC_FPSCR | 64 1785 PPC | KVM_REG_PPC_VSCR | 32 1786 PPC | KVM_REG_PPC_VPA_ADDR | 64 1787 PPC | KVM_REG_PPC_VPA_SLB | 128 1788 PPC | KVM_REG_PPC_VPA_DTL | 128 1789 PPC | KVM_REG_PPC_EPCR | 32 1790 PPC | KVM_REG_PPC_EPR | 32 1791 PPC | KVM_REG_PPC_TCR | 32 1792 PPC | KVM_REG_PPC_TSR | 32 1793 PPC | KVM_REG_PPC_OR_TSR | 32 1794 PPC | KVM_REG_PPC_CLEAR_TSR | 32 1795 PPC | KVM_REG_PPC_MAS0 | 32 1796 PPC | KVM_REG_PPC_MAS1 | 32 1797 PPC | KVM_REG_PPC_MAS2 | 64 1798 PPC | KVM_REG_PPC_MAS7_3 | 64 1799 PPC | KVM_REG_PPC_MAS4 | 32 1800 PPC | KVM_REG_PPC_MAS6 | 32 1801 PPC | KVM_REG_PPC_MMUCFG | 32 1802 PPC | KVM_REG_PPC_TLB0CFG | 32 1803 PPC | KVM_REG_PPC_TLB1CFG | 32 1804 PPC | KVM_REG_PPC_TLB2CFG | 32 1805 PPC | KVM_REG_PPC_TLB3CFG | 32 1806 PPC | KVM_REG_PPC_TLB0PS | 32 1807 PPC | KVM_REG_PPC_TLB1PS | 32 1808 PPC | KVM_REG_PPC_TLB2PS | 32 1809 PPC | KVM_REG_PPC_TLB3PS | 32 1810 PPC | KVM_REG_PPC_EPTCFG | 32 1811 PPC | KVM_REG_PPC_ICP_STATE | 64 1812 1813ARM registers are mapped using the lower 32 bits. The upper 16 of that 1814is the register group type, or coprocessor number: 1815 1816ARM core registers have the following id bit patterns: 1817 0x4020 0000 0010 <index into the kvm_regs struct:16> 1818 1819ARM 32-bit CP15 registers have the following id bit patterns: 1820 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 1821 1822ARM 64-bit CP15 registers have the following id bit patterns: 1823 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 1824 1825ARM CCSIDR registers are demultiplexed by CSSELR value: 1826 0x4020 0000 0011 00 <csselr:8> 1827 1828ARM 32-bit VFP control registers have the following id bit patterns: 1829 0x4020 0000 0012 1 <regno:12> 1830 1831ARM 64-bit FP registers have the following id bit patterns: 1832 0x4030 0000 0012 0 <regno:12> 1833 18344.69 KVM_GET_ONE_REG 1835 1836Capability: KVM_CAP_ONE_REG 1837Architectures: all 1838Type: vcpu ioctl 1839Parameters: struct kvm_one_reg (in and out) 1840Returns: 0 on success, negative value on failure 1841 1842This ioctl allows to receive the value of a single register implemented 1843in a vcpu. The register to read is indicated by the "id" field of the 1844kvm_one_reg struct passed in. On success, the register value can be found 1845at the memory location pointed to by "addr". 1846 1847The list of registers accessible using this interface is identical to the 1848list in 4.68. 1849 1850 18514.70 KVM_KVMCLOCK_CTRL 1852 1853Capability: KVM_CAP_KVMCLOCK_CTRL 1854Architectures: Any that implement pvclocks (currently x86 only) 1855Type: vcpu ioctl 1856Parameters: None 1857Returns: 0 on success, -1 on error 1858 1859This signals to the host kernel that the specified guest is being paused by 1860userspace. The host will set a flag in the pvclock structure that is checked 1861from the soft lockup watchdog. The flag is part of the pvclock structure that 1862is shared between guest and host, specifically the second bit of the flags 1863field of the pvclock_vcpu_time_info structure. It will be set exclusively by 1864the host and read/cleared exclusively by the guest. The guest operation of 1865checking and clearing the flag must an atomic operation so 1866load-link/store-conditional, or equivalent must be used. There are two cases 1867where the guest will clear the flag: when the soft lockup watchdog timer resets 1868itself or when a soft lockup is detected. This ioctl can be called any time 1869after pausing the vcpu, but before it is resumed. 1870 1871 18724.71 KVM_SIGNAL_MSI 1873 1874Capability: KVM_CAP_SIGNAL_MSI 1875Architectures: x86 1876Type: vm ioctl 1877Parameters: struct kvm_msi (in) 1878Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 1879 1880Directly inject a MSI message. Only valid with in-kernel irqchip that handles 1881MSI messages. 1882 1883struct kvm_msi { 1884 __u32 address_lo; 1885 __u32 address_hi; 1886 __u32 data; 1887 __u32 flags; 1888 __u8 pad[16]; 1889}; 1890 1891No flags are defined so far. The corresponding field must be 0. 1892 1893 18944.71 KVM_CREATE_PIT2 1895 1896Capability: KVM_CAP_PIT2 1897Architectures: x86 1898Type: vm ioctl 1899Parameters: struct kvm_pit_config (in) 1900Returns: 0 on success, -1 on error 1901 1902Creates an in-kernel device model for the i8254 PIT. This call is only valid 1903after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 1904parameters have to be passed: 1905 1906struct kvm_pit_config { 1907 __u32 flags; 1908 __u32 pad[15]; 1909}; 1910 1911Valid flags are: 1912 1913#define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 1914 1915PIT timer interrupts may use a per-VM kernel thread for injection. If it 1916exists, this thread will have a name of the following pattern: 1917 1918kvm-pit/<owner-process-pid> 1919 1920When running a guest with elevated priorities, the scheduling parameters of 1921this thread may have to be adjusted accordingly. 1922 1923This IOCTL replaces the obsolete KVM_CREATE_PIT. 1924 1925 19264.72 KVM_GET_PIT2 1927 1928Capability: KVM_CAP_PIT_STATE2 1929Architectures: x86 1930Type: vm ioctl 1931Parameters: struct kvm_pit_state2 (out) 1932Returns: 0 on success, -1 on error 1933 1934Retrieves the state of the in-kernel PIT model. Only valid after 1935KVM_CREATE_PIT2. The state is returned in the following structure: 1936 1937struct kvm_pit_state2 { 1938 struct kvm_pit_channel_state channels[3]; 1939 __u32 flags; 1940 __u32 reserved[9]; 1941}; 1942 1943Valid flags are: 1944 1945/* disable PIT in HPET legacy mode */ 1946#define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 1947 1948This IOCTL replaces the obsolete KVM_GET_PIT. 1949 1950 19514.73 KVM_SET_PIT2 1952 1953Capability: KVM_CAP_PIT_STATE2 1954Architectures: x86 1955Type: vm ioctl 1956Parameters: struct kvm_pit_state2 (in) 1957Returns: 0 on success, -1 on error 1958 1959Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 1960See KVM_GET_PIT2 for details on struct kvm_pit_state2. 1961 1962This IOCTL replaces the obsolete KVM_SET_PIT. 1963 1964 19654.74 KVM_PPC_GET_SMMU_INFO 1966 1967Capability: KVM_CAP_PPC_GET_SMMU_INFO 1968Architectures: powerpc 1969Type: vm ioctl 1970Parameters: None 1971Returns: 0 on success, -1 on error 1972 1973This populates and returns a structure describing the features of 1974the "Server" class MMU emulation supported by KVM. 1975This can in turn be used by userspace to generate the appropariate 1976device-tree properties for the guest operating system. 1977 1978The structure contains some global informations, followed by an 1979array of supported segment page sizes: 1980 1981 struct kvm_ppc_smmu_info { 1982 __u64 flags; 1983 __u32 slb_size; 1984 __u32 pad; 1985 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 1986 }; 1987 1988The supported flags are: 1989 1990 - KVM_PPC_PAGE_SIZES_REAL: 1991 When that flag is set, guest page sizes must "fit" the backing 1992 store page sizes. When not set, any page size in the list can 1993 be used regardless of how they are backed by userspace. 1994 1995 - KVM_PPC_1T_SEGMENTS 1996 The emulated MMU supports 1T segments in addition to the 1997 standard 256M ones. 1998 1999The "slb_size" field indicates how many SLB entries are supported 2000 2001The "sps" array contains 8 entries indicating the supported base 2002page sizes for a segment in increasing order. Each entry is defined 2003as follow: 2004 2005 struct kvm_ppc_one_seg_page_size { 2006 __u32 page_shift; /* Base page shift of segment (or 0) */ 2007 __u32 slb_enc; /* SLB encoding for BookS */ 2008 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 2009 }; 2010 2011An entry with a "page_shift" of 0 is unused. Because the array is 2012organized in increasing order, a lookup can stop when encoutering 2013such an entry. 2014 2015The "slb_enc" field provides the encoding to use in the SLB for the 2016page size. The bits are in positions such as the value can directly 2017be OR'ed into the "vsid" argument of the slbmte instruction. 2018 2019The "enc" array is a list which for each of those segment base page 2020size provides the list of supported actual page sizes (which can be 2021only larger or equal to the base page size), along with the 2022corresponding encoding in the hash PTE. Similarily, the array is 20238 entries sorted by increasing sizes and an entry with a "0" shift 2024is an empty entry and a terminator: 2025 2026 struct kvm_ppc_one_page_size { 2027 __u32 page_shift; /* Page shift (or 0) */ 2028 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 2029 }; 2030 2031The "pte_enc" field provides a value that can OR'ed into the hash 2032PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 2033into the hash PTE second double word). 2034 20354.75 KVM_IRQFD 2036 2037Capability: KVM_CAP_IRQFD 2038Architectures: x86 2039Type: vm ioctl 2040Parameters: struct kvm_irqfd (in) 2041Returns: 0 on success, -1 on error 2042 2043Allows setting an eventfd to directly trigger a guest interrupt. 2044kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 2045kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 2046an event is tiggered on the eventfd, an interrupt is injected into 2047the guest using the specified gsi pin. The irqfd is removed using 2048the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 2049and kvm_irqfd.gsi. 2050 2051With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 2052mechanism allowing emulation of level-triggered, irqfd-based 2053interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 2054additional eventfd in the kvm_irqfd.resamplefd field. When operating 2055in resample mode, posting of an interrupt through kvm_irq.fd asserts 2056the specified gsi in the irqchip. When the irqchip is resampled, such 2057as from an EOI, the gsi is de-asserted and the user is notifed via 2058kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 2059the interrupt if the device making use of it still requires service. 2060Note that closing the resamplefd is not sufficient to disable the 2061irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 2062and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 2063 20644.76 KVM_PPC_ALLOCATE_HTAB 2065 2066Capability: KVM_CAP_PPC_ALLOC_HTAB 2067Architectures: powerpc 2068Type: vm ioctl 2069Parameters: Pointer to u32 containing hash table order (in/out) 2070Returns: 0 on success, -1 on error 2071 2072This requests the host kernel to allocate an MMU hash table for a 2073guest using the PAPR paravirtualization interface. This only does 2074anything if the kernel is configured to use the Book 3S HV style of 2075virtualization. Otherwise the capability doesn't exist and the ioctl 2076returns an ENOTTY error. The rest of this description assumes Book 3S 2077HV. 2078 2079There must be no vcpus running when this ioctl is called; if there 2080are, it will do nothing and return an EBUSY error. 2081 2082The parameter is a pointer to a 32-bit unsigned integer variable 2083containing the order (log base 2) of the desired size of the hash 2084table, which must be between 18 and 46. On successful return from the 2085ioctl, it will have been updated with the order of the hash table that 2086was allocated. 2087 2088If no hash table has been allocated when any vcpu is asked to run 2089(with the KVM_RUN ioctl), the host kernel will allocate a 2090default-sized hash table (16 MB). 2091 2092If this ioctl is called when a hash table has already been allocated, 2093the kernel will clear out the existing hash table (zero all HPTEs) and 2094return the hash table order in the parameter. (If the guest is using 2095the virtualized real-mode area (VRMA) facility, the kernel will 2096re-create the VMRA HPTEs on the next KVM_RUN of any vcpu.) 2097 20984.77 KVM_S390_INTERRUPT 2099 2100Capability: basic 2101Architectures: s390 2102Type: vm ioctl, vcpu ioctl 2103Parameters: struct kvm_s390_interrupt (in) 2104Returns: 0 on success, -1 on error 2105 2106Allows to inject an interrupt to the guest. Interrupts can be floating 2107(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 2108 2109Interrupt parameters are passed via kvm_s390_interrupt: 2110 2111struct kvm_s390_interrupt { 2112 __u32 type; 2113 __u32 parm; 2114 __u64 parm64; 2115}; 2116 2117type can be one of the following: 2118 2119KVM_S390_SIGP_STOP (vcpu) - sigp restart 2120KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm 2121KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm 2122KVM_S390_RESTART (vcpu) - restart 2123KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt 2124 parameters in parm and parm64 2125KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm 2126KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm 2127KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm 2128KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an 2129 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 2130 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 2131 interruption subclass) 2132KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm, 2133 machine check interrupt code in parm64 (note that 2134 machine checks needing further payload are not 2135 supported by this ioctl) 2136 2137Note that the vcpu ioctl is asynchronous to vcpu execution. 2138 21394.78 KVM_PPC_GET_HTAB_FD 2140 2141Capability: KVM_CAP_PPC_HTAB_FD 2142Architectures: powerpc 2143Type: vm ioctl 2144Parameters: Pointer to struct kvm_get_htab_fd (in) 2145Returns: file descriptor number (>= 0) on success, -1 on error 2146 2147This returns a file descriptor that can be used either to read out the 2148entries in the guest's hashed page table (HPT), or to write entries to 2149initialize the HPT. The returned fd can only be written to if the 2150KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 2151can only be read if that bit is clear. The argument struct looks like 2152this: 2153 2154/* For KVM_PPC_GET_HTAB_FD */ 2155struct kvm_get_htab_fd { 2156 __u64 flags; 2157 __u64 start_index; 2158 __u64 reserved[2]; 2159}; 2160 2161/* Values for kvm_get_htab_fd.flags */ 2162#define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 2163#define KVM_GET_HTAB_WRITE ((__u64)0x2) 2164 2165The `start_index' field gives the index in the HPT of the entry at 2166which to start reading. It is ignored when writing. 2167 2168Reads on the fd will initially supply information about all 2169"interesting" HPT entries. Interesting entries are those with the 2170bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 2171all entries. When the end of the HPT is reached, the read() will 2172return. If read() is called again on the fd, it will start again from 2173the beginning of the HPT, but will only return HPT entries that have 2174changed since they were last read. 2175 2176Data read or written is structured as a header (8 bytes) followed by a 2177series of valid HPT entries (16 bytes) each. The header indicates how 2178many valid HPT entries there are and how many invalid entries follow 2179the valid entries. The invalid entries are not represented explicitly 2180in the stream. The header format is: 2181 2182struct kvm_get_htab_header { 2183 __u32 index; 2184 __u16 n_valid; 2185 __u16 n_invalid; 2186}; 2187 2188Writes to the fd create HPT entries starting at the index given in the 2189header; first `n_valid' valid entries with contents from the data 2190written, then `n_invalid' invalid entries, invalidating any previously 2191valid entries found. 2192 21934.79 KVM_CREATE_DEVICE 2194 2195Capability: KVM_CAP_DEVICE_CTRL 2196Type: vm ioctl 2197Parameters: struct kvm_create_device (in/out) 2198Returns: 0 on success, -1 on error 2199Errors: 2200 ENODEV: The device type is unknown or unsupported 2201 EEXIST: Device already created, and this type of device may not 2202 be instantiated multiple times 2203 2204 Other error conditions may be defined by individual device types or 2205 have their standard meanings. 2206 2207Creates an emulated device in the kernel. The file descriptor returned 2208in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 2209 2210If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 2211device type is supported (not necessarily whether it can be created 2212in the current vm). 2213 2214Individual devices should not define flags. Attributes should be used 2215for specifying any behavior that is not implied by the device type 2216number. 2217 2218struct kvm_create_device { 2219 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 2220 __u32 fd; /* out: device handle */ 2221 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 2222}; 2223 22244.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 2225 2226Capability: KVM_CAP_DEVICE_CTRL 2227Type: device ioctl 2228Parameters: struct kvm_device_attr 2229Returns: 0 on success, -1 on error 2230Errors: 2231 ENXIO: The group or attribute is unknown/unsupported for this device 2232 EPERM: The attribute cannot (currently) be accessed this way 2233 (e.g. read-only attribute, or attribute that only makes 2234 sense when the device is in a different state) 2235 2236 Other error conditions may be defined by individual device types. 2237 2238Gets/sets a specified piece of device configuration and/or state. The 2239semantics are device-specific. See individual device documentation in 2240the "devices" directory. As with ONE_REG, the size of the data 2241transferred is defined by the particular attribute. 2242 2243struct kvm_device_attr { 2244 __u32 flags; /* no flags currently defined */ 2245 __u32 group; /* device-defined */ 2246 __u64 attr; /* group-defined */ 2247 __u64 addr; /* userspace address of attr data */ 2248}; 2249 22504.81 KVM_HAS_DEVICE_ATTR 2251 2252Capability: KVM_CAP_DEVICE_CTRL 2253Type: device ioctl 2254Parameters: struct kvm_device_attr 2255Returns: 0 on success, -1 on error 2256Errors: 2257 ENXIO: The group or attribute is unknown/unsupported for this device 2258 2259Tests whether a device supports a particular attribute. A successful 2260return indicates the attribute is implemented. It does not necessarily 2261indicate that the attribute can be read or written in the device's 2262current state. "addr" is ignored. 2263 22644.77 KVM_ARM_VCPU_INIT 2265 2266Capability: basic 2267Architectures: arm 2268Type: vcpu ioctl 2269Parameters: struct struct kvm_vcpu_init (in) 2270Returns: 0 on success; -1 on error 2271Errors: 2272 EINVAL: the target is unknown, or the combination of features is invalid. 2273 ENOENT: a features bit specified is unknown. 2274 2275This tells KVM what type of CPU to present to the guest, and what 2276optional features it should have. This will cause a reset of the cpu 2277registers to their initial values. If this is not called, KVM_RUN will 2278return ENOEXEC for that vcpu. 2279 2280Note that because some registers reflect machine topology, all vcpus 2281should be created before this ioctl is invoked. 2282 2283Possible features: 2284 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 2285 Depends on KVM_CAP_ARM_PSCI. 2286 2287 22884.78 KVM_GET_REG_LIST 2289 2290Capability: basic 2291Architectures: arm 2292Type: vcpu ioctl 2293Parameters: struct kvm_reg_list (in/out) 2294Returns: 0 on success; -1 on error 2295Errors: 2296 E2BIG: the reg index list is too big to fit in the array specified by 2297 the user (the number required will be written into n). 2298 2299struct kvm_reg_list { 2300 __u64 n; /* number of registers in reg[] */ 2301 __u64 reg[0]; 2302}; 2303 2304This ioctl returns the guest registers that are supported for the 2305KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 2306 2307 23084.80 KVM_ARM_SET_DEVICE_ADDR 2309 2310Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 2311Architectures: arm 2312Type: vm ioctl 2313Parameters: struct kvm_arm_device_address (in) 2314Returns: 0 on success, -1 on error 2315Errors: 2316 ENODEV: The device id is unknown 2317 ENXIO: Device not supported on current system 2318 EEXIST: Address already set 2319 E2BIG: Address outside guest physical address space 2320 EBUSY: Address overlaps with other device range 2321 2322struct kvm_arm_device_addr { 2323 __u64 id; 2324 __u64 addr; 2325}; 2326 2327Specify a device address in the guest's physical address space where guests 2328can access emulated or directly exposed devices, which the host kernel needs 2329to know about. The id field is an architecture specific identifier for a 2330specific device. 2331 2332ARM divides the id field into two parts, a device id and an address type id 2333specific to the individual device. 2334 2335 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 2336 field: | 0x00000000 | device id | addr type id | 2337 2338ARM currently only require this when using the in-kernel GIC support for the 2339hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 as the device id. When 2340setting the base address for the guest's mapping of the VGIC virtual CPU 2341and distributor interface, the ioctl must be called after calling 2342KVM_CREATE_IRQCHIP, but before calling KVM_RUN on any of the VCPUs. Calling 2343this ioctl twice for any of the base addresses will return -EEXIST. 2344 23454.82 KVM_PPC_RTAS_DEFINE_TOKEN 2346 2347Capability: KVM_CAP_PPC_RTAS 2348Architectures: ppc 2349Type: vm ioctl 2350Parameters: struct kvm_rtas_token_args 2351Returns: 0 on success, -1 on error 2352 2353Defines a token value for a RTAS (Run Time Abstraction Services) 2354service in order to allow it to be handled in the kernel. The 2355argument struct gives the name of the service, which must be the name 2356of a service that has a kernel-side implementation. If the token 2357value is non-zero, it will be associated with that service, and 2358subsequent RTAS calls by the guest specifying that token will be 2359handled by the kernel. If the token value is 0, then any token 2360associated with the service will be forgotten, and subsequent RTAS 2361calls by the guest for that service will be passed to userspace to be 2362handled. 2363 2364 23655. The kvm_run structure 2366------------------------ 2367 2368Application code obtains a pointer to the kvm_run structure by 2369mmap()ing a vcpu fd. From that point, application code can control 2370execution by changing fields in kvm_run prior to calling the KVM_RUN 2371ioctl, and obtain information about the reason KVM_RUN returned by 2372looking up structure members. 2373 2374struct kvm_run { 2375 /* in */ 2376 __u8 request_interrupt_window; 2377 2378Request that KVM_RUN return when it becomes possible to inject external 2379interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 2380 2381 __u8 padding1[7]; 2382 2383 /* out */ 2384 __u32 exit_reason; 2385 2386When KVM_RUN has returned successfully (return value 0), this informs 2387application code why KVM_RUN has returned. Allowable values for this 2388field are detailed below. 2389 2390 __u8 ready_for_interrupt_injection; 2391 2392If request_interrupt_window has been specified, this field indicates 2393an interrupt can be injected now with KVM_INTERRUPT. 2394 2395 __u8 if_flag; 2396 2397The value of the current interrupt flag. Only valid if in-kernel 2398local APIC is not used. 2399 2400 __u8 padding2[2]; 2401 2402 /* in (pre_kvm_run), out (post_kvm_run) */ 2403 __u64 cr8; 2404 2405The value of the cr8 register. Only valid if in-kernel local APIC is 2406not used. Both input and output. 2407 2408 __u64 apic_base; 2409 2410The value of the APIC BASE msr. Only valid if in-kernel local 2411APIC is not used. Both input and output. 2412 2413 union { 2414 /* KVM_EXIT_UNKNOWN */ 2415 struct { 2416 __u64 hardware_exit_reason; 2417 } hw; 2418 2419If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 2420reasons. Further architecture-specific information is available in 2421hardware_exit_reason. 2422 2423 /* KVM_EXIT_FAIL_ENTRY */ 2424 struct { 2425 __u64 hardware_entry_failure_reason; 2426 } fail_entry; 2427 2428If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 2429to unknown reasons. Further architecture-specific information is 2430available in hardware_entry_failure_reason. 2431 2432 /* KVM_EXIT_EXCEPTION */ 2433 struct { 2434 __u32 exception; 2435 __u32 error_code; 2436 } ex; 2437 2438Unused. 2439 2440 /* KVM_EXIT_IO */ 2441 struct { 2442#define KVM_EXIT_IO_IN 0 2443#define KVM_EXIT_IO_OUT 1 2444 __u8 direction; 2445 __u8 size; /* bytes */ 2446 __u16 port; 2447 __u32 count; 2448 __u64 data_offset; /* relative to kvm_run start */ 2449 } io; 2450 2451If exit_reason is KVM_EXIT_IO, then the vcpu has 2452executed a port I/O instruction which could not be satisfied by kvm. 2453data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 2454where kvm expects application code to place the data for the next 2455KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 2456 2457 struct { 2458 struct kvm_debug_exit_arch arch; 2459 } debug; 2460 2461Unused. 2462 2463 /* KVM_EXIT_MMIO */ 2464 struct { 2465 __u64 phys_addr; 2466 __u8 data[8]; 2467 __u32 len; 2468 __u8 is_write; 2469 } mmio; 2470 2471If exit_reason is KVM_EXIT_MMIO, then the vcpu has 2472executed a memory-mapped I/O instruction which could not be satisfied 2473by kvm. The 'data' member contains the written data if 'is_write' is 2474true, and should be filled by application code otherwise. 2475 2476NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_DCR, 2477 KVM_EXIT_PAPR and KVM_EXIT_EPR the corresponding 2478operations are complete (and guest state is consistent) only after userspace 2479has re-entered the kernel with KVM_RUN. The kernel side will first finish 2480incomplete operations and then check for pending signals. Userspace 2481can re-enter the guest with an unmasked signal pending to complete 2482pending operations. 2483 2484 /* KVM_EXIT_HYPERCALL */ 2485 struct { 2486 __u64 nr; 2487 __u64 args[6]; 2488 __u64 ret; 2489 __u32 longmode; 2490 __u32 pad; 2491 } hypercall; 2492 2493Unused. This was once used for 'hypercall to userspace'. To implement 2494such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). 2495Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 2496 2497 /* KVM_EXIT_TPR_ACCESS */ 2498 struct { 2499 __u64 rip; 2500 __u32 is_write; 2501 __u32 pad; 2502 } tpr_access; 2503 2504To be documented (KVM_TPR_ACCESS_REPORTING). 2505 2506 /* KVM_EXIT_S390_SIEIC */ 2507 struct { 2508 __u8 icptcode; 2509 __u64 mask; /* psw upper half */ 2510 __u64 addr; /* psw lower half */ 2511 __u16 ipa; 2512 __u32 ipb; 2513 } s390_sieic; 2514 2515s390 specific. 2516 2517 /* KVM_EXIT_S390_RESET */ 2518#define KVM_S390_RESET_POR 1 2519#define KVM_S390_RESET_CLEAR 2 2520#define KVM_S390_RESET_SUBSYSTEM 4 2521#define KVM_S390_RESET_CPU_INIT 8 2522#define KVM_S390_RESET_IPL 16 2523 __u64 s390_reset_flags; 2524 2525s390 specific. 2526 2527 /* KVM_EXIT_S390_UCONTROL */ 2528 struct { 2529 __u64 trans_exc_code; 2530 __u32 pgm_code; 2531 } s390_ucontrol; 2532 2533s390 specific. A page fault has occurred for a user controlled virtual 2534machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be 2535resolved by the kernel. 2536The program code and the translation exception code that were placed 2537in the cpu's lowcore are presented here as defined by the z Architecture 2538Principles of Operation Book in the Chapter for Dynamic Address Translation 2539(DAT) 2540 2541 /* KVM_EXIT_DCR */ 2542 struct { 2543 __u32 dcrn; 2544 __u32 data; 2545 __u8 is_write; 2546 } dcr; 2547 2548powerpc specific. 2549 2550 /* KVM_EXIT_OSI */ 2551 struct { 2552 __u64 gprs[32]; 2553 } osi; 2554 2555MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 2556hypercalls and exit with this exit struct that contains all the guest gprs. 2557 2558If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 2559Userspace can now handle the hypercall and when it's done modify the gprs as 2560necessary. Upon guest entry all guest GPRs will then be replaced by the values 2561in this struct. 2562 2563 /* KVM_EXIT_PAPR_HCALL */ 2564 struct { 2565 __u64 nr; 2566 __u64 ret; 2567 __u64 args[9]; 2568 } papr_hcall; 2569 2570This is used on 64-bit PowerPC when emulating a pSeries partition, 2571e.g. with the 'pseries' machine type in qemu. It occurs when the 2572guest does a hypercall using the 'sc 1' instruction. The 'nr' field 2573contains the hypercall number (from the guest R3), and 'args' contains 2574the arguments (from the guest R4 - R12). Userspace should put the 2575return code in 'ret' and any extra returned values in args[]. 2576The possible hypercalls are defined in the Power Architecture Platform 2577Requirements (PAPR) document available from www.power.org (free 2578developer registration required to access it). 2579 2580 /* KVM_EXIT_S390_TSCH */ 2581 struct { 2582 __u16 subchannel_id; 2583 __u16 subchannel_nr; 2584 __u32 io_int_parm; 2585 __u32 io_int_word; 2586 __u32 ipb; 2587 __u8 dequeued; 2588 } s390_tsch; 2589 2590s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 2591and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 2592interrupt for the target subchannel has been dequeued and subchannel_id, 2593subchannel_nr, io_int_parm and io_int_word contain the parameters for that 2594interrupt. ipb is needed for instruction parameter decoding. 2595 2596 /* KVM_EXIT_EPR */ 2597 struct { 2598 __u32 epr; 2599 } epr; 2600 2601On FSL BookE PowerPC chips, the interrupt controller has a fast patch 2602interrupt acknowledge path to the core. When the core successfully 2603delivers an interrupt, it automatically populates the EPR register with 2604the interrupt vector number and acknowledges the interrupt inside 2605the interrupt controller. 2606 2607In case the interrupt controller lives in user space, we need to do 2608the interrupt acknowledge cycle through it to fetch the next to be 2609delivered interrupt vector using this exit. 2610 2611It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 2612external interrupt has just been delivered into the guest. User space 2613should put the acknowledged interrupt vector into the 'epr' field. 2614 2615 /* Fix the size of the union. */ 2616 char padding[256]; 2617 }; 2618 2619 /* 2620 * shared registers between kvm and userspace. 2621 * kvm_valid_regs specifies the register classes set by the host 2622 * kvm_dirty_regs specified the register classes dirtied by userspace 2623 * struct kvm_sync_regs is architecture specific, as well as the 2624 * bits for kvm_valid_regs and kvm_dirty_regs 2625 */ 2626 __u64 kvm_valid_regs; 2627 __u64 kvm_dirty_regs; 2628 union { 2629 struct kvm_sync_regs regs; 2630 char padding[1024]; 2631 } s; 2632 2633If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 2634certain guest registers without having to call SET/GET_*REGS. Thus we can 2635avoid some system call overhead if userspace has to handle the exit. 2636Userspace can query the validity of the structure by checking 2637kvm_valid_regs for specific bits. These bits are architecture specific 2638and usually define the validity of a groups of registers. (e.g. one bit 2639 for general purpose registers) 2640 2641}; 2642 2643 26446. Capabilities that can be enabled 2645----------------------------------- 2646 2647There are certain capabilities that change the behavior of the virtual CPU when 2648enabled. To enable them, please see section 4.37. Below you can find a list of 2649capabilities and what their effect on the vCPU is when enabling them. 2650 2651The following information is provided along with the description: 2652 2653 Architectures: which instruction set architectures provide this ioctl. 2654 x86 includes both i386 and x86_64. 2655 2656 Parameters: what parameters are accepted by the capability. 2657 2658 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) 2659 are not detailed, but errors with specific meanings are. 2660 2661 26626.1 KVM_CAP_PPC_OSI 2663 2664Architectures: ppc 2665Parameters: none 2666Returns: 0 on success; -1 on error 2667 2668This capability enables interception of OSI hypercalls that otherwise would 2669be treated as normal system calls to be injected into the guest. OSI hypercalls 2670were invented by Mac-on-Linux to have a standardized communication mechanism 2671between the guest and the host. 2672 2673When this capability is enabled, KVM_EXIT_OSI can occur. 2674 2675 26766.2 KVM_CAP_PPC_PAPR 2677 2678Architectures: ppc 2679Parameters: none 2680Returns: 0 on success; -1 on error 2681 2682This capability enables interception of PAPR hypercalls. PAPR hypercalls are 2683done using the hypercall instruction "sc 1". 2684 2685It also sets the guest privilege level to "supervisor" mode. Usually the guest 2686runs in "hypervisor" privilege mode with a few missing features. 2687 2688In addition to the above, it changes the semantics of SDR1. In this mode, the 2689HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 2690HTAB invisible to the guest. 2691 2692When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 2693 2694 26956.3 KVM_CAP_SW_TLB 2696 2697Architectures: ppc 2698Parameters: args[0] is the address of a struct kvm_config_tlb 2699Returns: 0 on success; -1 on error 2700 2701struct kvm_config_tlb { 2702 __u64 params; 2703 __u64 array; 2704 __u32 mmu_type; 2705 __u32 array_len; 2706}; 2707 2708Configures the virtual CPU's TLB array, establishing a shared memory area 2709between userspace and KVM. The "params" and "array" fields are userspace 2710addresses of mmu-type-specific data structures. The "array_len" field is an 2711safety mechanism, and should be set to the size in bytes of the memory that 2712userspace has reserved for the array. It must be at least the size dictated 2713by "mmu_type" and "params". 2714 2715While KVM_RUN is active, the shared region is under control of KVM. Its 2716contents are undefined, and any modification by userspace results in 2717boundedly undefined behavior. 2718 2719On return from KVM_RUN, the shared region will reflect the current state of 2720the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 2721to tell KVM which entries have been changed, prior to calling KVM_RUN again 2722on this vcpu. 2723 2724For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 2725 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 2726 - The "array" field points to an array of type "struct 2727 kvm_book3e_206_tlb_entry". 2728 - The array consists of all entries in the first TLB, followed by all 2729 entries in the second TLB. 2730 - Within a TLB, entries are ordered first by increasing set number. Within a 2731 set, entries are ordered by way (increasing ESEL). 2732 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 2733 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 2734 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 2735 hardware ignores this value for TLB0. 2736 27376.4 KVM_CAP_S390_CSS_SUPPORT 2738 2739Architectures: s390 2740Parameters: none 2741Returns: 0 on success; -1 on error 2742 2743This capability enables support for handling of channel I/O instructions. 2744 2745TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 2746handled in-kernel, while the other I/O instructions are passed to userspace. 2747 2748When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 2749SUBCHANNEL intercepts. 2750 27516.5 KVM_CAP_PPC_EPR 2752 2753Architectures: ppc 2754Parameters: args[0] defines whether the proxy facility is active 2755Returns: 0 on success; -1 on error 2756 2757This capability enables or disables the delivery of interrupts through the 2758external proxy facility. 2759 2760When enabled (args[0] != 0), every time the guest gets an external interrupt 2761delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 2762to receive the topmost interrupt vector. 2763 2764When disabled (args[0] == 0), behavior is as if this facility is unsupported. 2765 2766When this capability is enabled, KVM_EXIT_EPR can occur. 2767 27686.6 KVM_CAP_IRQ_MPIC 2769 2770Architectures: ppc 2771Parameters: args[0] is the MPIC device fd 2772 args[1] is the MPIC CPU number for this vcpu 2773 2774This capability connects the vcpu to an in-kernel MPIC device. 2775 27766.7 KVM_CAP_IRQ_XICS 2777 2778Architectures: ppc 2779Parameters: args[0] is the XICS device fd 2780 args[1] is the XICS CPU number (server ID) for this vcpu 2781 2782This capability connects the vcpu to an in-kernel XICS device. 2783