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 although 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 154On arm64, the physical address size for a VM (IPA Size limit) is limited 155to 40bits by default. The limit can be configured if the host supports the 156extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use 157KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type 158identifier, where IPA_Bits is the maximum width of any physical 159address used by the VM. The IPA_Bits is encoded in bits[7-0] of the 160machine type identifier. 161 162e.g, to configure a guest to use 48bit physical address size:: 163 164 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48)); 165 166The requested size (IPA_Bits) must be: 167 168 == ========================================================= 169 0 Implies default size, 40bits (for backward compatibility) 170 N Implies N bits, where N is a positive integer such that, 171 32 <= N <= Host_IPA_Limit 172 == ========================================================= 173 174Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and 175is dependent on the CPU capability and the kernel configuration. The limit can 176be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION 177ioctl() at run-time. 178 179Creation of the VM will fail if the requested IPA size (whether it is 180implicit or explicit) is unsupported on the host. 181 182Please note that configuring the IPA size does not affect the capability 183exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects 184size of the address translated by the stage2 level (guest physical to 185host physical address translations). 186 187 1884.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST 189---------------------------------------------------------- 190 191:Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST 192:Architectures: x86 193:Type: system ioctl 194:Parameters: struct kvm_msr_list (in/out) 195:Returns: 0 on success; -1 on error 196 197Errors: 198 199 ====== ============================================================ 200 EFAULT the msr index list cannot be read from or written to 201 E2BIG the msr index list is too big to fit in the array specified by 202 the user. 203 ====== ============================================================ 204 205:: 206 207 struct kvm_msr_list { 208 __u32 nmsrs; /* number of msrs in entries */ 209 __u32 indices[0]; 210 }; 211 212The user fills in the size of the indices array in nmsrs, and in return 213kvm adjusts nmsrs to reflect the actual number of msrs and fills in the 214indices array with their numbers. 215 216KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list 217varies by kvm version and host processor, but does not change otherwise. 218 219Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 220not returned in the MSR list, as different vcpus can have a different number 221of banks, as set via the KVM_X86_SETUP_MCE ioctl. 222 223KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed 224to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities 225and processor features that are exposed via MSRs (e.g., VMX capabilities). 226This list also varies by kvm version and host processor, but does not change 227otherwise. 228 229 2304.4 KVM_CHECK_EXTENSION 231----------------------- 232 233:Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl 234:Architectures: all 235:Type: system ioctl, vm ioctl 236:Parameters: extension identifier (KVM_CAP_*) 237:Returns: 0 if unsupported; 1 (or some other positive integer) if supported 238 239The API allows the application to query about extensions to the core 240kvm API. Userspace passes an extension identifier (an integer) and 241receives an integer that describes the extension availability. 242Generally 0 means no and 1 means yes, but some extensions may report 243additional information in the integer return value. 244 245Based on their initialization different VMs may have different capabilities. 246It is thus encouraged to use the vm ioctl to query for capabilities (available 247with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 248 2494.5 KVM_GET_VCPU_MMAP_SIZE 250-------------------------- 251 252:Capability: basic 253:Architectures: all 254:Type: system ioctl 255:Parameters: none 256:Returns: size of vcpu mmap area, in bytes 257 258The KVM_RUN ioctl (cf.) communicates with userspace via a shared 259memory region. This ioctl returns the size of that region. See the 260KVM_RUN documentation for details. 261 262Besides the size of the KVM_RUN communication region, other areas of 263the VCPU file descriptor can be mmap-ed, including: 264 265- if KVM_CAP_COALESCED_MMIO is available, a page at 266 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons, 267 this page is included in the result of KVM_GET_VCPU_MMAP_SIZE. 268 KVM_CAP_COALESCED_MMIO is not documented yet. 269 270- if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at 271 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For more information on 272 KVM_CAP_DIRTY_LOG_RING, see section 8.3. 273 274 2754.6 KVM_SET_MEMORY_REGION 276------------------------- 277 278:Capability: basic 279:Architectures: all 280:Type: vm ioctl 281:Parameters: struct kvm_memory_region (in) 282:Returns: 0 on success, -1 on error 283 284This ioctl is obsolete and has been removed. 285 286 2874.7 KVM_CREATE_VCPU 288------------------- 289 290:Capability: basic 291:Architectures: all 292:Type: vm ioctl 293:Parameters: vcpu id (apic id on x86) 294:Returns: vcpu fd on success, -1 on error 295 296This API adds a vcpu to a virtual machine. No more than max_vcpus may be added. 297The vcpu id is an integer in the range [0, max_vcpu_id). 298 299The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 300the KVM_CHECK_EXTENSION ioctl() at run-time. 301The maximum possible value for max_vcpus can be retrieved using the 302KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 303 304If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 305cpus max. 306If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 307same as the value returned from KVM_CAP_NR_VCPUS. 308 309The maximum possible value for max_vcpu_id can be retrieved using the 310KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time. 311 312If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id 313is the same as the value returned from KVM_CAP_MAX_VCPUS. 314 315On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 316threads in one or more virtual CPU cores. (This is because the 317hardware requires all the hardware threads in a CPU core to be in the 318same partition.) The KVM_CAP_PPC_SMT capability indicates the number 319of vcpus per virtual core (vcore). The vcore id is obtained by 320dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 321given vcore will always be in the same physical core as each other 322(though that might be a different physical core from time to time). 323Userspace can control the threading (SMT) mode of the guest by its 324allocation of vcpu ids. For example, if userspace wants 325single-threaded guest vcpus, it should make all vcpu ids be a multiple 326of the number of vcpus per vcore. 327 328For virtual cpus that have been created with S390 user controlled virtual 329machines, the resulting vcpu fd can be memory mapped at page offset 330KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 331cpu's hardware control block. 332 333 3344.8 KVM_GET_DIRTY_LOG (vm ioctl) 335-------------------------------- 336 337:Capability: basic 338:Architectures: all 339:Type: vm ioctl 340:Parameters: struct kvm_dirty_log (in/out) 341:Returns: 0 on success, -1 on error 342 343:: 344 345 /* for KVM_GET_DIRTY_LOG */ 346 struct kvm_dirty_log { 347 __u32 slot; 348 __u32 padding; 349 union { 350 void __user *dirty_bitmap; /* one bit per page */ 351 __u64 padding; 352 }; 353 }; 354 355Given a memory slot, return a bitmap containing any pages dirtied 356since the last call to this ioctl. Bit 0 is the first page in the 357memory slot. Ensure the entire structure is cleared to avoid padding 358issues. 359 360If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 361the address space for which you want to return the dirty bitmap. See 362KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 363 364The bits in the dirty bitmap are cleared before the ioctl returns, unless 365KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information, 366see the description of the capability. 367 368Note that the Xen shared info page, if configured, shall always be assumed 369to be dirty. KVM will not explicitly mark it such. 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 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 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, riscv 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 601RISC-V: 602^^^^^^^ 603 604Queues an external interrupt to be injected into the virutal CPU. This ioctl 605is overloaded with 2 different irq values: 606 607a) KVM_INTERRUPT_SET 608 609 This sets external interrupt for a virtual CPU and it will receive 610 once it is ready. 611 612b) KVM_INTERRUPT_UNSET 613 614 This clears pending external interrupt for a virtual CPU. 615 616This is an asynchronous vcpu ioctl and can be invoked from any thread. 617 618 6194.17 KVM_DEBUG_GUEST 620-------------------- 621 622:Capability: basic 623:Architectures: none 624:Type: vcpu ioctl 625:Parameters: none) 626:Returns: -1 on error 627 628Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. 629 630 6314.18 KVM_GET_MSRS 632----------------- 633 634:Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system) 635:Architectures: x86 636:Type: system ioctl, vcpu ioctl 637:Parameters: struct kvm_msrs (in/out) 638:Returns: number of msrs successfully returned; 639 -1 on error 640 641When used as a system ioctl: 642Reads the values of MSR-based features that are available for the VM. This 643is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values. 644The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST 645in a system ioctl. 646 647When used as a vcpu ioctl: 648Reads model-specific registers from the vcpu. Supported msr indices can 649be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl. 650 651:: 652 653 struct kvm_msrs { 654 __u32 nmsrs; /* number of msrs in entries */ 655 __u32 pad; 656 657 struct kvm_msr_entry entries[0]; 658 }; 659 660 struct kvm_msr_entry { 661 __u32 index; 662 __u32 reserved; 663 __u64 data; 664 }; 665 666Application code should set the 'nmsrs' member (which indicates the 667size of the entries array) and the 'index' member of each array entry. 668kvm will fill in the 'data' member. 669 670 6714.19 KVM_SET_MSRS 672----------------- 673 674:Capability: basic 675:Architectures: x86 676:Type: vcpu ioctl 677:Parameters: struct kvm_msrs (in) 678:Returns: number of msrs successfully set (see below), -1 on error 679 680Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 681data structures. 682 683Application code should set the 'nmsrs' member (which indicates the 684size of the entries array), and the 'index' and 'data' members of each 685array entry. 686 687It tries to set the MSRs in array entries[] one by one. If setting an MSR 688fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated 689by KVM, etc..., it stops processing the MSR list and returns the number of 690MSRs that have been set successfully. 691 692 6934.20 KVM_SET_CPUID 694------------------ 695 696:Capability: basic 697:Architectures: x86 698:Type: vcpu ioctl 699:Parameters: struct kvm_cpuid (in) 700:Returns: 0 on success, -1 on error 701 702Defines the vcpu responses to the cpuid instruction. Applications 703should use the KVM_SET_CPUID2 ioctl if available. 704 705Caveat emptor: 706 - If this IOCTL fails, KVM gives no guarantees that previous valid CPUID 707 configuration (if there is) is not corrupted. Userspace can get a copy 708 of the resulting CPUID configuration through KVM_GET_CPUID2 in case. 709 - Using KVM_SET_CPUID{,2} after KVM_RUN, i.e. changing the guest vCPU model 710 after running the guest, may cause guest instability. 711 - Using heterogeneous CPUID configurations, modulo APIC IDs, topology, etc... 712 may cause guest instability. 713 714:: 715 716 struct kvm_cpuid_entry { 717 __u32 function; 718 __u32 eax; 719 __u32 ebx; 720 __u32 ecx; 721 __u32 edx; 722 __u32 padding; 723 }; 724 725 /* for KVM_SET_CPUID */ 726 struct kvm_cpuid { 727 __u32 nent; 728 __u32 padding; 729 struct kvm_cpuid_entry entries[0]; 730 }; 731 732 7334.21 KVM_SET_SIGNAL_MASK 734------------------------ 735 736:Capability: basic 737:Architectures: all 738:Type: vcpu ioctl 739:Parameters: struct kvm_signal_mask (in) 740:Returns: 0 on success, -1 on error 741 742Defines which signals are blocked during execution of KVM_RUN. This 743signal mask temporarily overrides the threads signal mask. Any 744unblocked signal received (except SIGKILL and SIGSTOP, which retain 745their traditional behaviour) will cause KVM_RUN to return with -EINTR. 746 747Note the signal will only be delivered if not blocked by the original 748signal mask. 749 750:: 751 752 /* for KVM_SET_SIGNAL_MASK */ 753 struct kvm_signal_mask { 754 __u32 len; 755 __u8 sigset[0]; 756 }; 757 758 7594.22 KVM_GET_FPU 760---------------- 761 762:Capability: basic 763:Architectures: x86 764:Type: vcpu ioctl 765:Parameters: struct kvm_fpu (out) 766:Returns: 0 on success, -1 on error 767 768Reads the floating point state from the vcpu. 769 770:: 771 772 /* for KVM_GET_FPU and KVM_SET_FPU */ 773 struct kvm_fpu { 774 __u8 fpr[8][16]; 775 __u16 fcw; 776 __u16 fsw; 777 __u8 ftwx; /* in fxsave format */ 778 __u8 pad1; 779 __u16 last_opcode; 780 __u64 last_ip; 781 __u64 last_dp; 782 __u8 xmm[16][16]; 783 __u32 mxcsr; 784 __u32 pad2; 785 }; 786 787 7884.23 KVM_SET_FPU 789---------------- 790 791:Capability: basic 792:Architectures: x86 793:Type: vcpu ioctl 794:Parameters: struct kvm_fpu (in) 795:Returns: 0 on success, -1 on error 796 797Writes the floating point state to the vcpu. 798 799:: 800 801 /* for KVM_GET_FPU and KVM_SET_FPU */ 802 struct kvm_fpu { 803 __u8 fpr[8][16]; 804 __u16 fcw; 805 __u16 fsw; 806 __u8 ftwx; /* in fxsave format */ 807 __u8 pad1; 808 __u16 last_opcode; 809 __u64 last_ip; 810 __u64 last_dp; 811 __u8 xmm[16][16]; 812 __u32 mxcsr; 813 __u32 pad2; 814 }; 815 816 8174.24 KVM_CREATE_IRQCHIP 818----------------------- 819 820:Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) 821:Architectures: x86, arm64, s390 822:Type: vm ioctl 823:Parameters: none 824:Returns: 0 on success, -1 on error 825 826Creates an interrupt controller model in the kernel. 827On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up 828future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both 829PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. 830On arm64, a GICv2 is created. Any other GIC versions require the usage of 831KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 832KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 833On s390, a dummy irq routing table is created. 834 835Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 836before KVM_CREATE_IRQCHIP can be used. 837 838 8394.25 KVM_IRQ_LINE 840----------------- 841 842:Capability: KVM_CAP_IRQCHIP 843:Architectures: x86, arm64 844:Type: vm ioctl 845:Parameters: struct kvm_irq_level 846:Returns: 0 on success, -1 on error 847 848Sets the level of a GSI input to the interrupt controller model in the kernel. 849On some architectures it is required that an interrupt controller model has 850been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 851interrupts require the level to be set to 1 and then back to 0. 852 853On real hardware, interrupt pins can be active-low or active-high. This 854does not matter for the level field of struct kvm_irq_level: 1 always 855means active (asserted), 0 means inactive (deasserted). 856 857x86 allows the operating system to program the interrupt polarity 858(active-low/active-high) for level-triggered interrupts, and KVM used 859to consider the polarity. However, due to bitrot in the handling of 860active-low interrupts, the above convention is now valid on x86 too. 861This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 862should not present interrupts to the guest as active-low unless this 863capability is present (or unless it is not using the in-kernel irqchip, 864of course). 865 866 867arm64 can signal an interrupt either at the CPU level, or at the 868in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 869use PPIs designated for specific cpus. The irq field is interpreted 870like this:: 871 872 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 | 873 field: | vcpu2_index | irq_type | vcpu_index | irq_id | 874 875The irq_type field has the following values: 876 877- irq_type[0]: 878 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 879- irq_type[1]: 880 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 881 (the vcpu_index field is ignored) 882- irq_type[2]: 883 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 884 885(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 886 887In both cases, level is used to assert/deassert the line. 888 889When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is 890identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index 891must be zero. 892 893Note that on arm64, the KVM_CAP_IRQCHIP capability only conditions 894injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always 895be used for a userspace interrupt controller. 896 897:: 898 899 struct kvm_irq_level { 900 union { 901 __u32 irq; /* GSI */ 902 __s32 status; /* not used for KVM_IRQ_LEVEL */ 903 }; 904 __u32 level; /* 0 or 1 */ 905 }; 906 907 9084.26 KVM_GET_IRQCHIP 909-------------------- 910 911:Capability: KVM_CAP_IRQCHIP 912:Architectures: x86 913:Type: vm ioctl 914:Parameters: struct kvm_irqchip (in/out) 915:Returns: 0 on success, -1 on error 916 917Reads the state of a kernel interrupt controller created with 918KVM_CREATE_IRQCHIP into a buffer provided by the caller. 919 920:: 921 922 struct kvm_irqchip { 923 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 924 __u32 pad; 925 union { 926 char dummy[512]; /* reserving space */ 927 struct kvm_pic_state pic; 928 struct kvm_ioapic_state ioapic; 929 } chip; 930 }; 931 932 9334.27 KVM_SET_IRQCHIP 934-------------------- 935 936:Capability: KVM_CAP_IRQCHIP 937:Architectures: x86 938:Type: vm ioctl 939:Parameters: struct kvm_irqchip (in) 940:Returns: 0 on success, -1 on error 941 942Sets the state of a kernel interrupt controller created with 943KVM_CREATE_IRQCHIP from a buffer provided by the caller. 944 945:: 946 947 struct kvm_irqchip { 948 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 949 __u32 pad; 950 union { 951 char dummy[512]; /* reserving space */ 952 struct kvm_pic_state pic; 953 struct kvm_ioapic_state ioapic; 954 } chip; 955 }; 956 957 9584.28 KVM_XEN_HVM_CONFIG 959----------------------- 960 961:Capability: KVM_CAP_XEN_HVM 962:Architectures: x86 963:Type: vm ioctl 964:Parameters: struct kvm_xen_hvm_config (in) 965:Returns: 0 on success, -1 on error 966 967Sets the MSR that the Xen HVM guest uses to initialize its hypercall 968page, and provides the starting address and size of the hypercall 969blobs in userspace. When the guest writes the MSR, kvm copies one 970page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 971memory. 972 973:: 974 975 struct kvm_xen_hvm_config { 976 __u32 flags; 977 __u32 msr; 978 __u64 blob_addr_32; 979 __u64 blob_addr_64; 980 __u8 blob_size_32; 981 __u8 blob_size_64; 982 __u8 pad2[30]; 983 }; 984 985If the KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag is returned from the 986KVM_CAP_XEN_HVM check, it may be set in the flags field of this ioctl. 987This requests KVM to generate the contents of the hypercall page 988automatically; hypercalls will be intercepted and passed to userspace 989through KVM_EXIT_XEN. In this case, all of the blob size and address 990fields must be zero. 991 992No other flags are currently valid in the struct kvm_xen_hvm_config. 993 9944.29 KVM_GET_CLOCK 995------------------ 996 997:Capability: KVM_CAP_ADJUST_CLOCK 998:Architectures: x86 999:Type: vm ioctl 1000:Parameters: struct kvm_clock_data (out) 1001:Returns: 0 on success, -1 on error 1002 1003Gets the current timestamp of kvmclock as seen by the current guest. In 1004conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 1005such as migration. 1006 1007When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 1008set of bits that KVM can return in struct kvm_clock_data's flag member. 1009 1010The following flags are defined: 1011 1012KVM_CLOCK_TSC_STABLE 1013 If set, the returned value is the exact kvmclock 1014 value seen by all VCPUs at the instant when KVM_GET_CLOCK was called. 1015 If clear, the returned value is simply CLOCK_MONOTONIC plus a constant 1016 offset; the offset can be modified with KVM_SET_CLOCK. KVM will try 1017 to make all VCPUs follow this clock, but the exact value read by each 1018 VCPU could differ, because the host TSC is not stable. 1019 1020KVM_CLOCK_REALTIME 1021 If set, the `realtime` field in the kvm_clock_data 1022 structure is populated with the value of the host's real time 1023 clocksource at the instant when KVM_GET_CLOCK was called. If clear, 1024 the `realtime` field does not contain a value. 1025 1026KVM_CLOCK_HOST_TSC 1027 If set, the `host_tsc` field in the kvm_clock_data 1028 structure is populated with the value of the host's timestamp counter (TSC) 1029 at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field 1030 does not contain a value. 1031 1032:: 1033 1034 struct kvm_clock_data { 1035 __u64 clock; /* kvmclock current value */ 1036 __u32 flags; 1037 __u32 pad0; 1038 __u64 realtime; 1039 __u64 host_tsc; 1040 __u32 pad[4]; 1041 }; 1042 1043 10444.30 KVM_SET_CLOCK 1045------------------ 1046 1047:Capability: KVM_CAP_ADJUST_CLOCK 1048:Architectures: x86 1049:Type: vm ioctl 1050:Parameters: struct kvm_clock_data (in) 1051:Returns: 0 on success, -1 on error 1052 1053Sets the current timestamp of kvmclock to the value specified in its parameter. 1054In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 1055such as migration. 1056 1057The following flags can be passed: 1058 1059KVM_CLOCK_REALTIME 1060 If set, KVM will compare the value of the `realtime` field 1061 with the value of the host's real time clocksource at the instant when 1062 KVM_SET_CLOCK was called. The difference in elapsed time is added to the final 1063 kvmclock value that will be provided to guests. 1064 1065Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored. 1066 1067:: 1068 1069 struct kvm_clock_data { 1070 __u64 clock; /* kvmclock current value */ 1071 __u32 flags; 1072 __u32 pad0; 1073 __u64 realtime; 1074 __u64 host_tsc; 1075 __u32 pad[4]; 1076 }; 1077 1078 10794.31 KVM_GET_VCPU_EVENTS 1080------------------------ 1081 1082:Capability: KVM_CAP_VCPU_EVENTS 1083:Extended by: KVM_CAP_INTR_SHADOW 1084:Architectures: x86, arm64 1085:Type: vcpu ioctl 1086:Parameters: struct kvm_vcpu_event (out) 1087:Returns: 0 on success, -1 on error 1088 1089X86: 1090^^^^ 1091 1092Gets currently pending exceptions, interrupts, and NMIs as well as related 1093states of the vcpu. 1094 1095:: 1096 1097 struct kvm_vcpu_events { 1098 struct { 1099 __u8 injected; 1100 __u8 nr; 1101 __u8 has_error_code; 1102 __u8 pending; 1103 __u32 error_code; 1104 } exception; 1105 struct { 1106 __u8 injected; 1107 __u8 nr; 1108 __u8 soft; 1109 __u8 shadow; 1110 } interrupt; 1111 struct { 1112 __u8 injected; 1113 __u8 pending; 1114 __u8 masked; 1115 __u8 pad; 1116 } nmi; 1117 __u32 sipi_vector; 1118 __u32 flags; 1119 struct { 1120 __u8 smm; 1121 __u8 pending; 1122 __u8 smm_inside_nmi; 1123 __u8 latched_init; 1124 } smi; 1125 __u8 reserved[27]; 1126 __u8 exception_has_payload; 1127 __u64 exception_payload; 1128 }; 1129 1130The following bits are defined in the flags field: 1131 1132- KVM_VCPUEVENT_VALID_SHADOW may be set to signal that 1133 interrupt.shadow contains a valid state. 1134 1135- KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a 1136 valid state. 1137 1138- KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the 1139 exception_has_payload, exception_payload, and exception.pending 1140 fields contain a valid state. This bit will be set whenever 1141 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1142 1143ARM64: 1144^^^^^^ 1145 1146If the guest accesses a device that is being emulated by the host kernel in 1147such a way that a real device would generate a physical SError, KVM may make 1148a virtual SError pending for that VCPU. This system error interrupt remains 1149pending until the guest takes the exception by unmasking PSTATE.A. 1150 1151Running the VCPU may cause it to take a pending SError, or make an access that 1152causes an SError to become pending. The event's description is only valid while 1153the VPCU is not running. 1154 1155This API provides a way to read and write the pending 'event' state that is not 1156visible to the guest. To save, restore or migrate a VCPU the struct representing 1157the state can be read then written using this GET/SET API, along with the other 1158guest-visible registers. It is not possible to 'cancel' an SError that has been 1159made pending. 1160 1161A device being emulated in user-space may also wish to generate an SError. To do 1162this the events structure can be populated by user-space. The current state 1163should be read first, to ensure no existing SError is pending. If an existing 1164SError is pending, the architecture's 'Multiple SError interrupts' rules should 1165be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and 1166Serviceability (RAS) Specification"). 1167 1168SError exceptions always have an ESR value. Some CPUs have the ability to 1169specify what the virtual SError's ESR value should be. These systems will 1170advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will 1171always have a non-zero value when read, and the agent making an SError pending 1172should specify the ISS field in the lower 24 bits of exception.serror_esr. If 1173the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events 1174with exception.has_esr as zero, KVM will choose an ESR. 1175 1176Specifying exception.has_esr on a system that does not support it will return 1177-EINVAL. Setting anything other than the lower 24bits of exception.serror_esr 1178will return -EINVAL. 1179 1180It is not possible to read back a pending external abort (injected via 1181KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered 1182directly to the virtual CPU). 1183 1184:: 1185 1186 struct kvm_vcpu_events { 1187 struct { 1188 __u8 serror_pending; 1189 __u8 serror_has_esr; 1190 __u8 ext_dabt_pending; 1191 /* Align it to 8 bytes */ 1192 __u8 pad[5]; 1193 __u64 serror_esr; 1194 } exception; 1195 __u32 reserved[12]; 1196 }; 1197 11984.32 KVM_SET_VCPU_EVENTS 1199------------------------ 1200 1201:Capability: KVM_CAP_VCPU_EVENTS 1202:Extended by: KVM_CAP_INTR_SHADOW 1203:Architectures: x86, arm64 1204:Type: vcpu ioctl 1205:Parameters: struct kvm_vcpu_event (in) 1206:Returns: 0 on success, -1 on error 1207 1208X86: 1209^^^^ 1210 1211Set pending exceptions, interrupts, and NMIs as well as related states of the 1212vcpu. 1213 1214See KVM_GET_VCPU_EVENTS for the data structure. 1215 1216Fields that may be modified asynchronously by running VCPUs can be excluded 1217from the update. These fields are nmi.pending, sipi_vector, smi.smm, 1218smi.pending. Keep the corresponding bits in the flags field cleared to 1219suppress overwriting the current in-kernel state. The bits are: 1220 1221=============================== ================================== 1222KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel 1223KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector 1224KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct. 1225=============================== ================================== 1226 1227If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 1228the flags field to signal that interrupt.shadow contains a valid state and 1229shall be written into the VCPU. 1230 1231KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 1232 1233If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD 1234can be set in the flags field to signal that the 1235exception_has_payload, exception_payload, and exception.pending fields 1236contain a valid state and shall be written into the VCPU. 1237 1238ARM64: 1239^^^^^^ 1240 1241User space may need to inject several types of events to the guest. 1242 1243Set the pending SError exception state for this VCPU. It is not possible to 1244'cancel' an Serror that has been made pending. 1245 1246If the guest performed an access to I/O memory which could not be handled by 1247userspace, for example because of missing instruction syndrome decode 1248information or because there is no device mapped at the accessed IPA, then 1249userspace can ask the kernel to inject an external abort using the address 1250from the exiting fault on the VCPU. It is a programming error to set 1251ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or 1252KVM_EXIT_ARM_NISV. This feature is only available if the system supports 1253KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in 1254how userspace reports accesses for the above cases to guests, across different 1255userspace implementations. Nevertheless, userspace can still emulate all Arm 1256exceptions by manipulating individual registers using the KVM_SET_ONE_REG API. 1257 1258See KVM_GET_VCPU_EVENTS for the data structure. 1259 1260 12614.33 KVM_GET_DEBUGREGS 1262---------------------- 1263 1264:Capability: KVM_CAP_DEBUGREGS 1265:Architectures: x86 1266:Type: vm ioctl 1267:Parameters: struct kvm_debugregs (out) 1268:Returns: 0 on success, -1 on error 1269 1270Reads debug registers from the vcpu. 1271 1272:: 1273 1274 struct kvm_debugregs { 1275 __u64 db[4]; 1276 __u64 dr6; 1277 __u64 dr7; 1278 __u64 flags; 1279 __u64 reserved[9]; 1280 }; 1281 1282 12834.34 KVM_SET_DEBUGREGS 1284---------------------- 1285 1286:Capability: KVM_CAP_DEBUGREGS 1287:Architectures: x86 1288:Type: vm ioctl 1289:Parameters: struct kvm_debugregs (in) 1290:Returns: 0 on success, -1 on error 1291 1292Writes debug registers into the vcpu. 1293 1294See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 1295yet and must be cleared on entry. 1296 1297 12984.35 KVM_SET_USER_MEMORY_REGION 1299------------------------------- 1300 1301:Capability: KVM_CAP_USER_MEMORY 1302:Architectures: all 1303:Type: vm ioctl 1304:Parameters: struct kvm_userspace_memory_region (in) 1305:Returns: 0 on success, -1 on error 1306 1307:: 1308 1309 struct kvm_userspace_memory_region { 1310 __u32 slot; 1311 __u32 flags; 1312 __u64 guest_phys_addr; 1313 __u64 memory_size; /* bytes */ 1314 __u64 userspace_addr; /* start of the userspace allocated memory */ 1315 }; 1316 1317 /* for kvm_memory_region::flags */ 1318 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 1319 #define KVM_MEM_READONLY (1UL << 1) 1320 1321This ioctl allows the user to create, modify or delete a guest physical 1322memory slot. Bits 0-15 of "slot" specify the slot id and this value 1323should be less than the maximum number of user memory slots supported per 1324VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS. 1325Slots may not overlap in guest physical address space. 1326 1327If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 1328specifies the address space which is being modified. They must be 1329less than the value that KVM_CHECK_EXTENSION returns for the 1330KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 1331are unrelated; the restriction on overlapping slots only applies within 1332each address space. 1333 1334Deleting a slot is done by passing zero for memory_size. When changing 1335an existing slot, it may be moved in the guest physical memory space, 1336or its flags may be modified, but it may not be resized. 1337 1338Memory for the region is taken starting at the address denoted by the 1339field userspace_addr, which must point at user addressable memory for 1340the entire memory slot size. Any object may back this memory, including 1341anonymous memory, ordinary files, and hugetlbfs. 1342 1343On architectures that support a form of address tagging, userspace_addr must 1344be an untagged address. 1345 1346It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 1347be identical. This allows large pages in the guest to be backed by large 1348pages in the host. 1349 1350The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 1351KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 1352writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 1353use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 1354to make a new slot read-only. In this case, writes to this memory will be 1355posted to userspace as KVM_EXIT_MMIO exits. 1356 1357When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 1358the memory region are automatically reflected into the guest. For example, an 1359mmap() that affects the region will be made visible immediately. Another 1360example is madvise(MADV_DROP). 1361 1362It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. 1363The KVM_SET_MEMORY_REGION does not allow fine grained control over memory 1364allocation and is deprecated. 1365 1366 13674.36 KVM_SET_TSS_ADDR 1368--------------------- 1369 1370:Capability: KVM_CAP_SET_TSS_ADDR 1371:Architectures: x86 1372:Type: vm ioctl 1373:Parameters: unsigned long tss_address (in) 1374:Returns: 0 on success, -1 on error 1375 1376This ioctl defines the physical address of a three-page region in the guest 1377physical address space. The region must be within the first 4GB of the 1378guest physical address space and must not conflict with any memory slot 1379or any mmio address. The guest may malfunction if it accesses this memory 1380region. 1381 1382This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1383because of a quirk in the virtualization implementation (see the internals 1384documentation when it pops into existence). 1385 1386 13874.37 KVM_ENABLE_CAP 1388------------------- 1389 1390:Capability: KVM_CAP_ENABLE_CAP 1391:Architectures: mips, ppc, s390, x86 1392:Type: vcpu ioctl 1393:Parameters: struct kvm_enable_cap (in) 1394:Returns: 0 on success; -1 on error 1395 1396:Capability: KVM_CAP_ENABLE_CAP_VM 1397:Architectures: all 1398:Type: vm ioctl 1399:Parameters: struct kvm_enable_cap (in) 1400:Returns: 0 on success; -1 on error 1401 1402.. note:: 1403 1404 Not all extensions are enabled by default. Using this ioctl the application 1405 can enable an extension, making it available to the guest. 1406 1407On systems that do not support this ioctl, it always fails. On systems that 1408do support it, it only works for extensions that are supported for enablement. 1409 1410To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1411be used. 1412 1413:: 1414 1415 struct kvm_enable_cap { 1416 /* in */ 1417 __u32 cap; 1418 1419The capability that is supposed to get enabled. 1420 1421:: 1422 1423 __u32 flags; 1424 1425A bitfield indicating future enhancements. Has to be 0 for now. 1426 1427:: 1428 1429 __u64 args[4]; 1430 1431Arguments for enabling a feature. If a feature needs initial values to 1432function properly, this is the place to put them. 1433 1434:: 1435 1436 __u8 pad[64]; 1437 }; 1438 1439The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1440for vm-wide capabilities. 1441 14424.38 KVM_GET_MP_STATE 1443--------------------- 1444 1445:Capability: KVM_CAP_MP_STATE 1446:Architectures: x86, s390, arm64, riscv 1447:Type: vcpu ioctl 1448:Parameters: struct kvm_mp_state (out) 1449:Returns: 0 on success; -1 on error 1450 1451:: 1452 1453 struct kvm_mp_state { 1454 __u32 mp_state; 1455 }; 1456 1457Returns the vcpu's current "multiprocessing state" (though also valid on 1458uniprocessor guests). 1459 1460Possible values are: 1461 1462 ========================== =============================================== 1463 KVM_MP_STATE_RUNNABLE the vcpu is currently running 1464 [x86,arm64,riscv] 1465 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP) 1466 which has not yet received an INIT signal [x86] 1467 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is 1468 now ready for a SIPI [x86] 1469 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and 1470 is waiting for an interrupt [x86] 1471 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector 1472 accessible via KVM_GET_VCPU_EVENTS) [x86] 1473 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm64,riscv] 1474 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390] 1475 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted) 1476 [s390] 1477 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state 1478 [s390] 1479 ========================== =============================================== 1480 1481On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1482in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1483these architectures. 1484 1485For arm64/riscv: 1486^^^^^^^^^^^^^^^^ 1487 1488The only states that are valid are KVM_MP_STATE_STOPPED and 1489KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1490 14914.39 KVM_SET_MP_STATE 1492--------------------- 1493 1494:Capability: KVM_CAP_MP_STATE 1495:Architectures: x86, s390, arm64, riscv 1496:Type: vcpu ioctl 1497:Parameters: struct kvm_mp_state (in) 1498:Returns: 0 on success; -1 on error 1499 1500Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1501arguments. 1502 1503On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1504in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1505these architectures. 1506 1507For arm64/riscv: 1508^^^^^^^^^^^^^^^^ 1509 1510The only states that are valid are KVM_MP_STATE_STOPPED and 1511KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1512 15134.40 KVM_SET_IDENTITY_MAP_ADDR 1514------------------------------ 1515 1516:Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1517:Architectures: x86 1518:Type: vm ioctl 1519:Parameters: unsigned long identity (in) 1520:Returns: 0 on success, -1 on error 1521 1522This ioctl defines the physical address of a one-page region in the guest 1523physical address space. The region must be within the first 4GB of the 1524guest physical address space and must not conflict with any memory slot 1525or any mmio address. The guest may malfunction if it accesses this memory 1526region. 1527 1528Setting the address to 0 will result in resetting the address to its default 1529(0xfffbc000). 1530 1531This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1532because of a quirk in the virtualization implementation (see the internals 1533documentation when it pops into existence). 1534 1535Fails if any VCPU has already been created. 1536 15374.41 KVM_SET_BOOT_CPU_ID 1538------------------------ 1539 1540:Capability: KVM_CAP_SET_BOOT_CPU_ID 1541:Architectures: x86 1542:Type: vm ioctl 1543:Parameters: unsigned long vcpu_id 1544:Returns: 0 on success, -1 on error 1545 1546Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1547as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1548is vcpu 0. This ioctl has to be called before vcpu creation, 1549otherwise it will return EBUSY error. 1550 1551 15524.42 KVM_GET_XSAVE 1553------------------ 1554 1555:Capability: KVM_CAP_XSAVE 1556:Architectures: x86 1557:Type: vcpu ioctl 1558:Parameters: struct kvm_xsave (out) 1559:Returns: 0 on success, -1 on error 1560 1561 1562:: 1563 1564 struct kvm_xsave { 1565 __u32 region[1024]; 1566 __u32 extra[0]; 1567 }; 1568 1569This ioctl would copy current vcpu's xsave struct to the userspace. 1570 1571 15724.43 KVM_SET_XSAVE 1573------------------ 1574 1575:Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2 1576:Architectures: x86 1577:Type: vcpu ioctl 1578:Parameters: struct kvm_xsave (in) 1579:Returns: 0 on success, -1 on error 1580 1581:: 1582 1583 1584 struct kvm_xsave { 1585 __u32 region[1024]; 1586 __u32 extra[0]; 1587 }; 1588 1589This ioctl would copy userspace's xsave struct to the kernel. It copies 1590as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2), 1591when invoked on the vm file descriptor. The size value returned by 1592KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 1593Currently, it is only greater than 4096 if a dynamic feature has been 1594enabled with ``arch_prctl()``, but this may change in the future. 1595 1596The offsets of the state save areas in struct kvm_xsave follow the 1597contents of CPUID leaf 0xD on the host. 1598 1599 16004.44 KVM_GET_XCRS 1601----------------- 1602 1603:Capability: KVM_CAP_XCRS 1604:Architectures: x86 1605:Type: vcpu ioctl 1606:Parameters: struct kvm_xcrs (out) 1607:Returns: 0 on success, -1 on error 1608 1609:: 1610 1611 struct kvm_xcr { 1612 __u32 xcr; 1613 __u32 reserved; 1614 __u64 value; 1615 }; 1616 1617 struct kvm_xcrs { 1618 __u32 nr_xcrs; 1619 __u32 flags; 1620 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1621 __u64 padding[16]; 1622 }; 1623 1624This ioctl would copy current vcpu's xcrs to the userspace. 1625 1626 16274.45 KVM_SET_XCRS 1628----------------- 1629 1630:Capability: KVM_CAP_XCRS 1631:Architectures: x86 1632:Type: vcpu ioctl 1633:Parameters: struct kvm_xcrs (in) 1634:Returns: 0 on success, -1 on error 1635 1636:: 1637 1638 struct kvm_xcr { 1639 __u32 xcr; 1640 __u32 reserved; 1641 __u64 value; 1642 }; 1643 1644 struct kvm_xcrs { 1645 __u32 nr_xcrs; 1646 __u32 flags; 1647 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1648 __u64 padding[16]; 1649 }; 1650 1651This ioctl would set vcpu's xcr to the value userspace specified. 1652 1653 16544.46 KVM_GET_SUPPORTED_CPUID 1655---------------------------- 1656 1657:Capability: KVM_CAP_EXT_CPUID 1658:Architectures: x86 1659:Type: system ioctl 1660:Parameters: struct kvm_cpuid2 (in/out) 1661:Returns: 0 on success, -1 on error 1662 1663:: 1664 1665 struct kvm_cpuid2 { 1666 __u32 nent; 1667 __u32 padding; 1668 struct kvm_cpuid_entry2 entries[0]; 1669 }; 1670 1671 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1672 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 1673 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 1674 1675 struct kvm_cpuid_entry2 { 1676 __u32 function; 1677 __u32 index; 1678 __u32 flags; 1679 __u32 eax; 1680 __u32 ebx; 1681 __u32 ecx; 1682 __u32 edx; 1683 __u32 padding[3]; 1684 }; 1685 1686This ioctl returns x86 cpuid features which are supported by both the 1687hardware and kvm in its default configuration. Userspace can use the 1688information returned by this ioctl to construct cpuid information (for 1689KVM_SET_CPUID2) that is consistent with hardware, kernel, and 1690userspace capabilities, and with user requirements (for example, the 1691user may wish to constrain cpuid to emulate older hardware, or for 1692feature consistency across a cluster). 1693 1694Dynamically-enabled feature bits need to be requested with 1695``arch_prctl()`` before calling this ioctl. Feature bits that have not 1696been requested are excluded from the result. 1697 1698Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may 1699expose cpuid features (e.g. MONITOR) which are not supported by kvm in 1700its default configuration. If userspace enables such capabilities, it 1701is responsible for modifying the results of this ioctl appropriately. 1702 1703Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1704with the 'nent' field indicating the number of entries in the variable-size 1705array 'entries'. If the number of entries is too low to describe the cpu 1706capabilities, an error (E2BIG) is returned. If the number is too high, 1707the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1708number is just right, the 'nent' field is adjusted to the number of valid 1709entries in the 'entries' array, which is then filled. 1710 1711The entries returned are the host cpuid as returned by the cpuid instruction, 1712with unknown or unsupported features masked out. Some features (for example, 1713x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1714emulate them efficiently. The fields in each entry are defined as follows: 1715 1716 function: 1717 the eax value used to obtain the entry 1718 1719 index: 1720 the ecx value used to obtain the entry (for entries that are 1721 affected by ecx) 1722 1723 flags: 1724 an OR of zero or more of the following: 1725 1726 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1727 if the index field is valid 1728 1729 eax, ebx, ecx, edx: 1730 the values returned by the cpuid instruction for 1731 this function/index combination 1732 1733The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned 1734as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC 1735support. Instead it is reported via:: 1736 1737 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1738 1739if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1740feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1741 1742 17434.47 KVM_PPC_GET_PVINFO 1744----------------------- 1745 1746:Capability: KVM_CAP_PPC_GET_PVINFO 1747:Architectures: ppc 1748:Type: vm ioctl 1749:Parameters: struct kvm_ppc_pvinfo (out) 1750:Returns: 0 on success, !0 on error 1751 1752:: 1753 1754 struct kvm_ppc_pvinfo { 1755 __u32 flags; 1756 __u32 hcall[4]; 1757 __u8 pad[108]; 1758 }; 1759 1760This ioctl fetches PV specific information that need to be passed to the guest 1761using the device tree or other means from vm context. 1762 1763The hcall array defines 4 instructions that make up a hypercall. 1764 1765If any additional field gets added to this structure later on, a bit for that 1766additional piece of information will be set in the flags bitmap. 1767 1768The flags bitmap is defined as:: 1769 1770 /* the host supports the ePAPR idle hcall 1771 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1772 17734.52 KVM_SET_GSI_ROUTING 1774------------------------ 1775 1776:Capability: KVM_CAP_IRQ_ROUTING 1777:Architectures: x86 s390 arm64 1778:Type: vm ioctl 1779:Parameters: struct kvm_irq_routing (in) 1780:Returns: 0 on success, -1 on error 1781 1782Sets the GSI routing table entries, overwriting any previously set entries. 1783 1784On arm64, GSI routing has the following limitation: 1785 1786- GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1787 1788:: 1789 1790 struct kvm_irq_routing { 1791 __u32 nr; 1792 __u32 flags; 1793 struct kvm_irq_routing_entry entries[0]; 1794 }; 1795 1796No flags are specified so far, the corresponding field must be set to zero. 1797 1798:: 1799 1800 struct kvm_irq_routing_entry { 1801 __u32 gsi; 1802 __u32 type; 1803 __u32 flags; 1804 __u32 pad; 1805 union { 1806 struct kvm_irq_routing_irqchip irqchip; 1807 struct kvm_irq_routing_msi msi; 1808 struct kvm_irq_routing_s390_adapter adapter; 1809 struct kvm_irq_routing_hv_sint hv_sint; 1810 struct kvm_irq_routing_xen_evtchn xen_evtchn; 1811 __u32 pad[8]; 1812 } u; 1813 }; 1814 1815 /* gsi routing entry types */ 1816 #define KVM_IRQ_ROUTING_IRQCHIP 1 1817 #define KVM_IRQ_ROUTING_MSI 2 1818 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1819 #define KVM_IRQ_ROUTING_HV_SINT 4 1820 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 1821 1822flags: 1823 1824- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1825 type, specifies that the devid field contains a valid value. The per-VM 1826 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1827 the device ID. If this capability is not available, userspace should 1828 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1829- zero otherwise 1830 1831:: 1832 1833 struct kvm_irq_routing_irqchip { 1834 __u32 irqchip; 1835 __u32 pin; 1836 }; 1837 1838 struct kvm_irq_routing_msi { 1839 __u32 address_lo; 1840 __u32 address_hi; 1841 __u32 data; 1842 union { 1843 __u32 pad; 1844 __u32 devid; 1845 }; 1846 }; 1847 1848If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1849for the device that wrote the MSI message. For PCI, this is usually a 1850BFD identifier in the lower 16 bits. 1851 1852On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1853feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1854address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1855address_hi must be zero. 1856 1857:: 1858 1859 struct kvm_irq_routing_s390_adapter { 1860 __u64 ind_addr; 1861 __u64 summary_addr; 1862 __u64 ind_offset; 1863 __u32 summary_offset; 1864 __u32 adapter_id; 1865 }; 1866 1867 struct kvm_irq_routing_hv_sint { 1868 __u32 vcpu; 1869 __u32 sint; 1870 }; 1871 1872 struct kvm_irq_routing_xen_evtchn { 1873 __u32 port; 1874 __u32 vcpu; 1875 __u32 priority; 1876 }; 1877 1878 1879When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL bit 1880in its indication of supported features, routing to Xen event channels 1881is supported. Although the priority field is present, only the value 1882KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported, which means delivery by 18832 level event channels. FIFO event channel support may be added in 1884the future. 1885 1886 18874.55 KVM_SET_TSC_KHZ 1888-------------------- 1889 1890:Capability: KVM_CAP_TSC_CONTROL 1891:Architectures: x86 1892:Type: vcpu ioctl 1893:Parameters: virtual tsc_khz 1894:Returns: 0 on success, -1 on error 1895 1896Specifies the tsc frequency for the virtual machine. The unit of the 1897frequency is KHz. 1898 1899 19004.56 KVM_GET_TSC_KHZ 1901-------------------- 1902 1903:Capability: KVM_CAP_GET_TSC_KHZ 1904:Architectures: x86 1905:Type: vcpu ioctl 1906:Parameters: none 1907:Returns: virtual tsc-khz on success, negative value on error 1908 1909Returns the tsc frequency of the guest. The unit of the return value is 1910KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 1911error. 1912 1913 19144.57 KVM_GET_LAPIC 1915------------------ 1916 1917:Capability: KVM_CAP_IRQCHIP 1918:Architectures: x86 1919:Type: vcpu ioctl 1920:Parameters: struct kvm_lapic_state (out) 1921:Returns: 0 on success, -1 on error 1922 1923:: 1924 1925 #define KVM_APIC_REG_SIZE 0x400 1926 struct kvm_lapic_state { 1927 char regs[KVM_APIC_REG_SIZE]; 1928 }; 1929 1930Reads the Local APIC registers and copies them into the input argument. The 1931data format and layout are the same as documented in the architecture manual. 1932 1933If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 1934enabled, then the format of APIC_ID register depends on the APIC mode 1935(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 1936the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 1937which is stored in bits 31-24 of the APIC register, or equivalently in 1938byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 1939be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 1940 1941If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 1942always uses xAPIC format. 1943 1944 19454.58 KVM_SET_LAPIC 1946------------------ 1947 1948:Capability: KVM_CAP_IRQCHIP 1949:Architectures: x86 1950:Type: vcpu ioctl 1951:Parameters: struct kvm_lapic_state (in) 1952:Returns: 0 on success, -1 on error 1953 1954:: 1955 1956 #define KVM_APIC_REG_SIZE 0x400 1957 struct kvm_lapic_state { 1958 char regs[KVM_APIC_REG_SIZE]; 1959 }; 1960 1961Copies the input argument into the Local APIC registers. The data format 1962and layout are the same as documented in the architecture manual. 1963 1964The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 1965regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 1966See the note in KVM_GET_LAPIC. 1967 1968 19694.59 KVM_IOEVENTFD 1970------------------ 1971 1972:Capability: KVM_CAP_IOEVENTFD 1973:Architectures: all 1974:Type: vm ioctl 1975:Parameters: struct kvm_ioeventfd (in) 1976:Returns: 0 on success, !0 on error 1977 1978This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 1979within the guest. A guest write in the registered address will signal the 1980provided event instead of triggering an exit. 1981 1982:: 1983 1984 struct kvm_ioeventfd { 1985 __u64 datamatch; 1986 __u64 addr; /* legal pio/mmio address */ 1987 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 1988 __s32 fd; 1989 __u32 flags; 1990 __u8 pad[36]; 1991 }; 1992 1993For the special case of virtio-ccw devices on s390, the ioevent is matched 1994to a subchannel/virtqueue tuple instead. 1995 1996The following flags are defined:: 1997 1998 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 1999 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 2000 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 2001 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 2002 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 2003 2004If datamatch flag is set, the event will be signaled only if the written value 2005to the registered address is equal to datamatch in struct kvm_ioeventfd. 2006 2007For virtio-ccw devices, addr contains the subchannel id and datamatch the 2008virtqueue index. 2009 2010With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 2011the kernel will ignore the length of guest write and may get a faster vmexit. 2012The speedup may only apply to specific architectures, but the ioeventfd will 2013work anyway. 2014 20154.60 KVM_DIRTY_TLB 2016------------------ 2017 2018:Capability: KVM_CAP_SW_TLB 2019:Architectures: ppc 2020:Type: vcpu ioctl 2021:Parameters: struct kvm_dirty_tlb (in) 2022:Returns: 0 on success, -1 on error 2023 2024:: 2025 2026 struct kvm_dirty_tlb { 2027 __u64 bitmap; 2028 __u32 num_dirty; 2029 }; 2030 2031This must be called whenever userspace has changed an entry in the shared 2032TLB, prior to calling KVM_RUN on the associated vcpu. 2033 2034The "bitmap" field is the userspace address of an array. This array 2035consists of a number of bits, equal to the total number of TLB entries as 2036determined by the last successful call to KVM_CONFIG_TLB, rounded up to the 2037nearest multiple of 64. 2038 2039Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 2040array. 2041 2042The array is little-endian: the bit 0 is the least significant bit of the 2043first byte, bit 8 is the least significant bit of the second byte, etc. 2044This avoids any complications with differing word sizes. 2045 2046The "num_dirty" field is a performance hint for KVM to determine whether it 2047should skip processing the bitmap and just invalidate everything. It must 2048be set to the number of set bits in the bitmap. 2049 2050 20514.62 KVM_CREATE_SPAPR_TCE 2052------------------------- 2053 2054:Capability: KVM_CAP_SPAPR_TCE 2055:Architectures: powerpc 2056:Type: vm ioctl 2057:Parameters: struct kvm_create_spapr_tce (in) 2058:Returns: file descriptor for manipulating the created TCE table 2059 2060This creates a virtual TCE (translation control entry) table, which 2061is an IOMMU for PAPR-style virtual I/O. It is used to translate 2062logical addresses used in virtual I/O into guest physical addresses, 2063and provides a scatter/gather capability for PAPR virtual I/O. 2064 2065:: 2066 2067 /* for KVM_CAP_SPAPR_TCE */ 2068 struct kvm_create_spapr_tce { 2069 __u64 liobn; 2070 __u32 window_size; 2071 }; 2072 2073The liobn field gives the logical IO bus number for which to create a 2074TCE table. The window_size field specifies the size of the DMA window 2075which this TCE table will translate - the table will contain one 64 2076bit TCE entry for every 4kiB of the DMA window. 2077 2078When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 2079table has been created using this ioctl(), the kernel will handle it 2080in real mode, updating the TCE table. H_PUT_TCE calls for other 2081liobns will cause a vm exit and must be handled by userspace. 2082 2083The return value is a file descriptor which can be passed to mmap(2) 2084to map the created TCE table into userspace. This lets userspace read 2085the entries written by kernel-handled H_PUT_TCE calls, and also lets 2086userspace update the TCE table directly which is useful in some 2087circumstances. 2088 2089 20904.63 KVM_ALLOCATE_RMA 2091--------------------- 2092 2093:Capability: KVM_CAP_PPC_RMA 2094:Architectures: powerpc 2095:Type: vm ioctl 2096:Parameters: struct kvm_allocate_rma (out) 2097:Returns: file descriptor for mapping the allocated RMA 2098 2099This allocates a Real Mode Area (RMA) from the pool allocated at boot 2100time by the kernel. An RMA is a physically-contiguous, aligned region 2101of memory used on older POWER processors to provide the memory which 2102will be accessed by real-mode (MMU off) accesses in a KVM guest. 2103POWER processors support a set of sizes for the RMA that usually 2104includes 64MB, 128MB, 256MB and some larger powers of two. 2105 2106:: 2107 2108 /* for KVM_ALLOCATE_RMA */ 2109 struct kvm_allocate_rma { 2110 __u64 rma_size; 2111 }; 2112 2113The return value is a file descriptor which can be passed to mmap(2) 2114to map the allocated RMA into userspace. The mapped area can then be 2115passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the 2116RMA for a virtual machine. The size of the RMA in bytes (which is 2117fixed at host kernel boot time) is returned in the rma_size field of 2118the argument structure. 2119 2120The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl 2121is supported; 2 if the processor requires all virtual machines to have 2122an RMA, or 1 if the processor can use an RMA but doesn't require it, 2123because it supports the Virtual RMA (VRMA) facility. 2124 2125 21264.64 KVM_NMI 2127------------ 2128 2129:Capability: KVM_CAP_USER_NMI 2130:Architectures: x86 2131:Type: vcpu ioctl 2132:Parameters: none 2133:Returns: 0 on success, -1 on error 2134 2135Queues an NMI on the thread's vcpu. Note this is well defined only 2136when KVM_CREATE_IRQCHIP has not been called, since this is an interface 2137between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 2138has been called, this interface is completely emulated within the kernel. 2139 2140To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 2141following algorithm: 2142 2143 - pause the vcpu 2144 - read the local APIC's state (KVM_GET_LAPIC) 2145 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 2146 - if so, issue KVM_NMI 2147 - resume the vcpu 2148 2149Some guests configure the LINT1 NMI input to cause a panic, aiding in 2150debugging. 2151 2152 21534.65 KVM_S390_UCAS_MAP 2154---------------------- 2155 2156:Capability: KVM_CAP_S390_UCONTROL 2157:Architectures: s390 2158:Type: vcpu ioctl 2159:Parameters: struct kvm_s390_ucas_mapping (in) 2160:Returns: 0 in case of success 2161 2162The parameter is defined like this:: 2163 2164 struct kvm_s390_ucas_mapping { 2165 __u64 user_addr; 2166 __u64 vcpu_addr; 2167 __u64 length; 2168 }; 2169 2170This ioctl maps the memory at "user_addr" with the length "length" to 2171the vcpu's address space starting at "vcpu_addr". All parameters need to 2172be aligned by 1 megabyte. 2173 2174 21754.66 KVM_S390_UCAS_UNMAP 2176------------------------ 2177 2178:Capability: KVM_CAP_S390_UCONTROL 2179:Architectures: s390 2180:Type: vcpu ioctl 2181:Parameters: struct kvm_s390_ucas_mapping (in) 2182:Returns: 0 in case of success 2183 2184The parameter is defined like this:: 2185 2186 struct kvm_s390_ucas_mapping { 2187 __u64 user_addr; 2188 __u64 vcpu_addr; 2189 __u64 length; 2190 }; 2191 2192This ioctl unmaps the memory in the vcpu's address space starting at 2193"vcpu_addr" with the length "length". The field "user_addr" is ignored. 2194All parameters need to be aligned by 1 megabyte. 2195 2196 21974.67 KVM_S390_VCPU_FAULT 2198------------------------ 2199 2200:Capability: KVM_CAP_S390_UCONTROL 2201:Architectures: s390 2202:Type: vcpu ioctl 2203:Parameters: vcpu absolute address (in) 2204:Returns: 0 in case of success 2205 2206This call creates a page table entry on the virtual cpu's address space 2207(for user controlled virtual machines) or the virtual machine's address 2208space (for regular virtual machines). This only works for minor faults, 2209thus it's recommended to access subject memory page via the user page 2210table upfront. This is useful to handle validity intercepts for user 2211controlled virtual machines to fault in the virtual cpu's lowcore pages 2212prior to calling the KVM_RUN ioctl. 2213 2214 22154.68 KVM_SET_ONE_REG 2216-------------------- 2217 2218:Capability: KVM_CAP_ONE_REG 2219:Architectures: all 2220:Type: vcpu ioctl 2221:Parameters: struct kvm_one_reg (in) 2222:Returns: 0 on success, negative value on failure 2223 2224Errors: 2225 2226 ====== ============================================================ 2227 ENOENT no such register 2228 EINVAL invalid register ID, or no such register or used with VMs in 2229 protected virtualization mode on s390 2230 EPERM (arm64) register access not allowed before vcpu finalization 2231 ====== ============================================================ 2232 2233(These error codes are indicative only: do not rely on a specific error 2234code being returned in a specific situation.) 2235 2236:: 2237 2238 struct kvm_one_reg { 2239 __u64 id; 2240 __u64 addr; 2241 }; 2242 2243Using this ioctl, a single vcpu register can be set to a specific value 2244defined by user space with the passed in struct kvm_one_reg, where id 2245refers to the register identifier as described below and addr is a pointer 2246to a variable with the respective size. There can be architecture agnostic 2247and architecture specific registers. Each have their own range of operation 2248and their own constants and width. To keep track of the implemented 2249registers, find a list below: 2250 2251 ======= =============================== ============ 2252 Arch Register Width (bits) 2253 ======= =============================== ============ 2254 PPC KVM_REG_PPC_HIOR 64 2255 PPC KVM_REG_PPC_IAC1 64 2256 PPC KVM_REG_PPC_IAC2 64 2257 PPC KVM_REG_PPC_IAC3 64 2258 PPC KVM_REG_PPC_IAC4 64 2259 PPC KVM_REG_PPC_DAC1 64 2260 PPC KVM_REG_PPC_DAC2 64 2261 PPC KVM_REG_PPC_DABR 64 2262 PPC KVM_REG_PPC_DSCR 64 2263 PPC KVM_REG_PPC_PURR 64 2264 PPC KVM_REG_PPC_SPURR 64 2265 PPC KVM_REG_PPC_DAR 64 2266 PPC KVM_REG_PPC_DSISR 32 2267 PPC KVM_REG_PPC_AMR 64 2268 PPC KVM_REG_PPC_UAMOR 64 2269 PPC KVM_REG_PPC_MMCR0 64 2270 PPC KVM_REG_PPC_MMCR1 64 2271 PPC KVM_REG_PPC_MMCRA 64 2272 PPC KVM_REG_PPC_MMCR2 64 2273 PPC KVM_REG_PPC_MMCRS 64 2274 PPC KVM_REG_PPC_MMCR3 64 2275 PPC KVM_REG_PPC_SIAR 64 2276 PPC KVM_REG_PPC_SDAR 64 2277 PPC KVM_REG_PPC_SIER 64 2278 PPC KVM_REG_PPC_SIER2 64 2279 PPC KVM_REG_PPC_SIER3 64 2280 PPC KVM_REG_PPC_PMC1 32 2281 PPC KVM_REG_PPC_PMC2 32 2282 PPC KVM_REG_PPC_PMC3 32 2283 PPC KVM_REG_PPC_PMC4 32 2284 PPC KVM_REG_PPC_PMC5 32 2285 PPC KVM_REG_PPC_PMC6 32 2286 PPC KVM_REG_PPC_PMC7 32 2287 PPC KVM_REG_PPC_PMC8 32 2288 PPC KVM_REG_PPC_FPR0 64 2289 ... 2290 PPC KVM_REG_PPC_FPR31 64 2291 PPC KVM_REG_PPC_VR0 128 2292 ... 2293 PPC KVM_REG_PPC_VR31 128 2294 PPC KVM_REG_PPC_VSR0 128 2295 ... 2296 PPC KVM_REG_PPC_VSR31 128 2297 PPC KVM_REG_PPC_FPSCR 64 2298 PPC KVM_REG_PPC_VSCR 32 2299 PPC KVM_REG_PPC_VPA_ADDR 64 2300 PPC KVM_REG_PPC_VPA_SLB 128 2301 PPC KVM_REG_PPC_VPA_DTL 128 2302 PPC KVM_REG_PPC_EPCR 32 2303 PPC KVM_REG_PPC_EPR 32 2304 PPC KVM_REG_PPC_TCR 32 2305 PPC KVM_REG_PPC_TSR 32 2306 PPC KVM_REG_PPC_OR_TSR 32 2307 PPC KVM_REG_PPC_CLEAR_TSR 32 2308 PPC KVM_REG_PPC_MAS0 32 2309 PPC KVM_REG_PPC_MAS1 32 2310 PPC KVM_REG_PPC_MAS2 64 2311 PPC KVM_REG_PPC_MAS7_3 64 2312 PPC KVM_REG_PPC_MAS4 32 2313 PPC KVM_REG_PPC_MAS6 32 2314 PPC KVM_REG_PPC_MMUCFG 32 2315 PPC KVM_REG_PPC_TLB0CFG 32 2316 PPC KVM_REG_PPC_TLB1CFG 32 2317 PPC KVM_REG_PPC_TLB2CFG 32 2318 PPC KVM_REG_PPC_TLB3CFG 32 2319 PPC KVM_REG_PPC_TLB0PS 32 2320 PPC KVM_REG_PPC_TLB1PS 32 2321 PPC KVM_REG_PPC_TLB2PS 32 2322 PPC KVM_REG_PPC_TLB3PS 32 2323 PPC KVM_REG_PPC_EPTCFG 32 2324 PPC KVM_REG_PPC_ICP_STATE 64 2325 PPC KVM_REG_PPC_VP_STATE 128 2326 PPC KVM_REG_PPC_TB_OFFSET 64 2327 PPC KVM_REG_PPC_SPMC1 32 2328 PPC KVM_REG_PPC_SPMC2 32 2329 PPC KVM_REG_PPC_IAMR 64 2330 PPC KVM_REG_PPC_TFHAR 64 2331 PPC KVM_REG_PPC_TFIAR 64 2332 PPC KVM_REG_PPC_TEXASR 64 2333 PPC KVM_REG_PPC_FSCR 64 2334 PPC KVM_REG_PPC_PSPB 32 2335 PPC KVM_REG_PPC_EBBHR 64 2336 PPC KVM_REG_PPC_EBBRR 64 2337 PPC KVM_REG_PPC_BESCR 64 2338 PPC KVM_REG_PPC_TAR 64 2339 PPC KVM_REG_PPC_DPDES 64 2340 PPC KVM_REG_PPC_DAWR 64 2341 PPC KVM_REG_PPC_DAWRX 64 2342 PPC KVM_REG_PPC_CIABR 64 2343 PPC KVM_REG_PPC_IC 64 2344 PPC KVM_REG_PPC_VTB 64 2345 PPC KVM_REG_PPC_CSIGR 64 2346 PPC KVM_REG_PPC_TACR 64 2347 PPC KVM_REG_PPC_TCSCR 64 2348 PPC KVM_REG_PPC_PID 64 2349 PPC KVM_REG_PPC_ACOP 64 2350 PPC KVM_REG_PPC_VRSAVE 32 2351 PPC KVM_REG_PPC_LPCR 32 2352 PPC KVM_REG_PPC_LPCR_64 64 2353 PPC KVM_REG_PPC_PPR 64 2354 PPC KVM_REG_PPC_ARCH_COMPAT 32 2355 PPC KVM_REG_PPC_DABRX 32 2356 PPC KVM_REG_PPC_WORT 64 2357 PPC KVM_REG_PPC_SPRG9 64 2358 PPC KVM_REG_PPC_DBSR 32 2359 PPC KVM_REG_PPC_TIDR 64 2360 PPC KVM_REG_PPC_PSSCR 64 2361 PPC KVM_REG_PPC_DEC_EXPIRY 64 2362 PPC KVM_REG_PPC_PTCR 64 2363 PPC KVM_REG_PPC_DAWR1 64 2364 PPC KVM_REG_PPC_DAWRX1 64 2365 PPC KVM_REG_PPC_TM_GPR0 64 2366 ... 2367 PPC KVM_REG_PPC_TM_GPR31 64 2368 PPC KVM_REG_PPC_TM_VSR0 128 2369 ... 2370 PPC KVM_REG_PPC_TM_VSR63 128 2371 PPC KVM_REG_PPC_TM_CR 64 2372 PPC KVM_REG_PPC_TM_LR 64 2373 PPC KVM_REG_PPC_TM_CTR 64 2374 PPC KVM_REG_PPC_TM_FPSCR 64 2375 PPC KVM_REG_PPC_TM_AMR 64 2376 PPC KVM_REG_PPC_TM_PPR 64 2377 PPC KVM_REG_PPC_TM_VRSAVE 64 2378 PPC KVM_REG_PPC_TM_VSCR 32 2379 PPC KVM_REG_PPC_TM_DSCR 64 2380 PPC KVM_REG_PPC_TM_TAR 64 2381 PPC KVM_REG_PPC_TM_XER 64 2382 2383 MIPS KVM_REG_MIPS_R0 64 2384 ... 2385 MIPS KVM_REG_MIPS_R31 64 2386 MIPS KVM_REG_MIPS_HI 64 2387 MIPS KVM_REG_MIPS_LO 64 2388 MIPS KVM_REG_MIPS_PC 64 2389 MIPS KVM_REG_MIPS_CP0_INDEX 32 2390 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2391 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2392 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2393 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2394 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2395 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2396 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2397 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2398 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2399 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2400 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2401 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2402 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2403 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2404 MIPS KVM_REG_MIPS_CP0_WIRED 32 2405 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2406 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2407 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2408 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2409 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2410 MIPS KVM_REG_MIPS_CP0_COUNT 32 2411 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2412 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2413 MIPS KVM_REG_MIPS_CP0_STATUS 32 2414 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2415 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2416 MIPS KVM_REG_MIPS_CP0_EPC 64 2417 MIPS KVM_REG_MIPS_CP0_PRID 32 2418 MIPS KVM_REG_MIPS_CP0_EBASE 64 2419 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2420 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2421 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2422 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2423 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2424 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2425 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2426 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2427 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2428 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2429 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2430 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2431 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2432 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2433 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2434 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2435 MIPS KVM_REG_MIPS_COUNT_CTL 64 2436 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2437 MIPS KVM_REG_MIPS_COUNT_HZ 64 2438 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2439 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2440 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2441 MIPS KVM_REG_MIPS_FCR_IR 32 2442 MIPS KVM_REG_MIPS_FCR_CSR 32 2443 MIPS KVM_REG_MIPS_MSA_IR 32 2444 MIPS KVM_REG_MIPS_MSA_CSR 32 2445 ======= =============================== ============ 2446 2447ARM registers are mapped using the lower 32 bits. The upper 16 of that 2448is the register group type, or coprocessor number: 2449 2450ARM core registers have the following id bit patterns:: 2451 2452 0x4020 0000 0010 <index into the kvm_regs struct:16> 2453 2454ARM 32-bit CP15 registers have the following id bit patterns:: 2455 2456 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 2457 2458ARM 64-bit CP15 registers have the following id bit patterns:: 2459 2460 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 2461 2462ARM CCSIDR registers are demultiplexed by CSSELR value:: 2463 2464 0x4020 0000 0011 00 <csselr:8> 2465 2466ARM 32-bit VFP control registers have the following id bit patterns:: 2467 2468 0x4020 0000 0012 1 <regno:12> 2469 2470ARM 64-bit FP registers have the following id bit patterns:: 2471 2472 0x4030 0000 0012 0 <regno:12> 2473 2474ARM firmware pseudo-registers have the following bit pattern:: 2475 2476 0x4030 0000 0014 <regno:16> 2477 2478 2479arm64 registers are mapped using the lower 32 bits. The upper 16 of 2480that is the register group type, or coprocessor number: 2481 2482arm64 core/FP-SIMD registers have the following id bit patterns. Note 2483that the size of the access is variable, as the kvm_regs structure 2484contains elements ranging from 32 to 128 bits. The index is a 32bit 2485value in the kvm_regs structure seen as a 32bit array:: 2486 2487 0x60x0 0000 0010 <index into the kvm_regs struct:16> 2488 2489Specifically: 2490 2491======================= ========= ===== ======================================= 2492 Encoding Register Bits kvm_regs member 2493======================= ========= ===== ======================================= 2494 0x6030 0000 0010 0000 X0 64 regs.regs[0] 2495 0x6030 0000 0010 0002 X1 64 regs.regs[1] 2496 ... 2497 0x6030 0000 0010 003c X30 64 regs.regs[30] 2498 0x6030 0000 0010 003e SP 64 regs.sp 2499 0x6030 0000 0010 0040 PC 64 regs.pc 2500 0x6030 0000 0010 0042 PSTATE 64 regs.pstate 2501 0x6030 0000 0010 0044 SP_EL1 64 sp_el1 2502 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1 2503 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC) 2504 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT] 2505 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND] 2506 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ] 2507 0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ] 2508 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_ 2509 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_ 2510 ... 2511 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_ 2512 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr 2513 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr 2514======================= ========= ===== ======================================= 2515 2516.. [1] These encodings are not accepted for SVE-enabled vcpus. See 2517 KVM_ARM_VCPU_INIT. 2518 2519 The equivalent register content can be accessed via bits [127:0] of 2520 the corresponding SVE Zn registers instead for vcpus that have SVE 2521 enabled (see below). 2522 2523arm64 CCSIDR registers are demultiplexed by CSSELR value:: 2524 2525 0x6020 0000 0011 00 <csselr:8> 2526 2527arm64 system registers have the following id bit patterns:: 2528 2529 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 2530 2531.. warning:: 2532 2533 Two system register IDs do not follow the specified pattern. These 2534 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to 2535 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These 2536 two had their values accidentally swapped, which means TIMER_CVAL is 2537 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is 2538 derived from the register encoding for CNTV_CVAL_EL0. As this is 2539 API, it must remain this way. 2540 2541arm64 firmware pseudo-registers have the following bit pattern:: 2542 2543 0x6030 0000 0014 <regno:16> 2544 2545arm64 SVE registers have the following bit patterns:: 2546 2547 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice] 2548 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice] 2549 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice] 2550 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register 2551 2552Access to register IDs where 2048 * slice >= 128 * max_vq will fail with 2553ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit 2554quadwords: see [2]_ below. 2555 2556These registers are only accessible on vcpus for which SVE is enabled. 2557See KVM_ARM_VCPU_INIT for details. 2558 2559In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not 2560accessible until the vcpu's SVE configuration has been finalized 2561using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT 2562and KVM_ARM_VCPU_FINALIZE for more information about this procedure. 2563 2564KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector 2565lengths supported by the vcpu to be discovered and configured by 2566userspace. When transferred to or from user memory via KVM_GET_ONE_REG 2567or KVM_SET_ONE_REG, the value of this register is of type 2568__u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as 2569follows:: 2570 2571 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS]; 2572 2573 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2574 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >> 2575 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1)) 2576 /* Vector length vq * 16 bytes supported */ 2577 else 2578 /* Vector length vq * 16 bytes not supported */ 2579 2580.. [2] The maximum value vq for which the above condition is true is 2581 max_vq. This is the maximum vector length available to the guest on 2582 this vcpu, and determines which register slices are visible through 2583 this ioctl interface. 2584 2585(See Documentation/arm64/sve.rst for an explanation of the "vq" 2586nomenclature.) 2587 2588KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT. 2589KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that 2590the host supports. 2591 2592Userspace may subsequently modify it if desired until the vcpu's SVE 2593configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). 2594 2595Apart from simply removing all vector lengths from the host set that 2596exceed some value, support for arbitrarily chosen sets of vector lengths 2597is hardware-dependent and may not be available. Attempting to configure 2598an invalid set of vector lengths via KVM_SET_ONE_REG will fail with 2599EINVAL. 2600 2601After the vcpu's SVE configuration is finalized, further attempts to 2602write this register will fail with EPERM. 2603 2604arm64 bitmap feature firmware pseudo-registers have the following bit pattern:: 2605 2606 0x6030 0000 0016 <regno:16> 2607 2608The bitmap feature firmware registers exposes the hypercall services that 2609are available for userspace to configure. The set bits corresponds to the 2610services that are available for the guests to access. By default, KVM 2611sets all the supported bits during VM initialization. The userspace can 2612discover the available services via KVM_GET_ONE_REG, and write back the 2613bitmap corresponding to the features that it wishes guests to see via 2614KVM_SET_ONE_REG. 2615 2616Note: These registers are immutable once any of the vCPUs of the VM has 2617run at least once. A KVM_SET_ONE_REG in such a scenario will return 2618a -EBUSY to userspace. 2619 2620(See Documentation/virt/kvm/arm/hypercalls.rst for more details.) 2621 2622 2623MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 2624the register group type: 2625 2626MIPS core registers (see above) have the following id bit patterns:: 2627 2628 0x7030 0000 0000 <reg:16> 2629 2630MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 2631patterns depending on whether they're 32-bit or 64-bit registers:: 2632 2633 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 2634 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2635 2636Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 2637versions of the EntryLo registers regardless of the word size of the host 2638hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 2639with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 2640the PFNX field starting at bit 30. 2641 2642MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 2643patterns:: 2644 2645 0x7030 0000 0001 01 <reg:8> 2646 2647MIPS KVM control registers (see above) have the following id bit patterns:: 2648 2649 0x7030 0000 0002 <reg:16> 2650 2651MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2652id bit patterns depending on the size of the register being accessed. They are 2653always accessed according to the current guest FPU mode (Status.FR and 2654Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2655if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2656registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2657overlap the FPU registers:: 2658 2659 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2660 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2661 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2662 2663MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2664following id bit patterns:: 2665 2666 0x7020 0000 0003 01 <0:3> <reg:5> 2667 2668MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2669following id bit patterns:: 2670 2671 0x7020 0000 0003 02 <0:3> <reg:5> 2672 2673RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of 2674that is the register group type. 2675 2676RISC-V config registers are meant for configuring a Guest VCPU and it has 2677the following id bit patterns:: 2678 2679 0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host) 2680 0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host) 2681 2682Following are the RISC-V config registers: 2683 2684======================= ========= ============================================= 2685 Encoding Register Description 2686======================= ========= ============================================= 2687 0x80x0 0000 0100 0000 isa ISA feature bitmap of Guest VCPU 2688======================= ========= ============================================= 2689 2690The isa config register can be read anytime but can only be written before 2691a Guest VCPU runs. It will have ISA feature bits matching underlying host 2692set by default. 2693 2694RISC-V core registers represent the general excution state of a Guest VCPU 2695and it has the following id bit patterns:: 2696 2697 0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host) 2698 0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host) 2699 2700Following are the RISC-V core registers: 2701 2702======================= ========= ============================================= 2703 Encoding Register Description 2704======================= ========= ============================================= 2705 0x80x0 0000 0200 0000 regs.pc Program counter 2706 0x80x0 0000 0200 0001 regs.ra Return address 2707 0x80x0 0000 0200 0002 regs.sp Stack pointer 2708 0x80x0 0000 0200 0003 regs.gp Global pointer 2709 0x80x0 0000 0200 0004 regs.tp Task pointer 2710 0x80x0 0000 0200 0005 regs.t0 Caller saved register 0 2711 0x80x0 0000 0200 0006 regs.t1 Caller saved register 1 2712 0x80x0 0000 0200 0007 regs.t2 Caller saved register 2 2713 0x80x0 0000 0200 0008 regs.s0 Callee saved register 0 2714 0x80x0 0000 0200 0009 regs.s1 Callee saved register 1 2715 0x80x0 0000 0200 000a regs.a0 Function argument (or return value) 0 2716 0x80x0 0000 0200 000b regs.a1 Function argument (or return value) 1 2717 0x80x0 0000 0200 000c regs.a2 Function argument 2 2718 0x80x0 0000 0200 000d regs.a3 Function argument 3 2719 0x80x0 0000 0200 000e regs.a4 Function argument 4 2720 0x80x0 0000 0200 000f regs.a5 Function argument 5 2721 0x80x0 0000 0200 0010 regs.a6 Function argument 6 2722 0x80x0 0000 0200 0011 regs.a7 Function argument 7 2723 0x80x0 0000 0200 0012 regs.s2 Callee saved register 2 2724 0x80x0 0000 0200 0013 regs.s3 Callee saved register 3 2725 0x80x0 0000 0200 0014 regs.s4 Callee saved register 4 2726 0x80x0 0000 0200 0015 regs.s5 Callee saved register 5 2727 0x80x0 0000 0200 0016 regs.s6 Callee saved register 6 2728 0x80x0 0000 0200 0017 regs.s7 Callee saved register 7 2729 0x80x0 0000 0200 0018 regs.s8 Callee saved register 8 2730 0x80x0 0000 0200 0019 regs.s9 Callee saved register 9 2731 0x80x0 0000 0200 001a regs.s10 Callee saved register 10 2732 0x80x0 0000 0200 001b regs.s11 Callee saved register 11 2733 0x80x0 0000 0200 001c regs.t3 Caller saved register 3 2734 0x80x0 0000 0200 001d regs.t4 Caller saved register 4 2735 0x80x0 0000 0200 001e regs.t5 Caller saved register 5 2736 0x80x0 0000 0200 001f regs.t6 Caller saved register 6 2737 0x80x0 0000 0200 0020 mode Privilege mode (1 = S-mode or 0 = U-mode) 2738======================= ========= ============================================= 2739 2740RISC-V csr registers represent the supervisor mode control/status registers 2741of a Guest VCPU and it has the following id bit patterns:: 2742 2743 0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host) 2744 0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host) 2745 2746Following are the RISC-V csr registers: 2747 2748======================= ========= ============================================= 2749 Encoding Register Description 2750======================= ========= ============================================= 2751 0x80x0 0000 0300 0000 sstatus Supervisor status 2752 0x80x0 0000 0300 0001 sie Supervisor interrupt enable 2753 0x80x0 0000 0300 0002 stvec Supervisor trap vector base 2754 0x80x0 0000 0300 0003 sscratch Supervisor scratch register 2755 0x80x0 0000 0300 0004 sepc Supervisor exception program counter 2756 0x80x0 0000 0300 0005 scause Supervisor trap cause 2757 0x80x0 0000 0300 0006 stval Supervisor bad address or instruction 2758 0x80x0 0000 0300 0007 sip Supervisor interrupt pending 2759 0x80x0 0000 0300 0008 satp Supervisor address translation and protection 2760======================= ========= ============================================= 2761 2762RISC-V timer registers represent the timer state of a Guest VCPU and it has 2763the following id bit patterns:: 2764 2765 0x8030 0000 04 <index into the kvm_riscv_timer struct:24> 2766 2767Following are the RISC-V timer registers: 2768 2769======================= ========= ============================================= 2770 Encoding Register Description 2771======================= ========= ============================================= 2772 0x8030 0000 0400 0000 frequency Time base frequency (read-only) 2773 0x8030 0000 0400 0001 time Time value visible to Guest 2774 0x8030 0000 0400 0002 compare Time compare programmed by Guest 2775 0x8030 0000 0400 0003 state Time compare state (1 = ON or 0 = OFF) 2776======================= ========= ============================================= 2777 2778RISC-V F-extension registers represent the single precision floating point 2779state of a Guest VCPU and it has the following id bit patterns:: 2780 2781 0x8020 0000 05 <index into the __riscv_f_ext_state struct:24> 2782 2783Following are the RISC-V F-extension registers: 2784 2785======================= ========= ============================================= 2786 Encoding Register Description 2787======================= ========= ============================================= 2788 0x8020 0000 0500 0000 f[0] Floating point register 0 2789 ... 2790 0x8020 0000 0500 001f f[31] Floating point register 31 2791 0x8020 0000 0500 0020 fcsr Floating point control and status register 2792======================= ========= ============================================= 2793 2794RISC-V D-extension registers represent the double precision floating point 2795state of a Guest VCPU and it has the following id bit patterns:: 2796 2797 0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr) 2798 0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr) 2799 2800Following are the RISC-V D-extension registers: 2801 2802======================= ========= ============================================= 2803 Encoding Register Description 2804======================= ========= ============================================= 2805 0x8030 0000 0600 0000 f[0] Floating point register 0 2806 ... 2807 0x8030 0000 0600 001f f[31] Floating point register 31 2808 0x8020 0000 0600 0020 fcsr Floating point control and status register 2809======================= ========= ============================================= 2810 2811 28124.69 KVM_GET_ONE_REG 2813-------------------- 2814 2815:Capability: KVM_CAP_ONE_REG 2816:Architectures: all 2817:Type: vcpu ioctl 2818:Parameters: struct kvm_one_reg (in and out) 2819:Returns: 0 on success, negative value on failure 2820 2821Errors include: 2822 2823 ======== ============================================================ 2824 ENOENT no such register 2825 EINVAL invalid register ID, or no such register or used with VMs in 2826 protected virtualization mode on s390 2827 EPERM (arm64) register access not allowed before vcpu finalization 2828 ======== ============================================================ 2829 2830(These error codes are indicative only: do not rely on a specific error 2831code being returned in a specific situation.) 2832 2833This ioctl allows to receive the value of a single register implemented 2834in a vcpu. The register to read is indicated by the "id" field of the 2835kvm_one_reg struct passed in. On success, the register value can be found 2836at the memory location pointed to by "addr". 2837 2838The list of registers accessible using this interface is identical to the 2839list in 4.68. 2840 2841 28424.70 KVM_KVMCLOCK_CTRL 2843---------------------- 2844 2845:Capability: KVM_CAP_KVMCLOCK_CTRL 2846:Architectures: Any that implement pvclocks (currently x86 only) 2847:Type: vcpu ioctl 2848:Parameters: None 2849:Returns: 0 on success, -1 on error 2850 2851This ioctl sets a flag accessible to the guest indicating that the specified 2852vCPU has been paused by the host userspace. 2853 2854The host will set a flag in the pvclock structure that is checked from the 2855soft lockup watchdog. The flag is part of the pvclock structure that is 2856shared between guest and host, specifically the second bit of the flags 2857field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2858the host and read/cleared exclusively by the guest. The guest operation of 2859checking and clearing the flag must be an atomic operation so 2860load-link/store-conditional, or equivalent must be used. There are two cases 2861where the guest will clear the flag: when the soft lockup watchdog timer resets 2862itself or when a soft lockup is detected. This ioctl can be called any time 2863after pausing the vcpu, but before it is resumed. 2864 2865 28664.71 KVM_SIGNAL_MSI 2867------------------- 2868 2869:Capability: KVM_CAP_SIGNAL_MSI 2870:Architectures: x86 arm64 2871:Type: vm ioctl 2872:Parameters: struct kvm_msi (in) 2873:Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2874 2875Directly inject a MSI message. Only valid with in-kernel irqchip that handles 2876MSI messages. 2877 2878:: 2879 2880 struct kvm_msi { 2881 __u32 address_lo; 2882 __u32 address_hi; 2883 __u32 data; 2884 __u32 flags; 2885 __u32 devid; 2886 __u8 pad[12]; 2887 }; 2888 2889flags: 2890 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 2891 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 2892 the device ID. If this capability is not available, userspace 2893 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 2894 2895If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 2896for the device that wrote the MSI message. For PCI, this is usually a 2897BFD identifier in the lower 16 bits. 2898 2899On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 2900feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 2901address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 2902address_hi must be zero. 2903 2904 29054.71 KVM_CREATE_PIT2 2906-------------------- 2907 2908:Capability: KVM_CAP_PIT2 2909:Architectures: x86 2910:Type: vm ioctl 2911:Parameters: struct kvm_pit_config (in) 2912:Returns: 0 on success, -1 on error 2913 2914Creates an in-kernel device model for the i8254 PIT. This call is only valid 2915after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 2916parameters have to be passed:: 2917 2918 struct kvm_pit_config { 2919 __u32 flags; 2920 __u32 pad[15]; 2921 }; 2922 2923Valid flags are:: 2924 2925 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 2926 2927PIT timer interrupts may use a per-VM kernel thread for injection. If it 2928exists, this thread will have a name of the following pattern:: 2929 2930 kvm-pit/<owner-process-pid> 2931 2932When running a guest with elevated priorities, the scheduling parameters of 2933this thread may have to be adjusted accordingly. 2934 2935This IOCTL replaces the obsolete KVM_CREATE_PIT. 2936 2937 29384.72 KVM_GET_PIT2 2939----------------- 2940 2941:Capability: KVM_CAP_PIT_STATE2 2942:Architectures: x86 2943:Type: vm ioctl 2944:Parameters: struct kvm_pit_state2 (out) 2945:Returns: 0 on success, -1 on error 2946 2947Retrieves the state of the in-kernel PIT model. Only valid after 2948KVM_CREATE_PIT2. The state is returned in the following structure:: 2949 2950 struct kvm_pit_state2 { 2951 struct kvm_pit_channel_state channels[3]; 2952 __u32 flags; 2953 __u32 reserved[9]; 2954 }; 2955 2956Valid flags are:: 2957 2958 /* disable PIT in HPET legacy mode */ 2959 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 2960 2961This IOCTL replaces the obsolete KVM_GET_PIT. 2962 2963 29644.73 KVM_SET_PIT2 2965----------------- 2966 2967:Capability: KVM_CAP_PIT_STATE2 2968:Architectures: x86 2969:Type: vm ioctl 2970:Parameters: struct kvm_pit_state2 (in) 2971:Returns: 0 on success, -1 on error 2972 2973Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 2974See KVM_GET_PIT2 for details on struct kvm_pit_state2. 2975 2976This IOCTL replaces the obsolete KVM_SET_PIT. 2977 2978 29794.74 KVM_PPC_GET_SMMU_INFO 2980-------------------------- 2981 2982:Capability: KVM_CAP_PPC_GET_SMMU_INFO 2983:Architectures: powerpc 2984:Type: vm ioctl 2985:Parameters: None 2986:Returns: 0 on success, -1 on error 2987 2988This populates and returns a structure describing the features of 2989the "Server" class MMU emulation supported by KVM. 2990This can in turn be used by userspace to generate the appropriate 2991device-tree properties for the guest operating system. 2992 2993The structure contains some global information, followed by an 2994array of supported segment page sizes:: 2995 2996 struct kvm_ppc_smmu_info { 2997 __u64 flags; 2998 __u32 slb_size; 2999 __u32 pad; 3000 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3001 }; 3002 3003The supported flags are: 3004 3005 - KVM_PPC_PAGE_SIZES_REAL: 3006 When that flag is set, guest page sizes must "fit" the backing 3007 store page sizes. When not set, any page size in the list can 3008 be used regardless of how they are backed by userspace. 3009 3010 - KVM_PPC_1T_SEGMENTS 3011 The emulated MMU supports 1T segments in addition to the 3012 standard 256M ones. 3013 3014 - KVM_PPC_NO_HASH 3015 This flag indicates that HPT guests are not supported by KVM, 3016 thus all guests must use radix MMU mode. 3017 3018The "slb_size" field indicates how many SLB entries are supported 3019 3020The "sps" array contains 8 entries indicating the supported base 3021page sizes for a segment in increasing order. Each entry is defined 3022as follow:: 3023 3024 struct kvm_ppc_one_seg_page_size { 3025 __u32 page_shift; /* Base page shift of segment (or 0) */ 3026 __u32 slb_enc; /* SLB encoding for BookS */ 3027 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3028 }; 3029 3030An entry with a "page_shift" of 0 is unused. Because the array is 3031organized in increasing order, a lookup can stop when encoutering 3032such an entry. 3033 3034The "slb_enc" field provides the encoding to use in the SLB for the 3035page size. The bits are in positions such as the value can directly 3036be OR'ed into the "vsid" argument of the slbmte instruction. 3037 3038The "enc" array is a list which for each of those segment base page 3039size provides the list of supported actual page sizes (which can be 3040only larger or equal to the base page size), along with the 3041corresponding encoding in the hash PTE. Similarly, the array is 30428 entries sorted by increasing sizes and an entry with a "0" shift 3043is an empty entry and a terminator:: 3044 3045 struct kvm_ppc_one_page_size { 3046 __u32 page_shift; /* Page shift (or 0) */ 3047 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 3048 }; 3049 3050The "pte_enc" field provides a value that can OR'ed into the hash 3051PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 3052into the hash PTE second double word). 3053 30544.75 KVM_IRQFD 3055-------------- 3056 3057:Capability: KVM_CAP_IRQFD 3058:Architectures: x86 s390 arm64 3059:Type: vm ioctl 3060:Parameters: struct kvm_irqfd (in) 3061:Returns: 0 on success, -1 on error 3062 3063Allows setting an eventfd to directly trigger a guest interrupt. 3064kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 3065kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 3066an event is triggered on the eventfd, an interrupt is injected into 3067the guest using the specified gsi pin. The irqfd is removed using 3068the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 3069and kvm_irqfd.gsi. 3070 3071With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 3072mechanism allowing emulation of level-triggered, irqfd-based 3073interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 3074additional eventfd in the kvm_irqfd.resamplefd field. When operating 3075in resample mode, posting of an interrupt through kvm_irq.fd asserts 3076the specified gsi in the irqchip. When the irqchip is resampled, such 3077as from an EOI, the gsi is de-asserted and the user is notified via 3078kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 3079the interrupt if the device making use of it still requires service. 3080Note that closing the resamplefd is not sufficient to disable the 3081irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 3082and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 3083 3084On arm64, gsi routing being supported, the following can happen: 3085 3086- in case no routing entry is associated to this gsi, injection fails 3087- in case the gsi is associated to an irqchip routing entry, 3088 irqchip.pin + 32 corresponds to the injected SPI ID. 3089- in case the gsi is associated to an MSI routing entry, the MSI 3090 message and device ID are translated into an LPI (support restricted 3091 to GICv3 ITS in-kernel emulation). 3092 30934.76 KVM_PPC_ALLOCATE_HTAB 3094-------------------------- 3095 3096:Capability: KVM_CAP_PPC_ALLOC_HTAB 3097:Architectures: powerpc 3098:Type: vm ioctl 3099:Parameters: Pointer to u32 containing hash table order (in/out) 3100:Returns: 0 on success, -1 on error 3101 3102This requests the host kernel to allocate an MMU hash table for a 3103guest using the PAPR paravirtualization interface. This only does 3104anything if the kernel is configured to use the Book 3S HV style of 3105virtualization. Otherwise the capability doesn't exist and the ioctl 3106returns an ENOTTY error. The rest of this description assumes Book 3S 3107HV. 3108 3109There must be no vcpus running when this ioctl is called; if there 3110are, it will do nothing and return an EBUSY error. 3111 3112The parameter is a pointer to a 32-bit unsigned integer variable 3113containing the order (log base 2) of the desired size of the hash 3114table, which must be between 18 and 46. On successful return from the 3115ioctl, the value will not be changed by the kernel. 3116 3117If no hash table has been allocated when any vcpu is asked to run 3118(with the KVM_RUN ioctl), the host kernel will allocate a 3119default-sized hash table (16 MB). 3120 3121If this ioctl is called when a hash table has already been allocated, 3122with a different order from the existing hash table, the existing hash 3123table will be freed and a new one allocated. If this is ioctl is 3124called when a hash table has already been allocated of the same order 3125as specified, the kernel will clear out the existing hash table (zero 3126all HPTEs). In either case, if the guest is using the virtualized 3127real-mode area (VRMA) facility, the kernel will re-create the VMRA 3128HPTEs on the next KVM_RUN of any vcpu. 3129 31304.77 KVM_S390_INTERRUPT 3131----------------------- 3132 3133:Capability: basic 3134:Architectures: s390 3135:Type: vm ioctl, vcpu ioctl 3136:Parameters: struct kvm_s390_interrupt (in) 3137:Returns: 0 on success, -1 on error 3138 3139Allows to inject an interrupt to the guest. Interrupts can be floating 3140(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 3141 3142Interrupt parameters are passed via kvm_s390_interrupt:: 3143 3144 struct kvm_s390_interrupt { 3145 __u32 type; 3146 __u32 parm; 3147 __u64 parm64; 3148 }; 3149 3150type can be one of the following: 3151 3152KVM_S390_SIGP_STOP (vcpu) 3153 - sigp stop; optional flags in parm 3154KVM_S390_PROGRAM_INT (vcpu) 3155 - program check; code in parm 3156KVM_S390_SIGP_SET_PREFIX (vcpu) 3157 - sigp set prefix; prefix address in parm 3158KVM_S390_RESTART (vcpu) 3159 - restart 3160KVM_S390_INT_CLOCK_COMP (vcpu) 3161 - clock comparator interrupt 3162KVM_S390_INT_CPU_TIMER (vcpu) 3163 - CPU timer interrupt 3164KVM_S390_INT_VIRTIO (vm) 3165 - virtio external interrupt; external interrupt 3166 parameters in parm and parm64 3167KVM_S390_INT_SERVICE (vm) 3168 - sclp external interrupt; sclp parameter in parm 3169KVM_S390_INT_EMERGENCY (vcpu) 3170 - sigp emergency; source cpu in parm 3171KVM_S390_INT_EXTERNAL_CALL (vcpu) 3172 - sigp external call; source cpu in parm 3173KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 3174 - compound value to indicate an 3175 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 3176 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 3177 interruption subclass) 3178KVM_S390_MCHK (vm, vcpu) 3179 - machine check interrupt; cr 14 bits in parm, machine check interrupt 3180 code in parm64 (note that machine checks needing further payload are not 3181 supported by this ioctl) 3182 3183This is an asynchronous vcpu ioctl and can be invoked from any thread. 3184 31854.78 KVM_PPC_GET_HTAB_FD 3186------------------------ 3187 3188:Capability: KVM_CAP_PPC_HTAB_FD 3189:Architectures: powerpc 3190:Type: vm ioctl 3191:Parameters: Pointer to struct kvm_get_htab_fd (in) 3192:Returns: file descriptor number (>= 0) on success, -1 on error 3193 3194This returns a file descriptor that can be used either to read out the 3195entries in the guest's hashed page table (HPT), or to write entries to 3196initialize the HPT. The returned fd can only be written to if the 3197KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 3198can only be read if that bit is clear. The argument struct looks like 3199this:: 3200 3201 /* For KVM_PPC_GET_HTAB_FD */ 3202 struct kvm_get_htab_fd { 3203 __u64 flags; 3204 __u64 start_index; 3205 __u64 reserved[2]; 3206 }; 3207 3208 /* Values for kvm_get_htab_fd.flags */ 3209 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 3210 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 3211 3212The 'start_index' field gives the index in the HPT of the entry at 3213which to start reading. It is ignored when writing. 3214 3215Reads on the fd will initially supply information about all 3216"interesting" HPT entries. Interesting entries are those with the 3217bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 3218all entries. When the end of the HPT is reached, the read() will 3219return. If read() is called again on the fd, it will start again from 3220the beginning of the HPT, but will only return HPT entries that have 3221changed since they were last read. 3222 3223Data read or written is structured as a header (8 bytes) followed by a 3224series of valid HPT entries (16 bytes) each. The header indicates how 3225many valid HPT entries there are and how many invalid entries follow 3226the valid entries. The invalid entries are not represented explicitly 3227in the stream. The header format is:: 3228 3229 struct kvm_get_htab_header { 3230 __u32 index; 3231 __u16 n_valid; 3232 __u16 n_invalid; 3233 }; 3234 3235Writes to the fd create HPT entries starting at the index given in the 3236header; first 'n_valid' valid entries with contents from the data 3237written, then 'n_invalid' invalid entries, invalidating any previously 3238valid entries found. 3239 32404.79 KVM_CREATE_DEVICE 3241---------------------- 3242 3243:Capability: KVM_CAP_DEVICE_CTRL 3244:Type: vm ioctl 3245:Parameters: struct kvm_create_device (in/out) 3246:Returns: 0 on success, -1 on error 3247 3248Errors: 3249 3250 ====== ======================================================= 3251 ENODEV The device type is unknown or unsupported 3252 EEXIST Device already created, and this type of device may not 3253 be instantiated multiple times 3254 ====== ======================================================= 3255 3256 Other error conditions may be defined by individual device types or 3257 have their standard meanings. 3258 3259Creates an emulated device in the kernel. The file descriptor returned 3260in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 3261 3262If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 3263device type is supported (not necessarily whether it can be created 3264in the current vm). 3265 3266Individual devices should not define flags. Attributes should be used 3267for specifying any behavior that is not implied by the device type 3268number. 3269 3270:: 3271 3272 struct kvm_create_device { 3273 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 3274 __u32 fd; /* out: device handle */ 3275 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 3276 }; 3277 32784.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3279-------------------------------------------- 3280 3281:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3282 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3283 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device (no set) 3284:Type: device ioctl, vm ioctl, vcpu ioctl 3285:Parameters: struct kvm_device_attr 3286:Returns: 0 on success, -1 on error 3287 3288Errors: 3289 3290 ===== ============================================================= 3291 ENXIO The group or attribute is unknown/unsupported for this device 3292 or hardware support is missing. 3293 EPERM The attribute cannot (currently) be accessed this way 3294 (e.g. read-only attribute, or attribute that only makes 3295 sense when the device is in a different state) 3296 ===== ============================================================= 3297 3298 Other error conditions may be defined by individual device types. 3299 3300Gets/sets a specified piece of device configuration and/or state. The 3301semantics are device-specific. See individual device documentation in 3302the "devices" directory. As with ONE_REG, the size of the data 3303transferred is defined by the particular attribute. 3304 3305:: 3306 3307 struct kvm_device_attr { 3308 __u32 flags; /* no flags currently defined */ 3309 __u32 group; /* device-defined */ 3310 __u64 attr; /* group-defined */ 3311 __u64 addr; /* userspace address of attr data */ 3312 }; 3313 33144.81 KVM_HAS_DEVICE_ATTR 3315------------------------ 3316 3317:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3318 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3319 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device 3320:Type: device ioctl, vm ioctl, vcpu ioctl 3321:Parameters: struct kvm_device_attr 3322:Returns: 0 on success, -1 on error 3323 3324Errors: 3325 3326 ===== ============================================================= 3327 ENXIO The group or attribute is unknown/unsupported for this device 3328 or hardware support is missing. 3329 ===== ============================================================= 3330 3331Tests whether a device supports a particular attribute. A successful 3332return indicates the attribute is implemented. It does not necessarily 3333indicate that the attribute can be read or written in the device's 3334current state. "addr" is ignored. 3335 33364.82 KVM_ARM_VCPU_INIT 3337---------------------- 3338 3339:Capability: basic 3340:Architectures: arm64 3341:Type: vcpu ioctl 3342:Parameters: struct kvm_vcpu_init (in) 3343:Returns: 0 on success; -1 on error 3344 3345Errors: 3346 3347 ====== ================================================================= 3348 EINVAL the target is unknown, or the combination of features is invalid. 3349 ENOENT a features bit specified is unknown. 3350 ====== ================================================================= 3351 3352This tells KVM what type of CPU to present to the guest, and what 3353optional features it should have. This will cause a reset of the cpu 3354registers to their initial values. If this is not called, KVM_RUN will 3355return ENOEXEC for that vcpu. 3356 3357The initial values are defined as: 3358 - Processor state: 3359 * AArch64: EL1h, D, A, I and F bits set. All other bits 3360 are cleared. 3361 * AArch32: SVC, A, I and F bits set. All other bits are 3362 cleared. 3363 - General Purpose registers, including PC and SP: set to 0 3364 - FPSIMD/NEON registers: set to 0 3365 - SVE registers: set to 0 3366 - System registers: Reset to their architecturally defined 3367 values as for a warm reset to EL1 (resp. SVC) 3368 3369Note that because some registers reflect machine topology, all vcpus 3370should be created before this ioctl is invoked. 3371 3372Userspace can call this function multiple times for a given vcpu, including 3373after the vcpu has been run. This will reset the vcpu to its initial 3374state. All calls to this function after the initial call must use the same 3375target and same set of feature flags, otherwise EINVAL will be returned. 3376 3377Possible features: 3378 3379 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 3380 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 3381 and execute guest code when KVM_RUN is called. 3382 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 3383 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 3384 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision 3385 backward compatible with v0.2) for the CPU. 3386 Depends on KVM_CAP_ARM_PSCI_0_2. 3387 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 3388 Depends on KVM_CAP_ARM_PMU_V3. 3389 3390 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication 3391 for arm64 only. 3392 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS. 3393 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3394 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3395 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3396 requested. 3397 3398 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication 3399 for arm64 only. 3400 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC. 3401 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3402 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3403 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3404 requested. 3405 3406 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only). 3407 Depends on KVM_CAP_ARM_SVE. 3408 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3409 3410 * After KVM_ARM_VCPU_INIT: 3411 3412 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the 3413 initial value of this pseudo-register indicates the best set of 3414 vector lengths possible for a vcpu on this host. 3415 3416 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3417 3418 - KVM_RUN and KVM_GET_REG_LIST are not available; 3419 3420 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access 3421 the scalable archietctural SVE registers 3422 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or 3423 KVM_REG_ARM64_SVE_FFR; 3424 3425 - KVM_REG_ARM64_SVE_VLS may optionally be written using 3426 KVM_SET_ONE_REG, to modify the set of vector lengths available 3427 for the vcpu. 3428 3429 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3430 3431 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can 3432 no longer be written using KVM_SET_ONE_REG. 3433 34344.83 KVM_ARM_PREFERRED_TARGET 3435----------------------------- 3436 3437:Capability: basic 3438:Architectures: arm64 3439:Type: vm ioctl 3440:Parameters: struct kvm_vcpu_init (out) 3441:Returns: 0 on success; -1 on error 3442 3443Errors: 3444 3445 ====== ========================================== 3446 ENODEV no preferred target available for the host 3447 ====== ========================================== 3448 3449This queries KVM for preferred CPU target type which can be emulated 3450by KVM on underlying host. 3451 3452The ioctl returns struct kvm_vcpu_init instance containing information 3453about preferred CPU target type and recommended features for it. The 3454kvm_vcpu_init->features bitmap returned will have feature bits set if 3455the preferred target recommends setting these features, but this is 3456not mandatory. 3457 3458The information returned by this ioctl can be used to prepare an instance 3459of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 3460VCPU matching underlying host. 3461 3462 34634.84 KVM_GET_REG_LIST 3464--------------------- 3465 3466:Capability: basic 3467:Architectures: arm64, mips 3468:Type: vcpu ioctl 3469:Parameters: struct kvm_reg_list (in/out) 3470:Returns: 0 on success; -1 on error 3471 3472Errors: 3473 3474 ===== ============================================================== 3475 E2BIG the reg index list is too big to fit in the array specified by 3476 the user (the number required will be written into n). 3477 ===== ============================================================== 3478 3479:: 3480 3481 struct kvm_reg_list { 3482 __u64 n; /* number of registers in reg[] */ 3483 __u64 reg[0]; 3484 }; 3485 3486This ioctl returns the guest registers that are supported for the 3487KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3488 3489 34904.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3491----------------------------------------- 3492 3493:Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3494:Architectures: arm64 3495:Type: vm ioctl 3496:Parameters: struct kvm_arm_device_address (in) 3497:Returns: 0 on success, -1 on error 3498 3499Errors: 3500 3501 ====== ============================================ 3502 ENODEV The device id is unknown 3503 ENXIO Device not supported on current system 3504 EEXIST Address already set 3505 E2BIG Address outside guest physical address space 3506 EBUSY Address overlaps with other device range 3507 ====== ============================================ 3508 3509:: 3510 3511 struct kvm_arm_device_addr { 3512 __u64 id; 3513 __u64 addr; 3514 }; 3515 3516Specify a device address in the guest's physical address space where guests 3517can access emulated or directly exposed devices, which the host kernel needs 3518to know about. The id field is an architecture specific identifier for a 3519specific device. 3520 3521arm64 divides the id field into two parts, a device id and an 3522address type id specific to the individual device:: 3523 3524 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 3525 field: | 0x00000000 | device id | addr type id | 3526 3527arm64 currently only require this when using the in-kernel GIC 3528support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 3529as the device id. When setting the base address for the guest's 3530mapping of the VGIC virtual CPU and distributor interface, the ioctl 3531must be called after calling KVM_CREATE_IRQCHIP, but before calling 3532KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 3533base addresses will return -EEXIST. 3534 3535Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 3536should be used instead. 3537 3538 35394.86 KVM_PPC_RTAS_DEFINE_TOKEN 3540------------------------------ 3541 3542:Capability: KVM_CAP_PPC_RTAS 3543:Architectures: ppc 3544:Type: vm ioctl 3545:Parameters: struct kvm_rtas_token_args 3546:Returns: 0 on success, -1 on error 3547 3548Defines a token value for a RTAS (Run Time Abstraction Services) 3549service in order to allow it to be handled in the kernel. The 3550argument struct gives the name of the service, which must be the name 3551of a service that has a kernel-side implementation. If the token 3552value is non-zero, it will be associated with that service, and 3553subsequent RTAS calls by the guest specifying that token will be 3554handled by the kernel. If the token value is 0, then any token 3555associated with the service will be forgotten, and subsequent RTAS 3556calls by the guest for that service will be passed to userspace to be 3557handled. 3558 35594.87 KVM_SET_GUEST_DEBUG 3560------------------------ 3561 3562:Capability: KVM_CAP_SET_GUEST_DEBUG 3563:Architectures: x86, s390, ppc, arm64 3564:Type: vcpu ioctl 3565:Parameters: struct kvm_guest_debug (in) 3566:Returns: 0 on success; -1 on error 3567 3568:: 3569 3570 struct kvm_guest_debug { 3571 __u32 control; 3572 __u32 pad; 3573 struct kvm_guest_debug_arch arch; 3574 }; 3575 3576Set up the processor specific debug registers and configure vcpu for 3577handling guest debug events. There are two parts to the structure, the 3578first a control bitfield indicates the type of debug events to handle 3579when running. Common control bits are: 3580 3581 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 3582 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 3583 3584The top 16 bits of the control field are architecture specific control 3585flags which can include the following: 3586 3587 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 3588 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390] 3589 - KVM_GUESTDBG_USE_HW: using hardware debug events [arm64] 3590 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 3591 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 3592 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 3593 - KVM_GUESTDBG_BLOCKIRQ: avoid injecting interrupts/NMI/SMI [x86] 3594 3595For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 3596are enabled in memory so we need to ensure breakpoint exceptions are 3597correctly trapped and the KVM run loop exits at the breakpoint and not 3598running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 3599we need to ensure the guest vCPUs architecture specific registers are 3600updated to the correct (supplied) values. 3601 3602The second part of the structure is architecture specific and 3603typically contains a set of debug registers. 3604 3605For arm64 the number of debug registers is implementation defined and 3606can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 3607KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 3608indicating the number of supported registers. 3609 3610For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether 3611the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported. 3612 3613Also when supported, KVM_CAP_SET_GUEST_DEBUG2 capability indicates the 3614supported KVM_GUESTDBG_* bits in the control field. 3615 3616When debug events exit the main run loop with the reason 3617KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 3618structure containing architecture specific debug information. 3619 36204.88 KVM_GET_EMULATED_CPUID 3621--------------------------- 3622 3623:Capability: KVM_CAP_EXT_EMUL_CPUID 3624:Architectures: x86 3625:Type: system ioctl 3626:Parameters: struct kvm_cpuid2 (in/out) 3627:Returns: 0 on success, -1 on error 3628 3629:: 3630 3631 struct kvm_cpuid2 { 3632 __u32 nent; 3633 __u32 flags; 3634 struct kvm_cpuid_entry2 entries[0]; 3635 }; 3636 3637The member 'flags' is used for passing flags from userspace. 3638 3639:: 3640 3641 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 3642 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 3643 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 3644 3645 struct kvm_cpuid_entry2 { 3646 __u32 function; 3647 __u32 index; 3648 __u32 flags; 3649 __u32 eax; 3650 __u32 ebx; 3651 __u32 ecx; 3652 __u32 edx; 3653 __u32 padding[3]; 3654 }; 3655 3656This ioctl returns x86 cpuid features which are emulated by 3657kvm.Userspace can use the information returned by this ioctl to query 3658which features are emulated by kvm instead of being present natively. 3659 3660Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 3661structure with the 'nent' field indicating the number of entries in 3662the variable-size array 'entries'. If the number of entries is too low 3663to describe the cpu capabilities, an error (E2BIG) is returned. If the 3664number is too high, the 'nent' field is adjusted and an error (ENOMEM) 3665is returned. If the number is just right, the 'nent' field is adjusted 3666to the number of valid entries in the 'entries' array, which is then 3667filled. 3668 3669The entries returned are the set CPUID bits of the respective features 3670which kvm emulates, as returned by the CPUID instruction, with unknown 3671or unsupported feature bits cleared. 3672 3673Features like x2apic, for example, may not be present in the host cpu 3674but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 3675emulated efficiently and thus not included here. 3676 3677The fields in each entry are defined as follows: 3678 3679 function: 3680 the eax value used to obtain the entry 3681 index: 3682 the ecx value used to obtain the entry (for entries that are 3683 affected by ecx) 3684 flags: 3685 an OR of zero or more of the following: 3686 3687 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3688 if the index field is valid 3689 3690 eax, ebx, ecx, edx: 3691 3692 the values returned by the cpuid instruction for 3693 this function/index combination 3694 36954.89 KVM_S390_MEM_OP 3696-------------------- 3697 3698:Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S390_PROTECTED, KVM_CAP_S390_MEM_OP_EXTENSION 3699:Architectures: s390 3700:Type: vm ioctl, vcpu ioctl 3701:Parameters: struct kvm_s390_mem_op (in) 3702:Returns: = 0 on success, 3703 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 3704 > 0 if an exception occurred while walking the page tables 3705 3706Read or write data from/to the VM's memory. 3707The KVM_CAP_S390_MEM_OP_EXTENSION capability specifies what functionality is 3708supported. 3709 3710Parameters are specified via the following structure:: 3711 3712 struct kvm_s390_mem_op { 3713 __u64 gaddr; /* the guest address */ 3714 __u64 flags; /* flags */ 3715 __u32 size; /* amount of bytes */ 3716 __u32 op; /* type of operation */ 3717 __u64 buf; /* buffer in userspace */ 3718 union { 3719 struct { 3720 __u8 ar; /* the access register number */ 3721 __u8 key; /* access key, ignored if flag unset */ 3722 }; 3723 __u32 sida_offset; /* offset into the sida */ 3724 __u8 reserved[32]; /* ignored */ 3725 }; 3726 }; 3727 3728The start address of the memory region has to be specified in the "gaddr" 3729field, and the length of the region in the "size" field (which must not 3730be 0). The maximum value for "size" can be obtained by checking the 3731KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the 3732userspace application where the read data should be written to for 3733a read access, or where the data that should be written is stored for 3734a write access. The "reserved" field is meant for future extensions. 3735Reserved and unused values are ignored. Future extension that add members must 3736introduce new flags. 3737 3738The type of operation is specified in the "op" field. Flags modifying 3739their behavior can be set in the "flags" field. Undefined flag bits must 3740be set to 0. 3741 3742Possible operations are: 3743 * ``KVM_S390_MEMOP_LOGICAL_READ`` 3744 * ``KVM_S390_MEMOP_LOGICAL_WRITE`` 3745 * ``KVM_S390_MEMOP_ABSOLUTE_READ`` 3746 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE`` 3747 * ``KVM_S390_MEMOP_SIDA_READ`` 3748 * ``KVM_S390_MEMOP_SIDA_WRITE`` 3749 3750Logical read/write: 3751^^^^^^^^^^^^^^^^^^^ 3752 3753Access logical memory, i.e. translate the given guest address to an absolute 3754address given the state of the VCPU and use the absolute address as target of 3755the access. "ar" designates the access register number to be used; the valid 3756range is 0..15. 3757Logical accesses are permitted for the VCPU ioctl only. 3758Logical accesses are permitted for non-protected guests only. 3759 3760Supported flags: 3761 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3762 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION`` 3763 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3764 3765The KVM_S390_MEMOP_F_CHECK_ONLY flag can be set to check whether the 3766corresponding memory access would cause an access exception; however, 3767no actual access to the data in memory at the destination is performed. 3768In this case, "buf" is unused and can be NULL. 3769 3770In case an access exception occurred during the access (or would occur 3771in case of KVM_S390_MEMOP_F_CHECK_ONLY), the ioctl returns a positive 3772error number indicating the type of exception. This exception is also 3773raised directly at the corresponding VCPU if the flag 3774KVM_S390_MEMOP_F_INJECT_EXCEPTION is set. 3775 3776If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key 3777protection is also in effect and may cause exceptions if accesses are 3778prohibited given the access key designated by "key"; the valid range is 0..15. 3779KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION 3780is > 0. 3781 3782Absolute read/write: 3783^^^^^^^^^^^^^^^^^^^^ 3784 3785Access absolute memory. This operation is intended to be used with the 3786KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to allow accessing memory and performing 3787the checks required for storage key protection as one operation (as opposed to 3788user space getting the storage keys, performing the checks, and accessing 3789memory thereafter, which could lead to a delay between check and access). 3790Absolute accesses are permitted for the VM ioctl if KVM_CAP_S390_MEM_OP_EXTENSION 3791is > 0. 3792Currently absolute accesses are not permitted for VCPU ioctls. 3793Absolute accesses are permitted for non-protected guests only. 3794 3795Supported flags: 3796 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3797 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3798 3799The semantics of the flags are as for logical accesses. 3800 3801SIDA read/write: 3802^^^^^^^^^^^^^^^^ 3803 3804Access the secure instruction data area which contains memory operands necessary 3805for instruction emulation for protected guests. 3806SIDA accesses are available if the KVM_CAP_S390_PROTECTED capability is available. 3807SIDA accesses are permitted for the VCPU ioctl only. 3808SIDA accesses are permitted for protected guests only. 3809 3810No flags are supported. 3811 38124.90 KVM_S390_GET_SKEYS 3813----------------------- 3814 3815:Capability: KVM_CAP_S390_SKEYS 3816:Architectures: s390 3817:Type: vm ioctl 3818:Parameters: struct kvm_s390_skeys 3819:Returns: 0 on success, KVM_S390_GET_SKEYS_NONE if guest is not using storage 3820 keys, negative value on error 3821 3822This ioctl is used to get guest storage key values on the s390 3823architecture. The ioctl takes parameters via the kvm_s390_skeys struct:: 3824 3825 struct kvm_s390_skeys { 3826 __u64 start_gfn; 3827 __u64 count; 3828 __u64 skeydata_addr; 3829 __u32 flags; 3830 __u32 reserved[9]; 3831 }; 3832 3833The start_gfn field is the number of the first guest frame whose storage keys 3834you want to get. 3835 3836The count field is the number of consecutive frames (starting from start_gfn) 3837whose storage keys to get. The count field must be at least 1 and the maximum 3838allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 3839will cause the ioctl to return -EINVAL. 3840 3841The skeydata_addr field is the address to a buffer large enough to hold count 3842bytes. This buffer will be filled with storage key data by the ioctl. 3843 38444.91 KVM_S390_SET_SKEYS 3845----------------------- 3846 3847:Capability: KVM_CAP_S390_SKEYS 3848:Architectures: s390 3849:Type: vm ioctl 3850:Parameters: struct kvm_s390_skeys 3851:Returns: 0 on success, negative value on error 3852 3853This ioctl is used to set guest storage key values on the s390 3854architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 3855See section on KVM_S390_GET_SKEYS for struct definition. 3856 3857The start_gfn field is the number of the first guest frame whose storage keys 3858you want to set. 3859 3860The count field is the number of consecutive frames (starting from start_gfn) 3861whose storage keys to get. The count field must be at least 1 and the maximum 3862allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 3863will cause the ioctl to return -EINVAL. 3864 3865The skeydata_addr field is the address to a buffer containing count bytes of 3866storage keys. Each byte in the buffer will be set as the storage key for a 3867single frame starting at start_gfn for count frames. 3868 3869Note: If any architecturally invalid key value is found in the given data then 3870the ioctl will return -EINVAL. 3871 38724.92 KVM_S390_IRQ 3873----------------- 3874 3875:Capability: KVM_CAP_S390_INJECT_IRQ 3876:Architectures: s390 3877:Type: vcpu ioctl 3878:Parameters: struct kvm_s390_irq (in) 3879:Returns: 0 on success, -1 on error 3880 3881Errors: 3882 3883 3884 ====== ================================================================= 3885 EINVAL interrupt type is invalid 3886 type is KVM_S390_SIGP_STOP and flag parameter is invalid value, 3887 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 3888 than the maximum of VCPUs 3889 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped, 3890 type is KVM_S390_SIGP_STOP and a stop irq is already pending, 3891 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 3892 is already pending 3893 ====== ================================================================= 3894 3895Allows to inject an interrupt to the guest. 3896 3897Using struct kvm_s390_irq as a parameter allows 3898to inject additional payload which is not 3899possible via KVM_S390_INTERRUPT. 3900 3901Interrupt parameters are passed via kvm_s390_irq:: 3902 3903 struct kvm_s390_irq { 3904 __u64 type; 3905 union { 3906 struct kvm_s390_io_info io; 3907 struct kvm_s390_ext_info ext; 3908 struct kvm_s390_pgm_info pgm; 3909 struct kvm_s390_emerg_info emerg; 3910 struct kvm_s390_extcall_info extcall; 3911 struct kvm_s390_prefix_info prefix; 3912 struct kvm_s390_stop_info stop; 3913 struct kvm_s390_mchk_info mchk; 3914 char reserved[64]; 3915 } u; 3916 }; 3917 3918type can be one of the following: 3919 3920- KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 3921- KVM_S390_PROGRAM_INT - program check; parameters in .pgm 3922- KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 3923- KVM_S390_RESTART - restart; no parameters 3924- KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 3925- KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 3926- KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 3927- KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 3928- KVM_S390_MCHK - machine check interrupt; parameters in .mchk 3929 3930This is an asynchronous vcpu ioctl and can be invoked from any thread. 3931 39324.94 KVM_S390_GET_IRQ_STATE 3933--------------------------- 3934 3935:Capability: KVM_CAP_S390_IRQ_STATE 3936:Architectures: s390 3937:Type: vcpu ioctl 3938:Parameters: struct kvm_s390_irq_state (out) 3939:Returns: >= number of bytes copied into buffer, 3940 -EINVAL if buffer size is 0, 3941 -ENOBUFS if buffer size is too small to fit all pending interrupts, 3942 -EFAULT if the buffer address was invalid 3943 3944This ioctl allows userspace to retrieve the complete state of all currently 3945pending interrupts in a single buffer. Use cases include migration 3946and introspection. The parameter structure contains the address of a 3947userspace buffer and its length:: 3948 3949 struct kvm_s390_irq_state { 3950 __u64 buf; 3951 __u32 flags; /* will stay unused for compatibility reasons */ 3952 __u32 len; 3953 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 3954 }; 3955 3956Userspace passes in the above struct and for each pending interrupt a 3957struct kvm_s390_irq is copied to the provided buffer. 3958 3959The structure contains a flags and a reserved field for future extensions. As 3960the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 3961reserved, these fields can not be used in the future without breaking 3962compatibility. 3963 3964If -ENOBUFS is returned the buffer provided was too small and userspace 3965may retry with a bigger buffer. 3966 39674.95 KVM_S390_SET_IRQ_STATE 3968--------------------------- 3969 3970:Capability: KVM_CAP_S390_IRQ_STATE 3971:Architectures: s390 3972:Type: vcpu ioctl 3973:Parameters: struct kvm_s390_irq_state (in) 3974:Returns: 0 on success, 3975 -EFAULT if the buffer address was invalid, 3976 -EINVAL for an invalid buffer length (see below), 3977 -EBUSY if there were already interrupts pending, 3978 errors occurring when actually injecting the 3979 interrupt. See KVM_S390_IRQ. 3980 3981This ioctl allows userspace to set the complete state of all cpu-local 3982interrupts currently pending for the vcpu. It is intended for restoring 3983interrupt state after a migration. The input parameter is a userspace buffer 3984containing a struct kvm_s390_irq_state:: 3985 3986 struct kvm_s390_irq_state { 3987 __u64 buf; 3988 __u32 flags; /* will stay unused for compatibility reasons */ 3989 __u32 len; 3990 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 3991 }; 3992 3993The restrictions for flags and reserved apply as well. 3994(see KVM_S390_GET_IRQ_STATE) 3995 3996The userspace memory referenced by buf contains a struct kvm_s390_irq 3997for each interrupt to be injected into the guest. 3998If one of the interrupts could not be injected for some reason the 3999ioctl aborts. 4000 4001len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 4002and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 4003which is the maximum number of possibly pending cpu-local interrupts. 4004 40054.96 KVM_SMI 4006------------ 4007 4008:Capability: KVM_CAP_X86_SMM 4009:Architectures: x86 4010:Type: vcpu ioctl 4011:Parameters: none 4012:Returns: 0 on success, -1 on error 4013 4014Queues an SMI on the thread's vcpu. 4015 40164.97 KVM_X86_SET_MSR_FILTER 4017---------------------------- 4018 4019:Capability: KVM_X86_SET_MSR_FILTER 4020:Architectures: x86 4021:Type: vm ioctl 4022:Parameters: struct kvm_msr_filter 4023:Returns: 0 on success, < 0 on error 4024 4025:: 4026 4027 struct kvm_msr_filter_range { 4028 #define KVM_MSR_FILTER_READ (1 << 0) 4029 #define KVM_MSR_FILTER_WRITE (1 << 1) 4030 __u32 flags; 4031 __u32 nmsrs; /* number of msrs in bitmap */ 4032 __u32 base; /* MSR index the bitmap starts at */ 4033 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 4034 }; 4035 4036 #define KVM_MSR_FILTER_MAX_RANGES 16 4037 struct kvm_msr_filter { 4038 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 4039 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 4040 __u32 flags; 4041 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 4042 }; 4043 4044flags values for ``struct kvm_msr_filter_range``: 4045 4046``KVM_MSR_FILTER_READ`` 4047 4048 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 4049 indicates that a read should immediately fail, while a 1 indicates that 4050 a read for a particular MSR should be handled regardless of the default 4051 filter action. 4052 4053``KVM_MSR_FILTER_WRITE`` 4054 4055 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 4056 indicates that a write should immediately fail, while a 1 indicates that 4057 a write for a particular MSR should be handled regardless of the default 4058 filter action. 4059 4060``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE`` 4061 4062 Filter both read and write accesses to MSRs using the given bitmap. A 0 4063 in the bitmap indicates that both reads and writes should immediately fail, 4064 while a 1 indicates that reads and writes for a particular MSR are not 4065 filtered by this range. 4066 4067flags values for ``struct kvm_msr_filter``: 4068 4069``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4070 4071 If no filter range matches an MSR index that is getting accessed, KVM will 4072 fall back to allowing access to the MSR. 4073 4074``KVM_MSR_FILTER_DEFAULT_DENY`` 4075 4076 If no filter range matches an MSR index that is getting accessed, KVM will 4077 fall back to rejecting access to the MSR. In this mode, all MSRs that should 4078 be processed by KVM need to explicitly be marked as allowed in the bitmaps. 4079 4080This ioctl allows user space to define up to 16 bitmaps of MSR ranges to 4081specify whether a certain MSR access should be explicitly filtered for or not. 4082 4083If this ioctl has never been invoked, MSR accesses are not guarded and the 4084default KVM in-kernel emulation behavior is fully preserved. 4085 4086Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 4087filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 4088an error. 4089 4090As soon as the filtering is in place, every MSR access is processed through 4091the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff); 4092x2APIC MSRs are always allowed, independent of the ``default_allow`` setting, 4093and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base 4094register. 4095 4096.. warning:: 4097 MSR accesses coming from nested vmentry/vmexit are not filtered. 4098 This includes both writes to individual VMCS fields and reads/writes 4099 through the MSR lists pointed to by the VMCS. 4100 4101If a bit is within one of the defined ranges, read and write accesses are 4102guarded by the bitmap's value for the MSR index if the kind of access 4103is included in the ``struct kvm_msr_filter_range`` flags. If no range 4104cover this particular access, the behavior is determined by the flags 4105field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4106and ``KVM_MSR_FILTER_DEFAULT_DENY``. 4107 4108Each bitmap range specifies a range of MSRs to potentially allow access on. 4109The range goes from MSR index [base .. base+nmsrs]. The flags field 4110indicates whether reads, writes or both reads and writes are filtered 4111by setting a 1 bit in the bitmap for the corresponding MSR index. 4112 4113If an MSR access is not permitted through the filtering, it generates a 4114#GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that 4115allows user space to deflect and potentially handle various MSR accesses 4116into user space. 4117 4118If a vCPU is in running state while this ioctl is invoked, the vCPU may 4119experience inconsistent filtering behavior on MSR accesses. 4120 41214.98 KVM_CREATE_SPAPR_TCE_64 4122---------------------------- 4123 4124:Capability: KVM_CAP_SPAPR_TCE_64 4125:Architectures: powerpc 4126:Type: vm ioctl 4127:Parameters: struct kvm_create_spapr_tce_64 (in) 4128:Returns: file descriptor for manipulating the created TCE table 4129 4130This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 4131windows, described in 4.62 KVM_CREATE_SPAPR_TCE 4132 4133This capability uses extended struct in ioctl interface:: 4134 4135 /* for KVM_CAP_SPAPR_TCE_64 */ 4136 struct kvm_create_spapr_tce_64 { 4137 __u64 liobn; 4138 __u32 page_shift; 4139 __u32 flags; 4140 __u64 offset; /* in pages */ 4141 __u64 size; /* in pages */ 4142 }; 4143 4144The aim of extension is to support an additional bigger DMA window with 4145a variable page size. 4146KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 4147a bus offset of the corresponding DMA window, @size and @offset are numbers 4148of IOMMU pages. 4149 4150@flags are not used at the moment. 4151 4152The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 4153 41544.99 KVM_REINJECT_CONTROL 4155------------------------- 4156 4157:Capability: KVM_CAP_REINJECT_CONTROL 4158:Architectures: x86 4159:Type: vm ioctl 4160:Parameters: struct kvm_reinject_control (in) 4161:Returns: 0 on success, 4162 -EFAULT if struct kvm_reinject_control cannot be read, 4163 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 4164 4165i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 4166where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 4167vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 4168interrupt whenever there isn't a pending interrupt from i8254. 4169!reinject mode injects an interrupt as soon as a tick arrives. 4170 4171:: 4172 4173 struct kvm_reinject_control { 4174 __u8 pit_reinject; 4175 __u8 reserved[31]; 4176 }; 4177 4178pit_reinject = 0 (!reinject mode) is recommended, unless running an old 4179operating system that uses the PIT for timing (e.g. Linux 2.4.x). 4180 41814.100 KVM_PPC_CONFIGURE_V3_MMU 4182------------------------------ 4183 4184:Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 4185:Architectures: ppc 4186:Type: vm ioctl 4187:Parameters: struct kvm_ppc_mmuv3_cfg (in) 4188:Returns: 0 on success, 4189 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 4190 -EINVAL if the configuration is invalid 4191 4192This ioctl controls whether the guest will use radix or HPT (hashed 4193page table) translation, and sets the pointer to the process table for 4194the guest. 4195 4196:: 4197 4198 struct kvm_ppc_mmuv3_cfg { 4199 __u64 flags; 4200 __u64 process_table; 4201 }; 4202 4203There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 4204KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 4205to use radix tree translation, and if clear, to use HPT translation. 4206KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 4207to be able to use the global TLB and SLB invalidation instructions; 4208if clear, the guest may not use these instructions. 4209 4210The process_table field specifies the address and size of the guest 4211process table, which is in the guest's space. This field is formatted 4212as the second doubleword of the partition table entry, as defined in 4213the Power ISA V3.00, Book III section 5.7.6.1. 4214 42154.101 KVM_PPC_GET_RMMU_INFO 4216--------------------------- 4217 4218:Capability: KVM_CAP_PPC_RADIX_MMU 4219:Architectures: ppc 4220:Type: vm ioctl 4221:Parameters: struct kvm_ppc_rmmu_info (out) 4222:Returns: 0 on success, 4223 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 4224 -EINVAL if no useful information can be returned 4225 4226This ioctl returns a structure containing two things: (a) a list 4227containing supported radix tree geometries, and (b) a list that maps 4228page sizes to put in the "AP" (actual page size) field for the tlbie 4229(TLB invalidate entry) instruction. 4230 4231:: 4232 4233 struct kvm_ppc_rmmu_info { 4234 struct kvm_ppc_radix_geom { 4235 __u8 page_shift; 4236 __u8 level_bits[4]; 4237 __u8 pad[3]; 4238 } geometries[8]; 4239 __u32 ap_encodings[8]; 4240 }; 4241 4242The geometries[] field gives up to 8 supported geometries for the 4243radix page table, in terms of the log base 2 of the smallest page 4244size, and the number of bits indexed at each level of the tree, from 4245the PTE level up to the PGD level in that order. Any unused entries 4246will have 0 in the page_shift field. 4247 4248The ap_encodings gives the supported page sizes and their AP field 4249encodings, encoded with the AP value in the top 3 bits and the log 4250base 2 of the page size in the bottom 6 bits. 4251 42524.102 KVM_PPC_RESIZE_HPT_PREPARE 4253-------------------------------- 4254 4255:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4256:Architectures: powerpc 4257:Type: vm ioctl 4258:Parameters: struct kvm_ppc_resize_hpt (in) 4259:Returns: 0 on successful completion, 4260 >0 if a new HPT is being prepared, the value is an estimated 4261 number of milliseconds until preparation is complete, 4262 -EFAULT if struct kvm_reinject_control cannot be read, 4263 -EINVAL if the supplied shift or flags are invalid, 4264 -ENOMEM if unable to allocate the new HPT, 4265 4266Used to implement the PAPR extension for runtime resizing of a guest's 4267Hashed Page Table (HPT). Specifically this starts, stops or monitors 4268the preparation of a new potential HPT for the guest, essentially 4269implementing the H_RESIZE_HPT_PREPARE hypercall. 4270 4271:: 4272 4273 struct kvm_ppc_resize_hpt { 4274 __u64 flags; 4275 __u32 shift; 4276 __u32 pad; 4277 }; 4278 4279If called with shift > 0 when there is no pending HPT for the guest, 4280this begins preparation of a new pending HPT of size 2^(shift) bytes. 4281It then returns a positive integer with the estimated number of 4282milliseconds until preparation is complete. 4283 4284If called when there is a pending HPT whose size does not match that 4285requested in the parameters, discards the existing pending HPT and 4286creates a new one as above. 4287 4288If called when there is a pending HPT of the size requested, will: 4289 4290 * If preparation of the pending HPT is already complete, return 0 4291 * If preparation of the pending HPT has failed, return an error 4292 code, then discard the pending HPT. 4293 * If preparation of the pending HPT is still in progress, return an 4294 estimated number of milliseconds until preparation is complete. 4295 4296If called with shift == 0, discards any currently pending HPT and 4297returns 0 (i.e. cancels any in-progress preparation). 4298 4299flags is reserved for future expansion, currently setting any bits in 4300flags will result in an -EINVAL. 4301 4302Normally this will be called repeatedly with the same parameters until 4303it returns <= 0. The first call will initiate preparation, subsequent 4304ones will monitor preparation until it completes or fails. 4305 43064.103 KVM_PPC_RESIZE_HPT_COMMIT 4307------------------------------- 4308 4309:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4310:Architectures: powerpc 4311:Type: vm ioctl 4312:Parameters: struct kvm_ppc_resize_hpt (in) 4313:Returns: 0 on successful completion, 4314 -EFAULT if struct kvm_reinject_control cannot be read, 4315 -EINVAL if the supplied shift or flags are invalid, 4316 -ENXIO is there is no pending HPT, or the pending HPT doesn't 4317 have the requested size, 4318 -EBUSY if the pending HPT is not fully prepared, 4319 -ENOSPC if there was a hash collision when moving existing 4320 HPT entries to the new HPT, 4321 -EIO on other error conditions 4322 4323Used to implement the PAPR extension for runtime resizing of a guest's 4324Hashed Page Table (HPT). Specifically this requests that the guest be 4325transferred to working with the new HPT, essentially implementing the 4326H_RESIZE_HPT_COMMIT hypercall. 4327 4328:: 4329 4330 struct kvm_ppc_resize_hpt { 4331 __u64 flags; 4332 __u32 shift; 4333 __u32 pad; 4334 }; 4335 4336This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 4337returned 0 with the same parameters. In other cases 4338KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 4339-EBUSY, though others may be possible if the preparation was started, 4340but failed). 4341 4342This will have undefined effects on the guest if it has not already 4343placed itself in a quiescent state where no vcpu will make MMU enabled 4344memory accesses. 4345 4346On succsful completion, the pending HPT will become the guest's active 4347HPT and the previous HPT will be discarded. 4348 4349On failure, the guest will still be operating on its previous HPT. 4350 43514.104 KVM_X86_GET_MCE_CAP_SUPPORTED 4352----------------------------------- 4353 4354:Capability: KVM_CAP_MCE 4355:Architectures: x86 4356:Type: system ioctl 4357:Parameters: u64 mce_cap (out) 4358:Returns: 0 on success, -1 on error 4359 4360Returns supported MCE capabilities. The u64 mce_cap parameter 4361has the same format as the MSR_IA32_MCG_CAP register. Supported 4362capabilities will have the corresponding bits set. 4363 43644.105 KVM_X86_SETUP_MCE 4365----------------------- 4366 4367:Capability: KVM_CAP_MCE 4368:Architectures: x86 4369:Type: vcpu ioctl 4370:Parameters: u64 mcg_cap (in) 4371:Returns: 0 on success, 4372 -EFAULT if u64 mcg_cap cannot be read, 4373 -EINVAL if the requested number of banks is invalid, 4374 -EINVAL if requested MCE capability is not supported. 4375 4376Initializes MCE support for use. The u64 mcg_cap parameter 4377has the same format as the MSR_IA32_MCG_CAP register and 4378specifies which capabilities should be enabled. The maximum 4379supported number of error-reporting banks can be retrieved when 4380checking for KVM_CAP_MCE. The supported capabilities can be 4381retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 4382 43834.106 KVM_X86_SET_MCE 4384--------------------- 4385 4386:Capability: KVM_CAP_MCE 4387:Architectures: x86 4388:Type: vcpu ioctl 4389:Parameters: struct kvm_x86_mce (in) 4390:Returns: 0 on success, 4391 -EFAULT if struct kvm_x86_mce cannot be read, 4392 -EINVAL if the bank number is invalid, 4393 -EINVAL if VAL bit is not set in status field. 4394 4395Inject a machine check error (MCE) into the guest. The input 4396parameter is:: 4397 4398 struct kvm_x86_mce { 4399 __u64 status; 4400 __u64 addr; 4401 __u64 misc; 4402 __u64 mcg_status; 4403 __u8 bank; 4404 __u8 pad1[7]; 4405 __u64 pad2[3]; 4406 }; 4407 4408If the MCE being reported is an uncorrected error, KVM will 4409inject it as an MCE exception into the guest. If the guest 4410MCG_STATUS register reports that an MCE is in progress, KVM 4411causes an KVM_EXIT_SHUTDOWN vmexit. 4412 4413Otherwise, if the MCE is a corrected error, KVM will just 4414store it in the corresponding bank (provided this bank is 4415not holding a previously reported uncorrected error). 4416 44174.107 KVM_S390_GET_CMMA_BITS 4418---------------------------- 4419 4420:Capability: KVM_CAP_S390_CMMA_MIGRATION 4421:Architectures: s390 4422:Type: vm ioctl 4423:Parameters: struct kvm_s390_cmma_log (in, out) 4424:Returns: 0 on success, a negative value on error 4425 4426This ioctl is used to get the values of the CMMA bits on the s390 4427architecture. It is meant to be used in two scenarios: 4428 4429- During live migration to save the CMMA values. Live migration needs 4430 to be enabled via the KVM_REQ_START_MIGRATION VM property. 4431- To non-destructively peek at the CMMA values, with the flag 4432 KVM_S390_CMMA_PEEK set. 4433 4434The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 4435values are written to a buffer whose location is indicated via the "values" 4436member in the kvm_s390_cmma_log struct. The values in the input struct are 4437also updated as needed. 4438 4439Each CMMA value takes up one byte. 4440 4441:: 4442 4443 struct kvm_s390_cmma_log { 4444 __u64 start_gfn; 4445 __u32 count; 4446 __u32 flags; 4447 union { 4448 __u64 remaining; 4449 __u64 mask; 4450 }; 4451 __u64 values; 4452 }; 4453 4454start_gfn is the number of the first guest frame whose CMMA values are 4455to be retrieved, 4456 4457count is the length of the buffer in bytes, 4458 4459values points to the buffer where the result will be written to. 4460 4461If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 4462KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 4463other ioctls. 4464 4465The result is written in the buffer pointed to by the field values, and 4466the values of the input parameter are updated as follows. 4467 4468Depending on the flags, different actions are performed. The only 4469supported flag so far is KVM_S390_CMMA_PEEK. 4470 4471The default behaviour if KVM_S390_CMMA_PEEK is not set is: 4472start_gfn will indicate the first page frame whose CMMA bits were dirty. 4473It is not necessarily the same as the one passed as input, as clean pages 4474are skipped. 4475 4476count will indicate the number of bytes actually written in the buffer. 4477It can (and very often will) be smaller than the input value, since the 4478buffer is only filled until 16 bytes of clean values are found (which 4479are then not copied in the buffer). Since a CMMA migration block needs 4480the base address and the length, for a total of 16 bytes, we will send 4481back some clean data if there is some dirty data afterwards, as long as 4482the size of the clean data does not exceed the size of the header. This 4483allows to minimize the amount of data to be saved or transferred over 4484the network at the expense of more roundtrips to userspace. The next 4485invocation of the ioctl will skip over all the clean values, saving 4486potentially more than just the 16 bytes we found. 4487 4488If KVM_S390_CMMA_PEEK is set: 4489the existing storage attributes are read even when not in migration 4490mode, and no other action is performed; 4491 4492the output start_gfn will be equal to the input start_gfn, 4493 4494the output count will be equal to the input count, except if the end of 4495memory has been reached. 4496 4497In both cases: 4498the field "remaining" will indicate the total number of dirty CMMA values 4499still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 4500not enabled. 4501 4502mask is unused. 4503 4504values points to the userspace buffer where the result will be stored. 4505 4506This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4507complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4508KVM_S390_CMMA_PEEK is not set but migration mode was not enabled, with 4509-EFAULT if the userspace address is invalid or if no page table is 4510present for the addresses (e.g. when using hugepages). 4511 45124.108 KVM_S390_SET_CMMA_BITS 4513---------------------------- 4514 4515:Capability: KVM_CAP_S390_CMMA_MIGRATION 4516:Architectures: s390 4517:Type: vm ioctl 4518:Parameters: struct kvm_s390_cmma_log (in) 4519:Returns: 0 on success, a negative value on error 4520 4521This ioctl is used to set the values of the CMMA bits on the s390 4522architecture. It is meant to be used during live migration to restore 4523the CMMA values, but there are no restrictions on its use. 4524The ioctl takes parameters via the kvm_s390_cmma_values struct. 4525Each CMMA value takes up one byte. 4526 4527:: 4528 4529 struct kvm_s390_cmma_log { 4530 __u64 start_gfn; 4531 __u32 count; 4532 __u32 flags; 4533 union { 4534 __u64 remaining; 4535 __u64 mask; 4536 }; 4537 __u64 values; 4538 }; 4539 4540start_gfn indicates the starting guest frame number, 4541 4542count indicates how many values are to be considered in the buffer, 4543 4544flags is not used and must be 0. 4545 4546mask indicates which PGSTE bits are to be considered. 4547 4548remaining is not used. 4549 4550values points to the buffer in userspace where to store the values. 4551 4552This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4553complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4554the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 4555if the flags field was not 0, with -EFAULT if the userspace address is 4556invalid, if invalid pages are written to (e.g. after the end of memory) 4557or if no page table is present for the addresses (e.g. when using 4558hugepages). 4559 45604.109 KVM_PPC_GET_CPU_CHAR 4561-------------------------- 4562 4563:Capability: KVM_CAP_PPC_GET_CPU_CHAR 4564:Architectures: powerpc 4565:Type: vm ioctl 4566:Parameters: struct kvm_ppc_cpu_char (out) 4567:Returns: 0 on successful completion, 4568 -EFAULT if struct kvm_ppc_cpu_char cannot be written 4569 4570This ioctl gives userspace information about certain characteristics 4571of the CPU relating to speculative execution of instructions and 4572possible information leakage resulting from speculative execution (see 4573CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 4574returned in struct kvm_ppc_cpu_char, which looks like this:: 4575 4576 struct kvm_ppc_cpu_char { 4577 __u64 character; /* characteristics of the CPU */ 4578 __u64 behaviour; /* recommended software behaviour */ 4579 __u64 character_mask; /* valid bits in character */ 4580 __u64 behaviour_mask; /* valid bits in behaviour */ 4581 }; 4582 4583For extensibility, the character_mask and behaviour_mask fields 4584indicate which bits of character and behaviour have been filled in by 4585the kernel. If the set of defined bits is extended in future then 4586userspace will be able to tell whether it is running on a kernel that 4587knows about the new bits. 4588 4589The character field describes attributes of the CPU which can help 4590with preventing inadvertent information disclosure - specifically, 4591whether there is an instruction to flash-invalidate the L1 data cache 4592(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 4593to a mode where entries can only be used by the thread that created 4594them, whether the bcctr[l] instruction prevents speculation, and 4595whether a speculation barrier instruction (ori 31,31,0) is provided. 4596 4597The behaviour field describes actions that software should take to 4598prevent inadvertent information disclosure, and thus describes which 4599vulnerabilities the hardware is subject to; specifically whether the 4600L1 data cache should be flushed when returning to user mode from the 4601kernel, and whether a speculation barrier should be placed between an 4602array bounds check and the array access. 4603 4604These fields use the same bit definitions as the new 4605H_GET_CPU_CHARACTERISTICS hypercall. 4606 46074.110 KVM_MEMORY_ENCRYPT_OP 4608--------------------------- 4609 4610:Capability: basic 4611:Architectures: x86 4612:Type: vm 4613:Parameters: an opaque platform specific structure (in/out) 4614:Returns: 0 on success; -1 on error 4615 4616If the platform supports creating encrypted VMs then this ioctl can be used 4617for issuing platform-specific memory encryption commands to manage those 4618encrypted VMs. 4619 4620Currently, this ioctl is used for issuing Secure Encrypted Virtualization 4621(SEV) commands on AMD Processors. The SEV commands are defined in 4622Documentation/virt/kvm/amd-memory-encryption.rst. 4623 46244.111 KVM_MEMORY_ENCRYPT_REG_REGION 4625----------------------------------- 4626 4627:Capability: basic 4628:Architectures: x86 4629:Type: system 4630:Parameters: struct kvm_enc_region (in) 4631:Returns: 0 on success; -1 on error 4632 4633This ioctl can be used to register a guest memory region which may 4634contain encrypted data (e.g. guest RAM, SMRAM etc). 4635 4636It is used in the SEV-enabled guest. When encryption is enabled, a guest 4637memory region may contain encrypted data. The SEV memory encryption 4638engine uses a tweak such that two identical plaintext pages, each at 4639different locations will have differing ciphertexts. So swapping or 4640moving ciphertext of those pages will not result in plaintext being 4641swapped. So relocating (or migrating) physical backing pages for the SEV 4642guest will require some additional steps. 4643 4644Note: The current SEV key management spec does not provide commands to 4645swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 4646memory region registered with the ioctl. 4647 46484.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4649------------------------------------- 4650 4651:Capability: basic 4652:Architectures: x86 4653:Type: system 4654:Parameters: struct kvm_enc_region (in) 4655:Returns: 0 on success; -1 on error 4656 4657This ioctl can be used to unregister the guest memory region registered 4658with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 4659 46604.113 KVM_HYPERV_EVENTFD 4661------------------------ 4662 4663:Capability: KVM_CAP_HYPERV_EVENTFD 4664:Architectures: x86 4665:Type: vm ioctl 4666:Parameters: struct kvm_hyperv_eventfd (in) 4667 4668This ioctl (un)registers an eventfd to receive notifications from the guest on 4669the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without 4670causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number 4671(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit. 4672 4673:: 4674 4675 struct kvm_hyperv_eventfd { 4676 __u32 conn_id; 4677 __s32 fd; 4678 __u32 flags; 4679 __u32 padding[3]; 4680 }; 4681 4682The conn_id field should fit within 24 bits:: 4683 4684 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff 4685 4686The acceptable values for the flags field are:: 4687 4688 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0) 4689 4690:Returns: 0 on success, 4691 -EINVAL if conn_id or flags is outside the allowed range, 4692 -ENOENT on deassign if the conn_id isn't registered, 4693 -EEXIST on assign if the conn_id is already registered 4694 46954.114 KVM_GET_NESTED_STATE 4696-------------------------- 4697 4698:Capability: KVM_CAP_NESTED_STATE 4699:Architectures: x86 4700:Type: vcpu ioctl 4701:Parameters: struct kvm_nested_state (in/out) 4702:Returns: 0 on success, -1 on error 4703 4704Errors: 4705 4706 ===== ============================================================= 4707 E2BIG the total state size exceeds the value of 'size' specified by 4708 the user; the size required will be written into size. 4709 ===== ============================================================= 4710 4711:: 4712 4713 struct kvm_nested_state { 4714 __u16 flags; 4715 __u16 format; 4716 __u32 size; 4717 4718 union { 4719 struct kvm_vmx_nested_state_hdr vmx; 4720 struct kvm_svm_nested_state_hdr svm; 4721 4722 /* Pad the header to 128 bytes. */ 4723 __u8 pad[120]; 4724 } hdr; 4725 4726 union { 4727 struct kvm_vmx_nested_state_data vmx[0]; 4728 struct kvm_svm_nested_state_data svm[0]; 4729 } data; 4730 }; 4731 4732 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001 4733 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002 4734 #define KVM_STATE_NESTED_EVMCS 0x00000004 4735 4736 #define KVM_STATE_NESTED_FORMAT_VMX 0 4737 #define KVM_STATE_NESTED_FORMAT_SVM 1 4738 4739 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 4740 4741 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001 4742 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002 4743 4744 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001 4745 4746 struct kvm_vmx_nested_state_hdr { 4747 __u64 vmxon_pa; 4748 __u64 vmcs12_pa; 4749 4750 struct { 4751 __u16 flags; 4752 } smm; 4753 4754 __u32 flags; 4755 __u64 preemption_timer_deadline; 4756 }; 4757 4758 struct kvm_vmx_nested_state_data { 4759 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4760 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4761 }; 4762 4763This ioctl copies the vcpu's nested virtualization state from the kernel to 4764userspace. 4765 4766The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE 4767to the KVM_CHECK_EXTENSION ioctl(). 4768 47694.115 KVM_SET_NESTED_STATE 4770-------------------------- 4771 4772:Capability: KVM_CAP_NESTED_STATE 4773:Architectures: x86 4774:Type: vcpu ioctl 4775:Parameters: struct kvm_nested_state (in) 4776:Returns: 0 on success, -1 on error 4777 4778This copies the vcpu's kvm_nested_state struct from userspace to the kernel. 4779For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE. 4780 47814.116 KVM_(UN)REGISTER_COALESCED_MMIO 4782------------------------------------- 4783 4784:Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio) 4785 KVM_CAP_COALESCED_PIO (for coalesced pio) 4786:Architectures: all 4787:Type: vm ioctl 4788:Parameters: struct kvm_coalesced_mmio_zone 4789:Returns: 0 on success, < 0 on error 4790 4791Coalesced I/O is a performance optimization that defers hardware 4792register write emulation so that userspace exits are avoided. It is 4793typically used to reduce the overhead of emulating frequently accessed 4794hardware registers. 4795 4796When a hardware register is configured for coalesced I/O, write accesses 4797do not exit to userspace and their value is recorded in a ring buffer 4798that is shared between kernel and userspace. 4799 4800Coalesced I/O is used if one or more write accesses to a hardware 4801register can be deferred until a read or a write to another hardware 4802register on the same device. This last access will cause a vmexit and 4803userspace will process accesses from the ring buffer before emulating 4804it. That will avoid exiting to userspace on repeated writes. 4805 4806Coalesced pio is based on coalesced mmio. There is little difference 4807between coalesced mmio and pio except that coalesced pio records accesses 4808to I/O ports. 4809 48104.117 KVM_CLEAR_DIRTY_LOG (vm ioctl) 4811------------------------------------ 4812 4813:Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4814:Architectures: x86, arm64, mips 4815:Type: vm ioctl 4816:Parameters: struct kvm_clear_dirty_log (in) 4817:Returns: 0 on success, -1 on error 4818 4819:: 4820 4821 /* for KVM_CLEAR_DIRTY_LOG */ 4822 struct kvm_clear_dirty_log { 4823 __u32 slot; 4824 __u32 num_pages; 4825 __u64 first_page; 4826 union { 4827 void __user *dirty_bitmap; /* one bit per page */ 4828 __u64 padding; 4829 }; 4830 }; 4831 4832The ioctl clears the dirty status of pages in a memory slot, according to 4833the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap 4834field. Bit 0 of the bitmap corresponds to page "first_page" in the 4835memory slot, and num_pages is the size in bits of the input bitmap. 4836first_page must be a multiple of 64; num_pages must also be a multiple of 483764 unless first_page + num_pages is the size of the memory slot. For each 4838bit that is set in the input bitmap, the corresponding page is marked "clean" 4839in KVM's dirty bitmap, and dirty tracking is re-enabled for that page 4840(for example via write-protection, or by clearing the dirty bit in 4841a page table entry). 4842 4843If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 4844the address space for which you want to clear the dirty status. See 4845KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 4846 4847This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4848is enabled; for more information, see the description of the capability. 4849However, it can always be used as long as KVM_CHECK_EXTENSION confirms 4850that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present. 4851 48524.118 KVM_GET_SUPPORTED_HV_CPUID 4853-------------------------------- 4854 4855:Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system) 4856:Architectures: x86 4857:Type: system ioctl, vcpu ioctl 4858:Parameters: struct kvm_cpuid2 (in/out) 4859:Returns: 0 on success, -1 on error 4860 4861:: 4862 4863 struct kvm_cpuid2 { 4864 __u32 nent; 4865 __u32 padding; 4866 struct kvm_cpuid_entry2 entries[0]; 4867 }; 4868 4869 struct kvm_cpuid_entry2 { 4870 __u32 function; 4871 __u32 index; 4872 __u32 flags; 4873 __u32 eax; 4874 __u32 ebx; 4875 __u32 ecx; 4876 __u32 edx; 4877 __u32 padding[3]; 4878 }; 4879 4880This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in 4881KVM. Userspace can use the information returned by this ioctl to construct 4882cpuid information presented to guests consuming Hyper-V enlightenments (e.g. 4883Windows or Hyper-V guests). 4884 4885CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level 4886Functional Specification (TLFS). These leaves can't be obtained with 4887KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature 4888leaves (0x40000000, 0x40000001). 4889 4890Currently, the following list of CPUID leaves are returned: 4891 4892 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 4893 - HYPERV_CPUID_INTERFACE 4894 - HYPERV_CPUID_VERSION 4895 - HYPERV_CPUID_FEATURES 4896 - HYPERV_CPUID_ENLIGHTMENT_INFO 4897 - HYPERV_CPUID_IMPLEMENT_LIMITS 4898 - HYPERV_CPUID_NESTED_FEATURES 4899 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS 4900 - HYPERV_CPUID_SYNDBG_INTERFACE 4901 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 4902 4903Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure 4904with the 'nent' field indicating the number of entries in the variable-size 4905array 'entries'. If the number of entries is too low to describe all Hyper-V 4906feature leaves, an error (E2BIG) is returned. If the number is more or equal 4907to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the 4908number of valid entries in the 'entries' array, which is then filled. 4909 4910'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved, 4911userspace should not expect to get any particular value there. 4912 4913Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike 4914system ioctl which exposes all supported feature bits unconditionally, vcpu 4915version has the following quirks: 4916 4917- HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED 4918 feature bit are only exposed when Enlightened VMCS was previously enabled 4919 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS). 4920- HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC. 4921 (presumes KVM_CREATE_IRQCHIP has already been called). 4922 49234.119 KVM_ARM_VCPU_FINALIZE 4924--------------------------- 4925 4926:Architectures: arm64 4927:Type: vcpu ioctl 4928:Parameters: int feature (in) 4929:Returns: 0 on success, -1 on error 4930 4931Errors: 4932 4933 ====== ============================================================== 4934 EPERM feature not enabled, needs configuration, or already finalized 4935 EINVAL feature unknown or not present 4936 ====== ============================================================== 4937 4938Recognised values for feature: 4939 4940 ===== =========================================== 4941 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE) 4942 ===== =========================================== 4943 4944Finalizes the configuration of the specified vcpu feature. 4945 4946The vcpu must already have been initialised, enabling the affected feature, by 4947means of a successful KVM_ARM_VCPU_INIT call with the appropriate flag set in 4948features[]. 4949 4950For affected vcpu features, this is a mandatory step that must be performed 4951before the vcpu is fully usable. 4952 4953Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be 4954configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration 4955that should be performaned and how to do it are feature-dependent. 4956 4957Other calls that depend on a particular feature being finalized, such as 4958KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with 4959-EPERM unless the feature has already been finalized by means of a 4960KVM_ARM_VCPU_FINALIZE call. 4961 4962See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization 4963using this ioctl. 4964 49654.120 KVM_SET_PMU_EVENT_FILTER 4966------------------------------ 4967 4968:Capability: KVM_CAP_PMU_EVENT_FILTER 4969:Architectures: x86 4970:Type: vm ioctl 4971:Parameters: struct kvm_pmu_event_filter (in) 4972:Returns: 0 on success, -1 on error 4973 4974:: 4975 4976 struct kvm_pmu_event_filter { 4977 __u32 action; 4978 __u32 nevents; 4979 __u32 fixed_counter_bitmap; 4980 __u32 flags; 4981 __u32 pad[4]; 4982 __u64 events[0]; 4983 }; 4984 4985This ioctl restricts the set of PMU events that the guest can program. 4986The argument holds a list of events which will be allowed or denied. 4987The eventsel+umask of each event the guest attempts to program is compared 4988against the events field to determine whether the guest should have access. 4989The events field only controls general purpose counters; fixed purpose 4990counters are controlled by the fixed_counter_bitmap. 4991 4992No flags are defined yet, the field must be zero. 4993 4994Valid values for 'action':: 4995 4996 #define KVM_PMU_EVENT_ALLOW 0 4997 #define KVM_PMU_EVENT_DENY 1 4998 49994.121 KVM_PPC_SVM_OFF 5000--------------------- 5001 5002:Capability: basic 5003:Architectures: powerpc 5004:Type: vm ioctl 5005:Parameters: none 5006:Returns: 0 on successful completion, 5007 5008Errors: 5009 5010 ====== ================================================================ 5011 EINVAL if ultravisor failed to terminate the secure guest 5012 ENOMEM if hypervisor failed to allocate new radix page tables for guest 5013 ====== ================================================================ 5014 5015This ioctl is used to turn off the secure mode of the guest or transition 5016the guest from secure mode to normal mode. This is invoked when the guest 5017is reset. This has no effect if called for a normal guest. 5018 5019This ioctl issues an ultravisor call to terminate the secure guest, 5020unpins the VPA pages and releases all the device pages that are used to 5021track the secure pages by hypervisor. 5022 50234.122 KVM_S390_NORMAL_RESET 5024--------------------------- 5025 5026:Capability: KVM_CAP_S390_VCPU_RESETS 5027:Architectures: s390 5028:Type: vcpu ioctl 5029:Parameters: none 5030:Returns: 0 5031 5032This ioctl resets VCPU registers and control structures according to 5033the cpu reset definition in the POP (Principles Of Operation). 5034 50354.123 KVM_S390_INITIAL_RESET 5036---------------------------- 5037 5038:Capability: none 5039:Architectures: s390 5040:Type: vcpu ioctl 5041:Parameters: none 5042:Returns: 0 5043 5044This ioctl resets VCPU registers and control structures according to 5045the initial cpu reset definition in the POP. However, the cpu is not 5046put into ESA mode. This reset is a superset of the normal reset. 5047 50484.124 KVM_S390_CLEAR_RESET 5049-------------------------- 5050 5051:Capability: KVM_CAP_S390_VCPU_RESETS 5052:Architectures: s390 5053:Type: vcpu ioctl 5054:Parameters: none 5055:Returns: 0 5056 5057This ioctl resets VCPU registers and control structures according to 5058the clear cpu reset definition in the POP. However, the cpu is not put 5059into ESA mode. This reset is a superset of the initial reset. 5060 5061 50624.125 KVM_S390_PV_COMMAND 5063------------------------- 5064 5065:Capability: KVM_CAP_S390_PROTECTED 5066:Architectures: s390 5067:Type: vm ioctl 5068:Parameters: struct kvm_pv_cmd 5069:Returns: 0 on success, < 0 on error 5070 5071:: 5072 5073 struct kvm_pv_cmd { 5074 __u32 cmd; /* Command to be executed */ 5075 __u16 rc; /* Ultravisor return code */ 5076 __u16 rrc; /* Ultravisor return reason code */ 5077 __u64 data; /* Data or address */ 5078 __u32 flags; /* flags for future extensions. Must be 0 for now */ 5079 __u32 reserved[3]; 5080 }; 5081 5082cmd values: 5083 5084KVM_PV_ENABLE 5085 Allocate memory and register the VM with the Ultravisor, thereby 5086 donating memory to the Ultravisor that will become inaccessible to 5087 KVM. All existing CPUs are converted to protected ones. After this 5088 command has succeeded, any CPU added via hotplug will become 5089 protected during its creation as well. 5090 5091 Errors: 5092 5093 ===== ============================= 5094 EINTR an unmasked signal is pending 5095 ===== ============================= 5096 5097KVM_PV_DISABLE 5098 5099 Deregister the VM from the Ultravisor and reclaim the memory that 5100 had been donated to the Ultravisor, making it usable by the kernel 5101 again. All registered VCPUs are converted back to non-protected 5102 ones. 5103 5104KVM_PV_VM_SET_SEC_PARMS 5105 Pass the image header from VM memory to the Ultravisor in 5106 preparation of image unpacking and verification. 5107 5108KVM_PV_VM_UNPACK 5109 Unpack (protect and decrypt) a page of the encrypted boot image. 5110 5111KVM_PV_VM_VERIFY 5112 Verify the integrity of the unpacked image. Only if this succeeds, 5113 KVM is allowed to start protected VCPUs. 5114 51154.126 KVM_X86_SET_MSR_FILTER 5116---------------------------- 5117 5118:Capability: KVM_CAP_X86_MSR_FILTER 5119:Architectures: x86 5120:Type: vm ioctl 5121:Parameters: struct kvm_msr_filter 5122:Returns: 0 on success, < 0 on error 5123 5124:: 5125 5126 struct kvm_msr_filter_range { 5127 #define KVM_MSR_FILTER_READ (1 << 0) 5128 #define KVM_MSR_FILTER_WRITE (1 << 1) 5129 __u32 flags; 5130 __u32 nmsrs; /* number of msrs in bitmap */ 5131 __u32 base; /* MSR index the bitmap starts at */ 5132 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 5133 }; 5134 5135 #define KVM_MSR_FILTER_MAX_RANGES 16 5136 struct kvm_msr_filter { 5137 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 5138 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 5139 __u32 flags; 5140 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 5141 }; 5142 5143flags values for ``struct kvm_msr_filter_range``: 5144 5145``KVM_MSR_FILTER_READ`` 5146 5147 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 5148 indicates that a read should immediately fail, while a 1 indicates that 5149 a read for a particular MSR should be handled regardless of the default 5150 filter action. 5151 5152``KVM_MSR_FILTER_WRITE`` 5153 5154 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 5155 indicates that a write should immediately fail, while a 1 indicates that 5156 a write for a particular MSR should be handled regardless of the default 5157 filter action. 5158 5159``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE`` 5160 5161 Filter both read and write accesses to MSRs using the given bitmap. A 0 5162 in the bitmap indicates that both reads and writes should immediately fail, 5163 while a 1 indicates that reads and writes for a particular MSR are not 5164 filtered by this range. 5165 5166flags values for ``struct kvm_msr_filter``: 5167 5168``KVM_MSR_FILTER_DEFAULT_ALLOW`` 5169 5170 If no filter range matches an MSR index that is getting accessed, KVM will 5171 fall back to allowing access to the MSR. 5172 5173``KVM_MSR_FILTER_DEFAULT_DENY`` 5174 5175 If no filter range matches an MSR index that is getting accessed, KVM will 5176 fall back to rejecting access to the MSR. In this mode, all MSRs that should 5177 be processed by KVM need to explicitly be marked as allowed in the bitmaps. 5178 5179This ioctl allows user space to define up to 16 bitmaps of MSR ranges to 5180specify whether a certain MSR access should be explicitly filtered for or not. 5181 5182If this ioctl has never been invoked, MSR accesses are not guarded and the 5183default KVM in-kernel emulation behavior is fully preserved. 5184 5185Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 5186filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 5187an error. 5188 5189As soon as the filtering is in place, every MSR access is processed through 5190the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff); 5191x2APIC MSRs are always allowed, independent of the ``default_allow`` setting, 5192and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base 5193register. 5194 5195If a bit is within one of the defined ranges, read and write accesses are 5196guarded by the bitmap's value for the MSR index if the kind of access 5197is included in the ``struct kvm_msr_filter_range`` flags. If no range 5198cover this particular access, the behavior is determined by the flags 5199field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 5200and ``KVM_MSR_FILTER_DEFAULT_DENY``. 5201 5202Each bitmap range specifies a range of MSRs to potentially allow access on. 5203The range goes from MSR index [base .. base+nmsrs]. The flags field 5204indicates whether reads, writes or both reads and writes are filtered 5205by setting a 1 bit in the bitmap for the corresponding MSR index. 5206 5207If an MSR access is not permitted through the filtering, it generates a 5208#GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that 5209allows user space to deflect and potentially handle various MSR accesses 5210into user space. 5211 5212Note, invoking this ioctl with a vCPU is running is inherently racy. However, 5213KVM does guarantee that vCPUs will see either the previous filter or the new 5214filter, e.g. MSRs with identical settings in both the old and new filter will 5215have deterministic behavior. 5216 52174.127 KVM_XEN_HVM_SET_ATTR 5218-------------------------- 5219 5220:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5221:Architectures: x86 5222:Type: vm ioctl 5223:Parameters: struct kvm_xen_hvm_attr 5224:Returns: 0 on success, < 0 on error 5225 5226:: 5227 5228 struct kvm_xen_hvm_attr { 5229 __u16 type; 5230 __u16 pad[3]; 5231 union { 5232 __u8 long_mode; 5233 __u8 vector; 5234 struct { 5235 __u64 gfn; 5236 } shared_info; 5237 __u64 pad[4]; 5238 } u; 5239 }; 5240 5241type values: 5242 5243KVM_XEN_ATTR_TYPE_LONG_MODE 5244 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This 5245 determines the layout of the shared info pages exposed to the VM. 5246 5247KVM_XEN_ATTR_TYPE_SHARED_INFO 5248 Sets the guest physical frame number at which the Xen "shared info" 5249 page resides. Note that although Xen places vcpu_info for the first 5250 32 vCPUs in the shared_info page, KVM does not automatically do so 5251 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO be used 5252 explicitly even when the vcpu_info for a given vCPU resides at the 5253 "default" location in the shared_info page. This is because KVM is 5254 not aware of the Xen CPU id which is used as the index into the 5255 vcpu_info[] array, so cannot know the correct default location. 5256 5257 Note that the shared info page may be constantly written to by KVM; 5258 it contains the event channel bitmap used to deliver interrupts to 5259 a Xen guest, amongst other things. It is exempt from dirty tracking 5260 mechanisms — KVM will not explicitly mark the page as dirty each 5261 time an event channel interrupt is delivered to the guest! Thus, 5262 userspace should always assume that the designated GFN is dirty if 5263 any vCPU has been running or any event channel interrupts can be 5264 routed to the guest. 5265 5266KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 5267 Sets the exception vector used to deliver Xen event channel upcalls. 5268 52694.127 KVM_XEN_HVM_GET_ATTR 5270-------------------------- 5271 5272:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5273:Architectures: x86 5274:Type: vm ioctl 5275:Parameters: struct kvm_xen_hvm_attr 5276:Returns: 0 on success, < 0 on error 5277 5278Allows Xen VM attributes to be read. For the structure and types, 5279see KVM_XEN_HVM_SET_ATTR above. 5280 52814.128 KVM_XEN_VCPU_SET_ATTR 5282--------------------------- 5283 5284:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5285:Architectures: x86 5286:Type: vcpu ioctl 5287:Parameters: struct kvm_xen_vcpu_attr 5288:Returns: 0 on success, < 0 on error 5289 5290:: 5291 5292 struct kvm_xen_vcpu_attr { 5293 __u16 type; 5294 __u16 pad[3]; 5295 union { 5296 __u64 gpa; 5297 __u64 pad[4]; 5298 struct { 5299 __u64 state; 5300 __u64 state_entry_time; 5301 __u64 time_running; 5302 __u64 time_runnable; 5303 __u64 time_blocked; 5304 __u64 time_offline; 5305 } runstate; 5306 } u; 5307 }; 5308 5309type values: 5310 5311KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 5312 Sets the guest physical address of the vcpu_info for a given vCPU. 5313 As with the shared_info page for the VM, the corresponding page may be 5314 dirtied at any time if event channel interrupt delivery is enabled, so 5315 userspace should always assume that the page is dirty without relying 5316 on dirty logging. 5317 5318KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 5319 Sets the guest physical address of an additional pvclock structure 5320 for a given vCPU. This is typically used for guest vsyscall support. 5321 5322KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 5323 Sets the guest physical address of the vcpu_runstate_info for a given 5324 vCPU. This is how a Xen guest tracks CPU state such as steal time. 5325 5326KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 5327 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of 5328 the given vCPU from the .u.runstate.state member of the structure. 5329 KVM automatically accounts running and runnable time but blocked 5330 and offline states are only entered explicitly. 5331 5332KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 5333 Sets all fields of the vCPU runstate data from the .u.runstate member 5334 of the structure, including the current runstate. The state_entry_time 5335 must equal the sum of the other four times. 5336 5337KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 5338 This *adds* the contents of the .u.runstate members of the structure 5339 to the corresponding members of the given vCPU's runstate data, thus 5340 permitting atomic adjustments to the runstate times. The adjustment 5341 to the state_entry_time must equal the sum of the adjustments to the 5342 other four times. The state field must be set to -1, or to a valid 5343 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked 5344 or RUNSTATE_offline) to set the current accounted state as of the 5345 adjusted state_entry_time. 5346 53474.129 KVM_XEN_VCPU_GET_ATTR 5348--------------------------- 5349 5350:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5351:Architectures: x86 5352:Type: vcpu ioctl 5353:Parameters: struct kvm_xen_vcpu_attr 5354:Returns: 0 on success, < 0 on error 5355 5356Allows Xen vCPU attributes to be read. For the structure and types, 5357see KVM_XEN_VCPU_SET_ATTR above. 5358 5359The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used 5360with the KVM_XEN_VCPU_GET_ATTR ioctl. 5361 53624.130 KVM_ARM_MTE_COPY_TAGS 5363--------------------------- 5364 5365:Capability: KVM_CAP_ARM_MTE 5366:Architectures: arm64 5367:Type: vm ioctl 5368:Parameters: struct kvm_arm_copy_mte_tags 5369:Returns: number of bytes copied, < 0 on error (-EINVAL for incorrect 5370 arguments, -EFAULT if memory cannot be accessed). 5371 5372:: 5373 5374 struct kvm_arm_copy_mte_tags { 5375 __u64 guest_ipa; 5376 __u64 length; 5377 void __user *addr; 5378 __u64 flags; 5379 __u64 reserved[2]; 5380 }; 5381 5382Copies Memory Tagging Extension (MTE) tags to/from guest tag memory. The 5383``guest_ipa`` and ``length`` fields must be ``PAGE_SIZE`` aligned. The ``addr`` 5384field must point to a buffer which the tags will be copied to or from. 5385 5386``flags`` specifies the direction of copy, either ``KVM_ARM_TAGS_TO_GUEST`` or 5387``KVM_ARM_TAGS_FROM_GUEST``. 5388 5389The size of the buffer to store the tags is ``(length / 16)`` bytes 5390(granules in MTE are 16 bytes long). Each byte contains a single tag 5391value. This matches the format of ``PTRACE_PEEKMTETAGS`` and 5392``PTRACE_POKEMTETAGS``. 5393 5394If an error occurs before any data is copied then a negative error code is 5395returned. If some tags have been copied before an error occurs then the number 5396of bytes successfully copied is returned. If the call completes successfully 5397then ``length`` is returned. 5398 53994.131 KVM_GET_SREGS2 5400-------------------- 5401 5402:Capability: KVM_CAP_SREGS2 5403:Architectures: x86 5404:Type: vcpu ioctl 5405:Parameters: struct kvm_sregs2 (out) 5406:Returns: 0 on success, -1 on error 5407 5408Reads special registers from the vcpu. 5409This ioctl (when supported) replaces the KVM_GET_SREGS. 5410 5411:: 5412 5413 struct kvm_sregs2 { 5414 /* out (KVM_GET_SREGS2) / in (KVM_SET_SREGS2) */ 5415 struct kvm_segment cs, ds, es, fs, gs, ss; 5416 struct kvm_segment tr, ldt; 5417 struct kvm_dtable gdt, idt; 5418 __u64 cr0, cr2, cr3, cr4, cr8; 5419 __u64 efer; 5420 __u64 apic_base; 5421 __u64 flags; 5422 __u64 pdptrs[4]; 5423 }; 5424 5425flags values for ``kvm_sregs2``: 5426 5427``KVM_SREGS2_FLAGS_PDPTRS_VALID`` 5428 5429 Indicates thats the struct contain valid PDPTR values. 5430 5431 54324.132 KVM_SET_SREGS2 5433-------------------- 5434 5435:Capability: KVM_CAP_SREGS2 5436:Architectures: x86 5437:Type: vcpu ioctl 5438:Parameters: struct kvm_sregs2 (in) 5439:Returns: 0 on success, -1 on error 5440 5441Writes special registers into the vcpu. 5442See KVM_GET_SREGS2 for the data structures. 5443This ioctl (when supported) replaces the KVM_SET_SREGS. 5444 54454.133 KVM_GET_STATS_FD 5446---------------------- 5447 5448:Capability: KVM_CAP_STATS_BINARY_FD 5449:Architectures: all 5450:Type: vm ioctl, vcpu ioctl 5451:Parameters: none 5452:Returns: statistics file descriptor on success, < 0 on error 5453 5454Errors: 5455 5456 ====== ====================================================== 5457 ENOMEM if the fd could not be created due to lack of memory 5458 EMFILE if the number of opened files exceeds the limit 5459 ====== ====================================================== 5460 5461The returned file descriptor can be used to read VM/vCPU statistics data in 5462binary format. The data in the file descriptor consists of four blocks 5463organized as follows: 5464 5465+-------------+ 5466| Header | 5467+-------------+ 5468| id string | 5469+-------------+ 5470| Descriptors | 5471+-------------+ 5472| Stats Data | 5473+-------------+ 5474 5475Apart from the header starting at offset 0, please be aware that it is 5476not guaranteed that the four blocks are adjacent or in the above order; 5477the offsets of the id, descriptors and data blocks are found in the 5478header. However, all four blocks are aligned to 64 bit offsets in the 5479file and they do not overlap. 5480 5481All blocks except the data block are immutable. Userspace can read them 5482only one time after retrieving the file descriptor, and then use ``pread`` or 5483``lseek`` to read the statistics repeatedly. 5484 5485All data is in system endianness. 5486 5487The format of the header is as follows:: 5488 5489 struct kvm_stats_header { 5490 __u32 flags; 5491 __u32 name_size; 5492 __u32 num_desc; 5493 __u32 id_offset; 5494 __u32 desc_offset; 5495 __u32 data_offset; 5496 }; 5497 5498The ``flags`` field is not used at the moment. It is always read as 0. 5499 5500The ``name_size`` field is the size (in byte) of the statistics name string 5501(including trailing '\0') which is contained in the "id string" block and 5502appended at the end of every descriptor. 5503 5504The ``num_desc`` field is the number of descriptors that are included in the 5505descriptor block. (The actual number of values in the data block may be 5506larger, since each descriptor may comprise more than one value). 5507 5508The ``id_offset`` field is the offset of the id string from the start of the 5509file indicated by the file descriptor. It is a multiple of 8. 5510 5511The ``desc_offset`` field is the offset of the Descriptors block from the start 5512of the file indicated by the file descriptor. It is a multiple of 8. 5513 5514The ``data_offset`` field is the offset of the Stats Data block from the start 5515of the file indicated by the file descriptor. It is a multiple of 8. 5516 5517The id string block contains a string which identifies the file descriptor on 5518which KVM_GET_STATS_FD was invoked. The size of the block, including the 5519trailing ``'\0'``, is indicated by the ``name_size`` field in the header. 5520 5521The descriptors block is only needed to be read once for the lifetime of the 5522file descriptor contains a sequence of ``struct kvm_stats_desc``, each followed 5523by a string of size ``name_size``. 5524:: 5525 5526 #define KVM_STATS_TYPE_SHIFT 0 5527 #define KVM_STATS_TYPE_MASK (0xF << KVM_STATS_TYPE_SHIFT) 5528 #define KVM_STATS_TYPE_CUMULATIVE (0x0 << KVM_STATS_TYPE_SHIFT) 5529 #define KVM_STATS_TYPE_INSTANT (0x1 << KVM_STATS_TYPE_SHIFT) 5530 #define KVM_STATS_TYPE_PEAK (0x2 << KVM_STATS_TYPE_SHIFT) 5531 #define KVM_STATS_TYPE_LINEAR_HIST (0x3 << KVM_STATS_TYPE_SHIFT) 5532 #define KVM_STATS_TYPE_LOG_HIST (0x4 << KVM_STATS_TYPE_SHIFT) 5533 #define KVM_STATS_TYPE_MAX KVM_STATS_TYPE_LOG_HIST 5534 5535 #define KVM_STATS_UNIT_SHIFT 4 5536 #define KVM_STATS_UNIT_MASK (0xF << KVM_STATS_UNIT_SHIFT) 5537 #define KVM_STATS_UNIT_NONE (0x0 << KVM_STATS_UNIT_SHIFT) 5538 #define KVM_STATS_UNIT_BYTES (0x1 << KVM_STATS_UNIT_SHIFT) 5539 #define KVM_STATS_UNIT_SECONDS (0x2 << KVM_STATS_UNIT_SHIFT) 5540 #define KVM_STATS_UNIT_CYCLES (0x3 << KVM_STATS_UNIT_SHIFT) 5541 #define KVM_STATS_UNIT_MAX KVM_STATS_UNIT_CYCLES 5542 5543 #define KVM_STATS_BASE_SHIFT 8 5544 #define KVM_STATS_BASE_MASK (0xF << KVM_STATS_BASE_SHIFT) 5545 #define KVM_STATS_BASE_POW10 (0x0 << KVM_STATS_BASE_SHIFT) 5546 #define KVM_STATS_BASE_POW2 (0x1 << KVM_STATS_BASE_SHIFT) 5547 #define KVM_STATS_BASE_MAX KVM_STATS_BASE_POW2 5548 5549 struct kvm_stats_desc { 5550 __u32 flags; 5551 __s16 exponent; 5552 __u16 size; 5553 __u32 offset; 5554 __u32 bucket_size; 5555 char name[]; 5556 }; 5557 5558The ``flags`` field contains the type and unit of the statistics data described 5559by this descriptor. Its endianness is CPU native. 5560The following flags are supported: 5561 5562Bits 0-3 of ``flags`` encode the type: 5563 5564 * ``KVM_STATS_TYPE_CUMULATIVE`` 5565 The statistics reports a cumulative count. The value of data can only be increased. 5566 Most of the counters used in KVM are of this type. 5567 The corresponding ``size`` field for this type is always 1. 5568 All cumulative statistics data are read/write. 5569 * ``KVM_STATS_TYPE_INSTANT`` 5570 The statistics reports an instantaneous value. Its value can be increased or 5571 decreased. This type is usually used as a measurement of some resources, 5572 like the number of dirty pages, the number of large pages, etc. 5573 All instant statistics are read only. 5574 The corresponding ``size`` field for this type is always 1. 5575 * ``KVM_STATS_TYPE_PEAK`` 5576 The statistics data reports a peak value, for example the maximum number 5577 of items in a hash table bucket, the longest time waited and so on. 5578 The value of data can only be increased. 5579 The corresponding ``size`` field for this type is always 1. 5580 * ``KVM_STATS_TYPE_LINEAR_HIST`` 5581 The statistic is reported as a linear histogram. The number of 5582 buckets is specified by the ``size`` field. The size of buckets is specified 5583 by the ``hist_param`` field. The range of the Nth bucket (1 <= N < ``size``) 5584 is [``hist_param``*(N-1), ``hist_param``*N), while the range of the last 5585 bucket is [``hist_param``*(``size``-1), +INF). (+INF means positive infinity 5586 value.) The bucket value indicates how many samples fell in the bucket's range. 5587 * ``KVM_STATS_TYPE_LOG_HIST`` 5588 The statistic is reported as a logarithmic histogram. The number of 5589 buckets is specified by the ``size`` field. The range of the first bucket is 5590 [0, 1), while the range of the last bucket is [pow(2, ``size``-2), +INF). 5591 Otherwise, The Nth bucket (1 < N < ``size``) covers 5592 [pow(2, N-2), pow(2, N-1)). The bucket value indicates how many samples fell 5593 in the bucket's range. 5594 5595Bits 4-7 of ``flags`` encode the unit: 5596 5597 * ``KVM_STATS_UNIT_NONE`` 5598 There is no unit for the value of statistics data. This usually means that 5599 the value is a simple counter of an event. 5600 * ``KVM_STATS_UNIT_BYTES`` 5601 It indicates that the statistics data is used to measure memory size, in the 5602 unit of Byte, KiByte, MiByte, GiByte, etc. The unit of the data is 5603 determined by the ``exponent`` field in the descriptor. 5604 * ``KVM_STATS_UNIT_SECONDS`` 5605 It indicates that the statistics data is used to measure time or latency. 5606 * ``KVM_STATS_UNIT_CYCLES`` 5607 It indicates that the statistics data is used to measure CPU clock cycles. 5608 5609Bits 8-11 of ``flags``, together with ``exponent``, encode the scale of the 5610unit: 5611 5612 * ``KVM_STATS_BASE_POW10`` 5613 The scale is based on power of 10. It is used for measurement of time and 5614 CPU clock cycles. For example, an exponent of -9 can be used with 5615 ``KVM_STATS_UNIT_SECONDS`` to express that the unit is nanoseconds. 5616 * ``KVM_STATS_BASE_POW2`` 5617 The scale is based on power of 2. It is used for measurement of memory size. 5618 For example, an exponent of 20 can be used with ``KVM_STATS_UNIT_BYTES`` to 5619 express that the unit is MiB. 5620 5621The ``size`` field is the number of values of this statistics data. Its 5622value is usually 1 for most of simple statistics. 1 means it contains an 5623unsigned 64bit data. 5624 5625The ``offset`` field is the offset from the start of Data Block to the start of 5626the corresponding statistics data. 5627 5628The ``bucket_size`` field is used as a parameter for histogram statistics data. 5629It is only used by linear histogram statistics data, specifying the size of a 5630bucket. 5631 5632The ``name`` field is the name string of the statistics data. The name string 5633starts at the end of ``struct kvm_stats_desc``. The maximum length including 5634the trailing ``'\0'``, is indicated by ``name_size`` in the header. 5635 5636The Stats Data block contains an array of 64-bit values in the same order 5637as the descriptors in Descriptors block. 5638 56394.134 KVM_GET_XSAVE2 5640-------------------- 5641 5642:Capability: KVM_CAP_XSAVE2 5643:Architectures: x86 5644:Type: vcpu ioctl 5645:Parameters: struct kvm_xsave (out) 5646:Returns: 0 on success, -1 on error 5647 5648 5649:: 5650 5651 struct kvm_xsave { 5652 __u32 region[1024]; 5653 __u32 extra[0]; 5654 }; 5655 5656This ioctl would copy current vcpu's xsave struct to the userspace. It 5657copies as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) 5658when invoked on the vm file descriptor. The size value returned by 5659KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 5660Currently, it is only greater than 4096 if a dynamic feature has been 5661enabled with ``arch_prctl()``, but this may change in the future. 5662 5663The offsets of the state save areas in struct kvm_xsave follow the contents 5664of CPUID leaf 0xD on the host. 5665 5666 56675. The kvm_run structure 5668======================== 5669 5670Application code obtains a pointer to the kvm_run structure by 5671mmap()ing a vcpu fd. From that point, application code can control 5672execution by changing fields in kvm_run prior to calling the KVM_RUN 5673ioctl, and obtain information about the reason KVM_RUN returned by 5674looking up structure members. 5675 5676:: 5677 5678 struct kvm_run { 5679 /* in */ 5680 __u8 request_interrupt_window; 5681 5682Request that KVM_RUN return when it becomes possible to inject external 5683interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 5684 5685:: 5686 5687 __u8 immediate_exit; 5688 5689This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 5690exits immediately, returning -EINTR. In the common scenario where a 5691signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 5692to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 5693Rather than blocking the signal outside KVM_RUN, userspace can set up 5694a signal handler that sets run->immediate_exit to a non-zero value. 5695 5696This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 5697 5698:: 5699 5700 __u8 padding1[6]; 5701 5702 /* out */ 5703 __u32 exit_reason; 5704 5705When KVM_RUN has returned successfully (return value 0), this informs 5706application code why KVM_RUN has returned. Allowable values for this 5707field are detailed below. 5708 5709:: 5710 5711 __u8 ready_for_interrupt_injection; 5712 5713If request_interrupt_window has been specified, this field indicates 5714an interrupt can be injected now with KVM_INTERRUPT. 5715 5716:: 5717 5718 __u8 if_flag; 5719 5720The value of the current interrupt flag. Only valid if in-kernel 5721local APIC is not used. 5722 5723:: 5724 5725 __u16 flags; 5726 5727More architecture-specific flags detailing state of the VCPU that may 5728affect the device's behavior. Current defined flags:: 5729 5730 /* x86, set if the VCPU is in system management mode */ 5731 #define KVM_RUN_X86_SMM (1 << 0) 5732 /* x86, set if bus lock detected in VM */ 5733 #define KVM_RUN_BUS_LOCK (1 << 1) 5734 5735:: 5736 5737 /* in (pre_kvm_run), out (post_kvm_run) */ 5738 __u64 cr8; 5739 5740The value of the cr8 register. Only valid if in-kernel local APIC is 5741not used. Both input and output. 5742 5743:: 5744 5745 __u64 apic_base; 5746 5747The value of the APIC BASE msr. Only valid if in-kernel local 5748APIC is not used. Both input and output. 5749 5750:: 5751 5752 union { 5753 /* KVM_EXIT_UNKNOWN */ 5754 struct { 5755 __u64 hardware_exit_reason; 5756 } hw; 5757 5758If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 5759reasons. Further architecture-specific information is available in 5760hardware_exit_reason. 5761 5762:: 5763 5764 /* KVM_EXIT_FAIL_ENTRY */ 5765 struct { 5766 __u64 hardware_entry_failure_reason; 5767 __u32 cpu; /* if KVM_LAST_CPU */ 5768 } fail_entry; 5769 5770If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 5771to unknown reasons. Further architecture-specific information is 5772available in hardware_entry_failure_reason. 5773 5774:: 5775 5776 /* KVM_EXIT_EXCEPTION */ 5777 struct { 5778 __u32 exception; 5779 __u32 error_code; 5780 } ex; 5781 5782Unused. 5783 5784:: 5785 5786 /* KVM_EXIT_IO */ 5787 struct { 5788 #define KVM_EXIT_IO_IN 0 5789 #define KVM_EXIT_IO_OUT 1 5790 __u8 direction; 5791 __u8 size; /* bytes */ 5792 __u16 port; 5793 __u32 count; 5794 __u64 data_offset; /* relative to kvm_run start */ 5795 } io; 5796 5797If exit_reason is KVM_EXIT_IO, then the vcpu has 5798executed a port I/O instruction which could not be satisfied by kvm. 5799data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 5800where kvm expects application code to place the data for the next 5801KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 5802 5803:: 5804 5805 /* KVM_EXIT_DEBUG */ 5806 struct { 5807 struct kvm_debug_exit_arch arch; 5808 } debug; 5809 5810If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 5811for which architecture specific information is returned. 5812 5813:: 5814 5815 /* KVM_EXIT_MMIO */ 5816 struct { 5817 __u64 phys_addr; 5818 __u8 data[8]; 5819 __u32 len; 5820 __u8 is_write; 5821 } mmio; 5822 5823If exit_reason is KVM_EXIT_MMIO, then the vcpu has 5824executed a memory-mapped I/O instruction which could not be satisfied 5825by kvm. The 'data' member contains the written data if 'is_write' is 5826true, and should be filled by application code otherwise. 5827 5828The 'data' member contains, in its first 'len' bytes, the value as it would 5829appear if the VCPU performed a load or store of the appropriate width directly 5830to the byte array. 5831 5832.. note:: 5833 5834 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN, 5835 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding 5836 operations are complete (and guest state is consistent) only after userspace 5837 has re-entered the kernel with KVM_RUN. The kernel side will first finish 5838 incomplete operations and then check for pending signals. 5839 5840 The pending state of the operation is not preserved in state which is 5841 visible to userspace, thus userspace should ensure that the operation is 5842 completed before performing a live migration. Userspace can re-enter the 5843 guest with an unmasked signal pending or with the immediate_exit field set 5844 to complete pending operations without allowing any further instructions 5845 to be executed. 5846 5847:: 5848 5849 /* KVM_EXIT_HYPERCALL */ 5850 struct { 5851 __u64 nr; 5852 __u64 args[6]; 5853 __u64 ret; 5854 __u32 longmode; 5855 __u32 pad; 5856 } hypercall; 5857 5858Unused. This was once used for 'hypercall to userspace'. To implement 5859such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). 5860 5861.. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 5862 5863:: 5864 5865 /* KVM_EXIT_TPR_ACCESS */ 5866 struct { 5867 __u64 rip; 5868 __u32 is_write; 5869 __u32 pad; 5870 } tpr_access; 5871 5872To be documented (KVM_TPR_ACCESS_REPORTING). 5873 5874:: 5875 5876 /* KVM_EXIT_S390_SIEIC */ 5877 struct { 5878 __u8 icptcode; 5879 __u64 mask; /* psw upper half */ 5880 __u64 addr; /* psw lower half */ 5881 __u16 ipa; 5882 __u32 ipb; 5883 } s390_sieic; 5884 5885s390 specific. 5886 5887:: 5888 5889 /* KVM_EXIT_S390_RESET */ 5890 #define KVM_S390_RESET_POR 1 5891 #define KVM_S390_RESET_CLEAR 2 5892 #define KVM_S390_RESET_SUBSYSTEM 4 5893 #define KVM_S390_RESET_CPU_INIT 8 5894 #define KVM_S390_RESET_IPL 16 5895 __u64 s390_reset_flags; 5896 5897s390 specific. 5898 5899:: 5900 5901 /* KVM_EXIT_S390_UCONTROL */ 5902 struct { 5903 __u64 trans_exc_code; 5904 __u32 pgm_code; 5905 } s390_ucontrol; 5906 5907s390 specific. A page fault has occurred for a user controlled virtual 5908machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be 5909resolved by the kernel. 5910The program code and the translation exception code that were placed 5911in the cpu's lowcore are presented here as defined by the z Architecture 5912Principles of Operation Book in the Chapter for Dynamic Address Translation 5913(DAT) 5914 5915:: 5916 5917 /* KVM_EXIT_DCR */ 5918 struct { 5919 __u32 dcrn; 5920 __u32 data; 5921 __u8 is_write; 5922 } dcr; 5923 5924Deprecated - was used for 440 KVM. 5925 5926:: 5927 5928 /* KVM_EXIT_OSI */ 5929 struct { 5930 __u64 gprs[32]; 5931 } osi; 5932 5933MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 5934hypercalls and exit with this exit struct that contains all the guest gprs. 5935 5936If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 5937Userspace can now handle the hypercall and when it's done modify the gprs as 5938necessary. Upon guest entry all guest GPRs will then be replaced by the values 5939in this struct. 5940 5941:: 5942 5943 /* KVM_EXIT_PAPR_HCALL */ 5944 struct { 5945 __u64 nr; 5946 __u64 ret; 5947 __u64 args[9]; 5948 } papr_hcall; 5949 5950This is used on 64-bit PowerPC when emulating a pSeries partition, 5951e.g. with the 'pseries' machine type in qemu. It occurs when the 5952guest does a hypercall using the 'sc 1' instruction. The 'nr' field 5953contains the hypercall number (from the guest R3), and 'args' contains 5954the arguments (from the guest R4 - R12). Userspace should put the 5955return code in 'ret' and any extra returned values in args[]. 5956The possible hypercalls are defined in the Power Architecture Platform 5957Requirements (PAPR) document available from www.power.org (free 5958developer registration required to access it). 5959 5960:: 5961 5962 /* KVM_EXIT_S390_TSCH */ 5963 struct { 5964 __u16 subchannel_id; 5965 __u16 subchannel_nr; 5966 __u32 io_int_parm; 5967 __u32 io_int_word; 5968 __u32 ipb; 5969 __u8 dequeued; 5970 } s390_tsch; 5971 5972s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 5973and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 5974interrupt for the target subchannel has been dequeued and subchannel_id, 5975subchannel_nr, io_int_parm and io_int_word contain the parameters for that 5976interrupt. ipb is needed for instruction parameter decoding. 5977 5978:: 5979 5980 /* KVM_EXIT_EPR */ 5981 struct { 5982 __u32 epr; 5983 } epr; 5984 5985On FSL BookE PowerPC chips, the interrupt controller has a fast patch 5986interrupt acknowledge path to the core. When the core successfully 5987delivers an interrupt, it automatically populates the EPR register with 5988the interrupt vector number and acknowledges the interrupt inside 5989the interrupt controller. 5990 5991In case the interrupt controller lives in user space, we need to do 5992the interrupt acknowledge cycle through it to fetch the next to be 5993delivered interrupt vector using this exit. 5994 5995It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 5996external interrupt has just been delivered into the guest. User space 5997should put the acknowledged interrupt vector into the 'epr' field. 5998 5999:: 6000 6001 /* KVM_EXIT_SYSTEM_EVENT */ 6002 struct { 6003 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 6004 #define KVM_SYSTEM_EVENT_RESET 2 6005 #define KVM_SYSTEM_EVENT_CRASH 3 6006 __u32 type; 6007 __u64 flags; 6008 } system_event; 6009 6010If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 6011a system-level event using some architecture specific mechanism (hypercall 6012or some special instruction). In case of ARM64, this is triggered using 6013HVC instruction based PSCI call from the vcpu. The 'type' field describes 6014the system-level event type. The 'flags' field describes architecture 6015specific flags for the system-level event. 6016 6017Valid values for 'type' are: 6018 6019 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 6020 VM. Userspace is not obliged to honour this, and if it does honour 6021 this does not need to destroy the VM synchronously (ie it may call 6022 KVM_RUN again before shutdown finally occurs). 6023 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 6024 As with SHUTDOWN, userspace can choose to ignore the request, or 6025 to schedule the reset to occur in the future and may call KVM_RUN again. 6026 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 6027 has requested a crash condition maintenance. Userspace can choose 6028 to ignore the request, or to gather VM memory core dump and/or 6029 reset/shutdown of the VM. 6030 6031Valid flags are: 6032 6033 - KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2 (arm64 only) -- the guest issued 6034 a SYSTEM_RESET2 call according to v1.1 of the PSCI specification. 6035 6036:: 6037 6038 /* KVM_EXIT_IOAPIC_EOI */ 6039 struct { 6040 __u8 vector; 6041 } eoi; 6042 6043Indicates that the VCPU's in-kernel local APIC received an EOI for a 6044level-triggered IOAPIC interrupt. This exit only triggers when the 6045IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 6046the userspace IOAPIC should process the EOI and retrigger the interrupt if 6047it is still asserted. Vector is the LAPIC interrupt vector for which the 6048EOI was received. 6049 6050:: 6051 6052 struct kvm_hyperv_exit { 6053 #define KVM_EXIT_HYPERV_SYNIC 1 6054 #define KVM_EXIT_HYPERV_HCALL 2 6055 #define KVM_EXIT_HYPERV_SYNDBG 3 6056 __u32 type; 6057 __u32 pad1; 6058 union { 6059 struct { 6060 __u32 msr; 6061 __u32 pad2; 6062 __u64 control; 6063 __u64 evt_page; 6064 __u64 msg_page; 6065 } synic; 6066 struct { 6067 __u64 input; 6068 __u64 result; 6069 __u64 params[2]; 6070 } hcall; 6071 struct { 6072 __u32 msr; 6073 __u32 pad2; 6074 __u64 control; 6075 __u64 status; 6076 __u64 send_page; 6077 __u64 recv_page; 6078 __u64 pending_page; 6079 } syndbg; 6080 } u; 6081 }; 6082 /* KVM_EXIT_HYPERV */ 6083 struct kvm_hyperv_exit hyperv; 6084 6085Indicates that the VCPU exits into userspace to process some tasks 6086related to Hyper-V emulation. 6087 6088Valid values for 'type' are: 6089 6090 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 6091 6092Hyper-V SynIC state change. Notification is used to remap SynIC 6093event/message pages and to enable/disable SynIC messages/events processing 6094in userspace. 6095 6096 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about 6097 6098Hyper-V Synthetic debugger state change. Notification is used to either update 6099the pending_page location or to send a control command (send the buffer located 6100in send_page or recv a buffer to recv_page). 6101 6102:: 6103 6104 /* KVM_EXIT_ARM_NISV */ 6105 struct { 6106 __u64 esr_iss; 6107 __u64 fault_ipa; 6108 } arm_nisv; 6109 6110Used on arm64 systems. If a guest accesses memory not in a memslot, 6111KVM will typically return to userspace and ask it to do MMIO emulation on its 6112behalf. However, for certain classes of instructions, no instruction decode 6113(direction, length of memory access) is provided, and fetching and decoding 6114the instruction from the VM is overly complicated to live in the kernel. 6115 6116Historically, when this situation occurred, KVM would print a warning and kill 6117the VM. KVM assumed that if the guest accessed non-memslot memory, it was 6118trying to do I/O, which just couldn't be emulated, and the warning message was 6119phrased accordingly. However, what happened more often was that a guest bug 6120caused access outside the guest memory areas which should lead to a more 6121meaningful warning message and an external abort in the guest, if the access 6122did not fall within an I/O window. 6123 6124Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable 6125this capability at VM creation. Once this is done, these types of errors will 6126instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from 6127the ESR_EL2 in the esr_iss field, and the faulting IPA in the fault_ipa field. 6128Userspace can either fix up the access if it's actually an I/O access by 6129decoding the instruction from guest memory (if it's very brave) and continue 6130executing the guest, or it can decide to suspend, dump, or restart the guest. 6131 6132Note that KVM does not skip the faulting instruction as it does for 6133KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state 6134if it decides to decode and emulate the instruction. 6135 6136:: 6137 6138 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */ 6139 struct { 6140 __u8 error; /* user -> kernel */ 6141 __u8 pad[7]; 6142 __u32 reason; /* kernel -> user */ 6143 __u32 index; /* kernel -> user */ 6144 __u64 data; /* kernel <-> user */ 6145 } msr; 6146 6147Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is 6148enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code 6149will instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR 6150exit for writes. 6151 6152The "reason" field specifies why the MSR trap occurred. User space will only 6153receive MSR exit traps when a particular reason was requested during through 6154ENABLE_CAP. Currently valid exit reasons are: 6155 6156 KVM_MSR_EXIT_REASON_UNKNOWN - access to MSR that is unknown to KVM 6157 KVM_MSR_EXIT_REASON_INVAL - access to invalid MSRs or reserved bits 6158 KVM_MSR_EXIT_REASON_FILTER - access blocked by KVM_X86_SET_MSR_FILTER 6159 6160For KVM_EXIT_X86_RDMSR, the "index" field tells user space which MSR the guest 6161wants to read. To respond to this request with a successful read, user space 6162writes the respective data into the "data" field and must continue guest 6163execution to ensure the read data is transferred into guest register state. 6164 6165If the RDMSR request was unsuccessful, user space indicates that with a "1" in 6166the "error" field. This will inject a #GP into the guest when the VCPU is 6167executed again. 6168 6169For KVM_EXIT_X86_WRMSR, the "index" field tells user space which MSR the guest 6170wants to write. Once finished processing the event, user space must continue 6171vCPU execution. If the MSR write was unsuccessful, user space also sets the 6172"error" field to "1". 6173 6174:: 6175 6176 6177 struct kvm_xen_exit { 6178 #define KVM_EXIT_XEN_HCALL 1 6179 __u32 type; 6180 union { 6181 struct { 6182 __u32 longmode; 6183 __u32 cpl; 6184 __u64 input; 6185 __u64 result; 6186 __u64 params[6]; 6187 } hcall; 6188 } u; 6189 }; 6190 /* KVM_EXIT_XEN */ 6191 struct kvm_hyperv_exit xen; 6192 6193Indicates that the VCPU exits into userspace to process some tasks 6194related to Xen emulation. 6195 6196Valid values for 'type' are: 6197 6198 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall. 6199 Userspace is expected to place the hypercall result into the appropriate 6200 field before invoking KVM_RUN again. 6201 6202:: 6203 6204 /* KVM_EXIT_RISCV_SBI */ 6205 struct { 6206 unsigned long extension_id; 6207 unsigned long function_id; 6208 unsigned long args[6]; 6209 unsigned long ret[2]; 6210 } riscv_sbi; 6211 6212If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has 6213done a SBI call which is not handled by KVM RISC-V kernel module. The details 6214of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The 6215'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the 6216'function_id' field represents function ID of given SBI extension. The 'args' 6217array field of 'riscv_sbi' represents parameters for the SBI call and 'ret' 6218array field represents return values. The userspace should update the return 6219values of SBI call before resuming the VCPU. For more details on RISC-V SBI 6220spec refer, https://github.com/riscv/riscv-sbi-doc. 6221 6222:: 6223 6224 /* Fix the size of the union. */ 6225 char padding[256]; 6226 }; 6227 6228 /* 6229 * shared registers between kvm and userspace. 6230 * kvm_valid_regs specifies the register classes set by the host 6231 * kvm_dirty_regs specified the register classes dirtied by userspace 6232 * struct kvm_sync_regs is architecture specific, as well as the 6233 * bits for kvm_valid_regs and kvm_dirty_regs 6234 */ 6235 __u64 kvm_valid_regs; 6236 __u64 kvm_dirty_regs; 6237 union { 6238 struct kvm_sync_regs regs; 6239 char padding[SYNC_REGS_SIZE_BYTES]; 6240 } s; 6241 6242If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 6243certain guest registers without having to call SET/GET_*REGS. Thus we can 6244avoid some system call overhead if userspace has to handle the exit. 6245Userspace can query the validity of the structure by checking 6246kvm_valid_regs for specific bits. These bits are architecture specific 6247and usually define the validity of a groups of registers. (e.g. one bit 6248for general purpose registers) 6249 6250Please note that the kernel is allowed to use the kvm_run structure as the 6251primary storage for certain register types. Therefore, the kernel may use the 6252values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 6253 6254:: 6255 6256 }; 6257 6258 6259 62606. Capabilities that can be enabled on vCPUs 6261============================================ 6262 6263There are certain capabilities that change the behavior of the virtual CPU or 6264the virtual machine when enabled. To enable them, please see section 4.37. 6265Below you can find a list of capabilities and what their effect on the vCPU or 6266the virtual machine is when enabling them. 6267 6268The following information is provided along with the description: 6269 6270 Architectures: 6271 which instruction set architectures provide this ioctl. 6272 x86 includes both i386 and x86_64. 6273 6274 Target: 6275 whether this is a per-vcpu or per-vm capability. 6276 6277 Parameters: 6278 what parameters are accepted by the capability. 6279 6280 Returns: 6281 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 6282 are not detailed, but errors with specific meanings are. 6283 6284 62856.1 KVM_CAP_PPC_OSI 6286------------------- 6287 6288:Architectures: ppc 6289:Target: vcpu 6290:Parameters: none 6291:Returns: 0 on success; -1 on error 6292 6293This capability enables interception of OSI hypercalls that otherwise would 6294be treated as normal system calls to be injected into the guest. OSI hypercalls 6295were invented by Mac-on-Linux to have a standardized communication mechanism 6296between the guest and the host. 6297 6298When this capability is enabled, KVM_EXIT_OSI can occur. 6299 6300 63016.2 KVM_CAP_PPC_PAPR 6302-------------------- 6303 6304:Architectures: ppc 6305:Target: vcpu 6306:Parameters: none 6307:Returns: 0 on success; -1 on error 6308 6309This capability enables interception of PAPR hypercalls. PAPR hypercalls are 6310done using the hypercall instruction "sc 1". 6311 6312It also sets the guest privilege level to "supervisor" mode. Usually the guest 6313runs in "hypervisor" privilege mode with a few missing features. 6314 6315In addition to the above, it changes the semantics of SDR1. In this mode, the 6316HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 6317HTAB invisible to the guest. 6318 6319When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 6320 6321 63226.3 KVM_CAP_SW_TLB 6323------------------ 6324 6325:Architectures: ppc 6326:Target: vcpu 6327:Parameters: args[0] is the address of a struct kvm_config_tlb 6328:Returns: 0 on success; -1 on error 6329 6330:: 6331 6332 struct kvm_config_tlb { 6333 __u64 params; 6334 __u64 array; 6335 __u32 mmu_type; 6336 __u32 array_len; 6337 }; 6338 6339Configures the virtual CPU's TLB array, establishing a shared memory area 6340between userspace and KVM. The "params" and "array" fields are userspace 6341addresses of mmu-type-specific data structures. The "array_len" field is an 6342safety mechanism, and should be set to the size in bytes of the memory that 6343userspace has reserved for the array. It must be at least the size dictated 6344by "mmu_type" and "params". 6345 6346While KVM_RUN is active, the shared region is under control of KVM. Its 6347contents are undefined, and any modification by userspace results in 6348boundedly undefined behavior. 6349 6350On return from KVM_RUN, the shared region will reflect the current state of 6351the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 6352to tell KVM which entries have been changed, prior to calling KVM_RUN again 6353on this vcpu. 6354 6355For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 6356 6357 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 6358 - The "array" field points to an array of type "struct 6359 kvm_book3e_206_tlb_entry". 6360 - The array consists of all entries in the first TLB, followed by all 6361 entries in the second TLB. 6362 - Within a TLB, entries are ordered first by increasing set number. Within a 6363 set, entries are ordered by way (increasing ESEL). 6364 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 6365 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 6366 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 6367 hardware ignores this value for TLB0. 6368 63696.4 KVM_CAP_S390_CSS_SUPPORT 6370---------------------------- 6371 6372:Architectures: s390 6373:Target: vcpu 6374:Parameters: none 6375:Returns: 0 on success; -1 on error 6376 6377This capability enables support for handling of channel I/O instructions. 6378 6379TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 6380handled in-kernel, while the other I/O instructions are passed to userspace. 6381 6382When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 6383SUBCHANNEL intercepts. 6384 6385Note that even though this capability is enabled per-vcpu, the complete 6386virtual machine is affected. 6387 63886.5 KVM_CAP_PPC_EPR 6389------------------- 6390 6391:Architectures: ppc 6392:Target: vcpu 6393:Parameters: args[0] defines whether the proxy facility is active 6394:Returns: 0 on success; -1 on error 6395 6396This capability enables or disables the delivery of interrupts through the 6397external proxy facility. 6398 6399When enabled (args[0] != 0), every time the guest gets an external interrupt 6400delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 6401to receive the topmost interrupt vector. 6402 6403When disabled (args[0] == 0), behavior is as if this facility is unsupported. 6404 6405When this capability is enabled, KVM_EXIT_EPR can occur. 6406 64076.6 KVM_CAP_IRQ_MPIC 6408-------------------- 6409 6410:Architectures: ppc 6411:Parameters: args[0] is the MPIC device fd; 6412 args[1] is the MPIC CPU number for this vcpu 6413 6414This capability connects the vcpu to an in-kernel MPIC device. 6415 64166.7 KVM_CAP_IRQ_XICS 6417-------------------- 6418 6419:Architectures: ppc 6420:Target: vcpu 6421:Parameters: args[0] is the XICS device fd; 6422 args[1] is the XICS CPU number (server ID) for this vcpu 6423 6424This capability connects the vcpu to an in-kernel XICS device. 6425 64266.8 KVM_CAP_S390_IRQCHIP 6427------------------------ 6428 6429:Architectures: s390 6430:Target: vm 6431:Parameters: none 6432 6433This capability enables the in-kernel irqchip for s390. Please refer to 6434"4.24 KVM_CREATE_IRQCHIP" for details. 6435 64366.9 KVM_CAP_MIPS_FPU 6437-------------------- 6438 6439:Architectures: mips 6440:Target: vcpu 6441:Parameters: args[0] is reserved for future use (should be 0). 6442 6443This capability allows the use of the host Floating Point Unit by the guest. It 6444allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 6445done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be 6446accessed (depending on the current guest FPU register mode), and the Status.FR, 6447Config5.FRE bits are accessible via the KVM API and also from the guest, 6448depending on them being supported by the FPU. 6449 64506.10 KVM_CAP_MIPS_MSA 6451--------------------- 6452 6453:Architectures: mips 6454:Target: vcpu 6455:Parameters: args[0] is reserved for future use (should be 0). 6456 6457This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 6458It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 6459Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*`` 6460registers can be accessed, and the Config5.MSAEn bit is accessible via the 6461KVM API and also from the guest. 6462 64636.74 KVM_CAP_SYNC_REGS 6464---------------------- 6465 6466:Architectures: s390, x86 6467:Target: s390: always enabled, x86: vcpu 6468:Parameters: none 6469:Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register 6470 sets are supported 6471 (bitfields defined in arch/x86/include/uapi/asm/kvm.h). 6472 6473As described above in the kvm_sync_regs struct info in section 5 (kvm_run): 6474KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers 6475without having to call SET/GET_*REGS". This reduces overhead by eliminating 6476repeated ioctl calls for setting and/or getting register values. This is 6477particularly important when userspace is making synchronous guest state 6478modifications, e.g. when emulating and/or intercepting instructions in 6479userspace. 6480 6481For s390 specifics, please refer to the source code. 6482 6483For x86: 6484 6485- the register sets to be copied out to kvm_run are selectable 6486 by userspace (rather that all sets being copied out for every exit). 6487- vcpu_events are available in addition to regs and sregs. 6488 6489For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to 6490function as an input bit-array field set by userspace to indicate the 6491specific register sets to be copied out on the next exit. 6492 6493To indicate when userspace has modified values that should be copied into 6494the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set. 6495This is done using the same bitflags as for the 'kvm_valid_regs' field. 6496If the dirty bit is not set, then the register set values will not be copied 6497into the vCPU even if they've been modified. 6498 6499Unused bitfields in the bitarrays must be set to zero. 6500 6501:: 6502 6503 struct kvm_sync_regs { 6504 struct kvm_regs regs; 6505 struct kvm_sregs sregs; 6506 struct kvm_vcpu_events events; 6507 }; 6508 65096.75 KVM_CAP_PPC_IRQ_XIVE 6510------------------------- 6511 6512:Architectures: ppc 6513:Target: vcpu 6514:Parameters: args[0] is the XIVE device fd; 6515 args[1] is the XIVE CPU number (server ID) for this vcpu 6516 6517This capability connects the vcpu to an in-kernel XIVE device. 6518 65197. Capabilities that can be enabled on VMs 6520========================================== 6521 6522There are certain capabilities that change the behavior of the virtual 6523machine when enabled. To enable them, please see section 4.37. Below 6524you can find a list of capabilities and what their effect on the VM 6525is when enabling them. 6526 6527The following information is provided along with the description: 6528 6529 Architectures: 6530 which instruction set architectures provide this ioctl. 6531 x86 includes both i386 and x86_64. 6532 6533 Parameters: 6534 what parameters are accepted by the capability. 6535 6536 Returns: 6537 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 6538 are not detailed, but errors with specific meanings are. 6539 6540 65417.1 KVM_CAP_PPC_ENABLE_HCALL 6542---------------------------- 6543 6544:Architectures: ppc 6545:Parameters: args[0] is the sPAPR hcall number; 6546 args[1] is 0 to disable, 1 to enable in-kernel handling 6547 6548This capability controls whether individual sPAPR hypercalls (hcalls) 6549get handled by the kernel or not. Enabling or disabling in-kernel 6550handling of an hcall is effective across the VM. On creation, an 6551initial set of hcalls are enabled for in-kernel handling, which 6552consists of those hcalls for which in-kernel handlers were implemented 6553before this capability was implemented. If disabled, the kernel will 6554not to attempt to handle the hcall, but will always exit to userspace 6555to handle it. Note that it may not make sense to enable some and 6556disable others of a group of related hcalls, but KVM does not prevent 6557userspace from doing that. 6558 6559If the hcall number specified is not one that has an in-kernel 6560implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 6561error. 6562 65637.2 KVM_CAP_S390_USER_SIGP 6564-------------------------- 6565 6566:Architectures: s390 6567:Parameters: none 6568 6569This capability controls which SIGP orders will be handled completely in user 6570space. With this capability enabled, all fast orders will be handled completely 6571in the kernel: 6572 6573- SENSE 6574- SENSE RUNNING 6575- EXTERNAL CALL 6576- EMERGENCY SIGNAL 6577- CONDITIONAL EMERGENCY SIGNAL 6578 6579All other orders will be handled completely in user space. 6580 6581Only privileged operation exceptions will be checked for in the kernel (or even 6582in the hardware prior to interception). If this capability is not enabled, the 6583old way of handling SIGP orders is used (partially in kernel and user space). 6584 65857.3 KVM_CAP_S390_VECTOR_REGISTERS 6586--------------------------------- 6587 6588:Architectures: s390 6589:Parameters: none 6590:Returns: 0 on success, negative value on error 6591 6592Allows use of the vector registers introduced with z13 processor, and 6593provides for the synchronization between host and user space. Will 6594return -EINVAL if the machine does not support vectors. 6595 65967.4 KVM_CAP_S390_USER_STSI 6597-------------------------- 6598 6599:Architectures: s390 6600:Parameters: none 6601 6602This capability allows post-handlers for the STSI instruction. After 6603initial handling in the kernel, KVM exits to user space with 6604KVM_EXIT_S390_STSI to allow user space to insert further data. 6605 6606Before exiting to userspace, kvm handlers should fill in s390_stsi field of 6607vcpu->run:: 6608 6609 struct { 6610 __u64 addr; 6611 __u8 ar; 6612 __u8 reserved; 6613 __u8 fc; 6614 __u8 sel1; 6615 __u16 sel2; 6616 } s390_stsi; 6617 6618 @addr - guest address of STSI SYSIB 6619 @fc - function code 6620 @sel1 - selector 1 6621 @sel2 - selector 2 6622 @ar - access register number 6623 6624KVM handlers should exit to userspace with rc = -EREMOTE. 6625 66267.5 KVM_CAP_SPLIT_IRQCHIP 6627------------------------- 6628 6629:Architectures: x86 6630:Parameters: args[0] - number of routes reserved for userspace IOAPICs 6631:Returns: 0 on success, -1 on error 6632 6633Create a local apic for each processor in the kernel. This can be used 6634instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 6635IOAPIC and PIC (and also the PIT, even though this has to be enabled 6636separately). 6637 6638This capability also enables in kernel routing of interrupt requests; 6639when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 6640used in the IRQ routing table. The first args[0] MSI routes are reserved 6641for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 6642a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 6643 6644Fails if VCPU has already been created, or if the irqchip is already in the 6645kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 6646 66477.6 KVM_CAP_S390_RI 6648------------------- 6649 6650:Architectures: s390 6651:Parameters: none 6652 6653Allows use of runtime-instrumentation introduced with zEC12 processor. 6654Will return -EINVAL if the machine does not support runtime-instrumentation. 6655Will return -EBUSY if a VCPU has already been created. 6656 66577.7 KVM_CAP_X2APIC_API 6658---------------------- 6659 6660:Architectures: x86 6661:Parameters: args[0] - features that should be enabled 6662:Returns: 0 on success, -EINVAL when args[0] contains invalid features 6663 6664Valid feature flags in args[0] are:: 6665 6666 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 6667 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 6668 6669Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 6670KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 6671allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 6672respective sections. 6673 6674KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 6675in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 6676as a broadcast even in x2APIC mode in order to support physical x2APIC 6677without interrupt remapping. This is undesirable in logical mode, 6678where 0xff represents CPUs 0-7 in cluster 0. 6679 66807.8 KVM_CAP_S390_USER_INSTR0 6681---------------------------- 6682 6683:Architectures: s390 6684:Parameters: none 6685 6686With this capability enabled, all illegal instructions 0x0000 (2 bytes) will 6687be intercepted and forwarded to user space. User space can use this 6688mechanism e.g. to realize 2-byte software breakpoints. The kernel will 6689not inject an operating exception for these instructions, user space has 6690to take care of that. 6691 6692This capability can be enabled dynamically even if VCPUs were already 6693created and are running. 6694 66957.9 KVM_CAP_S390_GS 6696------------------- 6697 6698:Architectures: s390 6699:Parameters: none 6700:Returns: 0 on success; -EINVAL if the machine does not support 6701 guarded storage; -EBUSY if a VCPU has already been created. 6702 6703Allows use of guarded storage for the KVM guest. 6704 67057.10 KVM_CAP_S390_AIS 6706--------------------- 6707 6708:Architectures: s390 6709:Parameters: none 6710 6711Allow use of adapter-interruption suppression. 6712:Returns: 0 on success; -EBUSY if a VCPU has already been created. 6713 67147.11 KVM_CAP_PPC_SMT 6715-------------------- 6716 6717:Architectures: ppc 6718:Parameters: vsmt_mode, flags 6719 6720Enabling this capability on a VM provides userspace with a way to set 6721the desired virtual SMT mode (i.e. the number of virtual CPUs per 6722virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 6723between 1 and 8. On POWER8, vsmt_mode must also be no greater than 6724the number of threads per subcore for the host. Currently flags must 6725be 0. A successful call to enable this capability will result in 6726vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 6727subsequently queried for the VM. This capability is only supported by 6728HV KVM, and can only be set before any VCPUs have been created. 6729The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 6730modes are available. 6731 67327.12 KVM_CAP_PPC_FWNMI 6733---------------------- 6734 6735:Architectures: ppc 6736:Parameters: none 6737 6738With this capability a machine check exception in the guest address 6739space will cause KVM to exit the guest with NMI exit reason. This 6740enables QEMU to build error log and branch to guest kernel registered 6741machine check handling routine. Without this capability KVM will 6742branch to guests' 0x200 interrupt vector. 6743 67447.13 KVM_CAP_X86_DISABLE_EXITS 6745------------------------------ 6746 6747:Architectures: x86 6748:Parameters: args[0] defines which exits are disabled 6749:Returns: 0 on success, -EINVAL when args[0] contains invalid exits 6750 6751Valid bits in args[0] are:: 6752 6753 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) 6754 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1) 6755 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) 6756 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3) 6757 6758Enabling this capability on a VM provides userspace with a way to no 6759longer intercept some instructions for improved latency in some 6760workloads, and is suggested when vCPUs are associated to dedicated 6761physical CPUs. More bits can be added in the future; userspace can 6762just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable 6763all such vmexits. 6764 6765Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits. 6766 67677.14 KVM_CAP_S390_HPAGE_1M 6768-------------------------- 6769 6770:Architectures: s390 6771:Parameters: none 6772:Returns: 0 on success, -EINVAL if hpage module parameter was not set 6773 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 6774 flag set 6775 6776With this capability the KVM support for memory backing with 1m pages 6777through hugetlbfs can be enabled for a VM. After the capability is 6778enabled, cmma can't be enabled anymore and pfmfi and the storage key 6779interpretation are disabled. If cmma has already been enabled or the 6780hpage module parameter is not set to 1, -EINVAL is returned. 6781 6782While it is generally possible to create a huge page backed VM without 6783this capability, the VM will not be able to run. 6784 67857.15 KVM_CAP_MSR_PLATFORM_INFO 6786------------------------------ 6787 6788:Architectures: x86 6789:Parameters: args[0] whether feature should be enabled or not 6790 6791With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise, 6792a #GP would be raised when the guest tries to access. Currently, this 6793capability does not enable write permissions of this MSR for the guest. 6794 67957.16 KVM_CAP_PPC_NESTED_HV 6796-------------------------- 6797 6798:Architectures: ppc 6799:Parameters: none 6800:Returns: 0 on success, -EINVAL when the implementation doesn't support 6801 nested-HV virtualization. 6802 6803HV-KVM on POWER9 and later systems allows for "nested-HV" 6804virtualization, which provides a way for a guest VM to run guests that 6805can run using the CPU's supervisor mode (privileged non-hypervisor 6806state). Enabling this capability on a VM depends on the CPU having 6807the necessary functionality and on the facility being enabled with a 6808kvm-hv module parameter. 6809 68107.17 KVM_CAP_EXCEPTION_PAYLOAD 6811------------------------------ 6812 6813:Architectures: x86 6814:Parameters: args[0] whether feature should be enabled or not 6815 6816With this capability enabled, CR2 will not be modified prior to the 6817emulated VM-exit when L1 intercepts a #PF exception that occurs in 6818L2. Similarly, for kvm-intel only, DR6 will not be modified prior to 6819the emulated VM-exit when L1 intercepts a #DB exception that occurs in 6820L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or 6821#DB) exception for L2, exception.has_payload will be set and the 6822faulting address (or the new DR6 bits*) will be reported in the 6823exception_payload field. Similarly, when userspace injects a #PF (or 6824#DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set 6825exception.has_payload and to put the faulting address - or the new DR6 6826bits\ [#]_ - in the exception_payload field. 6827 6828This capability also enables exception.pending in struct 6829kvm_vcpu_events, which allows userspace to distinguish between pending 6830and injected exceptions. 6831 6832 6833.. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception 6834 will clear DR6.RTM. 6835 68367.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 6837 6838:Architectures: x86, arm64, mips 6839:Parameters: args[0] whether feature should be enabled or not 6840 6841Valid flags are:: 6842 6843 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0) 6844 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1) 6845 6846With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not 6847automatically clear and write-protect all pages that are returned as dirty. 6848Rather, userspace will have to do this operation separately using 6849KVM_CLEAR_DIRTY_LOG. 6850 6851At the cost of a slightly more complicated operation, this provides better 6852scalability and responsiveness for two reasons. First, 6853KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather 6854than requiring to sync a full memslot; this ensures that KVM does not 6855take spinlocks for an extended period of time. Second, in some cases a 6856large amount of time can pass between a call to KVM_GET_DIRTY_LOG and 6857userspace actually using the data in the page. Pages can be modified 6858during this time, which is inefficient for both the guest and userspace: 6859the guest will incur a higher penalty due to write protection faults, 6860while userspace can see false reports of dirty pages. Manual reprotection 6861helps reducing this time, improving guest performance and reducing the 6862number of dirty log false positives. 6863 6864With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap 6865will be initialized to 1 when created. This also improves performance because 6866dirty logging can be enabled gradually in small chunks on the first call 6867to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on 6868KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on 6869x86 and arm64 for now). 6870 6871KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name 6872KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make 6873it hard or impossible to use it correctly. The availability of 6874KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed. 6875Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT. 6876 68777.19 KVM_CAP_PPC_SECURE_GUEST 6878------------------------------ 6879 6880:Architectures: ppc 6881 6882This capability indicates that KVM is running on a host that has 6883ultravisor firmware and thus can support a secure guest. On such a 6884system, a guest can ask the ultravisor to make it a secure guest, 6885one whose memory is inaccessible to the host except for pages which 6886are explicitly requested to be shared with the host. The ultravisor 6887notifies KVM when a guest requests to become a secure guest, and KVM 6888has the opportunity to veto the transition. 6889 6890If present, this capability can be enabled for a VM, meaning that KVM 6891will allow the transition to secure guest mode. Otherwise KVM will 6892veto the transition. 6893 68947.20 KVM_CAP_HALT_POLL 6895---------------------- 6896 6897:Architectures: all 6898:Target: VM 6899:Parameters: args[0] is the maximum poll time in nanoseconds 6900:Returns: 0 on success; -1 on error 6901 6902This capability overrides the kvm module parameter halt_poll_ns for the 6903target VM. 6904 6905VCPU polling allows a VCPU to poll for wakeup events instead of immediately 6906scheduling during guest halts. The maximum time a VCPU can spend polling is 6907controlled by the kvm module parameter halt_poll_ns. This capability allows 6908the maximum halt time to specified on a per-VM basis, effectively overriding 6909the module parameter for the target VM. 6910 69117.21 KVM_CAP_X86_USER_SPACE_MSR 6912------------------------------- 6913 6914:Architectures: x86 6915:Target: VM 6916:Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report 6917:Returns: 0 on success; -1 on error 6918 6919This capability enables trapping of #GP invoking RDMSR and WRMSR instructions 6920into user space. 6921 6922When a guest requests to read or write an MSR, KVM may not implement all MSRs 6923that are relevant to a respective system. It also does not differentiate by 6924CPU type. 6925 6926To allow more fine grained control over MSR handling, user space may enable 6927this capability. With it enabled, MSR accesses that match the mask specified in 6928args[0] and trigger a #GP event inside the guest by KVM will instead trigger 6929KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications which user space 6930can then handle to implement model specific MSR handling and/or user notifications 6931to inform a user that an MSR was not handled. 6932 69337.22 KVM_CAP_X86_BUS_LOCK_EXIT 6934------------------------------- 6935 6936:Architectures: x86 6937:Target: VM 6938:Parameters: args[0] defines the policy used when bus locks detected in guest 6939:Returns: 0 on success, -EINVAL when args[0] contains invalid bits 6940 6941Valid bits in args[0] are:: 6942 6943 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0) 6944 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1) 6945 6946Enabling this capability on a VM provides userspace with a way to select 6947a policy to handle the bus locks detected in guest. Userspace can obtain 6948the supported modes from the result of KVM_CHECK_EXTENSION and define it 6949through the KVM_ENABLE_CAP. 6950 6951KVM_BUS_LOCK_DETECTION_OFF and KVM_BUS_LOCK_DETECTION_EXIT are supported 6952currently and mutually exclusive with each other. More bits can be added in 6953the future. 6954 6955With KVM_BUS_LOCK_DETECTION_OFF set, bus locks in guest will not cause vm exits 6956so that no additional actions are needed. This is the default mode. 6957 6958With KVM_BUS_LOCK_DETECTION_EXIT set, vm exits happen when bus lock detected 6959in VM. KVM just exits to userspace when handling them. Userspace can enforce 6960its own throttling or other policy based mitigations. 6961 6962This capability is aimed to address the thread that VM can exploit bus locks to 6963degree the performance of the whole system. Once the userspace enable this 6964capability and select the KVM_BUS_LOCK_DETECTION_EXIT mode, KVM will set the 6965KVM_RUN_BUS_LOCK flag in vcpu-run->flags field and exit to userspace. Concerning 6966the bus lock vm exit can be preempted by a higher priority VM exit, the exit 6967notifications to userspace can be KVM_EXIT_BUS_LOCK or other reasons. 6968KVM_RUN_BUS_LOCK flag is used to distinguish between them. 6969 69707.23 KVM_CAP_PPC_DAWR1 6971---------------------- 6972 6973:Architectures: ppc 6974:Parameters: none 6975:Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR 6976 6977This capability can be used to check / enable 2nd DAWR feature provided 6978by POWER10 processor. 6979 6980 69817.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM 6982------------------------------------- 6983 6984Architectures: x86 SEV enabled 6985Type: vm 6986Parameters: args[0] is the fd of the source vm 6987Returns: 0 on success; ENOTTY on error 6988 6989This capability enables userspace to copy encryption context from the vm 6990indicated by the fd to the vm this is called on. 6991 6992This is intended to support in-guest workloads scheduled by the host. This 6993allows the in-guest workload to maintain its own NPTs and keeps the two vms 6994from accidentally clobbering each other with interrupts and the like (separate 6995APIC/MSRs/etc). 6996 69977.25 KVM_CAP_SGX_ATTRIBUTE 6998-------------------------- 6999 7000:Architectures: x86 7001:Target: VM 7002:Parameters: args[0] is a file handle of a SGX attribute file in securityfs 7003:Returns: 0 on success, -EINVAL if the file handle is invalid or if a requested 7004 attribute is not supported by KVM. 7005 7006KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM to grant a VM access to one or 7007more priveleged enclave attributes. args[0] must hold a file handle to a valid 7008SGX attribute file corresponding to an attribute that is supported/restricted 7009by KVM (currently only PROVISIONKEY). 7010 7011The SGX subsystem restricts access to a subset of enclave attributes to provide 7012additional security for an uncompromised kernel, e.g. use of the PROVISIONKEY 7013is restricted to deter malware from using the PROVISIONKEY to obtain a stable 7014system fingerprint. To prevent userspace from circumventing such restrictions 7015by running an enclave in a VM, KVM prevents access to privileged attributes by 7016default. 7017 7018See Documentation/x86/sgx.rst for more details. 7019 70207.26 KVM_CAP_PPC_RPT_INVALIDATE 7021------------------------------- 7022 7023:Capability: KVM_CAP_PPC_RPT_INVALIDATE 7024:Architectures: ppc 7025:Type: vm 7026 7027This capability indicates that the kernel is capable of handling 7028H_RPT_INVALIDATE hcall. 7029 7030In order to enable the use of H_RPT_INVALIDATE in the guest, 7031user space might have to advertise it for the guest. For example, 7032IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is 7033present in the "ibm,hypertas-functions" device-tree property. 7034 7035This capability is enabled for hypervisors on platforms like POWER9 7036that support radix MMU. 7037 70387.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE 7039-------------------------------------- 7040 7041:Architectures: x86 7042:Parameters: args[0] whether the feature should be enabled or not 7043 7044When this capability is enabled, an emulation failure will result in an exit 7045to userspace with KVM_INTERNAL_ERROR (except when the emulator was invoked 7046to handle a VMware backdoor instruction). Furthermore, KVM will now provide up 7047to 15 instruction bytes for any exit to userspace resulting from an emulation 7048failure. When these exits to userspace occur use the emulation_failure struct 7049instead of the internal struct. They both have the same layout, but the 7050emulation_failure struct matches the content better. It also explicitly 7051defines the 'flags' field which is used to describe the fields in the struct 7052that are valid (ie: if KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES is 7053set in the 'flags' field then both 'insn_size' and 'insn_bytes' have valid data 7054in them.) 7055 70567.28 KVM_CAP_ARM_MTE 7057-------------------- 7058 7059:Architectures: arm64 7060:Parameters: none 7061 7062This capability indicates that KVM (and the hardware) supports exposing the 7063Memory Tagging Extensions (MTE) to the guest. It must also be enabled by the 7064VMM before creating any VCPUs to allow the guest access. Note that MTE is only 7065available to a guest running in AArch64 mode and enabling this capability will 7066cause attempts to create AArch32 VCPUs to fail. 7067 7068When enabled the guest is able to access tags associated with any memory given 7069to the guest. KVM will ensure that the tags are maintained during swap or 7070hibernation of the host; however the VMM needs to manually save/restore the 7071tags as appropriate if the VM is migrated. 7072 7073When this capability is enabled all memory in memslots must be mapped as 7074not-shareable (no MAP_SHARED), attempts to create a memslot with a 7075MAP_SHARED mmap will result in an -EINVAL return. 7076 7077When enabled the VMM may make use of the ``KVM_ARM_MTE_COPY_TAGS`` ioctl to 7078perform a bulk copy of tags to/from the guest. 7079 70807.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 7081------------------------------------- 7082 7083Architectures: x86 SEV enabled 7084Type: vm 7085Parameters: args[0] is the fd of the source vm 7086Returns: 0 on success 7087 7088This capability enables userspace to migrate the encryption context from the VM 7089indicated by the fd to the VM this is called on. 7090 7091This is intended to support intra-host migration of VMs between userspace VMMs, 7092upgrading the VMM process without interrupting the guest. 7093 70947.30 KVM_CAP_PPC_AIL_MODE_3 7095------------------------------- 7096 7097:Capability: KVM_CAP_PPC_AIL_MODE_3 7098:Architectures: ppc 7099:Type: vm 7100 7101This capability indicates that the kernel supports the mode 3 setting for the 7102"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location" 7103resource that is controlled with the H_SET_MODE hypercall. 7104 7105This capability allows a guest kernel to use a better-performance mode for 7106handling interrupts and system calls. 7107 71087.31 KVM_CAP_DISABLE_QUIRKS2 7109---------------------------- 7110 7111:Capability: KVM_CAP_DISABLE_QUIRKS2 7112:Parameters: args[0] - set of KVM quirks to disable 7113:Architectures: x86 7114:Type: vm 7115 7116This capability, if enabled, will cause KVM to disable some behavior 7117quirks. 7118 7119Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 7120quirks that can be disabled in KVM. 7121 7122The argument to KVM_ENABLE_CAP for this capability is a bitmask of 7123quirks to disable, and must be a subset of the bitmask returned by 7124KVM_CHECK_EXTENSION. 7125 7126The valid bits in cap.args[0] are: 7127 7128=================================== ============================================ 7129 KVM_X86_QUIRK_LINT0_REENABLED By default, the reset value for the LVT 7130 LINT0 register is 0x700 (APIC_MODE_EXTINT). 7131 When this quirk is disabled, the reset value 7132 is 0x10000 (APIC_LVT_MASKED). 7133 7134 KVM_X86_QUIRK_CD_NW_CLEARED By default, KVM clears CR0.CD and CR0.NW. 7135 When this quirk is disabled, KVM does not 7136 change the value of CR0.CD and CR0.NW. 7137 7138 KVM_X86_QUIRK_LAPIC_MMIO_HOLE By default, the MMIO LAPIC interface is 7139 available even when configured for x2APIC 7140 mode. When this quirk is disabled, KVM 7141 disables the MMIO LAPIC interface if the 7142 LAPIC is in x2APIC mode. 7143 7144 KVM_X86_QUIRK_OUT_7E_INC_RIP By default, KVM pre-increments %rip before 7145 exiting to userspace for an OUT instruction 7146 to port 0x7e. When this quirk is disabled, 7147 KVM does not pre-increment %rip before 7148 exiting to userspace. 7149 7150 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this quirk is disabled, KVM sets 7151 CPUID.01H:ECX[bit 3] (MONITOR/MWAIT) if 7152 IA32_MISC_ENABLE[bit 18] (MWAIT) is set. 7153 Additionally, when this quirk is disabled, 7154 KVM clears CPUID.01H:ECX[bit 3] if 7155 IA32_MISC_ENABLE[bit 18] is cleared. 7156=================================== ============================================ 7157 71588. Other capabilities. 7159====================== 7160 7161This section lists capabilities that give information about other 7162features of the KVM implementation. 7163 71648.1 KVM_CAP_PPC_HWRNG 7165--------------------- 7166 7167:Architectures: ppc 7168 7169This capability, if KVM_CHECK_EXTENSION indicates that it is 7170available, means that the kernel has an implementation of the 7171H_RANDOM hypercall backed by a hardware random-number generator. 7172If present, the kernel H_RANDOM handler can be enabled for guest use 7173with the KVM_CAP_PPC_ENABLE_HCALL capability. 7174 71758.2 KVM_CAP_HYPERV_SYNIC 7176------------------------ 7177 7178:Architectures: x86 7179 7180This capability, if KVM_CHECK_EXTENSION indicates that it is 7181available, means that the kernel has an implementation of the 7182Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 7183used to support Windows Hyper-V based guest paravirt drivers(VMBus). 7184 7185In order to use SynIC, it has to be activated by setting this 7186capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 7187will disable the use of APIC hardware virtualization even if supported 7188by the CPU, as it's incompatible with SynIC auto-EOI behavior. 7189 71908.3 KVM_CAP_PPC_RADIX_MMU 7191------------------------- 7192 7193:Architectures: ppc 7194 7195This capability, if KVM_CHECK_EXTENSION indicates that it is 7196available, means that the kernel can support guests using the 7197radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 7198processor). 7199 72008.4 KVM_CAP_PPC_HASH_MMU_V3 7201--------------------------- 7202 7203:Architectures: ppc 7204 7205This capability, if KVM_CHECK_EXTENSION indicates that it is 7206available, means that the kernel can support guests using the 7207hashed page table MMU defined in Power ISA V3.00 (as implemented in 7208the POWER9 processor), including in-memory segment tables. 7209 72108.5 KVM_CAP_MIPS_VZ 7211------------------- 7212 7213:Architectures: mips 7214 7215This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 7216it is available, means that full hardware assisted virtualization capabilities 7217of the hardware are available for use through KVM. An appropriate 7218KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 7219utilises it. 7220 7221If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 7222available, it means that the VM is using full hardware assisted virtualization 7223capabilities of the hardware. This is useful to check after creating a VM with 7224KVM_VM_MIPS_DEFAULT. 7225 7226The value returned by KVM_CHECK_EXTENSION should be compared against known 7227values (see below). All other values are reserved. This is to allow for the 7228possibility of other hardware assisted virtualization implementations which 7229may be incompatible with the MIPS VZ ASE. 7230 7231== ========================================================================== 7232 0 The trap & emulate implementation is in use to run guest code in user 7233 mode. Guest virtual memory segments are rearranged to fit the guest in the 7234 user mode address space. 7235 7236 1 The MIPS VZ ASE is in use, providing full hardware assisted 7237 virtualization, including standard guest virtual memory segments. 7238== ========================================================================== 7239 72408.6 KVM_CAP_MIPS_TE 7241------------------- 7242 7243:Architectures: mips 7244 7245This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 7246it is available, means that the trap & emulate implementation is available to 7247run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware 7248assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed 7249to KVM_CREATE_VM to create a VM which utilises it. 7250 7251If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 7252available, it means that the VM is using trap & emulate. 7253 72548.7 KVM_CAP_MIPS_64BIT 7255---------------------- 7256 7257:Architectures: mips 7258 7259This capability indicates the supported architecture type of the guest, i.e. the 7260supported register and address width. 7261 7262The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 7263kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 7264be checked specifically against known values (see below). All other values are 7265reserved. 7266 7267== ======================================================================== 7268 0 MIPS32 or microMIPS32. 7269 Both registers and addresses are 32-bits wide. 7270 It will only be possible to run 32-bit guest code. 7271 7272 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 7273 Registers are 64-bits wide, but addresses are 32-bits wide. 7274 64-bit guest code may run but cannot access MIPS64 memory segments. 7275 It will also be possible to run 32-bit guest code. 7276 7277 2 MIPS64 or microMIPS64 with access to all address segments. 7278 Both registers and addresses are 64-bits wide. 7279 It will be possible to run 64-bit or 32-bit guest code. 7280== ======================================================================== 7281 72828.9 KVM_CAP_ARM_USER_IRQ 7283------------------------ 7284 7285:Architectures: arm64 7286 7287This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 7288that if userspace creates a VM without an in-kernel interrupt controller, it 7289will be notified of changes to the output level of in-kernel emulated devices, 7290which can generate virtual interrupts, presented to the VM. 7291For such VMs, on every return to userspace, the kernel 7292updates the vcpu's run->s.regs.device_irq_level field to represent the actual 7293output level of the device. 7294 7295Whenever kvm detects a change in the device output level, kvm guarantees at 7296least one return to userspace before running the VM. This exit could either 7297be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 7298userspace can always sample the device output level and re-compute the state of 7299the userspace interrupt controller. Userspace should always check the state 7300of run->s.regs.device_irq_level on every kvm exit. 7301The value in run->s.regs.device_irq_level can represent both level and edge 7302triggered interrupt signals, depending on the device. Edge triggered interrupt 7303signals will exit to userspace with the bit in run->s.regs.device_irq_level 7304set exactly once per edge signal. 7305 7306The field run->s.regs.device_irq_level is available independent of 7307run->kvm_valid_regs or run->kvm_dirty_regs bits. 7308 7309If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 7310number larger than 0 indicating the version of this capability is implemented 7311and thereby which bits in run->s.regs.device_irq_level can signal values. 7312 7313Currently the following bits are defined for the device_irq_level bitmap:: 7314 7315 KVM_CAP_ARM_USER_IRQ >= 1: 7316 7317 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 7318 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 7319 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 7320 7321Future versions of kvm may implement additional events. These will get 7322indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 7323listed above. 7324 73258.10 KVM_CAP_PPC_SMT_POSSIBLE 7326----------------------------- 7327 7328:Architectures: ppc 7329 7330Querying this capability returns a bitmap indicating the possible 7331virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 7332(counting from the right) is set, then a virtual SMT mode of 2^N is 7333available. 7334 73358.11 KVM_CAP_HYPERV_SYNIC2 7336-------------------------- 7337 7338:Architectures: x86 7339 7340This capability enables a newer version of Hyper-V Synthetic interrupt 7341controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 7342doesn't clear SynIC message and event flags pages when they are enabled by 7343writing to the respective MSRs. 7344 73458.12 KVM_CAP_HYPERV_VP_INDEX 7346---------------------------- 7347 7348:Architectures: x86 7349 7350This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 7351value is used to denote the target vcpu for a SynIC interrupt. For 7352compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this 7353capability is absent, userspace can still query this msr's value. 7354 73558.13 KVM_CAP_S390_AIS_MIGRATION 7356------------------------------- 7357 7358:Architectures: s390 7359:Parameters: none 7360 7361This capability indicates if the flic device will be able to get/set the 7362AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 7363to discover this without having to create a flic device. 7364 73658.14 KVM_CAP_S390_PSW 7366--------------------- 7367 7368:Architectures: s390 7369 7370This capability indicates that the PSW is exposed via the kvm_run structure. 7371 73728.15 KVM_CAP_S390_GMAP 7373---------------------- 7374 7375:Architectures: s390 7376 7377This capability indicates that the user space memory used as guest mapping can 7378be anywhere in the user memory address space, as long as the memory slots are 7379aligned and sized to a segment (1MB) boundary. 7380 73818.16 KVM_CAP_S390_COW 7382--------------------- 7383 7384:Architectures: s390 7385 7386This capability indicates that the user space memory used as guest mapping can 7387use copy-on-write semantics as well as dirty pages tracking via read-only page 7388tables. 7389 73908.17 KVM_CAP_S390_BPB 7391--------------------- 7392 7393:Architectures: s390 7394 7395This capability indicates that kvm will implement the interfaces to handle 7396reset, migration and nested KVM for branch prediction blocking. The stfle 7397facility 82 should not be provided to the guest without this capability. 7398 73998.18 KVM_CAP_HYPERV_TLBFLUSH 7400---------------------------- 7401 7402:Architectures: x86 7403 7404This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush 7405hypercalls: 7406HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx, 7407HvFlushVirtualAddressList, HvFlushVirtualAddressListEx. 7408 74098.19 KVM_CAP_ARM_INJECT_SERROR_ESR 7410---------------------------------- 7411 7412:Architectures: arm64 7413 7414This capability indicates that userspace can specify (via the 7415KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it 7416takes a virtual SError interrupt exception. 7417If KVM advertises this capability, userspace can only specify the ISS field for 7418the ESR syndrome. Other parts of the ESR, such as the EC are generated by the 7419CPU when the exception is taken. If this virtual SError is taken to EL1 using 7420AArch64, this value will be reported in the ISS field of ESR_ELx. 7421 7422See KVM_CAP_VCPU_EVENTS for more details. 7423 74248.20 KVM_CAP_HYPERV_SEND_IPI 7425---------------------------- 7426 7427:Architectures: x86 7428 7429This capability indicates that KVM supports paravirtualized Hyper-V IPI send 7430hypercalls: 7431HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx. 7432 74338.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 7434----------------------------------- 7435 7436:Architectures: x86 7437 7438This capability indicates that KVM running on top of Hyper-V hypervisor 7439enables Direct TLB flush for its guests meaning that TLB flush 7440hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM. 7441Due to the different ABI for hypercall parameters between Hyper-V and 7442KVM, enabling this capability effectively disables all hypercall 7443handling by KVM (as some KVM hypercall may be mistakenly treated as TLB 7444flush hypercalls by Hyper-V) so userspace should disable KVM identification 7445in CPUID and only exposes Hyper-V identification. In this case, guest 7446thinks it's running on Hyper-V and only use Hyper-V hypercalls. 7447 74488.22 KVM_CAP_S390_VCPU_RESETS 7449----------------------------- 7450 7451:Architectures: s390 7452 7453This capability indicates that the KVM_S390_NORMAL_RESET and 7454KVM_S390_CLEAR_RESET ioctls are available. 7455 74568.23 KVM_CAP_S390_PROTECTED 7457--------------------------- 7458 7459:Architectures: s390 7460 7461This capability indicates that the Ultravisor has been initialized and 7462KVM can therefore start protected VMs. 7463This capability governs the KVM_S390_PV_COMMAND ioctl and the 7464KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected 7465guests when the state change is invalid. 7466 74678.24 KVM_CAP_STEAL_TIME 7468----------------------- 7469 7470:Architectures: arm64, x86 7471 7472This capability indicates that KVM supports steal time accounting. 7473When steal time accounting is supported it may be enabled with 7474architecture-specific interfaces. This capability and the architecture- 7475specific interfaces must be consistent, i.e. if one says the feature 7476is supported, than the other should as well and vice versa. For arm64 7477see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL". 7478For x86 see Documentation/virt/kvm/msr.rst "MSR_KVM_STEAL_TIME". 7479 74808.25 KVM_CAP_S390_DIAG318 7481------------------------- 7482 7483:Architectures: s390 7484 7485This capability enables a guest to set information about its control program 7486(i.e. guest kernel type and version). The information is helpful during 7487system/firmware service events, providing additional data about the guest 7488environments running on the machine. 7489 7490The information is associated with the DIAGNOSE 0x318 instruction, which sets 7491an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and 7492a 7-byte Control Program Version Code (CPVC). The CPNC determines what 7493environment the control program is running in (e.g. Linux, z/VM...), and the 7494CPVC is used for information specific to OS (e.g. Linux version, Linux 7495distribution...) 7496 7497If this capability is available, then the CPNC and CPVC can be synchronized 7498between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318). 7499 75008.26 KVM_CAP_X86_USER_SPACE_MSR 7501------------------------------- 7502 7503:Architectures: x86 7504 7505This capability indicates that KVM supports deflection of MSR reads and 7506writes to user space. It can be enabled on a VM level. If enabled, MSR 7507accesses that would usually trigger a #GP by KVM into the guest will 7508instead get bounced to user space through the KVM_EXIT_X86_RDMSR and 7509KVM_EXIT_X86_WRMSR exit notifications. 7510 75118.27 KVM_CAP_X86_MSR_FILTER 7512--------------------------- 7513 7514:Architectures: x86 7515 7516This capability indicates that KVM supports that accesses to user defined MSRs 7517may be rejected. With this capability exposed, KVM exports new VM ioctl 7518KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR 7519ranges that KVM should reject access to. 7520 7521In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to 7522trap and emulate MSRs that are outside of the scope of KVM as well as 7523limit the attack surface on KVM's MSR emulation code. 7524 75258.28 KVM_CAP_ENFORCE_PV_FEATURE_CPUID 7526------------------------------------- 7527 7528Architectures: x86 7529 7530When enabled, KVM will disable paravirtual features provided to the 7531guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf 7532(0x40000001). Otherwise, a guest may use the paravirtual features 7533regardless of what has actually been exposed through the CPUID leaf. 7534 75358.29 KVM_CAP_DIRTY_LOG_RING 7536--------------------------- 7537 7538:Architectures: x86 7539:Parameters: args[0] - size of the dirty log ring 7540 7541KVM is capable of tracking dirty memory using ring buffers that are 7542mmaped into userspace; there is one dirty ring per vcpu. 7543 7544The dirty ring is available to userspace as an array of 7545``struct kvm_dirty_gfn``. Each dirty entry it's defined as:: 7546 7547 struct kvm_dirty_gfn { 7548 __u32 flags; 7549 __u32 slot; /* as_id | slot_id */ 7550 __u64 offset; 7551 }; 7552 7553The following values are defined for the flags field to define the 7554current state of the entry:: 7555 7556 #define KVM_DIRTY_GFN_F_DIRTY BIT(0) 7557 #define KVM_DIRTY_GFN_F_RESET BIT(1) 7558 #define KVM_DIRTY_GFN_F_MASK 0x3 7559 7560Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM 7561ioctl to enable this capability for the new guest and set the size of 7562the rings. Enabling the capability is only allowed before creating any 7563vCPU, and the size of the ring must be a power of two. The larger the 7564ring buffer, the less likely the ring is full and the VM is forced to 7565exit to userspace. The optimal size depends on the workload, but it is 7566recommended that it be at least 64 KiB (4096 entries). 7567 7568Just like for dirty page bitmaps, the buffer tracks writes to 7569all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was 7570set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered 7571with the flag set, userspace can start harvesting dirty pages from the 7572ring buffer. 7573 7574An entry in the ring buffer can be unused (flag bits ``00``), 7575dirty (flag bits ``01``) or harvested (flag bits ``1X``). The 7576state machine for the entry is as follows:: 7577 7578 dirtied harvested reset 7579 00 -----------> 01 -------------> 1X -------+ 7580 ^ | 7581 | | 7582 +------------------------------------------+ 7583 7584To harvest the dirty pages, userspace accesses the mmaped ring buffer 7585to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage 7586the RESET bit must be cleared), then it means this GFN is a dirty GFN. 7587The userspace should harvest this GFN and mark the flags from state 7588``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set 7589to show that this GFN is harvested and waiting for a reset), and move 7590on to the next GFN. The userspace should continue to do this until the 7591flags of a GFN have the DIRTY bit cleared, meaning that it has harvested 7592all the dirty GFNs that were available. 7593 7594It's not necessary for userspace to harvest the all dirty GFNs at once. 7595However it must collect the dirty GFNs in sequence, i.e., the userspace 7596program cannot skip one dirty GFN to collect the one next to it. 7597 7598After processing one or more entries in the ring buffer, userspace 7599calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about 7600it, so that the kernel will reprotect those collected GFNs. 7601Therefore, the ioctl must be called *before* reading the content of 7602the dirty pages. 7603 7604The dirty ring can get full. When it happens, the KVM_RUN of the 7605vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL. 7606 7607The dirty ring interface has a major difference comparing to the 7608KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from 7609userspace, it's still possible that the kernel has not yet flushed the 7610processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the 7611flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one 7612needs to kick the vcpu out of KVM_RUN using a signal. The resulting 7613vmexit ensures that all dirty GFNs are flushed to the dirty rings. 7614 7615NOTE: the capability KVM_CAP_DIRTY_LOG_RING and the corresponding 7616ioctl KVM_RESET_DIRTY_RINGS are mutual exclusive to the existing ioctls 7617KVM_GET_DIRTY_LOG and KVM_CLEAR_DIRTY_LOG. After enabling 7618KVM_CAP_DIRTY_LOG_RING with an acceptable dirty ring size, the virtual 7619machine will switch to ring-buffer dirty page tracking and further 7620KVM_GET_DIRTY_LOG or KVM_CLEAR_DIRTY_LOG ioctls will fail. 7621 76228.30 KVM_CAP_XEN_HVM 7623-------------------- 7624 7625:Architectures: x86 7626 7627This capability indicates the features that Xen supports for hosting Xen 7628PVHVM guests. Valid flags are:: 7629 7630 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0) 7631 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1) 7632 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2) 7633 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 2) 7634 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL (1 << 3) 7635 7636The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG 7637ioctl is available, for the guest to set its hypercall page. 7638 7639If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be 7640provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page 7641contents, to request that KVM generate hypercall page content automatically 7642and also enable interception of guest hypercalls with KVM_EXIT_XEN. 7643 7644The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the 7645KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and 7646KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors 7647for event channel upcalls when the evtchn_upcall_pending field of a vcpu's 7648vcpu_info is set. 7649 7650The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related 7651features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are 7652supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls. 7653 7654The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries 7655of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority 7656field set to indicate 2 level event channel delivery. 7657 76588.31 KVM_CAP_PPC_MULTITCE 7659------------------------- 7660 7661:Capability: KVM_CAP_PPC_MULTITCE 7662:Architectures: ppc 7663:Type: vm 7664 7665This capability means the kernel is capable of handling hypercalls 7666H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 7667space. This significantly accelerates DMA operations for PPC KVM guests. 7668User space should expect that its handlers for these hypercalls 7669are not going to be called if user space previously registered LIOBN 7670in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 7671 7672In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 7673user space might have to advertise it for the guest. For example, 7674IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 7675present in the "ibm,hypertas-functions" device-tree property. 7676 7677The hypercalls mentioned above may or may not be processed successfully 7678in the kernel based fast path. If they can not be handled by the kernel, 7679they will get passed on to user space. So user space still has to have 7680an implementation for these despite the in kernel acceleration. 7681 7682This capability is always enabled. 7683 76848.32 KVM_CAP_PTP_KVM 7685-------------------- 7686 7687:Architectures: arm64 7688 7689This capability indicates that the KVM virtual PTP service is 7690supported in the host. A VMM can check whether the service is 7691available to the guest on migration. 7692 76938.33 KVM_CAP_HYPERV_ENFORCE_CPUID 7694--------------------------------- 7695 7696Architectures: x86 7697 7698When enabled, KVM will disable emulated Hyper-V features provided to the 7699guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all 7700currently implmented Hyper-V features are provided unconditionally when 7701Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001) 7702leaf. 7703 77048.34 KVM_CAP_EXIT_HYPERCALL 7705--------------------------- 7706 7707:Capability: KVM_CAP_EXIT_HYPERCALL 7708:Architectures: x86 7709:Type: vm 7710 7711This capability, if enabled, will cause KVM to exit to userspace 7712with KVM_EXIT_HYPERCALL exit reason to process some hypercalls. 7713 7714Calling KVM_CHECK_EXTENSION for this capability will return a bitmask 7715of hypercalls that can be configured to exit to userspace. 7716Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE. 7717 7718The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset 7719of the result of KVM_CHECK_EXTENSION. KVM will forward to userspace 7720the hypercalls whose corresponding bit is in the argument, and return 7721ENOSYS for the others. 7722 77238.35 KVM_CAP_PMU_CAPABILITY 7724--------------------------- 7725 7726:Capability KVM_CAP_PMU_CAPABILITY 7727:Architectures: x86 7728:Type: vm 7729:Parameters: arg[0] is bitmask of PMU virtualization capabilities. 7730:Returns 0 on success, -EINVAL when arg[0] contains invalid bits 7731 7732This capability alters PMU virtualization in KVM. 7733 7734Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 7735PMU virtualization capabilities that can be adjusted on a VM. 7736 7737The argument to KVM_ENABLE_CAP is also a bitmask and selects specific 7738PMU virtualization capabilities to be applied to the VM. This can 7739only be invoked on a VM prior to the creation of VCPUs. 7740 7741At this time, KVM_PMU_CAP_DISABLE is the only capability. Setting 7742this capability will disable PMU virtualization for that VM. Usermode 7743should adjust CPUID leaf 0xA to reflect that the PMU is disabled. 7744 77459. Known KVM API problems 7746========================= 7747 7748In some cases, KVM's API has some inconsistencies or common pitfalls 7749that userspace need to be aware of. This section details some of 7750these issues. 7751 7752Most of them are architecture specific, so the section is split by 7753architecture. 7754 77559.1. x86 7756-------- 7757 7758``KVM_GET_SUPPORTED_CPUID`` issues 7759^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 7760 7761In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible 7762to take its result and pass it directly to ``KVM_SET_CPUID2``. This section 7763documents some cases in which that requires some care. 7764 7765Local APIC features 7766~~~~~~~~~~~~~~~~~~~ 7767 7768CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``, 7769but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or 7770``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of 7771the local APIC. 7772 7773The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature. 7774 7775CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by ``KVM_GET_SUPPORTED_CPUID``. 7776It can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` is present and the kernel 7777has enabled in-kernel emulation of the local APIC. 7778 7779Obsolete ioctls and capabilities 7780^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 7781 7782KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually 7783available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if 7784available. 7785 7786Ordering of KVM_GET_*/KVM_SET_* ioctls 7787^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 7788 7789TBD 7790