1.. SPDX-License-Identifier: GPL-2.0 2 3=================================================================== 4The Definitive KVM (Kernel-based Virtual Machine) API Documentation 5=================================================================== 6 71. General description 8====================== 9 10The kvm API is a set of ioctls that are issued to control various aspects 11of a virtual machine. The ioctls belong to the following classes: 12 13 - System ioctls: These query and set global attributes which affect the 14 whole kvm subsystem. In addition a system ioctl is used to create 15 virtual machines. 16 17 - VM ioctls: These query and set attributes that affect an entire virtual 18 machine, for example memory layout. In addition a VM ioctl is used to 19 create virtual cpus (vcpus) and devices. 20 21 VM ioctls must be issued from the same process (address space) that was 22 used to create the VM. 23 24 - vcpu ioctls: These query and set attributes that control the operation 25 of a single virtual cpu. 26 27 vcpu ioctls should be issued from the same thread that was used to create 28 the vcpu, except for asynchronous vcpu ioctl that are marked as such in 29 the documentation. Otherwise, the first ioctl after switching threads 30 could see a performance impact. 31 32 - device ioctls: These query and set attributes that control the operation 33 of a single device. 34 35 device ioctls must be issued from the same process (address space) that 36 was used to create the VM. 37 382. File descriptors 39=================== 40 41The kvm API is centered around file descriptors. An initial 42open("/dev/kvm") obtains a handle to the kvm subsystem; this handle 43can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this 44handle will create a VM file descriptor which can be used to issue VM 45ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVICE ioctl on a VM fd will 46create a virtual cpu or device and return a file descriptor pointing to 47the new resource. Finally, ioctls on a vcpu or device fd can be used 48to control the vcpu or device. For vcpus, this includes the important 49task of actually running guest code. 50 51In general file descriptors can be migrated among processes by means 52of fork() and the SCM_RIGHTS facility of unix domain socket. These 53kinds of tricks are explicitly not supported by kvm. While they will 54not cause harm to the host, their actual behavior is not guaranteed by 55the API. See "General description" for details on the ioctl usage 56model that is supported by KVM. 57 58It is important to note that althought VM ioctls may only be issued from 59the process that created the VM, a VM's lifecycle is associated with its 60file descriptor, not its creator (process). In other words, the VM and 61its resources, *including the associated address space*, are not freed 62until the last reference to the VM's file descriptor has been released. 63For example, if fork() is issued after ioctl(KVM_CREATE_VM), the VM will 64not be freed until both the parent (original) process and its child have 65put their references to the VM's file descriptor. 66 67Because a VM's resources are not freed until the last reference to its 68file descriptor is released, creating additional references to a VM 69via fork(), dup(), etc... without careful consideration is strongly 70discouraged and may have unwanted side effects, e.g. memory allocated 71by and on behalf of the VM's process may not be freed/unaccounted when 72the VM is shut down. 73 74 753. Extensions 76============= 77 78As of Linux 2.6.22, the KVM ABI has been stabilized: no backward 79incompatible change are allowed. However, there is an extension 80facility that allows backward-compatible extensions to the API to be 81queried and used. 82 83The extension mechanism is not based on the Linux version number. 84Instead, kvm defines extension identifiers and a facility to query 85whether a particular extension identifier is available. If it is, a 86set of ioctls is available for application use. 87 88 894. API description 90================== 91 92This section describes ioctls that can be used to control kvm guests. 93For each ioctl, the following information is provided along with a 94description: 95 96 Capability: 97 which KVM extension provides this ioctl. Can be 'basic', 98 which means that is will be provided by any kernel that supports 99 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which 100 means availability needs to be checked with KVM_CHECK_EXTENSION 101 (see section 4.4), or 'none' which means that while not all kernels 102 support this ioctl, there's no capability bit to check its 103 availability: for kernels that don't support the ioctl, 104 the ioctl returns -ENOTTY. 105 106 Architectures: 107 which instruction set architectures provide this ioctl. 108 x86 includes both i386 and x86_64. 109 110 Type: 111 system, vm, or vcpu. 112 113 Parameters: 114 what parameters are accepted by the ioctl. 115 116 Returns: 117 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 118 are not detailed, but errors with specific meanings are. 119 120 1214.1 KVM_GET_API_VERSION 122----------------------- 123 124:Capability: basic 125:Architectures: all 126:Type: system ioctl 127:Parameters: none 128:Returns: the constant KVM_API_VERSION (=12) 129 130This identifies the API version as the stable kvm API. It is not 131expected that this number will change. However, Linux 2.6.20 and 1322.6.21 report earlier versions; these are not documented and not 133supported. Applications should refuse to run if KVM_GET_API_VERSION 134returns a value other than 12. If this check passes, all ioctls 135described as 'basic' will be available. 136 137 1384.2 KVM_CREATE_VM 139----------------- 140 141:Capability: basic 142:Architectures: all 143:Type: system ioctl 144:Parameters: machine type identifier (KVM_VM_*) 145:Returns: a VM fd that can be used to control the new virtual machine. 146 147The new VM has no virtual cpus and no memory. 148You probably want to use 0 as machine type. 149 150In order to create user controlled virtual machines on S390, check 151KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as 152privileged user (CAP_SYS_ADMIN). 153 154To use hardware assisted virtualization on MIPS (VZ ASE) rather than 155the default trap & emulate implementation (which changes the virtual 156memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the 157flag KVM_VM_MIPS_VZ. 158 159 160On arm64, the physical address size for a VM (IPA Size limit) is limited 161to 40bits by default. The limit can be configured if the host supports the 162extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use 163KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type 164identifier, where IPA_Bits is the maximum width of any physical 165address used by the VM. The IPA_Bits is encoded in bits[7-0] of the 166machine type identifier. 167 168e.g, to configure a guest to use 48bit physical address size:: 169 170 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48)); 171 172The requested size (IPA_Bits) must be: 173 174 == ========================================================= 175 0 Implies default size, 40bits (for backward compatibility) 176 N Implies N bits, where N is a positive integer such that, 177 32 <= N <= Host_IPA_Limit 178 == ========================================================= 179 180Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and 181is dependent on the CPU capability and the kernel configuration. The limit can 182be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION 183ioctl() at run-time. 184 185Please note that configuring the IPA size does not affect the capability 186exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects 187size of the address translated by the stage2 level (guest physical to 188host physical address translations). 189 190 1914.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST 192---------------------------------------------------------- 193 194:Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST 195:Architectures: x86 196:Type: system ioctl 197:Parameters: struct kvm_msr_list (in/out) 198:Returns: 0 on success; -1 on error 199 200Errors: 201 202 ====== ============================================================ 203 EFAULT the msr index list cannot be read from or written to 204 E2BIG the msr index list is to be to fit in the array specified by 205 the user. 206 ====== ============================================================ 207 208:: 209 210 struct kvm_msr_list { 211 __u32 nmsrs; /* number of msrs in entries */ 212 __u32 indices[0]; 213 }; 214 215The user fills in the size of the indices array in nmsrs, and in return 216kvm adjusts nmsrs to reflect the actual number of msrs and fills in the 217indices array with their numbers. 218 219KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list 220varies by kvm version and host processor, but does not change otherwise. 221 222Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 223not returned in the MSR list, as different vcpus can have a different number 224of banks, as set via the KVM_X86_SETUP_MCE ioctl. 225 226KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed 227to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities 228and processor features that are exposed via MSRs (e.g., VMX capabilities). 229This list also varies by kvm version and host processor, but does not change 230otherwise. 231 232 2334.4 KVM_CHECK_EXTENSION 234----------------------- 235 236:Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl 237:Architectures: all 238:Type: system ioctl, vm ioctl 239:Parameters: extension identifier (KVM_CAP_*) 240:Returns: 0 if unsupported; 1 (or some other positive integer) if supported 241 242The API allows the application to query about extensions to the core 243kvm API. Userspace passes an extension identifier (an integer) and 244receives an integer that describes the extension availability. 245Generally 0 means no and 1 means yes, but some extensions may report 246additional information in the integer return value. 247 248Based on their initialization different VMs may have different capabilities. 249It is thus encouraged to use the vm ioctl to query for capabilities (available 250with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 251 2524.5 KVM_GET_VCPU_MMAP_SIZE 253-------------------------- 254 255:Capability: basic 256:Architectures: all 257:Type: system ioctl 258:Parameters: none 259:Returns: size of vcpu mmap area, in bytes 260 261The KVM_RUN ioctl (cf.) communicates with userspace via a shared 262memory region. This ioctl returns the size of that region. See the 263KVM_RUN documentation for details. 264 265Besides the size of the KVM_RUN communication region, other areas of 266the VCPU file descriptor can be mmap-ed, including: 267 268- if KVM_CAP_COALESCED_MMIO is available, a page at 269 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons, 270 this page is included in the result of KVM_GET_VCPU_MMAP_SIZE. 271 KVM_CAP_COALESCED_MMIO is not documented yet. 272 273- if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at 274 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For more information on 275 KVM_CAP_DIRTY_LOG_RING, see section 8.3. 276 277 2784.6 KVM_SET_MEMORY_REGION 279------------------------- 280 281:Capability: basic 282:Architectures: all 283:Type: vm ioctl 284:Parameters: struct kvm_memory_region (in) 285:Returns: 0 on success, -1 on error 286 287This ioctl is obsolete and has been removed. 288 289 2904.7 KVM_CREATE_VCPU 291------------------- 292 293:Capability: basic 294:Architectures: all 295:Type: vm ioctl 296:Parameters: vcpu id (apic id on x86) 297:Returns: vcpu fd on success, -1 on error 298 299This API adds a vcpu to a virtual machine. No more than max_vcpus may be added. 300The vcpu id is an integer in the range [0, max_vcpu_id). 301 302The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 303the KVM_CHECK_EXTENSION ioctl() at run-time. 304The maximum possible value for max_vcpus can be retrieved using the 305KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 306 307If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 308cpus max. 309If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 310same as the value returned from KVM_CAP_NR_VCPUS. 311 312The maximum possible value for max_vcpu_id can be retrieved using the 313KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time. 314 315If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id 316is the same as the value returned from KVM_CAP_MAX_VCPUS. 317 318On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 319threads in one or more virtual CPU cores. (This is because the 320hardware requires all the hardware threads in a CPU core to be in the 321same partition.) The KVM_CAP_PPC_SMT capability indicates the number 322of vcpus per virtual core (vcore). The vcore id is obtained by 323dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 324given vcore will always be in the same physical core as each other 325(though that might be a different physical core from time to time). 326Userspace can control the threading (SMT) mode of the guest by its 327allocation of vcpu ids. For example, if userspace wants 328single-threaded guest vcpus, it should make all vcpu ids be a multiple 329of the number of vcpus per vcore. 330 331For virtual cpus that have been created with S390 user controlled virtual 332machines, the resulting vcpu fd can be memory mapped at page offset 333KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 334cpu's hardware control block. 335 336 3374.8 KVM_GET_DIRTY_LOG (vm ioctl) 338-------------------------------- 339 340:Capability: basic 341:Architectures: all 342:Type: vm ioctl 343:Parameters: struct kvm_dirty_log (in/out) 344:Returns: 0 on success, -1 on error 345 346:: 347 348 /* for KVM_GET_DIRTY_LOG */ 349 struct kvm_dirty_log { 350 __u32 slot; 351 __u32 padding; 352 union { 353 void __user *dirty_bitmap; /* one bit per page */ 354 __u64 padding; 355 }; 356 }; 357 358Given a memory slot, return a bitmap containing any pages dirtied 359since the last call to this ioctl. Bit 0 is the first page in the 360memory slot. Ensure the entire structure is cleared to avoid padding 361issues. 362 363If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 364the address space for which you want to return the dirty bitmap. See 365KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 366 367The bits in the dirty bitmap are cleared before the ioctl returns, unless 368KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information, 369see the description of the capability. 370 3714.9 KVM_SET_MEMORY_ALIAS 372------------------------ 373 374:Capability: basic 375:Architectures: x86 376:Type: vm ioctl 377:Parameters: struct kvm_memory_alias (in) 378:Returns: 0 (success), -1 (error) 379 380This ioctl is obsolete and has been removed. 381 382 3834.10 KVM_RUN 384------------ 385 386:Capability: basic 387:Architectures: all 388:Type: vcpu ioctl 389:Parameters: none 390:Returns: 0 on success, -1 on error 391 392Errors: 393 394 ======= ============================================================== 395 EINTR an unmasked signal is pending 396 ENOEXEC the vcpu hasn't been initialized or the guest tried to execute 397 instructions from device memory (arm64) 398 ENOSYS data abort outside memslots with no syndrome info and 399 KVM_CAP_ARM_NISV_TO_USER not enabled (arm64) 400 EPERM SVE feature set but not finalized (arm64) 401 ======= ============================================================== 402 403This ioctl is used to run a guest virtual cpu. While there are no 404explicit parameters, there is an implicit parameter block that can be 405obtained by mmap()ing the vcpu fd at offset 0, with the size given by 406KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct 407kvm_run' (see below). 408 409 4104.11 KVM_GET_REGS 411----------------- 412 413:Capability: basic 414:Architectures: all except ARM, arm64 415:Type: vcpu ioctl 416:Parameters: struct kvm_regs (out) 417:Returns: 0 on success, -1 on error 418 419Reads the general purpose registers from the vcpu. 420 421:: 422 423 /* x86 */ 424 struct kvm_regs { 425 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 426 __u64 rax, rbx, rcx, rdx; 427 __u64 rsi, rdi, rsp, rbp; 428 __u64 r8, r9, r10, r11; 429 __u64 r12, r13, r14, r15; 430 __u64 rip, rflags; 431 }; 432 433 /* mips */ 434 struct kvm_regs { 435 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 436 __u64 gpr[32]; 437 __u64 hi; 438 __u64 lo; 439 __u64 pc; 440 }; 441 442 4434.12 KVM_SET_REGS 444----------------- 445 446:Capability: basic 447:Architectures: all except ARM, arm64 448:Type: vcpu ioctl 449:Parameters: struct kvm_regs (in) 450:Returns: 0 on success, -1 on error 451 452Writes the general purpose registers into the vcpu. 453 454See KVM_GET_REGS for the data structure. 455 456 4574.13 KVM_GET_SREGS 458------------------ 459 460:Capability: basic 461:Architectures: x86, ppc 462:Type: vcpu ioctl 463:Parameters: struct kvm_sregs (out) 464:Returns: 0 on success, -1 on error 465 466Reads special registers from the vcpu. 467 468:: 469 470 /* x86 */ 471 struct kvm_sregs { 472 struct kvm_segment cs, ds, es, fs, gs, ss; 473 struct kvm_segment tr, ldt; 474 struct kvm_dtable gdt, idt; 475 __u64 cr0, cr2, cr3, cr4, cr8; 476 __u64 efer; 477 __u64 apic_base; 478 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; 479 }; 480 481 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ 482 483interrupt_bitmap is a bitmap of pending external interrupts. At most 484one bit may be set. This interrupt has been acknowledged by the APIC 485but not yet injected into the cpu core. 486 487 4884.14 KVM_SET_SREGS 489------------------ 490 491:Capability: basic 492:Architectures: x86, ppc 493:Type: vcpu ioctl 494:Parameters: struct kvm_sregs (in) 495:Returns: 0 on success, -1 on error 496 497Writes special registers into the vcpu. See KVM_GET_SREGS for the 498data structures. 499 500 5014.15 KVM_TRANSLATE 502------------------ 503 504:Capability: basic 505:Architectures: x86 506:Type: vcpu ioctl 507:Parameters: struct kvm_translation (in/out) 508:Returns: 0 on success, -1 on error 509 510Translates a virtual address according to the vcpu's current address 511translation mode. 512 513:: 514 515 struct kvm_translation { 516 /* in */ 517 __u64 linear_address; 518 519 /* out */ 520 __u64 physical_address; 521 __u8 valid; 522 __u8 writeable; 523 __u8 usermode; 524 __u8 pad[5]; 525 }; 526 527 5284.16 KVM_INTERRUPT 529------------------ 530 531:Capability: basic 532:Architectures: x86, ppc, mips 533:Type: vcpu ioctl 534:Parameters: struct kvm_interrupt (in) 535:Returns: 0 on success, negative on failure. 536 537Queues a hardware interrupt vector to be injected. 538 539:: 540 541 /* for KVM_INTERRUPT */ 542 struct kvm_interrupt { 543 /* in */ 544 __u32 irq; 545 }; 546 547X86: 548^^^^ 549 550:Returns: 551 552 ========= =================================== 553 0 on success, 554 -EEXIST if an interrupt is already enqueued 555 -EINVAL the irq number is invalid 556 -ENXIO if the PIC is in the kernel 557 -EFAULT if the pointer is invalid 558 ========= =================================== 559 560Note 'irq' is an interrupt vector, not an interrupt pin or line. This 561ioctl is useful if the in-kernel PIC is not used. 562 563PPC: 564^^^^ 565 566Queues an external interrupt to be injected. This ioctl is overleaded 567with 3 different irq values: 568 569a) KVM_INTERRUPT_SET 570 571 This injects an edge type external interrupt into the guest once it's ready 572 to receive interrupts. When injected, the interrupt is done. 573 574b) KVM_INTERRUPT_UNSET 575 576 This unsets any pending interrupt. 577 578 Only available with KVM_CAP_PPC_UNSET_IRQ. 579 580c) KVM_INTERRUPT_SET_LEVEL 581 582 This injects a level type external interrupt into the guest context. The 583 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET 584 is triggered. 585 586 Only available with KVM_CAP_PPC_IRQ_LEVEL. 587 588Note that any value for 'irq' other than the ones stated above is invalid 589and incurs unexpected behavior. 590 591This is an asynchronous vcpu ioctl and can be invoked from any thread. 592 593MIPS: 594^^^^^ 595 596Queues an external interrupt to be injected into the virtual CPU. A negative 597interrupt number dequeues the interrupt. 598 599This is an asynchronous vcpu ioctl and can be invoked from any thread. 600 601 6024.17 KVM_DEBUG_GUEST 603-------------------- 604 605:Capability: basic 606:Architectures: none 607:Type: vcpu ioctl 608:Parameters: none) 609:Returns: -1 on error 610 611Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. 612 613 6144.18 KVM_GET_MSRS 615----------------- 616 617:Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system) 618:Architectures: x86 619:Type: system ioctl, vcpu ioctl 620:Parameters: struct kvm_msrs (in/out) 621:Returns: number of msrs successfully returned; 622 -1 on error 623 624When used as a system ioctl: 625Reads the values of MSR-based features that are available for the VM. This 626is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values. 627The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST 628in a system ioctl. 629 630When used as a vcpu ioctl: 631Reads model-specific registers from the vcpu. Supported msr indices can 632be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl. 633 634:: 635 636 struct kvm_msrs { 637 __u32 nmsrs; /* number of msrs in entries */ 638 __u32 pad; 639 640 struct kvm_msr_entry entries[0]; 641 }; 642 643 struct kvm_msr_entry { 644 __u32 index; 645 __u32 reserved; 646 __u64 data; 647 }; 648 649Application code should set the 'nmsrs' member (which indicates the 650size of the entries array) and the 'index' member of each array entry. 651kvm will fill in the 'data' member. 652 653 6544.19 KVM_SET_MSRS 655----------------- 656 657:Capability: basic 658:Architectures: x86 659:Type: vcpu ioctl 660:Parameters: struct kvm_msrs (in) 661:Returns: number of msrs successfully set (see below), -1 on error 662 663Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 664data structures. 665 666Application code should set the 'nmsrs' member (which indicates the 667size of the entries array), and the 'index' and 'data' members of each 668array entry. 669 670It tries to set the MSRs in array entries[] one by one. If setting an MSR 671fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated 672by KVM, etc..., it stops processing the MSR list and returns the number of 673MSRs that have been set successfully. 674 675 6764.20 KVM_SET_CPUID 677------------------ 678 679:Capability: basic 680:Architectures: x86 681:Type: vcpu ioctl 682:Parameters: struct kvm_cpuid (in) 683:Returns: 0 on success, -1 on error 684 685Defines the vcpu responses to the cpuid instruction. Applications 686should use the KVM_SET_CPUID2 ioctl if available. 687 688Note, when this IOCTL fails, KVM gives no guarantees that previous valid CPUID 689configuration (if there is) is not corrupted. Userspace can get a copy of the 690resulting CPUID configuration through KVM_GET_CPUID2 in case. 691 692:: 693 694 struct kvm_cpuid_entry { 695 __u32 function; 696 __u32 eax; 697 __u32 ebx; 698 __u32 ecx; 699 __u32 edx; 700 __u32 padding; 701 }; 702 703 /* for KVM_SET_CPUID */ 704 struct kvm_cpuid { 705 __u32 nent; 706 __u32 padding; 707 struct kvm_cpuid_entry entries[0]; 708 }; 709 710 7114.21 KVM_SET_SIGNAL_MASK 712------------------------ 713 714:Capability: basic 715:Architectures: all 716:Type: vcpu ioctl 717:Parameters: struct kvm_signal_mask (in) 718:Returns: 0 on success, -1 on error 719 720Defines which signals are blocked during execution of KVM_RUN. This 721signal mask temporarily overrides the threads signal mask. Any 722unblocked signal received (except SIGKILL and SIGSTOP, which retain 723their traditional behaviour) will cause KVM_RUN to return with -EINTR. 724 725Note the signal will only be delivered if not blocked by the original 726signal mask. 727 728:: 729 730 /* for KVM_SET_SIGNAL_MASK */ 731 struct kvm_signal_mask { 732 __u32 len; 733 __u8 sigset[0]; 734 }; 735 736 7374.22 KVM_GET_FPU 738---------------- 739 740:Capability: basic 741:Architectures: x86 742:Type: vcpu ioctl 743:Parameters: struct kvm_fpu (out) 744:Returns: 0 on success, -1 on error 745 746Reads the floating point state from the vcpu. 747 748:: 749 750 /* for KVM_GET_FPU and KVM_SET_FPU */ 751 struct kvm_fpu { 752 __u8 fpr[8][16]; 753 __u16 fcw; 754 __u16 fsw; 755 __u8 ftwx; /* in fxsave format */ 756 __u8 pad1; 757 __u16 last_opcode; 758 __u64 last_ip; 759 __u64 last_dp; 760 __u8 xmm[16][16]; 761 __u32 mxcsr; 762 __u32 pad2; 763 }; 764 765 7664.23 KVM_SET_FPU 767---------------- 768 769:Capability: basic 770:Architectures: x86 771:Type: vcpu ioctl 772:Parameters: struct kvm_fpu (in) 773:Returns: 0 on success, -1 on error 774 775Writes the floating point state to the vcpu. 776 777:: 778 779 /* for KVM_GET_FPU and KVM_SET_FPU */ 780 struct kvm_fpu { 781 __u8 fpr[8][16]; 782 __u16 fcw; 783 __u16 fsw; 784 __u8 ftwx; /* in fxsave format */ 785 __u8 pad1; 786 __u16 last_opcode; 787 __u64 last_ip; 788 __u64 last_dp; 789 __u8 xmm[16][16]; 790 __u32 mxcsr; 791 __u32 pad2; 792 }; 793 794 7954.24 KVM_CREATE_IRQCHIP 796----------------------- 797 798:Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) 799:Architectures: x86, ARM, arm64, s390 800:Type: vm ioctl 801:Parameters: none 802:Returns: 0 on success, -1 on error 803 804Creates an interrupt controller model in the kernel. 805On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up 806future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both 807PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. 808On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of 809KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 810KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 811On s390, a dummy irq routing table is created. 812 813Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 814before KVM_CREATE_IRQCHIP can be used. 815 816 8174.25 KVM_IRQ_LINE 818----------------- 819 820:Capability: KVM_CAP_IRQCHIP 821:Architectures: x86, arm, arm64 822:Type: vm ioctl 823:Parameters: struct kvm_irq_level 824:Returns: 0 on success, -1 on error 825 826Sets the level of a GSI input to the interrupt controller model in the kernel. 827On some architectures it is required that an interrupt controller model has 828been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 829interrupts require the level to be set to 1 and then back to 0. 830 831On real hardware, interrupt pins can be active-low or active-high. This 832does not matter for the level field of struct kvm_irq_level: 1 always 833means active (asserted), 0 means inactive (deasserted). 834 835x86 allows the operating system to program the interrupt polarity 836(active-low/active-high) for level-triggered interrupts, and KVM used 837to consider the polarity. However, due to bitrot in the handling of 838active-low interrupts, the above convention is now valid on x86 too. 839This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 840should not present interrupts to the guest as active-low unless this 841capability is present (or unless it is not using the in-kernel irqchip, 842of course). 843 844 845ARM/arm64 can signal an interrupt either at the CPU level, or at the 846in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 847use PPIs designated for specific cpus. The irq field is interpreted 848like this:: 849 850 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 | 851 field: | vcpu2_index | irq_type | vcpu_index | irq_id | 852 853The irq_type field has the following values: 854 855- irq_type[0]: 856 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 857- irq_type[1]: 858 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 859 (the vcpu_index field is ignored) 860- irq_type[2]: 861 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 862 863(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 864 865In both cases, level is used to assert/deassert the line. 866 867When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is 868identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index 869must be zero. 870 871Note that on arm/arm64, the KVM_CAP_IRQCHIP capability only conditions 872injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always 873be used for a userspace interrupt controller. 874 875:: 876 877 struct kvm_irq_level { 878 union { 879 __u32 irq; /* GSI */ 880 __s32 status; /* not used for KVM_IRQ_LEVEL */ 881 }; 882 __u32 level; /* 0 or 1 */ 883 }; 884 885 8864.26 KVM_GET_IRQCHIP 887-------------------- 888 889:Capability: KVM_CAP_IRQCHIP 890:Architectures: x86 891:Type: vm ioctl 892:Parameters: struct kvm_irqchip (in/out) 893:Returns: 0 on success, -1 on error 894 895Reads the state of a kernel interrupt controller created with 896KVM_CREATE_IRQCHIP into a buffer provided by the caller. 897 898:: 899 900 struct kvm_irqchip { 901 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 902 __u32 pad; 903 union { 904 char dummy[512]; /* reserving space */ 905 struct kvm_pic_state pic; 906 struct kvm_ioapic_state ioapic; 907 } chip; 908 }; 909 910 9114.27 KVM_SET_IRQCHIP 912-------------------- 913 914:Capability: KVM_CAP_IRQCHIP 915:Architectures: x86 916:Type: vm ioctl 917:Parameters: struct kvm_irqchip (in) 918:Returns: 0 on success, -1 on error 919 920Sets the state of a kernel interrupt controller created with 921KVM_CREATE_IRQCHIP from a buffer provided by the caller. 922 923:: 924 925 struct kvm_irqchip { 926 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 927 __u32 pad; 928 union { 929 char dummy[512]; /* reserving space */ 930 struct kvm_pic_state pic; 931 struct kvm_ioapic_state ioapic; 932 } chip; 933 }; 934 935 9364.28 KVM_XEN_HVM_CONFIG 937----------------------- 938 939:Capability: KVM_CAP_XEN_HVM 940:Architectures: x86 941:Type: vm ioctl 942:Parameters: struct kvm_xen_hvm_config (in) 943:Returns: 0 on success, -1 on error 944 945Sets the MSR that the Xen HVM guest uses to initialize its hypercall 946page, and provides the starting address and size of the hypercall 947blobs in userspace. When the guest writes the MSR, kvm copies one 948page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 949memory. 950 951:: 952 953 struct kvm_xen_hvm_config { 954 __u32 flags; 955 __u32 msr; 956 __u64 blob_addr_32; 957 __u64 blob_addr_64; 958 __u8 blob_size_32; 959 __u8 blob_size_64; 960 __u8 pad2[30]; 961 }; 962 963If the KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag is returned from the 964KVM_CAP_XEN_HVM check, it may be set in the flags field of this ioctl. 965This requests KVM to generate the contents of the hypercall page 966automatically; hypercalls will be intercepted and passed to userspace 967through KVM_EXIT_XEN. In this case, all of the blob size and address 968fields must be zero. 969 970No other flags are currently valid in the struct kvm_xen_hvm_config. 971 9724.29 KVM_GET_CLOCK 973------------------ 974 975:Capability: KVM_CAP_ADJUST_CLOCK 976:Architectures: x86 977:Type: vm ioctl 978:Parameters: struct kvm_clock_data (out) 979:Returns: 0 on success, -1 on error 980 981Gets the current timestamp of kvmclock as seen by the current guest. In 982conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 983such as migration. 984 985When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 986set of bits that KVM can return in struct kvm_clock_data's flag member. 987 988The only flag defined now is KVM_CLOCK_TSC_STABLE. If set, the returned 989value is the exact kvmclock value seen by all VCPUs at the instant 990when KVM_GET_CLOCK was called. If clear, the returned value is simply 991CLOCK_MONOTONIC plus a constant offset; the offset can be modified 992with KVM_SET_CLOCK. KVM will try to make all VCPUs follow this clock, 993but the exact value read by each VCPU could differ, because the host 994TSC is not stable. 995 996:: 997 998 struct kvm_clock_data { 999 __u64 clock; /* kvmclock current value */ 1000 __u32 flags; 1001 __u32 pad[9]; 1002 }; 1003 1004 10054.30 KVM_SET_CLOCK 1006------------------ 1007 1008:Capability: KVM_CAP_ADJUST_CLOCK 1009:Architectures: x86 1010:Type: vm ioctl 1011:Parameters: struct kvm_clock_data (in) 1012:Returns: 0 on success, -1 on error 1013 1014Sets the current timestamp of kvmclock to the value specified in its parameter. 1015In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 1016such as migration. 1017 1018:: 1019 1020 struct kvm_clock_data { 1021 __u64 clock; /* kvmclock current value */ 1022 __u32 flags; 1023 __u32 pad[9]; 1024 }; 1025 1026 10274.31 KVM_GET_VCPU_EVENTS 1028------------------------ 1029 1030:Capability: KVM_CAP_VCPU_EVENTS 1031:Extended by: KVM_CAP_INTR_SHADOW 1032:Architectures: x86, arm, arm64 1033:Type: vcpu ioctl 1034:Parameters: struct kvm_vcpu_event (out) 1035:Returns: 0 on success, -1 on error 1036 1037X86: 1038^^^^ 1039 1040Gets currently pending exceptions, interrupts, and NMIs as well as related 1041states of the vcpu. 1042 1043:: 1044 1045 struct kvm_vcpu_events { 1046 struct { 1047 __u8 injected; 1048 __u8 nr; 1049 __u8 has_error_code; 1050 __u8 pending; 1051 __u32 error_code; 1052 } exception; 1053 struct { 1054 __u8 injected; 1055 __u8 nr; 1056 __u8 soft; 1057 __u8 shadow; 1058 } interrupt; 1059 struct { 1060 __u8 injected; 1061 __u8 pending; 1062 __u8 masked; 1063 __u8 pad; 1064 } nmi; 1065 __u32 sipi_vector; 1066 __u32 flags; 1067 struct { 1068 __u8 smm; 1069 __u8 pending; 1070 __u8 smm_inside_nmi; 1071 __u8 latched_init; 1072 } smi; 1073 __u8 reserved[27]; 1074 __u8 exception_has_payload; 1075 __u64 exception_payload; 1076 }; 1077 1078The following bits are defined in the flags field: 1079 1080- KVM_VCPUEVENT_VALID_SHADOW may be set to signal that 1081 interrupt.shadow contains a valid state. 1082 1083- KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a 1084 valid state. 1085 1086- KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the 1087 exception_has_payload, exception_payload, and exception.pending 1088 fields contain a valid state. This bit will be set whenever 1089 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1090 1091ARM/ARM64: 1092^^^^^^^^^^ 1093 1094If the guest accesses a device that is being emulated by the host kernel in 1095such a way that a real device would generate a physical SError, KVM may make 1096a virtual SError pending for that VCPU. This system error interrupt remains 1097pending until the guest takes the exception by unmasking PSTATE.A. 1098 1099Running the VCPU may cause it to take a pending SError, or make an access that 1100causes an SError to become pending. The event's description is only valid while 1101the VPCU is not running. 1102 1103This API provides a way to read and write the pending 'event' state that is not 1104visible to the guest. To save, restore or migrate a VCPU the struct representing 1105the state can be read then written using this GET/SET API, along with the other 1106guest-visible registers. It is not possible to 'cancel' an SError that has been 1107made pending. 1108 1109A device being emulated in user-space may also wish to generate an SError. To do 1110this the events structure can be populated by user-space. The current state 1111should be read first, to ensure no existing SError is pending. If an existing 1112SError is pending, the architecture's 'Multiple SError interrupts' rules should 1113be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and 1114Serviceability (RAS) Specification"). 1115 1116SError exceptions always have an ESR value. Some CPUs have the ability to 1117specify what the virtual SError's ESR value should be. These systems will 1118advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will 1119always have a non-zero value when read, and the agent making an SError pending 1120should specify the ISS field in the lower 24 bits of exception.serror_esr. If 1121the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events 1122with exception.has_esr as zero, KVM will choose an ESR. 1123 1124Specifying exception.has_esr on a system that does not support it will return 1125-EINVAL. Setting anything other than the lower 24bits of exception.serror_esr 1126will return -EINVAL. 1127 1128It is not possible to read back a pending external abort (injected via 1129KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered 1130directly to the virtual CPU). 1131 1132:: 1133 1134 struct kvm_vcpu_events { 1135 struct { 1136 __u8 serror_pending; 1137 __u8 serror_has_esr; 1138 __u8 ext_dabt_pending; 1139 /* Align it to 8 bytes */ 1140 __u8 pad[5]; 1141 __u64 serror_esr; 1142 } exception; 1143 __u32 reserved[12]; 1144 }; 1145 11464.32 KVM_SET_VCPU_EVENTS 1147------------------------ 1148 1149:Capability: KVM_CAP_VCPU_EVENTS 1150:Extended by: KVM_CAP_INTR_SHADOW 1151:Architectures: x86, arm, arm64 1152:Type: vcpu ioctl 1153:Parameters: struct kvm_vcpu_event (in) 1154:Returns: 0 on success, -1 on error 1155 1156X86: 1157^^^^ 1158 1159Set pending exceptions, interrupts, and NMIs as well as related states of the 1160vcpu. 1161 1162See KVM_GET_VCPU_EVENTS for the data structure. 1163 1164Fields that may be modified asynchronously by running VCPUs can be excluded 1165from the update. These fields are nmi.pending, sipi_vector, smi.smm, 1166smi.pending. Keep the corresponding bits in the flags field cleared to 1167suppress overwriting the current in-kernel state. The bits are: 1168 1169=============================== ================================== 1170KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel 1171KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector 1172KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct. 1173=============================== ================================== 1174 1175If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 1176the flags field to signal that interrupt.shadow contains a valid state and 1177shall be written into the VCPU. 1178 1179KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 1180 1181If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD 1182can be set in the flags field to signal that the 1183exception_has_payload, exception_payload, and exception.pending fields 1184contain a valid state and shall be written into the VCPU. 1185 1186ARM/ARM64: 1187^^^^^^^^^^ 1188 1189User space may need to inject several types of events to the guest. 1190 1191Set the pending SError exception state for this VCPU. It is not possible to 1192'cancel' an Serror that has been made pending. 1193 1194If the guest performed an access to I/O memory which could not be handled by 1195userspace, for example because of missing instruction syndrome decode 1196information or because there is no device mapped at the accessed IPA, then 1197userspace can ask the kernel to inject an external abort using the address 1198from the exiting fault on the VCPU. It is a programming error to set 1199ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or 1200KVM_EXIT_ARM_NISV. This feature is only available if the system supports 1201KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in 1202how userspace reports accesses for the above cases to guests, across different 1203userspace implementations. Nevertheless, userspace can still emulate all Arm 1204exceptions by manipulating individual registers using the KVM_SET_ONE_REG API. 1205 1206See KVM_GET_VCPU_EVENTS for the data structure. 1207 1208 12094.33 KVM_GET_DEBUGREGS 1210---------------------- 1211 1212:Capability: KVM_CAP_DEBUGREGS 1213:Architectures: x86 1214:Type: vm ioctl 1215:Parameters: struct kvm_debugregs (out) 1216:Returns: 0 on success, -1 on error 1217 1218Reads debug registers from the vcpu. 1219 1220:: 1221 1222 struct kvm_debugregs { 1223 __u64 db[4]; 1224 __u64 dr6; 1225 __u64 dr7; 1226 __u64 flags; 1227 __u64 reserved[9]; 1228 }; 1229 1230 12314.34 KVM_SET_DEBUGREGS 1232---------------------- 1233 1234:Capability: KVM_CAP_DEBUGREGS 1235:Architectures: x86 1236:Type: vm ioctl 1237:Parameters: struct kvm_debugregs (in) 1238:Returns: 0 on success, -1 on error 1239 1240Writes debug registers into the vcpu. 1241 1242See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 1243yet and must be cleared on entry. 1244 1245 12464.35 KVM_SET_USER_MEMORY_REGION 1247------------------------------- 1248 1249:Capability: KVM_CAP_USER_MEMORY 1250:Architectures: all 1251:Type: vm ioctl 1252:Parameters: struct kvm_userspace_memory_region (in) 1253:Returns: 0 on success, -1 on error 1254 1255:: 1256 1257 struct kvm_userspace_memory_region { 1258 __u32 slot; 1259 __u32 flags; 1260 __u64 guest_phys_addr; 1261 __u64 memory_size; /* bytes */ 1262 __u64 userspace_addr; /* start of the userspace allocated memory */ 1263 }; 1264 1265 /* for kvm_memory_region::flags */ 1266 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 1267 #define KVM_MEM_READONLY (1UL << 1) 1268 1269This ioctl allows the user to create, modify or delete a guest physical 1270memory slot. Bits 0-15 of "slot" specify the slot id and this value 1271should be less than the maximum number of user memory slots supported per 1272VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS. 1273Slots may not overlap in guest physical address space. 1274 1275If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 1276specifies the address space which is being modified. They must be 1277less than the value that KVM_CHECK_EXTENSION returns for the 1278KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 1279are unrelated; the restriction on overlapping slots only applies within 1280each address space. 1281 1282Deleting a slot is done by passing zero for memory_size. When changing 1283an existing slot, it may be moved in the guest physical memory space, 1284or its flags may be modified, but it may not be resized. 1285 1286Memory for the region is taken starting at the address denoted by the 1287field userspace_addr, which must point at user addressable memory for 1288the entire memory slot size. Any object may back this memory, including 1289anonymous memory, ordinary files, and hugetlbfs. 1290 1291On architectures that support a form of address tagging, userspace_addr must 1292be an untagged address. 1293 1294It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 1295be identical. This allows large pages in the guest to be backed by large 1296pages in the host. 1297 1298The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 1299KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 1300writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 1301use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 1302to make a new slot read-only. In this case, writes to this memory will be 1303posted to userspace as KVM_EXIT_MMIO exits. 1304 1305When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 1306the memory region are automatically reflected into the guest. For example, an 1307mmap() that affects the region will be made visible immediately. Another 1308example is madvise(MADV_DROP). 1309 1310It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. 1311The KVM_SET_MEMORY_REGION does not allow fine grained control over memory 1312allocation and is deprecated. 1313 1314 13154.36 KVM_SET_TSS_ADDR 1316--------------------- 1317 1318:Capability: KVM_CAP_SET_TSS_ADDR 1319:Architectures: x86 1320:Type: vm ioctl 1321:Parameters: unsigned long tss_address (in) 1322:Returns: 0 on success, -1 on error 1323 1324This ioctl defines the physical address of a three-page region in the guest 1325physical address space. The region must be within the first 4GB of the 1326guest physical address space and must not conflict with any memory slot 1327or any mmio address. The guest may malfunction if it accesses this memory 1328region. 1329 1330This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1331because of a quirk in the virtualization implementation (see the internals 1332documentation when it pops into existence). 1333 1334 13354.37 KVM_ENABLE_CAP 1336------------------- 1337 1338:Capability: KVM_CAP_ENABLE_CAP 1339:Architectures: mips, ppc, s390 1340:Type: vcpu ioctl 1341:Parameters: struct kvm_enable_cap (in) 1342:Returns: 0 on success; -1 on error 1343 1344:Capability: KVM_CAP_ENABLE_CAP_VM 1345:Architectures: all 1346:Type: vm ioctl 1347:Parameters: struct kvm_enable_cap (in) 1348:Returns: 0 on success; -1 on error 1349 1350.. note:: 1351 1352 Not all extensions are enabled by default. Using this ioctl the application 1353 can enable an extension, making it available to the guest. 1354 1355On systems that do not support this ioctl, it always fails. On systems that 1356do support it, it only works for extensions that are supported for enablement. 1357 1358To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1359be used. 1360 1361:: 1362 1363 struct kvm_enable_cap { 1364 /* in */ 1365 __u32 cap; 1366 1367The capability that is supposed to get enabled. 1368 1369:: 1370 1371 __u32 flags; 1372 1373A bitfield indicating future enhancements. Has to be 0 for now. 1374 1375:: 1376 1377 __u64 args[4]; 1378 1379Arguments for enabling a feature. If a feature needs initial values to 1380function properly, this is the place to put them. 1381 1382:: 1383 1384 __u8 pad[64]; 1385 }; 1386 1387The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1388for vm-wide capabilities. 1389 13904.38 KVM_GET_MP_STATE 1391--------------------- 1392 1393:Capability: KVM_CAP_MP_STATE 1394:Architectures: x86, s390, arm, arm64 1395:Type: vcpu ioctl 1396:Parameters: struct kvm_mp_state (out) 1397:Returns: 0 on success; -1 on error 1398 1399:: 1400 1401 struct kvm_mp_state { 1402 __u32 mp_state; 1403 }; 1404 1405Returns the vcpu's current "multiprocessing state" (though also valid on 1406uniprocessor guests). 1407 1408Possible values are: 1409 1410 ========================== =============================================== 1411 KVM_MP_STATE_RUNNABLE the vcpu is currently running [x86,arm/arm64] 1412 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP) 1413 which has not yet received an INIT signal [x86] 1414 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is 1415 now ready for a SIPI [x86] 1416 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and 1417 is waiting for an interrupt [x86] 1418 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector 1419 accessible via KVM_GET_VCPU_EVENTS) [x86] 1420 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm/arm64] 1421 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390] 1422 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted) 1423 [s390] 1424 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state 1425 [s390] 1426 ========================== =============================================== 1427 1428On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1429in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1430these architectures. 1431 1432For arm/arm64: 1433^^^^^^^^^^^^^^ 1434 1435The only states that are valid are KVM_MP_STATE_STOPPED and 1436KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1437 14384.39 KVM_SET_MP_STATE 1439--------------------- 1440 1441:Capability: KVM_CAP_MP_STATE 1442:Architectures: x86, s390, arm, arm64 1443:Type: vcpu ioctl 1444:Parameters: struct kvm_mp_state (in) 1445:Returns: 0 on success; -1 on error 1446 1447Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1448arguments. 1449 1450On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1451in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1452these architectures. 1453 1454For arm/arm64: 1455^^^^^^^^^^^^^^ 1456 1457The only states that are valid are KVM_MP_STATE_STOPPED and 1458KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1459 14604.40 KVM_SET_IDENTITY_MAP_ADDR 1461------------------------------ 1462 1463:Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1464:Architectures: x86 1465:Type: vm ioctl 1466:Parameters: unsigned long identity (in) 1467:Returns: 0 on success, -1 on error 1468 1469This ioctl defines the physical address of a one-page region in the guest 1470physical address space. The region must be within the first 4GB of the 1471guest physical address space and must not conflict with any memory slot 1472or any mmio address. The guest may malfunction if it accesses this memory 1473region. 1474 1475Setting the address to 0 will result in resetting the address to its default 1476(0xfffbc000). 1477 1478This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1479because of a quirk in the virtualization implementation (see the internals 1480documentation when it pops into existence). 1481 1482Fails if any VCPU has already been created. 1483 14844.41 KVM_SET_BOOT_CPU_ID 1485------------------------ 1486 1487:Capability: KVM_CAP_SET_BOOT_CPU_ID 1488:Architectures: x86 1489:Type: vm ioctl 1490:Parameters: unsigned long vcpu_id 1491:Returns: 0 on success, -1 on error 1492 1493Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1494as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1495is vcpu 0. 1496 1497 14984.42 KVM_GET_XSAVE 1499------------------ 1500 1501:Capability: KVM_CAP_XSAVE 1502:Architectures: x86 1503:Type: vcpu ioctl 1504:Parameters: struct kvm_xsave (out) 1505:Returns: 0 on success, -1 on error 1506 1507 1508:: 1509 1510 struct kvm_xsave { 1511 __u32 region[1024]; 1512 }; 1513 1514This ioctl would copy current vcpu's xsave struct to the userspace. 1515 1516 15174.43 KVM_SET_XSAVE 1518------------------ 1519 1520:Capability: KVM_CAP_XSAVE 1521:Architectures: x86 1522:Type: vcpu ioctl 1523:Parameters: struct kvm_xsave (in) 1524:Returns: 0 on success, -1 on error 1525 1526:: 1527 1528 1529 struct kvm_xsave { 1530 __u32 region[1024]; 1531 }; 1532 1533This ioctl would copy userspace's xsave struct to the kernel. 1534 1535 15364.44 KVM_GET_XCRS 1537----------------- 1538 1539:Capability: KVM_CAP_XCRS 1540:Architectures: x86 1541:Type: vcpu ioctl 1542:Parameters: struct kvm_xcrs (out) 1543:Returns: 0 on success, -1 on error 1544 1545:: 1546 1547 struct kvm_xcr { 1548 __u32 xcr; 1549 __u32 reserved; 1550 __u64 value; 1551 }; 1552 1553 struct kvm_xcrs { 1554 __u32 nr_xcrs; 1555 __u32 flags; 1556 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1557 __u64 padding[16]; 1558 }; 1559 1560This ioctl would copy current vcpu's xcrs to the userspace. 1561 1562 15634.45 KVM_SET_XCRS 1564----------------- 1565 1566:Capability: KVM_CAP_XCRS 1567:Architectures: x86 1568:Type: vcpu ioctl 1569:Parameters: struct kvm_xcrs (in) 1570:Returns: 0 on success, -1 on error 1571 1572:: 1573 1574 struct kvm_xcr { 1575 __u32 xcr; 1576 __u32 reserved; 1577 __u64 value; 1578 }; 1579 1580 struct kvm_xcrs { 1581 __u32 nr_xcrs; 1582 __u32 flags; 1583 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1584 __u64 padding[16]; 1585 }; 1586 1587This ioctl would set vcpu's xcr to the value userspace specified. 1588 1589 15904.46 KVM_GET_SUPPORTED_CPUID 1591---------------------------- 1592 1593:Capability: KVM_CAP_EXT_CPUID 1594:Architectures: x86 1595:Type: system ioctl 1596:Parameters: struct kvm_cpuid2 (in/out) 1597:Returns: 0 on success, -1 on error 1598 1599:: 1600 1601 struct kvm_cpuid2 { 1602 __u32 nent; 1603 __u32 padding; 1604 struct kvm_cpuid_entry2 entries[0]; 1605 }; 1606 1607 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1608 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 1609 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 1610 1611 struct kvm_cpuid_entry2 { 1612 __u32 function; 1613 __u32 index; 1614 __u32 flags; 1615 __u32 eax; 1616 __u32 ebx; 1617 __u32 ecx; 1618 __u32 edx; 1619 __u32 padding[3]; 1620 }; 1621 1622This ioctl returns x86 cpuid features which are supported by both the 1623hardware and kvm in its default configuration. Userspace can use the 1624information returned by this ioctl to construct cpuid information (for 1625KVM_SET_CPUID2) that is consistent with hardware, kernel, and 1626userspace capabilities, and with user requirements (for example, the 1627user may wish to constrain cpuid to emulate older hardware, or for 1628feature consistency across a cluster). 1629 1630Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may 1631expose cpuid features (e.g. MONITOR) which are not supported by kvm in 1632its default configuration. If userspace enables such capabilities, it 1633is responsible for modifying the results of this ioctl appropriately. 1634 1635Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1636with the 'nent' field indicating the number of entries in the variable-size 1637array 'entries'. If the number of entries is too low to describe the cpu 1638capabilities, an error (E2BIG) is returned. If the number is too high, 1639the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1640number is just right, the 'nent' field is adjusted to the number of valid 1641entries in the 'entries' array, which is then filled. 1642 1643The entries returned are the host cpuid as returned by the cpuid instruction, 1644with unknown or unsupported features masked out. Some features (for example, 1645x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1646emulate them efficiently. The fields in each entry are defined as follows: 1647 1648 function: 1649 the eax value used to obtain the entry 1650 1651 index: 1652 the ecx value used to obtain the entry (for entries that are 1653 affected by ecx) 1654 1655 flags: 1656 an OR of zero or more of the following: 1657 1658 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1659 if the index field is valid 1660 1661 eax, ebx, ecx, edx: 1662 the values returned by the cpuid instruction for 1663 this function/index combination 1664 1665The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned 1666as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC 1667support. Instead it is reported via:: 1668 1669 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1670 1671if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1672feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1673 1674 16754.47 KVM_PPC_GET_PVINFO 1676----------------------- 1677 1678:Capability: KVM_CAP_PPC_GET_PVINFO 1679:Architectures: ppc 1680:Type: vm ioctl 1681:Parameters: struct kvm_ppc_pvinfo (out) 1682:Returns: 0 on success, !0 on error 1683 1684:: 1685 1686 struct kvm_ppc_pvinfo { 1687 __u32 flags; 1688 __u32 hcall[4]; 1689 __u8 pad[108]; 1690 }; 1691 1692This ioctl fetches PV specific information that need to be passed to the guest 1693using the device tree or other means from vm context. 1694 1695The hcall array defines 4 instructions that make up a hypercall. 1696 1697If any additional field gets added to this structure later on, a bit for that 1698additional piece of information will be set in the flags bitmap. 1699 1700The flags bitmap is defined as:: 1701 1702 /* the host supports the ePAPR idle hcall 1703 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1704 17054.52 KVM_SET_GSI_ROUTING 1706------------------------ 1707 1708:Capability: KVM_CAP_IRQ_ROUTING 1709:Architectures: x86 s390 arm arm64 1710:Type: vm ioctl 1711:Parameters: struct kvm_irq_routing (in) 1712:Returns: 0 on success, -1 on error 1713 1714Sets the GSI routing table entries, overwriting any previously set entries. 1715 1716On arm/arm64, GSI routing has the following limitation: 1717 1718- GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1719 1720:: 1721 1722 struct kvm_irq_routing { 1723 __u32 nr; 1724 __u32 flags; 1725 struct kvm_irq_routing_entry entries[0]; 1726 }; 1727 1728No flags are specified so far, the corresponding field must be set to zero. 1729 1730:: 1731 1732 struct kvm_irq_routing_entry { 1733 __u32 gsi; 1734 __u32 type; 1735 __u32 flags; 1736 __u32 pad; 1737 union { 1738 struct kvm_irq_routing_irqchip irqchip; 1739 struct kvm_irq_routing_msi msi; 1740 struct kvm_irq_routing_s390_adapter adapter; 1741 struct kvm_irq_routing_hv_sint hv_sint; 1742 __u32 pad[8]; 1743 } u; 1744 }; 1745 1746 /* gsi routing entry types */ 1747 #define KVM_IRQ_ROUTING_IRQCHIP 1 1748 #define KVM_IRQ_ROUTING_MSI 2 1749 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1750 #define KVM_IRQ_ROUTING_HV_SINT 4 1751 1752flags: 1753 1754- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1755 type, specifies that the devid field contains a valid value. The per-VM 1756 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1757 the device ID. If this capability is not available, userspace should 1758 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1759- zero otherwise 1760 1761:: 1762 1763 struct kvm_irq_routing_irqchip { 1764 __u32 irqchip; 1765 __u32 pin; 1766 }; 1767 1768 struct kvm_irq_routing_msi { 1769 __u32 address_lo; 1770 __u32 address_hi; 1771 __u32 data; 1772 union { 1773 __u32 pad; 1774 __u32 devid; 1775 }; 1776 }; 1777 1778If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1779for the device that wrote the MSI message. For PCI, this is usually a 1780BFD identifier in the lower 16 bits. 1781 1782On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1783feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1784address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1785address_hi must be zero. 1786 1787:: 1788 1789 struct kvm_irq_routing_s390_adapter { 1790 __u64 ind_addr; 1791 __u64 summary_addr; 1792 __u64 ind_offset; 1793 __u32 summary_offset; 1794 __u32 adapter_id; 1795 }; 1796 1797 struct kvm_irq_routing_hv_sint { 1798 __u32 vcpu; 1799 __u32 sint; 1800 }; 1801 1802 18034.55 KVM_SET_TSC_KHZ 1804-------------------- 1805 1806:Capability: KVM_CAP_TSC_CONTROL 1807:Architectures: x86 1808:Type: vcpu ioctl 1809:Parameters: virtual tsc_khz 1810:Returns: 0 on success, -1 on error 1811 1812Specifies the tsc frequency for the virtual machine. The unit of the 1813frequency is KHz. 1814 1815 18164.56 KVM_GET_TSC_KHZ 1817-------------------- 1818 1819:Capability: KVM_CAP_GET_TSC_KHZ 1820:Architectures: x86 1821:Type: vcpu ioctl 1822:Parameters: none 1823:Returns: virtual tsc-khz on success, negative value on error 1824 1825Returns the tsc frequency of the guest. The unit of the return value is 1826KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 1827error. 1828 1829 18304.57 KVM_GET_LAPIC 1831------------------ 1832 1833:Capability: KVM_CAP_IRQCHIP 1834:Architectures: x86 1835:Type: vcpu ioctl 1836:Parameters: struct kvm_lapic_state (out) 1837:Returns: 0 on success, -1 on error 1838 1839:: 1840 1841 #define KVM_APIC_REG_SIZE 0x400 1842 struct kvm_lapic_state { 1843 char regs[KVM_APIC_REG_SIZE]; 1844 }; 1845 1846Reads the Local APIC registers and copies them into the input argument. The 1847data format and layout are the same as documented in the architecture manual. 1848 1849If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 1850enabled, then the format of APIC_ID register depends on the APIC mode 1851(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 1852the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 1853which is stored in bits 31-24 of the APIC register, or equivalently in 1854byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 1855be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 1856 1857If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 1858always uses xAPIC format. 1859 1860 18614.58 KVM_SET_LAPIC 1862------------------ 1863 1864:Capability: KVM_CAP_IRQCHIP 1865:Architectures: x86 1866:Type: vcpu ioctl 1867:Parameters: struct kvm_lapic_state (in) 1868:Returns: 0 on success, -1 on error 1869 1870:: 1871 1872 #define KVM_APIC_REG_SIZE 0x400 1873 struct kvm_lapic_state { 1874 char regs[KVM_APIC_REG_SIZE]; 1875 }; 1876 1877Copies the input argument into the Local APIC registers. The data format 1878and layout are the same as documented in the architecture manual. 1879 1880The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 1881regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 1882See the note in KVM_GET_LAPIC. 1883 1884 18854.59 KVM_IOEVENTFD 1886------------------ 1887 1888:Capability: KVM_CAP_IOEVENTFD 1889:Architectures: all 1890:Type: vm ioctl 1891:Parameters: struct kvm_ioeventfd (in) 1892:Returns: 0 on success, !0 on error 1893 1894This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 1895within the guest. A guest write in the registered address will signal the 1896provided event instead of triggering an exit. 1897 1898:: 1899 1900 struct kvm_ioeventfd { 1901 __u64 datamatch; 1902 __u64 addr; /* legal pio/mmio address */ 1903 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 1904 __s32 fd; 1905 __u32 flags; 1906 __u8 pad[36]; 1907 }; 1908 1909For the special case of virtio-ccw devices on s390, the ioevent is matched 1910to a subchannel/virtqueue tuple instead. 1911 1912The following flags are defined:: 1913 1914 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 1915 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 1916 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 1917 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 1918 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 1919 1920If datamatch flag is set, the event will be signaled only if the written value 1921to the registered address is equal to datamatch in struct kvm_ioeventfd. 1922 1923For virtio-ccw devices, addr contains the subchannel id and datamatch the 1924virtqueue index. 1925 1926With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 1927the kernel will ignore the length of guest write and may get a faster vmexit. 1928The speedup may only apply to specific architectures, but the ioeventfd will 1929work anyway. 1930 19314.60 KVM_DIRTY_TLB 1932------------------ 1933 1934:Capability: KVM_CAP_SW_TLB 1935:Architectures: ppc 1936:Type: vcpu ioctl 1937:Parameters: struct kvm_dirty_tlb (in) 1938:Returns: 0 on success, -1 on error 1939 1940:: 1941 1942 struct kvm_dirty_tlb { 1943 __u64 bitmap; 1944 __u32 num_dirty; 1945 }; 1946 1947This must be called whenever userspace has changed an entry in the shared 1948TLB, prior to calling KVM_RUN on the associated vcpu. 1949 1950The "bitmap" field is the userspace address of an array. This array 1951consists of a number of bits, equal to the total number of TLB entries as 1952determined by the last successful call to KVM_CONFIG_TLB, rounded up to the 1953nearest multiple of 64. 1954 1955Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 1956array. 1957 1958The array is little-endian: the bit 0 is the least significant bit of the 1959first byte, bit 8 is the least significant bit of the second byte, etc. 1960This avoids any complications with differing word sizes. 1961 1962The "num_dirty" field is a performance hint for KVM to determine whether it 1963should skip processing the bitmap and just invalidate everything. It must 1964be set to the number of set bits in the bitmap. 1965 1966 19674.62 KVM_CREATE_SPAPR_TCE 1968------------------------- 1969 1970:Capability: KVM_CAP_SPAPR_TCE 1971:Architectures: powerpc 1972:Type: vm ioctl 1973:Parameters: struct kvm_create_spapr_tce (in) 1974:Returns: file descriptor for manipulating the created TCE table 1975 1976This creates a virtual TCE (translation control entry) table, which 1977is an IOMMU for PAPR-style virtual I/O. It is used to translate 1978logical addresses used in virtual I/O into guest physical addresses, 1979and provides a scatter/gather capability for PAPR virtual I/O. 1980 1981:: 1982 1983 /* for KVM_CAP_SPAPR_TCE */ 1984 struct kvm_create_spapr_tce { 1985 __u64 liobn; 1986 __u32 window_size; 1987 }; 1988 1989The liobn field gives the logical IO bus number for which to create a 1990TCE table. The window_size field specifies the size of the DMA window 1991which this TCE table will translate - the table will contain one 64 1992bit TCE entry for every 4kiB of the DMA window. 1993 1994When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 1995table has been created using this ioctl(), the kernel will handle it 1996in real mode, updating the TCE table. H_PUT_TCE calls for other 1997liobns will cause a vm exit and must be handled by userspace. 1998 1999The return value is a file descriptor which can be passed to mmap(2) 2000to map the created TCE table into userspace. This lets userspace read 2001the entries written by kernel-handled H_PUT_TCE calls, and also lets 2002userspace update the TCE table directly which is useful in some 2003circumstances. 2004 2005 20064.63 KVM_ALLOCATE_RMA 2007--------------------- 2008 2009:Capability: KVM_CAP_PPC_RMA 2010:Architectures: powerpc 2011:Type: vm ioctl 2012:Parameters: struct kvm_allocate_rma (out) 2013:Returns: file descriptor for mapping the allocated RMA 2014 2015This allocates a Real Mode Area (RMA) from the pool allocated at boot 2016time by the kernel. An RMA is a physically-contiguous, aligned region 2017of memory used on older POWER processors to provide the memory which 2018will be accessed by real-mode (MMU off) accesses in a KVM guest. 2019POWER processors support a set of sizes for the RMA that usually 2020includes 64MB, 128MB, 256MB and some larger powers of two. 2021 2022:: 2023 2024 /* for KVM_ALLOCATE_RMA */ 2025 struct kvm_allocate_rma { 2026 __u64 rma_size; 2027 }; 2028 2029The return value is a file descriptor which can be passed to mmap(2) 2030to map the allocated RMA into userspace. The mapped area can then be 2031passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the 2032RMA for a virtual machine. The size of the RMA in bytes (which is 2033fixed at host kernel boot time) is returned in the rma_size field of 2034the argument structure. 2035 2036The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl 2037is supported; 2 if the processor requires all virtual machines to have 2038an RMA, or 1 if the processor can use an RMA but doesn't require it, 2039because it supports the Virtual RMA (VRMA) facility. 2040 2041 20424.64 KVM_NMI 2043------------ 2044 2045:Capability: KVM_CAP_USER_NMI 2046:Architectures: x86 2047:Type: vcpu ioctl 2048:Parameters: none 2049:Returns: 0 on success, -1 on error 2050 2051Queues an NMI on the thread's vcpu. Note this is well defined only 2052when KVM_CREATE_IRQCHIP has not been called, since this is an interface 2053between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 2054has been called, this interface is completely emulated within the kernel. 2055 2056To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 2057following algorithm: 2058 2059 - pause the vcpu 2060 - read the local APIC's state (KVM_GET_LAPIC) 2061 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 2062 - if so, issue KVM_NMI 2063 - resume the vcpu 2064 2065Some guests configure the LINT1 NMI input to cause a panic, aiding in 2066debugging. 2067 2068 20694.65 KVM_S390_UCAS_MAP 2070---------------------- 2071 2072:Capability: KVM_CAP_S390_UCONTROL 2073:Architectures: s390 2074:Type: vcpu ioctl 2075:Parameters: struct kvm_s390_ucas_mapping (in) 2076:Returns: 0 in case of success 2077 2078The parameter is defined like this:: 2079 2080 struct kvm_s390_ucas_mapping { 2081 __u64 user_addr; 2082 __u64 vcpu_addr; 2083 __u64 length; 2084 }; 2085 2086This ioctl maps the memory at "user_addr" with the length "length" to 2087the vcpu's address space starting at "vcpu_addr". All parameters need to 2088be aligned by 1 megabyte. 2089 2090 20914.66 KVM_S390_UCAS_UNMAP 2092------------------------ 2093 2094:Capability: KVM_CAP_S390_UCONTROL 2095:Architectures: s390 2096:Type: vcpu ioctl 2097:Parameters: struct kvm_s390_ucas_mapping (in) 2098:Returns: 0 in case of success 2099 2100The parameter is defined like this:: 2101 2102 struct kvm_s390_ucas_mapping { 2103 __u64 user_addr; 2104 __u64 vcpu_addr; 2105 __u64 length; 2106 }; 2107 2108This ioctl unmaps the memory in the vcpu's address space starting at 2109"vcpu_addr" with the length "length". The field "user_addr" is ignored. 2110All parameters need to be aligned by 1 megabyte. 2111 2112 21134.67 KVM_S390_VCPU_FAULT 2114------------------------ 2115 2116:Capability: KVM_CAP_S390_UCONTROL 2117:Architectures: s390 2118:Type: vcpu ioctl 2119:Parameters: vcpu absolute address (in) 2120:Returns: 0 in case of success 2121 2122This call creates a page table entry on the virtual cpu's address space 2123(for user controlled virtual machines) or the virtual machine's address 2124space (for regular virtual machines). This only works for minor faults, 2125thus it's recommended to access subject memory page via the user page 2126table upfront. This is useful to handle validity intercepts for user 2127controlled virtual machines to fault in the virtual cpu's lowcore pages 2128prior to calling the KVM_RUN ioctl. 2129 2130 21314.68 KVM_SET_ONE_REG 2132-------------------- 2133 2134:Capability: KVM_CAP_ONE_REG 2135:Architectures: all 2136:Type: vcpu ioctl 2137:Parameters: struct kvm_one_reg (in) 2138:Returns: 0 on success, negative value on failure 2139 2140Errors: 2141 2142 ====== ============================================================ 2143 ENOENT no such register 2144 EINVAL invalid register ID, or no such register or used with VMs in 2145 protected virtualization mode on s390 2146 EPERM (arm64) register access not allowed before vcpu finalization 2147 ====== ============================================================ 2148 2149(These error codes are indicative only: do not rely on a specific error 2150code being returned in a specific situation.) 2151 2152:: 2153 2154 struct kvm_one_reg { 2155 __u64 id; 2156 __u64 addr; 2157 }; 2158 2159Using this ioctl, a single vcpu register can be set to a specific value 2160defined by user space with the passed in struct kvm_one_reg, where id 2161refers to the register identifier as described below and addr is a pointer 2162to a variable with the respective size. There can be architecture agnostic 2163and architecture specific registers. Each have their own range of operation 2164and their own constants and width. To keep track of the implemented 2165registers, find a list below: 2166 2167 ======= =============================== ============ 2168 Arch Register Width (bits) 2169 ======= =============================== ============ 2170 PPC KVM_REG_PPC_HIOR 64 2171 PPC KVM_REG_PPC_IAC1 64 2172 PPC KVM_REG_PPC_IAC2 64 2173 PPC KVM_REG_PPC_IAC3 64 2174 PPC KVM_REG_PPC_IAC4 64 2175 PPC KVM_REG_PPC_DAC1 64 2176 PPC KVM_REG_PPC_DAC2 64 2177 PPC KVM_REG_PPC_DABR 64 2178 PPC KVM_REG_PPC_DSCR 64 2179 PPC KVM_REG_PPC_PURR 64 2180 PPC KVM_REG_PPC_SPURR 64 2181 PPC KVM_REG_PPC_DAR 64 2182 PPC KVM_REG_PPC_DSISR 32 2183 PPC KVM_REG_PPC_AMR 64 2184 PPC KVM_REG_PPC_UAMOR 64 2185 PPC KVM_REG_PPC_MMCR0 64 2186 PPC KVM_REG_PPC_MMCR1 64 2187 PPC KVM_REG_PPC_MMCRA 64 2188 PPC KVM_REG_PPC_MMCR2 64 2189 PPC KVM_REG_PPC_MMCRS 64 2190 PPC KVM_REG_PPC_MMCR3 64 2191 PPC KVM_REG_PPC_SIAR 64 2192 PPC KVM_REG_PPC_SDAR 64 2193 PPC KVM_REG_PPC_SIER 64 2194 PPC KVM_REG_PPC_SIER2 64 2195 PPC KVM_REG_PPC_SIER3 64 2196 PPC KVM_REG_PPC_PMC1 32 2197 PPC KVM_REG_PPC_PMC2 32 2198 PPC KVM_REG_PPC_PMC3 32 2199 PPC KVM_REG_PPC_PMC4 32 2200 PPC KVM_REG_PPC_PMC5 32 2201 PPC KVM_REG_PPC_PMC6 32 2202 PPC KVM_REG_PPC_PMC7 32 2203 PPC KVM_REG_PPC_PMC8 32 2204 PPC KVM_REG_PPC_FPR0 64 2205 ... 2206 PPC KVM_REG_PPC_FPR31 64 2207 PPC KVM_REG_PPC_VR0 128 2208 ... 2209 PPC KVM_REG_PPC_VR31 128 2210 PPC KVM_REG_PPC_VSR0 128 2211 ... 2212 PPC KVM_REG_PPC_VSR31 128 2213 PPC KVM_REG_PPC_FPSCR 64 2214 PPC KVM_REG_PPC_VSCR 32 2215 PPC KVM_REG_PPC_VPA_ADDR 64 2216 PPC KVM_REG_PPC_VPA_SLB 128 2217 PPC KVM_REG_PPC_VPA_DTL 128 2218 PPC KVM_REG_PPC_EPCR 32 2219 PPC KVM_REG_PPC_EPR 32 2220 PPC KVM_REG_PPC_TCR 32 2221 PPC KVM_REG_PPC_TSR 32 2222 PPC KVM_REG_PPC_OR_TSR 32 2223 PPC KVM_REG_PPC_CLEAR_TSR 32 2224 PPC KVM_REG_PPC_MAS0 32 2225 PPC KVM_REG_PPC_MAS1 32 2226 PPC KVM_REG_PPC_MAS2 64 2227 PPC KVM_REG_PPC_MAS7_3 64 2228 PPC KVM_REG_PPC_MAS4 32 2229 PPC KVM_REG_PPC_MAS6 32 2230 PPC KVM_REG_PPC_MMUCFG 32 2231 PPC KVM_REG_PPC_TLB0CFG 32 2232 PPC KVM_REG_PPC_TLB1CFG 32 2233 PPC KVM_REG_PPC_TLB2CFG 32 2234 PPC KVM_REG_PPC_TLB3CFG 32 2235 PPC KVM_REG_PPC_TLB0PS 32 2236 PPC KVM_REG_PPC_TLB1PS 32 2237 PPC KVM_REG_PPC_TLB2PS 32 2238 PPC KVM_REG_PPC_TLB3PS 32 2239 PPC KVM_REG_PPC_EPTCFG 32 2240 PPC KVM_REG_PPC_ICP_STATE 64 2241 PPC KVM_REG_PPC_VP_STATE 128 2242 PPC KVM_REG_PPC_TB_OFFSET 64 2243 PPC KVM_REG_PPC_SPMC1 32 2244 PPC KVM_REG_PPC_SPMC2 32 2245 PPC KVM_REG_PPC_IAMR 64 2246 PPC KVM_REG_PPC_TFHAR 64 2247 PPC KVM_REG_PPC_TFIAR 64 2248 PPC KVM_REG_PPC_TEXASR 64 2249 PPC KVM_REG_PPC_FSCR 64 2250 PPC KVM_REG_PPC_PSPB 32 2251 PPC KVM_REG_PPC_EBBHR 64 2252 PPC KVM_REG_PPC_EBBRR 64 2253 PPC KVM_REG_PPC_BESCR 64 2254 PPC KVM_REG_PPC_TAR 64 2255 PPC KVM_REG_PPC_DPDES 64 2256 PPC KVM_REG_PPC_DAWR 64 2257 PPC KVM_REG_PPC_DAWRX 64 2258 PPC KVM_REG_PPC_CIABR 64 2259 PPC KVM_REG_PPC_IC 64 2260 PPC KVM_REG_PPC_VTB 64 2261 PPC KVM_REG_PPC_CSIGR 64 2262 PPC KVM_REG_PPC_TACR 64 2263 PPC KVM_REG_PPC_TCSCR 64 2264 PPC KVM_REG_PPC_PID 64 2265 PPC KVM_REG_PPC_ACOP 64 2266 PPC KVM_REG_PPC_VRSAVE 32 2267 PPC KVM_REG_PPC_LPCR 32 2268 PPC KVM_REG_PPC_LPCR_64 64 2269 PPC KVM_REG_PPC_PPR 64 2270 PPC KVM_REG_PPC_ARCH_COMPAT 32 2271 PPC KVM_REG_PPC_DABRX 32 2272 PPC KVM_REG_PPC_WORT 64 2273 PPC KVM_REG_PPC_SPRG9 64 2274 PPC KVM_REG_PPC_DBSR 32 2275 PPC KVM_REG_PPC_TIDR 64 2276 PPC KVM_REG_PPC_PSSCR 64 2277 PPC KVM_REG_PPC_DEC_EXPIRY 64 2278 PPC KVM_REG_PPC_PTCR 64 2279 PPC KVM_REG_PPC_DAWR1 64 2280 PPC KVM_REG_PPC_DAWRX1 64 2281 PPC KVM_REG_PPC_TM_GPR0 64 2282 ... 2283 PPC KVM_REG_PPC_TM_GPR31 64 2284 PPC KVM_REG_PPC_TM_VSR0 128 2285 ... 2286 PPC KVM_REG_PPC_TM_VSR63 128 2287 PPC KVM_REG_PPC_TM_CR 64 2288 PPC KVM_REG_PPC_TM_LR 64 2289 PPC KVM_REG_PPC_TM_CTR 64 2290 PPC KVM_REG_PPC_TM_FPSCR 64 2291 PPC KVM_REG_PPC_TM_AMR 64 2292 PPC KVM_REG_PPC_TM_PPR 64 2293 PPC KVM_REG_PPC_TM_VRSAVE 64 2294 PPC KVM_REG_PPC_TM_VSCR 32 2295 PPC KVM_REG_PPC_TM_DSCR 64 2296 PPC KVM_REG_PPC_TM_TAR 64 2297 PPC KVM_REG_PPC_TM_XER 64 2298 2299 MIPS KVM_REG_MIPS_R0 64 2300 ... 2301 MIPS KVM_REG_MIPS_R31 64 2302 MIPS KVM_REG_MIPS_HI 64 2303 MIPS KVM_REG_MIPS_LO 64 2304 MIPS KVM_REG_MIPS_PC 64 2305 MIPS KVM_REG_MIPS_CP0_INDEX 32 2306 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2307 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2308 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2309 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2310 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2311 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2312 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2313 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2314 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2315 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2316 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2317 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2318 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2319 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2320 MIPS KVM_REG_MIPS_CP0_WIRED 32 2321 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2322 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2323 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2324 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2325 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2326 MIPS KVM_REG_MIPS_CP0_COUNT 32 2327 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2328 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2329 MIPS KVM_REG_MIPS_CP0_STATUS 32 2330 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2331 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2332 MIPS KVM_REG_MIPS_CP0_EPC 64 2333 MIPS KVM_REG_MIPS_CP0_PRID 32 2334 MIPS KVM_REG_MIPS_CP0_EBASE 64 2335 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2336 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2337 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2338 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2339 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2340 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2341 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2342 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2343 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2344 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2345 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2346 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2347 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2348 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2349 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2350 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2351 MIPS KVM_REG_MIPS_COUNT_CTL 64 2352 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2353 MIPS KVM_REG_MIPS_COUNT_HZ 64 2354 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2355 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2356 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2357 MIPS KVM_REG_MIPS_FCR_IR 32 2358 MIPS KVM_REG_MIPS_FCR_CSR 32 2359 MIPS KVM_REG_MIPS_MSA_IR 32 2360 MIPS KVM_REG_MIPS_MSA_CSR 32 2361 ======= =============================== ============ 2362 2363ARM registers are mapped using the lower 32 bits. The upper 16 of that 2364is the register group type, or coprocessor number: 2365 2366ARM core registers have the following id bit patterns:: 2367 2368 0x4020 0000 0010 <index into the kvm_regs struct:16> 2369 2370ARM 32-bit CP15 registers have the following id bit patterns:: 2371 2372 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 2373 2374ARM 64-bit CP15 registers have the following id bit patterns:: 2375 2376 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 2377 2378ARM CCSIDR registers are demultiplexed by CSSELR value:: 2379 2380 0x4020 0000 0011 00 <csselr:8> 2381 2382ARM 32-bit VFP control registers have the following id bit patterns:: 2383 2384 0x4020 0000 0012 1 <regno:12> 2385 2386ARM 64-bit FP registers have the following id bit patterns:: 2387 2388 0x4030 0000 0012 0 <regno:12> 2389 2390ARM firmware pseudo-registers have the following bit pattern:: 2391 2392 0x4030 0000 0014 <regno:16> 2393 2394 2395arm64 registers are mapped using the lower 32 bits. The upper 16 of 2396that is the register group type, or coprocessor number: 2397 2398arm64 core/FP-SIMD registers have the following id bit patterns. Note 2399that the size of the access is variable, as the kvm_regs structure 2400contains elements ranging from 32 to 128 bits. The index is a 32bit 2401value in the kvm_regs structure seen as a 32bit array:: 2402 2403 0x60x0 0000 0010 <index into the kvm_regs struct:16> 2404 2405Specifically: 2406 2407======================= ========= ===== ======================================= 2408 Encoding Register Bits kvm_regs member 2409======================= ========= ===== ======================================= 2410 0x6030 0000 0010 0000 X0 64 regs.regs[0] 2411 0x6030 0000 0010 0002 X1 64 regs.regs[1] 2412 ... 2413 0x6030 0000 0010 003c X30 64 regs.regs[30] 2414 0x6030 0000 0010 003e SP 64 regs.sp 2415 0x6030 0000 0010 0040 PC 64 regs.pc 2416 0x6030 0000 0010 0042 PSTATE 64 regs.pstate 2417 0x6030 0000 0010 0044 SP_EL1 64 sp_el1 2418 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1 2419 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC) 2420 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT] 2421 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND] 2422 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ] 2423 0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ] 2424 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_ 2425 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_ 2426 ... 2427 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_ 2428 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr 2429 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr 2430======================= ========= ===== ======================================= 2431 2432.. [1] These encodings are not accepted for SVE-enabled vcpus. See 2433 KVM_ARM_VCPU_INIT. 2434 2435 The equivalent register content can be accessed via bits [127:0] of 2436 the corresponding SVE Zn registers instead for vcpus that have SVE 2437 enabled (see below). 2438 2439arm64 CCSIDR registers are demultiplexed by CSSELR value:: 2440 2441 0x6020 0000 0011 00 <csselr:8> 2442 2443arm64 system registers have the following id bit patterns:: 2444 2445 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 2446 2447.. warning:: 2448 2449 Two system register IDs do not follow the specified pattern. These 2450 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to 2451 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These 2452 two had their values accidentally swapped, which means TIMER_CVAL is 2453 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is 2454 derived from the register encoding for CNTV_CVAL_EL0. As this is 2455 API, it must remain this way. 2456 2457arm64 firmware pseudo-registers have the following bit pattern:: 2458 2459 0x6030 0000 0014 <regno:16> 2460 2461arm64 SVE registers have the following bit patterns:: 2462 2463 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice] 2464 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice] 2465 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice] 2466 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register 2467 2468Access to register IDs where 2048 * slice >= 128 * max_vq will fail with 2469ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit 2470quadwords: see [2]_ below. 2471 2472These registers are only accessible on vcpus for which SVE is enabled. 2473See KVM_ARM_VCPU_INIT for details. 2474 2475In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not 2476accessible until the vcpu's SVE configuration has been finalized 2477using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT 2478and KVM_ARM_VCPU_FINALIZE for more information about this procedure. 2479 2480KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector 2481lengths supported by the vcpu to be discovered and configured by 2482userspace. When transferred to or from user memory via KVM_GET_ONE_REG 2483or KVM_SET_ONE_REG, the value of this register is of type 2484__u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as 2485follows:: 2486 2487 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS]; 2488 2489 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2490 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >> 2491 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1)) 2492 /* Vector length vq * 16 bytes supported */ 2493 else 2494 /* Vector length vq * 16 bytes not supported */ 2495 2496.. [2] The maximum value vq for which the above condition is true is 2497 max_vq. This is the maximum vector length available to the guest on 2498 this vcpu, and determines which register slices are visible through 2499 this ioctl interface. 2500 2501(See Documentation/arm64/sve.rst for an explanation of the "vq" 2502nomenclature.) 2503 2504KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT. 2505KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that 2506the host supports. 2507 2508Userspace may subsequently modify it if desired until the vcpu's SVE 2509configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). 2510 2511Apart from simply removing all vector lengths from the host set that 2512exceed some value, support for arbitrarily chosen sets of vector lengths 2513is hardware-dependent and may not be available. Attempting to configure 2514an invalid set of vector lengths via KVM_SET_ONE_REG will fail with 2515EINVAL. 2516 2517After the vcpu's SVE configuration is finalized, further attempts to 2518write this register will fail with EPERM. 2519 2520 2521MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 2522the register group type: 2523 2524MIPS core registers (see above) have the following id bit patterns:: 2525 2526 0x7030 0000 0000 <reg:16> 2527 2528MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 2529patterns depending on whether they're 32-bit or 64-bit registers:: 2530 2531 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 2532 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2533 2534Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 2535versions of the EntryLo registers regardless of the word size of the host 2536hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 2537with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 2538the PFNX field starting at bit 30. 2539 2540MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 2541patterns:: 2542 2543 0x7030 0000 0001 01 <reg:8> 2544 2545MIPS KVM control registers (see above) have the following id bit patterns:: 2546 2547 0x7030 0000 0002 <reg:16> 2548 2549MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2550id bit patterns depending on the size of the register being accessed. They are 2551always accessed according to the current guest FPU mode (Status.FR and 2552Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2553if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2554registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2555overlap the FPU registers:: 2556 2557 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2558 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2559 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2560 2561MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2562following id bit patterns:: 2563 2564 0x7020 0000 0003 01 <0:3> <reg:5> 2565 2566MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2567following id bit patterns:: 2568 2569 0x7020 0000 0003 02 <0:3> <reg:5> 2570 2571 25724.69 KVM_GET_ONE_REG 2573-------------------- 2574 2575:Capability: KVM_CAP_ONE_REG 2576:Architectures: all 2577:Type: vcpu ioctl 2578:Parameters: struct kvm_one_reg (in and out) 2579:Returns: 0 on success, negative value on failure 2580 2581Errors include: 2582 2583 ======== ============================================================ 2584 ENOENT no such register 2585 EINVAL invalid register ID, or no such register or used with VMs in 2586 protected virtualization mode on s390 2587 EPERM (arm64) register access not allowed before vcpu finalization 2588 ======== ============================================================ 2589 2590(These error codes are indicative only: do not rely on a specific error 2591code being returned in a specific situation.) 2592 2593This ioctl allows to receive the value of a single register implemented 2594in a vcpu. The register to read is indicated by the "id" field of the 2595kvm_one_reg struct passed in. On success, the register value can be found 2596at the memory location pointed to by "addr". 2597 2598The list of registers accessible using this interface is identical to the 2599list in 4.68. 2600 2601 26024.70 KVM_KVMCLOCK_CTRL 2603---------------------- 2604 2605:Capability: KVM_CAP_KVMCLOCK_CTRL 2606:Architectures: Any that implement pvclocks (currently x86 only) 2607:Type: vcpu ioctl 2608:Parameters: None 2609:Returns: 0 on success, -1 on error 2610 2611This ioctl sets a flag accessible to the guest indicating that the specified 2612vCPU has been paused by the host userspace. 2613 2614The host will set a flag in the pvclock structure that is checked from the 2615soft lockup watchdog. The flag is part of the pvclock structure that is 2616shared between guest and host, specifically the second bit of the flags 2617field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2618the host and read/cleared exclusively by the guest. The guest operation of 2619checking and clearing the flag must be an atomic operation so 2620load-link/store-conditional, or equivalent must be used. There are two cases 2621where the guest will clear the flag: when the soft lockup watchdog timer resets 2622itself or when a soft lockup is detected. This ioctl can be called any time 2623after pausing the vcpu, but before it is resumed. 2624 2625 26264.71 KVM_SIGNAL_MSI 2627------------------- 2628 2629:Capability: KVM_CAP_SIGNAL_MSI 2630:Architectures: x86 arm arm64 2631:Type: vm ioctl 2632:Parameters: struct kvm_msi (in) 2633:Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2634 2635Directly inject a MSI message. Only valid with in-kernel irqchip that handles 2636MSI messages. 2637 2638:: 2639 2640 struct kvm_msi { 2641 __u32 address_lo; 2642 __u32 address_hi; 2643 __u32 data; 2644 __u32 flags; 2645 __u32 devid; 2646 __u8 pad[12]; 2647 }; 2648 2649flags: 2650 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 2651 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 2652 the device ID. If this capability is not available, userspace 2653 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 2654 2655If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 2656for the device that wrote the MSI message. For PCI, this is usually a 2657BFD identifier in the lower 16 bits. 2658 2659On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 2660feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 2661address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 2662address_hi must be zero. 2663 2664 26654.71 KVM_CREATE_PIT2 2666-------------------- 2667 2668:Capability: KVM_CAP_PIT2 2669:Architectures: x86 2670:Type: vm ioctl 2671:Parameters: struct kvm_pit_config (in) 2672:Returns: 0 on success, -1 on error 2673 2674Creates an in-kernel device model for the i8254 PIT. This call is only valid 2675after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 2676parameters have to be passed:: 2677 2678 struct kvm_pit_config { 2679 __u32 flags; 2680 __u32 pad[15]; 2681 }; 2682 2683Valid flags are:: 2684 2685 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 2686 2687PIT timer interrupts may use a per-VM kernel thread for injection. If it 2688exists, this thread will have a name of the following pattern:: 2689 2690 kvm-pit/<owner-process-pid> 2691 2692When running a guest with elevated priorities, the scheduling parameters of 2693this thread may have to be adjusted accordingly. 2694 2695This IOCTL replaces the obsolete KVM_CREATE_PIT. 2696 2697 26984.72 KVM_GET_PIT2 2699----------------- 2700 2701:Capability: KVM_CAP_PIT_STATE2 2702:Architectures: x86 2703:Type: vm ioctl 2704:Parameters: struct kvm_pit_state2 (out) 2705:Returns: 0 on success, -1 on error 2706 2707Retrieves the state of the in-kernel PIT model. Only valid after 2708KVM_CREATE_PIT2. The state is returned in the following structure:: 2709 2710 struct kvm_pit_state2 { 2711 struct kvm_pit_channel_state channels[3]; 2712 __u32 flags; 2713 __u32 reserved[9]; 2714 }; 2715 2716Valid flags are:: 2717 2718 /* disable PIT in HPET legacy mode */ 2719 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 2720 2721This IOCTL replaces the obsolete KVM_GET_PIT. 2722 2723 27244.73 KVM_SET_PIT2 2725----------------- 2726 2727:Capability: KVM_CAP_PIT_STATE2 2728:Architectures: x86 2729:Type: vm ioctl 2730:Parameters: struct kvm_pit_state2 (in) 2731:Returns: 0 on success, -1 on error 2732 2733Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 2734See KVM_GET_PIT2 for details on struct kvm_pit_state2. 2735 2736This IOCTL replaces the obsolete KVM_SET_PIT. 2737 2738 27394.74 KVM_PPC_GET_SMMU_INFO 2740-------------------------- 2741 2742:Capability: KVM_CAP_PPC_GET_SMMU_INFO 2743:Architectures: powerpc 2744:Type: vm ioctl 2745:Parameters: None 2746:Returns: 0 on success, -1 on error 2747 2748This populates and returns a structure describing the features of 2749the "Server" class MMU emulation supported by KVM. 2750This can in turn be used by userspace to generate the appropriate 2751device-tree properties for the guest operating system. 2752 2753The structure contains some global information, followed by an 2754array of supported segment page sizes:: 2755 2756 struct kvm_ppc_smmu_info { 2757 __u64 flags; 2758 __u32 slb_size; 2759 __u32 pad; 2760 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 2761 }; 2762 2763The supported flags are: 2764 2765 - KVM_PPC_PAGE_SIZES_REAL: 2766 When that flag is set, guest page sizes must "fit" the backing 2767 store page sizes. When not set, any page size in the list can 2768 be used regardless of how they are backed by userspace. 2769 2770 - KVM_PPC_1T_SEGMENTS 2771 The emulated MMU supports 1T segments in addition to the 2772 standard 256M ones. 2773 2774 - KVM_PPC_NO_HASH 2775 This flag indicates that HPT guests are not supported by KVM, 2776 thus all guests must use radix MMU mode. 2777 2778The "slb_size" field indicates how many SLB entries are supported 2779 2780The "sps" array contains 8 entries indicating the supported base 2781page sizes for a segment in increasing order. Each entry is defined 2782as follow:: 2783 2784 struct kvm_ppc_one_seg_page_size { 2785 __u32 page_shift; /* Base page shift of segment (or 0) */ 2786 __u32 slb_enc; /* SLB encoding for BookS */ 2787 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 2788 }; 2789 2790An entry with a "page_shift" of 0 is unused. Because the array is 2791organized in increasing order, a lookup can stop when encoutering 2792such an entry. 2793 2794The "slb_enc" field provides the encoding to use in the SLB for the 2795page size. The bits are in positions such as the value can directly 2796be OR'ed into the "vsid" argument of the slbmte instruction. 2797 2798The "enc" array is a list which for each of those segment base page 2799size provides the list of supported actual page sizes (which can be 2800only larger or equal to the base page size), along with the 2801corresponding encoding in the hash PTE. Similarly, the array is 28028 entries sorted by increasing sizes and an entry with a "0" shift 2803is an empty entry and a terminator:: 2804 2805 struct kvm_ppc_one_page_size { 2806 __u32 page_shift; /* Page shift (or 0) */ 2807 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 2808 }; 2809 2810The "pte_enc" field provides a value that can OR'ed into the hash 2811PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 2812into the hash PTE second double word). 2813 28144.75 KVM_IRQFD 2815-------------- 2816 2817:Capability: KVM_CAP_IRQFD 2818:Architectures: x86 s390 arm arm64 2819:Type: vm ioctl 2820:Parameters: struct kvm_irqfd (in) 2821:Returns: 0 on success, -1 on error 2822 2823Allows setting an eventfd to directly trigger a guest interrupt. 2824kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 2825kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 2826an event is triggered on the eventfd, an interrupt is injected into 2827the guest using the specified gsi pin. The irqfd is removed using 2828the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 2829and kvm_irqfd.gsi. 2830 2831With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 2832mechanism allowing emulation of level-triggered, irqfd-based 2833interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 2834additional eventfd in the kvm_irqfd.resamplefd field. When operating 2835in resample mode, posting of an interrupt through kvm_irq.fd asserts 2836the specified gsi in the irqchip. When the irqchip is resampled, such 2837as from an EOI, the gsi is de-asserted and the user is notified via 2838kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 2839the interrupt if the device making use of it still requires service. 2840Note that closing the resamplefd is not sufficient to disable the 2841irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 2842and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 2843 2844On arm/arm64, gsi routing being supported, the following can happen: 2845 2846- in case no routing entry is associated to this gsi, injection fails 2847- in case the gsi is associated to an irqchip routing entry, 2848 irqchip.pin + 32 corresponds to the injected SPI ID. 2849- in case the gsi is associated to an MSI routing entry, the MSI 2850 message and device ID are translated into an LPI (support restricted 2851 to GICv3 ITS in-kernel emulation). 2852 28534.76 KVM_PPC_ALLOCATE_HTAB 2854-------------------------- 2855 2856:Capability: KVM_CAP_PPC_ALLOC_HTAB 2857:Architectures: powerpc 2858:Type: vm ioctl 2859:Parameters: Pointer to u32 containing hash table order (in/out) 2860:Returns: 0 on success, -1 on error 2861 2862This requests the host kernel to allocate an MMU hash table for a 2863guest using the PAPR paravirtualization interface. This only does 2864anything if the kernel is configured to use the Book 3S HV style of 2865virtualization. Otherwise the capability doesn't exist and the ioctl 2866returns an ENOTTY error. The rest of this description assumes Book 3S 2867HV. 2868 2869There must be no vcpus running when this ioctl is called; if there 2870are, it will do nothing and return an EBUSY error. 2871 2872The parameter is a pointer to a 32-bit unsigned integer variable 2873containing the order (log base 2) of the desired size of the hash 2874table, which must be between 18 and 46. On successful return from the 2875ioctl, the value will not be changed by the kernel. 2876 2877If no hash table has been allocated when any vcpu is asked to run 2878(with the KVM_RUN ioctl), the host kernel will allocate a 2879default-sized hash table (16 MB). 2880 2881If this ioctl is called when a hash table has already been allocated, 2882with a different order from the existing hash table, the existing hash 2883table will be freed and a new one allocated. If this is ioctl is 2884called when a hash table has already been allocated of the same order 2885as specified, the kernel will clear out the existing hash table (zero 2886all HPTEs). In either case, if the guest is using the virtualized 2887real-mode area (VRMA) facility, the kernel will re-create the VMRA 2888HPTEs on the next KVM_RUN of any vcpu. 2889 28904.77 KVM_S390_INTERRUPT 2891----------------------- 2892 2893:Capability: basic 2894:Architectures: s390 2895:Type: vm ioctl, vcpu ioctl 2896:Parameters: struct kvm_s390_interrupt (in) 2897:Returns: 0 on success, -1 on error 2898 2899Allows to inject an interrupt to the guest. Interrupts can be floating 2900(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 2901 2902Interrupt parameters are passed via kvm_s390_interrupt:: 2903 2904 struct kvm_s390_interrupt { 2905 __u32 type; 2906 __u32 parm; 2907 __u64 parm64; 2908 }; 2909 2910type can be one of the following: 2911 2912KVM_S390_SIGP_STOP (vcpu) 2913 - sigp stop; optional flags in parm 2914KVM_S390_PROGRAM_INT (vcpu) 2915 - program check; code in parm 2916KVM_S390_SIGP_SET_PREFIX (vcpu) 2917 - sigp set prefix; prefix address in parm 2918KVM_S390_RESTART (vcpu) 2919 - restart 2920KVM_S390_INT_CLOCK_COMP (vcpu) 2921 - clock comparator interrupt 2922KVM_S390_INT_CPU_TIMER (vcpu) 2923 - CPU timer interrupt 2924KVM_S390_INT_VIRTIO (vm) 2925 - virtio external interrupt; external interrupt 2926 parameters in parm and parm64 2927KVM_S390_INT_SERVICE (vm) 2928 - sclp external interrupt; sclp parameter in parm 2929KVM_S390_INT_EMERGENCY (vcpu) 2930 - sigp emergency; source cpu in parm 2931KVM_S390_INT_EXTERNAL_CALL (vcpu) 2932 - sigp external call; source cpu in parm 2933KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 2934 - compound value to indicate an 2935 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 2936 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 2937 interruption subclass) 2938KVM_S390_MCHK (vm, vcpu) 2939 - machine check interrupt; cr 14 bits in parm, machine check interrupt 2940 code in parm64 (note that machine checks needing further payload are not 2941 supported by this ioctl) 2942 2943This is an asynchronous vcpu ioctl and can be invoked from any thread. 2944 29454.78 KVM_PPC_GET_HTAB_FD 2946------------------------ 2947 2948:Capability: KVM_CAP_PPC_HTAB_FD 2949:Architectures: powerpc 2950:Type: vm ioctl 2951:Parameters: Pointer to struct kvm_get_htab_fd (in) 2952:Returns: file descriptor number (>= 0) on success, -1 on error 2953 2954This returns a file descriptor that can be used either to read out the 2955entries in the guest's hashed page table (HPT), or to write entries to 2956initialize the HPT. The returned fd can only be written to if the 2957KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 2958can only be read if that bit is clear. The argument struct looks like 2959this:: 2960 2961 /* For KVM_PPC_GET_HTAB_FD */ 2962 struct kvm_get_htab_fd { 2963 __u64 flags; 2964 __u64 start_index; 2965 __u64 reserved[2]; 2966 }; 2967 2968 /* Values for kvm_get_htab_fd.flags */ 2969 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 2970 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 2971 2972The 'start_index' field gives the index in the HPT of the entry at 2973which to start reading. It is ignored when writing. 2974 2975Reads on the fd will initially supply information about all 2976"interesting" HPT entries. Interesting entries are those with the 2977bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 2978all entries. When the end of the HPT is reached, the read() will 2979return. If read() is called again on the fd, it will start again from 2980the beginning of the HPT, but will only return HPT entries that have 2981changed since they were last read. 2982 2983Data read or written is structured as a header (8 bytes) followed by a 2984series of valid HPT entries (16 bytes) each. The header indicates how 2985many valid HPT entries there are and how many invalid entries follow 2986the valid entries. The invalid entries are not represented explicitly 2987in the stream. The header format is:: 2988 2989 struct kvm_get_htab_header { 2990 __u32 index; 2991 __u16 n_valid; 2992 __u16 n_invalid; 2993 }; 2994 2995Writes to the fd create HPT entries starting at the index given in the 2996header; first 'n_valid' valid entries with contents from the data 2997written, then 'n_invalid' invalid entries, invalidating any previously 2998valid entries found. 2999 30004.79 KVM_CREATE_DEVICE 3001---------------------- 3002 3003:Capability: KVM_CAP_DEVICE_CTRL 3004:Type: vm ioctl 3005:Parameters: struct kvm_create_device (in/out) 3006:Returns: 0 on success, -1 on error 3007 3008Errors: 3009 3010 ====== ======================================================= 3011 ENODEV The device type is unknown or unsupported 3012 EEXIST Device already created, and this type of device may not 3013 be instantiated multiple times 3014 ====== ======================================================= 3015 3016 Other error conditions may be defined by individual device types or 3017 have their standard meanings. 3018 3019Creates an emulated device in the kernel. The file descriptor returned 3020in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 3021 3022If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 3023device type is supported (not necessarily whether it can be created 3024in the current vm). 3025 3026Individual devices should not define flags. Attributes should be used 3027for specifying any behavior that is not implied by the device type 3028number. 3029 3030:: 3031 3032 struct kvm_create_device { 3033 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 3034 __u32 fd; /* out: device handle */ 3035 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 3036 }; 3037 30384.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3039-------------------------------------------- 3040 3041:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3042 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3043:Type: device ioctl, vm ioctl, vcpu ioctl 3044:Parameters: struct kvm_device_attr 3045:Returns: 0 on success, -1 on error 3046 3047Errors: 3048 3049 ===== ============================================================= 3050 ENXIO The group or attribute is unknown/unsupported for this device 3051 or hardware support is missing. 3052 EPERM The attribute cannot (currently) be accessed this way 3053 (e.g. read-only attribute, or attribute that only makes 3054 sense when the device is in a different state) 3055 ===== ============================================================= 3056 3057 Other error conditions may be defined by individual device types. 3058 3059Gets/sets a specified piece of device configuration and/or state. The 3060semantics are device-specific. See individual device documentation in 3061the "devices" directory. As with ONE_REG, the size of the data 3062transferred is defined by the particular attribute. 3063 3064:: 3065 3066 struct kvm_device_attr { 3067 __u32 flags; /* no flags currently defined */ 3068 __u32 group; /* device-defined */ 3069 __u64 attr; /* group-defined */ 3070 __u64 addr; /* userspace address of attr data */ 3071 }; 3072 30734.81 KVM_HAS_DEVICE_ATTR 3074------------------------ 3075 3076:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3077 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3078:Type: device ioctl, vm ioctl, vcpu ioctl 3079:Parameters: struct kvm_device_attr 3080:Returns: 0 on success, -1 on error 3081 3082Errors: 3083 3084 ===== ============================================================= 3085 ENXIO The group or attribute is unknown/unsupported for this device 3086 or hardware support is missing. 3087 ===== ============================================================= 3088 3089Tests whether a device supports a particular attribute. A successful 3090return indicates the attribute is implemented. It does not necessarily 3091indicate that the attribute can be read or written in the device's 3092current state. "addr" is ignored. 3093 30944.82 KVM_ARM_VCPU_INIT 3095---------------------- 3096 3097:Capability: basic 3098:Architectures: arm, arm64 3099:Type: vcpu ioctl 3100:Parameters: struct kvm_vcpu_init (in) 3101:Returns: 0 on success; -1 on error 3102 3103Errors: 3104 3105 ====== ================================================================= 3106 EINVAL the target is unknown, or the combination of features is invalid. 3107 ENOENT a features bit specified is unknown. 3108 ====== ================================================================= 3109 3110This tells KVM what type of CPU to present to the guest, and what 3111optional features it should have. This will cause a reset of the cpu 3112registers to their initial values. If this is not called, KVM_RUN will 3113return ENOEXEC for that vcpu. 3114 3115Note that because some registers reflect machine topology, all vcpus 3116should be created before this ioctl is invoked. 3117 3118Userspace can call this function multiple times for a given vcpu, including 3119after the vcpu has been run. This will reset the vcpu to its initial 3120state. All calls to this function after the initial call must use the same 3121target and same set of feature flags, otherwise EINVAL will be returned. 3122 3123Possible features: 3124 3125 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 3126 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 3127 and execute guest code when KVM_RUN is called. 3128 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 3129 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 3130 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision 3131 backward compatible with v0.2) for the CPU. 3132 Depends on KVM_CAP_ARM_PSCI_0_2. 3133 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 3134 Depends on KVM_CAP_ARM_PMU_V3. 3135 3136 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication 3137 for arm64 only. 3138 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS. 3139 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3140 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3141 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3142 requested. 3143 3144 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication 3145 for arm64 only. 3146 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC. 3147 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3148 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3149 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3150 requested. 3151 3152 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only). 3153 Depends on KVM_CAP_ARM_SVE. 3154 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3155 3156 * After KVM_ARM_VCPU_INIT: 3157 3158 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the 3159 initial value of this pseudo-register indicates the best set of 3160 vector lengths possible for a vcpu on this host. 3161 3162 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3163 3164 - KVM_RUN and KVM_GET_REG_LIST are not available; 3165 3166 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access 3167 the scalable archietctural SVE registers 3168 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or 3169 KVM_REG_ARM64_SVE_FFR; 3170 3171 - KVM_REG_ARM64_SVE_VLS may optionally be written using 3172 KVM_SET_ONE_REG, to modify the set of vector lengths available 3173 for the vcpu. 3174 3175 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3176 3177 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can 3178 no longer be written using KVM_SET_ONE_REG. 3179 31804.83 KVM_ARM_PREFERRED_TARGET 3181----------------------------- 3182 3183:Capability: basic 3184:Architectures: arm, arm64 3185:Type: vm ioctl 3186:Parameters: struct kvm_vcpu_init (out) 3187:Returns: 0 on success; -1 on error 3188 3189Errors: 3190 3191 ====== ========================================== 3192 ENODEV no preferred target available for the host 3193 ====== ========================================== 3194 3195This queries KVM for preferred CPU target type which can be emulated 3196by KVM on underlying host. 3197 3198The ioctl returns struct kvm_vcpu_init instance containing information 3199about preferred CPU target type and recommended features for it. The 3200kvm_vcpu_init->features bitmap returned will have feature bits set if 3201the preferred target recommends setting these features, but this is 3202not mandatory. 3203 3204The information returned by this ioctl can be used to prepare an instance 3205of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 3206VCPU matching underlying host. 3207 3208 32094.84 KVM_GET_REG_LIST 3210--------------------- 3211 3212:Capability: basic 3213:Architectures: arm, arm64, mips 3214:Type: vcpu ioctl 3215:Parameters: struct kvm_reg_list (in/out) 3216:Returns: 0 on success; -1 on error 3217 3218Errors: 3219 3220 ===== ============================================================== 3221 E2BIG the reg index list is too big to fit in the array specified by 3222 the user (the number required will be written into n). 3223 ===== ============================================================== 3224 3225:: 3226 3227 struct kvm_reg_list { 3228 __u64 n; /* number of registers in reg[] */ 3229 __u64 reg[0]; 3230 }; 3231 3232This ioctl returns the guest registers that are supported for the 3233KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3234 3235 32364.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3237----------------------------------------- 3238 3239:Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3240:Architectures: arm, arm64 3241:Type: vm ioctl 3242:Parameters: struct kvm_arm_device_address (in) 3243:Returns: 0 on success, -1 on error 3244 3245Errors: 3246 3247 ====== ============================================ 3248 ENODEV The device id is unknown 3249 ENXIO Device not supported on current system 3250 EEXIST Address already set 3251 E2BIG Address outside guest physical address space 3252 EBUSY Address overlaps with other device range 3253 ====== ============================================ 3254 3255:: 3256 3257 struct kvm_arm_device_addr { 3258 __u64 id; 3259 __u64 addr; 3260 }; 3261 3262Specify a device address in the guest's physical address space where guests 3263can access emulated or directly exposed devices, which the host kernel needs 3264to know about. The id field is an architecture specific identifier for a 3265specific device. 3266 3267ARM/arm64 divides the id field into two parts, a device id and an 3268address type id specific to the individual device:: 3269 3270 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 3271 field: | 0x00000000 | device id | addr type id | 3272 3273ARM/arm64 currently only require this when using the in-kernel GIC 3274support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 3275as the device id. When setting the base address for the guest's 3276mapping of the VGIC virtual CPU and distributor interface, the ioctl 3277must be called after calling KVM_CREATE_IRQCHIP, but before calling 3278KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 3279base addresses will return -EEXIST. 3280 3281Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 3282should be used instead. 3283 3284 32854.86 KVM_PPC_RTAS_DEFINE_TOKEN 3286------------------------------ 3287 3288:Capability: KVM_CAP_PPC_RTAS 3289:Architectures: ppc 3290:Type: vm ioctl 3291:Parameters: struct kvm_rtas_token_args 3292:Returns: 0 on success, -1 on error 3293 3294Defines a token value for a RTAS (Run Time Abstraction Services) 3295service in order to allow it to be handled in the kernel. The 3296argument struct gives the name of the service, which must be the name 3297of a service that has a kernel-side implementation. If the token 3298value is non-zero, it will be associated with that service, and 3299subsequent RTAS calls by the guest specifying that token will be 3300handled by the kernel. If the token value is 0, then any token 3301associated with the service will be forgotten, and subsequent RTAS 3302calls by the guest for that service will be passed to userspace to be 3303handled. 3304 33054.87 KVM_SET_GUEST_DEBUG 3306------------------------ 3307 3308:Capability: KVM_CAP_SET_GUEST_DEBUG 3309:Architectures: x86, s390, ppc, arm64 3310:Type: vcpu ioctl 3311:Parameters: struct kvm_guest_debug (in) 3312:Returns: 0 on success; -1 on error 3313 3314:: 3315 3316 struct kvm_guest_debug { 3317 __u32 control; 3318 __u32 pad; 3319 struct kvm_guest_debug_arch arch; 3320 }; 3321 3322Set up the processor specific debug registers and configure vcpu for 3323handling guest debug events. There are two parts to the structure, the 3324first a control bitfield indicates the type of debug events to handle 3325when running. Common control bits are: 3326 3327 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 3328 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 3329 3330The top 16 bits of the control field are architecture specific control 3331flags which can include the following: 3332 3333 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 3334 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64] 3335 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 3336 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 3337 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 3338 3339For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 3340are enabled in memory so we need to ensure breakpoint exceptions are 3341correctly trapped and the KVM run loop exits at the breakpoint and not 3342running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 3343we need to ensure the guest vCPUs architecture specific registers are 3344updated to the correct (supplied) values. 3345 3346The second part of the structure is architecture specific and 3347typically contains a set of debug registers. 3348 3349For arm64 the number of debug registers is implementation defined and 3350can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 3351KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 3352indicating the number of supported registers. 3353 3354For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether 3355the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported. 3356 3357When debug events exit the main run loop with the reason 3358KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 3359structure containing architecture specific debug information. 3360 33614.88 KVM_GET_EMULATED_CPUID 3362--------------------------- 3363 3364:Capability: KVM_CAP_EXT_EMUL_CPUID 3365:Architectures: x86 3366:Type: system ioctl 3367:Parameters: struct kvm_cpuid2 (in/out) 3368:Returns: 0 on success, -1 on error 3369 3370:: 3371 3372 struct kvm_cpuid2 { 3373 __u32 nent; 3374 __u32 flags; 3375 struct kvm_cpuid_entry2 entries[0]; 3376 }; 3377 3378The member 'flags' is used for passing flags from userspace. 3379 3380:: 3381 3382 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 3383 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 3384 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 3385 3386 struct kvm_cpuid_entry2 { 3387 __u32 function; 3388 __u32 index; 3389 __u32 flags; 3390 __u32 eax; 3391 __u32 ebx; 3392 __u32 ecx; 3393 __u32 edx; 3394 __u32 padding[3]; 3395 }; 3396 3397This ioctl returns x86 cpuid features which are emulated by 3398kvm.Userspace can use the information returned by this ioctl to query 3399which features are emulated by kvm instead of being present natively. 3400 3401Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 3402structure with the 'nent' field indicating the number of entries in 3403the variable-size array 'entries'. If the number of entries is too low 3404to describe the cpu capabilities, an error (E2BIG) is returned. If the 3405number is too high, the 'nent' field is adjusted and an error (ENOMEM) 3406is returned. If the number is just right, the 'nent' field is adjusted 3407to the number of valid entries in the 'entries' array, which is then 3408filled. 3409 3410The entries returned are the set CPUID bits of the respective features 3411which kvm emulates, as returned by the CPUID instruction, with unknown 3412or unsupported feature bits cleared. 3413 3414Features like x2apic, for example, may not be present in the host cpu 3415but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 3416emulated efficiently and thus not included here. 3417 3418The fields in each entry are defined as follows: 3419 3420 function: 3421 the eax value used to obtain the entry 3422 index: 3423 the ecx value used to obtain the entry (for entries that are 3424 affected by ecx) 3425 flags: 3426 an OR of zero or more of the following: 3427 3428 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3429 if the index field is valid 3430 3431 eax, ebx, ecx, edx: 3432 3433 the values returned by the cpuid instruction for 3434 this function/index combination 3435 34364.89 KVM_S390_MEM_OP 3437-------------------- 3438 3439:Capability: KVM_CAP_S390_MEM_OP 3440:Architectures: s390 3441:Type: vcpu ioctl 3442:Parameters: struct kvm_s390_mem_op (in) 3443:Returns: = 0 on success, 3444 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 3445 > 0 if an exception occurred while walking the page tables 3446 3447Read or write data from/to the logical (virtual) memory of a VCPU. 3448 3449Parameters are specified via the following structure:: 3450 3451 struct kvm_s390_mem_op { 3452 __u64 gaddr; /* the guest address */ 3453 __u64 flags; /* flags */ 3454 __u32 size; /* amount of bytes */ 3455 __u32 op; /* type of operation */ 3456 __u64 buf; /* buffer in userspace */ 3457 __u8 ar; /* the access register number */ 3458 __u8 reserved[31]; /* should be set to 0 */ 3459 }; 3460 3461The type of operation is specified in the "op" field. It is either 3462KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or 3463KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The 3464KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check 3465whether the corresponding memory access would create an access exception 3466(without touching the data in the memory at the destination). In case an 3467access exception occurred while walking the MMU tables of the guest, the 3468ioctl returns a positive error number to indicate the type of exception. 3469This exception is also raised directly at the corresponding VCPU if the 3470flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field. 3471 3472The start address of the memory region has to be specified in the "gaddr" 3473field, and the length of the region in the "size" field (which must not 3474be 0). The maximum value for "size" can be obtained by checking the 3475KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the 3476userspace application where the read data should be written to for 3477KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written is 3478stored for a KVM_S390_MEMOP_LOGICAL_WRITE. When KVM_S390_MEMOP_F_CHECK_ONLY 3479is specified, "buf" is unused and can be NULL. "ar" designates the access 3480register number to be used; the valid range is 0..15. 3481 3482The "reserved" field is meant for future extensions. It is not used by 3483KVM with the currently defined set of flags. 3484 34854.90 KVM_S390_GET_SKEYS 3486----------------------- 3487 3488:Capability: KVM_CAP_S390_SKEYS 3489:Architectures: s390 3490:Type: vm ioctl 3491:Parameters: struct kvm_s390_skeys 3492:Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage 3493 keys, negative value on error 3494 3495This ioctl is used to get guest storage key values on the s390 3496architecture. The ioctl takes parameters via the kvm_s390_skeys struct:: 3497 3498 struct kvm_s390_skeys { 3499 __u64 start_gfn; 3500 __u64 count; 3501 __u64 skeydata_addr; 3502 __u32 flags; 3503 __u32 reserved[9]; 3504 }; 3505 3506The start_gfn field is the number of the first guest frame whose storage keys 3507you want to get. 3508 3509The count field is the number of consecutive frames (starting from start_gfn) 3510whose storage keys to get. The count field must be at least 1 and the maximum 3511allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range 3512will cause the ioctl to return -EINVAL. 3513 3514The skeydata_addr field is the address to a buffer large enough to hold count 3515bytes. This buffer will be filled with storage key data by the ioctl. 3516 35174.91 KVM_S390_SET_SKEYS 3518----------------------- 3519 3520:Capability: KVM_CAP_S390_SKEYS 3521:Architectures: s390 3522:Type: vm ioctl 3523:Parameters: struct kvm_s390_skeys 3524:Returns: 0 on success, negative value on error 3525 3526This ioctl is used to set guest storage key values on the s390 3527architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 3528See section on KVM_S390_GET_SKEYS for struct definition. 3529 3530The start_gfn field is the number of the first guest frame whose storage keys 3531you want to set. 3532 3533The count field is the number of consecutive frames (starting from start_gfn) 3534whose storage keys to get. The count field must be at least 1 and the maximum 3535allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range 3536will cause the ioctl to return -EINVAL. 3537 3538The skeydata_addr field is the address to a buffer containing count bytes of 3539storage keys. Each byte in the buffer will be set as the storage key for a 3540single frame starting at start_gfn for count frames. 3541 3542Note: If any architecturally invalid key value is found in the given data then 3543the ioctl will return -EINVAL. 3544 35454.92 KVM_S390_IRQ 3546----------------- 3547 3548:Capability: KVM_CAP_S390_INJECT_IRQ 3549:Architectures: s390 3550:Type: vcpu ioctl 3551:Parameters: struct kvm_s390_irq (in) 3552:Returns: 0 on success, -1 on error 3553 3554Errors: 3555 3556 3557 ====== ================================================================= 3558 EINVAL interrupt type is invalid 3559 type is KVM_S390_SIGP_STOP and flag parameter is invalid value, 3560 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 3561 than the maximum of VCPUs 3562 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped, 3563 type is KVM_S390_SIGP_STOP and a stop irq is already pending, 3564 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 3565 is already pending 3566 ====== ================================================================= 3567 3568Allows to inject an interrupt to the guest. 3569 3570Using struct kvm_s390_irq as a parameter allows 3571to inject additional payload which is not 3572possible via KVM_S390_INTERRUPT. 3573 3574Interrupt parameters are passed via kvm_s390_irq:: 3575 3576 struct kvm_s390_irq { 3577 __u64 type; 3578 union { 3579 struct kvm_s390_io_info io; 3580 struct kvm_s390_ext_info ext; 3581 struct kvm_s390_pgm_info pgm; 3582 struct kvm_s390_emerg_info emerg; 3583 struct kvm_s390_extcall_info extcall; 3584 struct kvm_s390_prefix_info prefix; 3585 struct kvm_s390_stop_info stop; 3586 struct kvm_s390_mchk_info mchk; 3587 char reserved[64]; 3588 } u; 3589 }; 3590 3591type can be one of the following: 3592 3593- KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 3594- KVM_S390_PROGRAM_INT - program check; parameters in .pgm 3595- KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 3596- KVM_S390_RESTART - restart; no parameters 3597- KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 3598- KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 3599- KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 3600- KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 3601- KVM_S390_MCHK - machine check interrupt; parameters in .mchk 3602 3603This is an asynchronous vcpu ioctl and can be invoked from any thread. 3604 36054.94 KVM_S390_GET_IRQ_STATE 3606--------------------------- 3607 3608:Capability: KVM_CAP_S390_IRQ_STATE 3609:Architectures: s390 3610:Type: vcpu ioctl 3611:Parameters: struct kvm_s390_irq_state (out) 3612:Returns: >= number of bytes copied into buffer, 3613 -EINVAL if buffer size is 0, 3614 -ENOBUFS if buffer size is too small to fit all pending interrupts, 3615 -EFAULT if the buffer address was invalid 3616 3617This ioctl allows userspace to retrieve the complete state of all currently 3618pending interrupts in a single buffer. Use cases include migration 3619and introspection. The parameter structure contains the address of a 3620userspace buffer and its length:: 3621 3622 struct kvm_s390_irq_state { 3623 __u64 buf; 3624 __u32 flags; /* will stay unused for compatibility reasons */ 3625 __u32 len; 3626 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 3627 }; 3628 3629Userspace passes in the above struct and for each pending interrupt a 3630struct kvm_s390_irq is copied to the provided buffer. 3631 3632The structure contains a flags and a reserved field for future extensions. As 3633the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 3634reserved, these fields can not be used in the future without breaking 3635compatibility. 3636 3637If -ENOBUFS is returned the buffer provided was too small and userspace 3638may retry with a bigger buffer. 3639 36404.95 KVM_S390_SET_IRQ_STATE 3641--------------------------- 3642 3643:Capability: KVM_CAP_S390_IRQ_STATE 3644:Architectures: s390 3645:Type: vcpu ioctl 3646:Parameters: struct kvm_s390_irq_state (in) 3647:Returns: 0 on success, 3648 -EFAULT if the buffer address was invalid, 3649 -EINVAL for an invalid buffer length (see below), 3650 -EBUSY if there were already interrupts pending, 3651 errors occurring when actually injecting the 3652 interrupt. See KVM_S390_IRQ. 3653 3654This ioctl allows userspace to set the complete state of all cpu-local 3655interrupts currently pending for the vcpu. It is intended for restoring 3656interrupt state after a migration. The input parameter is a userspace buffer 3657containing a struct kvm_s390_irq_state:: 3658 3659 struct kvm_s390_irq_state { 3660 __u64 buf; 3661 __u32 flags; /* will stay unused for compatibility reasons */ 3662 __u32 len; 3663 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 3664 }; 3665 3666The restrictions for flags and reserved apply as well. 3667(see KVM_S390_GET_IRQ_STATE) 3668 3669The userspace memory referenced by buf contains a struct kvm_s390_irq 3670for each interrupt to be injected into the guest. 3671If one of the interrupts could not be injected for some reason the 3672ioctl aborts. 3673 3674len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 3675and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 3676which is the maximum number of possibly pending cpu-local interrupts. 3677 36784.96 KVM_SMI 3679------------ 3680 3681:Capability: KVM_CAP_X86_SMM 3682:Architectures: x86 3683:Type: vcpu ioctl 3684:Parameters: none 3685:Returns: 0 on success, -1 on error 3686 3687Queues an SMI on the thread's vcpu. 3688 36894.97 KVM_CAP_PPC_MULTITCE 3690------------------------- 3691 3692:Capability: KVM_CAP_PPC_MULTITCE 3693:Architectures: ppc 3694:Type: vm 3695 3696This capability means the kernel is capable of handling hypercalls 3697H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 3698space. This significantly accelerates DMA operations for PPC KVM guests. 3699User space should expect that its handlers for these hypercalls 3700are not going to be called if user space previously registered LIOBN 3701in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 3702 3703In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 3704user space might have to advertise it for the guest. For example, 3705IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 3706present in the "ibm,hypertas-functions" device-tree property. 3707 3708The hypercalls mentioned above may or may not be processed successfully 3709in the kernel based fast path. If they can not be handled by the kernel, 3710they will get passed on to user space. So user space still has to have 3711an implementation for these despite the in kernel acceleration. 3712 3713This capability is always enabled. 3714 37154.98 KVM_CREATE_SPAPR_TCE_64 3716---------------------------- 3717 3718:Capability: KVM_CAP_SPAPR_TCE_64 3719:Architectures: powerpc 3720:Type: vm ioctl 3721:Parameters: struct kvm_create_spapr_tce_64 (in) 3722:Returns: file descriptor for manipulating the created TCE table 3723 3724This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 3725windows, described in 4.62 KVM_CREATE_SPAPR_TCE 3726 3727This capability uses extended struct in ioctl interface:: 3728 3729 /* for KVM_CAP_SPAPR_TCE_64 */ 3730 struct kvm_create_spapr_tce_64 { 3731 __u64 liobn; 3732 __u32 page_shift; 3733 __u32 flags; 3734 __u64 offset; /* in pages */ 3735 __u64 size; /* in pages */ 3736 }; 3737 3738The aim of extension is to support an additional bigger DMA window with 3739a variable page size. 3740KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 3741a bus offset of the corresponding DMA window, @size and @offset are numbers 3742of IOMMU pages. 3743 3744@flags are not used at the moment. 3745 3746The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 3747 37484.99 KVM_REINJECT_CONTROL 3749------------------------- 3750 3751:Capability: KVM_CAP_REINJECT_CONTROL 3752:Architectures: x86 3753:Type: vm ioctl 3754:Parameters: struct kvm_reinject_control (in) 3755:Returns: 0 on success, 3756 -EFAULT if struct kvm_reinject_control cannot be read, 3757 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 3758 3759i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 3760where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 3761vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 3762interrupt whenever there isn't a pending interrupt from i8254. 3763!reinject mode injects an interrupt as soon as a tick arrives. 3764 3765:: 3766 3767 struct kvm_reinject_control { 3768 __u8 pit_reinject; 3769 __u8 reserved[31]; 3770 }; 3771 3772pit_reinject = 0 (!reinject mode) is recommended, unless running an old 3773operating system that uses the PIT for timing (e.g. Linux 2.4.x). 3774 37754.100 KVM_PPC_CONFIGURE_V3_MMU 3776------------------------------ 3777 3778:Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 3779:Architectures: ppc 3780:Type: vm ioctl 3781:Parameters: struct kvm_ppc_mmuv3_cfg (in) 3782:Returns: 0 on success, 3783 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 3784 -EINVAL if the configuration is invalid 3785 3786This ioctl controls whether the guest will use radix or HPT (hashed 3787page table) translation, and sets the pointer to the process table for 3788the guest. 3789 3790:: 3791 3792 struct kvm_ppc_mmuv3_cfg { 3793 __u64 flags; 3794 __u64 process_table; 3795 }; 3796 3797There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 3798KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 3799to use radix tree translation, and if clear, to use HPT translation. 3800KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 3801to be able to use the global TLB and SLB invalidation instructions; 3802if clear, the guest may not use these instructions. 3803 3804The process_table field specifies the address and size of the guest 3805process table, which is in the guest's space. This field is formatted 3806as the second doubleword of the partition table entry, as defined in 3807the Power ISA V3.00, Book III section 5.7.6.1. 3808 38094.101 KVM_PPC_GET_RMMU_INFO 3810--------------------------- 3811 3812:Capability: KVM_CAP_PPC_RADIX_MMU 3813:Architectures: ppc 3814:Type: vm ioctl 3815:Parameters: struct kvm_ppc_rmmu_info (out) 3816:Returns: 0 on success, 3817 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 3818 -EINVAL if no useful information can be returned 3819 3820This ioctl returns a structure containing two things: (a) a list 3821containing supported radix tree geometries, and (b) a list that maps 3822page sizes to put in the "AP" (actual page size) field for the tlbie 3823(TLB invalidate entry) instruction. 3824 3825:: 3826 3827 struct kvm_ppc_rmmu_info { 3828 struct kvm_ppc_radix_geom { 3829 __u8 page_shift; 3830 __u8 level_bits[4]; 3831 __u8 pad[3]; 3832 } geometries[8]; 3833 __u32 ap_encodings[8]; 3834 }; 3835 3836The geometries[] field gives up to 8 supported geometries for the 3837radix page table, in terms of the log base 2 of the smallest page 3838size, and the number of bits indexed at each level of the tree, from 3839the PTE level up to the PGD level in that order. Any unused entries 3840will have 0 in the page_shift field. 3841 3842The ap_encodings gives the supported page sizes and their AP field 3843encodings, encoded with the AP value in the top 3 bits and the log 3844base 2 of the page size in the bottom 6 bits. 3845 38464.102 KVM_PPC_RESIZE_HPT_PREPARE 3847-------------------------------- 3848 3849:Capability: KVM_CAP_SPAPR_RESIZE_HPT 3850:Architectures: powerpc 3851:Type: vm ioctl 3852:Parameters: struct kvm_ppc_resize_hpt (in) 3853:Returns: 0 on successful completion, 3854 >0 if a new HPT is being prepared, the value is an estimated 3855 number of milliseconds until preparation is complete, 3856 -EFAULT if struct kvm_reinject_control cannot be read, 3857 -EINVAL if the supplied shift or flags are invalid, 3858 -ENOMEM if unable to allocate the new HPT, 3859 3860Used to implement the PAPR extension for runtime resizing of a guest's 3861Hashed Page Table (HPT). Specifically this starts, stops or monitors 3862the preparation of a new potential HPT for the guest, essentially 3863implementing the H_RESIZE_HPT_PREPARE hypercall. 3864 3865:: 3866 3867 struct kvm_ppc_resize_hpt { 3868 __u64 flags; 3869 __u32 shift; 3870 __u32 pad; 3871 }; 3872 3873If called with shift > 0 when there is no pending HPT for the guest, 3874this begins preparation of a new pending HPT of size 2^(shift) bytes. 3875It then returns a positive integer with the estimated number of 3876milliseconds until preparation is complete. 3877 3878If called when there is a pending HPT whose size does not match that 3879requested in the parameters, discards the existing pending HPT and 3880creates a new one as above. 3881 3882If called when there is a pending HPT of the size requested, will: 3883 3884 * If preparation of the pending HPT is already complete, return 0 3885 * If preparation of the pending HPT has failed, return an error 3886 code, then discard the pending HPT. 3887 * If preparation of the pending HPT is still in progress, return an 3888 estimated number of milliseconds until preparation is complete. 3889 3890If called with shift == 0, discards any currently pending HPT and 3891returns 0 (i.e. cancels any in-progress preparation). 3892 3893flags is reserved for future expansion, currently setting any bits in 3894flags will result in an -EINVAL. 3895 3896Normally this will be called repeatedly with the same parameters until 3897it returns <= 0. The first call will initiate preparation, subsequent 3898ones will monitor preparation until it completes or fails. 3899 39004.103 KVM_PPC_RESIZE_HPT_COMMIT 3901------------------------------- 3902 3903:Capability: KVM_CAP_SPAPR_RESIZE_HPT 3904:Architectures: powerpc 3905:Type: vm ioctl 3906:Parameters: struct kvm_ppc_resize_hpt (in) 3907:Returns: 0 on successful completion, 3908 -EFAULT if struct kvm_reinject_control cannot be read, 3909 -EINVAL if the supplied shift or flags are invalid, 3910 -ENXIO is there is no pending HPT, or the pending HPT doesn't 3911 have the requested size, 3912 -EBUSY if the pending HPT is not fully prepared, 3913 -ENOSPC if there was a hash collision when moving existing 3914 HPT entries to the new HPT, 3915 -EIO on other error conditions 3916 3917Used to implement the PAPR extension for runtime resizing of a guest's 3918Hashed Page Table (HPT). Specifically this requests that the guest be 3919transferred to working with the new HPT, essentially implementing the 3920H_RESIZE_HPT_COMMIT hypercall. 3921 3922:: 3923 3924 struct kvm_ppc_resize_hpt { 3925 __u64 flags; 3926 __u32 shift; 3927 __u32 pad; 3928 }; 3929 3930This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 3931returned 0 with the same parameters. In other cases 3932KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 3933-EBUSY, though others may be possible if the preparation was started, 3934but failed). 3935 3936This will have undefined effects on the guest if it has not already 3937placed itself in a quiescent state where no vcpu will make MMU enabled 3938memory accesses. 3939 3940On succsful completion, the pending HPT will become the guest's active 3941HPT and the previous HPT will be discarded. 3942 3943On failure, the guest will still be operating on its previous HPT. 3944 39454.104 KVM_X86_GET_MCE_CAP_SUPPORTED 3946----------------------------------- 3947 3948:Capability: KVM_CAP_MCE 3949:Architectures: x86 3950:Type: system ioctl 3951:Parameters: u64 mce_cap (out) 3952:Returns: 0 on success, -1 on error 3953 3954Returns supported MCE capabilities. The u64 mce_cap parameter 3955has the same format as the MSR_IA32_MCG_CAP register. Supported 3956capabilities will have the corresponding bits set. 3957 39584.105 KVM_X86_SETUP_MCE 3959----------------------- 3960 3961:Capability: KVM_CAP_MCE 3962:Architectures: x86 3963:Type: vcpu ioctl 3964:Parameters: u64 mcg_cap (in) 3965:Returns: 0 on success, 3966 -EFAULT if u64 mcg_cap cannot be read, 3967 -EINVAL if the requested number of banks is invalid, 3968 -EINVAL if requested MCE capability is not supported. 3969 3970Initializes MCE support for use. The u64 mcg_cap parameter 3971has the same format as the MSR_IA32_MCG_CAP register and 3972specifies which capabilities should be enabled. The maximum 3973supported number of error-reporting banks can be retrieved when 3974checking for KVM_CAP_MCE. The supported capabilities can be 3975retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 3976 39774.106 KVM_X86_SET_MCE 3978--------------------- 3979 3980:Capability: KVM_CAP_MCE 3981:Architectures: x86 3982:Type: vcpu ioctl 3983:Parameters: struct kvm_x86_mce (in) 3984:Returns: 0 on success, 3985 -EFAULT if struct kvm_x86_mce cannot be read, 3986 -EINVAL if the bank number is invalid, 3987 -EINVAL if VAL bit is not set in status field. 3988 3989Inject a machine check error (MCE) into the guest. The input 3990parameter is:: 3991 3992 struct kvm_x86_mce { 3993 __u64 status; 3994 __u64 addr; 3995 __u64 misc; 3996 __u64 mcg_status; 3997 __u8 bank; 3998 __u8 pad1[7]; 3999 __u64 pad2[3]; 4000 }; 4001 4002If the MCE being reported is an uncorrected error, KVM will 4003inject it as an MCE exception into the guest. If the guest 4004MCG_STATUS register reports that an MCE is in progress, KVM 4005causes an KVM_EXIT_SHUTDOWN vmexit. 4006 4007Otherwise, if the MCE is a corrected error, KVM will just 4008store it in the corresponding bank (provided this bank is 4009not holding a previously reported uncorrected error). 4010 40114.107 KVM_S390_GET_CMMA_BITS 4012---------------------------- 4013 4014:Capability: KVM_CAP_S390_CMMA_MIGRATION 4015:Architectures: s390 4016:Type: vm ioctl 4017:Parameters: struct kvm_s390_cmma_log (in, out) 4018:Returns: 0 on success, a negative value on error 4019 4020This ioctl is used to get the values of the CMMA bits on the s390 4021architecture. It is meant to be used in two scenarios: 4022 4023- During live migration to save the CMMA values. Live migration needs 4024 to be enabled via the KVM_REQ_START_MIGRATION VM property. 4025- To non-destructively peek at the CMMA values, with the flag 4026 KVM_S390_CMMA_PEEK set. 4027 4028The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 4029values are written to a buffer whose location is indicated via the "values" 4030member in the kvm_s390_cmma_log struct. The values in the input struct are 4031also updated as needed. 4032 4033Each CMMA value takes up one byte. 4034 4035:: 4036 4037 struct kvm_s390_cmma_log { 4038 __u64 start_gfn; 4039 __u32 count; 4040 __u32 flags; 4041 union { 4042 __u64 remaining; 4043 __u64 mask; 4044 }; 4045 __u64 values; 4046 }; 4047 4048start_gfn is the number of the first guest frame whose CMMA values are 4049to be retrieved, 4050 4051count is the length of the buffer in bytes, 4052 4053values points to the buffer where the result will be written to. 4054 4055If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 4056KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 4057other ioctls. 4058 4059The result is written in the buffer pointed to by the field values, and 4060the values of the input parameter are updated as follows. 4061 4062Depending on the flags, different actions are performed. The only 4063supported flag so far is KVM_S390_CMMA_PEEK. 4064 4065The default behaviour if KVM_S390_CMMA_PEEK is not set is: 4066start_gfn will indicate the first page frame whose CMMA bits were dirty. 4067It is not necessarily the same as the one passed as input, as clean pages 4068are skipped. 4069 4070count will indicate the number of bytes actually written in the buffer. 4071It can (and very often will) be smaller than the input value, since the 4072buffer is only filled until 16 bytes of clean values are found (which 4073are then not copied in the buffer). Since a CMMA migration block needs 4074the base address and the length, for a total of 16 bytes, we will send 4075back some clean data if there is some dirty data afterwards, as long as 4076the size of the clean data does not exceed the size of the header. This 4077allows to minimize the amount of data to be saved or transferred over 4078the network at the expense of more roundtrips to userspace. The next 4079invocation of the ioctl will skip over all the clean values, saving 4080potentially more than just the 16 bytes we found. 4081 4082If KVM_S390_CMMA_PEEK is set: 4083the existing storage attributes are read even when not in migration 4084mode, and no other action is performed; 4085 4086the output start_gfn will be equal to the input start_gfn, 4087 4088the output count will be equal to the input count, except if the end of 4089memory has been reached. 4090 4091In both cases: 4092the field "remaining" will indicate the total number of dirty CMMA values 4093still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 4094not enabled. 4095 4096mask is unused. 4097 4098values points to the userspace buffer where the result will be stored. 4099 4100This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4101complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4102KVM_S390_CMMA_PEEK is not set but migration mode was not enabled, with 4103-EFAULT if the userspace address is invalid or if no page table is 4104present for the addresses (e.g. when using hugepages). 4105 41064.108 KVM_S390_SET_CMMA_BITS 4107---------------------------- 4108 4109:Capability: KVM_CAP_S390_CMMA_MIGRATION 4110:Architectures: s390 4111:Type: vm ioctl 4112:Parameters: struct kvm_s390_cmma_log (in) 4113:Returns: 0 on success, a negative value on error 4114 4115This ioctl is used to set the values of the CMMA bits on the s390 4116architecture. It is meant to be used during live migration to restore 4117the CMMA values, but there are no restrictions on its use. 4118The ioctl takes parameters via the kvm_s390_cmma_values struct. 4119Each CMMA value takes up one byte. 4120 4121:: 4122 4123 struct kvm_s390_cmma_log { 4124 __u64 start_gfn; 4125 __u32 count; 4126 __u32 flags; 4127 union { 4128 __u64 remaining; 4129 __u64 mask; 4130 }; 4131 __u64 values; 4132 }; 4133 4134start_gfn indicates the starting guest frame number, 4135 4136count indicates how many values are to be considered in the buffer, 4137 4138flags is not used and must be 0. 4139 4140mask indicates which PGSTE bits are to be considered. 4141 4142remaining is not used. 4143 4144values points to the buffer in userspace where to store the values. 4145 4146This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4147complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4148the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 4149if the flags field was not 0, with -EFAULT if the userspace address is 4150invalid, if invalid pages are written to (e.g. after the end of memory) 4151or if no page table is present for the addresses (e.g. when using 4152hugepages). 4153 41544.109 KVM_PPC_GET_CPU_CHAR 4155-------------------------- 4156 4157:Capability: KVM_CAP_PPC_GET_CPU_CHAR 4158:Architectures: powerpc 4159:Type: vm ioctl 4160:Parameters: struct kvm_ppc_cpu_char (out) 4161:Returns: 0 on successful completion, 4162 -EFAULT if struct kvm_ppc_cpu_char cannot be written 4163 4164This ioctl gives userspace information about certain characteristics 4165of the CPU relating to speculative execution of instructions and 4166possible information leakage resulting from speculative execution (see 4167CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 4168returned in struct kvm_ppc_cpu_char, which looks like this:: 4169 4170 struct kvm_ppc_cpu_char { 4171 __u64 character; /* characteristics of the CPU */ 4172 __u64 behaviour; /* recommended software behaviour */ 4173 __u64 character_mask; /* valid bits in character */ 4174 __u64 behaviour_mask; /* valid bits in behaviour */ 4175 }; 4176 4177For extensibility, the character_mask and behaviour_mask fields 4178indicate which bits of character and behaviour have been filled in by 4179the kernel. If the set of defined bits is extended in future then 4180userspace will be able to tell whether it is running on a kernel that 4181knows about the new bits. 4182 4183The character field describes attributes of the CPU which can help 4184with preventing inadvertent information disclosure - specifically, 4185whether there is an instruction to flash-invalidate the L1 data cache 4186(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 4187to a mode where entries can only be used by the thread that created 4188them, whether the bcctr[l] instruction prevents speculation, and 4189whether a speculation barrier instruction (ori 31,31,0) is provided. 4190 4191The behaviour field describes actions that software should take to 4192prevent inadvertent information disclosure, and thus describes which 4193vulnerabilities the hardware is subject to; specifically whether the 4194L1 data cache should be flushed when returning to user mode from the 4195kernel, and whether a speculation barrier should be placed between an 4196array bounds check and the array access. 4197 4198These fields use the same bit definitions as the new 4199H_GET_CPU_CHARACTERISTICS hypercall. 4200 42014.110 KVM_MEMORY_ENCRYPT_OP 4202--------------------------- 4203 4204:Capability: basic 4205:Architectures: x86 4206:Type: vm 4207:Parameters: an opaque platform specific structure (in/out) 4208:Returns: 0 on success; -1 on error 4209 4210If the platform supports creating encrypted VMs then this ioctl can be used 4211for issuing platform-specific memory encryption commands to manage those 4212encrypted VMs. 4213 4214Currently, this ioctl is used for issuing Secure Encrypted Virtualization 4215(SEV) commands on AMD Processors. The SEV commands are defined in 4216Documentation/virt/kvm/amd-memory-encryption.rst. 4217 42184.111 KVM_MEMORY_ENCRYPT_REG_REGION 4219----------------------------------- 4220 4221:Capability: basic 4222:Architectures: x86 4223:Type: system 4224:Parameters: struct kvm_enc_region (in) 4225:Returns: 0 on success; -1 on error 4226 4227This ioctl can be used to register a guest memory region which may 4228contain encrypted data (e.g. guest RAM, SMRAM etc). 4229 4230It is used in the SEV-enabled guest. When encryption is enabled, a guest 4231memory region may contain encrypted data. The SEV memory encryption 4232engine uses a tweak such that two identical plaintext pages, each at 4233different locations will have differing ciphertexts. So swapping or 4234moving ciphertext of those pages will not result in plaintext being 4235swapped. So relocating (or migrating) physical backing pages for the SEV 4236guest will require some additional steps. 4237 4238Note: The current SEV key management spec does not provide commands to 4239swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 4240memory region registered with the ioctl. 4241 42424.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4243------------------------------------- 4244 4245:Capability: basic 4246:Architectures: x86 4247:Type: system 4248:Parameters: struct kvm_enc_region (in) 4249:Returns: 0 on success; -1 on error 4250 4251This ioctl can be used to unregister the guest memory region registered 4252with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 4253 42544.113 KVM_HYPERV_EVENTFD 4255------------------------ 4256 4257:Capability: KVM_CAP_HYPERV_EVENTFD 4258:Architectures: x86 4259:Type: vm ioctl 4260:Parameters: struct kvm_hyperv_eventfd (in) 4261 4262This ioctl (un)registers an eventfd to receive notifications from the guest on 4263the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without 4264causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number 4265(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit. 4266 4267:: 4268 4269 struct kvm_hyperv_eventfd { 4270 __u32 conn_id; 4271 __s32 fd; 4272 __u32 flags; 4273 __u32 padding[3]; 4274 }; 4275 4276The conn_id field should fit within 24 bits:: 4277 4278 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff 4279 4280The acceptable values for the flags field are:: 4281 4282 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0) 4283 4284:Returns: 0 on success, 4285 -EINVAL if conn_id or flags is outside the allowed range, 4286 -ENOENT on deassign if the conn_id isn't registered, 4287 -EEXIST on assign if the conn_id is already registered 4288 42894.114 KVM_GET_NESTED_STATE 4290-------------------------- 4291 4292:Capability: KVM_CAP_NESTED_STATE 4293:Architectures: x86 4294:Type: vcpu ioctl 4295:Parameters: struct kvm_nested_state (in/out) 4296:Returns: 0 on success, -1 on error 4297 4298Errors: 4299 4300 ===== ============================================================= 4301 E2BIG the total state size exceeds the value of 'size' specified by 4302 the user; the size required will be written into size. 4303 ===== ============================================================= 4304 4305:: 4306 4307 struct kvm_nested_state { 4308 __u16 flags; 4309 __u16 format; 4310 __u32 size; 4311 4312 union { 4313 struct kvm_vmx_nested_state_hdr vmx; 4314 struct kvm_svm_nested_state_hdr svm; 4315 4316 /* Pad the header to 128 bytes. */ 4317 __u8 pad[120]; 4318 } hdr; 4319 4320 union { 4321 struct kvm_vmx_nested_state_data vmx[0]; 4322 struct kvm_svm_nested_state_data svm[0]; 4323 } data; 4324 }; 4325 4326 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001 4327 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002 4328 #define KVM_STATE_NESTED_EVMCS 0x00000004 4329 4330 #define KVM_STATE_NESTED_FORMAT_VMX 0 4331 #define KVM_STATE_NESTED_FORMAT_SVM 1 4332 4333 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 4334 4335 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001 4336 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002 4337 4338 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001 4339 4340 struct kvm_vmx_nested_state_hdr { 4341 __u64 vmxon_pa; 4342 __u64 vmcs12_pa; 4343 4344 struct { 4345 __u16 flags; 4346 } smm; 4347 4348 __u32 flags; 4349 __u64 preemption_timer_deadline; 4350 }; 4351 4352 struct kvm_vmx_nested_state_data { 4353 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4354 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4355 }; 4356 4357This ioctl copies the vcpu's nested virtualization state from the kernel to 4358userspace. 4359 4360The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE 4361to the KVM_CHECK_EXTENSION ioctl(). 4362 43634.115 KVM_SET_NESTED_STATE 4364-------------------------- 4365 4366:Capability: KVM_CAP_NESTED_STATE 4367:Architectures: x86 4368:Type: vcpu ioctl 4369:Parameters: struct kvm_nested_state (in) 4370:Returns: 0 on success, -1 on error 4371 4372This copies the vcpu's kvm_nested_state struct from userspace to the kernel. 4373For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE. 4374 43754.116 KVM_(UN)REGISTER_COALESCED_MMIO 4376------------------------------------- 4377 4378:Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio) 4379 KVM_CAP_COALESCED_PIO (for coalesced pio) 4380:Architectures: all 4381:Type: vm ioctl 4382:Parameters: struct kvm_coalesced_mmio_zone 4383:Returns: 0 on success, < 0 on error 4384 4385Coalesced I/O is a performance optimization that defers hardware 4386register write emulation so that userspace exits are avoided. It is 4387typically used to reduce the overhead of emulating frequently accessed 4388hardware registers. 4389 4390When a hardware register is configured for coalesced I/O, write accesses 4391do not exit to userspace and their value is recorded in a ring buffer 4392that is shared between kernel and userspace. 4393 4394Coalesced I/O is used if one or more write accesses to a hardware 4395register can be deferred until a read or a write to another hardware 4396register on the same device. This last access will cause a vmexit and 4397userspace will process accesses from the ring buffer before emulating 4398it. That will avoid exiting to userspace on repeated writes. 4399 4400Coalesced pio is based on coalesced mmio. There is little difference 4401between coalesced mmio and pio except that coalesced pio records accesses 4402to I/O ports. 4403 44044.117 KVM_CLEAR_DIRTY_LOG (vm ioctl) 4405------------------------------------ 4406 4407:Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4408:Architectures: x86, arm, arm64, mips 4409:Type: vm ioctl 4410:Parameters: struct kvm_clear_dirty_log (in) 4411:Returns: 0 on success, -1 on error 4412 4413:: 4414 4415 /* for KVM_CLEAR_DIRTY_LOG */ 4416 struct kvm_clear_dirty_log { 4417 __u32 slot; 4418 __u32 num_pages; 4419 __u64 first_page; 4420 union { 4421 void __user *dirty_bitmap; /* one bit per page */ 4422 __u64 padding; 4423 }; 4424 }; 4425 4426The ioctl clears the dirty status of pages in a memory slot, according to 4427the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap 4428field. Bit 0 of the bitmap corresponds to page "first_page" in the 4429memory slot, and num_pages is the size in bits of the input bitmap. 4430first_page must be a multiple of 64; num_pages must also be a multiple of 443164 unless first_page + num_pages is the size of the memory slot. For each 4432bit that is set in the input bitmap, the corresponding page is marked "clean" 4433in KVM's dirty bitmap, and dirty tracking is re-enabled for that page 4434(for example via write-protection, or by clearing the dirty bit in 4435a page table entry). 4436 4437If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 4438the address space for which you want to clear the dirty status. See 4439KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 4440 4441This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4442is enabled; for more information, see the description of the capability. 4443However, it can always be used as long as KVM_CHECK_EXTENSION confirms 4444that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present. 4445 44464.118 KVM_GET_SUPPORTED_HV_CPUID 4447-------------------------------- 4448 4449:Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system) 4450:Architectures: x86 4451:Type: system ioctl, vcpu ioctl 4452:Parameters: struct kvm_cpuid2 (in/out) 4453:Returns: 0 on success, -1 on error 4454 4455:: 4456 4457 struct kvm_cpuid2 { 4458 __u32 nent; 4459 __u32 padding; 4460 struct kvm_cpuid_entry2 entries[0]; 4461 }; 4462 4463 struct kvm_cpuid_entry2 { 4464 __u32 function; 4465 __u32 index; 4466 __u32 flags; 4467 __u32 eax; 4468 __u32 ebx; 4469 __u32 ecx; 4470 __u32 edx; 4471 __u32 padding[3]; 4472 }; 4473 4474This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in 4475KVM. Userspace can use the information returned by this ioctl to construct 4476cpuid information presented to guests consuming Hyper-V enlightenments (e.g. 4477Windows or Hyper-V guests). 4478 4479CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level 4480Functional Specification (TLFS). These leaves can't be obtained with 4481KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature 4482leaves (0x40000000, 0x40000001). 4483 4484Currently, the following list of CPUID leaves are returned: 4485 4486 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 4487 - HYPERV_CPUID_INTERFACE 4488 - HYPERV_CPUID_VERSION 4489 - HYPERV_CPUID_FEATURES 4490 - HYPERV_CPUID_ENLIGHTMENT_INFO 4491 - HYPERV_CPUID_IMPLEMENT_LIMITS 4492 - HYPERV_CPUID_NESTED_FEATURES 4493 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS 4494 - HYPERV_CPUID_SYNDBG_INTERFACE 4495 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 4496 4497Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure 4498with the 'nent' field indicating the number of entries in the variable-size 4499array 'entries'. If the number of entries is too low to describe all Hyper-V 4500feature leaves, an error (E2BIG) is returned. If the number is more or equal 4501to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the 4502number of valid entries in the 'entries' array, which is then filled. 4503 4504'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved, 4505userspace should not expect to get any particular value there. 4506 4507Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike 4508system ioctl which exposes all supported feature bits unconditionally, vcpu 4509version has the following quirks: 4510 4511- HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED 4512 feature bit are only exposed when Enlightened VMCS was previously enabled 4513 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS). 4514- HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC. 4515 (presumes KVM_CREATE_IRQCHIP has already been called). 4516 45174.119 KVM_ARM_VCPU_FINALIZE 4518--------------------------- 4519 4520:Architectures: arm, arm64 4521:Type: vcpu ioctl 4522:Parameters: int feature (in) 4523:Returns: 0 on success, -1 on error 4524 4525Errors: 4526 4527 ====== ============================================================== 4528 EPERM feature not enabled, needs configuration, or already finalized 4529 EINVAL feature unknown or not present 4530 ====== ============================================================== 4531 4532Recognised values for feature: 4533 4534 ===== =========================================== 4535 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE) 4536 ===== =========================================== 4537 4538Finalizes the configuration of the specified vcpu feature. 4539 4540The vcpu must already have been initialised, enabling the affected feature, by 4541means of a successful KVM_ARM_VCPU_INIT call with the appropriate flag set in 4542features[]. 4543 4544For affected vcpu features, this is a mandatory step that must be performed 4545before the vcpu is fully usable. 4546 4547Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be 4548configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration 4549that should be performaned and how to do it are feature-dependent. 4550 4551Other calls that depend on a particular feature being finalized, such as 4552KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with 4553-EPERM unless the feature has already been finalized by means of a 4554KVM_ARM_VCPU_FINALIZE call. 4555 4556See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization 4557using this ioctl. 4558 45594.120 KVM_SET_PMU_EVENT_FILTER 4560------------------------------ 4561 4562:Capability: KVM_CAP_PMU_EVENT_FILTER 4563:Architectures: x86 4564:Type: vm ioctl 4565:Parameters: struct kvm_pmu_event_filter (in) 4566:Returns: 0 on success, -1 on error 4567 4568:: 4569 4570 struct kvm_pmu_event_filter { 4571 __u32 action; 4572 __u32 nevents; 4573 __u32 fixed_counter_bitmap; 4574 __u32 flags; 4575 __u32 pad[4]; 4576 __u64 events[0]; 4577 }; 4578 4579This ioctl restricts the set of PMU events that the guest can program. 4580The argument holds a list of events which will be allowed or denied. 4581The eventsel+umask of each event the guest attempts to program is compared 4582against the events field to determine whether the guest should have access. 4583The events field only controls general purpose counters; fixed purpose 4584counters are controlled by the fixed_counter_bitmap. 4585 4586No flags are defined yet, the field must be zero. 4587 4588Valid values for 'action':: 4589 4590 #define KVM_PMU_EVENT_ALLOW 0 4591 #define KVM_PMU_EVENT_DENY 1 4592 45934.121 KVM_PPC_SVM_OFF 4594--------------------- 4595 4596:Capability: basic 4597:Architectures: powerpc 4598:Type: vm ioctl 4599:Parameters: none 4600:Returns: 0 on successful completion, 4601 4602Errors: 4603 4604 ====== ================================================================ 4605 EINVAL if ultravisor failed to terminate the secure guest 4606 ENOMEM if hypervisor failed to allocate new radix page tables for guest 4607 ====== ================================================================ 4608 4609This ioctl is used to turn off the secure mode of the guest or transition 4610the guest from secure mode to normal mode. This is invoked when the guest 4611is reset. This has no effect if called for a normal guest. 4612 4613This ioctl issues an ultravisor call to terminate the secure guest, 4614unpins the VPA pages and releases all the device pages that are used to 4615track the secure pages by hypervisor. 4616 46174.122 KVM_S390_NORMAL_RESET 4618--------------------------- 4619 4620:Capability: KVM_CAP_S390_VCPU_RESETS 4621:Architectures: s390 4622:Type: vcpu ioctl 4623:Parameters: none 4624:Returns: 0 4625 4626This ioctl resets VCPU registers and control structures according to 4627the cpu reset definition in the POP (Principles Of Operation). 4628 46294.123 KVM_S390_INITIAL_RESET 4630---------------------------- 4631 4632:Capability: none 4633:Architectures: s390 4634:Type: vcpu ioctl 4635:Parameters: none 4636:Returns: 0 4637 4638This ioctl resets VCPU registers and control structures according to 4639the initial cpu reset definition in the POP. However, the cpu is not 4640put into ESA mode. This reset is a superset of the normal reset. 4641 46424.124 KVM_S390_CLEAR_RESET 4643-------------------------- 4644 4645:Capability: KVM_CAP_S390_VCPU_RESETS 4646:Architectures: s390 4647:Type: vcpu ioctl 4648:Parameters: none 4649:Returns: 0 4650 4651This ioctl resets VCPU registers and control structures according to 4652the clear cpu reset definition in the POP. However, the cpu is not put 4653into ESA mode. This reset is a superset of the initial reset. 4654 4655 46564.125 KVM_S390_PV_COMMAND 4657------------------------- 4658 4659:Capability: KVM_CAP_S390_PROTECTED 4660:Architectures: s390 4661:Type: vm ioctl 4662:Parameters: struct kvm_pv_cmd 4663:Returns: 0 on success, < 0 on error 4664 4665:: 4666 4667 struct kvm_pv_cmd { 4668 __u32 cmd; /* Command to be executed */ 4669 __u16 rc; /* Ultravisor return code */ 4670 __u16 rrc; /* Ultravisor return reason code */ 4671 __u64 data; /* Data or address */ 4672 __u32 flags; /* flags for future extensions. Must be 0 for now */ 4673 __u32 reserved[3]; 4674 }; 4675 4676cmd values: 4677 4678KVM_PV_ENABLE 4679 Allocate memory and register the VM with the Ultravisor, thereby 4680 donating memory to the Ultravisor that will become inaccessible to 4681 KVM. All existing CPUs are converted to protected ones. After this 4682 command has succeeded, any CPU added via hotplug will become 4683 protected during its creation as well. 4684 4685 Errors: 4686 4687 ===== ============================= 4688 EINTR an unmasked signal is pending 4689 ===== ============================= 4690 4691KVM_PV_DISABLE 4692 4693 Deregister the VM from the Ultravisor and reclaim the memory that 4694 had been donated to the Ultravisor, making it usable by the kernel 4695 again. All registered VCPUs are converted back to non-protected 4696 ones. 4697 4698KVM_PV_VM_SET_SEC_PARMS 4699 Pass the image header from VM memory to the Ultravisor in 4700 preparation of image unpacking and verification. 4701 4702KVM_PV_VM_UNPACK 4703 Unpack (protect and decrypt) a page of the encrypted boot image. 4704 4705KVM_PV_VM_VERIFY 4706 Verify the integrity of the unpacked image. Only if this succeeds, 4707 KVM is allowed to start protected VCPUs. 4708 47094.126 KVM_X86_SET_MSR_FILTER 4710---------------------------- 4711 4712:Capability: KVM_X86_SET_MSR_FILTER 4713:Architectures: x86 4714:Type: vm ioctl 4715:Parameters: struct kvm_msr_filter 4716:Returns: 0 on success, < 0 on error 4717 4718:: 4719 4720 struct kvm_msr_filter_range { 4721 #define KVM_MSR_FILTER_READ (1 << 0) 4722 #define KVM_MSR_FILTER_WRITE (1 << 1) 4723 __u32 flags; 4724 __u32 nmsrs; /* number of msrs in bitmap */ 4725 __u32 base; /* MSR index the bitmap starts at */ 4726 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 4727 }; 4728 4729 #define KVM_MSR_FILTER_MAX_RANGES 16 4730 struct kvm_msr_filter { 4731 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 4732 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 4733 __u32 flags; 4734 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 4735 }; 4736 4737flags values for ``struct kvm_msr_filter_range``: 4738 4739``KVM_MSR_FILTER_READ`` 4740 4741 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 4742 indicates that a read should immediately fail, while a 1 indicates that 4743 a read for a particular MSR should be handled regardless of the default 4744 filter action. 4745 4746``KVM_MSR_FILTER_WRITE`` 4747 4748 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 4749 indicates that a write should immediately fail, while a 1 indicates that 4750 a write for a particular MSR should be handled regardless of the default 4751 filter action. 4752 4753``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE`` 4754 4755 Filter both read and write accesses to MSRs using the given bitmap. A 0 4756 in the bitmap indicates that both reads and writes should immediately fail, 4757 while a 1 indicates that reads and writes for a particular MSR are not 4758 filtered by this range. 4759 4760flags values for ``struct kvm_msr_filter``: 4761 4762``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4763 4764 If no filter range matches an MSR index that is getting accessed, KVM will 4765 fall back to allowing access to the MSR. 4766 4767``KVM_MSR_FILTER_DEFAULT_DENY`` 4768 4769 If no filter range matches an MSR index that is getting accessed, KVM will 4770 fall back to rejecting access to the MSR. In this mode, all MSRs that should 4771 be processed by KVM need to explicitly be marked as allowed in the bitmaps. 4772 4773This ioctl allows user space to define up to 16 bitmaps of MSR ranges to 4774specify whether a certain MSR access should be explicitly filtered for or not. 4775 4776If this ioctl has never been invoked, MSR accesses are not guarded and the 4777default KVM in-kernel emulation behavior is fully preserved. 4778 4779Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 4780filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 4781an error. 4782 4783As soon as the filtering is in place, every MSR access is processed through 4784the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff); 4785x2APIC MSRs are always allowed, independent of the ``default_allow`` setting, 4786and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base 4787register. 4788 4789If a bit is within one of the defined ranges, read and write accesses are 4790guarded by the bitmap's value for the MSR index if the kind of access 4791is included in the ``struct kvm_msr_filter_range`` flags. If no range 4792cover this particular access, the behavior is determined by the flags 4793field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4794and ``KVM_MSR_FILTER_DEFAULT_DENY``. 4795 4796Each bitmap range specifies a range of MSRs to potentially allow access on. 4797The range goes from MSR index [base .. base+nmsrs]. The flags field 4798indicates whether reads, writes or both reads and writes are filtered 4799by setting a 1 bit in the bitmap for the corresponding MSR index. 4800 4801If an MSR access is not permitted through the filtering, it generates a 4802#GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that 4803allows user space to deflect and potentially handle various MSR accesses 4804into user space. 4805 4806If a vCPU is in running state while this ioctl is invoked, the vCPU may 4807experience inconsistent filtering behavior on MSR accesses. 4808 48094.127 KVM_XEN_HVM_SET_ATTR 4810-------------------------- 4811 4812:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 4813:Architectures: x86 4814:Type: vm ioctl 4815:Parameters: struct kvm_xen_hvm_attr 4816:Returns: 0 on success, < 0 on error 4817 4818:: 4819 4820 struct kvm_xen_hvm_attr { 4821 __u16 type; 4822 __u16 pad[3]; 4823 union { 4824 __u8 long_mode; 4825 __u8 vector; 4826 struct { 4827 __u64 gfn; 4828 } shared_info; 4829 __u64 pad[4]; 4830 } u; 4831 }; 4832 4833type values: 4834 4835KVM_XEN_ATTR_TYPE_LONG_MODE 4836 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This 4837 determines the layout of the shared info pages exposed to the VM. 4838 4839KVM_XEN_ATTR_TYPE_SHARED_INFO 4840 Sets the guest physical frame number at which the Xen "shared info" 4841 page resides. Note that although Xen places vcpu_info for the first 4842 32 vCPUs in the shared_info page, KVM does not automatically do so 4843 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO be used 4844 explicitly even when the vcpu_info for a given vCPU resides at the 4845 "default" location in the shared_info page. This is because KVM is 4846 not aware of the Xen CPU id which is used as the index into the 4847 vcpu_info[] array, so cannot know the correct default location. 4848 4849KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 4850 Sets the exception vector used to deliver Xen event channel upcalls. 4851 48524.128 KVM_XEN_HVM_GET_ATTR 4853-------------------------- 4854 4855:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 4856:Architectures: x86 4857:Type: vm ioctl 4858:Parameters: struct kvm_xen_hvm_attr 4859:Returns: 0 on success, < 0 on error 4860 4861Allows Xen VM attributes to be read. For the structure and types, 4862see KVM_XEN_HVM_SET_ATTR above. 4863 48644.129 KVM_XEN_VCPU_SET_ATTR 4865--------------------------- 4866 4867:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 4868:Architectures: x86 4869:Type: vcpu ioctl 4870:Parameters: struct kvm_xen_vcpu_attr 4871:Returns: 0 on success, < 0 on error 4872 4873:: 4874 4875 struct kvm_xen_vcpu_attr { 4876 __u16 type; 4877 __u16 pad[3]; 4878 union { 4879 __u64 gpa; 4880 __u64 pad[4]; 4881 struct { 4882 __u64 state; 4883 __u64 state_entry_time; 4884 __u64 time_running; 4885 __u64 time_runnable; 4886 __u64 time_blocked; 4887 __u64 time_offline; 4888 } runstate; 4889 } u; 4890 }; 4891 4892type values: 4893 4894KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 4895 Sets the guest physical address of the vcpu_info for a given vCPU. 4896 4897KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 4898 Sets the guest physical address of an additional pvclock structure 4899 for a given vCPU. This is typically used for guest vsyscall support. 4900 4901KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 4902 Sets the guest physical address of the vcpu_runstate_info for a given 4903 vCPU. This is how a Xen guest tracks CPU state such as steal time. 4904 4905KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 4906 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of 4907 the given vCPU from the .u.runstate.state member of the structure. 4908 KVM automatically accounts running and runnable time but blocked 4909 and offline states are only entered explicitly. 4910 4911KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 4912 Sets all fields of the vCPU runstate data from the .u.runstate member 4913 of the structure, including the current runstate. The state_entry_time 4914 must equal the sum of the other four times. 4915 4916KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 4917 This *adds* the contents of the .u.runstate members of the structure 4918 to the corresponding members of the given vCPU's runstate data, thus 4919 permitting atomic adjustments to the runstate times. The adjustment 4920 to the state_entry_time must equal the sum of the adjustments to the 4921 other four times. The state field must be set to -1, or to a valid 4922 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked 4923 or RUNSTATE_offline) to set the current accounted state as of the 4924 adjusted state_entry_time. 4925 49264.130 KVM_XEN_VCPU_GET_ATTR 4927--------------------------- 4928 4929:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 4930:Architectures: x86 4931:Type: vcpu ioctl 4932:Parameters: struct kvm_xen_vcpu_attr 4933:Returns: 0 on success, < 0 on error 4934 4935Allows Xen vCPU attributes to be read. For the structure and types, 4936see KVM_XEN_VCPU_SET_ATTR above. 4937 4938The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used 4939with the KVM_XEN_VCPU_GET_ATTR ioctl. 4940 49415. The kvm_run structure 4942======================== 4943 4944Application code obtains a pointer to the kvm_run structure by 4945mmap()ing a vcpu fd. From that point, application code can control 4946execution by changing fields in kvm_run prior to calling the KVM_RUN 4947ioctl, and obtain information about the reason KVM_RUN returned by 4948looking up structure members. 4949 4950:: 4951 4952 struct kvm_run { 4953 /* in */ 4954 __u8 request_interrupt_window; 4955 4956Request that KVM_RUN return when it becomes possible to inject external 4957interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 4958 4959:: 4960 4961 __u8 immediate_exit; 4962 4963This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 4964exits immediately, returning -EINTR. In the common scenario where a 4965signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 4966to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 4967Rather than blocking the signal outside KVM_RUN, userspace can set up 4968a signal handler that sets run->immediate_exit to a non-zero value. 4969 4970This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 4971 4972:: 4973 4974 __u8 padding1[6]; 4975 4976 /* out */ 4977 __u32 exit_reason; 4978 4979When KVM_RUN has returned successfully (return value 0), this informs 4980application code why KVM_RUN has returned. Allowable values for this 4981field are detailed below. 4982 4983:: 4984 4985 __u8 ready_for_interrupt_injection; 4986 4987If request_interrupt_window has been specified, this field indicates 4988an interrupt can be injected now with KVM_INTERRUPT. 4989 4990:: 4991 4992 __u8 if_flag; 4993 4994The value of the current interrupt flag. Only valid if in-kernel 4995local APIC is not used. 4996 4997:: 4998 4999 __u16 flags; 5000 5001More architecture-specific flags detailing state of the VCPU that may 5002affect the device's behavior. Current defined flags:: 5003 5004 /* x86, set if the VCPU is in system management mode */ 5005 #define KVM_RUN_X86_SMM (1 << 0) 5006 /* x86, set if bus lock detected in VM */ 5007 #define KVM_RUN_BUS_LOCK (1 << 1) 5008 5009:: 5010 5011 /* in (pre_kvm_run), out (post_kvm_run) */ 5012 __u64 cr8; 5013 5014The value of the cr8 register. Only valid if in-kernel local APIC is 5015not used. Both input and output. 5016 5017:: 5018 5019 __u64 apic_base; 5020 5021The value of the APIC BASE msr. Only valid if in-kernel local 5022APIC is not used. Both input and output. 5023 5024:: 5025 5026 union { 5027 /* KVM_EXIT_UNKNOWN */ 5028 struct { 5029 __u64 hardware_exit_reason; 5030 } hw; 5031 5032If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 5033reasons. Further architecture-specific information is available in 5034hardware_exit_reason. 5035 5036:: 5037 5038 /* KVM_EXIT_FAIL_ENTRY */ 5039 struct { 5040 __u64 hardware_entry_failure_reason; 5041 __u32 cpu; /* if KVM_LAST_CPU */ 5042 } fail_entry; 5043 5044If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 5045to unknown reasons. Further architecture-specific information is 5046available in hardware_entry_failure_reason. 5047 5048:: 5049 5050 /* KVM_EXIT_EXCEPTION */ 5051 struct { 5052 __u32 exception; 5053 __u32 error_code; 5054 } ex; 5055 5056Unused. 5057 5058:: 5059 5060 /* KVM_EXIT_IO */ 5061 struct { 5062 #define KVM_EXIT_IO_IN 0 5063 #define KVM_EXIT_IO_OUT 1 5064 __u8 direction; 5065 __u8 size; /* bytes */ 5066 __u16 port; 5067 __u32 count; 5068 __u64 data_offset; /* relative to kvm_run start */ 5069 } io; 5070 5071If exit_reason is KVM_EXIT_IO, then the vcpu has 5072executed a port I/O instruction which could not be satisfied by kvm. 5073data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 5074where kvm expects application code to place the data for the next 5075KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 5076 5077:: 5078 5079 /* KVM_EXIT_DEBUG */ 5080 struct { 5081 struct kvm_debug_exit_arch arch; 5082 } debug; 5083 5084If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 5085for which architecture specific information is returned. 5086 5087:: 5088 5089 /* KVM_EXIT_MMIO */ 5090 struct { 5091 __u64 phys_addr; 5092 __u8 data[8]; 5093 __u32 len; 5094 __u8 is_write; 5095 } mmio; 5096 5097If exit_reason is KVM_EXIT_MMIO, then the vcpu has 5098executed a memory-mapped I/O instruction which could not be satisfied 5099by kvm. The 'data' member contains the written data if 'is_write' is 5100true, and should be filled by application code otherwise. 5101 5102The 'data' member contains, in its first 'len' bytes, the value as it would 5103appear if the VCPU performed a load or store of the appropriate width directly 5104to the byte array. 5105 5106.. note:: 5107 5108 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN, 5109 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding 5110 operations are complete (and guest state is consistent) only after userspace 5111 has re-entered the kernel with KVM_RUN. The kernel side will first finish 5112 incomplete operations and then check for pending signals. 5113 5114 The pending state of the operation is not preserved in state which is 5115 visible to userspace, thus userspace should ensure that the operation is 5116 completed before performing a live migration. Userspace can re-enter the 5117 guest with an unmasked signal pending or with the immediate_exit field set 5118 to complete pending operations without allowing any further instructions 5119 to be executed. 5120 5121:: 5122 5123 /* KVM_EXIT_HYPERCALL */ 5124 struct { 5125 __u64 nr; 5126 __u64 args[6]; 5127 __u64 ret; 5128 __u32 longmode; 5129 __u32 pad; 5130 } hypercall; 5131 5132Unused. This was once used for 'hypercall to userspace'. To implement 5133such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). 5134 5135.. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 5136 5137:: 5138 5139 /* KVM_EXIT_TPR_ACCESS */ 5140 struct { 5141 __u64 rip; 5142 __u32 is_write; 5143 __u32 pad; 5144 } tpr_access; 5145 5146To be documented (KVM_TPR_ACCESS_REPORTING). 5147 5148:: 5149 5150 /* KVM_EXIT_S390_SIEIC */ 5151 struct { 5152 __u8 icptcode; 5153 __u64 mask; /* psw upper half */ 5154 __u64 addr; /* psw lower half */ 5155 __u16 ipa; 5156 __u32 ipb; 5157 } s390_sieic; 5158 5159s390 specific. 5160 5161:: 5162 5163 /* KVM_EXIT_S390_RESET */ 5164 #define KVM_S390_RESET_POR 1 5165 #define KVM_S390_RESET_CLEAR 2 5166 #define KVM_S390_RESET_SUBSYSTEM 4 5167 #define KVM_S390_RESET_CPU_INIT 8 5168 #define KVM_S390_RESET_IPL 16 5169 __u64 s390_reset_flags; 5170 5171s390 specific. 5172 5173:: 5174 5175 /* KVM_EXIT_S390_UCONTROL */ 5176 struct { 5177 __u64 trans_exc_code; 5178 __u32 pgm_code; 5179 } s390_ucontrol; 5180 5181s390 specific. A page fault has occurred for a user controlled virtual 5182machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be 5183resolved by the kernel. 5184The program code and the translation exception code that were placed 5185in the cpu's lowcore are presented here as defined by the z Architecture 5186Principles of Operation Book in the Chapter for Dynamic Address Translation 5187(DAT) 5188 5189:: 5190 5191 /* KVM_EXIT_DCR */ 5192 struct { 5193 __u32 dcrn; 5194 __u32 data; 5195 __u8 is_write; 5196 } dcr; 5197 5198Deprecated - was used for 440 KVM. 5199 5200:: 5201 5202 /* KVM_EXIT_OSI */ 5203 struct { 5204 __u64 gprs[32]; 5205 } osi; 5206 5207MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 5208hypercalls and exit with this exit struct that contains all the guest gprs. 5209 5210If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 5211Userspace can now handle the hypercall and when it's done modify the gprs as 5212necessary. Upon guest entry all guest GPRs will then be replaced by the values 5213in this struct. 5214 5215:: 5216 5217 /* KVM_EXIT_PAPR_HCALL */ 5218 struct { 5219 __u64 nr; 5220 __u64 ret; 5221 __u64 args[9]; 5222 } papr_hcall; 5223 5224This is used on 64-bit PowerPC when emulating a pSeries partition, 5225e.g. with the 'pseries' machine type in qemu. It occurs when the 5226guest does a hypercall using the 'sc 1' instruction. The 'nr' field 5227contains the hypercall number (from the guest R3), and 'args' contains 5228the arguments (from the guest R4 - R12). Userspace should put the 5229return code in 'ret' and any extra returned values in args[]. 5230The possible hypercalls are defined in the Power Architecture Platform 5231Requirements (PAPR) document available from www.power.org (free 5232developer registration required to access it). 5233 5234:: 5235 5236 /* KVM_EXIT_S390_TSCH */ 5237 struct { 5238 __u16 subchannel_id; 5239 __u16 subchannel_nr; 5240 __u32 io_int_parm; 5241 __u32 io_int_word; 5242 __u32 ipb; 5243 __u8 dequeued; 5244 } s390_tsch; 5245 5246s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 5247and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 5248interrupt for the target subchannel has been dequeued and subchannel_id, 5249subchannel_nr, io_int_parm and io_int_word contain the parameters for that 5250interrupt. ipb is needed for instruction parameter decoding. 5251 5252:: 5253 5254 /* KVM_EXIT_EPR */ 5255 struct { 5256 __u32 epr; 5257 } epr; 5258 5259On FSL BookE PowerPC chips, the interrupt controller has a fast patch 5260interrupt acknowledge path to the core. When the core successfully 5261delivers an interrupt, it automatically populates the EPR register with 5262the interrupt vector number and acknowledges the interrupt inside 5263the interrupt controller. 5264 5265In case the interrupt controller lives in user space, we need to do 5266the interrupt acknowledge cycle through it to fetch the next to be 5267delivered interrupt vector using this exit. 5268 5269It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 5270external interrupt has just been delivered into the guest. User space 5271should put the acknowledged interrupt vector into the 'epr' field. 5272 5273:: 5274 5275 /* KVM_EXIT_SYSTEM_EVENT */ 5276 struct { 5277 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 5278 #define KVM_SYSTEM_EVENT_RESET 2 5279 #define KVM_SYSTEM_EVENT_CRASH 3 5280 __u32 type; 5281 __u64 flags; 5282 } system_event; 5283 5284If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 5285a system-level event using some architecture specific mechanism (hypercall 5286or some special instruction). In case of ARM/ARM64, this is triggered using 5287HVC instruction based PSCI call from the vcpu. The 'type' field describes 5288the system-level event type. The 'flags' field describes architecture 5289specific flags for the system-level event. 5290 5291Valid values for 'type' are: 5292 5293 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 5294 VM. Userspace is not obliged to honour this, and if it does honour 5295 this does not need to destroy the VM synchronously (ie it may call 5296 KVM_RUN again before shutdown finally occurs). 5297 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 5298 As with SHUTDOWN, userspace can choose to ignore the request, or 5299 to schedule the reset to occur in the future and may call KVM_RUN again. 5300 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 5301 has requested a crash condition maintenance. Userspace can choose 5302 to ignore the request, or to gather VM memory core dump and/or 5303 reset/shutdown of the VM. 5304 5305:: 5306 5307 /* KVM_EXIT_IOAPIC_EOI */ 5308 struct { 5309 __u8 vector; 5310 } eoi; 5311 5312Indicates that the VCPU's in-kernel local APIC received an EOI for a 5313level-triggered IOAPIC interrupt. This exit only triggers when the 5314IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 5315the userspace IOAPIC should process the EOI and retrigger the interrupt if 5316it is still asserted. Vector is the LAPIC interrupt vector for which the 5317EOI was received. 5318 5319:: 5320 5321 struct kvm_hyperv_exit { 5322 #define KVM_EXIT_HYPERV_SYNIC 1 5323 #define KVM_EXIT_HYPERV_HCALL 2 5324 #define KVM_EXIT_HYPERV_SYNDBG 3 5325 __u32 type; 5326 __u32 pad1; 5327 union { 5328 struct { 5329 __u32 msr; 5330 __u32 pad2; 5331 __u64 control; 5332 __u64 evt_page; 5333 __u64 msg_page; 5334 } synic; 5335 struct { 5336 __u64 input; 5337 __u64 result; 5338 __u64 params[2]; 5339 } hcall; 5340 struct { 5341 __u32 msr; 5342 __u32 pad2; 5343 __u64 control; 5344 __u64 status; 5345 __u64 send_page; 5346 __u64 recv_page; 5347 __u64 pending_page; 5348 } syndbg; 5349 } u; 5350 }; 5351 /* KVM_EXIT_HYPERV */ 5352 struct kvm_hyperv_exit hyperv; 5353 5354Indicates that the VCPU exits into userspace to process some tasks 5355related to Hyper-V emulation. 5356 5357Valid values for 'type' are: 5358 5359 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 5360 5361Hyper-V SynIC state change. Notification is used to remap SynIC 5362event/message pages and to enable/disable SynIC messages/events processing 5363in userspace. 5364 5365 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about 5366 5367Hyper-V Synthetic debugger state change. Notification is used to either update 5368the pending_page location or to send a control command (send the buffer located 5369in send_page or recv a buffer to recv_page). 5370 5371:: 5372 5373 /* KVM_EXIT_ARM_NISV */ 5374 struct { 5375 __u64 esr_iss; 5376 __u64 fault_ipa; 5377 } arm_nisv; 5378 5379Used on arm and arm64 systems. If a guest accesses memory not in a memslot, 5380KVM will typically return to userspace and ask it to do MMIO emulation on its 5381behalf. However, for certain classes of instructions, no instruction decode 5382(direction, length of memory access) is provided, and fetching and decoding 5383the instruction from the VM is overly complicated to live in the kernel. 5384 5385Historically, when this situation occurred, KVM would print a warning and kill 5386the VM. KVM assumed that if the guest accessed non-memslot memory, it was 5387trying to do I/O, which just couldn't be emulated, and the warning message was 5388phrased accordingly. However, what happened more often was that a guest bug 5389caused access outside the guest memory areas which should lead to a more 5390meaningful warning message and an external abort in the guest, if the access 5391did not fall within an I/O window. 5392 5393Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable 5394this capability at VM creation. Once this is done, these types of errors will 5395instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from 5396the HSR (arm) and ESR_EL2 (arm64) in the esr_iss field, and the faulting IPA 5397in the fault_ipa field. Userspace can either fix up the access if it's 5398actually an I/O access by decoding the instruction from guest memory (if it's 5399very brave) and continue executing the guest, or it can decide to suspend, 5400dump, or restart the guest. 5401 5402Note that KVM does not skip the faulting instruction as it does for 5403KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state 5404if it decides to decode and emulate the instruction. 5405 5406:: 5407 5408 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */ 5409 struct { 5410 __u8 error; /* user -> kernel */ 5411 __u8 pad[7]; 5412 __u32 reason; /* kernel -> user */ 5413 __u32 index; /* kernel -> user */ 5414 __u64 data; /* kernel <-> user */ 5415 } msr; 5416 5417Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is 5418enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code 5419will instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR 5420exit for writes. 5421 5422The "reason" field specifies why the MSR trap occurred. User space will only 5423receive MSR exit traps when a particular reason was requested during through 5424ENABLE_CAP. Currently valid exit reasons are: 5425 5426 KVM_MSR_EXIT_REASON_UNKNOWN - access to MSR that is unknown to KVM 5427 KVM_MSR_EXIT_REASON_INVAL - access to invalid MSRs or reserved bits 5428 KVM_MSR_EXIT_REASON_FILTER - access blocked by KVM_X86_SET_MSR_FILTER 5429 5430For KVM_EXIT_X86_RDMSR, the "index" field tells user space which MSR the guest 5431wants to read. To respond to this request with a successful read, user space 5432writes the respective data into the "data" field and must continue guest 5433execution to ensure the read data is transferred into guest register state. 5434 5435If the RDMSR request was unsuccessful, user space indicates that with a "1" in 5436the "error" field. This will inject a #GP into the guest when the VCPU is 5437executed again. 5438 5439For KVM_EXIT_X86_WRMSR, the "index" field tells user space which MSR the guest 5440wants to write. Once finished processing the event, user space must continue 5441vCPU execution. If the MSR write was unsuccessful, user space also sets the 5442"error" field to "1". 5443 5444:: 5445 5446 5447 struct kvm_xen_exit { 5448 #define KVM_EXIT_XEN_HCALL 1 5449 __u32 type; 5450 union { 5451 struct { 5452 __u32 longmode; 5453 __u32 cpl; 5454 __u64 input; 5455 __u64 result; 5456 __u64 params[6]; 5457 } hcall; 5458 } u; 5459 }; 5460 /* KVM_EXIT_XEN */ 5461 struct kvm_hyperv_exit xen; 5462 5463Indicates that the VCPU exits into userspace to process some tasks 5464related to Xen emulation. 5465 5466Valid values for 'type' are: 5467 5468 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall. 5469 Userspace is expected to place the hypercall result into the appropriate 5470 field before invoking KVM_RUN again. 5471 5472:: 5473 5474 /* Fix the size of the union. */ 5475 char padding[256]; 5476 }; 5477 5478 /* 5479 * shared registers between kvm and userspace. 5480 * kvm_valid_regs specifies the register classes set by the host 5481 * kvm_dirty_regs specified the register classes dirtied by userspace 5482 * struct kvm_sync_regs is architecture specific, as well as the 5483 * bits for kvm_valid_regs and kvm_dirty_regs 5484 */ 5485 __u64 kvm_valid_regs; 5486 __u64 kvm_dirty_regs; 5487 union { 5488 struct kvm_sync_regs regs; 5489 char padding[SYNC_REGS_SIZE_BYTES]; 5490 } s; 5491 5492If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 5493certain guest registers without having to call SET/GET_*REGS. Thus we can 5494avoid some system call overhead if userspace has to handle the exit. 5495Userspace can query the validity of the structure by checking 5496kvm_valid_regs for specific bits. These bits are architecture specific 5497and usually define the validity of a groups of registers. (e.g. one bit 5498for general purpose registers) 5499 5500Please note that the kernel is allowed to use the kvm_run structure as the 5501primary storage for certain register types. Therefore, the kernel may use the 5502values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 5503 5504:: 5505 5506 }; 5507 5508 5509 55106. Capabilities that can be enabled on vCPUs 5511============================================ 5512 5513There are certain capabilities that change the behavior of the virtual CPU or 5514the virtual machine when enabled. To enable them, please see section 4.37. 5515Below you can find a list of capabilities and what their effect on the vCPU or 5516the virtual machine is when enabling them. 5517 5518The following information is provided along with the description: 5519 5520 Architectures: 5521 which instruction set architectures provide this ioctl. 5522 x86 includes both i386 and x86_64. 5523 5524 Target: 5525 whether this is a per-vcpu or per-vm capability. 5526 5527 Parameters: 5528 what parameters are accepted by the capability. 5529 5530 Returns: 5531 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 5532 are not detailed, but errors with specific meanings are. 5533 5534 55356.1 KVM_CAP_PPC_OSI 5536------------------- 5537 5538:Architectures: ppc 5539:Target: vcpu 5540:Parameters: none 5541:Returns: 0 on success; -1 on error 5542 5543This capability enables interception of OSI hypercalls that otherwise would 5544be treated as normal system calls to be injected into the guest. OSI hypercalls 5545were invented by Mac-on-Linux to have a standardized communication mechanism 5546between the guest and the host. 5547 5548When this capability is enabled, KVM_EXIT_OSI can occur. 5549 5550 55516.2 KVM_CAP_PPC_PAPR 5552-------------------- 5553 5554:Architectures: ppc 5555:Target: vcpu 5556:Parameters: none 5557:Returns: 0 on success; -1 on error 5558 5559This capability enables interception of PAPR hypercalls. PAPR hypercalls are 5560done using the hypercall instruction "sc 1". 5561 5562It also sets the guest privilege level to "supervisor" mode. Usually the guest 5563runs in "hypervisor" privilege mode with a few missing features. 5564 5565In addition to the above, it changes the semantics of SDR1. In this mode, the 5566HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 5567HTAB invisible to the guest. 5568 5569When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 5570 5571 55726.3 KVM_CAP_SW_TLB 5573------------------ 5574 5575:Architectures: ppc 5576:Target: vcpu 5577:Parameters: args[0] is the address of a struct kvm_config_tlb 5578:Returns: 0 on success; -1 on error 5579 5580:: 5581 5582 struct kvm_config_tlb { 5583 __u64 params; 5584 __u64 array; 5585 __u32 mmu_type; 5586 __u32 array_len; 5587 }; 5588 5589Configures the virtual CPU's TLB array, establishing a shared memory area 5590between userspace and KVM. The "params" and "array" fields are userspace 5591addresses of mmu-type-specific data structures. The "array_len" field is an 5592safety mechanism, and should be set to the size in bytes of the memory that 5593userspace has reserved for the array. It must be at least the size dictated 5594by "mmu_type" and "params". 5595 5596While KVM_RUN is active, the shared region is under control of KVM. Its 5597contents are undefined, and any modification by userspace results in 5598boundedly undefined behavior. 5599 5600On return from KVM_RUN, the shared region will reflect the current state of 5601the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 5602to tell KVM which entries have been changed, prior to calling KVM_RUN again 5603on this vcpu. 5604 5605For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 5606 5607 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 5608 - The "array" field points to an array of type "struct 5609 kvm_book3e_206_tlb_entry". 5610 - The array consists of all entries in the first TLB, followed by all 5611 entries in the second TLB. 5612 - Within a TLB, entries are ordered first by increasing set number. Within a 5613 set, entries are ordered by way (increasing ESEL). 5614 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 5615 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 5616 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 5617 hardware ignores this value for TLB0. 5618 56196.4 KVM_CAP_S390_CSS_SUPPORT 5620---------------------------- 5621 5622:Architectures: s390 5623:Target: vcpu 5624:Parameters: none 5625:Returns: 0 on success; -1 on error 5626 5627This capability enables support for handling of channel I/O instructions. 5628 5629TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 5630handled in-kernel, while the other I/O instructions are passed to userspace. 5631 5632When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 5633SUBCHANNEL intercepts. 5634 5635Note that even though this capability is enabled per-vcpu, the complete 5636virtual machine is affected. 5637 56386.5 KVM_CAP_PPC_EPR 5639------------------- 5640 5641:Architectures: ppc 5642:Target: vcpu 5643:Parameters: args[0] defines whether the proxy facility is active 5644:Returns: 0 on success; -1 on error 5645 5646This capability enables or disables the delivery of interrupts through the 5647external proxy facility. 5648 5649When enabled (args[0] != 0), every time the guest gets an external interrupt 5650delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 5651to receive the topmost interrupt vector. 5652 5653When disabled (args[0] == 0), behavior is as if this facility is unsupported. 5654 5655When this capability is enabled, KVM_EXIT_EPR can occur. 5656 56576.6 KVM_CAP_IRQ_MPIC 5658-------------------- 5659 5660:Architectures: ppc 5661:Parameters: args[0] is the MPIC device fd; 5662 args[1] is the MPIC CPU number for this vcpu 5663 5664This capability connects the vcpu to an in-kernel MPIC device. 5665 56666.7 KVM_CAP_IRQ_XICS 5667-------------------- 5668 5669:Architectures: ppc 5670:Target: vcpu 5671:Parameters: args[0] is the XICS device fd; 5672 args[1] is the XICS CPU number (server ID) for this vcpu 5673 5674This capability connects the vcpu to an in-kernel XICS device. 5675 56766.8 KVM_CAP_S390_IRQCHIP 5677------------------------ 5678 5679:Architectures: s390 5680:Target: vm 5681:Parameters: none 5682 5683This capability enables the in-kernel irqchip for s390. Please refer to 5684"4.24 KVM_CREATE_IRQCHIP" for details. 5685 56866.9 KVM_CAP_MIPS_FPU 5687-------------------- 5688 5689:Architectures: mips 5690:Target: vcpu 5691:Parameters: args[0] is reserved for future use (should be 0). 5692 5693This capability allows the use of the host Floating Point Unit by the guest. It 5694allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 5695done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be 5696accessed (depending on the current guest FPU register mode), and the Status.FR, 5697Config5.FRE bits are accessible via the KVM API and also from the guest, 5698depending on them being supported by the FPU. 5699 57006.10 KVM_CAP_MIPS_MSA 5701--------------------- 5702 5703:Architectures: mips 5704:Target: vcpu 5705:Parameters: args[0] is reserved for future use (should be 0). 5706 5707This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 5708It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 5709Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*`` 5710registers can be accessed, and the Config5.MSAEn bit is accessible via the 5711KVM API and also from the guest. 5712 57136.74 KVM_CAP_SYNC_REGS 5714---------------------- 5715 5716:Architectures: s390, x86 5717:Target: s390: always enabled, x86: vcpu 5718:Parameters: none 5719:Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register 5720 sets are supported 5721 (bitfields defined in arch/x86/include/uapi/asm/kvm.h). 5722 5723As described above in the kvm_sync_regs struct info in section 5 (kvm_run): 5724KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers 5725without having to call SET/GET_*REGS". This reduces overhead by eliminating 5726repeated ioctl calls for setting and/or getting register values. This is 5727particularly important when userspace is making synchronous guest state 5728modifications, e.g. when emulating and/or intercepting instructions in 5729userspace. 5730 5731For s390 specifics, please refer to the source code. 5732 5733For x86: 5734 5735- the register sets to be copied out to kvm_run are selectable 5736 by userspace (rather that all sets being copied out for every exit). 5737- vcpu_events are available in addition to regs and sregs. 5738 5739For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to 5740function as an input bit-array field set by userspace to indicate the 5741specific register sets to be copied out on the next exit. 5742 5743To indicate when userspace has modified values that should be copied into 5744the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set. 5745This is done using the same bitflags as for the 'kvm_valid_regs' field. 5746If the dirty bit is not set, then the register set values will not be copied 5747into the vCPU even if they've been modified. 5748 5749Unused bitfields in the bitarrays must be set to zero. 5750 5751:: 5752 5753 struct kvm_sync_regs { 5754 struct kvm_regs regs; 5755 struct kvm_sregs sregs; 5756 struct kvm_vcpu_events events; 5757 }; 5758 57596.75 KVM_CAP_PPC_IRQ_XIVE 5760------------------------- 5761 5762:Architectures: ppc 5763:Target: vcpu 5764:Parameters: args[0] is the XIVE device fd; 5765 args[1] is the XIVE CPU number (server ID) for this vcpu 5766 5767This capability connects the vcpu to an in-kernel XIVE device. 5768 57697. Capabilities that can be enabled on VMs 5770========================================== 5771 5772There are certain capabilities that change the behavior of the virtual 5773machine when enabled. To enable them, please see section 4.37. Below 5774you can find a list of capabilities and what their effect on the VM 5775is when enabling them. 5776 5777The following information is provided along with the description: 5778 5779 Architectures: 5780 which instruction set architectures provide this ioctl. 5781 x86 includes both i386 and x86_64. 5782 5783 Parameters: 5784 what parameters are accepted by the capability. 5785 5786 Returns: 5787 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 5788 are not detailed, but errors with specific meanings are. 5789 5790 57917.1 KVM_CAP_PPC_ENABLE_HCALL 5792---------------------------- 5793 5794:Architectures: ppc 5795:Parameters: args[0] is the sPAPR hcall number; 5796 args[1] is 0 to disable, 1 to enable in-kernel handling 5797 5798This capability controls whether individual sPAPR hypercalls (hcalls) 5799get handled by the kernel or not. Enabling or disabling in-kernel 5800handling of an hcall is effective across the VM. On creation, an 5801initial set of hcalls are enabled for in-kernel handling, which 5802consists of those hcalls for which in-kernel handlers were implemented 5803before this capability was implemented. If disabled, the kernel will 5804not to attempt to handle the hcall, but will always exit to userspace 5805to handle it. Note that it may not make sense to enable some and 5806disable others of a group of related hcalls, but KVM does not prevent 5807userspace from doing that. 5808 5809If the hcall number specified is not one that has an in-kernel 5810implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 5811error. 5812 58137.2 KVM_CAP_S390_USER_SIGP 5814-------------------------- 5815 5816:Architectures: s390 5817:Parameters: none 5818 5819This capability controls which SIGP orders will be handled completely in user 5820space. With this capability enabled, all fast orders will be handled completely 5821in the kernel: 5822 5823- SENSE 5824- SENSE RUNNING 5825- EXTERNAL CALL 5826- EMERGENCY SIGNAL 5827- CONDITIONAL EMERGENCY SIGNAL 5828 5829All other orders will be handled completely in user space. 5830 5831Only privileged operation exceptions will be checked for in the kernel (or even 5832in the hardware prior to interception). If this capability is not enabled, the 5833old way of handling SIGP orders is used (partially in kernel and user space). 5834 58357.3 KVM_CAP_S390_VECTOR_REGISTERS 5836--------------------------------- 5837 5838:Architectures: s390 5839:Parameters: none 5840:Returns: 0 on success, negative value on error 5841 5842Allows use of the vector registers introduced with z13 processor, and 5843provides for the synchronization between host and user space. Will 5844return -EINVAL if the machine does not support vectors. 5845 58467.4 KVM_CAP_S390_USER_STSI 5847-------------------------- 5848 5849:Architectures: s390 5850:Parameters: none 5851 5852This capability allows post-handlers for the STSI instruction. After 5853initial handling in the kernel, KVM exits to user space with 5854KVM_EXIT_S390_STSI to allow user space to insert further data. 5855 5856Before exiting to userspace, kvm handlers should fill in s390_stsi field of 5857vcpu->run:: 5858 5859 struct { 5860 __u64 addr; 5861 __u8 ar; 5862 __u8 reserved; 5863 __u8 fc; 5864 __u8 sel1; 5865 __u16 sel2; 5866 } s390_stsi; 5867 5868 @addr - guest address of STSI SYSIB 5869 @fc - function code 5870 @sel1 - selector 1 5871 @sel2 - selector 2 5872 @ar - access register number 5873 5874KVM handlers should exit to userspace with rc = -EREMOTE. 5875 58767.5 KVM_CAP_SPLIT_IRQCHIP 5877------------------------- 5878 5879:Architectures: x86 5880:Parameters: args[0] - number of routes reserved for userspace IOAPICs 5881:Returns: 0 on success, -1 on error 5882 5883Create a local apic for each processor in the kernel. This can be used 5884instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 5885IOAPIC and PIC (and also the PIT, even though this has to be enabled 5886separately). 5887 5888This capability also enables in kernel routing of interrupt requests; 5889when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 5890used in the IRQ routing table. The first args[0] MSI routes are reserved 5891for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 5892a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 5893 5894Fails if VCPU has already been created, or if the irqchip is already in the 5895kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 5896 58977.6 KVM_CAP_S390_RI 5898------------------- 5899 5900:Architectures: s390 5901:Parameters: none 5902 5903Allows use of runtime-instrumentation introduced with zEC12 processor. 5904Will return -EINVAL if the machine does not support runtime-instrumentation. 5905Will return -EBUSY if a VCPU has already been created. 5906 59077.7 KVM_CAP_X2APIC_API 5908---------------------- 5909 5910:Architectures: x86 5911:Parameters: args[0] - features that should be enabled 5912:Returns: 0 on success, -EINVAL when args[0] contains invalid features 5913 5914Valid feature flags in args[0] are:: 5915 5916 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 5917 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 5918 5919Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 5920KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 5921allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 5922respective sections. 5923 5924KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 5925in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 5926as a broadcast even in x2APIC mode in order to support physical x2APIC 5927without interrupt remapping. This is undesirable in logical mode, 5928where 0xff represents CPUs 0-7 in cluster 0. 5929 59307.8 KVM_CAP_S390_USER_INSTR0 5931---------------------------- 5932 5933:Architectures: s390 5934:Parameters: none 5935 5936With this capability enabled, all illegal instructions 0x0000 (2 bytes) will 5937be intercepted and forwarded to user space. User space can use this 5938mechanism e.g. to realize 2-byte software breakpoints. The kernel will 5939not inject an operating exception for these instructions, user space has 5940to take care of that. 5941 5942This capability can be enabled dynamically even if VCPUs were already 5943created and are running. 5944 59457.9 KVM_CAP_S390_GS 5946------------------- 5947 5948:Architectures: s390 5949:Parameters: none 5950:Returns: 0 on success; -EINVAL if the machine does not support 5951 guarded storage; -EBUSY if a VCPU has already been created. 5952 5953Allows use of guarded storage for the KVM guest. 5954 59557.10 KVM_CAP_S390_AIS 5956--------------------- 5957 5958:Architectures: s390 5959:Parameters: none 5960 5961Allow use of adapter-interruption suppression. 5962:Returns: 0 on success; -EBUSY if a VCPU has already been created. 5963 59647.11 KVM_CAP_PPC_SMT 5965-------------------- 5966 5967:Architectures: ppc 5968:Parameters: vsmt_mode, flags 5969 5970Enabling this capability on a VM provides userspace with a way to set 5971the desired virtual SMT mode (i.e. the number of virtual CPUs per 5972virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 5973between 1 and 8. On POWER8, vsmt_mode must also be no greater than 5974the number of threads per subcore for the host. Currently flags must 5975be 0. A successful call to enable this capability will result in 5976vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 5977subsequently queried for the VM. This capability is only supported by 5978HV KVM, and can only be set before any VCPUs have been created. 5979The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 5980modes are available. 5981 59827.12 KVM_CAP_PPC_FWNMI 5983---------------------- 5984 5985:Architectures: ppc 5986:Parameters: none 5987 5988With this capability a machine check exception in the guest address 5989space will cause KVM to exit the guest with NMI exit reason. This 5990enables QEMU to build error log and branch to guest kernel registered 5991machine check handling routine. Without this capability KVM will 5992branch to guests' 0x200 interrupt vector. 5993 59947.13 KVM_CAP_X86_DISABLE_EXITS 5995------------------------------ 5996 5997:Architectures: x86 5998:Parameters: args[0] defines which exits are disabled 5999:Returns: 0 on success, -EINVAL when args[0] contains invalid exits 6000 6001Valid bits in args[0] are:: 6002 6003 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) 6004 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1) 6005 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) 6006 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3) 6007 6008Enabling this capability on a VM provides userspace with a way to no 6009longer intercept some instructions for improved latency in some 6010workloads, and is suggested when vCPUs are associated to dedicated 6011physical CPUs. More bits can be added in the future; userspace can 6012just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable 6013all such vmexits. 6014 6015Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits. 6016 60177.14 KVM_CAP_S390_HPAGE_1M 6018-------------------------- 6019 6020:Architectures: s390 6021:Parameters: none 6022:Returns: 0 on success, -EINVAL if hpage module parameter was not set 6023 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 6024 flag set 6025 6026With this capability the KVM support for memory backing with 1m pages 6027through hugetlbfs can be enabled for a VM. After the capability is 6028enabled, cmma can't be enabled anymore and pfmfi and the storage key 6029interpretation are disabled. If cmma has already been enabled or the 6030hpage module parameter is not set to 1, -EINVAL is returned. 6031 6032While it is generally possible to create a huge page backed VM without 6033this capability, the VM will not be able to run. 6034 60357.15 KVM_CAP_MSR_PLATFORM_INFO 6036------------------------------ 6037 6038:Architectures: x86 6039:Parameters: args[0] whether feature should be enabled or not 6040 6041With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise, 6042a #GP would be raised when the guest tries to access. Currently, this 6043capability does not enable write permissions of this MSR for the guest. 6044 60457.16 KVM_CAP_PPC_NESTED_HV 6046-------------------------- 6047 6048:Architectures: ppc 6049:Parameters: none 6050:Returns: 0 on success, -EINVAL when the implementation doesn't support 6051 nested-HV virtualization. 6052 6053HV-KVM on POWER9 and later systems allows for "nested-HV" 6054virtualization, which provides a way for a guest VM to run guests that 6055can run using the CPU's supervisor mode (privileged non-hypervisor 6056state). Enabling this capability on a VM depends on the CPU having 6057the necessary functionality and on the facility being enabled with a 6058kvm-hv module parameter. 6059 60607.17 KVM_CAP_EXCEPTION_PAYLOAD 6061------------------------------ 6062 6063:Architectures: x86 6064:Parameters: args[0] whether feature should be enabled or not 6065 6066With this capability enabled, CR2 will not be modified prior to the 6067emulated VM-exit when L1 intercepts a #PF exception that occurs in 6068L2. Similarly, for kvm-intel only, DR6 will not be modified prior to 6069the emulated VM-exit when L1 intercepts a #DB exception that occurs in 6070L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or 6071#DB) exception for L2, exception.has_payload will be set and the 6072faulting address (or the new DR6 bits*) will be reported in the 6073exception_payload field. Similarly, when userspace injects a #PF (or 6074#DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set 6075exception.has_payload and to put the faulting address - or the new DR6 6076bits\ [#]_ - in the exception_payload field. 6077 6078This capability also enables exception.pending in struct 6079kvm_vcpu_events, which allows userspace to distinguish between pending 6080and injected exceptions. 6081 6082 6083.. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception 6084 will clear DR6.RTM. 6085 60867.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 6087 6088:Architectures: x86, arm, arm64, mips 6089:Parameters: args[0] whether feature should be enabled or not 6090 6091Valid flags are:: 6092 6093 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0) 6094 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1) 6095 6096With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not 6097automatically clear and write-protect all pages that are returned as dirty. 6098Rather, userspace will have to do this operation separately using 6099KVM_CLEAR_DIRTY_LOG. 6100 6101At the cost of a slightly more complicated operation, this provides better 6102scalability and responsiveness for two reasons. First, 6103KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather 6104than requiring to sync a full memslot; this ensures that KVM does not 6105take spinlocks for an extended period of time. Second, in some cases a 6106large amount of time can pass between a call to KVM_GET_DIRTY_LOG and 6107userspace actually using the data in the page. Pages can be modified 6108during this time, which is inefficient for both the guest and userspace: 6109the guest will incur a higher penalty due to write protection faults, 6110while userspace can see false reports of dirty pages. Manual reprotection 6111helps reducing this time, improving guest performance and reducing the 6112number of dirty log false positives. 6113 6114With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap 6115will be initialized to 1 when created. This also improves performance because 6116dirty logging can be enabled gradually in small chunks on the first call 6117to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on 6118KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on 6119x86 and arm64 for now). 6120 6121KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name 6122KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make 6123it hard or impossible to use it correctly. The availability of 6124KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed. 6125Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT. 6126 61277.19 KVM_CAP_PPC_SECURE_GUEST 6128------------------------------ 6129 6130:Architectures: ppc 6131 6132This capability indicates that KVM is running on a host that has 6133ultravisor firmware and thus can support a secure guest. On such a 6134system, a guest can ask the ultravisor to make it a secure guest, 6135one whose memory is inaccessible to the host except for pages which 6136are explicitly requested to be shared with the host. The ultravisor 6137notifies KVM when a guest requests to become a secure guest, and KVM 6138has the opportunity to veto the transition. 6139 6140If present, this capability can be enabled for a VM, meaning that KVM 6141will allow the transition to secure guest mode. Otherwise KVM will 6142veto the transition. 6143 61447.20 KVM_CAP_HALT_POLL 6145---------------------- 6146 6147:Architectures: all 6148:Target: VM 6149:Parameters: args[0] is the maximum poll time in nanoseconds 6150:Returns: 0 on success; -1 on error 6151 6152This capability overrides the kvm module parameter halt_poll_ns for the 6153target VM. 6154 6155VCPU polling allows a VCPU to poll for wakeup events instead of immediately 6156scheduling during guest halts. The maximum time a VCPU can spend polling is 6157controlled by the kvm module parameter halt_poll_ns. This capability allows 6158the maximum halt time to specified on a per-VM basis, effectively overriding 6159the module parameter for the target VM. 6160 61617.21 KVM_CAP_X86_USER_SPACE_MSR 6162------------------------------- 6163 6164:Architectures: x86 6165:Target: VM 6166:Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report 6167:Returns: 0 on success; -1 on error 6168 6169This capability enables trapping of #GP invoking RDMSR and WRMSR instructions 6170into user space. 6171 6172When a guest requests to read or write an MSR, KVM may not implement all MSRs 6173that are relevant to a respective system. It also does not differentiate by 6174CPU type. 6175 6176To allow more fine grained control over MSR handling, user space may enable 6177this capability. With it enabled, MSR accesses that match the mask specified in 6178args[0] and trigger a #GP event inside the guest by KVM will instead trigger 6179KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications which user space 6180can then handle to implement model specific MSR handling and/or user notifications 6181to inform a user that an MSR was not handled. 6182 61837.22 KVM_CAP_X86_BUS_LOCK_EXIT 6184------------------------------- 6185 6186:Architectures: x86 6187:Target: VM 6188:Parameters: args[0] defines the policy used when bus locks detected in guest 6189:Returns: 0 on success, -EINVAL when args[0] contains invalid bits 6190 6191Valid bits in args[0] are:: 6192 6193 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0) 6194 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1) 6195 6196Enabling this capability on a VM provides userspace with a way to select 6197a policy to handle the bus locks detected in guest. Userspace can obtain 6198the supported modes from the result of KVM_CHECK_EXTENSION and define it 6199through the KVM_ENABLE_CAP. 6200 6201KVM_BUS_LOCK_DETECTION_OFF and KVM_BUS_LOCK_DETECTION_EXIT are supported 6202currently and mutually exclusive with each other. More bits can be added in 6203the future. 6204 6205With KVM_BUS_LOCK_DETECTION_OFF set, bus locks in guest will not cause vm exits 6206so that no additional actions are needed. This is the default mode. 6207 6208With KVM_BUS_LOCK_DETECTION_EXIT set, vm exits happen when bus lock detected 6209in VM. KVM just exits to userspace when handling them. Userspace can enforce 6210its own throttling or other policy based mitigations. 6211 6212This capability is aimed to address the thread that VM can exploit bus locks to 6213degree the performance of the whole system. Once the userspace enable this 6214capability and select the KVM_BUS_LOCK_DETECTION_EXIT mode, KVM will set the 6215KVM_RUN_BUS_LOCK flag in vcpu-run->flags field and exit to userspace. Concerning 6216the bus lock vm exit can be preempted by a higher priority VM exit, the exit 6217notifications to userspace can be KVM_EXIT_BUS_LOCK or other reasons. 6218KVM_RUN_BUS_LOCK flag is used to distinguish between them. 6219 62207.23 KVM_CAP_PPC_DAWR1 6221---------------------- 6222 6223:Architectures: ppc 6224:Parameters: none 6225:Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR 6226 6227This capability can be used to check / enable 2nd DAWR feature provided 6228by POWER10 processor. 6229 62308. Other capabilities. 6231====================== 6232 6233This section lists capabilities that give information about other 6234features of the KVM implementation. 6235 62368.1 KVM_CAP_PPC_HWRNG 6237--------------------- 6238 6239:Architectures: ppc 6240 6241This capability, if KVM_CHECK_EXTENSION indicates that it is 6242available, means that the kernel has an implementation of the 6243H_RANDOM hypercall backed by a hardware random-number generator. 6244If present, the kernel H_RANDOM handler can be enabled for guest use 6245with the KVM_CAP_PPC_ENABLE_HCALL capability. 6246 62478.2 KVM_CAP_HYPERV_SYNIC 6248------------------------ 6249 6250:Architectures: x86 6251 6252This capability, if KVM_CHECK_EXTENSION indicates that it is 6253available, means that the kernel has an implementation of the 6254Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 6255used to support Windows Hyper-V based guest paravirt drivers(VMBus). 6256 6257In order to use SynIC, it has to be activated by setting this 6258capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 6259will disable the use of APIC hardware virtualization even if supported 6260by the CPU, as it's incompatible with SynIC auto-EOI behavior. 6261 62628.3 KVM_CAP_PPC_RADIX_MMU 6263------------------------- 6264 6265:Architectures: ppc 6266 6267This capability, if KVM_CHECK_EXTENSION indicates that it is 6268available, means that the kernel can support guests using the 6269radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 6270processor). 6271 62728.4 KVM_CAP_PPC_HASH_MMU_V3 6273--------------------------- 6274 6275:Architectures: ppc 6276 6277This capability, if KVM_CHECK_EXTENSION indicates that it is 6278available, means that the kernel can support guests using the 6279hashed page table MMU defined in Power ISA V3.00 (as implemented in 6280the POWER9 processor), including in-memory segment tables. 6281 62828.5 KVM_CAP_MIPS_VZ 6283------------------- 6284 6285:Architectures: mips 6286 6287This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 6288it is available, means that full hardware assisted virtualization capabilities 6289of the hardware are available for use through KVM. An appropriate 6290KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 6291utilises it. 6292 6293If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 6294available, it means that the VM is using full hardware assisted virtualization 6295capabilities of the hardware. This is useful to check after creating a VM with 6296KVM_VM_MIPS_DEFAULT. 6297 6298The value returned by KVM_CHECK_EXTENSION should be compared against known 6299values (see below). All other values are reserved. This is to allow for the 6300possibility of other hardware assisted virtualization implementations which 6301may be incompatible with the MIPS VZ ASE. 6302 6303== ========================================================================== 6304 0 The trap & emulate implementation is in use to run guest code in user 6305 mode. Guest virtual memory segments are rearranged to fit the guest in the 6306 user mode address space. 6307 6308 1 The MIPS VZ ASE is in use, providing full hardware assisted 6309 virtualization, including standard guest virtual memory segments. 6310== ========================================================================== 6311 63128.6 KVM_CAP_MIPS_TE 6313------------------- 6314 6315:Architectures: mips 6316 6317This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 6318it is available, means that the trap & emulate implementation is available to 6319run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware 6320assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed 6321to KVM_CREATE_VM to create a VM which utilises it. 6322 6323If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 6324available, it means that the VM is using trap & emulate. 6325 63268.7 KVM_CAP_MIPS_64BIT 6327---------------------- 6328 6329:Architectures: mips 6330 6331This capability indicates the supported architecture type of the guest, i.e. the 6332supported register and address width. 6333 6334The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 6335kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 6336be checked specifically against known values (see below). All other values are 6337reserved. 6338 6339== ======================================================================== 6340 0 MIPS32 or microMIPS32. 6341 Both registers and addresses are 32-bits wide. 6342 It will only be possible to run 32-bit guest code. 6343 6344 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 6345 Registers are 64-bits wide, but addresses are 32-bits wide. 6346 64-bit guest code may run but cannot access MIPS64 memory segments. 6347 It will also be possible to run 32-bit guest code. 6348 6349 2 MIPS64 or microMIPS64 with access to all address segments. 6350 Both registers and addresses are 64-bits wide. 6351 It will be possible to run 64-bit or 32-bit guest code. 6352== ======================================================================== 6353 63548.9 KVM_CAP_ARM_USER_IRQ 6355------------------------ 6356 6357:Architectures: arm, arm64 6358 6359This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 6360that if userspace creates a VM without an in-kernel interrupt controller, it 6361will be notified of changes to the output level of in-kernel emulated devices, 6362which can generate virtual interrupts, presented to the VM. 6363For such VMs, on every return to userspace, the kernel 6364updates the vcpu's run->s.regs.device_irq_level field to represent the actual 6365output level of the device. 6366 6367Whenever kvm detects a change in the device output level, kvm guarantees at 6368least one return to userspace before running the VM. This exit could either 6369be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 6370userspace can always sample the device output level and re-compute the state of 6371the userspace interrupt controller. Userspace should always check the state 6372of run->s.regs.device_irq_level on every kvm exit. 6373The value in run->s.regs.device_irq_level can represent both level and edge 6374triggered interrupt signals, depending on the device. Edge triggered interrupt 6375signals will exit to userspace with the bit in run->s.regs.device_irq_level 6376set exactly once per edge signal. 6377 6378The field run->s.regs.device_irq_level is available independent of 6379run->kvm_valid_regs or run->kvm_dirty_regs bits. 6380 6381If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 6382number larger than 0 indicating the version of this capability is implemented 6383and thereby which bits in run->s.regs.device_irq_level can signal values. 6384 6385Currently the following bits are defined for the device_irq_level bitmap:: 6386 6387 KVM_CAP_ARM_USER_IRQ >= 1: 6388 6389 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 6390 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 6391 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 6392 6393Future versions of kvm may implement additional events. These will get 6394indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 6395listed above. 6396 63978.10 KVM_CAP_PPC_SMT_POSSIBLE 6398----------------------------- 6399 6400:Architectures: ppc 6401 6402Querying this capability returns a bitmap indicating the possible 6403virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 6404(counting from the right) is set, then a virtual SMT mode of 2^N is 6405available. 6406 64078.11 KVM_CAP_HYPERV_SYNIC2 6408-------------------------- 6409 6410:Architectures: x86 6411 6412This capability enables a newer version of Hyper-V Synthetic interrupt 6413controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 6414doesn't clear SynIC message and event flags pages when they are enabled by 6415writing to the respective MSRs. 6416 64178.12 KVM_CAP_HYPERV_VP_INDEX 6418---------------------------- 6419 6420:Architectures: x86 6421 6422This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 6423value is used to denote the target vcpu for a SynIC interrupt. For 6424compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this 6425capability is absent, userspace can still query this msr's value. 6426 64278.13 KVM_CAP_S390_AIS_MIGRATION 6428------------------------------- 6429 6430:Architectures: s390 6431:Parameters: none 6432 6433This capability indicates if the flic device will be able to get/set the 6434AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 6435to discover this without having to create a flic device. 6436 64378.14 KVM_CAP_S390_PSW 6438--------------------- 6439 6440:Architectures: s390 6441 6442This capability indicates that the PSW is exposed via the kvm_run structure. 6443 64448.15 KVM_CAP_S390_GMAP 6445---------------------- 6446 6447:Architectures: s390 6448 6449This capability indicates that the user space memory used as guest mapping can 6450be anywhere in the user memory address space, as long as the memory slots are 6451aligned and sized to a segment (1MB) boundary. 6452 64538.16 KVM_CAP_S390_COW 6454--------------------- 6455 6456:Architectures: s390 6457 6458This capability indicates that the user space memory used as guest mapping can 6459use copy-on-write semantics as well as dirty pages tracking via read-only page 6460tables. 6461 64628.17 KVM_CAP_S390_BPB 6463--------------------- 6464 6465:Architectures: s390 6466 6467This capability indicates that kvm will implement the interfaces to handle 6468reset, migration and nested KVM for branch prediction blocking. The stfle 6469facility 82 should not be provided to the guest without this capability. 6470 64718.18 KVM_CAP_HYPERV_TLBFLUSH 6472---------------------------- 6473 6474:Architectures: x86 6475 6476This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush 6477hypercalls: 6478HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx, 6479HvFlushVirtualAddressList, HvFlushVirtualAddressListEx. 6480 64818.19 KVM_CAP_ARM_INJECT_SERROR_ESR 6482---------------------------------- 6483 6484:Architectures: arm, arm64 6485 6486This capability indicates that userspace can specify (via the 6487KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it 6488takes a virtual SError interrupt exception. 6489If KVM advertises this capability, userspace can only specify the ISS field for 6490the ESR syndrome. Other parts of the ESR, such as the EC are generated by the 6491CPU when the exception is taken. If this virtual SError is taken to EL1 using 6492AArch64, this value will be reported in the ISS field of ESR_ELx. 6493 6494See KVM_CAP_VCPU_EVENTS for more details. 6495 64968.20 KVM_CAP_HYPERV_SEND_IPI 6497---------------------------- 6498 6499:Architectures: x86 6500 6501This capability indicates that KVM supports paravirtualized Hyper-V IPI send 6502hypercalls: 6503HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx. 6504 65058.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 6506----------------------------------- 6507 6508:Architectures: x86 6509 6510This capability indicates that KVM running on top of Hyper-V hypervisor 6511enables Direct TLB flush for its guests meaning that TLB flush 6512hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM. 6513Due to the different ABI for hypercall parameters between Hyper-V and 6514KVM, enabling this capability effectively disables all hypercall 6515handling by KVM (as some KVM hypercall may be mistakenly treated as TLB 6516flush hypercalls by Hyper-V) so userspace should disable KVM identification 6517in CPUID and only exposes Hyper-V identification. In this case, guest 6518thinks it's running on Hyper-V and only use Hyper-V hypercalls. 6519 65208.22 KVM_CAP_S390_VCPU_RESETS 6521----------------------------- 6522 6523:Architectures: s390 6524 6525This capability indicates that the KVM_S390_NORMAL_RESET and 6526KVM_S390_CLEAR_RESET ioctls are available. 6527 65288.23 KVM_CAP_S390_PROTECTED 6529--------------------------- 6530 6531:Architectures: s390 6532 6533This capability indicates that the Ultravisor has been initialized and 6534KVM can therefore start protected VMs. 6535This capability governs the KVM_S390_PV_COMMAND ioctl and the 6536KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected 6537guests when the state change is invalid. 6538 65398.24 KVM_CAP_STEAL_TIME 6540----------------------- 6541 6542:Architectures: arm64, x86 6543 6544This capability indicates that KVM supports steal time accounting. 6545When steal time accounting is supported it may be enabled with 6546architecture-specific interfaces. This capability and the architecture- 6547specific interfaces must be consistent, i.e. if one says the feature 6548is supported, than the other should as well and vice versa. For arm64 6549see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL". 6550For x86 see Documentation/virt/kvm/msr.rst "MSR_KVM_STEAL_TIME". 6551 65528.25 KVM_CAP_S390_DIAG318 6553------------------------- 6554 6555:Architectures: s390 6556 6557This capability enables a guest to set information about its control program 6558(i.e. guest kernel type and version). The information is helpful during 6559system/firmware service events, providing additional data about the guest 6560environments running on the machine. 6561 6562The information is associated with the DIAGNOSE 0x318 instruction, which sets 6563an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and 6564a 7-byte Control Program Version Code (CPVC). The CPNC determines what 6565environment the control program is running in (e.g. Linux, z/VM...), and the 6566CPVC is used for information specific to OS (e.g. Linux version, Linux 6567distribution...) 6568 6569If this capability is available, then the CPNC and CPVC can be synchronized 6570between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318). 6571 65728.26 KVM_CAP_X86_USER_SPACE_MSR 6573------------------------------- 6574 6575:Architectures: x86 6576 6577This capability indicates that KVM supports deflection of MSR reads and 6578writes to user space. It can be enabled on a VM level. If enabled, MSR 6579accesses that would usually trigger a #GP by KVM into the guest will 6580instead get bounced to user space through the KVM_EXIT_X86_RDMSR and 6581KVM_EXIT_X86_WRMSR exit notifications. 6582 65838.27 KVM_X86_SET_MSR_FILTER 6584--------------------------- 6585 6586:Architectures: x86 6587 6588This capability indicates that KVM supports that accesses to user defined MSRs 6589may be rejected. With this capability exposed, KVM exports new VM ioctl 6590KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR 6591ranges that KVM should reject access to. 6592 6593In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to 6594trap and emulate MSRs that are outside of the scope of KVM as well as 6595limit the attack surface on KVM's MSR emulation code. 6596 65978.28 KVM_CAP_ENFORCE_PV_CPUID 6598----------------------------- 6599 6600Architectures: x86 6601 6602When enabled, KVM will disable paravirtual features provided to the 6603guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf 6604(0x40000001). Otherwise, a guest may use the paravirtual features 6605regardless of what has actually been exposed through the CPUID leaf. 6606 66078.29 KVM_CAP_DIRTY_LOG_RING 6608--------------------------- 6609 6610:Architectures: x86 6611:Parameters: args[0] - size of the dirty log ring 6612 6613KVM is capable of tracking dirty memory using ring buffers that are 6614mmaped into userspace; there is one dirty ring per vcpu. 6615 6616The dirty ring is available to userspace as an array of 6617``struct kvm_dirty_gfn``. Each dirty entry it's defined as:: 6618 6619 struct kvm_dirty_gfn { 6620 __u32 flags; 6621 __u32 slot; /* as_id | slot_id */ 6622 __u64 offset; 6623 }; 6624 6625The following values are defined for the flags field to define the 6626current state of the entry:: 6627 6628 #define KVM_DIRTY_GFN_F_DIRTY BIT(0) 6629 #define KVM_DIRTY_GFN_F_RESET BIT(1) 6630 #define KVM_DIRTY_GFN_F_MASK 0x3 6631 6632Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM 6633ioctl to enable this capability for the new guest and set the size of 6634the rings. Enabling the capability is only allowed before creating any 6635vCPU, and the size of the ring must be a power of two. The larger the 6636ring buffer, the less likely the ring is full and the VM is forced to 6637exit to userspace. The optimal size depends on the workload, but it is 6638recommended that it be at least 64 KiB (4096 entries). 6639 6640Just like for dirty page bitmaps, the buffer tracks writes to 6641all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was 6642set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered 6643with the flag set, userspace can start harvesting dirty pages from the 6644ring buffer. 6645 6646An entry in the ring buffer can be unused (flag bits ``00``), 6647dirty (flag bits ``01``) or harvested (flag bits ``1X``). The 6648state machine for the entry is as follows:: 6649 6650 dirtied harvested reset 6651 00 -----------> 01 -------------> 1X -------+ 6652 ^ | 6653 | | 6654 +------------------------------------------+ 6655 6656To harvest the dirty pages, userspace accesses the mmaped ring buffer 6657to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage 6658the RESET bit must be cleared), then it means this GFN is a dirty GFN. 6659The userspace should harvest this GFN and mark the flags from state 6660``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set 6661to show that this GFN is harvested and waiting for a reset), and move 6662on to the next GFN. The userspace should continue to do this until the 6663flags of a GFN have the DIRTY bit cleared, meaning that it has harvested 6664all the dirty GFNs that were available. 6665 6666It's not necessary for userspace to harvest the all dirty GFNs at once. 6667However it must collect the dirty GFNs in sequence, i.e., the userspace 6668program cannot skip one dirty GFN to collect the one next to it. 6669 6670After processing one or more entries in the ring buffer, userspace 6671calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about 6672it, so that the kernel will reprotect those collected GFNs. 6673Therefore, the ioctl must be called *before* reading the content of 6674the dirty pages. 6675 6676The dirty ring can get full. When it happens, the KVM_RUN of the 6677vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL. 6678 6679The dirty ring interface has a major difference comparing to the 6680KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from 6681userspace, it's still possible that the kernel has not yet flushed the 6682processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the 6683flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one 6684needs to kick the vcpu out of KVM_RUN using a signal. The resulting 6685vmexit ensures that all dirty GFNs are flushed to the dirty rings. 6686 6687NOTE: the capability KVM_CAP_DIRTY_LOG_RING and the corresponding 6688ioctl KVM_RESET_DIRTY_RINGS are mutual exclusive to the existing ioctls 6689KVM_GET_DIRTY_LOG and KVM_CLEAR_DIRTY_LOG. After enabling 6690KVM_CAP_DIRTY_LOG_RING with an acceptable dirty ring size, the virtual 6691machine will switch to ring-buffer dirty page tracking and further 6692KVM_GET_DIRTY_LOG or KVM_CLEAR_DIRTY_LOG ioctls will fail. 6693 66948.30 KVM_CAP_XEN_HVM 6695-------------------- 6696 6697:Architectures: x86 6698 6699This capability indicates the features that Xen supports for hosting Xen 6700PVHVM guests. Valid flags are:: 6701 6702 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0) 6703 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1) 6704 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2) 6705 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 2) 6706 6707The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG 6708ioctl is available, for the guest to set its hypercall page. 6709 6710If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be 6711provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page 6712contents, to request that KVM generate hypercall page content automatically 6713and also enable interception of guest hypercalls with KVM_EXIT_XEN. 6714 6715The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the 6716KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and 6717KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors 6718for event channel upcalls when the evtchn_upcall_pending field of a vcpu's 6719vcpu_info is set. 6720 6721The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related 6722features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are 6723supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls. 6724