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 centered around different kinds of file descriptors 11and ioctls that can be issued to these file descriptors. An initial 12open("/dev/kvm") obtains a handle to the kvm subsystem; this handle 13can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this 14handle will create a VM file descriptor which can be used to issue VM 15ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVICE ioctl on a VM fd will 16create a virtual cpu or device and return a file descriptor pointing to 17the new resource. 18 19In other words, the kvm API is a set of ioctls that are issued to 20different kinds of file descriptor in order to control various aspects of 21a virtual machine. Depending on the file descriptor that accepts them, 22ioctls belong to the following classes: 23 24 - System ioctls: These query and set global attributes which affect the 25 whole kvm subsystem. In addition a system ioctl is used to create 26 virtual machines. 27 28 - VM ioctls: These query and set attributes that affect an entire virtual 29 machine, for example memory layout. In addition a VM ioctl is used to 30 create virtual cpus (vcpus) and devices. 31 32 VM ioctls must be issued from the same process (address space) that was 33 used to create the VM. 34 35 - vcpu ioctls: These query and set attributes that control the operation 36 of a single virtual cpu. 37 38 vcpu ioctls should be issued from the same thread that was used to create 39 the vcpu, except for asynchronous vcpu ioctl that are marked as such in 40 the documentation. Otherwise, the first ioctl after switching threads 41 could see a performance impact. 42 43 - device ioctls: These query and set attributes that control the operation 44 of a single device. 45 46 device ioctls must be issued from the same process (address space) that 47 was used to create the VM. 48 49While most ioctls are specific to one kind of file descriptor, in some 50cases the same ioctl can belong to more than one class. 51 52The KVM API grew over time. For this reason, KVM defines many constants 53of the form ``KVM_CAP_*``, each corresponding to a set of functionality 54provided by one or more ioctls. Availability of these "capabilities" can 55be checked with :ref:`KVM_CHECK_EXTENSION <KVM_CHECK_EXTENSION>`. Some 56capabilities also need to be enabled for VMs or VCPUs where their 57functionality is desired (see :ref:`cap_enable` and :ref:`cap_enable_vm`). 58 59 602. Restrictions 61=============== 62 63In general file descriptors can be migrated among processes by means 64of fork() and the SCM_RIGHTS facility of unix domain socket. These 65kinds of tricks are explicitly not supported by kvm. While they will 66not cause harm to the host, their actual behavior is not guaranteed by 67the API. See "General description" for details on the ioctl usage 68model that is supported by KVM. 69 70It is important to note that although VM ioctls may only be issued from 71the process that created the VM, a VM's lifecycle is associated with its 72file descriptor, not its creator (process). In other words, the VM and 73its resources, *including the associated address space*, are not freed 74until the last reference to the VM's file descriptor has been released. 75For example, if fork() is issued after ioctl(KVM_CREATE_VM), the VM will 76not be freed until both the parent (original) process and its child have 77put their references to the VM's file descriptor. 78 79Because a VM's resources are not freed until the last reference to its 80file descriptor is released, creating additional references to a VM 81via fork(), dup(), etc... without careful consideration is strongly 82discouraged and may have unwanted side effects, e.g. memory allocated 83by and on behalf of the VM's process may not be freed/unaccounted when 84the VM is shut down. 85 86 873. Extensions 88============= 89 90As of Linux 2.6.22, the KVM ABI has been stabilized: no backward 91incompatible change are allowed. However, there is an extension 92facility that allows backward-compatible extensions to the API to be 93queried and used. 94 95The extension mechanism is not based on the Linux version number. 96Instead, kvm defines extension identifiers and a facility to query 97whether a particular extension identifier is available. If it is, a 98set of ioctls is available for application use. 99 100 1014. API description 102================== 103 104This section describes ioctls that can be used to control kvm guests. 105For each ioctl, the following information is provided along with a 106description: 107 108 Capability: 109 which KVM extension provides this ioctl. Can be 'basic', 110 which means that is will be provided by any kernel that supports 111 API version 12 (see :ref:`KVM_GET_API_VERSION <KVM_GET_API_VERSION>`), 112 or a KVM_CAP_xyz constant that can be checked with 113 :ref:`KVM_CHECK_EXTENSION <KVM_CHECK_EXTENSION>`. 114 115 Architectures: 116 which instruction set architectures provide this ioctl. 117 x86 includes both i386 and x86_64. 118 119 Type: 120 system, vm, or vcpu. 121 122 Parameters: 123 what parameters are accepted by the ioctl. 124 125 Returns: 126 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 127 are not detailed, but errors with specific meanings are. 128 129 130.. _KVM_GET_API_VERSION: 131 1324.1 KVM_GET_API_VERSION 133----------------------- 134 135:Capability: basic 136:Architectures: all 137:Type: system ioctl 138:Parameters: none 139:Returns: the constant KVM_API_VERSION (=12) 140 141This identifies the API version as the stable kvm API. It is not 142expected that this number will change. However, Linux 2.6.20 and 1432.6.21 report earlier versions; these are not documented and not 144supported. Applications should refuse to run if KVM_GET_API_VERSION 145returns a value other than 12. If this check passes, all ioctls 146described as 'basic' will be available. 147 148 1494.2 KVM_CREATE_VM 150----------------- 151 152:Capability: basic 153:Architectures: all 154:Type: system ioctl 155:Parameters: machine type identifier (KVM_VM_*) 156:Returns: a VM fd that can be used to control the new virtual machine. 157 158The new VM has no virtual cpus and no memory. 159You probably want to use 0 as machine type. 160 161X86: 162^^^^ 163 164Supported X86 VM types can be queried via KVM_CAP_VM_TYPES. 165 166S390: 167^^^^^ 168 169In order to create user controlled virtual machines on S390, check 170KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as 171privileged user (CAP_SYS_ADMIN). 172 173MIPS: 174^^^^^ 175 176To use hardware assisted virtualization on MIPS (VZ ASE) rather than 177the default trap & emulate implementation (which changes the virtual 178memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the 179flag KVM_VM_MIPS_VZ. 180 181ARM64: 182^^^^^^ 183 184On arm64, the physical address size for a VM (IPA Size limit) is limited 185to 40bits by default. The limit can be configured if the host supports the 186extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use 187KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type 188identifier, where IPA_Bits is the maximum width of any physical 189address used by the VM. The IPA_Bits is encoded in bits[7-0] of the 190machine type identifier. 191 192e.g, to configure a guest to use 48bit physical address size:: 193 194 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48)); 195 196The requested size (IPA_Bits) must be: 197 198 == ========================================================= 199 0 Implies default size, 40bits (for backward compatibility) 200 N Implies N bits, where N is a positive integer such that, 201 32 <= N <= Host_IPA_Limit 202 == ========================================================= 203 204Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and 205is dependent on the CPU capability and the kernel configuration. The limit can 206be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION 207ioctl() at run-time. 208 209Creation of the VM will fail if the requested IPA size (whether it is 210implicit or explicit) is unsupported on the host. 211 212Please note that configuring the IPA size does not affect the capability 213exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects 214size of the address translated by the stage2 level (guest physical to 215host physical address translations). 216 217 2184.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST 219---------------------------------------------------------- 220 221:Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST 222:Architectures: x86 223:Type: system ioctl 224:Parameters: struct kvm_msr_list (in/out) 225:Returns: 0 on success; -1 on error 226 227Errors: 228 229 ====== ============================================================ 230 EFAULT the msr index list cannot be read from or written to 231 E2BIG the msr index list is too big to fit in the array specified by 232 the user. 233 ====== ============================================================ 234 235:: 236 237 struct kvm_msr_list { 238 __u32 nmsrs; /* number of msrs in entries */ 239 __u32 indices[0]; 240 }; 241 242The user fills in the size of the indices array in nmsrs, and in return 243kvm adjusts nmsrs to reflect the actual number of msrs and fills in the 244indices array with their numbers. 245 246KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list 247varies by kvm version and host processor, but does not change otherwise. 248 249Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 250not returned in the MSR list, as different vcpus can have a different number 251of banks, as set via the KVM_X86_SETUP_MCE ioctl. 252 253KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed 254to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities 255and processor features that are exposed via MSRs (e.g., VMX capabilities). 256This list also varies by kvm version and host processor, but does not change 257otherwise. 258 259 260.. _KVM_CHECK_EXTENSION: 261 2624.4 KVM_CHECK_EXTENSION 263----------------------- 264 265:Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl 266:Architectures: all 267:Type: system ioctl, vm ioctl 268:Parameters: extension identifier (KVM_CAP_*) 269:Returns: 0 if unsupported; 1 (or some other positive integer) if supported 270 271The API allows the application to query about extensions to the core 272kvm API. Userspace passes an extension identifier (an integer) and 273receives an integer that describes the extension availability. 274Generally 0 means no and 1 means yes, but some extensions may report 275additional information in the integer return value. 276 277Based on their initialization different VMs may have different capabilities. 278It is thus encouraged to use the vm ioctl to query for capabilities (available 279with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 280 2814.5 KVM_GET_VCPU_MMAP_SIZE 282-------------------------- 283 284:Capability: basic 285:Architectures: all 286:Type: system ioctl 287:Parameters: none 288:Returns: size of vcpu mmap area, in bytes 289 290The KVM_RUN ioctl (cf.) communicates with userspace via a shared 291memory region. This ioctl returns the size of that region. See the 292KVM_RUN documentation for details. 293 294Besides the size of the KVM_RUN communication region, other areas of 295the VCPU file descriptor can be mmap-ed, including: 296 297- if KVM_CAP_COALESCED_MMIO is available, a page at 298 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons, 299 this page is included in the result of KVM_GET_VCPU_MMAP_SIZE. 300 KVM_CAP_COALESCED_MMIO is not documented yet. 301 302- if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at 303 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For more information on 304 KVM_CAP_DIRTY_LOG_RING, see :ref:`KVM_CAP_DIRTY_LOG_RING`. 305 306 3074.7 KVM_CREATE_VCPU 308------------------- 309 310:Capability: basic 311:Architectures: all 312:Type: vm ioctl 313:Parameters: vcpu id (apic id on x86) 314:Returns: vcpu fd on success, -1 on error 315 316This API adds a vcpu to a virtual machine. No more than max_vcpus may be added. 317The vcpu id is an integer in the range [0, max_vcpu_id). 318 319The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 320the KVM_CHECK_EXTENSION ioctl() at run-time. 321The maximum possible value for max_vcpus can be retrieved using the 322KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 323 324If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 325cpus max. 326If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 327same as the value returned from KVM_CAP_NR_VCPUS. 328 329The maximum possible value for max_vcpu_id can be retrieved using the 330KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time. 331 332If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id 333is the same as the value returned from KVM_CAP_MAX_VCPUS. 334 335On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 336threads in one or more virtual CPU cores. (This is because the 337hardware requires all the hardware threads in a CPU core to be in the 338same partition.) The KVM_CAP_PPC_SMT capability indicates the number 339of vcpus per virtual core (vcore). The vcore id is obtained by 340dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 341given vcore will always be in the same physical core as each other 342(though that might be a different physical core from time to time). 343Userspace can control the threading (SMT) mode of the guest by its 344allocation of vcpu ids. For example, if userspace wants 345single-threaded guest vcpus, it should make all vcpu ids be a multiple 346of the number of vcpus per vcore. 347 348For virtual cpus that have been created with S390 user controlled virtual 349machines, the resulting vcpu fd can be memory mapped at page offset 350KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 351cpu's hardware control block. 352 353 3544.8 KVM_GET_DIRTY_LOG 355--------------------- 356 357:Capability: basic 358:Architectures: all 359:Type: vm ioctl 360:Parameters: struct kvm_dirty_log (in/out) 361:Returns: 0 on success, -1 on error 362 363:: 364 365 /* for KVM_GET_DIRTY_LOG */ 366 struct kvm_dirty_log { 367 __u32 slot; 368 __u32 padding; 369 union { 370 void __user *dirty_bitmap; /* one bit per page */ 371 __u64 padding; 372 }; 373 }; 374 375Given a memory slot, return a bitmap containing any pages dirtied 376since the last call to this ioctl. Bit 0 is the first page in the 377memory slot. Ensure the entire structure is cleared to avoid padding 378issues. 379 380If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 381the address space for which you want to return the dirty bitmap. See 382KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 383 384The bits in the dirty bitmap are cleared before the ioctl returns, unless 385KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information, 386see the description of the capability. 387 388Note that the Xen shared_info page, if configured, shall always be assumed 389to be dirty. KVM will not explicitly mark it such. 390 391 3924.10 KVM_RUN 393------------ 394 395:Capability: basic 396:Architectures: all 397:Type: vcpu ioctl 398:Parameters: none 399:Returns: 0 on success, -1 on error 400 401Errors: 402 403 ======= ============================================================== 404 EINTR an unmasked signal is pending 405 ENOEXEC the vcpu hasn't been initialized or the guest tried to execute 406 instructions from device memory (arm64) 407 ENOSYS data abort outside memslots with no syndrome info and 408 KVM_CAP_ARM_NISV_TO_USER not enabled (arm64) 409 EPERM SVE feature set but not finalized (arm64) 410 ======= ============================================================== 411 412This ioctl is used to run a guest virtual cpu. While there are no 413explicit parameters, there is an implicit parameter block that can be 414obtained by mmap()ing the vcpu fd at offset 0, with the size given by 415KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct 416kvm_run' (see below). 417 418 4194.11 KVM_GET_REGS 420----------------- 421 422:Capability: basic 423:Architectures: all except arm64 424:Type: vcpu ioctl 425:Parameters: struct kvm_regs (out) 426:Returns: 0 on success, -1 on error 427 428Reads the general purpose registers from the vcpu. 429 430:: 431 432 /* x86 */ 433 struct kvm_regs { 434 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 435 __u64 rax, rbx, rcx, rdx; 436 __u64 rsi, rdi, rsp, rbp; 437 __u64 r8, r9, r10, r11; 438 __u64 r12, r13, r14, r15; 439 __u64 rip, rflags; 440 }; 441 442 /* mips */ 443 struct kvm_regs { 444 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 445 __u64 gpr[32]; 446 __u64 hi; 447 __u64 lo; 448 __u64 pc; 449 }; 450 451 /* LoongArch */ 452 struct kvm_regs { 453 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 454 unsigned long gpr[32]; 455 unsigned long pc; 456 }; 457 458 4594.12 KVM_SET_REGS 460----------------- 461 462:Capability: basic 463:Architectures: all except arm64 464:Type: vcpu ioctl 465:Parameters: struct kvm_regs (in) 466:Returns: 0 on success, -1 on error 467 468Writes the general purpose registers into the vcpu. 469 470See KVM_GET_REGS for the data structure. 471 472 4734.13 KVM_GET_SREGS 474------------------ 475 476:Capability: basic 477:Architectures: x86, ppc 478:Type: vcpu ioctl 479:Parameters: struct kvm_sregs (out) 480:Returns: 0 on success, -1 on error 481 482Reads special registers from the vcpu. 483 484:: 485 486 /* x86 */ 487 struct kvm_sregs { 488 struct kvm_segment cs, ds, es, fs, gs, ss; 489 struct kvm_segment tr, ldt; 490 struct kvm_dtable gdt, idt; 491 __u64 cr0, cr2, cr3, cr4, cr8; 492 __u64 efer; 493 __u64 apic_base; 494 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; 495 }; 496 497 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ 498 499interrupt_bitmap is a bitmap of pending external interrupts. At most 500one bit may be set. This interrupt has been acknowledged by the APIC 501but not yet injected into the cpu core. 502 503 5044.14 KVM_SET_SREGS 505------------------ 506 507:Capability: basic 508:Architectures: x86, ppc 509:Type: vcpu ioctl 510:Parameters: struct kvm_sregs (in) 511:Returns: 0 on success, -1 on error 512 513Writes special registers into the vcpu. See KVM_GET_SREGS for the 514data structures. 515 516 5174.15 KVM_TRANSLATE 518------------------ 519 520:Capability: basic 521:Architectures: x86 522:Type: vcpu ioctl 523:Parameters: struct kvm_translation (in/out) 524:Returns: 0 on success, -1 on error 525 526Translates a virtual address according to the vcpu's current address 527translation mode. 528 529:: 530 531 struct kvm_translation { 532 /* in */ 533 __u64 linear_address; 534 535 /* out */ 536 __u64 physical_address; 537 __u8 valid; 538 __u8 writeable; 539 __u8 usermode; 540 __u8 pad[5]; 541 }; 542 543 5444.16 KVM_INTERRUPT 545------------------ 546 547:Capability: basic 548:Architectures: x86, ppc, mips, riscv, loongarch 549:Type: vcpu ioctl 550:Parameters: struct kvm_interrupt (in) 551:Returns: 0 on success, negative on failure. 552 553Queues a hardware interrupt vector to be injected. 554 555:: 556 557 /* for KVM_INTERRUPT */ 558 struct kvm_interrupt { 559 /* in */ 560 __u32 irq; 561 }; 562 563X86: 564^^^^ 565 566:Returns: 567 568 ========= =================================== 569 0 on success, 570 -EEXIST if an interrupt is already enqueued 571 -EINVAL the irq number is invalid 572 -ENXIO if the PIC is in the kernel 573 -EFAULT if the pointer is invalid 574 ========= =================================== 575 576Note 'irq' is an interrupt vector, not an interrupt pin or line. This 577ioctl is useful if the in-kernel PIC is not used. 578 579PPC: 580^^^^ 581 582Queues an external interrupt to be injected. This ioctl is overloaded 583with 3 different irq values: 584 585a) KVM_INTERRUPT_SET 586 587 This injects an edge type external interrupt into the guest once it's ready 588 to receive interrupts. When injected, the interrupt is done. 589 590b) KVM_INTERRUPT_UNSET 591 592 This unsets any pending interrupt. 593 594 Only available with KVM_CAP_PPC_UNSET_IRQ. 595 596c) KVM_INTERRUPT_SET_LEVEL 597 598 This injects a level type external interrupt into the guest context. The 599 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET 600 is triggered. 601 602 Only available with KVM_CAP_PPC_IRQ_LEVEL. 603 604Note that any value for 'irq' other than the ones stated above is invalid 605and incurs unexpected behavior. 606 607This is an asynchronous vcpu ioctl and can be invoked from any thread. 608 609MIPS: 610^^^^^ 611 612Queues an external interrupt to be injected into the virtual CPU. A negative 613interrupt number dequeues the interrupt. 614 615This is an asynchronous vcpu ioctl and can be invoked from any thread. 616 617RISC-V: 618^^^^^^^ 619 620Queues an external interrupt to be injected into the virtual CPU. This ioctl 621is overloaded with 2 different irq values: 622 623a) KVM_INTERRUPT_SET 624 625 This sets external interrupt for a virtual CPU and it will receive 626 once it is ready. 627 628b) KVM_INTERRUPT_UNSET 629 630 This clears pending external interrupt for a virtual CPU. 631 632This is an asynchronous vcpu ioctl and can be invoked from any thread. 633 634LOONGARCH: 635^^^^^^^^^^ 636 637Queues an external interrupt to be injected into the virtual CPU. A negative 638interrupt number dequeues the interrupt. 639 640This is an asynchronous vcpu ioctl and can be invoked from any thread. 641 642 6434.18 KVM_GET_MSRS 644----------------- 645 646:Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system) 647:Architectures: x86 648:Type: system ioctl, vcpu ioctl 649:Parameters: struct kvm_msrs (in/out) 650:Returns: number of msrs successfully returned; 651 -1 on error 652 653When used as a system ioctl: 654Reads the values of MSR-based features that are available for the VM. This 655is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values. 656The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST 657in a system ioctl. 658 659When used as a vcpu ioctl: 660Reads model-specific registers from the vcpu. Supported msr indices can 661be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl. 662 663:: 664 665 struct kvm_msrs { 666 __u32 nmsrs; /* number of msrs in entries */ 667 __u32 pad; 668 669 struct kvm_msr_entry entries[0]; 670 }; 671 672 struct kvm_msr_entry { 673 __u32 index; 674 __u32 reserved; 675 __u64 data; 676 }; 677 678Application code should set the 'nmsrs' member (which indicates the 679size of the entries array) and the 'index' member of each array entry. 680kvm will fill in the 'data' member. 681 682 6834.19 KVM_SET_MSRS 684----------------- 685 686:Capability: basic 687:Architectures: x86 688:Type: vcpu ioctl 689:Parameters: struct kvm_msrs (in) 690:Returns: number of msrs successfully set (see below), -1 on error 691 692Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 693data structures. 694 695Application code should set the 'nmsrs' member (which indicates the 696size of the entries array), and the 'index' and 'data' members of each 697array entry. 698 699It tries to set the MSRs in array entries[] one by one. If setting an MSR 700fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated 701by KVM, etc..., it stops processing the MSR list and returns the number of 702MSRs that have been set successfully. 703 704 7054.20 KVM_SET_CPUID 706------------------ 707 708:Capability: basic 709:Architectures: x86 710:Type: vcpu ioctl 711:Parameters: struct kvm_cpuid (in) 712:Returns: 0 on success, -1 on error 713 714Defines the vcpu responses to the cpuid instruction. Applications 715should use the KVM_SET_CPUID2 ioctl if available. 716 717Caveat emptor: 718 - If this IOCTL fails, KVM gives no guarantees that previous valid CPUID 719 configuration (if there is) is not corrupted. Userspace can get a copy 720 of the resulting CPUID configuration through KVM_GET_CPUID2 in case. 721 - Using KVM_SET_CPUID{,2} after KVM_RUN, i.e. changing the guest vCPU model 722 after running the guest, may cause guest instability. 723 - Using heterogeneous CPUID configurations, modulo APIC IDs, topology, etc... 724 may cause guest instability. 725 726:: 727 728 struct kvm_cpuid_entry { 729 __u32 function; 730 __u32 eax; 731 __u32 ebx; 732 __u32 ecx; 733 __u32 edx; 734 __u32 padding; 735 }; 736 737 /* for KVM_SET_CPUID */ 738 struct kvm_cpuid { 739 __u32 nent; 740 __u32 padding; 741 struct kvm_cpuid_entry entries[0]; 742 }; 743 744 7454.21 KVM_SET_SIGNAL_MASK 746------------------------ 747 748:Capability: basic 749:Architectures: all 750:Type: vcpu ioctl 751:Parameters: struct kvm_signal_mask (in) 752:Returns: 0 on success, -1 on error 753 754Defines which signals are blocked during execution of KVM_RUN. This 755signal mask temporarily overrides the threads signal mask. Any 756unblocked signal received (except SIGKILL and SIGSTOP, which retain 757their traditional behaviour) will cause KVM_RUN to return with -EINTR. 758 759Note the signal will only be delivered if not blocked by the original 760signal mask. 761 762:: 763 764 /* for KVM_SET_SIGNAL_MASK */ 765 struct kvm_signal_mask { 766 __u32 len; 767 __u8 sigset[0]; 768 }; 769 770 7714.22 KVM_GET_FPU 772---------------- 773 774:Capability: basic 775:Architectures: x86, loongarch 776:Type: vcpu ioctl 777:Parameters: struct kvm_fpu (out) 778:Returns: 0 on success, -1 on error 779 780Reads the floating point state from the vcpu. 781 782:: 783 784 /* x86: for KVM_GET_FPU and KVM_SET_FPU */ 785 struct kvm_fpu { 786 __u8 fpr[8][16]; 787 __u16 fcw; 788 __u16 fsw; 789 __u8 ftwx; /* in fxsave format */ 790 __u8 pad1; 791 __u16 last_opcode; 792 __u64 last_ip; 793 __u64 last_dp; 794 __u8 xmm[16][16]; 795 __u32 mxcsr; 796 __u32 pad2; 797 }; 798 799 /* LoongArch: for KVM_GET_FPU and KVM_SET_FPU */ 800 struct kvm_fpu { 801 __u32 fcsr; 802 __u64 fcc; 803 struct kvm_fpureg { 804 __u64 val64[4]; 805 }fpr[32]; 806 }; 807 808 8094.23 KVM_SET_FPU 810---------------- 811 812:Capability: basic 813:Architectures: x86, loongarch 814:Type: vcpu ioctl 815:Parameters: struct kvm_fpu (in) 816:Returns: 0 on success, -1 on error 817 818Writes the floating point state to the vcpu. 819 820:: 821 822 /* x86: for KVM_GET_FPU and KVM_SET_FPU */ 823 struct kvm_fpu { 824 __u8 fpr[8][16]; 825 __u16 fcw; 826 __u16 fsw; 827 __u8 ftwx; /* in fxsave format */ 828 __u8 pad1; 829 __u16 last_opcode; 830 __u64 last_ip; 831 __u64 last_dp; 832 __u8 xmm[16][16]; 833 __u32 mxcsr; 834 __u32 pad2; 835 }; 836 837 /* LoongArch: for KVM_GET_FPU and KVM_SET_FPU */ 838 struct kvm_fpu { 839 __u32 fcsr; 840 __u64 fcc; 841 struct kvm_fpureg { 842 __u64 val64[4]; 843 }fpr[32]; 844 }; 845 846 8474.24 KVM_CREATE_IRQCHIP 848----------------------- 849 850:Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) 851:Architectures: x86, arm64, s390 852:Type: vm ioctl 853:Parameters: none 854:Returns: 0 on success, -1 on error 855 856Creates an interrupt controller model in the kernel. 857On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up 858future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both 859PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. 860On arm64, a GICv2 is created. Any other GIC versions require the usage of 861KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 862KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 863On s390, a dummy irq routing table is created. 864 865Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 866before KVM_CREATE_IRQCHIP can be used. 867 868 8694.25 KVM_IRQ_LINE 870----------------- 871 872:Capability: KVM_CAP_IRQCHIP 873:Architectures: x86, arm64 874:Type: vm ioctl 875:Parameters: struct kvm_irq_level 876:Returns: 0 on success, -1 on error 877 878Sets the level of a GSI input to the interrupt controller model in the kernel. 879On some architectures it is required that an interrupt controller model has 880been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 881interrupts require the level to be set to 1 and then back to 0. 882 883On real hardware, interrupt pins can be active-low or active-high. This 884does not matter for the level field of struct kvm_irq_level: 1 always 885means active (asserted), 0 means inactive (deasserted). 886 887x86 allows the operating system to program the interrupt polarity 888(active-low/active-high) for level-triggered interrupts, and KVM used 889to consider the polarity. However, due to bitrot in the handling of 890active-low interrupts, the above convention is now valid on x86 too. 891This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 892should not present interrupts to the guest as active-low unless this 893capability is present (or unless it is not using the in-kernel irqchip, 894of course). 895 896 897arm64 can signal an interrupt either at the CPU level, or at the 898in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 899use PPIs designated for specific cpus. The irq field is interpreted 900like this:: 901 902 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 | 903 field: | vcpu2_index | irq_type | vcpu_index | irq_id | 904 905The irq_type field has the following values: 906 907- KVM_ARM_IRQ_TYPE_CPU: 908 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 909- KVM_ARM_IRQ_TYPE_SPI: 910 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 911 (the vcpu_index field is ignored) 912- KVM_ARM_IRQ_TYPE_PPI: 913 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 914 915(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 916 917In both cases, level is used to assert/deassert the line. 918 919When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is 920identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index 921must be zero. 922 923Note that on arm64, the KVM_CAP_IRQCHIP capability only conditions 924injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always 925be used for a userspace interrupt controller. 926 927:: 928 929 struct kvm_irq_level { 930 union { 931 __u32 irq; /* GSI */ 932 __s32 status; /* not used for KVM_IRQ_LEVEL */ 933 }; 934 __u32 level; /* 0 or 1 */ 935 }; 936 937 9384.26 KVM_GET_IRQCHIP 939-------------------- 940 941:Capability: KVM_CAP_IRQCHIP 942:Architectures: x86 943:Type: vm ioctl 944:Parameters: struct kvm_irqchip (in/out) 945:Returns: 0 on success, -1 on error 946 947Reads the state of a kernel interrupt controller created with 948KVM_CREATE_IRQCHIP into a buffer provided by the caller. 949 950:: 951 952 struct kvm_irqchip { 953 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 954 __u32 pad; 955 union { 956 char dummy[512]; /* reserving space */ 957 struct kvm_pic_state pic; 958 struct kvm_ioapic_state ioapic; 959 } chip; 960 }; 961 962 9634.27 KVM_SET_IRQCHIP 964-------------------- 965 966:Capability: KVM_CAP_IRQCHIP 967:Architectures: x86 968:Type: vm ioctl 969:Parameters: struct kvm_irqchip (in) 970:Returns: 0 on success, -1 on error 971 972Sets the state of a kernel interrupt controller created with 973KVM_CREATE_IRQCHIP from a buffer provided by the caller. 974 975:: 976 977 struct kvm_irqchip { 978 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 979 __u32 pad; 980 union { 981 char dummy[512]; /* reserving space */ 982 struct kvm_pic_state pic; 983 struct kvm_ioapic_state ioapic; 984 } chip; 985 }; 986 987 9884.28 KVM_XEN_HVM_CONFIG 989----------------------- 990 991:Capability: KVM_CAP_XEN_HVM 992:Architectures: x86 993:Type: vm ioctl 994:Parameters: struct kvm_xen_hvm_config (in) 995:Returns: 0 on success, -1 on error 996 997Sets the MSR that the Xen HVM guest uses to initialize its hypercall 998page, and provides the starting address and size of the hypercall 999blobs in userspace. When the guest writes the MSR, kvm copies one 1000page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 1001memory. 1002 1003The MSR index must be in the range [0x40000000, 0x4fffffff], i.e. must reside 1004in the range that is unofficially reserved for use by hypervisors. The min/max 1005values are enumerated via KVM_XEN_MSR_MIN_INDEX and KVM_XEN_MSR_MAX_INDEX. 1006 1007:: 1008 1009 struct kvm_xen_hvm_config { 1010 __u32 flags; 1011 __u32 msr; 1012 __u64 blob_addr_32; 1013 __u64 blob_addr_64; 1014 __u8 blob_size_32; 1015 __u8 blob_size_64; 1016 __u8 pad2[30]; 1017 }; 1018 1019If certain flags are returned from the KVM_CAP_XEN_HVM check, they may 1020be set in the flags field of this ioctl: 1021 1022The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag requests KVM to generate 1023the contents of the hypercall page automatically; hypercalls will be 1024intercepted and passed to userspace through KVM_EXIT_XEN. In this 1025case, all of the blob size and address fields must be zero. 1026 1027The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates to KVM that userspace 1028will always use the KVM_XEN_HVM_EVTCHN_SEND ioctl to deliver event 1029channel interrupts rather than manipulating the guest's shared_info 1030structures directly. This, in turn, may allow KVM to enable features 1031such as intercepting the SCHEDOP_poll hypercall to accelerate PV 1032spinlock operation for the guest. Userspace may still use the ioctl 1033to deliver events if it was advertised, even if userspace does not 1034send this indication that it will always do so 1035 1036No other flags are currently valid in the struct kvm_xen_hvm_config. 1037 10384.29 KVM_GET_CLOCK 1039------------------ 1040 1041:Capability: KVM_CAP_ADJUST_CLOCK 1042:Architectures: x86 1043:Type: vm ioctl 1044:Parameters: struct kvm_clock_data (out) 1045:Returns: 0 on success, -1 on error 1046 1047Gets the current timestamp of kvmclock as seen by the current guest. In 1048conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 1049such as migration. 1050 1051When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 1052set of bits that KVM can return in struct kvm_clock_data's flag member. 1053 1054The following flags are defined: 1055 1056KVM_CLOCK_TSC_STABLE 1057 If set, the returned value is the exact kvmclock 1058 value seen by all VCPUs at the instant when KVM_GET_CLOCK was called. 1059 If clear, the returned value is simply CLOCK_MONOTONIC plus a constant 1060 offset; the offset can be modified with KVM_SET_CLOCK. KVM will try 1061 to make all VCPUs follow this clock, but the exact value read by each 1062 VCPU could differ, because the host TSC is not stable. 1063 1064KVM_CLOCK_REALTIME 1065 If set, the `realtime` field in the kvm_clock_data 1066 structure is populated with the value of the host's real time 1067 clocksource at the instant when KVM_GET_CLOCK was called. If clear, 1068 the `realtime` field does not contain a value. 1069 1070KVM_CLOCK_HOST_TSC 1071 If set, the `host_tsc` field in the kvm_clock_data 1072 structure is populated with the value of the host's timestamp counter (TSC) 1073 at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field 1074 does not contain a value. 1075 1076:: 1077 1078 struct kvm_clock_data { 1079 __u64 clock; /* kvmclock current value */ 1080 __u32 flags; 1081 __u32 pad0; 1082 __u64 realtime; 1083 __u64 host_tsc; 1084 __u32 pad[4]; 1085 }; 1086 1087 10884.30 KVM_SET_CLOCK 1089------------------ 1090 1091:Capability: KVM_CAP_ADJUST_CLOCK 1092:Architectures: x86 1093:Type: vm ioctl 1094:Parameters: struct kvm_clock_data (in) 1095:Returns: 0 on success, -1 on error 1096 1097Sets the current timestamp of kvmclock to the value specified in its parameter. 1098In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 1099such as migration. 1100 1101The following flags can be passed: 1102 1103KVM_CLOCK_REALTIME 1104 If set, KVM will compare the value of the `realtime` field 1105 with the value of the host's real time clocksource at the instant when 1106 KVM_SET_CLOCK was called. The difference in elapsed time is added to the final 1107 kvmclock value that will be provided to guests. 1108 1109Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored. 1110 1111:: 1112 1113 struct kvm_clock_data { 1114 __u64 clock; /* kvmclock current value */ 1115 __u32 flags; 1116 __u32 pad0; 1117 __u64 realtime; 1118 __u64 host_tsc; 1119 __u32 pad[4]; 1120 }; 1121 1122 11234.31 KVM_GET_VCPU_EVENTS 1124------------------------ 1125 1126:Capability: KVM_CAP_VCPU_EVENTS 1127:Extended by: KVM_CAP_INTR_SHADOW 1128:Architectures: x86, arm64 1129:Type: vcpu ioctl 1130:Parameters: struct kvm_vcpu_events (out) 1131:Returns: 0 on success, -1 on error 1132 1133X86: 1134^^^^ 1135 1136Gets currently pending exceptions, interrupts, and NMIs as well as related 1137states of the vcpu. 1138 1139:: 1140 1141 struct kvm_vcpu_events { 1142 struct { 1143 __u8 injected; 1144 __u8 nr; 1145 __u8 has_error_code; 1146 __u8 pending; 1147 __u32 error_code; 1148 } exception; 1149 struct { 1150 __u8 injected; 1151 __u8 nr; 1152 __u8 soft; 1153 __u8 shadow; 1154 } interrupt; 1155 struct { 1156 __u8 injected; 1157 __u8 pending; 1158 __u8 masked; 1159 __u8 pad; 1160 } nmi; 1161 __u32 sipi_vector; 1162 __u32 flags; 1163 struct { 1164 __u8 smm; 1165 __u8 pending; 1166 __u8 smm_inside_nmi; 1167 __u8 latched_init; 1168 } smi; 1169 __u8 reserved[27]; 1170 __u8 exception_has_payload; 1171 __u64 exception_payload; 1172 }; 1173 1174The following bits are defined in the flags field: 1175 1176- KVM_VCPUEVENT_VALID_SHADOW may be set to signal that 1177 interrupt.shadow contains a valid state. 1178 1179- KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a 1180 valid state. 1181 1182- KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the 1183 exception_has_payload, exception_payload, and exception.pending 1184 fields contain a valid state. This bit will be set whenever 1185 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1186 1187- KVM_VCPUEVENT_VALID_TRIPLE_FAULT may be set to signal that the 1188 triple_fault_pending field contains a valid state. This bit will 1189 be set whenever KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled. 1190 1191ARM64: 1192^^^^^^ 1193 1194If the guest accesses a device that is being emulated by the host kernel in 1195such a way that a real device would generate a physical SError, KVM may make 1196a virtual SError pending for that VCPU. This system error interrupt remains 1197pending until the guest takes the exception by unmasking PSTATE.A. 1198 1199Running the VCPU may cause it to take a pending SError, or make an access that 1200causes an SError to become pending. The event's description is only valid while 1201the VPCU is not running. 1202 1203This API provides a way to read and write the pending 'event' state that is not 1204visible to the guest. To save, restore or migrate a VCPU the struct representing 1205the state can be read then written using this GET/SET API, along with the other 1206guest-visible registers. It is not possible to 'cancel' an SError that has been 1207made pending. 1208 1209A device being emulated in user-space may also wish to generate an SError. To do 1210this the events structure can be populated by user-space. The current state 1211should be read first, to ensure no existing SError is pending. If an existing 1212SError is pending, the architecture's 'Multiple SError interrupts' rules should 1213be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and 1214Serviceability (RAS) Specification"). 1215 1216SError exceptions always have an ESR value. Some CPUs have the ability to 1217specify what the virtual SError's ESR value should be. These systems will 1218advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will 1219always have a non-zero value when read, and the agent making an SError pending 1220should specify the ISS field in the lower 24 bits of exception.serror_esr. If 1221the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events 1222with exception.has_esr as zero, KVM will choose an ESR. 1223 1224Specifying exception.has_esr on a system that does not support it will return 1225-EINVAL. Setting anything other than the lower 24bits of exception.serror_esr 1226will return -EINVAL. 1227 1228It is not possible to read back a pending external abort (injected via 1229KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered 1230directly to the virtual CPU). 1231 1232:: 1233 1234 struct kvm_vcpu_events { 1235 struct { 1236 __u8 serror_pending; 1237 __u8 serror_has_esr; 1238 __u8 ext_dabt_pending; 1239 /* Align it to 8 bytes */ 1240 __u8 pad[5]; 1241 __u64 serror_esr; 1242 } exception; 1243 __u32 reserved[12]; 1244 }; 1245 12464.32 KVM_SET_VCPU_EVENTS 1247------------------------ 1248 1249:Capability: KVM_CAP_VCPU_EVENTS 1250:Extended by: KVM_CAP_INTR_SHADOW 1251:Architectures: x86, arm64 1252:Type: vcpu ioctl 1253:Parameters: struct kvm_vcpu_events (in) 1254:Returns: 0 on success, -1 on error 1255 1256X86: 1257^^^^ 1258 1259Set pending exceptions, interrupts, and NMIs as well as related states of the 1260vcpu. 1261 1262See KVM_GET_VCPU_EVENTS for the data structure. 1263 1264Fields that may be modified asynchronously by running VCPUs can be excluded 1265from the update. These fields are nmi.pending, sipi_vector, smi.smm, 1266smi.pending. Keep the corresponding bits in the flags field cleared to 1267suppress overwriting the current in-kernel state. The bits are: 1268 1269=============================== ================================== 1270KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel 1271KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector 1272KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct. 1273=============================== ================================== 1274 1275If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 1276the flags field to signal that interrupt.shadow contains a valid state and 1277shall be written into the VCPU. 1278 1279KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 1280 1281If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD 1282can be set in the flags field to signal that the 1283exception_has_payload, exception_payload, and exception.pending fields 1284contain a valid state and shall be written into the VCPU. 1285 1286If KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled, KVM_VCPUEVENT_VALID_TRIPLE_FAULT 1287can be set in flags field to signal that the triple_fault field contains 1288a valid state and shall be written into the VCPU. 1289 1290ARM64: 1291^^^^^^ 1292 1293User space may need to inject several types of events to the guest. 1294 1295Set the pending SError exception state for this VCPU. It is not possible to 1296'cancel' an Serror that has been made pending. 1297 1298If the guest performed an access to I/O memory which could not be handled by 1299userspace, for example because of missing instruction syndrome decode 1300information or because there is no device mapped at the accessed IPA, then 1301userspace can ask the kernel to inject an external abort using the address 1302from the exiting fault on the VCPU. It is a programming error to set 1303ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or 1304KVM_EXIT_ARM_NISV. This feature is only available if the system supports 1305KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in 1306how userspace reports accesses for the above cases to guests, across different 1307userspace implementations. Nevertheless, userspace can still emulate all Arm 1308exceptions by manipulating individual registers using the KVM_SET_ONE_REG API. 1309 1310See KVM_GET_VCPU_EVENTS for the data structure. 1311 1312 13134.33 KVM_GET_DEBUGREGS 1314---------------------- 1315 1316:Capability: KVM_CAP_DEBUGREGS 1317:Architectures: x86 1318:Type: vcpu ioctl 1319:Parameters: struct kvm_debugregs (out) 1320:Returns: 0 on success, -1 on error 1321 1322Reads debug registers from the vcpu. 1323 1324:: 1325 1326 struct kvm_debugregs { 1327 __u64 db[4]; 1328 __u64 dr6; 1329 __u64 dr7; 1330 __u64 flags; 1331 __u64 reserved[9]; 1332 }; 1333 1334 13354.34 KVM_SET_DEBUGREGS 1336---------------------- 1337 1338:Capability: KVM_CAP_DEBUGREGS 1339:Architectures: x86 1340:Type: vcpu ioctl 1341:Parameters: struct kvm_debugregs (in) 1342:Returns: 0 on success, -1 on error 1343 1344Writes debug registers into the vcpu. 1345 1346See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 1347yet and must be cleared on entry. 1348 1349 13504.35 KVM_SET_USER_MEMORY_REGION 1351------------------------------- 1352 1353:Capability: KVM_CAP_USER_MEMORY 1354:Architectures: all 1355:Type: vm ioctl 1356:Parameters: struct kvm_userspace_memory_region (in) 1357:Returns: 0 on success, -1 on error 1358 1359:: 1360 1361 struct kvm_userspace_memory_region { 1362 __u32 slot; 1363 __u32 flags; 1364 __u64 guest_phys_addr; 1365 __u64 memory_size; /* bytes */ 1366 __u64 userspace_addr; /* start of the userspace allocated memory */ 1367 }; 1368 1369 /* for kvm_userspace_memory_region::flags */ 1370 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 1371 #define KVM_MEM_READONLY (1UL << 1) 1372 1373This ioctl allows the user to create, modify or delete a guest physical 1374memory slot. Bits 0-15 of "slot" specify the slot id and this value 1375should be less than the maximum number of user memory slots supported per 1376VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS. 1377Slots may not overlap in guest physical address space. 1378 1379If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 1380specifies the address space which is being modified. They must be 1381less than the value that KVM_CHECK_EXTENSION returns for the 1382KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 1383are unrelated; the restriction on overlapping slots only applies within 1384each address space. 1385 1386Deleting a slot is done by passing zero for memory_size. When changing 1387an existing slot, it may be moved in the guest physical memory space, 1388or its flags may be modified, but it may not be resized. 1389 1390Memory for the region is taken starting at the address denoted by the 1391field userspace_addr, which must point at user addressable memory for 1392the entire memory slot size. Any object may back this memory, including 1393anonymous memory, ordinary files, and hugetlbfs. 1394 1395On architectures that support a form of address tagging, userspace_addr must 1396be an untagged address. 1397 1398It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 1399be identical. This allows large pages in the guest to be backed by large 1400pages in the host. 1401 1402The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 1403KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 1404writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 1405use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 1406to make a new slot read-only. In this case, writes to this memory will be 1407posted to userspace as KVM_EXIT_MMIO exits. 1408 1409When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 1410the memory region are automatically reflected into the guest. For example, an 1411mmap() that affects the region will be made visible immediately. Another 1412example is madvise(MADV_DROP). 1413 1414For TDX guest, deleting/moving memory region loses guest memory contents. 1415Read only region isn't supported. Only as-id 0 is supported. 1416 1417Note: On arm64, a write generated by the page-table walker (to update 1418the Access and Dirty flags, for example) never results in a 1419KVM_EXIT_MMIO exit when the slot has the KVM_MEM_READONLY flag. This 1420is because KVM cannot provide the data that would be written by the 1421page-table walker, making it impossible to emulate the access. 1422Instead, an abort (data abort if the cause of the page-table update 1423was a load or a store, instruction abort if it was an instruction 1424fetch) is injected in the guest. 1425 1426S390: 1427^^^^^ 1428 1429Returns -EINVAL or -EEXIST if the VM has the KVM_VM_S390_UCONTROL flag set. 1430Returns -EINVAL if called on a protected VM. 1431 14324.36 KVM_SET_TSS_ADDR 1433--------------------- 1434 1435:Capability: KVM_CAP_SET_TSS_ADDR 1436:Architectures: x86 1437:Type: vm ioctl 1438:Parameters: unsigned long tss_address (in) 1439:Returns: 0 on success, -1 on error 1440 1441This ioctl defines the physical address of a three-page region in the guest 1442physical address space. The region must be within the first 4GB of the 1443guest physical address space and must not conflict with any memory slot 1444or any mmio address. The guest may malfunction if it accesses this memory 1445region. 1446 1447This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1448because of a quirk in the virtualization implementation (see the internals 1449documentation when it pops into existence). 1450 1451 1452.. _KVM_ENABLE_CAP: 1453 14544.37 KVM_ENABLE_CAP 1455------------------- 1456 1457:Capability: KVM_CAP_ENABLE_CAP 1458:Architectures: mips, ppc, s390, x86, loongarch 1459:Type: vcpu ioctl 1460:Parameters: struct kvm_enable_cap (in) 1461:Returns: 0 on success; -1 on error 1462 1463:Capability: KVM_CAP_ENABLE_CAP_VM 1464:Architectures: all 1465:Type: vm ioctl 1466:Parameters: struct kvm_enable_cap (in) 1467:Returns: 0 on success; -1 on error 1468 1469.. note:: 1470 1471 Not all extensions are enabled by default. Using this ioctl the application 1472 can enable an extension, making it available to the guest. 1473 1474On systems that do not support this ioctl, it always fails. On systems that 1475do support it, it only works for extensions that are supported for enablement. 1476 1477To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1478be used. 1479 1480:: 1481 1482 struct kvm_enable_cap { 1483 /* in */ 1484 __u32 cap; 1485 1486The capability that is supposed to get enabled. 1487 1488:: 1489 1490 __u32 flags; 1491 1492A bitfield indicating future enhancements. Has to be 0 for now. 1493 1494:: 1495 1496 __u64 args[4]; 1497 1498Arguments for enabling a feature. If a feature needs initial values to 1499function properly, this is the place to put them. 1500 1501:: 1502 1503 __u8 pad[64]; 1504 }; 1505 1506The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1507for vm-wide capabilities. 1508 15094.38 KVM_GET_MP_STATE 1510--------------------- 1511 1512:Capability: KVM_CAP_MP_STATE 1513:Architectures: x86, s390, arm64, riscv, loongarch 1514:Type: vcpu ioctl 1515:Parameters: struct kvm_mp_state (out) 1516:Returns: 0 on success; -1 on error 1517 1518:: 1519 1520 struct kvm_mp_state { 1521 __u32 mp_state; 1522 }; 1523 1524Returns the vcpu's current "multiprocessing state" (though also valid on 1525uniprocessor guests). 1526 1527Possible values are: 1528 1529 ========================== =============================================== 1530 KVM_MP_STATE_RUNNABLE the vcpu is currently running 1531 [x86,arm64,riscv,loongarch] 1532 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP) 1533 which has not yet received an INIT signal [x86] 1534 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is 1535 now ready for a SIPI [x86] 1536 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and 1537 is waiting for an interrupt [x86] 1538 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector 1539 accessible via KVM_GET_VCPU_EVENTS) [x86] 1540 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm64,riscv] 1541 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390] 1542 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted) 1543 [s390] 1544 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state 1545 [s390] 1546 KVM_MP_STATE_SUSPENDED the vcpu is in a suspend state and is waiting 1547 for a wakeup event [arm64] 1548 ========================== =============================================== 1549 1550On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1551in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1552these architectures. 1553 1554For arm64: 1555^^^^^^^^^^ 1556 1557If a vCPU is in the KVM_MP_STATE_SUSPENDED state, KVM will emulate the 1558architectural execution of a WFI instruction. 1559 1560If a wakeup event is recognized, KVM will exit to userspace with a 1561KVM_SYSTEM_EVENT exit, where the event type is KVM_SYSTEM_EVENT_WAKEUP. If 1562userspace wants to honor the wakeup, it must set the vCPU's MP state to 1563KVM_MP_STATE_RUNNABLE. If it does not, KVM will continue to await a wakeup 1564event in subsequent calls to KVM_RUN. 1565 1566.. warning:: 1567 1568 If userspace intends to keep the vCPU in a SUSPENDED state, it is 1569 strongly recommended that userspace take action to suppress the 1570 wakeup event (such as masking an interrupt). Otherwise, subsequent 1571 calls to KVM_RUN will immediately exit with a KVM_SYSTEM_EVENT_WAKEUP 1572 event and inadvertently waste CPU cycles. 1573 1574 Additionally, if userspace takes action to suppress a wakeup event, 1575 it is strongly recommended that it also restores the vCPU to its 1576 original state when the vCPU is made RUNNABLE again. For example, 1577 if userspace masked a pending interrupt to suppress the wakeup, 1578 the interrupt should be unmasked before returning control to the 1579 guest. 1580 1581For riscv: 1582^^^^^^^^^^ 1583 1584The only states that are valid are KVM_MP_STATE_STOPPED and 1585KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1586 1587On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect 1588whether the vcpu is runnable. 1589 15904.39 KVM_SET_MP_STATE 1591--------------------- 1592 1593:Capability: KVM_CAP_MP_STATE 1594:Architectures: x86, s390, arm64, riscv, loongarch 1595:Type: vcpu ioctl 1596:Parameters: struct kvm_mp_state (in) 1597:Returns: 0 on success; -1 on error 1598 1599Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1600arguments. 1601 1602On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1603in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1604these architectures. 1605 1606For arm64/riscv: 1607^^^^^^^^^^^^^^^^ 1608 1609The only states that are valid are KVM_MP_STATE_STOPPED and 1610KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1611 1612On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect 1613whether the vcpu is runnable. 1614 16154.40 KVM_SET_IDENTITY_MAP_ADDR 1616------------------------------ 1617 1618:Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1619:Architectures: x86 1620:Type: vm ioctl 1621:Parameters: unsigned long identity (in) 1622:Returns: 0 on success, -1 on error 1623 1624This ioctl defines the physical address of a one-page region in the guest 1625physical address space. The region must be within the first 4GB of the 1626guest physical address space and must not conflict with any memory slot 1627or any mmio address. The guest may malfunction if it accesses this memory 1628region. 1629 1630Setting the address to 0 will result in resetting the address to its default 1631(0xfffbc000). 1632 1633This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1634because of a quirk in the virtualization implementation (see the internals 1635documentation when it pops into existence). 1636 1637Fails if any VCPU has already been created. 1638 16394.41 KVM_SET_BOOT_CPU_ID 1640------------------------ 1641 1642:Capability: KVM_CAP_SET_BOOT_CPU_ID 1643:Architectures: x86 1644:Type: vm ioctl 1645:Parameters: unsigned long vcpu_id 1646:Returns: 0 on success, -1 on error 1647 1648Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1649as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1650is vcpu 0. This ioctl has to be called before vcpu creation, 1651otherwise it will return EBUSY error. 1652 1653 16544.42 KVM_GET_XSAVE 1655------------------ 1656 1657:Capability: KVM_CAP_XSAVE 1658:Architectures: x86 1659:Type: vcpu ioctl 1660:Parameters: struct kvm_xsave (out) 1661:Returns: 0 on success, -1 on error 1662 1663 1664:: 1665 1666 struct kvm_xsave { 1667 __u32 region[1024]; 1668 __u32 extra[0]; 1669 }; 1670 1671This ioctl would copy current vcpu's xsave struct to the userspace. 1672 1673 16744.43 KVM_SET_XSAVE 1675------------------ 1676 1677:Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2 1678:Architectures: x86 1679:Type: vcpu ioctl 1680:Parameters: struct kvm_xsave (in) 1681:Returns: 0 on success, -1 on error 1682 1683:: 1684 1685 1686 struct kvm_xsave { 1687 __u32 region[1024]; 1688 __u32 extra[0]; 1689 }; 1690 1691This ioctl would copy userspace's xsave struct to the kernel. It copies 1692as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2), 1693when invoked on the vm file descriptor. The size value returned by 1694KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 1695Currently, it is only greater than 4096 if a dynamic feature has been 1696enabled with ``arch_prctl()``, but this may change in the future. 1697 1698The offsets of the state save areas in struct kvm_xsave follow the 1699contents of CPUID leaf 0xD on the host. 1700 1701 17024.44 KVM_GET_XCRS 1703----------------- 1704 1705:Capability: KVM_CAP_XCRS 1706:Architectures: x86 1707:Type: vcpu ioctl 1708:Parameters: struct kvm_xcrs (out) 1709:Returns: 0 on success, -1 on error 1710 1711:: 1712 1713 struct kvm_xcr { 1714 __u32 xcr; 1715 __u32 reserved; 1716 __u64 value; 1717 }; 1718 1719 struct kvm_xcrs { 1720 __u32 nr_xcrs; 1721 __u32 flags; 1722 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1723 __u64 padding[16]; 1724 }; 1725 1726This ioctl would copy current vcpu's xcrs to the userspace. 1727 1728 17294.45 KVM_SET_XCRS 1730----------------- 1731 1732:Capability: KVM_CAP_XCRS 1733:Architectures: x86 1734:Type: vcpu ioctl 1735:Parameters: struct kvm_xcrs (in) 1736:Returns: 0 on success, -1 on error 1737 1738:: 1739 1740 struct kvm_xcr { 1741 __u32 xcr; 1742 __u32 reserved; 1743 __u64 value; 1744 }; 1745 1746 struct kvm_xcrs { 1747 __u32 nr_xcrs; 1748 __u32 flags; 1749 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1750 __u64 padding[16]; 1751 }; 1752 1753This ioctl would set vcpu's xcr to the value userspace specified. 1754 1755 17564.46 KVM_GET_SUPPORTED_CPUID 1757---------------------------- 1758 1759:Capability: KVM_CAP_EXT_CPUID 1760:Architectures: x86 1761:Type: system ioctl 1762:Parameters: struct kvm_cpuid2 (in/out) 1763:Returns: 0 on success, -1 on error 1764 1765:: 1766 1767 struct kvm_cpuid2 { 1768 __u32 nent; 1769 __u32 padding; 1770 struct kvm_cpuid_entry2 entries[0]; 1771 }; 1772 1773 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1774 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 1775 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 1776 1777 struct kvm_cpuid_entry2 { 1778 __u32 function; 1779 __u32 index; 1780 __u32 flags; 1781 __u32 eax; 1782 __u32 ebx; 1783 __u32 ecx; 1784 __u32 edx; 1785 __u32 padding[3]; 1786 }; 1787 1788This ioctl returns x86 cpuid features which are supported by both the 1789hardware and kvm in its default configuration. Userspace can use the 1790information returned by this ioctl to construct cpuid information (for 1791KVM_SET_CPUID2) that is consistent with hardware, kernel, and 1792userspace capabilities, and with user requirements (for example, the 1793user may wish to constrain cpuid to emulate older hardware, or for 1794feature consistency across a cluster). 1795 1796Dynamically-enabled feature bits need to be requested with 1797``arch_prctl()`` before calling this ioctl. Feature bits that have not 1798been requested are excluded from the result. 1799 1800Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may 1801expose cpuid features (e.g. MONITOR) which are not supported by kvm in 1802its default configuration. If userspace enables such capabilities, it 1803is responsible for modifying the results of this ioctl appropriately. 1804 1805Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1806with the 'nent' field indicating the number of entries in the variable-size 1807array 'entries'. If the number of entries is too low to describe the cpu 1808capabilities, an error (E2BIG) is returned. If the number is too high, 1809the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1810number is just right, the 'nent' field is adjusted to the number of valid 1811entries in the 'entries' array, which is then filled. 1812 1813The entries returned are the host cpuid as returned by the cpuid instruction, 1814with unknown or unsupported features masked out. Some features (for example, 1815x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1816emulate them efficiently. The fields in each entry are defined as follows: 1817 1818 function: 1819 the eax value used to obtain the entry 1820 1821 index: 1822 the ecx value used to obtain the entry (for entries that are 1823 affected by ecx) 1824 1825 flags: 1826 an OR of zero or more of the following: 1827 1828 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1829 if the index field is valid 1830 1831 eax, ebx, ecx, edx: 1832 the values returned by the cpuid instruction for 1833 this function/index combination 1834 1835x2APIC (CPUID leaf 1, ecx[21) and TSC deadline timer (CPUID leaf 1, ecx[24]) 1836may be returned as true, but they depend on KVM_CREATE_IRQCHIP for in-kernel 1837emulation of the local APIC. TSC deadline timer support is also reported via:: 1838 1839 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1840 1841if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1842feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1843 1844Enabling x2APIC in KVM_SET_CPUID2 requires KVM_CREATE_IRQCHIP as KVM doesn't 1845support forwarding x2APIC MSR accesses to userspace, i.e. KVM does not support 1846emulating x2APIC in userspace. 1847 18484.47 KVM_PPC_GET_PVINFO 1849----------------------- 1850 1851:Capability: KVM_CAP_PPC_GET_PVINFO 1852:Architectures: ppc 1853:Type: vm ioctl 1854:Parameters: struct kvm_ppc_pvinfo (out) 1855:Returns: 0 on success, !0 on error 1856 1857:: 1858 1859 struct kvm_ppc_pvinfo { 1860 __u32 flags; 1861 __u32 hcall[4]; 1862 __u8 pad[108]; 1863 }; 1864 1865This ioctl fetches PV specific information that need to be passed to the guest 1866using the device tree or other means from vm context. 1867 1868The hcall array defines 4 instructions that make up a hypercall. 1869 1870If any additional field gets added to this structure later on, a bit for that 1871additional piece of information will be set in the flags bitmap. 1872 1873The flags bitmap is defined as:: 1874 1875 /* the host supports the ePAPR idle hcall 1876 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1877 18784.52 KVM_SET_GSI_ROUTING 1879------------------------ 1880 1881:Capability: KVM_CAP_IRQ_ROUTING 1882:Architectures: x86 s390 arm64 1883:Type: vm ioctl 1884:Parameters: struct kvm_irq_routing (in) 1885:Returns: 0 on success, -1 on error 1886 1887Sets the GSI routing table entries, overwriting any previously set entries. 1888 1889On arm64, GSI routing has the following limitation: 1890 1891- GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1892 1893:: 1894 1895 struct kvm_irq_routing { 1896 __u32 nr; 1897 __u32 flags; 1898 struct kvm_irq_routing_entry entries[0]; 1899 }; 1900 1901No flags are specified so far, the corresponding field must be set to zero. 1902 1903:: 1904 1905 struct kvm_irq_routing_entry { 1906 __u32 gsi; 1907 __u32 type; 1908 __u32 flags; 1909 __u32 pad; 1910 union { 1911 struct kvm_irq_routing_irqchip irqchip; 1912 struct kvm_irq_routing_msi msi; 1913 struct kvm_irq_routing_s390_adapter adapter; 1914 struct kvm_irq_routing_hv_sint hv_sint; 1915 struct kvm_irq_routing_xen_evtchn xen_evtchn; 1916 __u32 pad[8]; 1917 } u; 1918 }; 1919 1920 /* gsi routing entry types */ 1921 #define KVM_IRQ_ROUTING_IRQCHIP 1 1922 #define KVM_IRQ_ROUTING_MSI 2 1923 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1924 #define KVM_IRQ_ROUTING_HV_SINT 4 1925 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 1926 1927On s390, adding a KVM_IRQ_ROUTING_S390_ADAPTER is rejected on ucontrol VMs with 1928error -EINVAL. 1929 1930flags: 1931 1932- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1933 type, specifies that the devid field contains a valid value. The per-VM 1934 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1935 the device ID. If this capability is not available, userspace should 1936 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1937- zero otherwise 1938 1939:: 1940 1941 struct kvm_irq_routing_irqchip { 1942 __u32 irqchip; 1943 __u32 pin; 1944 }; 1945 1946 struct kvm_irq_routing_msi { 1947 __u32 address_lo; 1948 __u32 address_hi; 1949 __u32 data; 1950 union { 1951 __u32 pad; 1952 __u32 devid; 1953 }; 1954 }; 1955 1956If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1957for the device that wrote the MSI message. For PCI, this is usually a 1958BDF identifier in the lower 16 bits. 1959 1960On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1961feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1962address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1963address_hi must be zero. 1964 1965:: 1966 1967 struct kvm_irq_routing_s390_adapter { 1968 __u64 ind_addr; 1969 __u64 summary_addr; 1970 __u64 ind_offset; 1971 __u32 summary_offset; 1972 __u32 adapter_id; 1973 }; 1974 1975 struct kvm_irq_routing_hv_sint { 1976 __u32 vcpu; 1977 __u32 sint; 1978 }; 1979 1980 struct kvm_irq_routing_xen_evtchn { 1981 __u32 port; 1982 __u32 vcpu; 1983 __u32 priority; 1984 }; 1985 1986 1987When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL bit 1988in its indication of supported features, routing to Xen event channels 1989is supported. Although the priority field is present, only the value 1990KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported, which means delivery by 19912 level event channels. FIFO event channel support may be added in 1992the future. 1993 1994 19954.55 KVM_SET_TSC_KHZ 1996-------------------- 1997 1998:Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM_TSC_CONTROL 1999:Architectures: x86 2000:Type: vcpu ioctl / vm ioctl 2001:Parameters: virtual tsc_khz 2002:Returns: 0 on success, -1 on error 2003 2004Specifies the tsc frequency for the virtual machine. The unit of the 2005frequency is KHz. 2006 2007If the KVM_CAP_VM_TSC_CONTROL capability is advertised, this can also 2008be used as a vm ioctl to set the initial tsc frequency of subsequently 2009created vCPUs. Note, the vm ioctl is only allowed prior to creating vCPUs. 2010 2011For TSC protected Confidential Computing (CoCo) VMs where TSC frequency 2012is configured once at VM scope and remains unchanged during VM's 2013lifetime, the vm ioctl should be used to configure the TSC frequency 2014and the vcpu ioctl is not supported. 2015 2016Example of such CoCo VMs: TDX guests. 2017 20184.56 KVM_GET_TSC_KHZ 2019-------------------- 2020 2021:Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM_TSC_CONTROL 2022:Architectures: x86 2023:Type: vcpu ioctl / vm ioctl 2024:Parameters: none 2025:Returns: virtual tsc-khz on success, negative value on error 2026 2027Returns the tsc frequency of the guest. The unit of the return value is 2028KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 2029error. 2030 2031 20324.57 KVM_GET_LAPIC 2033------------------ 2034 2035:Capability: KVM_CAP_IRQCHIP 2036:Architectures: x86 2037:Type: vcpu ioctl 2038:Parameters: struct kvm_lapic_state (out) 2039:Returns: 0 on success, -1 on error 2040 2041:: 2042 2043 #define KVM_APIC_REG_SIZE 0x400 2044 struct kvm_lapic_state { 2045 char regs[KVM_APIC_REG_SIZE]; 2046 }; 2047 2048Reads the Local APIC registers and copies them into the input argument. The 2049data format and layout are the same as documented in the architecture manual. 2050 2051If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 2052enabled, then the format of APIC_ID register depends on the APIC mode 2053(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 2054the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 2055which is stored in bits 31-24 of the APIC register, or equivalently in 2056byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 2057be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 2058 2059If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 2060always uses xAPIC format. 2061 2062 20634.58 KVM_SET_LAPIC 2064------------------ 2065 2066:Capability: KVM_CAP_IRQCHIP 2067:Architectures: x86 2068:Type: vcpu ioctl 2069:Parameters: struct kvm_lapic_state (in) 2070:Returns: 0 on success, -1 on error 2071 2072:: 2073 2074 #define KVM_APIC_REG_SIZE 0x400 2075 struct kvm_lapic_state { 2076 char regs[KVM_APIC_REG_SIZE]; 2077 }; 2078 2079Copies the input argument into the Local APIC registers. The data format 2080and layout are the same as documented in the architecture manual. 2081 2082The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 2083regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 2084See the note in KVM_GET_LAPIC. 2085 2086 20874.59 KVM_IOEVENTFD 2088------------------ 2089 2090:Capability: KVM_CAP_IOEVENTFD 2091:Architectures: all 2092:Type: vm ioctl 2093:Parameters: struct kvm_ioeventfd (in) 2094:Returns: 0 on success, !0 on error 2095 2096This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 2097within the guest. A guest write in the registered address will signal the 2098provided event instead of triggering an exit. 2099 2100:: 2101 2102 struct kvm_ioeventfd { 2103 __u64 datamatch; 2104 __u64 addr; /* legal pio/mmio address */ 2105 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 2106 __s32 fd; 2107 __u32 flags; 2108 __u8 pad[36]; 2109 }; 2110 2111For the special case of virtio-ccw devices on s390, the ioevent is matched 2112to a subchannel/virtqueue tuple instead. 2113 2114The following flags are defined:: 2115 2116 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 2117 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 2118 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 2119 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 2120 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 2121 2122If datamatch flag is set, the event will be signaled only if the written value 2123to the registered address is equal to datamatch in struct kvm_ioeventfd. 2124 2125For virtio-ccw devices, addr contains the subchannel id and datamatch the 2126virtqueue index. 2127 2128With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 2129the kernel will ignore the length of guest write and may get a faster vmexit. 2130The speedup may only apply to specific architectures, but the ioeventfd will 2131work anyway. 2132 21334.60 KVM_DIRTY_TLB 2134------------------ 2135 2136:Capability: KVM_CAP_SW_TLB 2137:Architectures: ppc 2138:Type: vcpu ioctl 2139:Parameters: struct kvm_dirty_tlb (in) 2140:Returns: 0 on success, -1 on error 2141 2142:: 2143 2144 struct kvm_dirty_tlb { 2145 __u64 bitmap; 2146 __u32 num_dirty; 2147 }; 2148 2149This must be called whenever userspace has changed an entry in the shared 2150TLB, prior to calling KVM_RUN on the associated vcpu. 2151 2152The "bitmap" field is the userspace address of an array. This array 2153consists of a number of bits, equal to the total number of TLB entries as 2154determined by the last successful call to ``KVM_ENABLE_CAP(KVM_CAP_SW_TLB)``, 2155rounded up to the nearest multiple of 64. 2156 2157Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 2158array. 2159 2160The array is little-endian: the bit 0 is the least significant bit of the 2161first byte, bit 8 is the least significant bit of the second byte, etc. 2162This avoids any complications with differing word sizes. 2163 2164The "num_dirty" field is a performance hint for KVM to determine whether it 2165should skip processing the bitmap and just invalidate everything. It must 2166be set to the number of set bits in the bitmap. 2167 2168 21694.62 KVM_CREATE_SPAPR_TCE 2170------------------------- 2171 2172:Capability: KVM_CAP_SPAPR_TCE 2173:Architectures: powerpc 2174:Type: vm ioctl 2175:Parameters: struct kvm_create_spapr_tce (in) 2176:Returns: file descriptor for manipulating the created TCE table 2177 2178This creates a virtual TCE (translation control entry) table, which 2179is an IOMMU for PAPR-style virtual I/O. It is used to translate 2180logical addresses used in virtual I/O into guest physical addresses, 2181and provides a scatter/gather capability for PAPR virtual I/O. 2182 2183:: 2184 2185 /* for KVM_CAP_SPAPR_TCE */ 2186 struct kvm_create_spapr_tce { 2187 __u64 liobn; 2188 __u32 window_size; 2189 }; 2190 2191The liobn field gives the logical IO bus number for which to create a 2192TCE table. The window_size field specifies the size of the DMA window 2193which this TCE table will translate - the table will contain one 64 2194bit TCE entry for every 4kiB of the DMA window. 2195 2196When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 2197table has been created using this ioctl(), the kernel will handle it 2198in real mode, updating the TCE table. H_PUT_TCE calls for other 2199liobns will cause a vm exit and must be handled by userspace. 2200 2201The return value is a file descriptor which can be passed to mmap(2) 2202to map the created TCE table into userspace. This lets userspace read 2203the entries written by kernel-handled H_PUT_TCE calls, and also lets 2204userspace update the TCE table directly which is useful in some 2205circumstances. 2206 2207 22084.64 KVM_NMI 2209------------ 2210 2211:Capability: KVM_CAP_USER_NMI 2212:Architectures: x86 2213:Type: vcpu ioctl 2214:Parameters: none 2215:Returns: 0 on success, -1 on error 2216 2217Queues an NMI on the thread's vcpu. Note this is well defined only 2218when KVM_CREATE_IRQCHIP has not been called, since this is an interface 2219between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 2220has been called, this interface is completely emulated within the kernel. 2221 2222To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 2223following algorithm: 2224 2225 - pause the vcpu 2226 - read the local APIC's state (KVM_GET_LAPIC) 2227 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 2228 - if so, issue KVM_NMI 2229 - resume the vcpu 2230 2231Some guests configure the LINT1 NMI input to cause a panic, aiding in 2232debugging. 2233 2234 22354.65 KVM_S390_UCAS_MAP 2236---------------------- 2237 2238:Capability: KVM_CAP_S390_UCONTROL 2239:Architectures: s390 2240:Type: vcpu ioctl 2241:Parameters: struct kvm_s390_ucas_mapping (in) 2242:Returns: 0 in case of success 2243 2244The parameter is defined like this:: 2245 2246 struct kvm_s390_ucas_mapping { 2247 __u64 user_addr; 2248 __u64 vcpu_addr; 2249 __u64 length; 2250 }; 2251 2252This ioctl maps the memory at "user_addr" with the length "length" to 2253the vcpu's address space starting at "vcpu_addr". All parameters need to 2254be aligned by 1 megabyte. 2255 2256 22574.66 KVM_S390_UCAS_UNMAP 2258------------------------ 2259 2260:Capability: KVM_CAP_S390_UCONTROL 2261:Architectures: s390 2262:Type: vcpu ioctl 2263:Parameters: struct kvm_s390_ucas_mapping (in) 2264:Returns: 0 in case of success 2265 2266The parameter is defined like this:: 2267 2268 struct kvm_s390_ucas_mapping { 2269 __u64 user_addr; 2270 __u64 vcpu_addr; 2271 __u64 length; 2272 }; 2273 2274This ioctl unmaps the memory in the vcpu's address space starting at 2275"vcpu_addr" with the length "length". The field "user_addr" is ignored. 2276All parameters need to be aligned by 1 megabyte. 2277 2278 22794.67 KVM_S390_VCPU_FAULT 2280------------------------ 2281 2282:Capability: KVM_CAP_S390_UCONTROL 2283:Architectures: s390 2284:Type: vcpu ioctl 2285:Parameters: vcpu absolute address (in) 2286:Returns: 0 in case of success 2287 2288This call creates a page table entry on the virtual cpu's address space 2289(for user controlled virtual machines) or the virtual machine's address 2290space (for regular virtual machines). This only works for minor faults, 2291thus it's recommended to access subject memory page via the user page 2292table upfront. This is useful to handle validity intercepts for user 2293controlled virtual machines to fault in the virtual cpu's lowcore pages 2294prior to calling the KVM_RUN ioctl. 2295 2296 22974.68 KVM_SET_ONE_REG 2298-------------------- 2299 2300:Capability: KVM_CAP_ONE_REG 2301:Architectures: all 2302:Type: vcpu ioctl 2303:Parameters: struct kvm_one_reg (in) 2304:Returns: 0 on success, negative value on failure 2305 2306Errors: 2307 2308 ====== ============================================================ 2309 ENOENT no such register 2310 EINVAL invalid register ID, or no such register or used with VMs in 2311 protected virtualization mode on s390 2312 EPERM (arm64) register access not allowed before vcpu finalization 2313 EBUSY (riscv) changing register value not allowed after the vcpu 2314 has run at least once 2315 ====== ============================================================ 2316 2317(These error codes are indicative only: do not rely on a specific error 2318code being returned in a specific situation.) 2319 2320:: 2321 2322 struct kvm_one_reg { 2323 __u64 id; 2324 __u64 addr; 2325 }; 2326 2327Using this ioctl, a single vcpu register can be set to a specific value 2328defined by user space with the passed in struct kvm_one_reg, where id 2329refers to the register identifier as described below and addr is a pointer 2330to a variable with the respective size. There can be architecture agnostic 2331and architecture specific registers. Each have their own range of operation 2332and their own constants and width. To keep track of the implemented 2333registers, find a list below: 2334 2335 ======= =============================== ============ 2336 Arch Register Width (bits) 2337 ======= =============================== ============ 2338 PPC KVM_REG_PPC_HIOR 64 2339 PPC KVM_REG_PPC_IAC1 64 2340 PPC KVM_REG_PPC_IAC2 64 2341 PPC KVM_REG_PPC_IAC3 64 2342 PPC KVM_REG_PPC_IAC4 64 2343 PPC KVM_REG_PPC_DAC1 64 2344 PPC KVM_REG_PPC_DAC2 64 2345 PPC KVM_REG_PPC_DABR 64 2346 PPC KVM_REG_PPC_DSCR 64 2347 PPC KVM_REG_PPC_PURR 64 2348 PPC KVM_REG_PPC_SPURR 64 2349 PPC KVM_REG_PPC_DAR 64 2350 PPC KVM_REG_PPC_DSISR 32 2351 PPC KVM_REG_PPC_AMR 64 2352 PPC KVM_REG_PPC_UAMOR 64 2353 PPC KVM_REG_PPC_MMCR0 64 2354 PPC KVM_REG_PPC_MMCR1 64 2355 PPC KVM_REG_PPC_MMCRA 64 2356 PPC KVM_REG_PPC_MMCR2 64 2357 PPC KVM_REG_PPC_MMCRS 64 2358 PPC KVM_REG_PPC_MMCR3 64 2359 PPC KVM_REG_PPC_SIAR 64 2360 PPC KVM_REG_PPC_SDAR 64 2361 PPC KVM_REG_PPC_SIER 64 2362 PPC KVM_REG_PPC_SIER2 64 2363 PPC KVM_REG_PPC_SIER3 64 2364 PPC KVM_REG_PPC_PMC1 32 2365 PPC KVM_REG_PPC_PMC2 32 2366 PPC KVM_REG_PPC_PMC3 32 2367 PPC KVM_REG_PPC_PMC4 32 2368 PPC KVM_REG_PPC_PMC5 32 2369 PPC KVM_REG_PPC_PMC6 32 2370 PPC KVM_REG_PPC_PMC7 32 2371 PPC KVM_REG_PPC_PMC8 32 2372 PPC KVM_REG_PPC_FPR0 64 2373 ... 2374 PPC KVM_REG_PPC_FPR31 64 2375 PPC KVM_REG_PPC_VR0 128 2376 ... 2377 PPC KVM_REG_PPC_VR31 128 2378 PPC KVM_REG_PPC_VSR0 128 2379 ... 2380 PPC KVM_REG_PPC_VSR31 128 2381 PPC KVM_REG_PPC_FPSCR 64 2382 PPC KVM_REG_PPC_VSCR 32 2383 PPC KVM_REG_PPC_VPA_ADDR 64 2384 PPC KVM_REG_PPC_VPA_SLB 128 2385 PPC KVM_REG_PPC_VPA_DTL 128 2386 PPC KVM_REG_PPC_EPCR 32 2387 PPC KVM_REG_PPC_EPR 32 2388 PPC KVM_REG_PPC_TCR 32 2389 PPC KVM_REG_PPC_TSR 32 2390 PPC KVM_REG_PPC_OR_TSR 32 2391 PPC KVM_REG_PPC_CLEAR_TSR 32 2392 PPC KVM_REG_PPC_MAS0 32 2393 PPC KVM_REG_PPC_MAS1 32 2394 PPC KVM_REG_PPC_MAS2 64 2395 PPC KVM_REG_PPC_MAS7_3 64 2396 PPC KVM_REG_PPC_MAS4 32 2397 PPC KVM_REG_PPC_MAS6 32 2398 PPC KVM_REG_PPC_MMUCFG 32 2399 PPC KVM_REG_PPC_TLB0CFG 32 2400 PPC KVM_REG_PPC_TLB1CFG 32 2401 PPC KVM_REG_PPC_TLB2CFG 32 2402 PPC KVM_REG_PPC_TLB3CFG 32 2403 PPC KVM_REG_PPC_TLB0PS 32 2404 PPC KVM_REG_PPC_TLB1PS 32 2405 PPC KVM_REG_PPC_TLB2PS 32 2406 PPC KVM_REG_PPC_TLB3PS 32 2407 PPC KVM_REG_PPC_EPTCFG 32 2408 PPC KVM_REG_PPC_ICP_STATE 64 2409 PPC KVM_REG_PPC_VP_STATE 128 2410 PPC KVM_REG_PPC_TB_OFFSET 64 2411 PPC KVM_REG_PPC_SPMC1 32 2412 PPC KVM_REG_PPC_SPMC2 32 2413 PPC KVM_REG_PPC_IAMR 64 2414 PPC KVM_REG_PPC_TFHAR 64 2415 PPC KVM_REG_PPC_TFIAR 64 2416 PPC KVM_REG_PPC_TEXASR 64 2417 PPC KVM_REG_PPC_FSCR 64 2418 PPC KVM_REG_PPC_PSPB 32 2419 PPC KVM_REG_PPC_EBBHR 64 2420 PPC KVM_REG_PPC_EBBRR 64 2421 PPC KVM_REG_PPC_BESCR 64 2422 PPC KVM_REG_PPC_TAR 64 2423 PPC KVM_REG_PPC_DPDES 64 2424 PPC KVM_REG_PPC_DAWR 64 2425 PPC KVM_REG_PPC_DAWRX 64 2426 PPC KVM_REG_PPC_CIABR 64 2427 PPC KVM_REG_PPC_IC 64 2428 PPC KVM_REG_PPC_VTB 64 2429 PPC KVM_REG_PPC_CSIGR 64 2430 PPC KVM_REG_PPC_TACR 64 2431 PPC KVM_REG_PPC_TCSCR 64 2432 PPC KVM_REG_PPC_PID 64 2433 PPC KVM_REG_PPC_ACOP 64 2434 PPC KVM_REG_PPC_VRSAVE 32 2435 PPC KVM_REG_PPC_LPCR 32 2436 PPC KVM_REG_PPC_LPCR_64 64 2437 PPC KVM_REG_PPC_PPR 64 2438 PPC KVM_REG_PPC_ARCH_COMPAT 32 2439 PPC KVM_REG_PPC_DABRX 32 2440 PPC KVM_REG_PPC_WORT 64 2441 PPC KVM_REG_PPC_SPRG9 64 2442 PPC KVM_REG_PPC_DBSR 32 2443 PPC KVM_REG_PPC_TIDR 64 2444 PPC KVM_REG_PPC_PSSCR 64 2445 PPC KVM_REG_PPC_DEC_EXPIRY 64 2446 PPC KVM_REG_PPC_PTCR 64 2447 PPC KVM_REG_PPC_HASHKEYR 64 2448 PPC KVM_REG_PPC_HASHPKEYR 64 2449 PPC KVM_REG_PPC_DAWR1 64 2450 PPC KVM_REG_PPC_DAWRX1 64 2451 PPC KVM_REG_PPC_DEXCR 64 2452 PPC KVM_REG_PPC_TM_GPR0 64 2453 ... 2454 PPC KVM_REG_PPC_TM_GPR31 64 2455 PPC KVM_REG_PPC_TM_VSR0 128 2456 ... 2457 PPC KVM_REG_PPC_TM_VSR63 128 2458 PPC KVM_REG_PPC_TM_CR 64 2459 PPC KVM_REG_PPC_TM_LR 64 2460 PPC KVM_REG_PPC_TM_CTR 64 2461 PPC KVM_REG_PPC_TM_FPSCR 64 2462 PPC KVM_REG_PPC_TM_AMR 64 2463 PPC KVM_REG_PPC_TM_PPR 64 2464 PPC KVM_REG_PPC_TM_VRSAVE 64 2465 PPC KVM_REG_PPC_TM_VSCR 32 2466 PPC KVM_REG_PPC_TM_DSCR 64 2467 PPC KVM_REG_PPC_TM_TAR 64 2468 PPC KVM_REG_PPC_TM_XER 64 2469 2470 MIPS KVM_REG_MIPS_R0 64 2471 ... 2472 MIPS KVM_REG_MIPS_R31 64 2473 MIPS KVM_REG_MIPS_HI 64 2474 MIPS KVM_REG_MIPS_LO 64 2475 MIPS KVM_REG_MIPS_PC 64 2476 MIPS KVM_REG_MIPS_CP0_INDEX 32 2477 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2478 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2479 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2480 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2481 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2482 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2483 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2484 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2485 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2486 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2487 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2488 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2489 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2490 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2491 MIPS KVM_REG_MIPS_CP0_WIRED 32 2492 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2493 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2494 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2495 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2496 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2497 MIPS KVM_REG_MIPS_CP0_COUNT 32 2498 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2499 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2500 MIPS KVM_REG_MIPS_CP0_STATUS 32 2501 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2502 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2503 MIPS KVM_REG_MIPS_CP0_EPC 64 2504 MIPS KVM_REG_MIPS_CP0_PRID 32 2505 MIPS KVM_REG_MIPS_CP0_EBASE 64 2506 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2507 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2508 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2509 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2510 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2511 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2512 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2513 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2514 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2515 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2516 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2517 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2518 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2519 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2520 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2521 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2522 MIPS KVM_REG_MIPS_COUNT_CTL 64 2523 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2524 MIPS KVM_REG_MIPS_COUNT_HZ 64 2525 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2526 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2527 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2528 MIPS KVM_REG_MIPS_FCR_IR 32 2529 MIPS KVM_REG_MIPS_FCR_CSR 32 2530 MIPS KVM_REG_MIPS_MSA_IR 32 2531 MIPS KVM_REG_MIPS_MSA_CSR 32 2532 ======= =============================== ============ 2533 2534ARM registers are mapped using the lower 32 bits. The upper 16 of that 2535is the register group type, or coprocessor number: 2536 2537ARM core registers have the following id bit patterns:: 2538 2539 0x4020 0000 0010 <index into the kvm_regs struct:16> 2540 2541ARM 32-bit CP15 registers have the following id bit patterns:: 2542 2543 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 2544 2545ARM 64-bit CP15 registers have the following id bit patterns:: 2546 2547 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 2548 2549ARM CCSIDR registers are demultiplexed by CSSELR value:: 2550 2551 0x4020 0000 0011 00 <csselr:8> 2552 2553ARM 32-bit VFP control registers have the following id bit patterns:: 2554 2555 0x4020 0000 0012 1 <regno:12> 2556 2557ARM 64-bit FP registers have the following id bit patterns:: 2558 2559 0x4030 0000 0012 0 <regno:12> 2560 2561ARM firmware pseudo-registers have the following bit pattern:: 2562 2563 0x4030 0000 0014 <regno:16> 2564 2565 2566arm64 registers are mapped using the lower 32 bits. The upper 16 of 2567that is the register group type, or coprocessor number: 2568 2569arm64 core/FP-SIMD registers have the following id bit patterns. Note 2570that the size of the access is variable, as the kvm_regs structure 2571contains elements ranging from 32 to 128 bits. The index is a 32bit 2572value in the kvm_regs structure seen as a 32bit array:: 2573 2574 0x60x0 0000 0010 <index into the kvm_regs struct:16> 2575 2576Specifically: 2577 2578======================= ========= ===== ======================================= 2579 Encoding Register Bits kvm_regs member 2580======================= ========= ===== ======================================= 2581 0x6030 0000 0010 0000 X0 64 regs.regs[0] 2582 0x6030 0000 0010 0002 X1 64 regs.regs[1] 2583 ... 2584 0x6030 0000 0010 003c X30 64 regs.regs[30] 2585 0x6030 0000 0010 003e SP 64 regs.sp 2586 0x6030 0000 0010 0040 PC 64 regs.pc 2587 0x6030 0000 0010 0042 PSTATE 64 regs.pstate 2588 0x6030 0000 0010 0044 SP_EL1 64 sp_el1 2589 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1 2590 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC) 2591 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT] 2592 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND] 2593 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ] 2594 0x6030 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ] 2595 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_ 2596 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_ 2597 ... 2598 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_ 2599 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr 2600 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr 2601======================= ========= ===== ======================================= 2602 2603.. [1] These encodings are not accepted for SVE-enabled vcpus. See 2604 :ref:`KVM_ARM_VCPU_INIT`. 2605 2606 The equivalent register content can be accessed via bits [127:0] of 2607 the corresponding SVE Zn registers instead for vcpus that have SVE 2608 enabled (see below). 2609 2610arm64 CCSIDR registers are demultiplexed by CSSELR value:: 2611 2612 0x6020 0000 0011 00 <csselr:8> 2613 2614arm64 system registers have the following id bit patterns:: 2615 2616 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 2617 2618.. warning:: 2619 2620 Two system register IDs do not follow the specified pattern. These 2621 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to 2622 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These 2623 two had their values accidentally swapped, which means TIMER_CVAL is 2624 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is 2625 derived from the register encoding for CNTV_CVAL_EL0. As this is 2626 API, it must remain this way. 2627 2628arm64 firmware pseudo-registers have the following bit pattern:: 2629 2630 0x6030 0000 0014 <regno:16> 2631 2632arm64 SVE registers have the following bit patterns:: 2633 2634 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice] 2635 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice] 2636 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice] 2637 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register 2638 2639Access to register IDs where 2048 * slice >= 128 * max_vq will fail with 2640ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit 2641quadwords: see [2]_ below. 2642 2643These registers are only accessible on vcpus for which SVE is enabled. 2644See KVM_ARM_VCPU_INIT for details. 2645 2646In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not 2647accessible until the vcpu's SVE configuration has been finalized 2648using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT 2649and KVM_ARM_VCPU_FINALIZE for more information about this procedure. 2650 2651KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector 2652lengths supported by the vcpu to be discovered and configured by 2653userspace. When transferred to or from user memory via KVM_GET_ONE_REG 2654or KVM_SET_ONE_REG, the value of this register is of type 2655__u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as 2656follows:: 2657 2658 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS]; 2659 2660 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2661 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >> 2662 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1)) 2663 /* Vector length vq * 16 bytes supported */ 2664 else 2665 /* Vector length vq * 16 bytes not supported */ 2666 2667.. [2] The maximum value vq for which the above condition is true is 2668 max_vq. This is the maximum vector length available to the guest on 2669 this vcpu, and determines which register slices are visible through 2670 this ioctl interface. 2671 2672(See Documentation/arch/arm64/sve.rst for an explanation of the "vq" 2673nomenclature.) 2674 2675KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT. 2676KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that 2677the host supports. 2678 2679Userspace may subsequently modify it if desired until the vcpu's SVE 2680configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). 2681 2682Apart from simply removing all vector lengths from the host set that 2683exceed some value, support for arbitrarily chosen sets of vector lengths 2684is hardware-dependent and may not be available. Attempting to configure 2685an invalid set of vector lengths via KVM_SET_ONE_REG will fail with 2686EINVAL. 2687 2688After the vcpu's SVE configuration is finalized, further attempts to 2689write this register will fail with EPERM. 2690 2691arm64 bitmap feature firmware pseudo-registers have the following bit pattern:: 2692 2693 0x6030 0000 0016 <regno:16> 2694 2695The bitmap feature firmware registers exposes the hypercall services that 2696are available for userspace to configure. The set bits corresponds to the 2697services that are available for the guests to access. By default, KVM 2698sets all the supported bits during VM initialization. The userspace can 2699discover the available services via KVM_GET_ONE_REG, and write back the 2700bitmap corresponding to the features that it wishes guests to see via 2701KVM_SET_ONE_REG. 2702 2703Note: These registers are immutable once any of the vCPUs of the VM has 2704run at least once. A KVM_SET_ONE_REG in such a scenario will return 2705a -EBUSY to userspace. 2706 2707(See Documentation/virt/kvm/arm/hypercalls.rst for more details.) 2708 2709 2710MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 2711the register group type: 2712 2713MIPS core registers (see above) have the following id bit patterns:: 2714 2715 0x7030 0000 0000 <reg:16> 2716 2717MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 2718patterns depending on whether they're 32-bit or 64-bit registers:: 2719 2720 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 2721 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2722 2723Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 2724versions of the EntryLo registers regardless of the word size of the host 2725hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 2726with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 2727the PFNX field starting at bit 30. 2728 2729MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 2730patterns:: 2731 2732 0x7030 0000 0001 01 <reg:8> 2733 2734MIPS KVM control registers (see above) have the following id bit patterns:: 2735 2736 0x7030 0000 0002 <reg:16> 2737 2738MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2739id bit patterns depending on the size of the register being accessed. They are 2740always accessed according to the current guest FPU mode (Status.FR and 2741Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2742if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2743registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2744overlap the FPU registers:: 2745 2746 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2747 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2748 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2749 2750MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2751following id bit patterns:: 2752 2753 0x7020 0000 0003 01 <0:3> <reg:5> 2754 2755MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2756following id bit patterns:: 2757 2758 0x7020 0000 0003 02 <0:3> <reg:5> 2759 2760RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of 2761that is the register group type. 2762 2763RISC-V config registers are meant for configuring a Guest VCPU and it has 2764the following id bit patterns:: 2765 2766 0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host) 2767 0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host) 2768 2769Following are the RISC-V config registers: 2770 2771======================= ========= ============================================= 2772 Encoding Register Description 2773======================= ========= ============================================= 2774 0x80x0 0000 0100 0000 isa ISA feature bitmap of Guest VCPU 2775======================= ========= ============================================= 2776 2777The isa config register can be read anytime but can only be written before 2778a Guest VCPU runs. It will have ISA feature bits matching underlying host 2779set by default. 2780 2781RISC-V core registers represent the general execution state of a Guest VCPU 2782and it has the following id bit patterns:: 2783 2784 0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host) 2785 0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host) 2786 2787Following are the RISC-V core registers: 2788 2789======================= ========= ============================================= 2790 Encoding Register Description 2791======================= ========= ============================================= 2792 0x80x0 0000 0200 0000 regs.pc Program counter 2793 0x80x0 0000 0200 0001 regs.ra Return address 2794 0x80x0 0000 0200 0002 regs.sp Stack pointer 2795 0x80x0 0000 0200 0003 regs.gp Global pointer 2796 0x80x0 0000 0200 0004 regs.tp Task pointer 2797 0x80x0 0000 0200 0005 regs.t0 Caller saved register 0 2798 0x80x0 0000 0200 0006 regs.t1 Caller saved register 1 2799 0x80x0 0000 0200 0007 regs.t2 Caller saved register 2 2800 0x80x0 0000 0200 0008 regs.s0 Callee saved register 0 2801 0x80x0 0000 0200 0009 regs.s1 Callee saved register 1 2802 0x80x0 0000 0200 000a regs.a0 Function argument (or return value) 0 2803 0x80x0 0000 0200 000b regs.a1 Function argument (or return value) 1 2804 0x80x0 0000 0200 000c regs.a2 Function argument 2 2805 0x80x0 0000 0200 000d regs.a3 Function argument 3 2806 0x80x0 0000 0200 000e regs.a4 Function argument 4 2807 0x80x0 0000 0200 000f regs.a5 Function argument 5 2808 0x80x0 0000 0200 0010 regs.a6 Function argument 6 2809 0x80x0 0000 0200 0011 regs.a7 Function argument 7 2810 0x80x0 0000 0200 0012 regs.s2 Callee saved register 2 2811 0x80x0 0000 0200 0013 regs.s3 Callee saved register 3 2812 0x80x0 0000 0200 0014 regs.s4 Callee saved register 4 2813 0x80x0 0000 0200 0015 regs.s5 Callee saved register 5 2814 0x80x0 0000 0200 0016 regs.s6 Callee saved register 6 2815 0x80x0 0000 0200 0017 regs.s7 Callee saved register 7 2816 0x80x0 0000 0200 0018 regs.s8 Callee saved register 8 2817 0x80x0 0000 0200 0019 regs.s9 Callee saved register 9 2818 0x80x0 0000 0200 001a regs.s10 Callee saved register 10 2819 0x80x0 0000 0200 001b regs.s11 Callee saved register 11 2820 0x80x0 0000 0200 001c regs.t3 Caller saved register 3 2821 0x80x0 0000 0200 001d regs.t4 Caller saved register 4 2822 0x80x0 0000 0200 001e regs.t5 Caller saved register 5 2823 0x80x0 0000 0200 001f regs.t6 Caller saved register 6 2824 0x80x0 0000 0200 0020 mode Privilege mode (1 = S-mode or 0 = U-mode) 2825======================= ========= ============================================= 2826 2827RISC-V csr registers represent the supervisor mode control/status registers 2828of a Guest VCPU and it has the following id bit patterns:: 2829 2830 0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host) 2831 0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host) 2832 2833Following are the RISC-V csr registers: 2834 2835======================= ========= ============================================= 2836 Encoding Register Description 2837======================= ========= ============================================= 2838 0x80x0 0000 0300 0000 sstatus Supervisor status 2839 0x80x0 0000 0300 0001 sie Supervisor interrupt enable 2840 0x80x0 0000 0300 0002 stvec Supervisor trap vector base 2841 0x80x0 0000 0300 0003 sscratch Supervisor scratch register 2842 0x80x0 0000 0300 0004 sepc Supervisor exception program counter 2843 0x80x0 0000 0300 0005 scause Supervisor trap cause 2844 0x80x0 0000 0300 0006 stval Supervisor bad address or instruction 2845 0x80x0 0000 0300 0007 sip Supervisor interrupt pending 2846 0x80x0 0000 0300 0008 satp Supervisor address translation and protection 2847======================= ========= ============================================= 2848 2849RISC-V timer registers represent the timer state of a Guest VCPU and it has 2850the following id bit patterns:: 2851 2852 0x8030 0000 04 <index into the kvm_riscv_timer struct:24> 2853 2854Following are the RISC-V timer registers: 2855 2856======================= ========= ============================================= 2857 Encoding Register Description 2858======================= ========= ============================================= 2859 0x8030 0000 0400 0000 frequency Time base frequency (read-only) 2860 0x8030 0000 0400 0001 time Time value visible to Guest 2861 0x8030 0000 0400 0002 compare Time compare programmed by Guest 2862 0x8030 0000 0400 0003 state Time compare state (1 = ON or 0 = OFF) 2863======================= ========= ============================================= 2864 2865RISC-V F-extension registers represent the single precision floating point 2866state of a Guest VCPU and it has the following id bit patterns:: 2867 2868 0x8020 0000 05 <index into the __riscv_f_ext_state struct:24> 2869 2870Following are the RISC-V F-extension registers: 2871 2872======================= ========= ============================================= 2873 Encoding Register Description 2874======================= ========= ============================================= 2875 0x8020 0000 0500 0000 f[0] Floating point register 0 2876 ... 2877 0x8020 0000 0500 001f f[31] Floating point register 31 2878 0x8020 0000 0500 0020 fcsr Floating point control and status register 2879======================= ========= ============================================= 2880 2881RISC-V D-extension registers represent the double precision floating point 2882state of a Guest VCPU and it has the following id bit patterns:: 2883 2884 0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr) 2885 0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr) 2886 2887Following are the RISC-V D-extension registers: 2888 2889======================= ========= ============================================= 2890 Encoding Register Description 2891======================= ========= ============================================= 2892 0x8030 0000 0600 0000 f[0] Floating point register 0 2893 ... 2894 0x8030 0000 0600 001f f[31] Floating point register 31 2895 0x8020 0000 0600 0020 fcsr Floating point control and status register 2896======================= ========= ============================================= 2897 2898LoongArch registers are mapped using the lower 32 bits. The upper 16 bits of 2899that is the register group type. 2900 2901LoongArch csr registers are used to control guest cpu or get status of guest 2902cpu, and they have the following id bit patterns:: 2903 2904 0x9030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2905 2906LoongArch KVM control registers are used to implement some new defined functions 2907such as set vcpu counter or reset vcpu, and they have the following id bit patterns:: 2908 2909 0x9030 0000 0002 <reg:16> 2910 2911x86 MSR registers have the following id bit patterns:: 2912 0x2030 0002 <msr number:32> 2913 2914Following are the KVM-defined registers for x86: 2915 2916======================= ========= ============================================= 2917 Encoding Register Description 2918======================= ========= ============================================= 2919 0x2030 0003 0000 0000 SSP Shadow Stack Pointer 2920======================= ========= ============================================= 2921 29224.69 KVM_GET_ONE_REG 2923-------------------- 2924 2925:Capability: KVM_CAP_ONE_REG 2926:Architectures: all 2927:Type: vcpu ioctl 2928:Parameters: struct kvm_one_reg (in and out) 2929:Returns: 0 on success, negative value on failure 2930 2931Errors include: 2932 2933 ======== ============================================================ 2934 ENOENT no such register 2935 EINVAL invalid register ID, or no such register or used with VMs in 2936 protected virtualization mode on s390 2937 EPERM (arm64) register access not allowed before vcpu finalization 2938 ======== ============================================================ 2939 2940(These error codes are indicative only: do not rely on a specific error 2941code being returned in a specific situation.) 2942 2943This ioctl allows to receive the value of a single register implemented 2944in a vcpu. The register to read is indicated by the "id" field of the 2945kvm_one_reg struct passed in. On success, the register value can be found 2946at the memory location pointed to by "addr". 2947 2948The list of registers accessible using this interface is identical to the 2949list in 4.68. 2950 2951 29524.70 KVM_KVMCLOCK_CTRL 2953---------------------- 2954 2955:Capability: KVM_CAP_KVMCLOCK_CTRL 2956:Architectures: Any that implement pvclocks (currently x86 only) 2957:Type: vcpu ioctl 2958:Parameters: None 2959:Returns: 0 on success, -1 on error 2960 2961This ioctl sets a flag accessible to the guest indicating that the specified 2962vCPU has been paused by the host userspace. 2963 2964The host will set a flag in the pvclock structure that is checked from the 2965soft lockup watchdog. The flag is part of the pvclock structure that is 2966shared between guest and host, specifically the second bit of the flags 2967field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2968the host and read/cleared exclusively by the guest. The guest operation of 2969checking and clearing the flag must be an atomic operation so 2970load-link/store-conditional, or equivalent must be used. There are two cases 2971where the guest will clear the flag: when the soft lockup watchdog timer resets 2972itself or when a soft lockup is detected. This ioctl can be called any time 2973after pausing the vcpu, but before it is resumed. 2974 2975 29764.71 KVM_SIGNAL_MSI 2977------------------- 2978 2979:Capability: KVM_CAP_SIGNAL_MSI 2980:Architectures: x86 arm64 2981:Type: vm ioctl 2982:Parameters: struct kvm_msi (in) 2983:Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2984 2985Directly inject a MSI message. Only valid with in-kernel irqchip that handles 2986MSI messages. 2987 2988:: 2989 2990 struct kvm_msi { 2991 __u32 address_lo; 2992 __u32 address_hi; 2993 __u32 data; 2994 __u32 flags; 2995 __u32 devid; 2996 __u8 pad[12]; 2997 }; 2998 2999flags: 3000 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 3001 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 3002 the device ID. If this capability is not available, userspace 3003 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 3004 3005If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 3006for the device that wrote the MSI message. For PCI, this is usually a 3007BDF identifier in the lower 16 bits. 3008 3009On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 3010feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 3011address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 3012address_hi must be zero. 3013 3014 30154.71 KVM_CREATE_PIT2 3016-------------------- 3017 3018:Capability: KVM_CAP_PIT2 3019:Architectures: x86 3020:Type: vm ioctl 3021:Parameters: struct kvm_pit_config (in) 3022:Returns: 0 on success, -1 on error 3023 3024Creates an in-kernel device model for the i8254 PIT. This call is only valid 3025after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 3026parameters have to be passed:: 3027 3028 struct kvm_pit_config { 3029 __u32 flags; 3030 __u32 pad[15]; 3031 }; 3032 3033Valid flags are:: 3034 3035 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 3036 3037PIT timer interrupts may use a per-VM kernel thread for injection. If it 3038exists, this thread will have a name of the following pattern:: 3039 3040 kvm-pit/<owner-process-pid> 3041 3042When running a guest with elevated priorities, the scheduling parameters of 3043this thread may have to be adjusted accordingly. 3044 3045This IOCTL replaces the obsolete KVM_CREATE_PIT. 3046 3047 30484.72 KVM_GET_PIT2 3049----------------- 3050 3051:Capability: KVM_CAP_PIT_STATE2 3052:Architectures: x86 3053:Type: vm ioctl 3054:Parameters: struct kvm_pit_state2 (out) 3055:Returns: 0 on success, -1 on error 3056 3057Retrieves the state of the in-kernel PIT model. Only valid after 3058KVM_CREATE_PIT2. The state is returned in the following structure:: 3059 3060 struct kvm_pit_state2 { 3061 struct kvm_pit_channel_state channels[3]; 3062 __u32 flags; 3063 __u32 reserved[9]; 3064 }; 3065 3066Valid flags are:: 3067 3068 /* disable PIT in HPET legacy mode */ 3069 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 3070 /* speaker port data bit enabled */ 3071 #define KVM_PIT_FLAGS_SPEAKER_DATA_ON 0x00000002 3072 3073This IOCTL replaces the obsolete KVM_GET_PIT. 3074 3075 30764.73 KVM_SET_PIT2 3077----------------- 3078 3079:Capability: KVM_CAP_PIT_STATE2 3080:Architectures: x86 3081:Type: vm ioctl 3082:Parameters: struct kvm_pit_state2 (in) 3083:Returns: 0 on success, -1 on error 3084 3085Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 3086See KVM_GET_PIT2 for details on struct kvm_pit_state2. 3087 3088.. Tip:: 3089 ``KVM_SET_PIT2`` strictly adheres to the spec of Intel 8254 PIT. For example, 3090 a ``count`` value of 0 in ``struct kvm_pit_channel_state`` is interpreted as 3091 65536, which is the maximum count value. Refer to `Intel 8254 programmable 3092 interval timer <https://www.scs.stanford.edu/10wi-cs140/pintos/specs/8254.pdf>`_. 3093 3094This IOCTL replaces the obsolete KVM_SET_PIT. 3095 3096 30974.74 KVM_PPC_GET_SMMU_INFO 3098-------------------------- 3099 3100:Capability: KVM_CAP_PPC_GET_SMMU_INFO 3101:Architectures: powerpc 3102:Type: vm ioctl 3103:Parameters: None 3104:Returns: 0 on success, -1 on error 3105 3106This populates and returns a structure describing the features of 3107the "Server" class MMU emulation supported by KVM. 3108This can in turn be used by userspace to generate the appropriate 3109device-tree properties for the guest operating system. 3110 3111The structure contains some global information, followed by an 3112array of supported segment page sizes:: 3113 3114 struct kvm_ppc_smmu_info { 3115 __u64 flags; 3116 __u32 slb_size; 3117 __u32 pad; 3118 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3119 }; 3120 3121The supported flags are: 3122 3123 - KVM_PPC_PAGE_SIZES_REAL: 3124 When that flag is set, guest page sizes must "fit" the backing 3125 store page sizes. When not set, any page size in the list can 3126 be used regardless of how they are backed by userspace. 3127 3128 - KVM_PPC_1T_SEGMENTS 3129 The emulated MMU supports 1T segments in addition to the 3130 standard 256M ones. 3131 3132 - KVM_PPC_NO_HASH 3133 This flag indicates that HPT guests are not supported by KVM, 3134 thus all guests must use radix MMU mode. 3135 3136The "slb_size" field indicates how many SLB entries are supported 3137 3138The "sps" array contains 8 entries indicating the supported base 3139page sizes for a segment in increasing order. Each entry is defined 3140as follow:: 3141 3142 struct kvm_ppc_one_seg_page_size { 3143 __u32 page_shift; /* Base page shift of segment (or 0) */ 3144 __u32 slb_enc; /* SLB encoding for BookS */ 3145 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3146 }; 3147 3148An entry with a "page_shift" of 0 is unused. Because the array is 3149organized in increasing order, a lookup can stop when encountering 3150such an entry. 3151 3152The "slb_enc" field provides the encoding to use in the SLB for the 3153page size. The bits are in positions such as the value can directly 3154be OR'ed into the "vsid" argument of the slbmte instruction. 3155 3156The "enc" array is a list which for each of those segment base page 3157size provides the list of supported actual page sizes (which can be 3158only larger or equal to the base page size), along with the 3159corresponding encoding in the hash PTE. Similarly, the array is 31608 entries sorted by increasing sizes and an entry with a "0" shift 3161is an empty entry and a terminator:: 3162 3163 struct kvm_ppc_one_page_size { 3164 __u32 page_shift; /* Page shift (or 0) */ 3165 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 3166 }; 3167 3168The "pte_enc" field provides a value that can OR'ed into the hash 3169PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 3170into the hash PTE second double word). 3171 31724.75 KVM_IRQFD 3173-------------- 3174 3175:Capability: KVM_CAP_IRQFD 3176:Architectures: x86 s390 arm64 3177:Type: vm ioctl 3178:Parameters: struct kvm_irqfd (in) 3179:Returns: 0 on success, -1 on error 3180 3181Allows setting an eventfd to directly trigger a guest interrupt. 3182kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 3183kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 3184an event is triggered on the eventfd, an interrupt is injected into 3185the guest using the specified gsi pin. The irqfd is removed using 3186the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 3187and kvm_irqfd.gsi. 3188 3189With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 3190mechanism allowing emulation of level-triggered, irqfd-based 3191interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 3192additional eventfd in the kvm_irqfd.resamplefd field. When operating 3193in resample mode, posting of an interrupt through kvm_irq.fd asserts 3194the specified gsi in the irqchip. When the irqchip is resampled, such 3195as from an EOI, the gsi is de-asserted and the user is notified via 3196kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 3197the interrupt if the device making use of it still requires service. 3198Note that closing the resamplefd is not sufficient to disable the 3199irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 3200and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 3201 3202On arm64, gsi routing being supported, the following can happen: 3203 3204- in case no routing entry is associated to this gsi, injection fails 3205- in case the gsi is associated to an irqchip routing entry, 3206 irqchip.pin + 32 corresponds to the injected SPI ID. 3207- in case the gsi is associated to an MSI routing entry, the MSI 3208 message and device ID are translated into an LPI (support restricted 3209 to GICv3 ITS in-kernel emulation). 3210 32114.76 KVM_PPC_ALLOCATE_HTAB 3212-------------------------- 3213 3214:Capability: KVM_CAP_PPC_ALLOC_HTAB 3215:Architectures: powerpc 3216:Type: vm ioctl 3217:Parameters: Pointer to u32 containing hash table order (in/out) 3218:Returns: 0 on success, -1 on error 3219 3220This requests the host kernel to allocate an MMU hash table for a 3221guest using the PAPR paravirtualization interface. This only does 3222anything if the kernel is configured to use the Book 3S HV style of 3223virtualization. Otherwise the capability doesn't exist and the ioctl 3224returns an ENOTTY error. The rest of this description assumes Book 3S 3225HV. 3226 3227There must be no vcpus running when this ioctl is called; if there 3228are, it will do nothing and return an EBUSY error. 3229 3230The parameter is a pointer to a 32-bit unsigned integer variable 3231containing the order (log base 2) of the desired size of the hash 3232table, which must be between 18 and 46. On successful return from the 3233ioctl, the value will not be changed by the kernel. 3234 3235If no hash table has been allocated when any vcpu is asked to run 3236(with the KVM_RUN ioctl), the host kernel will allocate a 3237default-sized hash table (16 MB). 3238 3239If this ioctl is called when a hash table has already been allocated, 3240with a different order from the existing hash table, the existing hash 3241table will be freed and a new one allocated. If this is ioctl is 3242called when a hash table has already been allocated of the same order 3243as specified, the kernel will clear out the existing hash table (zero 3244all HPTEs). In either case, if the guest is using the virtualized 3245real-mode area (VRMA) facility, the kernel will re-create the VMRA 3246HPTEs on the next KVM_RUN of any vcpu. 3247 32484.77 KVM_S390_INTERRUPT 3249----------------------- 3250 3251:Capability: basic 3252:Architectures: s390 3253:Type: vm ioctl, vcpu ioctl 3254:Parameters: struct kvm_s390_interrupt (in) 3255:Returns: 0 on success, -1 on error 3256 3257Allows to inject an interrupt to the guest. Interrupts can be floating 3258(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 3259 3260Interrupt parameters are passed via kvm_s390_interrupt:: 3261 3262 struct kvm_s390_interrupt { 3263 __u32 type; 3264 __u32 parm; 3265 __u64 parm64; 3266 }; 3267 3268type can be one of the following: 3269 3270KVM_S390_SIGP_STOP (vcpu) 3271 - sigp stop; optional flags in parm 3272KVM_S390_PROGRAM_INT (vcpu) 3273 - program check; code in parm 3274KVM_S390_SIGP_SET_PREFIX (vcpu) 3275 - sigp set prefix; prefix address in parm 3276KVM_S390_RESTART (vcpu) 3277 - restart 3278KVM_S390_INT_CLOCK_COMP (vcpu) 3279 - clock comparator interrupt 3280KVM_S390_INT_CPU_TIMER (vcpu) 3281 - CPU timer interrupt 3282KVM_S390_INT_VIRTIO (vm) 3283 - virtio external interrupt; external interrupt 3284 parameters in parm and parm64 3285KVM_S390_INT_SERVICE (vm) 3286 - sclp external interrupt; sclp parameter in parm 3287KVM_S390_INT_EMERGENCY (vcpu) 3288 - sigp emergency; source cpu in parm 3289KVM_S390_INT_EXTERNAL_CALL (vcpu) 3290 - sigp external call; source cpu in parm 3291KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 3292 - compound value to indicate an 3293 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 3294 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 3295 interruption subclass) 3296KVM_S390_MCHK (vm, vcpu) 3297 - machine check interrupt; cr 14 bits in parm, machine check interrupt 3298 code in parm64 (note that machine checks needing further payload are not 3299 supported by this ioctl) 3300 3301This is an asynchronous vcpu ioctl and can be invoked from any thread. 3302 33034.78 KVM_PPC_GET_HTAB_FD 3304------------------------ 3305 3306:Capability: KVM_CAP_PPC_HTAB_FD 3307:Architectures: powerpc 3308:Type: vm ioctl 3309:Parameters: Pointer to struct kvm_get_htab_fd (in) 3310:Returns: file descriptor number (>= 0) on success, -1 on error 3311 3312This returns a file descriptor that can be used either to read out the 3313entries in the guest's hashed page table (HPT), or to write entries to 3314initialize the HPT. The returned fd can only be written to if the 3315KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 3316can only be read if that bit is clear. The argument struct looks like 3317this:: 3318 3319 /* For KVM_PPC_GET_HTAB_FD */ 3320 struct kvm_get_htab_fd { 3321 __u64 flags; 3322 __u64 start_index; 3323 __u64 reserved[2]; 3324 }; 3325 3326 /* Values for kvm_get_htab_fd.flags */ 3327 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 3328 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 3329 3330The 'start_index' field gives the index in the HPT of the entry at 3331which to start reading. It is ignored when writing. 3332 3333Reads on the fd will initially supply information about all 3334"interesting" HPT entries. Interesting entries are those with the 3335bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 3336all entries. When the end of the HPT is reached, the read() will 3337return. If read() is called again on the fd, it will start again from 3338the beginning of the HPT, but will only return HPT entries that have 3339changed since they were last read. 3340 3341Data read or written is structured as a header (8 bytes) followed by a 3342series of valid HPT entries (16 bytes) each. The header indicates how 3343many valid HPT entries there are and how many invalid entries follow 3344the valid entries. The invalid entries are not represented explicitly 3345in the stream. The header format is:: 3346 3347 struct kvm_get_htab_header { 3348 __u32 index; 3349 __u16 n_valid; 3350 __u16 n_invalid; 3351 }; 3352 3353Writes to the fd create HPT entries starting at the index given in the 3354header; first 'n_valid' valid entries with contents from the data 3355written, then 'n_invalid' invalid entries, invalidating any previously 3356valid entries found. 3357 33584.79 KVM_CREATE_DEVICE 3359---------------------- 3360 3361:Capability: KVM_CAP_DEVICE_CTRL 3362:Architectures: all 3363:Type: vm ioctl 3364:Parameters: struct kvm_create_device (in/out) 3365:Returns: 0 on success, -1 on error 3366 3367Errors: 3368 3369 ====== ======================================================= 3370 ENODEV The device type is unknown or unsupported 3371 EEXIST Device already created, and this type of device may not 3372 be instantiated multiple times 3373 ====== ======================================================= 3374 3375 Other error conditions may be defined by individual device types or 3376 have their standard meanings. 3377 3378Creates an emulated device in the kernel. The file descriptor returned 3379in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 3380 3381If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 3382device type is supported (not necessarily whether it can be created 3383in the current vm). 3384 3385Individual devices should not define flags. Attributes should be used 3386for specifying any behavior that is not implied by the device type 3387number. 3388 3389:: 3390 3391 struct kvm_create_device { 3392 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 3393 __u32 fd; /* out: device handle */ 3394 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 3395 }; 3396 33974.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3398-------------------------------------------- 3399 3400:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3401 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3402 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device (no set) 3403:Architectures: x86, arm64, s390 3404:Type: device ioctl, vm ioctl, vcpu ioctl 3405:Parameters: struct kvm_device_attr 3406:Returns: 0 on success, -1 on error 3407 3408Errors: 3409 3410 ===== ============================================================= 3411 ENXIO The group or attribute is unknown/unsupported for this device 3412 or hardware support is missing. 3413 EPERM The attribute cannot (currently) be accessed this way 3414 (e.g. read-only attribute, or attribute that only makes 3415 sense when the device is in a different state) 3416 ===== ============================================================= 3417 3418 Other error conditions may be defined by individual device types. 3419 3420Gets/sets a specified piece of device configuration and/or state. The 3421semantics are device-specific. See individual device documentation in 3422the "devices" directory. As with ONE_REG, the size of the data 3423transferred is defined by the particular attribute. 3424 3425:: 3426 3427 struct kvm_device_attr { 3428 __u32 flags; /* no flags currently defined */ 3429 __u32 group; /* device-defined */ 3430 __u64 attr; /* group-defined */ 3431 __u64 addr; /* userspace address of attr data */ 3432 }; 3433 34344.81 KVM_HAS_DEVICE_ATTR 3435------------------------ 3436 3437:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3438 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3439 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device 3440:Type: device ioctl, vm ioctl, vcpu ioctl 3441:Parameters: struct kvm_device_attr 3442:Returns: 0 on success, -1 on error 3443 3444Errors: 3445 3446 ===== ============================================================= 3447 ENXIO The group or attribute is unknown/unsupported for this device 3448 or hardware support is missing. 3449 ===== ============================================================= 3450 3451Tests whether a device supports a particular attribute. A successful 3452return indicates the attribute is implemented. It does not necessarily 3453indicate that the attribute can be read or written in the device's 3454current state. "addr" is ignored. 3455 3456.. _KVM_ARM_VCPU_INIT: 3457 34584.82 KVM_ARM_VCPU_INIT 3459---------------------- 3460 3461:Capability: basic 3462:Architectures: arm64 3463:Type: vcpu ioctl 3464:Parameters: struct kvm_vcpu_init (in) 3465:Returns: 0 on success; -1 on error 3466 3467Errors: 3468 3469 ====== ================================================================= 3470 EINVAL the target is unknown, or the combination of features is invalid. 3471 ENOENT a features bit specified is unknown. 3472 ====== ================================================================= 3473 3474This tells KVM what type of CPU to present to the guest, and what 3475optional features it should have. This will cause a reset of the cpu 3476registers to their initial values. If this is not called, KVM_RUN will 3477return ENOEXEC for that vcpu. 3478 3479The initial values are defined as: 3480 - Processor state: 3481 * AArch64: EL1h, D, A, I and F bits set. All other bits 3482 are cleared. 3483 * AArch32: SVC, A, I and F bits set. All other bits are 3484 cleared. 3485 - General Purpose registers, including PC and SP: set to 0 3486 - FPSIMD/NEON registers: set to 0 3487 - SVE registers: set to 0 3488 - System registers: Reset to their architecturally defined 3489 values as for a warm reset to EL1 (resp. SVC) or EL2 (in the 3490 case of EL2 being enabled). 3491 3492Note that because some registers reflect machine topology, all vcpus 3493should be created before this ioctl is invoked. 3494 3495Userspace can call this function multiple times for a given vcpu, including 3496after the vcpu has been run. This will reset the vcpu to its initial 3497state. All calls to this function after the initial call must use the same 3498target and same set of feature flags, otherwise EINVAL will be returned. 3499 3500Possible features: 3501 3502 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 3503 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 3504 and execute guest code when KVM_RUN is called. 3505 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 3506 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 3507 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision 3508 backward compatible with v0.2) for the CPU. 3509 Depends on KVM_CAP_ARM_PSCI_0_2. 3510 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 3511 Depends on KVM_CAP_ARM_PMU_V3. 3512 3513 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication 3514 for arm64 only. 3515 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS. 3516 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3517 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3518 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3519 requested. 3520 3521 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication 3522 for arm64 only. 3523 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC. 3524 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3525 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3526 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3527 requested. 3528 3529 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only). 3530 Depends on KVM_CAP_ARM_SVE. 3531 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3532 3533 * After KVM_ARM_VCPU_INIT: 3534 3535 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the 3536 initial value of this pseudo-register indicates the best set of 3537 vector lengths possible for a vcpu on this host. 3538 3539 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3540 3541 - KVM_RUN and KVM_GET_REG_LIST are not available; 3542 3543 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access 3544 the scalable architectural SVE registers 3545 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or 3546 KVM_REG_ARM64_SVE_FFR; 3547 3548 - KVM_REG_ARM64_SVE_VLS may optionally be written using 3549 KVM_SET_ONE_REG, to modify the set of vector lengths available 3550 for the vcpu. 3551 3552 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3553 3554 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can 3555 no longer be written using KVM_SET_ONE_REG. 3556 3557 - KVM_ARM_VCPU_HAS_EL2: Enable Nested Virtualisation support, 3558 booting the guest from EL2 instead of EL1. 3559 Depends on KVM_CAP_ARM_EL2. 3560 The VM is running with HCR_EL2.E2H being RES1 (VHE) unless 3561 KVM_ARM_VCPU_HAS_EL2_E2H0 is also set. 3562 3563 - KVM_ARM_VCPU_HAS_EL2_E2H0: Restrict Nested Virtualisation 3564 support to HCR_EL2.E2H being RES0 (non-VHE). 3565 Depends on KVM_CAP_ARM_EL2_E2H0. 3566 KVM_ARM_VCPU_HAS_EL2 must also be set. 3567 35684.83 KVM_ARM_PREFERRED_TARGET 3569----------------------------- 3570 3571:Capability: basic 3572:Architectures: arm64 3573:Type: vm ioctl 3574:Parameters: struct kvm_vcpu_init (out) 3575:Returns: 0 on success; -1 on error 3576 3577Errors: 3578 3579 ====== ========================================== 3580 ENODEV no preferred target available for the host 3581 ====== ========================================== 3582 3583This queries KVM for preferred CPU target type which can be emulated 3584by KVM on underlying host. 3585 3586The ioctl returns struct kvm_vcpu_init instance containing information 3587about preferred CPU target type and recommended features for it. The 3588kvm_vcpu_init->features bitmap returned will have feature bits set if 3589the preferred target recommends setting these features, but this is 3590not mandatory. 3591 3592The information returned by this ioctl can be used to prepare an instance 3593of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 3594VCPU matching underlying host. 3595 3596 35974.84 KVM_GET_REG_LIST 3598--------------------- 3599 3600:Capability: basic 3601:Architectures: arm64, mips, riscv, x86 (if KVM_CAP_ONE_REG) 3602:Type: vcpu ioctl 3603:Parameters: struct kvm_reg_list (in/out) 3604:Returns: 0 on success; -1 on error 3605 3606Errors: 3607 3608 ===== ============================================================== 3609 E2BIG the reg index list is too big to fit in the array specified by 3610 the user (the number required will be written into n). 3611 ===== ============================================================== 3612 3613:: 3614 3615 struct kvm_reg_list { 3616 __u64 n; /* number of registers in reg[] */ 3617 __u64 reg[0]; 3618 }; 3619 3620This ioctl returns the guest registers that are supported for the 3621KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3622 3623Note that s390 does not support KVM_GET_REG_LIST for historical reasons 3624(read: nobody cared). The set of registers in kernels 4.x and newer is: 3625 3626- KVM_REG_S390_TODPR 3627 3628- KVM_REG_S390_EPOCHDIFF 3629 3630- KVM_REG_S390_CPU_TIMER 3631 3632- KVM_REG_S390_CLOCK_COMP 3633 3634- KVM_REG_S390_PFTOKEN 3635 3636- KVM_REG_S390_PFCOMPARE 3637 3638- KVM_REG_S390_PFSELECT 3639 3640- KVM_REG_S390_PP 3641 3642- KVM_REG_S390_GBEA 3643 3644Note, for x86, all MSRs enumerated by KVM_GET_MSR_INDEX_LIST are supported as 3645type KVM_X86_REG_TYPE_MSR, but are NOT enumerated via KVM_GET_REG_LIST. 3646 36474.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3648----------------------------------------- 3649 3650:Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3651:Architectures: arm64 3652:Type: vm ioctl 3653:Parameters: struct kvm_arm_device_address (in) 3654:Returns: 0 on success, -1 on error 3655 3656Errors: 3657 3658 ====== ============================================ 3659 ENODEV The device id is unknown 3660 ENXIO Device not supported on current system 3661 EEXIST Address already set 3662 E2BIG Address outside guest physical address space 3663 EBUSY Address overlaps with other device range 3664 ====== ============================================ 3665 3666:: 3667 3668 struct kvm_arm_device_addr { 3669 __u64 id; 3670 __u64 addr; 3671 }; 3672 3673Specify a device address in the guest's physical address space where guests 3674can access emulated or directly exposed devices, which the host kernel needs 3675to know about. The id field is an architecture specific identifier for a 3676specific device. 3677 3678arm64 divides the id field into two parts, a device id and an 3679address type id specific to the individual device:: 3680 3681 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 3682 field: | 0x00000000 | device id | addr type id | 3683 3684arm64 currently only require this when using the in-kernel GIC 3685support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 3686as the device id. When setting the base address for the guest's 3687mapping of the VGIC virtual CPU and distributor interface, the ioctl 3688must be called after calling KVM_CREATE_IRQCHIP, but before calling 3689KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 3690base addresses will return -EEXIST. 3691 3692Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 3693should be used instead. 3694 3695 36964.86 KVM_PPC_RTAS_DEFINE_TOKEN 3697------------------------------ 3698 3699:Capability: KVM_CAP_PPC_RTAS 3700:Architectures: ppc 3701:Type: vm ioctl 3702:Parameters: struct kvm_rtas_token_args 3703:Returns: 0 on success, -1 on error 3704 3705Defines a token value for a RTAS (Run Time Abstraction Services) 3706service in order to allow it to be handled in the kernel. The 3707argument struct gives the name of the service, which must be the name 3708of a service that has a kernel-side implementation. If the token 3709value is non-zero, it will be associated with that service, and 3710subsequent RTAS calls by the guest specifying that token will be 3711handled by the kernel. If the token value is 0, then any token 3712associated with the service will be forgotten, and subsequent RTAS 3713calls by the guest for that service will be passed to userspace to be 3714handled. 3715 37164.87 KVM_SET_GUEST_DEBUG 3717------------------------ 3718 3719:Capability: KVM_CAP_SET_GUEST_DEBUG 3720:Architectures: x86, s390, ppc, arm64 3721:Type: vcpu ioctl 3722:Parameters: struct kvm_guest_debug (in) 3723:Returns: 0 on success; -1 on error 3724 3725:: 3726 3727 struct kvm_guest_debug { 3728 __u32 control; 3729 __u32 pad; 3730 struct kvm_guest_debug_arch arch; 3731 }; 3732 3733Set up the processor specific debug registers and configure vcpu for 3734handling guest debug events. There are two parts to the structure, the 3735first a control bitfield indicates the type of debug events to handle 3736when running. Common control bits are: 3737 3738 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 3739 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 3740 3741The top 16 bits of the control field are architecture specific control 3742flags which can include the following: 3743 3744 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 3745 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390] 3746 - KVM_GUESTDBG_USE_HW: using hardware debug events [arm64] 3747 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 3748 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 3749 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 3750 - KVM_GUESTDBG_BLOCKIRQ: avoid injecting interrupts/NMI/SMI [x86] 3751 3752For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 3753are enabled in memory so we need to ensure breakpoint exceptions are 3754correctly trapped and the KVM run loop exits at the breakpoint and not 3755running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 3756we need to ensure the guest vCPUs architecture specific registers are 3757updated to the correct (supplied) values. 3758 3759The second part of the structure is architecture specific and 3760typically contains a set of debug registers. 3761 3762For arm64 the number of debug registers is implementation defined and 3763can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 3764KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 3765indicating the number of supported registers. 3766 3767For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether 3768the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported. 3769 3770Also when supported, KVM_CAP_SET_GUEST_DEBUG2 capability indicates the 3771supported KVM_GUESTDBG_* bits in the control field. 3772 3773When debug events exit the main run loop with the reason 3774KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 3775structure containing architecture specific debug information. 3776 37774.88 KVM_GET_EMULATED_CPUID 3778--------------------------- 3779 3780:Capability: KVM_CAP_EXT_EMUL_CPUID 3781:Architectures: x86 3782:Type: system ioctl 3783:Parameters: struct kvm_cpuid2 (in/out) 3784:Returns: 0 on success, -1 on error 3785 3786:: 3787 3788 struct kvm_cpuid2 { 3789 __u32 nent; 3790 __u32 flags; 3791 struct kvm_cpuid_entry2 entries[0]; 3792 }; 3793 3794The member 'flags' is used for passing flags from userspace. 3795 3796:: 3797 3798 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 3799 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 3800 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 3801 3802 struct kvm_cpuid_entry2 { 3803 __u32 function; 3804 __u32 index; 3805 __u32 flags; 3806 __u32 eax; 3807 __u32 ebx; 3808 __u32 ecx; 3809 __u32 edx; 3810 __u32 padding[3]; 3811 }; 3812 3813This ioctl returns x86 cpuid features which are emulated by 3814kvm.Userspace can use the information returned by this ioctl to query 3815which features are emulated by kvm instead of being present natively. 3816 3817Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 3818structure with the 'nent' field indicating the number of entries in 3819the variable-size array 'entries'. If the number of entries is too low 3820to describe the cpu capabilities, an error (E2BIG) is returned. If the 3821number is too high, the 'nent' field is adjusted and an error (ENOMEM) 3822is returned. If the number is just right, the 'nent' field is adjusted 3823to the number of valid entries in the 'entries' array, which is then 3824filled. 3825 3826The entries returned are the set CPUID bits of the respective features 3827which kvm emulates, as returned by the CPUID instruction, with unknown 3828or unsupported feature bits cleared. 3829 3830Features like x2apic, for example, may not be present in the host cpu 3831but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 3832emulated efficiently and thus not included here. 3833 3834The fields in each entry are defined as follows: 3835 3836 function: 3837 the eax value used to obtain the entry 3838 index: 3839 the ecx value used to obtain the entry (for entries that are 3840 affected by ecx) 3841 flags: 3842 an OR of zero or more of the following: 3843 3844 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3845 if the index field is valid 3846 3847 eax, ebx, ecx, edx: 3848 3849 the values returned by the cpuid instruction for 3850 this function/index combination 3851 38524.89 KVM_S390_MEM_OP 3853-------------------- 3854 3855:Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S390_PROTECTED, KVM_CAP_S390_MEM_OP_EXTENSION 3856:Architectures: s390 3857:Type: vm ioctl, vcpu ioctl 3858:Parameters: struct kvm_s390_mem_op (in) 3859:Returns: = 0 on success, 3860 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 3861 16 bit program exception code if the access causes such an exception 3862 3863Read or write data from/to the VM's memory. 3864The KVM_CAP_S390_MEM_OP_EXTENSION capability specifies what functionality is 3865supported. 3866 3867Parameters are specified via the following structure:: 3868 3869 struct kvm_s390_mem_op { 3870 __u64 gaddr; /* the guest address */ 3871 __u64 flags; /* flags */ 3872 __u32 size; /* amount of bytes */ 3873 __u32 op; /* type of operation */ 3874 __u64 buf; /* buffer in userspace */ 3875 union { 3876 struct { 3877 __u8 ar; /* the access register number */ 3878 __u8 key; /* access key, ignored if flag unset */ 3879 __u8 pad1[6]; /* ignored */ 3880 __u64 old_addr; /* ignored if flag unset */ 3881 }; 3882 __u32 sida_offset; /* offset into the sida */ 3883 __u8 reserved[32]; /* ignored */ 3884 }; 3885 }; 3886 3887The start address of the memory region has to be specified in the "gaddr" 3888field, and the length of the region in the "size" field (which must not 3889be 0). The maximum value for "size" can be obtained by checking the 3890KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the 3891userspace application where the read data should be written to for 3892a read access, or where the data that should be written is stored for 3893a write access. The "reserved" field is meant for future extensions. 3894Reserved and unused values are ignored. Future extension that add members must 3895introduce new flags. 3896 3897The type of operation is specified in the "op" field. Flags modifying 3898their behavior can be set in the "flags" field. Undefined flag bits must 3899be set to 0. 3900 3901Possible operations are: 3902 * ``KVM_S390_MEMOP_LOGICAL_READ`` 3903 * ``KVM_S390_MEMOP_LOGICAL_WRITE`` 3904 * ``KVM_S390_MEMOP_ABSOLUTE_READ`` 3905 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE`` 3906 * ``KVM_S390_MEMOP_SIDA_READ`` 3907 * ``KVM_S390_MEMOP_SIDA_WRITE`` 3908 * ``KVM_S390_MEMOP_ABSOLUTE_CMPXCHG`` 3909 3910Logical read/write: 3911^^^^^^^^^^^^^^^^^^^ 3912 3913Access logical memory, i.e. translate the given guest address to an absolute 3914address given the state of the VCPU and use the absolute address as target of 3915the access. "ar" designates the access register number to be used; the valid 3916range is 0..15. 3917Logical accesses are permitted for the VCPU ioctl only. 3918Logical accesses are permitted for non-protected guests only. 3919 3920Supported flags: 3921 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3922 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION`` 3923 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3924 3925The KVM_S390_MEMOP_F_CHECK_ONLY flag can be set to check whether the 3926corresponding memory access would cause an access exception; however, 3927no actual access to the data in memory at the destination is performed. 3928In this case, "buf" is unused and can be NULL. 3929 3930In case an access exception occurred during the access (or would occur 3931in case of KVM_S390_MEMOP_F_CHECK_ONLY), the ioctl returns a positive 3932error number indicating the type of exception. This exception is also 3933raised directly at the corresponding VCPU if the flag 3934KVM_S390_MEMOP_F_INJECT_EXCEPTION is set. 3935On protection exceptions, unless specified otherwise, the injected 3936translation-exception identifier (TEID) indicates suppression. 3937 3938If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key 3939protection is also in effect and may cause exceptions if accesses are 3940prohibited given the access key designated by "key"; the valid range is 0..15. 3941KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION 3942is > 0. 3943Since the accessed memory may span multiple pages and those pages might have 3944different storage keys, it is possible that a protection exception occurs 3945after memory has been modified. In this case, if the exception is injected, 3946the TEID does not indicate suppression. 3947 3948Absolute read/write: 3949^^^^^^^^^^^^^^^^^^^^ 3950 3951Access absolute memory. This operation is intended to be used with the 3952KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to allow accessing memory and performing 3953the checks required for storage key protection as one operation (as opposed to 3954user space getting the storage keys, performing the checks, and accessing 3955memory thereafter, which could lead to a delay between check and access). 3956Absolute accesses are permitted for the VM ioctl if KVM_CAP_S390_MEM_OP_EXTENSION 3957has the KVM_S390_MEMOP_EXTENSION_CAP_BASE bit set. 3958Currently absolute accesses are not permitted for VCPU ioctls. 3959Absolute accesses are permitted for non-protected guests only. 3960 3961Supported flags: 3962 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3963 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3964 3965The semantics of the flags common with logical accesses are as for logical 3966accesses. 3967 3968Absolute cmpxchg: 3969^^^^^^^^^^^^^^^^^ 3970 3971Perform cmpxchg on absolute guest memory. Intended for use with the 3972KVM_S390_MEMOP_F_SKEY_PROTECTION flag. 3973Instead of doing an unconditional write, the access occurs only if the target 3974location contains the value pointed to by "old_addr". 3975This is performed as an atomic cmpxchg with the length specified by the "size" 3976parameter. "size" must be a power of two up to and including 16. 3977If the exchange did not take place because the target value doesn't match the 3978old value, the value "old_addr" points to is replaced by the target value. 3979User space can tell if an exchange took place by checking if this replacement 3980occurred. The cmpxchg op is permitted for the VM ioctl if 3981KVM_CAP_S390_MEM_OP_EXTENSION has flag KVM_S390_MEMOP_EXTENSION_CAP_CMPXCHG set. 3982 3983Supported flags: 3984 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3985 3986SIDA read/write: 3987^^^^^^^^^^^^^^^^ 3988 3989Access the secure instruction data area which contains memory operands necessary 3990for instruction emulation for protected guests. 3991SIDA accesses are available if the KVM_CAP_S390_PROTECTED capability is available. 3992SIDA accesses are permitted for the VCPU ioctl only. 3993SIDA accesses are permitted for protected guests only. 3994 3995No flags are supported. 3996 39974.90 KVM_S390_GET_SKEYS 3998----------------------- 3999 4000:Capability: KVM_CAP_S390_SKEYS 4001:Architectures: s390 4002:Type: vm ioctl 4003:Parameters: struct kvm_s390_skeys 4004:Returns: 0 on success, KVM_S390_GET_SKEYS_NONE if guest is not using storage 4005 keys, negative value on error 4006 4007This ioctl is used to get guest storage key values on the s390 4008architecture. The ioctl takes parameters via the kvm_s390_skeys struct:: 4009 4010 struct kvm_s390_skeys { 4011 __u64 start_gfn; 4012 __u64 count; 4013 __u64 skeydata_addr; 4014 __u32 flags; 4015 __u32 reserved[9]; 4016 }; 4017 4018The start_gfn field is the number of the first guest frame whose storage keys 4019you want to get. 4020 4021The count field is the number of consecutive frames (starting from start_gfn) 4022whose storage keys to get. The count field must be at least 1 and the maximum 4023allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 4024will cause the ioctl to return -EINVAL. 4025 4026The skeydata_addr field is the address to a buffer large enough to hold count 4027bytes. This buffer will be filled with storage key data by the ioctl. 4028 40294.91 KVM_S390_SET_SKEYS 4030----------------------- 4031 4032:Capability: KVM_CAP_S390_SKEYS 4033:Architectures: s390 4034:Type: vm ioctl 4035:Parameters: struct kvm_s390_skeys 4036:Returns: 0 on success, negative value on error 4037 4038This ioctl is used to set guest storage key values on the s390 4039architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 4040See section on KVM_S390_GET_SKEYS for struct definition. 4041 4042The start_gfn field is the number of the first guest frame whose storage keys 4043you want to set. 4044 4045The count field is the number of consecutive frames (starting from start_gfn) 4046whose storage keys to get. The count field must be at least 1 and the maximum 4047allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 4048will cause the ioctl to return -EINVAL. 4049 4050The skeydata_addr field is the address to a buffer containing count bytes of 4051storage keys. Each byte in the buffer will be set as the storage key for a 4052single frame starting at start_gfn for count frames. 4053 4054Note: If any architecturally invalid key value is found in the given data then 4055the ioctl will return -EINVAL. 4056 40574.92 KVM_S390_IRQ 4058----------------- 4059 4060:Capability: KVM_CAP_S390_INJECT_IRQ 4061:Architectures: s390 4062:Type: vcpu ioctl 4063:Parameters: struct kvm_s390_irq (in) 4064:Returns: 0 on success, -1 on error 4065 4066Errors: 4067 4068 4069 ====== ================================================================= 4070 EINVAL interrupt type is invalid 4071 type is KVM_S390_SIGP_STOP and flag parameter is invalid value, 4072 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 4073 than the maximum of VCPUs 4074 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped, 4075 type is KVM_S390_SIGP_STOP and a stop irq is already pending, 4076 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 4077 is already pending 4078 ====== ================================================================= 4079 4080Allows to inject an interrupt to the guest. 4081 4082Using struct kvm_s390_irq as a parameter allows 4083to inject additional payload which is not 4084possible via KVM_S390_INTERRUPT. 4085 4086Interrupt parameters are passed via kvm_s390_irq:: 4087 4088 struct kvm_s390_irq { 4089 __u64 type; 4090 union { 4091 struct kvm_s390_io_info io; 4092 struct kvm_s390_ext_info ext; 4093 struct kvm_s390_pgm_info pgm; 4094 struct kvm_s390_emerg_info emerg; 4095 struct kvm_s390_extcall_info extcall; 4096 struct kvm_s390_prefix_info prefix; 4097 struct kvm_s390_stop_info stop; 4098 struct kvm_s390_mchk_info mchk; 4099 char reserved[64]; 4100 } u; 4101 }; 4102 4103type can be one of the following: 4104 4105- KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 4106- KVM_S390_PROGRAM_INT - program check; parameters in .pgm 4107- KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 4108- KVM_S390_RESTART - restart; no parameters 4109- KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 4110- KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 4111- KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 4112- KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 4113- KVM_S390_MCHK - machine check interrupt; parameters in .mchk 4114 4115This is an asynchronous vcpu ioctl and can be invoked from any thread. 4116 41174.94 KVM_S390_GET_IRQ_STATE 4118--------------------------- 4119 4120:Capability: KVM_CAP_S390_IRQ_STATE 4121:Architectures: s390 4122:Type: vcpu ioctl 4123:Parameters: struct kvm_s390_irq_state (out) 4124:Returns: >= number of bytes copied into buffer, 4125 -EINVAL if buffer size is 0, 4126 -ENOBUFS if buffer size is too small to fit all pending interrupts, 4127 -EFAULT if the buffer address was invalid 4128 4129This ioctl allows userspace to retrieve the complete state of all currently 4130pending interrupts in a single buffer. Use cases include migration 4131and introspection. The parameter structure contains the address of a 4132userspace buffer and its length:: 4133 4134 struct kvm_s390_irq_state { 4135 __u64 buf; 4136 __u32 flags; /* will stay unused for compatibility reasons */ 4137 __u32 len; 4138 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4139 }; 4140 4141Userspace passes in the above struct and for each pending interrupt a 4142struct kvm_s390_irq is copied to the provided buffer. 4143 4144The structure contains a flags and a reserved field for future extensions. As 4145the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 4146reserved, these fields can not be used in the future without breaking 4147compatibility. 4148 4149If -ENOBUFS is returned the buffer provided was too small and userspace 4150may retry with a bigger buffer. 4151 41524.95 KVM_S390_SET_IRQ_STATE 4153--------------------------- 4154 4155:Capability: KVM_CAP_S390_IRQ_STATE 4156:Architectures: s390 4157:Type: vcpu ioctl 4158:Parameters: struct kvm_s390_irq_state (in) 4159:Returns: 0 on success, 4160 -EFAULT if the buffer address was invalid, 4161 -EINVAL for an invalid buffer length (see below), 4162 -EBUSY if there were already interrupts pending, 4163 errors occurring when actually injecting the 4164 interrupt. See KVM_S390_IRQ. 4165 4166This ioctl allows userspace to set the complete state of all cpu-local 4167interrupts currently pending for the vcpu. It is intended for restoring 4168interrupt state after a migration. The input parameter is a userspace buffer 4169containing a struct kvm_s390_irq_state:: 4170 4171 struct kvm_s390_irq_state { 4172 __u64 buf; 4173 __u32 flags; /* will stay unused for compatibility reasons */ 4174 __u32 len; 4175 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4176 }; 4177 4178The restrictions for flags and reserved apply as well. 4179(see KVM_S390_GET_IRQ_STATE) 4180 4181The userspace memory referenced by buf contains a struct kvm_s390_irq 4182for each interrupt to be injected into the guest. 4183If one of the interrupts could not be injected for some reason the 4184ioctl aborts. 4185 4186len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 4187and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 4188which is the maximum number of possibly pending cpu-local interrupts. 4189 41904.96 KVM_SMI 4191------------ 4192 4193:Capability: KVM_CAP_X86_SMM 4194:Architectures: x86 4195:Type: vcpu ioctl 4196:Parameters: none 4197:Returns: 0 on success, -1 on error 4198 4199Queues an SMI on the thread's vcpu. 4200 42014.97 KVM_X86_SET_MSR_FILTER 4202---------------------------- 4203 4204:Capability: KVM_CAP_X86_MSR_FILTER 4205:Architectures: x86 4206:Type: vm ioctl 4207:Parameters: struct kvm_msr_filter 4208:Returns: 0 on success, < 0 on error 4209 4210:: 4211 4212 struct kvm_msr_filter_range { 4213 #define KVM_MSR_FILTER_READ (1 << 0) 4214 #define KVM_MSR_FILTER_WRITE (1 << 1) 4215 __u32 flags; 4216 __u32 nmsrs; /* number of msrs in bitmap */ 4217 __u32 base; /* MSR index the bitmap starts at */ 4218 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 4219 }; 4220 4221 #define KVM_MSR_FILTER_MAX_RANGES 16 4222 struct kvm_msr_filter { 4223 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 4224 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 4225 __u32 flags; 4226 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 4227 }; 4228 4229flags values for ``struct kvm_msr_filter_range``: 4230 4231``KVM_MSR_FILTER_READ`` 4232 4233 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 4234 indicates that read accesses should be denied, while a 1 indicates that 4235 a read for a particular MSR should be allowed regardless of the default 4236 filter action. 4237 4238``KVM_MSR_FILTER_WRITE`` 4239 4240 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 4241 indicates that write accesses should be denied, while a 1 indicates that 4242 a write for a particular MSR should be allowed regardless of the default 4243 filter action. 4244 4245flags values for ``struct kvm_msr_filter``: 4246 4247``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4248 4249 If no filter range matches an MSR index that is getting accessed, KVM will 4250 allow accesses to all MSRs by default. 4251 4252``KVM_MSR_FILTER_DEFAULT_DENY`` 4253 4254 If no filter range matches an MSR index that is getting accessed, KVM will 4255 deny accesses to all MSRs by default. 4256 4257This ioctl allows userspace to define up to 16 bitmaps of MSR ranges to deny 4258guest MSR accesses that would normally be allowed by KVM. If an MSR is not 4259covered by a specific range, the "default" filtering behavior applies. Each 4260bitmap range covers MSRs from [base .. base+nmsrs). 4261 4262If an MSR access is denied by userspace, the resulting KVM behavior depends on 4263whether or not KVM_CAP_X86_USER_SPACE_MSR's KVM_MSR_EXIT_REASON_FILTER is 4264enabled. If KVM_MSR_EXIT_REASON_FILTER is enabled, KVM will exit to userspace 4265on denied accesses, i.e. userspace effectively intercepts the MSR access. If 4266KVM_MSR_EXIT_REASON_FILTER is not enabled, KVM will inject a #GP into the guest 4267on denied accesses. Note, if an MSR access is denied during emulation of MSR 4268load/stores during VMX transitions, KVM ignores KVM_MSR_EXIT_REASON_FILTER. 4269See the below warning for full details. 4270 4271If an MSR access is allowed by userspace, KVM will emulate and/or virtualize 4272the access in accordance with the vCPU model. Note, KVM may still ultimately 4273inject a #GP if an access is allowed by userspace, e.g. if KVM doesn't support 4274the MSR, or to follow architectural behavior for the MSR. 4275 4276By default, KVM operates in KVM_MSR_FILTER_DEFAULT_ALLOW mode with no MSR range 4277filters. 4278 4279Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 4280filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 4281an error. 4282 4283.. warning:: 4284 MSR accesses that are side effects of instruction execution (emulated or 4285 native) are not filtered as hardware does not honor MSR bitmaps outside of 4286 RDMSR and WRMSR, and KVM mimics that behavior when emulating instructions 4287 to avoid pointless divergence from hardware. E.g. RDPID reads MSR_TSC_AUX, 4288 SYSENTER reads the SYSENTER MSRs, etc. 4289 4290 MSRs that are loaded/stored via dedicated VMCS fields are not filtered as 4291 part of VM-Enter/VM-Exit emulation. 4292 4293 MSRs that are loaded/store via VMX's load/store lists _are_ filtered as part 4294 of VM-Enter/VM-Exit emulation. If an MSR access is denied on VM-Enter, KVM 4295 synthesizes a consistency check VM-Exit(EXIT_REASON_MSR_LOAD_FAIL). If an 4296 MSR access is denied on VM-Exit, KVM synthesizes a VM-Abort. In short, KVM 4297 extends Intel's architectural list of MSRs that cannot be loaded/saved via 4298 the VM-Enter/VM-Exit MSR list. It is platform owner's responsibility to 4299 to communicate any such restrictions to their end users. 4300 4301 x2APIC MSR accesses cannot be filtered (KVM silently ignores filters that 4302 cover any x2APIC MSRs). 4303 4304Note, invoking this ioctl while a vCPU is running is inherently racy. However, 4305KVM does guarantee that vCPUs will see either the previous filter or the new 4306filter, e.g. MSRs with identical settings in both the old and new filter will 4307have deterministic behavior. 4308 4309Similarly, if userspace wishes to intercept on denied accesses, 4310KVM_MSR_EXIT_REASON_FILTER must be enabled before activating any filters, and 4311left enabled until after all filters are deactivated. Failure to do so may 4312result in KVM injecting a #GP instead of exiting to userspace. 4313 43144.98 KVM_CREATE_SPAPR_TCE_64 4315---------------------------- 4316 4317:Capability: KVM_CAP_SPAPR_TCE_64 4318:Architectures: powerpc 4319:Type: vm ioctl 4320:Parameters: struct kvm_create_spapr_tce_64 (in) 4321:Returns: file descriptor for manipulating the created TCE table 4322 4323This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 4324windows, described in 4.62 KVM_CREATE_SPAPR_TCE 4325 4326This capability uses extended struct in ioctl interface:: 4327 4328 /* for KVM_CAP_SPAPR_TCE_64 */ 4329 struct kvm_create_spapr_tce_64 { 4330 __u64 liobn; 4331 __u32 page_shift; 4332 __u32 flags; 4333 __u64 offset; /* in pages */ 4334 __u64 size; /* in pages */ 4335 }; 4336 4337The aim of extension is to support an additional bigger DMA window with 4338a variable page size. 4339KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 4340a bus offset of the corresponding DMA window, @size and @offset are numbers 4341of IOMMU pages. 4342 4343@flags are not used at the moment. 4344 4345The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 4346 43474.99 KVM_REINJECT_CONTROL 4348------------------------- 4349 4350:Capability: KVM_CAP_REINJECT_CONTROL 4351:Architectures: x86 4352:Type: vm ioctl 4353:Parameters: struct kvm_reinject_control (in) 4354:Returns: 0 on success, 4355 -EFAULT if struct kvm_reinject_control cannot be read, 4356 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 4357 4358i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 4359where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 4360vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 4361interrupt whenever there isn't a pending interrupt from i8254. 4362!reinject mode injects an interrupt as soon as a tick arrives. 4363 4364:: 4365 4366 struct kvm_reinject_control { 4367 __u8 pit_reinject; 4368 __u8 reserved[31]; 4369 }; 4370 4371pit_reinject = 0 (!reinject mode) is recommended, unless running an old 4372operating system that uses the PIT for timing (e.g. Linux 2.4.x). 4373 43744.100 KVM_PPC_CONFIGURE_V3_MMU 4375------------------------------ 4376 4377:Capability: KVM_CAP_PPC_MMU_RADIX or KVM_CAP_PPC_MMU_HASH_V3 4378:Architectures: ppc 4379:Type: vm ioctl 4380:Parameters: struct kvm_ppc_mmuv3_cfg (in) 4381:Returns: 0 on success, 4382 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 4383 -EINVAL if the configuration is invalid 4384 4385This ioctl controls whether the guest will use radix or HPT (hashed 4386page table) translation, and sets the pointer to the process table for 4387the guest. 4388 4389:: 4390 4391 struct kvm_ppc_mmuv3_cfg { 4392 __u64 flags; 4393 __u64 process_table; 4394 }; 4395 4396There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 4397KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 4398to use radix tree translation, and if clear, to use HPT translation. 4399KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 4400to be able to use the global TLB and SLB invalidation instructions; 4401if clear, the guest may not use these instructions. 4402 4403The process_table field specifies the address and size of the guest 4404process table, which is in the guest's space. This field is formatted 4405as the second doubleword of the partition table entry, as defined in 4406the Power ISA V3.00, Book III section 5.7.6.1. 4407 44084.101 KVM_PPC_GET_RMMU_INFO 4409--------------------------- 4410 4411:Capability: KVM_CAP_PPC_MMU_RADIX 4412:Architectures: ppc 4413:Type: vm ioctl 4414:Parameters: struct kvm_ppc_rmmu_info (out) 4415:Returns: 0 on success, 4416 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 4417 -EINVAL if no useful information can be returned 4418 4419This ioctl returns a structure containing two things: (a) a list 4420containing supported radix tree geometries, and (b) a list that maps 4421page sizes to put in the "AP" (actual page size) field for the tlbie 4422(TLB invalidate entry) instruction. 4423 4424:: 4425 4426 struct kvm_ppc_rmmu_info { 4427 struct kvm_ppc_radix_geom { 4428 __u8 page_shift; 4429 __u8 level_bits[4]; 4430 __u8 pad[3]; 4431 } geometries[8]; 4432 __u32 ap_encodings[8]; 4433 }; 4434 4435The geometries[] field gives up to 8 supported geometries for the 4436radix page table, in terms of the log base 2 of the smallest page 4437size, and the number of bits indexed at each level of the tree, from 4438the PTE level up to the PGD level in that order. Any unused entries 4439will have 0 in the page_shift field. 4440 4441The ap_encodings gives the supported page sizes and their AP field 4442encodings, encoded with the AP value in the top 3 bits and the log 4443base 2 of the page size in the bottom 6 bits. 4444 44454.102 KVM_PPC_RESIZE_HPT_PREPARE 4446-------------------------------- 4447 4448:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4449:Architectures: powerpc 4450:Type: vm ioctl 4451:Parameters: struct kvm_ppc_resize_hpt (in) 4452:Returns: 0 on successful completion, 4453 >0 if a new HPT is being prepared, the value is an estimated 4454 number of milliseconds until preparation is complete, 4455 -EFAULT if struct kvm_reinject_control cannot be read, 4456 -EINVAL if the supplied shift or flags are invalid, 4457 -ENOMEM if unable to allocate the new HPT, 4458 4459Used to implement the PAPR extension for runtime resizing of a guest's 4460Hashed Page Table (HPT). Specifically this starts, stops or monitors 4461the preparation of a new potential HPT for the guest, essentially 4462implementing the H_RESIZE_HPT_PREPARE hypercall. 4463 4464:: 4465 4466 struct kvm_ppc_resize_hpt { 4467 __u64 flags; 4468 __u32 shift; 4469 __u32 pad; 4470 }; 4471 4472If called with shift > 0 when there is no pending HPT for the guest, 4473this begins preparation of a new pending HPT of size 2^(shift) bytes. 4474It then returns a positive integer with the estimated number of 4475milliseconds until preparation is complete. 4476 4477If called when there is a pending HPT whose size does not match that 4478requested in the parameters, discards the existing pending HPT and 4479creates a new one as above. 4480 4481If called when there is a pending HPT of the size requested, will: 4482 4483 * If preparation of the pending HPT is already complete, return 0 4484 * If preparation of the pending HPT has failed, return an error 4485 code, then discard the pending HPT. 4486 * If preparation of the pending HPT is still in progress, return an 4487 estimated number of milliseconds until preparation is complete. 4488 4489If called with shift == 0, discards any currently pending HPT and 4490returns 0 (i.e. cancels any in-progress preparation). 4491 4492flags is reserved for future expansion, currently setting any bits in 4493flags will result in an -EINVAL. 4494 4495Normally this will be called repeatedly with the same parameters until 4496it returns <= 0. The first call will initiate preparation, subsequent 4497ones will monitor preparation until it completes or fails. 4498 44994.103 KVM_PPC_RESIZE_HPT_COMMIT 4500------------------------------- 4501 4502:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4503:Architectures: powerpc 4504:Type: vm ioctl 4505:Parameters: struct kvm_ppc_resize_hpt (in) 4506:Returns: 0 on successful completion, 4507 -EFAULT if struct kvm_reinject_control cannot be read, 4508 -EINVAL if the supplied shift or flags are invalid, 4509 -ENXIO is there is no pending HPT, or the pending HPT doesn't 4510 have the requested size, 4511 -EBUSY if the pending HPT is not fully prepared, 4512 -ENOSPC if there was a hash collision when moving existing 4513 HPT entries to the new HPT, 4514 -EIO on other error conditions 4515 4516Used to implement the PAPR extension for runtime resizing of a guest's 4517Hashed Page Table (HPT). Specifically this requests that the guest be 4518transferred to working with the new HPT, essentially implementing the 4519H_RESIZE_HPT_COMMIT hypercall. 4520 4521:: 4522 4523 struct kvm_ppc_resize_hpt { 4524 __u64 flags; 4525 __u32 shift; 4526 __u32 pad; 4527 }; 4528 4529This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 4530returned 0 with the same parameters. In other cases 4531KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 4532-EBUSY, though others may be possible if the preparation was started, 4533but failed). 4534 4535This will have undefined effects on the guest if it has not already 4536placed itself in a quiescent state where no vcpu will make MMU enabled 4537memory accesses. 4538 4539On successful completion, the pending HPT will become the guest's active 4540HPT and the previous HPT will be discarded. 4541 4542On failure, the guest will still be operating on its previous HPT. 4543 45444.104 KVM_X86_GET_MCE_CAP_SUPPORTED 4545----------------------------------- 4546 4547:Capability: KVM_CAP_MCE 4548:Architectures: x86 4549:Type: system ioctl 4550:Parameters: u64 mce_cap (out) 4551:Returns: 0 on success, -1 on error 4552 4553Returns supported MCE capabilities. The u64 mce_cap parameter 4554has the same format as the MSR_IA32_MCG_CAP register. Supported 4555capabilities will have the corresponding bits set. 4556 45574.105 KVM_X86_SETUP_MCE 4558----------------------- 4559 4560:Capability: KVM_CAP_MCE 4561:Architectures: x86 4562:Type: vcpu ioctl 4563:Parameters: u64 mcg_cap (in) 4564:Returns: 0 on success, 4565 -EFAULT if u64 mcg_cap cannot be read, 4566 -EINVAL if the requested number of banks is invalid, 4567 -EINVAL if requested MCE capability is not supported. 4568 4569Initializes MCE support for use. The u64 mcg_cap parameter 4570has the same format as the MSR_IA32_MCG_CAP register and 4571specifies which capabilities should be enabled. The maximum 4572supported number of error-reporting banks can be retrieved when 4573checking for KVM_CAP_MCE. The supported capabilities can be 4574retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 4575 45764.106 KVM_X86_SET_MCE 4577--------------------- 4578 4579:Capability: KVM_CAP_MCE 4580:Architectures: x86 4581:Type: vcpu ioctl 4582:Parameters: struct kvm_x86_mce (in) 4583:Returns: 0 on success, 4584 -EFAULT if struct kvm_x86_mce cannot be read, 4585 -EINVAL if the bank number is invalid, 4586 -EINVAL if VAL bit is not set in status field. 4587 4588Inject a machine check error (MCE) into the guest. The input 4589parameter is:: 4590 4591 struct kvm_x86_mce { 4592 __u64 status; 4593 __u64 addr; 4594 __u64 misc; 4595 __u64 mcg_status; 4596 __u8 bank; 4597 __u8 pad1[7]; 4598 __u64 pad2[3]; 4599 }; 4600 4601If the MCE being reported is an uncorrected error, KVM will 4602inject it as an MCE exception into the guest. If the guest 4603MCG_STATUS register reports that an MCE is in progress, KVM 4604causes an KVM_EXIT_SHUTDOWN vmexit. 4605 4606Otherwise, if the MCE is a corrected error, KVM will just 4607store it in the corresponding bank (provided this bank is 4608not holding a previously reported uncorrected error). 4609 46104.107 KVM_S390_GET_CMMA_BITS 4611---------------------------- 4612 4613:Capability: KVM_CAP_S390_CMMA_MIGRATION 4614:Architectures: s390 4615:Type: vm ioctl 4616:Parameters: struct kvm_s390_cmma_log (in, out) 4617:Returns: 0 on success, a negative value on error 4618 4619Errors: 4620 4621 ====== ============================================================= 4622 ENOMEM not enough memory can be allocated to complete the task 4623 ENXIO if CMMA is not enabled 4624 EINVAL if KVM_S390_CMMA_PEEK is not set but migration mode was not enabled 4625 EINVAL if KVM_S390_CMMA_PEEK is not set but dirty tracking has been 4626 disabled (and thus migration mode was automatically disabled) 4627 EFAULT if the userspace address is invalid or if no page table is 4628 present for the addresses (e.g. when using hugepages). 4629 ====== ============================================================= 4630 4631This ioctl is used to get the values of the CMMA bits on the s390 4632architecture. It is meant to be used in two scenarios: 4633 4634- During live migration to save the CMMA values. Live migration needs 4635 to be enabled via the KVM_REQ_START_MIGRATION VM property. 4636- To non-destructively peek at the CMMA values, with the flag 4637 KVM_S390_CMMA_PEEK set. 4638 4639The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 4640values are written to a buffer whose location is indicated via the "values" 4641member in the kvm_s390_cmma_log struct. The values in the input struct are 4642also updated as needed. 4643 4644Each CMMA value takes up one byte. 4645 4646:: 4647 4648 struct kvm_s390_cmma_log { 4649 __u64 start_gfn; 4650 __u32 count; 4651 __u32 flags; 4652 union { 4653 __u64 remaining; 4654 __u64 mask; 4655 }; 4656 __u64 values; 4657 }; 4658 4659start_gfn is the number of the first guest frame whose CMMA values are 4660to be retrieved, 4661 4662count is the length of the buffer in bytes, 4663 4664values points to the buffer where the result will be written to. 4665 4666If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 4667KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 4668other ioctls. 4669 4670The result is written in the buffer pointed to by the field values, and 4671the values of the input parameter are updated as follows. 4672 4673Depending on the flags, different actions are performed. The only 4674supported flag so far is KVM_S390_CMMA_PEEK. 4675 4676The default behaviour if KVM_S390_CMMA_PEEK is not set is: 4677start_gfn will indicate the first page frame whose CMMA bits were dirty. 4678It is not necessarily the same as the one passed as input, as clean pages 4679are skipped. 4680 4681count will indicate the number of bytes actually written in the buffer. 4682It can (and very often will) be smaller than the input value, since the 4683buffer is only filled until 16 bytes of clean values are found (which 4684are then not copied in the buffer). Since a CMMA migration block needs 4685the base address and the length, for a total of 16 bytes, we will send 4686back some clean data if there is some dirty data afterwards, as long as 4687the size of the clean data does not exceed the size of the header. This 4688allows to minimize the amount of data to be saved or transferred over 4689the network at the expense of more roundtrips to userspace. The next 4690invocation of the ioctl will skip over all the clean values, saving 4691potentially more than just the 16 bytes we found. 4692 4693If KVM_S390_CMMA_PEEK is set: 4694the existing storage attributes are read even when not in migration 4695mode, and no other action is performed; 4696 4697the output start_gfn will be equal to the input start_gfn, 4698 4699the output count will be equal to the input count, except if the end of 4700memory has been reached. 4701 4702In both cases: 4703the field "remaining" will indicate the total number of dirty CMMA values 4704still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 4705not enabled. 4706 4707mask is unused. 4708 4709values points to the userspace buffer where the result will be stored. 4710 47114.108 KVM_S390_SET_CMMA_BITS 4712---------------------------- 4713 4714:Capability: KVM_CAP_S390_CMMA_MIGRATION 4715:Architectures: s390 4716:Type: vm ioctl 4717:Parameters: struct kvm_s390_cmma_log (in) 4718:Returns: 0 on success, a negative value on error 4719 4720This ioctl is used to set the values of the CMMA bits on the s390 4721architecture. It is meant to be used during live migration to restore 4722the CMMA values, but there are no restrictions on its use. 4723The ioctl takes parameters via the kvm_s390_cmma_values struct. 4724Each CMMA value takes up one byte. 4725 4726:: 4727 4728 struct kvm_s390_cmma_log { 4729 __u64 start_gfn; 4730 __u32 count; 4731 __u32 flags; 4732 union { 4733 __u64 remaining; 4734 __u64 mask; 4735 }; 4736 __u64 values; 4737 }; 4738 4739start_gfn indicates the starting guest frame number, 4740 4741count indicates how many values are to be considered in the buffer, 4742 4743flags is not used and must be 0. 4744 4745mask indicates which PGSTE bits are to be considered. 4746 4747remaining is not used. 4748 4749values points to the buffer in userspace where to store the values. 4750 4751This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4752complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4753the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 4754if the flags field was not 0, with -EFAULT if the userspace address is 4755invalid, if invalid pages are written to (e.g. after the end of memory) 4756or if no page table is present for the addresses (e.g. when using 4757hugepages). 4758 47594.109 KVM_PPC_GET_CPU_CHAR 4760-------------------------- 4761 4762:Capability: KVM_CAP_PPC_GET_CPU_CHAR 4763:Architectures: powerpc 4764:Type: vm ioctl 4765:Parameters: struct kvm_ppc_cpu_char (out) 4766:Returns: 0 on successful completion, 4767 -EFAULT if struct kvm_ppc_cpu_char cannot be written 4768 4769This ioctl gives userspace information about certain characteristics 4770of the CPU relating to speculative execution of instructions and 4771possible information leakage resulting from speculative execution (see 4772CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 4773returned in struct kvm_ppc_cpu_char, which looks like this:: 4774 4775 struct kvm_ppc_cpu_char { 4776 __u64 character; /* characteristics of the CPU */ 4777 __u64 behaviour; /* recommended software behaviour */ 4778 __u64 character_mask; /* valid bits in character */ 4779 __u64 behaviour_mask; /* valid bits in behaviour */ 4780 }; 4781 4782For extensibility, the character_mask and behaviour_mask fields 4783indicate which bits of character and behaviour have been filled in by 4784the kernel. If the set of defined bits is extended in future then 4785userspace will be able to tell whether it is running on a kernel that 4786knows about the new bits. 4787 4788The character field describes attributes of the CPU which can help 4789with preventing inadvertent information disclosure - specifically, 4790whether there is an instruction to flash-invalidate the L1 data cache 4791(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 4792to a mode where entries can only be used by the thread that created 4793them, whether the bcctr[l] instruction prevents speculation, and 4794whether a speculation barrier instruction (ori 31,31,0) is provided. 4795 4796The behaviour field describes actions that software should take to 4797prevent inadvertent information disclosure, and thus describes which 4798vulnerabilities the hardware is subject to; specifically whether the 4799L1 data cache should be flushed when returning to user mode from the 4800kernel, and whether a speculation barrier should be placed between an 4801array bounds check and the array access. 4802 4803These fields use the same bit definitions as the new 4804H_GET_CPU_CHARACTERISTICS hypercall. 4805 48064.110 KVM_MEMORY_ENCRYPT_OP 4807--------------------------- 4808 4809:Capability: basic 4810:Architectures: x86 4811:Type: vm ioctl, vcpu ioctl 4812:Parameters: an opaque platform specific structure (in/out) 4813:Returns: 0 on success; -1 on error 4814 4815If the platform supports creating encrypted VMs then this ioctl can be used 4816for issuing platform-specific memory encryption commands to manage those 4817encrypted VMs. 4818 4819Currently, this ioctl is used for issuing both Secure Encrypted Virtualization 4820(SEV) commands on AMD Processors and Trusted Domain Extensions (TDX) commands 4821on Intel Processors. The detailed commands are defined in 4822Documentation/virt/kvm/x86/amd-memory-encryption.rst and 4823Documentation/virt/kvm/x86/intel-tdx.rst. 4824 48254.111 KVM_MEMORY_ENCRYPT_REG_REGION 4826----------------------------------- 4827 4828:Capability: basic 4829:Architectures: x86 4830:Type: system 4831:Parameters: struct kvm_enc_region (in) 4832:Returns: 0 on success; -1 on error 4833 4834This ioctl can be used to register a guest memory region which may 4835contain encrypted data (e.g. guest RAM, SMRAM etc). 4836 4837It is used in the SEV-enabled guest. When encryption is enabled, a guest 4838memory region may contain encrypted data. The SEV memory encryption 4839engine uses a tweak such that two identical plaintext pages, each at 4840different locations will have differing ciphertexts. So swapping or 4841moving ciphertext of those pages will not result in plaintext being 4842swapped. So relocating (or migrating) physical backing pages for the SEV 4843guest will require some additional steps. 4844 4845Note: The current SEV key management spec does not provide commands to 4846swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 4847memory region registered with the ioctl. 4848 48494.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4850------------------------------------- 4851 4852:Capability: basic 4853:Architectures: x86 4854:Type: system 4855:Parameters: struct kvm_enc_region (in) 4856:Returns: 0 on success; -1 on error 4857 4858This ioctl can be used to unregister the guest memory region registered 4859with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 4860 48614.113 KVM_HYPERV_EVENTFD 4862------------------------ 4863 4864:Capability: KVM_CAP_HYPERV_EVENTFD 4865:Architectures: x86 4866:Type: vm ioctl 4867:Parameters: struct kvm_hyperv_eventfd (in) 4868 4869This ioctl (un)registers an eventfd to receive notifications from the guest on 4870the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without 4871causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number 4872(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit. 4873 4874:: 4875 4876 struct kvm_hyperv_eventfd { 4877 __u32 conn_id; 4878 __s32 fd; 4879 __u32 flags; 4880 __u32 padding[3]; 4881 }; 4882 4883The conn_id field should fit within 24 bits:: 4884 4885 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff 4886 4887The acceptable values for the flags field are:: 4888 4889 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0) 4890 4891:Returns: 0 on success, 4892 -EINVAL if conn_id or flags is outside the allowed range, 4893 -ENOENT on deassign if the conn_id isn't registered, 4894 -EEXIST on assign if the conn_id is already registered 4895 48964.114 KVM_GET_NESTED_STATE 4897-------------------------- 4898 4899:Capability: KVM_CAP_NESTED_STATE 4900:Architectures: x86 4901:Type: vcpu ioctl 4902:Parameters: struct kvm_nested_state (in/out) 4903:Returns: 0 on success, -1 on error 4904 4905Errors: 4906 4907 ===== ============================================================= 4908 E2BIG the total state size exceeds the value of 'size' specified by 4909 the user; the size required will be written into size. 4910 ===== ============================================================= 4911 4912:: 4913 4914 struct kvm_nested_state { 4915 __u16 flags; 4916 __u16 format; 4917 __u32 size; 4918 4919 union { 4920 struct kvm_vmx_nested_state_hdr vmx; 4921 struct kvm_svm_nested_state_hdr svm; 4922 4923 /* Pad the header to 128 bytes. */ 4924 __u8 pad[120]; 4925 } hdr; 4926 4927 union { 4928 struct kvm_vmx_nested_state_data vmx[0]; 4929 struct kvm_svm_nested_state_data svm[0]; 4930 } data; 4931 }; 4932 4933 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001 4934 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002 4935 #define KVM_STATE_NESTED_EVMCS 0x00000004 4936 4937 #define KVM_STATE_NESTED_FORMAT_VMX 0 4938 #define KVM_STATE_NESTED_FORMAT_SVM 1 4939 4940 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 4941 4942 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001 4943 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002 4944 4945 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001 4946 4947 struct kvm_vmx_nested_state_hdr { 4948 __u64 vmxon_pa; 4949 __u64 vmcs12_pa; 4950 4951 struct { 4952 __u16 flags; 4953 } smm; 4954 4955 __u32 flags; 4956 __u64 preemption_timer_deadline; 4957 }; 4958 4959 struct kvm_vmx_nested_state_data { 4960 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4961 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4962 }; 4963 4964This ioctl copies the vcpu's nested virtualization state from the kernel to 4965userspace. 4966 4967The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE 4968to the KVM_CHECK_EXTENSION ioctl(). 4969 49704.115 KVM_SET_NESTED_STATE 4971-------------------------- 4972 4973:Capability: KVM_CAP_NESTED_STATE 4974:Architectures: x86 4975:Type: vcpu ioctl 4976:Parameters: struct kvm_nested_state (in) 4977:Returns: 0 on success, -1 on error 4978 4979This copies the vcpu's kvm_nested_state struct from userspace to the kernel. 4980For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE. 4981 49824.116 KVM_(UN)REGISTER_COALESCED_MMIO 4983------------------------------------- 4984 4985:Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio) 4986 KVM_CAP_COALESCED_PIO (for coalesced pio) 4987:Architectures: all 4988:Type: vm ioctl 4989:Parameters: struct kvm_coalesced_mmio_zone 4990:Returns: 0 on success, < 0 on error 4991 4992Coalesced I/O is a performance optimization that defers hardware 4993register write emulation so that userspace exits are avoided. It is 4994typically used to reduce the overhead of emulating frequently accessed 4995hardware registers. 4996 4997When a hardware register is configured for coalesced I/O, write accesses 4998do not exit to userspace and their value is recorded in a ring buffer 4999that is shared between kernel and userspace. 5000 5001Coalesced I/O is used if one or more write accesses to a hardware 5002register can be deferred until a read or a write to another hardware 5003register on the same device. This last access will cause a vmexit and 5004userspace will process accesses from the ring buffer before emulating 5005it. That will avoid exiting to userspace on repeated writes. 5006 5007Coalesced pio is based on coalesced mmio. There is little difference 5008between coalesced mmio and pio except that coalesced pio records accesses 5009to I/O ports. 5010 50114.117 KVM_CLEAR_DIRTY_LOG 5012------------------------- 5013 5014:Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 5015:Architectures: x86, arm64, mips 5016:Type: vm ioctl 5017:Parameters: struct kvm_clear_dirty_log (in) 5018:Returns: 0 on success, -1 on error 5019 5020:: 5021 5022 /* for KVM_CLEAR_DIRTY_LOG */ 5023 struct kvm_clear_dirty_log { 5024 __u32 slot; 5025 __u32 num_pages; 5026 __u64 first_page; 5027 union { 5028 void __user *dirty_bitmap; /* one bit per page */ 5029 __u64 padding; 5030 }; 5031 }; 5032 5033The ioctl clears the dirty status of pages in a memory slot, according to 5034the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap 5035field. Bit 0 of the bitmap corresponds to page "first_page" in the 5036memory slot, and num_pages is the size in bits of the input bitmap. 5037first_page must be a multiple of 64; num_pages must also be a multiple of 503864 unless first_page + num_pages is the size of the memory slot. For each 5039bit that is set in the input bitmap, the corresponding page is marked "clean" 5040in KVM's dirty bitmap, and dirty tracking is re-enabled for that page 5041(for example via write-protection, or by clearing the dirty bit in 5042a page table entry). 5043 5044If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 5045the address space for which you want to clear the dirty status. See 5046KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 5047 5048This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 5049is enabled; for more information, see the description of the capability. 5050However, it can always be used as long as KVM_CHECK_EXTENSION confirms 5051that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present. 5052 50534.118 KVM_GET_SUPPORTED_HV_CPUID 5054-------------------------------- 5055 5056:Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system) 5057:Architectures: x86 5058:Type: system ioctl, vcpu ioctl 5059:Parameters: struct kvm_cpuid2 (in/out) 5060:Returns: 0 on success, -1 on error 5061 5062:: 5063 5064 struct kvm_cpuid2 { 5065 __u32 nent; 5066 __u32 padding; 5067 struct kvm_cpuid_entry2 entries[0]; 5068 }; 5069 5070 struct kvm_cpuid_entry2 { 5071 __u32 function; 5072 __u32 index; 5073 __u32 flags; 5074 __u32 eax; 5075 __u32 ebx; 5076 __u32 ecx; 5077 __u32 edx; 5078 __u32 padding[3]; 5079 }; 5080 5081This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in 5082KVM. Userspace can use the information returned by this ioctl to construct 5083cpuid information presented to guests consuming Hyper-V enlightenments (e.g. 5084Windows or Hyper-V guests). 5085 5086CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level 5087Functional Specification (TLFS). These leaves can't be obtained with 5088KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature 5089leaves (0x40000000, 0x40000001). 5090 5091Currently, the following list of CPUID leaves are returned: 5092 5093 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 5094 - HYPERV_CPUID_INTERFACE 5095 - HYPERV_CPUID_VERSION 5096 - HYPERV_CPUID_FEATURES 5097 - HYPERV_CPUID_ENLIGHTMENT_INFO 5098 - HYPERV_CPUID_IMPLEMENT_LIMITS 5099 - HYPERV_CPUID_NESTED_FEATURES 5100 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS 5101 - HYPERV_CPUID_SYNDBG_INTERFACE 5102 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 5103 5104Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure 5105with the 'nent' field indicating the number of entries in the variable-size 5106array 'entries'. If the number of entries is too low to describe all Hyper-V 5107feature leaves, an error (E2BIG) is returned. If the number is more or equal 5108to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the 5109number of valid entries in the 'entries' array, which is then filled. 5110 5111'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved, 5112userspace should not expect to get any particular value there. 5113 5114Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike 5115system ioctl which exposes all supported feature bits unconditionally, vcpu 5116version has the following quirks: 5117 5118- HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED 5119 feature bit are only exposed when Enlightened VMCS was previously enabled 5120 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS). 5121- HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC. 5122 (presumes KVM_CREATE_IRQCHIP has already been called). 5123 51244.119 KVM_ARM_VCPU_FINALIZE 5125--------------------------- 5126 5127:Architectures: arm64 5128:Type: vcpu ioctl 5129:Parameters: int feature (in) 5130:Returns: 0 on success, -1 on error 5131 5132Errors: 5133 5134 ====== ============================================================== 5135 EPERM feature not enabled, needs configuration, or already finalized 5136 EINVAL feature unknown or not present 5137 ====== ============================================================== 5138 5139Recognised values for feature: 5140 5141 ===== =========================================== 5142 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE) 5143 ===== =========================================== 5144 5145Finalizes the configuration of the specified vcpu feature. 5146 5147The vcpu must already have been initialised, enabling the affected feature, by 5148means of a successful :ref:`KVM_ARM_VCPU_INIT <KVM_ARM_VCPU_INIT>` call with the 5149appropriate flag set in features[]. 5150 5151For affected vcpu features, this is a mandatory step that must be performed 5152before the vcpu is fully usable. 5153 5154Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be 5155configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration 5156that should be performed and how to do it are feature-dependent. 5157 5158Other calls that depend on a particular feature being finalized, such as 5159KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with 5160-EPERM unless the feature has already been finalized by means of a 5161KVM_ARM_VCPU_FINALIZE call. 5162 5163See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization 5164using this ioctl. 5165 51664.120 KVM_SET_PMU_EVENT_FILTER 5167------------------------------ 5168 5169:Capability: KVM_CAP_PMU_EVENT_FILTER 5170:Architectures: x86 5171:Type: vm ioctl 5172:Parameters: struct kvm_pmu_event_filter (in) 5173:Returns: 0 on success, -1 on error 5174 5175Errors: 5176 5177 ====== ============================================================ 5178 EFAULT args[0] cannot be accessed 5179 EINVAL args[0] contains invalid data in the filter or filter events 5180 E2BIG nevents is too large 5181 EBUSY not enough memory to allocate the filter 5182 ====== ============================================================ 5183 5184:: 5185 5186 struct kvm_pmu_event_filter { 5187 __u32 action; 5188 __u32 nevents; 5189 __u32 fixed_counter_bitmap; 5190 __u32 flags; 5191 __u32 pad[4]; 5192 __u64 events[0]; 5193 }; 5194 5195This ioctl restricts the set of PMU events the guest can program by limiting 5196which event select and unit mask combinations are permitted. 5197 5198The argument holds a list of filter events which will be allowed or denied. 5199 5200Filter events only control general purpose counters; fixed purpose counters 5201are controlled by the fixed_counter_bitmap. 5202 5203Valid values for 'flags':: 5204 5205``0`` 5206 5207To use this mode, clear the 'flags' field. 5208 5209In this mode each event will contain an event select + unit mask. 5210 5211When the guest attempts to program the PMU the guest's event select + 5212unit mask is compared against the filter events to determine whether the 5213guest should have access. 5214 5215``KVM_PMU_EVENT_FLAG_MASKED_EVENTS`` 5216:Capability: KVM_CAP_PMU_EVENT_MASKED_EVENTS 5217 5218In this mode each filter event will contain an event select, mask, match, and 5219exclude value. To encode a masked event use:: 5220 5221 KVM_PMU_ENCODE_MASKED_ENTRY() 5222 5223An encoded event will follow this layout:: 5224 5225 Bits Description 5226 ---- ----------- 5227 7:0 event select (low bits) 5228 15:8 umask match 5229 31:16 unused 5230 35:32 event select (high bits) 5231 36:54 unused 5232 55 exclude bit 5233 63:56 umask mask 5234 5235When the guest attempts to program the PMU, these steps are followed in 5236determining if the guest should have access: 5237 5238 1. Match the event select from the guest against the filter events. 5239 2. If a match is found, match the guest's unit mask to the mask and match 5240 values of the included filter events. 5241 I.e. (unit mask & mask) == match && !exclude. 5242 3. If a match is found, match the guest's unit mask to the mask and match 5243 values of the excluded filter events. 5244 I.e. (unit mask & mask) == match && exclude. 5245 4. 5246 a. If an included match is found and an excluded match is not found, filter 5247 the event. 5248 b. For everything else, do not filter the event. 5249 5. 5250 a. If the event is filtered and it's an allow list, allow the guest to 5251 program the event. 5252 b. If the event is filtered and it's a deny list, do not allow the guest to 5253 program the event. 5254 5255When setting a new pmu event filter, -EINVAL will be returned if any of the 5256unused fields are set or if any of the high bits (35:32) in the event 5257select are set when called on Intel. 5258 5259Valid values for 'action':: 5260 5261 #define KVM_PMU_EVENT_ALLOW 0 5262 #define KVM_PMU_EVENT_DENY 1 5263 5264Via this API, KVM userspace can also control the behavior of the VM's fixed 5265counters (if any) by configuring the "action" and "fixed_counter_bitmap" fields. 5266 5267Specifically, KVM follows the following pseudo-code when determining whether to 5268allow the guest FixCtr[i] to count its pre-defined fixed event:: 5269 5270 FixCtr[i]_is_allowed = (action == ALLOW) && (bitmap & BIT(i)) || 5271 (action == DENY) && !(bitmap & BIT(i)); 5272 FixCtr[i]_is_denied = !FixCtr[i]_is_allowed; 5273 5274KVM always consumes fixed_counter_bitmap, it's userspace's responsibility to 5275ensure fixed_counter_bitmap is set correctly, e.g. if userspace wants to define 5276a filter that only affects general purpose counters. 5277 5278Note, the "events" field also applies to fixed counters' hardcoded event_select 5279and unit_mask values. "fixed_counter_bitmap" has higher priority than "events" 5280if there is a contradiction between the two. 5281 52824.121 KVM_PPC_SVM_OFF 5283--------------------- 5284 5285:Capability: basic 5286:Architectures: powerpc 5287:Type: vm ioctl 5288:Parameters: none 5289:Returns: 0 on successful completion, 5290 5291Errors: 5292 5293 ====== ================================================================ 5294 EINVAL if ultravisor failed to terminate the secure guest 5295 ENOMEM if hypervisor failed to allocate new radix page tables for guest 5296 ====== ================================================================ 5297 5298This ioctl is used to turn off the secure mode of the guest or transition 5299the guest from secure mode to normal mode. This is invoked when the guest 5300is reset. This has no effect if called for a normal guest. 5301 5302This ioctl issues an ultravisor call to terminate the secure guest, 5303unpins the VPA pages and releases all the device pages that are used to 5304track the secure pages by hypervisor. 5305 53064.122 KVM_S390_NORMAL_RESET 5307--------------------------- 5308 5309:Capability: KVM_CAP_S390_VCPU_RESETS 5310:Architectures: s390 5311:Type: vcpu ioctl 5312:Parameters: none 5313:Returns: 0 5314 5315This ioctl resets VCPU registers and control structures according to 5316the cpu reset definition in the POP (Principles Of Operation). 5317 53184.123 KVM_S390_INITIAL_RESET 5319---------------------------- 5320 5321:Capability: basic 5322:Architectures: s390 5323:Type: vcpu ioctl 5324:Parameters: none 5325:Returns: 0 5326 5327This ioctl resets VCPU registers and control structures according to 5328the initial cpu reset definition in the POP. However, the cpu is not 5329put into ESA mode. This reset is a superset of the normal reset. 5330 53314.124 KVM_S390_CLEAR_RESET 5332-------------------------- 5333 5334:Capability: KVM_CAP_S390_VCPU_RESETS 5335:Architectures: s390 5336:Type: vcpu ioctl 5337:Parameters: none 5338:Returns: 0 5339 5340This ioctl resets VCPU registers and control structures according to 5341the clear cpu reset definition in the POP. However, the cpu is not put 5342into ESA mode. This reset is a superset of the initial reset. 5343 5344 53454.125 KVM_S390_PV_COMMAND 5346------------------------- 5347 5348:Capability: KVM_CAP_S390_PROTECTED 5349:Architectures: s390 5350:Type: vm ioctl 5351:Parameters: struct kvm_pv_cmd 5352:Returns: 0 on success, < 0 on error 5353 5354:: 5355 5356 struct kvm_pv_cmd { 5357 __u32 cmd; /* Command to be executed */ 5358 __u16 rc; /* Ultravisor return code */ 5359 __u16 rrc; /* Ultravisor return reason code */ 5360 __u64 data; /* Data or address */ 5361 __u32 flags; /* flags for future extensions. Must be 0 for now */ 5362 __u32 reserved[3]; 5363 }; 5364 5365**Ultravisor return codes** 5366The Ultravisor return (reason) codes are provided by the kernel if a 5367Ultravisor call has been executed to achieve the results expected by 5368the command. Therefore they are independent of the IOCTL return 5369code. If KVM changes `rc`, its value will always be greater than 0 5370hence setting it to 0 before issuing a PV command is advised to be 5371able to detect a change of `rc`. 5372 5373**cmd values:** 5374 5375KVM_PV_ENABLE 5376 Allocate memory and register the VM with the Ultravisor, thereby 5377 donating memory to the Ultravisor that will become inaccessible to 5378 KVM. All existing CPUs are converted to protected ones. After this 5379 command has succeeded, any CPU added via hotplug will become 5380 protected during its creation as well. 5381 5382 Errors: 5383 5384 ===== ============================= 5385 EINTR an unmasked signal is pending 5386 ===== ============================= 5387 5388KVM_PV_DISABLE 5389 Deregister the VM from the Ultravisor and reclaim the memory that had 5390 been donated to the Ultravisor, making it usable by the kernel again. 5391 All registered VCPUs are converted back to non-protected ones. If a 5392 previous protected VM had been prepared for asynchronous teardown with 5393 KVM_PV_ASYNC_CLEANUP_PREPARE and not subsequently torn down with 5394 KVM_PV_ASYNC_CLEANUP_PERFORM, it will be torn down in this call 5395 together with the current protected VM. 5396 5397KVM_PV_VM_SET_SEC_PARMS 5398 Pass the image header from VM memory to the Ultravisor in 5399 preparation of image unpacking and verification. 5400 5401KVM_PV_VM_UNPACK 5402 Unpack (protect and decrypt) a page of the encrypted boot image. 5403 5404KVM_PV_VM_VERIFY 5405 Verify the integrity of the unpacked image. Only if this succeeds, 5406 KVM is allowed to start protected VCPUs. 5407 5408KVM_PV_INFO 5409 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5410 5411 Presents an API that provides Ultravisor related data to userspace 5412 via subcommands. len_max is the size of the user space buffer, 5413 len_written is KVM's indication of how much bytes of that buffer 5414 were actually written to. len_written can be used to determine the 5415 valid fields if more response fields are added in the future. 5416 5417 :: 5418 5419 enum pv_cmd_info_id { 5420 KVM_PV_INFO_VM, 5421 KVM_PV_INFO_DUMP, 5422 }; 5423 5424 struct kvm_s390_pv_info_header { 5425 __u32 id; 5426 __u32 len_max; 5427 __u32 len_written; 5428 __u32 reserved; 5429 }; 5430 5431 struct kvm_s390_pv_info { 5432 struct kvm_s390_pv_info_header header; 5433 struct kvm_s390_pv_info_dump dump; 5434 struct kvm_s390_pv_info_vm vm; 5435 }; 5436 5437**subcommands:** 5438 5439 KVM_PV_INFO_VM 5440 This subcommand provides basic Ultravisor information for PV 5441 hosts. These values are likely also exported as files in the sysfs 5442 firmware UV query interface but they are more easily available to 5443 programs in this API. 5444 5445 The installed calls and feature_indication members provide the 5446 installed UV calls and the UV's other feature indications. 5447 5448 The max_* members provide information about the maximum number of PV 5449 vcpus, PV guests and PV guest memory size. 5450 5451 :: 5452 5453 struct kvm_s390_pv_info_vm { 5454 __u64 inst_calls_list[4]; 5455 __u64 max_cpus; 5456 __u64 max_guests; 5457 __u64 max_guest_addr; 5458 __u64 feature_indication; 5459 }; 5460 5461 5462 KVM_PV_INFO_DUMP 5463 This subcommand provides information related to dumping PV guests. 5464 5465 :: 5466 5467 struct kvm_s390_pv_info_dump { 5468 __u64 dump_cpu_buffer_len; 5469 __u64 dump_config_mem_buffer_per_1m; 5470 __u64 dump_config_finalize_len; 5471 }; 5472 5473KVM_PV_DUMP 5474 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5475 5476 Presents an API that provides calls which facilitate dumping a 5477 protected VM. 5478 5479 :: 5480 5481 struct kvm_s390_pv_dmp { 5482 __u64 subcmd; 5483 __u64 buff_addr; 5484 __u64 buff_len; 5485 __u64 gaddr; /* For dump storage state */ 5486 }; 5487 5488 **subcommands:** 5489 5490 KVM_PV_DUMP_INIT 5491 Initializes the dump process of a protected VM. If this call does 5492 not succeed all other subcommands will fail with -EINVAL. This 5493 subcommand will return -EINVAL if a dump process has not yet been 5494 completed. 5495 5496 Not all PV vms can be dumped, the owner needs to set `dump 5497 allowed` PCF bit 34 in the SE header to allow dumping. 5498 5499 KVM_PV_DUMP_CONFIG_STOR_STATE 5500 Stores `buff_len` bytes of tweak component values starting with 5501 the 1MB block specified by the absolute guest address 5502 (`gaddr`). `buff_len` needs to be `conf_dump_storage_state_len` 5503 aligned and at least >= the `conf_dump_storage_state_len` value 5504 provided by the dump uv_info data. buff_user might be written to 5505 even if an error rc is returned. For instance if we encounter a 5506 fault after writing the first page of data. 5507 5508 KVM_PV_DUMP_COMPLETE 5509 If the subcommand succeeds it completes the dump process and lets 5510 KVM_PV_DUMP_INIT be called again. 5511 5512 On success `conf_dump_finalize_len` bytes of completion data will be 5513 stored to the `buff_addr`. The completion data contains a key 5514 derivation seed, IV, tweak nonce and encryption keys as well as an 5515 authentication tag all of which are needed to decrypt the dump at a 5516 later time. 5517 5518KVM_PV_ASYNC_CLEANUP_PREPARE 5519 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE 5520 5521 Prepare the current protected VM for asynchronous teardown. Most 5522 resources used by the current protected VM will be set aside for a 5523 subsequent asynchronous teardown. The current protected VM will then 5524 resume execution immediately as non-protected. There can be at most 5525 one protected VM prepared for asynchronous teardown at any time. If 5526 a protected VM had already been prepared for teardown without 5527 subsequently calling KVM_PV_ASYNC_CLEANUP_PERFORM, this call will 5528 fail. In that case, the userspace process should issue a normal 5529 KVM_PV_DISABLE. The resources set aside with this call will need to 5530 be cleaned up with a subsequent call to KVM_PV_ASYNC_CLEANUP_PERFORM 5531 or KVM_PV_DISABLE, otherwise they will be cleaned up when KVM 5532 terminates. KVM_PV_ASYNC_CLEANUP_PREPARE can be called again as soon 5533 as cleanup starts, i.e. before KVM_PV_ASYNC_CLEANUP_PERFORM finishes. 5534 5535KVM_PV_ASYNC_CLEANUP_PERFORM 5536 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE 5537 5538 Tear down the protected VM previously prepared for teardown with 5539 KVM_PV_ASYNC_CLEANUP_PREPARE. The resources that had been set aside 5540 will be freed during the execution of this command. This PV command 5541 should ideally be issued by userspace from a separate thread. If a 5542 fatal signal is received (or the process terminates naturally), the 5543 command will terminate immediately without completing, and the normal 5544 KVM shutdown procedure will take care of cleaning up all remaining 5545 protected VMs, including the ones whose teardown was interrupted by 5546 process termination. 5547 55484.126 KVM_XEN_HVM_SET_ATTR 5549-------------------------- 5550 5551:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5552:Architectures: x86 5553:Type: vm ioctl 5554:Parameters: struct kvm_xen_hvm_attr 5555:Returns: 0 on success, < 0 on error 5556 5557:: 5558 5559 struct kvm_xen_hvm_attr { 5560 __u16 type; 5561 __u16 pad[3]; 5562 union { 5563 __u8 long_mode; 5564 __u8 vector; 5565 __u8 runstate_update_flag; 5566 union { 5567 __u64 gfn; 5568 __u64 hva; 5569 } shared_info; 5570 struct { 5571 __u32 send_port; 5572 __u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */ 5573 __u32 flags; 5574 union { 5575 struct { 5576 __u32 port; 5577 __u32 vcpu; 5578 __u32 priority; 5579 } port; 5580 struct { 5581 __u32 port; /* Zero for eventfd */ 5582 __s32 fd; 5583 } eventfd; 5584 __u32 padding[4]; 5585 } deliver; 5586 } evtchn; 5587 __u32 xen_version; 5588 __u64 pad[8]; 5589 } u; 5590 }; 5591 5592type values: 5593 5594KVM_XEN_ATTR_TYPE_LONG_MODE 5595 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This 5596 determines the layout of the shared_info page exposed to the VM. 5597 5598KVM_XEN_ATTR_TYPE_SHARED_INFO 5599 Sets the guest physical frame number at which the Xen shared_info 5600 page resides. Note that although Xen places vcpu_info for the first 5601 32 vCPUs in the shared_info page, KVM does not automatically do so 5602 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO or 5603 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA be used explicitly even when 5604 the vcpu_info for a given vCPU resides at the "default" location 5605 in the shared_info page. This is because KVM may not be aware of 5606 the Xen CPU id which is used as the index into the vcpu_info[] 5607 array, so may know the correct default location. 5608 5609 Note that the shared_info page may be constantly written to by KVM; 5610 it contains the event channel bitmap used to deliver interrupts to 5611 a Xen guest, amongst other things. It is exempt from dirty tracking 5612 mechanisms — KVM will not explicitly mark the page as dirty each 5613 time an event channel interrupt is delivered to the guest! Thus, 5614 userspace should always assume that the designated GFN is dirty if 5615 any vCPU has been running or any event channel interrupts can be 5616 routed to the guest. 5617 5618 Setting the gfn to KVM_XEN_INVALID_GFN will disable the shared_info 5619 page. 5620 5621KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA 5622 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA flag is also set in the 5623 Xen capabilities, then this attribute may be used to set the 5624 userspace address at which the shared_info page resides, which 5625 will always be fixed in the VMM regardless of where it is mapped 5626 in guest physical address space. This attribute should be used in 5627 preference to KVM_XEN_ATTR_TYPE_SHARED_INFO as it avoids 5628 unnecessary invalidation of an internal cache when the page is 5629 re-mapped in guest physical address space. 5630 5631 Setting the hva to zero will disable the shared_info page. 5632 5633KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 5634 Sets the exception vector used to deliver Xen event channel upcalls. 5635 This is the HVM-wide vector injected directly by the hypervisor 5636 (not through the local APIC), typically configured by a guest via 5637 HVM_PARAM_CALLBACK_IRQ. This can be disabled again (e.g. for guest 5638 SHUTDOWN_soft_reset) by setting it to zero. 5639 5640KVM_XEN_ATTR_TYPE_EVTCHN 5641 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5642 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5643 an outbound port number for interception of EVTCHNOP_send requests 5644 from the guest. A given sending port number may be directed back to 5645 a specified vCPU (by APIC ID) / port / priority on the guest, or to 5646 trigger events on an eventfd. The vCPU and priority can be changed 5647 by setting KVM_XEN_EVTCHN_UPDATE in a subsequent call, but other 5648 fields cannot change for a given sending port. A port mapping is 5649 removed by using KVM_XEN_EVTCHN_DEASSIGN in the flags field. Passing 5650 KVM_XEN_EVTCHN_RESET in the flags field removes all interception of 5651 outbound event channels. The values of the flags field are mutually 5652 exclusive and cannot be combined as a bitmask. 5653 5654KVM_XEN_ATTR_TYPE_XEN_VERSION 5655 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5656 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5657 the 32-bit version code returned to the guest when it invokes the 5658 XENVER_version call; typically (XEN_MAJOR << 16 | XEN_MINOR). PV 5659 Xen guests will often use this to as a dummy hypercall to trigger 5660 event channel delivery, so responding within the kernel without 5661 exiting to userspace is beneficial. 5662 5663KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG 5664 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5665 support for KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG. It enables the 5666 XEN_RUNSTATE_UPDATE flag which allows guest vCPUs to safely read 5667 other vCPUs' vcpu_runstate_info. Xen guests enable this feature via 5668 the VMASST_TYPE_runstate_update_flag of the HYPERVISOR_vm_assist 5669 hypercall. 5670 56714.127 KVM_XEN_HVM_GET_ATTR 5672-------------------------- 5673 5674:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5675:Architectures: x86 5676:Type: vm ioctl 5677:Parameters: struct kvm_xen_hvm_attr 5678:Returns: 0 on success, < 0 on error 5679 5680Allows Xen VM attributes to be read. For the structure and types, 5681see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_ATTR_TYPE_EVTCHN 5682attribute cannot be read. 5683 56844.128 KVM_XEN_VCPU_SET_ATTR 5685--------------------------- 5686 5687:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5688:Architectures: x86 5689:Type: vcpu ioctl 5690:Parameters: struct kvm_xen_vcpu_attr 5691:Returns: 0 on success, < 0 on error 5692 5693:: 5694 5695 struct kvm_xen_vcpu_attr { 5696 __u16 type; 5697 __u16 pad[3]; 5698 union { 5699 __u64 gpa; 5700 __u64 pad[4]; 5701 struct { 5702 __u64 state; 5703 __u64 state_entry_time; 5704 __u64 time_running; 5705 __u64 time_runnable; 5706 __u64 time_blocked; 5707 __u64 time_offline; 5708 } runstate; 5709 __u32 vcpu_id; 5710 struct { 5711 __u32 port; 5712 __u32 priority; 5713 __u64 expires_ns; 5714 } timer; 5715 __u8 vector; 5716 } u; 5717 }; 5718 5719type values: 5720 5721KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 5722 Sets the guest physical address of the vcpu_info for a given vCPU. 5723 As with the shared_info page for the VM, the corresponding page may be 5724 dirtied at any time if event channel interrupt delivery is enabled, so 5725 userspace should always assume that the page is dirty without relying 5726 on dirty logging. Setting the gpa to KVM_XEN_INVALID_GPA will disable 5727 the vcpu_info. 5728 5729KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA 5730 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA flag is also set in the 5731 Xen capabilities, then this attribute may be used to set the 5732 userspace address of the vcpu_info for a given vCPU. It should 5733 only be used when the vcpu_info resides at the "default" location 5734 in the shared_info page. In this case it is safe to assume the 5735 userspace address will not change, because the shared_info page is 5736 an overlay on guest memory and remains at a fixed host address 5737 regardless of where it is mapped in guest physical address space 5738 and hence unnecessary invalidation of an internal cache may be 5739 avoided if the guest memory layout is modified. 5740 If the vcpu_info does not reside at the "default" location then 5741 it is not guaranteed to remain at the same host address and 5742 hence the aforementioned cache invalidation is required. 5743 5744KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 5745 Sets the guest physical address of an additional pvclock structure 5746 for a given vCPU. This is typically used for guest vsyscall support. 5747 Setting the gpa to KVM_XEN_INVALID_GPA will disable the structure. 5748 5749KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 5750 Sets the guest physical address of the vcpu_runstate_info for a given 5751 vCPU. This is how a Xen guest tracks CPU state such as steal time. 5752 Setting the gpa to KVM_XEN_INVALID_GPA will disable the runstate area. 5753 5754KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 5755 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of 5756 the given vCPU from the .u.runstate.state member of the structure. 5757 KVM automatically accounts running and runnable time but blocked 5758 and offline states are only entered explicitly. 5759 5760KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 5761 Sets all fields of the vCPU runstate data from the .u.runstate member 5762 of the structure, including the current runstate. The state_entry_time 5763 must equal the sum of the other four times. 5764 5765KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 5766 This *adds* the contents of the .u.runstate members of the structure 5767 to the corresponding members of the given vCPU's runstate data, thus 5768 permitting atomic adjustments to the runstate times. The adjustment 5769 to the state_entry_time must equal the sum of the adjustments to the 5770 other four times. The state field must be set to -1, or to a valid 5771 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked 5772 or RUNSTATE_offline) to set the current accounted state as of the 5773 adjusted state_entry_time. 5774 5775KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID 5776 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5777 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the Xen 5778 vCPU ID of the given vCPU, to allow timer-related VCPU operations to 5779 be intercepted by KVM. 5780 5781KVM_XEN_VCPU_ATTR_TYPE_TIMER 5782 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5783 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5784 event channel port/priority for the VIRQ_TIMER of the vCPU, as well 5785 as allowing a pending timer to be saved/restored. Setting the timer 5786 port to zero disables kernel handling of the singleshot timer. 5787 5788KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR 5789 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5790 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5791 per-vCPU local APIC upcall vector, configured by a Xen guest with 5792 the HVMOP_set_evtchn_upcall_vector hypercall. This is typically 5793 used by Windows guests, and is distinct from the HVM-wide upcall 5794 vector configured with HVM_PARAM_CALLBACK_IRQ. It is disabled by 5795 setting the vector to zero. 5796 5797 57984.129 KVM_XEN_VCPU_GET_ATTR 5799--------------------------- 5800 5801:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5802:Architectures: x86 5803:Type: vcpu ioctl 5804:Parameters: struct kvm_xen_vcpu_attr 5805:Returns: 0 on success, < 0 on error 5806 5807Allows Xen vCPU attributes to be read. For the structure and types, 5808see KVM_XEN_VCPU_SET_ATTR above. 5809 5810The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used 5811with the KVM_XEN_VCPU_GET_ATTR ioctl. 5812 58134.130 KVM_ARM_MTE_COPY_TAGS 5814--------------------------- 5815 5816:Capability: KVM_CAP_ARM_MTE 5817:Architectures: arm64 5818:Type: vm ioctl 5819:Parameters: struct kvm_arm_copy_mte_tags 5820:Returns: number of bytes copied, < 0 on error (-EINVAL for incorrect 5821 arguments, -EFAULT if memory cannot be accessed). 5822 5823:: 5824 5825 struct kvm_arm_copy_mte_tags { 5826 __u64 guest_ipa; 5827 __u64 length; 5828 void __user *addr; 5829 __u64 flags; 5830 __u64 reserved[2]; 5831 }; 5832 5833Copies Memory Tagging Extension (MTE) tags to/from guest tag memory. The 5834``guest_ipa`` and ``length`` fields must be ``PAGE_SIZE`` aligned. 5835``length`` must not be bigger than 2^31 - PAGE_SIZE bytes. The ``addr`` 5836field must point to a buffer which the tags will be copied to or from. 5837 5838``flags`` specifies the direction of copy, either ``KVM_ARM_TAGS_TO_GUEST`` or 5839``KVM_ARM_TAGS_FROM_GUEST``. 5840 5841The size of the buffer to store the tags is ``(length / 16)`` bytes 5842(granules in MTE are 16 bytes long). Each byte contains a single tag 5843value. This matches the format of ``PTRACE_PEEKMTETAGS`` and 5844``PTRACE_POKEMTETAGS``. 5845 5846If an error occurs before any data is copied then a negative error code is 5847returned. If some tags have been copied before an error occurs then the number 5848of bytes successfully copied is returned. If the call completes successfully 5849then ``length`` is returned. 5850 58514.131 KVM_GET_SREGS2 5852-------------------- 5853 5854:Capability: KVM_CAP_SREGS2 5855:Architectures: x86 5856:Type: vcpu ioctl 5857:Parameters: struct kvm_sregs2 (out) 5858:Returns: 0 on success, -1 on error 5859 5860Reads special registers from the vcpu. 5861This ioctl (when supported) replaces the KVM_GET_SREGS. 5862 5863:: 5864 5865 struct kvm_sregs2 { 5866 /* out (KVM_GET_SREGS2) / in (KVM_SET_SREGS2) */ 5867 struct kvm_segment cs, ds, es, fs, gs, ss; 5868 struct kvm_segment tr, ldt; 5869 struct kvm_dtable gdt, idt; 5870 __u64 cr0, cr2, cr3, cr4, cr8; 5871 __u64 efer; 5872 __u64 apic_base; 5873 __u64 flags; 5874 __u64 pdptrs[4]; 5875 }; 5876 5877flags values for ``kvm_sregs2``: 5878 5879``KVM_SREGS2_FLAGS_PDPTRS_VALID`` 5880 5881 Indicates that the struct contains valid PDPTR values. 5882 5883 58844.132 KVM_SET_SREGS2 5885-------------------- 5886 5887:Capability: KVM_CAP_SREGS2 5888:Architectures: x86 5889:Type: vcpu ioctl 5890:Parameters: struct kvm_sregs2 (in) 5891:Returns: 0 on success, -1 on error 5892 5893Writes special registers into the vcpu. 5894See KVM_GET_SREGS2 for the data structures. 5895This ioctl (when supported) replaces the KVM_SET_SREGS. 5896 58974.133 KVM_GET_STATS_FD 5898---------------------- 5899 5900:Capability: KVM_CAP_STATS_BINARY_FD 5901:Architectures: all 5902:Type: vm ioctl, vcpu ioctl 5903:Parameters: none 5904:Returns: statistics file descriptor on success, < 0 on error 5905 5906Errors: 5907 5908 ====== ====================================================== 5909 ENOMEM if the fd could not be created due to lack of memory 5910 EMFILE if the number of opened files exceeds the limit 5911 ====== ====================================================== 5912 5913The returned file descriptor can be used to read VM/vCPU statistics data in 5914binary format. The data in the file descriptor consists of four blocks 5915organized as follows: 5916 5917+-------------+ 5918| Header | 5919+-------------+ 5920| id string | 5921+-------------+ 5922| Descriptors | 5923+-------------+ 5924| Stats Data | 5925+-------------+ 5926 5927Apart from the header starting at offset 0, please be aware that it is 5928not guaranteed that the four blocks are adjacent or in the above order; 5929the offsets of the id, descriptors and data blocks are found in the 5930header. However, all four blocks are aligned to 64 bit offsets in the 5931file and they do not overlap. 5932 5933All blocks except the data block are immutable. Userspace can read them 5934only one time after retrieving the file descriptor, and then use ``pread`` or 5935``lseek`` to read the statistics repeatedly. 5936 5937All data is in system endianness. 5938 5939The format of the header is as follows:: 5940 5941 struct kvm_stats_header { 5942 __u32 flags; 5943 __u32 name_size; 5944 __u32 num_desc; 5945 __u32 id_offset; 5946 __u32 desc_offset; 5947 __u32 data_offset; 5948 }; 5949 5950The ``flags`` field is not used at the moment. It is always read as 0. 5951 5952The ``name_size`` field is the size (in byte) of the statistics name string 5953(including trailing '\0') which is contained in the "id string" block and 5954appended at the end of every descriptor. 5955 5956The ``num_desc`` field is the number of descriptors that are included in the 5957descriptor block. (The actual number of values in the data block may be 5958larger, since each descriptor may comprise more than one value). 5959 5960The ``id_offset`` field is the offset of the id string from the start of the 5961file indicated by the file descriptor. It is a multiple of 8. 5962 5963The ``desc_offset`` field is the offset of the Descriptors block from the start 5964of the file indicated by the file descriptor. It is a multiple of 8. 5965 5966The ``data_offset`` field is the offset of the Stats Data block from the start 5967of the file indicated by the file descriptor. It is a multiple of 8. 5968 5969The id string block contains a string which identifies the file descriptor on 5970which KVM_GET_STATS_FD was invoked. The size of the block, including the 5971trailing ``'\0'``, is indicated by the ``name_size`` field in the header. 5972 5973The descriptors block is only needed to be read once for the lifetime of the 5974file descriptor contains a sequence of ``struct kvm_stats_desc``, each followed 5975by a string of size ``name_size``. 5976:: 5977 5978 #define KVM_STATS_TYPE_SHIFT 0 5979 #define KVM_STATS_TYPE_MASK (0xF << KVM_STATS_TYPE_SHIFT) 5980 #define KVM_STATS_TYPE_CUMULATIVE (0x0 << KVM_STATS_TYPE_SHIFT) 5981 #define KVM_STATS_TYPE_INSTANT (0x1 << KVM_STATS_TYPE_SHIFT) 5982 #define KVM_STATS_TYPE_PEAK (0x2 << KVM_STATS_TYPE_SHIFT) 5983 #define KVM_STATS_TYPE_LINEAR_HIST (0x3 << KVM_STATS_TYPE_SHIFT) 5984 #define KVM_STATS_TYPE_LOG_HIST (0x4 << KVM_STATS_TYPE_SHIFT) 5985 #define KVM_STATS_TYPE_MAX KVM_STATS_TYPE_LOG_HIST 5986 5987 #define KVM_STATS_UNIT_SHIFT 4 5988 #define KVM_STATS_UNIT_MASK (0xF << KVM_STATS_UNIT_SHIFT) 5989 #define KVM_STATS_UNIT_NONE (0x0 << KVM_STATS_UNIT_SHIFT) 5990 #define KVM_STATS_UNIT_BYTES (0x1 << KVM_STATS_UNIT_SHIFT) 5991 #define KVM_STATS_UNIT_SECONDS (0x2 << KVM_STATS_UNIT_SHIFT) 5992 #define KVM_STATS_UNIT_CYCLES (0x3 << KVM_STATS_UNIT_SHIFT) 5993 #define KVM_STATS_UNIT_BOOLEAN (0x4 << KVM_STATS_UNIT_SHIFT) 5994 #define KVM_STATS_UNIT_MAX KVM_STATS_UNIT_BOOLEAN 5995 5996 #define KVM_STATS_BASE_SHIFT 8 5997 #define KVM_STATS_BASE_MASK (0xF << KVM_STATS_BASE_SHIFT) 5998 #define KVM_STATS_BASE_POW10 (0x0 << KVM_STATS_BASE_SHIFT) 5999 #define KVM_STATS_BASE_POW2 (0x1 << KVM_STATS_BASE_SHIFT) 6000 #define KVM_STATS_BASE_MAX KVM_STATS_BASE_POW2 6001 6002 struct kvm_stats_desc { 6003 __u32 flags; 6004 __s16 exponent; 6005 __u16 size; 6006 __u32 offset; 6007 __u32 bucket_size; 6008 char name[]; 6009 }; 6010 6011The ``flags`` field contains the type and unit of the statistics data described 6012by this descriptor. Its endianness is CPU native. 6013The following flags are supported: 6014 6015Bits 0-3 of ``flags`` encode the type: 6016 6017 * ``KVM_STATS_TYPE_CUMULATIVE`` 6018 The statistics reports a cumulative count. The value of data can only be increased. 6019 Most of the counters used in KVM are of this type. 6020 The corresponding ``size`` field for this type is always 1. 6021 All cumulative statistics data are read/write. 6022 * ``KVM_STATS_TYPE_INSTANT`` 6023 The statistics reports an instantaneous value. Its value can be increased or 6024 decreased. This type is usually used as a measurement of some resources, 6025 like the number of dirty pages, the number of large pages, etc. 6026 All instant statistics are read only. 6027 The corresponding ``size`` field for this type is always 1. 6028 * ``KVM_STATS_TYPE_PEAK`` 6029 The statistics data reports a peak value, for example the maximum number 6030 of items in a hash table bucket, the longest time waited and so on. 6031 The value of data can only be increased. 6032 The corresponding ``size`` field for this type is always 1. 6033 * ``KVM_STATS_TYPE_LINEAR_HIST`` 6034 The statistic is reported as a linear histogram. The number of 6035 buckets is specified by the ``size`` field. The size of buckets is specified 6036 by the ``hist_param`` field. The range of the Nth bucket (1 <= N < ``size``) 6037 is [``hist_param``*(N-1), ``hist_param``*N), while the range of the last 6038 bucket is [``hist_param``*(``size``-1), +INF). (+INF means positive infinity 6039 value.) 6040 * ``KVM_STATS_TYPE_LOG_HIST`` 6041 The statistic is reported as a logarithmic histogram. The number of 6042 buckets is specified by the ``size`` field. The range of the first bucket is 6043 [0, 1), while the range of the last bucket is [pow(2, ``size``-2), +INF). 6044 Otherwise, The Nth bucket (1 < N < ``size``) covers 6045 [pow(2, N-2), pow(2, N-1)). 6046 6047Bits 4-7 of ``flags`` encode the unit: 6048 6049 * ``KVM_STATS_UNIT_NONE`` 6050 There is no unit for the value of statistics data. This usually means that 6051 the value is a simple counter of an event. 6052 * ``KVM_STATS_UNIT_BYTES`` 6053 It indicates that the statistics data is used to measure memory size, in the 6054 unit of Byte, KiByte, MiByte, GiByte, etc. The unit of the data is 6055 determined by the ``exponent`` field in the descriptor. 6056 * ``KVM_STATS_UNIT_SECONDS`` 6057 It indicates that the statistics data is used to measure time or latency. 6058 * ``KVM_STATS_UNIT_CYCLES`` 6059 It indicates that the statistics data is used to measure CPU clock cycles. 6060 * ``KVM_STATS_UNIT_BOOLEAN`` 6061 It indicates that the statistic will always be either 0 or 1. Boolean 6062 statistics of "peak" type will never go back from 1 to 0. Boolean 6063 statistics can be linear histograms (with two buckets) but not logarithmic 6064 histograms. 6065 6066Note that, in the case of histograms, the unit applies to the bucket 6067ranges, while the bucket value indicates how many samples fell in the 6068bucket's range. 6069 6070Bits 8-11 of ``flags``, together with ``exponent``, encode the scale of the 6071unit: 6072 6073 * ``KVM_STATS_BASE_POW10`` 6074 The scale is based on power of 10. It is used for measurement of time and 6075 CPU clock cycles. For example, an exponent of -9 can be used with 6076 ``KVM_STATS_UNIT_SECONDS`` to express that the unit is nanoseconds. 6077 * ``KVM_STATS_BASE_POW2`` 6078 The scale is based on power of 2. It is used for measurement of memory size. 6079 For example, an exponent of 20 can be used with ``KVM_STATS_UNIT_BYTES`` to 6080 express that the unit is MiB. 6081 6082The ``size`` field is the number of values of this statistics data. Its 6083value is usually 1 for most of simple statistics. 1 means it contains an 6084unsigned 64bit data. 6085 6086The ``offset`` field is the offset from the start of Data Block to the start of 6087the corresponding statistics data. 6088 6089The ``bucket_size`` field is used as a parameter for histogram statistics data. 6090It is only used by linear histogram statistics data, specifying the size of a 6091bucket in the unit expressed by bits 4-11 of ``flags`` together with ``exponent``. 6092 6093The ``name`` field is the name string of the statistics data. The name string 6094starts at the end of ``struct kvm_stats_desc``. The maximum length including 6095the trailing ``'\0'``, is indicated by ``name_size`` in the header. 6096 6097The Stats Data block contains an array of 64-bit values in the same order 6098as the descriptors in Descriptors block. 6099 61004.134 KVM_GET_XSAVE2 6101-------------------- 6102 6103:Capability: KVM_CAP_XSAVE2 6104:Architectures: x86 6105:Type: vcpu ioctl 6106:Parameters: struct kvm_xsave (out) 6107:Returns: 0 on success, -1 on error 6108 6109 6110:: 6111 6112 struct kvm_xsave { 6113 __u32 region[1024]; 6114 __u32 extra[0]; 6115 }; 6116 6117This ioctl would copy current vcpu's xsave struct to the userspace. It 6118copies as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) 6119when invoked on the vm file descriptor. The size value returned by 6120KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 6121Currently, it is only greater than 4096 if a dynamic feature has been 6122enabled with ``arch_prctl()``, but this may change in the future. 6123 6124The offsets of the state save areas in struct kvm_xsave follow the contents 6125of CPUID leaf 0xD on the host. 6126 61274.135 KVM_XEN_HVM_EVTCHN_SEND 6128----------------------------- 6129 6130:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND 6131:Architectures: x86 6132:Type: vm ioctl 6133:Parameters: struct kvm_irq_routing_xen_evtchn 6134:Returns: 0 on success, < 0 on error 6135 6136 6137:: 6138 6139 struct kvm_irq_routing_xen_evtchn { 6140 __u32 port; 6141 __u32 vcpu; 6142 __u32 priority; 6143 }; 6144 6145This ioctl injects an event channel interrupt directly to the guest vCPU. 6146 61474.136 KVM_S390_PV_CPU_COMMAND 6148----------------------------- 6149 6150:Capability: KVM_CAP_S390_PROTECTED_DUMP 6151:Architectures: s390 6152:Type: vcpu ioctl 6153:Parameters: none 6154:Returns: 0 on success, < 0 on error 6155 6156This ioctl closely mirrors `KVM_S390_PV_COMMAND` but handles requests 6157for vcpus. It re-uses the kvm_s390_pv_dmp struct and hence also shares 6158the command ids. 6159 6160**command:** 6161 6162KVM_PV_DUMP 6163 Presents an API that provides calls which facilitate dumping a vcpu 6164 of a protected VM. 6165 6166**subcommand:** 6167 6168KVM_PV_DUMP_CPU 6169 Provides encrypted dump data like register values. 6170 The length of the returned data is provided by uv_info.guest_cpu_stor_len. 6171 61724.137 KVM_S390_ZPCI_OP 6173---------------------- 6174 6175:Capability: KVM_CAP_S390_ZPCI_OP 6176:Architectures: s390 6177:Type: vm ioctl 6178:Parameters: struct kvm_s390_zpci_op (in) 6179:Returns: 0 on success, <0 on error 6180 6181Used to manage hardware-assisted virtualization features for zPCI devices. 6182 6183Parameters are specified via the following structure:: 6184 6185 struct kvm_s390_zpci_op { 6186 /* in */ 6187 __u32 fh; /* target device */ 6188 __u8 op; /* operation to perform */ 6189 __u8 pad[3]; 6190 union { 6191 /* for KVM_S390_ZPCIOP_REG_AEN */ 6192 struct { 6193 __u64 ibv; /* Guest addr of interrupt bit vector */ 6194 __u64 sb; /* Guest addr of summary bit */ 6195 __u32 flags; 6196 __u32 noi; /* Number of interrupts */ 6197 __u8 isc; /* Guest interrupt subclass */ 6198 __u8 sbo; /* Offset of guest summary bit vector */ 6199 __u16 pad; 6200 } reg_aen; 6201 __u64 reserved[8]; 6202 } u; 6203 }; 6204 6205The type of operation is specified in the "op" field. 6206KVM_S390_ZPCIOP_REG_AEN is used to register the VM for adapter event 6207notification interpretation, which will allow firmware delivery of adapter 6208events directly to the vm, with KVM providing a backup delivery mechanism; 6209KVM_S390_ZPCIOP_DEREG_AEN is used to subsequently disable interpretation of 6210adapter event notifications. 6211 6212The target zPCI function must also be specified via the "fh" field. For the 6213KVM_S390_ZPCIOP_REG_AEN operation, additional information to establish firmware 6214delivery must be provided via the "reg_aen" struct. 6215 6216The "pad" and "reserved" fields may be used for future extensions and should be 6217set to 0s by userspace. 6218 62194.138 KVM_ARM_SET_COUNTER_OFFSET 6220-------------------------------- 6221 6222:Capability: KVM_CAP_COUNTER_OFFSET 6223:Architectures: arm64 6224:Type: vm ioctl 6225:Parameters: struct kvm_arm_counter_offset (in) 6226:Returns: 0 on success, < 0 on error 6227 6228This capability indicates that userspace is able to apply a single VM-wide 6229offset to both the virtual and physical counters as viewed by the guest 6230using the KVM_ARM_SET_CNT_OFFSET ioctl and the following data structure: 6231 6232:: 6233 6234 struct kvm_arm_counter_offset { 6235 __u64 counter_offset; 6236 __u64 reserved; 6237 }; 6238 6239The offset describes a number of counter cycles that are subtracted from 6240both virtual and physical counter views (similar to the effects of the 6241CNTVOFF_EL2 and CNTPOFF_EL2 system registers, but only global). The offset 6242always applies to all vcpus (already created or created after this ioctl) 6243for this VM. 6244 6245It is userspace's responsibility to compute the offset based, for example, 6246on previous values of the guest counters. 6247 6248Any value other than 0 for the "reserved" field may result in an error 6249(-EINVAL) being returned. This ioctl can also return -EBUSY if any vcpu 6250ioctl is issued concurrently. 6251 6252Note that using this ioctl results in KVM ignoring subsequent userspace 6253writes to the CNTVCT_EL0 and CNTPCT_EL0 registers using the SET_ONE_REG 6254interface. No error will be returned, but the resulting offset will not be 6255applied. 6256 6257.. _KVM_ARM_GET_REG_WRITABLE_MASKS: 6258 62594.139 KVM_ARM_GET_REG_WRITABLE_MASKS 6260------------------------------------ 6261 6262:Capability: KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES 6263:Architectures: arm64 6264:Type: vm ioctl 6265:Parameters: struct reg_mask_range (in/out) 6266:Returns: 0 on success, < 0 on error 6267 6268 6269:: 6270 6271 #define KVM_ARM_FEATURE_ID_RANGE 0 6272 #define KVM_ARM_FEATURE_ID_RANGE_SIZE (3 * 8 * 8) 6273 6274 struct reg_mask_range { 6275 __u64 addr; /* Pointer to mask array */ 6276 __u32 range; /* Requested range */ 6277 __u32 reserved[13]; 6278 }; 6279 6280This ioctl copies the writable masks for a selected range of registers to 6281userspace. 6282 6283The ``addr`` field is a pointer to the destination array where KVM copies 6284the writable masks. 6285 6286The ``range`` field indicates the requested range of registers. 6287``KVM_CHECK_EXTENSION`` for the ``KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES`` 6288capability returns the supported ranges, expressed as a set of flags. Each 6289flag's bit index represents a possible value for the ``range`` field. 6290All other values are reserved for future use and KVM may return an error. 6291 6292The ``reserved[13]`` array is reserved for future use and should be 0, or 6293KVM may return an error. 6294 6295KVM_ARM_FEATURE_ID_RANGE (0) 6296^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 6297 6298The Feature ID range is defined as the AArch64 System register space with 6299op0==3, op1=={0, 1, 3}, CRn==0, CRm=={0-7}, op2=={0-7}. 6300 6301The mask returned array pointed to by ``addr`` is indexed by the macro 6302``ARM64_FEATURE_ID_RANGE_IDX(op0, op1, crn, crm, op2)``, allowing userspace 6303to know what fields can be changed for the system register described by 6304``op0, op1, crn, crm, op2``. KVM rejects ID register values that describe a 6305superset of the features supported by the system. 6306 63074.140 KVM_SET_USER_MEMORY_REGION2 6308--------------------------------- 6309 6310:Capability: KVM_CAP_USER_MEMORY2 6311:Architectures: all 6312:Type: vm ioctl 6313:Parameters: struct kvm_userspace_memory_region2 (in) 6314:Returns: 0 on success, -1 on error 6315 6316KVM_SET_USER_MEMORY_REGION2 is an extension to KVM_SET_USER_MEMORY_REGION that 6317allows mapping guest_memfd memory into a guest. All fields shared with 6318KVM_SET_USER_MEMORY_REGION identically. Userspace can set KVM_MEM_GUEST_MEMFD 6319in flags to have KVM bind the memory region to a given guest_memfd range of 6320[guest_memfd_offset, guest_memfd_offset + memory_size]. The target guest_memfd 6321must point at a file created via KVM_CREATE_GUEST_MEMFD on the current VM, and 6322the target range must not be bound to any other memory region. All standard 6323bounds checks apply (use common sense). 6324 6325:: 6326 6327 struct kvm_userspace_memory_region2 { 6328 __u32 slot; 6329 __u32 flags; 6330 __u64 guest_phys_addr; 6331 __u64 memory_size; /* bytes */ 6332 __u64 userspace_addr; /* start of the userspace allocated memory */ 6333 __u64 guest_memfd_offset; 6334 __u32 guest_memfd; 6335 __u32 pad1; 6336 __u64 pad2[14]; 6337 }; 6338 6339A KVM_MEM_GUEST_MEMFD region _must_ have a valid guest_memfd (private memory) and 6340userspace_addr (shared memory). However, "valid" for userspace_addr simply 6341means that the address itself must be a legal userspace address. The backing 6342mapping for userspace_addr is not required to be valid/populated at the time of 6343KVM_SET_USER_MEMORY_REGION2, e.g. shared memory can be lazily mapped/allocated 6344on-demand. 6345 6346When mapping a gfn into the guest, KVM selects shared vs. private, i.e consumes 6347userspace_addr vs. guest_memfd, based on the gfn's KVM_MEMORY_ATTRIBUTE_PRIVATE 6348state. At VM creation time, all memory is shared, i.e. the PRIVATE attribute 6349is '0' for all gfns. Userspace can control whether memory is shared/private by 6350toggling KVM_MEMORY_ATTRIBUTE_PRIVATE via KVM_SET_MEMORY_ATTRIBUTES as needed. 6351 6352S390: 6353^^^^^ 6354 6355Returns -EINVAL if the VM has the KVM_VM_S390_UCONTROL flag set. 6356Returns -EINVAL if called on a protected VM. 6357 63584.141 KVM_SET_MEMORY_ATTRIBUTES 6359------------------------------- 6360 6361:Capability: KVM_CAP_MEMORY_ATTRIBUTES 6362:Architectures: x86 6363:Type: vm ioctl 6364:Parameters: struct kvm_memory_attributes (in) 6365:Returns: 0 on success, <0 on error 6366 6367KVM_SET_MEMORY_ATTRIBUTES allows userspace to set memory attributes for a range 6368of guest physical memory. 6369 6370:: 6371 6372 struct kvm_memory_attributes { 6373 __u64 address; 6374 __u64 size; 6375 __u64 attributes; 6376 __u64 flags; 6377 }; 6378 6379 #define KVM_MEMORY_ATTRIBUTE_PRIVATE (1ULL << 3) 6380 6381The address and size must be page aligned. The supported attributes can be 6382retrieved via ioctl(KVM_CHECK_EXTENSION) on KVM_CAP_MEMORY_ATTRIBUTES. If 6383executed on a VM, KVM_CAP_MEMORY_ATTRIBUTES precisely returns the attributes 6384supported by that VM. If executed at system scope, KVM_CAP_MEMORY_ATTRIBUTES 6385returns all attributes supported by KVM. The only attribute defined at this 6386time is KVM_MEMORY_ATTRIBUTE_PRIVATE, which marks the associated gfn as being 6387guest private memory. 6388 6389Note, there is no "get" API. Userspace is responsible for explicitly tracking 6390the state of a gfn/page as needed. 6391 6392The "flags" field is reserved for future extensions and must be '0'. 6393 63944.142 KVM_CREATE_GUEST_MEMFD 6395---------------------------- 6396 6397:Capability: KVM_CAP_GUEST_MEMFD 6398:Architectures: none 6399:Type: vm ioctl 6400:Parameters: struct kvm_create_guest_memfd(in) 6401:Returns: A file descriptor on success, <0 on error 6402 6403KVM_CREATE_GUEST_MEMFD creates an anonymous file and returns a file descriptor 6404that refers to it. guest_memfd files are roughly analogous to files created 6405via memfd_create(), e.g. guest_memfd files live in RAM, have volatile storage, 6406and are automatically released when the last reference is dropped. Unlike 6407"regular" memfd_create() files, guest_memfd files are bound to their owning 6408virtual machine (see below), cannot be mapped, read, or written by userspace, 6409and cannot be resized (guest_memfd files do however support PUNCH_HOLE). 6410 6411:: 6412 6413 struct kvm_create_guest_memfd { 6414 __u64 size; 6415 __u64 flags; 6416 __u64 reserved[6]; 6417 }; 6418 6419Conceptually, the inode backing a guest_memfd file represents physical memory, 6420i.e. is coupled to the virtual machine as a thing, not to a "struct kvm". The 6421file itself, which is bound to a "struct kvm", is that instance's view of the 6422underlying memory, e.g. effectively provides the translation of guest addresses 6423to host memory. This allows for use cases where multiple KVM structures are 6424used to manage a single virtual machine, e.g. when performing intrahost 6425migration of a virtual machine. 6426 6427KVM currently only supports mapping guest_memfd via KVM_SET_USER_MEMORY_REGION2, 6428and more specifically via the guest_memfd and guest_memfd_offset fields in 6429"struct kvm_userspace_memory_region2", where guest_memfd_offset is the offset 6430into the guest_memfd instance. For a given guest_memfd file, there can be at 6431most one mapping per page, i.e. binding multiple memory regions to a single 6432guest_memfd range is not allowed (any number of memory regions can be bound to 6433a single guest_memfd file, but the bound ranges must not overlap). 6434 6435When the capability KVM_CAP_GUEST_MEMFD_MMAP is supported, the 'flags' field 6436supports GUEST_MEMFD_FLAG_MMAP. Setting this flag on guest_memfd creation 6437enables mmap() and faulting of guest_memfd memory to host userspace. 6438 6439When the KVM MMU performs a PFN lookup to service a guest fault and the backing 6440guest_memfd has the GUEST_MEMFD_FLAG_MMAP set, then the fault will always be 6441consumed from guest_memfd, regardless of whether it is a shared or a private 6442fault. 6443 6444See KVM_SET_USER_MEMORY_REGION2 for additional details. 6445 64464.143 KVM_PRE_FAULT_MEMORY 6447--------------------------- 6448 6449:Capability: KVM_CAP_PRE_FAULT_MEMORY 6450:Architectures: none 6451:Type: vcpu ioctl 6452:Parameters: struct kvm_pre_fault_memory (in/out) 6453:Returns: 0 if at least one page is processed, < 0 on error 6454 6455Errors: 6456 6457 ========== =============================================================== 6458 EINVAL The specified `gpa` and `size` were invalid (e.g. not 6459 page aligned, causes an overflow, or size is zero). 6460 ENOENT The specified `gpa` is outside defined memslots. 6461 EINTR An unmasked signal is pending and no page was processed. 6462 EFAULT The parameter address was invalid. 6463 EOPNOTSUPP Mapping memory for a GPA is unsupported by the 6464 hypervisor, and/or for the current vCPU state/mode. 6465 EIO unexpected error conditions (also causes a WARN) 6466 ========== =============================================================== 6467 6468:: 6469 6470 struct kvm_pre_fault_memory { 6471 /* in/out */ 6472 __u64 gpa; 6473 __u64 size; 6474 /* in */ 6475 __u64 flags; 6476 __u64 padding[5]; 6477 }; 6478 6479KVM_PRE_FAULT_MEMORY populates KVM's stage-2 page tables used to map memory 6480for the current vCPU state. KVM maps memory as if the vCPU generated a 6481stage-2 read page fault, e.g. faults in memory as needed, but doesn't break 6482CoW. However, KVM does not mark any newly created stage-2 PTE as Accessed. 6483 6484In the case of confidential VM types where there is an initial set up of 6485private guest memory before the guest is 'finalized'/measured, this ioctl 6486should only be issued after completing all the necessary setup to put the 6487guest into a 'finalized' state so that the above semantics can be reliably 6488ensured. 6489 6490In some cases, multiple vCPUs might share the page tables. In this 6491case, the ioctl can be called in parallel. 6492 6493When the ioctl returns, the input values are updated to point to the 6494remaining range. If `size` > 0 on return, the caller can just issue 6495the ioctl again with the same `struct kvm_map_memory` argument. 6496 6497Shadow page tables cannot support this ioctl because they 6498are indexed by virtual address or nested guest physical address. 6499Calling this ioctl when the guest is using shadow page tables (for 6500example because it is running a nested guest with nested page tables) 6501will fail with `EOPNOTSUPP` even if `KVM_CHECK_EXTENSION` reports 6502the capability to be present. 6503 6504`flags` must currently be zero. 6505 6506 6507.. _kvm_run: 6508 65095. The kvm_run structure 6510======================== 6511 6512Application code obtains a pointer to the kvm_run structure by 6513mmap()ing a vcpu fd. From that point, application code can control 6514execution by changing fields in kvm_run prior to calling the KVM_RUN 6515ioctl, and obtain information about the reason KVM_RUN returned by 6516looking up structure members. 6517 6518:: 6519 6520 struct kvm_run { 6521 /* in */ 6522 __u8 request_interrupt_window; 6523 6524Request that KVM_RUN return when it becomes possible to inject external 6525interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 6526 6527:: 6528 6529 __u8 immediate_exit; 6530 6531This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 6532exits immediately, returning -EINTR. In the common scenario where a 6533signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 6534to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 6535Rather than blocking the signal outside KVM_RUN, userspace can set up 6536a signal handler that sets run->immediate_exit to a non-zero value. 6537 6538This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 6539 6540:: 6541 6542 __u8 padding1[6]; 6543 6544 /* out */ 6545 __u32 exit_reason; 6546 6547When KVM_RUN has returned successfully (return value 0), this informs 6548application code why KVM_RUN has returned. Allowable values for this 6549field are detailed below. 6550 6551:: 6552 6553 __u8 ready_for_interrupt_injection; 6554 6555If request_interrupt_window has been specified, this field indicates 6556an interrupt can be injected now with KVM_INTERRUPT. 6557 6558:: 6559 6560 __u8 if_flag; 6561 6562The value of the current interrupt flag. Only valid if in-kernel 6563local APIC is not used. 6564 6565:: 6566 6567 __u16 flags; 6568 6569More architecture-specific flags detailing state of the VCPU that may 6570affect the device's behavior. Current defined flags:: 6571 6572 /* x86, set if the VCPU is in system management mode */ 6573 #define KVM_RUN_X86_SMM (1 << 0) 6574 /* x86, set if bus lock detected in VM */ 6575 #define KVM_RUN_X86_BUS_LOCK (1 << 1) 6576 /* x86, set if the VCPU is executing a nested (L2) guest */ 6577 #define KVM_RUN_X86_GUEST_MODE (1 << 2) 6578 6579 /* arm64, set for KVM_EXIT_DEBUG */ 6580 #define KVM_DEBUG_ARCH_HSR_HIGH_VALID (1 << 0) 6581 6582:: 6583 6584 /* in (pre_kvm_run), out (post_kvm_run) */ 6585 __u64 cr8; 6586 6587The value of the cr8 register. Only valid if in-kernel local APIC is 6588not used. Both input and output. 6589 6590:: 6591 6592 __u64 apic_base; 6593 6594The value of the APIC BASE msr. Only valid if in-kernel local 6595APIC is not used. Both input and output. 6596 6597:: 6598 6599 union { 6600 /* KVM_EXIT_UNKNOWN */ 6601 struct { 6602 __u64 hardware_exit_reason; 6603 } hw; 6604 6605If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 6606reasons. Further architecture-specific information is available in 6607hardware_exit_reason. 6608 6609:: 6610 6611 /* KVM_EXIT_FAIL_ENTRY */ 6612 struct { 6613 __u64 hardware_entry_failure_reason; 6614 __u32 cpu; /* if KVM_LAST_CPU */ 6615 } fail_entry; 6616 6617If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 6618to unknown reasons. Further architecture-specific information is 6619available in hardware_entry_failure_reason. 6620 6621:: 6622 6623 /* KVM_EXIT_EXCEPTION */ 6624 struct { 6625 __u32 exception; 6626 __u32 error_code; 6627 } ex; 6628 6629Unused. 6630 6631:: 6632 6633 /* KVM_EXIT_IO */ 6634 struct { 6635 #define KVM_EXIT_IO_IN 0 6636 #define KVM_EXIT_IO_OUT 1 6637 __u8 direction; 6638 __u8 size; /* bytes */ 6639 __u16 port; 6640 __u32 count; 6641 __u64 data_offset; /* relative to kvm_run start */ 6642 } io; 6643 6644If exit_reason is KVM_EXIT_IO, then the vcpu has 6645executed a port I/O instruction which could not be satisfied by kvm. 6646data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 6647where kvm expects application code to place the data for the next 6648KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 6649 6650:: 6651 6652 /* KVM_EXIT_DEBUG */ 6653 struct { 6654 struct kvm_debug_exit_arch arch; 6655 } debug; 6656 6657If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 6658for which architecture specific information is returned. 6659 6660:: 6661 6662 /* KVM_EXIT_MMIO */ 6663 struct { 6664 __u64 phys_addr; 6665 __u8 data[8]; 6666 __u32 len; 6667 __u8 is_write; 6668 } mmio; 6669 6670If exit_reason is KVM_EXIT_MMIO, then the vcpu has 6671executed a memory-mapped I/O instruction which could not be satisfied 6672by kvm. The 'data' member contains the written data if 'is_write' is 6673true, and should be filled by application code otherwise. 6674 6675The 'data' member contains, in its first 'len' bytes, the value as it would 6676appear if the VCPU performed a load or store of the appropriate width directly 6677to the byte array. 6678 6679.. note:: 6680 6681 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN, 6682 KVM_EXIT_EPR, KVM_EXIT_HYPERCALL, KVM_EXIT_TDX, 6683 KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding 6684 operations are complete (and guest state is consistent) only after userspace 6685 has re-entered the kernel with KVM_RUN. The kernel side will first finish 6686 incomplete operations and then check for pending signals. 6687 6688 The pending state of the operation is not preserved in state which is 6689 visible to userspace, thus userspace should ensure that the operation is 6690 completed before performing a live migration. Userspace can re-enter the 6691 guest with an unmasked signal pending or with the immediate_exit field set 6692 to complete pending operations without allowing any further instructions 6693 to be executed. 6694 6695:: 6696 6697 /* KVM_EXIT_HYPERCALL */ 6698 struct { 6699 __u64 nr; 6700 __u64 args[6]; 6701 __u64 ret; 6702 __u64 flags; 6703 } hypercall; 6704 6705 6706It is strongly recommended that userspace use ``KVM_EXIT_IO`` (x86) or 6707``KVM_EXIT_MMIO`` (all except s390) to implement functionality that 6708requires a guest to interact with host userspace. 6709 6710.. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 6711 6712For arm64: 6713---------- 6714 6715SMCCC exits can be enabled depending on the configuration of the SMCCC 6716filter. See the Documentation/virt/kvm/devices/vm.rst 6717``KVM_ARM_SMCCC_FILTER`` for more details. 6718 6719``nr`` contains the function ID of the guest's SMCCC call. Userspace is 6720expected to use the ``KVM_GET_ONE_REG`` ioctl to retrieve the call 6721parameters from the vCPU's GPRs. 6722 6723Definition of ``flags``: 6724 - ``KVM_HYPERCALL_EXIT_SMC``: Indicates that the guest used the SMC 6725 conduit to initiate the SMCCC call. If this bit is 0 then the guest 6726 used the HVC conduit for the SMCCC call. 6727 6728 - ``KVM_HYPERCALL_EXIT_16BIT``: Indicates that the guest used a 16bit 6729 instruction to initiate the SMCCC call. If this bit is 0 then the 6730 guest used a 32bit instruction. An AArch64 guest always has this 6731 bit set to 0. 6732 6733At the point of exit, PC points to the instruction immediately following 6734the trapping instruction. 6735 6736:: 6737 6738 /* KVM_EXIT_TPR_ACCESS */ 6739 struct { 6740 __u64 rip; 6741 __u32 is_write; 6742 __u32 pad; 6743 } tpr_access; 6744 6745To be documented (KVM_TPR_ACCESS_REPORTING). 6746 6747:: 6748 6749 /* KVM_EXIT_S390_SIEIC */ 6750 struct { 6751 __u8 icptcode; 6752 __u64 mask; /* psw upper half */ 6753 __u64 addr; /* psw lower half */ 6754 __u16 ipa; 6755 __u32 ipb; 6756 } s390_sieic; 6757 6758s390 specific. 6759 6760:: 6761 6762 /* KVM_EXIT_S390_RESET */ 6763 #define KVM_S390_RESET_POR 1 6764 #define KVM_S390_RESET_CLEAR 2 6765 #define KVM_S390_RESET_SUBSYSTEM 4 6766 #define KVM_S390_RESET_CPU_INIT 8 6767 #define KVM_S390_RESET_IPL 16 6768 __u64 s390_reset_flags; 6769 6770s390 specific. 6771 6772:: 6773 6774 /* KVM_EXIT_S390_UCONTROL */ 6775 struct { 6776 __u64 trans_exc_code; 6777 __u32 pgm_code; 6778 } s390_ucontrol; 6779 6780s390 specific. A page fault has occurred for a user controlled virtual 6781machine (KVM_VM_S390_UNCONTROL) on its host page table that cannot be 6782resolved by the kernel. 6783The program code and the translation exception code that were placed 6784in the cpu's lowcore are presented here as defined by the z Architecture 6785Principles of Operation Book in the Chapter for Dynamic Address Translation 6786(DAT) 6787 6788:: 6789 6790 /* KVM_EXIT_DCR */ 6791 struct { 6792 __u32 dcrn; 6793 __u32 data; 6794 __u8 is_write; 6795 } dcr; 6796 6797Deprecated - was used for 440 KVM. 6798 6799:: 6800 6801 /* KVM_EXIT_OSI */ 6802 struct { 6803 __u64 gprs[32]; 6804 } osi; 6805 6806MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 6807hypercalls and exit with this exit struct that contains all the guest gprs. 6808 6809If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 6810Userspace can now handle the hypercall and when it's done modify the gprs as 6811necessary. Upon guest entry all guest GPRs will then be replaced by the values 6812in this struct. 6813 6814:: 6815 6816 /* KVM_EXIT_PAPR_HCALL */ 6817 struct { 6818 __u64 nr; 6819 __u64 ret; 6820 __u64 args[9]; 6821 } papr_hcall; 6822 6823This is used on 64-bit PowerPC when emulating a pSeries partition, 6824e.g. with the 'pseries' machine type in qemu. It occurs when the 6825guest does a hypercall using the 'sc 1' instruction. The 'nr' field 6826contains the hypercall number (from the guest R3), and 'args' contains 6827the arguments (from the guest R4 - R12). Userspace should put the 6828return code in 'ret' and any extra returned values in args[]. 6829The possible hypercalls are defined in the Power Architecture Platform 6830Requirements (PAPR) document available from www.power.org (free 6831developer registration required to access it). 6832 6833:: 6834 6835 /* KVM_EXIT_S390_TSCH */ 6836 struct { 6837 __u16 subchannel_id; 6838 __u16 subchannel_nr; 6839 __u32 io_int_parm; 6840 __u32 io_int_word; 6841 __u32 ipb; 6842 __u8 dequeued; 6843 } s390_tsch; 6844 6845s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 6846and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 6847interrupt for the target subchannel has been dequeued and subchannel_id, 6848subchannel_nr, io_int_parm and io_int_word contain the parameters for that 6849interrupt. ipb is needed for instruction parameter decoding. 6850 6851:: 6852 6853 /* KVM_EXIT_EPR */ 6854 struct { 6855 __u32 epr; 6856 } epr; 6857 6858On FSL BookE PowerPC chips, the interrupt controller has a fast patch 6859interrupt acknowledge path to the core. When the core successfully 6860delivers an interrupt, it automatically populates the EPR register with 6861the interrupt vector number and acknowledges the interrupt inside 6862the interrupt controller. 6863 6864In case the interrupt controller lives in user space, we need to do 6865the interrupt acknowledge cycle through it to fetch the next to be 6866delivered interrupt vector using this exit. 6867 6868It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 6869external interrupt has just been delivered into the guest. User space 6870should put the acknowledged interrupt vector into the 'epr' field. 6871 6872:: 6873 6874 /* KVM_EXIT_SYSTEM_EVENT */ 6875 struct { 6876 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 6877 #define KVM_SYSTEM_EVENT_RESET 2 6878 #define KVM_SYSTEM_EVENT_CRASH 3 6879 #define KVM_SYSTEM_EVENT_WAKEUP 4 6880 #define KVM_SYSTEM_EVENT_SUSPEND 5 6881 #define KVM_SYSTEM_EVENT_SEV_TERM 6 6882 #define KVM_SYSTEM_EVENT_TDX_FATAL 7 6883 __u32 type; 6884 __u32 ndata; 6885 __u64 data[16]; 6886 } system_event; 6887 6888If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 6889a system-level event using some architecture specific mechanism (hypercall 6890or some special instruction). In case of ARM64, this is triggered using 6891HVC instruction based PSCI call from the vcpu. 6892 6893The 'type' field describes the system-level event type. 6894Valid values for 'type' are: 6895 6896 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 6897 VM. Userspace is not obliged to honour this, and if it does honour 6898 this does not need to destroy the VM synchronously (ie it may call 6899 KVM_RUN again before shutdown finally occurs). 6900 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 6901 As with SHUTDOWN, userspace can choose to ignore the request, or 6902 to schedule the reset to occur in the future and may call KVM_RUN again. 6903 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 6904 has requested a crash condition maintenance. Userspace can choose 6905 to ignore the request, or to gather VM memory core dump and/or 6906 reset/shutdown of the VM. 6907 - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination. 6908 The guest physical address of the guest's GHCB is stored in `data[0]`. 6909 - KVM_SYSTEM_EVENT_TDX_FATAL -- a TDX guest reported a fatal error state. 6910 KVM doesn't do any parsing or conversion, it just dumps 16 general-purpose 6911 registers to userspace, in ascending order of the 4-bit indices for x86-64 6912 general-purpose registers in instruction encoding, as defined in the Intel 6913 SDM. 6914 - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and 6915 KVM has recognized a wakeup event. Userspace may honor this event by 6916 marking the exiting vCPU as runnable, or deny it and call KVM_RUN again. 6917 - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested a suspension of 6918 the VM. 6919 6920If KVM_CAP_SYSTEM_EVENT_DATA is present, the 'data' field can contain 6921architecture specific information for the system-level event. Only 6922the first `ndata` items (possibly zero) of the data array are valid. 6923 6924 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2 if 6925 the guest issued a SYSTEM_RESET2 call according to v1.1 of the PSCI 6926 specification. 6927 6928 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2 6929 if the guest issued a SYSTEM_OFF2 call according to v1.3 of the PSCI 6930 specification. 6931 6932 - for RISC-V, data[0] is set to the value of the second argument of the 6933 ``sbi_system_reset`` call. 6934 6935Previous versions of Linux defined a `flags` member in this struct. The 6936field is now aliased to `data[0]`. Userspace can assume that it is only 6937written if ndata is greater than 0. 6938 6939For arm/arm64: 6940-------------- 6941 6942KVM_SYSTEM_EVENT_SUSPEND exits are enabled with the 6943KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If a guest invokes the PSCI 6944SYSTEM_SUSPEND function, KVM will exit to userspace with this event 6945type. 6946 6947It is the sole responsibility of userspace to implement the PSCI 6948SYSTEM_SUSPEND call according to ARM DEN0022D.b 5.19 "SYSTEM_SUSPEND". 6949KVM does not change the vCPU's state before exiting to userspace, so 6950the call parameters are left in-place in the vCPU registers. 6951 6952Userspace is _required_ to take action for such an exit. It must 6953either: 6954 6955 - Honor the guest request to suspend the VM. Userspace can request 6956 in-kernel emulation of suspension by setting the calling vCPU's 6957 state to KVM_MP_STATE_SUSPENDED. Userspace must configure the vCPU's 6958 state according to the parameters passed to the PSCI function when 6959 the calling vCPU is resumed. See ARM DEN0022D.b 5.19.1 "Intended use" 6960 for details on the function parameters. 6961 6962 - Deny the guest request to suspend the VM. See ARM DEN0022D.b 5.19.2 6963 "Caller responsibilities" for possible return values. 6964 6965Hibernation using the PSCI SYSTEM_OFF2 call is enabled when PSCI v1.3 6966is enabled. If a guest invokes the PSCI SYSTEM_OFF2 function, KVM will 6967exit to userspace with the KVM_SYSTEM_EVENT_SHUTDOWN event type and with 6968data[0] set to KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2. The only 6969supported hibernate type for the SYSTEM_OFF2 function is HIBERNATE_OFF. 6970 6971:: 6972 6973 /* KVM_EXIT_IOAPIC_EOI */ 6974 struct { 6975 __u8 vector; 6976 } eoi; 6977 6978Indicates that the VCPU's in-kernel local APIC received an EOI for a 6979level-triggered IOAPIC interrupt. This exit only triggers when the 6980IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 6981the userspace IOAPIC should process the EOI and retrigger the interrupt if 6982it is still asserted. Vector is the LAPIC interrupt vector for which the 6983EOI was received. 6984 6985:: 6986 6987 struct kvm_hyperv_exit { 6988 #define KVM_EXIT_HYPERV_SYNIC 1 6989 #define KVM_EXIT_HYPERV_HCALL 2 6990 #define KVM_EXIT_HYPERV_SYNDBG 3 6991 __u32 type; 6992 __u32 pad1; 6993 union { 6994 struct { 6995 __u32 msr; 6996 __u32 pad2; 6997 __u64 control; 6998 __u64 evt_page; 6999 __u64 msg_page; 7000 } synic; 7001 struct { 7002 __u64 input; 7003 __u64 result; 7004 __u64 params[2]; 7005 } hcall; 7006 struct { 7007 __u32 msr; 7008 __u32 pad2; 7009 __u64 control; 7010 __u64 status; 7011 __u64 send_page; 7012 __u64 recv_page; 7013 __u64 pending_page; 7014 } syndbg; 7015 } u; 7016 }; 7017 /* KVM_EXIT_HYPERV */ 7018 struct kvm_hyperv_exit hyperv; 7019 7020Indicates that the VCPU exits into userspace to process some tasks 7021related to Hyper-V emulation. 7022 7023Valid values for 'type' are: 7024 7025 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 7026 7027Hyper-V SynIC state change. Notification is used to remap SynIC 7028event/message pages and to enable/disable SynIC messages/events processing 7029in userspace. 7030 7031 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about 7032 7033Hyper-V Synthetic debugger state change. Notification is used to either update 7034the pending_page location or to send a control command (send the buffer located 7035in send_page or recv a buffer to recv_page). 7036 7037:: 7038 7039 /* KVM_EXIT_ARM_NISV */ 7040 struct { 7041 __u64 esr_iss; 7042 __u64 fault_ipa; 7043 } arm_nisv; 7044 7045Used on arm64 systems. If a guest accesses memory not in a memslot, 7046KVM will typically return to userspace and ask it to do MMIO emulation on its 7047behalf. However, for certain classes of instructions, no instruction decode 7048(direction, length of memory access) is provided, and fetching and decoding 7049the instruction from the VM is overly complicated to live in the kernel. 7050 7051Historically, when this situation occurred, KVM would print a warning and kill 7052the VM. KVM assumed that if the guest accessed non-memslot memory, it was 7053trying to do I/O, which just couldn't be emulated, and the warning message was 7054phrased accordingly. However, what happened more often was that a guest bug 7055caused access outside the guest memory areas which should lead to a more 7056meaningful warning message and an external abort in the guest, if the access 7057did not fall within an I/O window. 7058 7059Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable 7060this capability at VM creation. Once this is done, these types of errors will 7061instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from 7062the ESR_EL2 in the esr_iss field, and the faulting IPA in the fault_ipa field. 7063Userspace can either fix up the access if it's actually an I/O access by 7064decoding the instruction from guest memory (if it's very brave) and continue 7065executing the guest, or it can decide to suspend, dump, or restart the guest. 7066 7067Note that KVM does not skip the faulting instruction as it does for 7068KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state 7069if it decides to decode and emulate the instruction. 7070 7071This feature isn't available to protected VMs, as userspace does not 7072have access to the state that is required to perform the emulation. 7073Instead, a data abort exception is directly injected in the guest. 7074Note that although KVM_CAP_ARM_NISV_TO_USER will be reported if 7075queried outside of a protected VM context, the feature will not be 7076exposed if queried on a protected VM file descriptor. 7077 7078:: 7079 7080 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */ 7081 struct { 7082 __u8 error; /* user -> kernel */ 7083 __u8 pad[7]; 7084 __u32 reason; /* kernel -> user */ 7085 __u32 index; /* kernel -> user */ 7086 __u64 data; /* kernel <-> user */ 7087 } msr; 7088 7089Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is 7090enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code 7091may instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR 7092exit for writes. 7093 7094The "reason" field specifies why the MSR interception occurred. Userspace will 7095only receive MSR exits when a particular reason was requested during through 7096ENABLE_CAP. Currently valid exit reasons are: 7097 7098============================ ======================================== 7099 KVM_MSR_EXIT_REASON_UNKNOWN access to MSR that is unknown to KVM 7100 KVM_MSR_EXIT_REASON_INVAL access to invalid MSRs or reserved bits 7101 KVM_MSR_EXIT_REASON_FILTER access blocked by KVM_X86_SET_MSR_FILTER 7102============================ ======================================== 7103 7104For KVM_EXIT_X86_RDMSR, the "index" field tells userspace which MSR the guest 7105wants to read. To respond to this request with a successful read, userspace 7106writes the respective data into the "data" field and must continue guest 7107execution to ensure the read data is transferred into guest register state. 7108 7109If the RDMSR request was unsuccessful, userspace indicates that with a "1" in 7110the "error" field. This will inject a #GP into the guest when the VCPU is 7111executed again. 7112 7113For KVM_EXIT_X86_WRMSR, the "index" field tells userspace which MSR the guest 7114wants to write. Once finished processing the event, userspace must continue 7115vCPU execution. If the MSR write was unsuccessful, userspace also sets the 7116"error" field to "1". 7117 7118See KVM_X86_SET_MSR_FILTER for details on the interaction with MSR filtering. 7119 7120:: 7121 7122 7123 struct kvm_xen_exit { 7124 #define KVM_EXIT_XEN_HCALL 1 7125 __u32 type; 7126 union { 7127 struct { 7128 __u32 longmode; 7129 __u32 cpl; 7130 __u64 input; 7131 __u64 result; 7132 __u64 params[6]; 7133 } hcall; 7134 } u; 7135 }; 7136 /* KVM_EXIT_XEN */ 7137 struct kvm_hyperv_exit xen; 7138 7139Indicates that the VCPU exits into userspace to process some tasks 7140related to Xen emulation. 7141 7142Valid values for 'type' are: 7143 7144 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall. 7145 Userspace is expected to place the hypercall result into the appropriate 7146 field before invoking KVM_RUN again. 7147 7148:: 7149 7150 /* KVM_EXIT_RISCV_SBI */ 7151 struct { 7152 unsigned long extension_id; 7153 unsigned long function_id; 7154 unsigned long args[6]; 7155 unsigned long ret[2]; 7156 } riscv_sbi; 7157 7158If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has 7159done a SBI call which is not handled by KVM RISC-V kernel module. The details 7160of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The 7161'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the 7162'function_id' field represents function ID of given SBI extension. The 'args' 7163array field of 'riscv_sbi' represents parameters for the SBI call and 'ret' 7164array field represents return values. The userspace should update the return 7165values of SBI call before resuming the VCPU. For more details on RISC-V SBI 7166spec refer, https://github.com/riscv/riscv-sbi-doc. 7167 7168:: 7169 7170 /* KVM_EXIT_MEMORY_FAULT */ 7171 struct { 7172 #define KVM_MEMORY_EXIT_FLAG_PRIVATE (1ULL << 3) 7173 __u64 flags; 7174 __u64 gpa; 7175 __u64 size; 7176 } memory_fault; 7177 7178KVM_EXIT_MEMORY_FAULT indicates the vCPU has encountered a memory fault that 7179could not be resolved by KVM. The 'gpa' and 'size' (in bytes) describe the 7180guest physical address range [gpa, gpa + size) of the fault. The 'flags' field 7181describes properties of the faulting access that are likely pertinent: 7182 7183 - KVM_MEMORY_EXIT_FLAG_PRIVATE - When set, indicates the memory fault occurred 7184 on a private memory access. When clear, indicates the fault occurred on a 7185 shared access. 7186 7187Note! KVM_EXIT_MEMORY_FAULT is unique among all KVM exit reasons in that it 7188accompanies a return code of '-1', not '0'! errno will always be set to EFAULT 7189or EHWPOISON when KVM exits with KVM_EXIT_MEMORY_FAULT, userspace should assume 7190kvm_run.exit_reason is stale/undefined for all other error numbers. 7191 7192:: 7193 7194 /* KVM_EXIT_NOTIFY */ 7195 struct { 7196 #define KVM_NOTIFY_CONTEXT_INVALID (1 << 0) 7197 __u32 flags; 7198 } notify; 7199 7200Used on x86 systems. When the VM capability KVM_CAP_X86_NOTIFY_VMEXIT is 7201enabled, a VM exit generated if no event window occurs in VM non-root mode 7202for a specified amount of time. Once KVM_X86_NOTIFY_VMEXIT_USER is set when 7203enabling the cap, it would exit to userspace with the exit reason 7204KVM_EXIT_NOTIFY for further handling. The "flags" field contains more 7205detailed info. 7206 7207The valid value for 'flags' is: 7208 7209 - KVM_NOTIFY_CONTEXT_INVALID -- the VM context is corrupted and not valid 7210 in VMCS. It would run into unknown result if resume the target VM. 7211 7212:: 7213 7214 /* KVM_EXIT_TDX */ 7215 struct { 7216 __u64 flags; 7217 __u64 nr; 7218 union { 7219 struct { 7220 u64 ret; 7221 u64 data[5]; 7222 } unknown; 7223 struct { 7224 u64 ret; 7225 u64 gpa; 7226 u64 size; 7227 } get_quote; 7228 struct { 7229 u64 ret; 7230 u64 leaf; 7231 u64 r11, r12, r13, r14; 7232 } get_tdvmcall_info; 7233 struct { 7234 u64 ret; 7235 u64 vector; 7236 } setup_event_notify; 7237 }; 7238 } tdx; 7239 7240Process a TDVMCALL from the guest. KVM forwards select TDVMCALL based 7241on the Guest-Hypervisor Communication Interface (GHCI) specification; 7242KVM bridges these requests to the userspace VMM with minimal changes, 7243placing the inputs in the union and copying them back to the guest 7244on re-entry. 7245 7246Flags are currently always zero, whereas ``nr`` contains the TDVMCALL 7247number from register R11. The remaining field of the union provide the 7248inputs and outputs of the TDVMCALL. Currently the following values of 7249``nr`` are defined: 7250 7251 * ``TDVMCALL_GET_QUOTE``: the guest has requested to generate a TD-Quote 7252 signed by a service hosting TD-Quoting Enclave operating on the host. 7253 Parameters and return value are in the ``get_quote`` field of the union. 7254 The ``gpa`` field and ``size`` specify the guest physical address 7255 (without the shared bit set) and the size of a shared-memory buffer, in 7256 which the TDX guest passes a TD Report. The ``ret`` field represents 7257 the return value of the GetQuote request. When the request has been 7258 queued successfully, the TDX guest can poll the status field in the 7259 shared-memory area to check whether the Quote generation is completed or 7260 not. When completed, the generated Quote is returned via the same buffer. 7261 7262 * ``TDVMCALL_GET_TD_VM_CALL_INFO``: the guest has requested the support 7263 status of TDVMCALLs. The output values for the given leaf should be 7264 placed in fields from ``r11`` to ``r14`` of the ``get_tdvmcall_info`` 7265 field of the union. 7266 7267 * ``TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT``: the guest has requested to 7268 set up a notification interrupt for vector ``vector``. 7269 7270KVM may add support for more values in the future that may cause a userspace 7271exit, even without calls to ``KVM_ENABLE_CAP`` or similar. In this case, 7272it will enter with output fields already valid; in the common case, the 7273``unknown.ret`` field of the union will be ``TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED``. 7274Userspace need not do anything if it does not wish to support a TDVMCALL. 7275:: 7276 7277 /* Fix the size of the union. */ 7278 char padding[256]; 7279 }; 7280 7281 /* 7282 * shared registers between kvm and userspace. 7283 * kvm_valid_regs specifies the register classes set by the host 7284 * kvm_dirty_regs specified the register classes dirtied by userspace 7285 * struct kvm_sync_regs is architecture specific, as well as the 7286 * bits for kvm_valid_regs and kvm_dirty_regs 7287 */ 7288 __u64 kvm_valid_regs; 7289 __u64 kvm_dirty_regs; 7290 union { 7291 struct kvm_sync_regs regs; 7292 char padding[SYNC_REGS_SIZE_BYTES]; 7293 } s; 7294 7295If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 7296certain guest registers without having to call SET/GET_*REGS. Thus we can 7297avoid some system call overhead if userspace has to handle the exit. 7298Userspace can query the validity of the structure by checking 7299kvm_valid_regs for specific bits. These bits are architecture specific 7300and usually define the validity of a groups of registers. (e.g. one bit 7301for general purpose registers) 7302 7303Please note that the kernel is allowed to use the kvm_run structure as the 7304primary storage for certain register types. Therefore, the kernel may use the 7305values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 7306 7307 7308.. _cap_enable: 7309 73106. Capabilities that can be enabled on vCPUs 7311============================================ 7312 7313There are certain capabilities that change the behavior of the virtual CPU or 7314the virtual machine when enabled. To enable them, please see 7315:ref:`KVM_ENABLE_CAP`. 7316 7317Below you can find a list of capabilities and what their effect on the vCPU or 7318the virtual machine is when enabling them. 7319 7320The following information is provided along with the description: 7321 7322 Architectures: 7323 which instruction set architectures provide this ioctl. 7324 x86 includes both i386 and x86_64. 7325 7326 Target: 7327 whether this is a per-vcpu or per-vm capability. 7328 7329 Parameters: 7330 what parameters are accepted by the capability. 7331 7332 Returns: 7333 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 7334 are not detailed, but errors with specific meanings are. 7335 7336 73376.1 KVM_CAP_PPC_OSI 7338------------------- 7339 7340:Architectures: ppc 7341:Target: vcpu 7342:Parameters: none 7343:Returns: 0 on success; -1 on error 7344 7345This capability enables interception of OSI hypercalls that otherwise would 7346be treated as normal system calls to be injected into the guest. OSI hypercalls 7347were invented by Mac-on-Linux to have a standardized communication mechanism 7348between the guest and the host. 7349 7350When this capability is enabled, KVM_EXIT_OSI can occur. 7351 7352 73536.2 KVM_CAP_PPC_PAPR 7354-------------------- 7355 7356:Architectures: ppc 7357:Target: vcpu 7358:Parameters: none 7359:Returns: 0 on success; -1 on error 7360 7361This capability enables interception of PAPR hypercalls. PAPR hypercalls are 7362done using the hypercall instruction "sc 1". 7363 7364It also sets the guest privilege level to "supervisor" mode. Usually the guest 7365runs in "hypervisor" privilege mode with a few missing features. 7366 7367In addition to the above, it changes the semantics of SDR1. In this mode, the 7368HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 7369HTAB invisible to the guest. 7370 7371When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 7372 7373 73746.3 KVM_CAP_SW_TLB 7375------------------ 7376 7377:Architectures: ppc 7378:Target: vcpu 7379:Parameters: args[0] is the address of a struct kvm_config_tlb 7380:Returns: 0 on success; -1 on error 7381 7382:: 7383 7384 struct kvm_config_tlb { 7385 __u64 params; 7386 __u64 array; 7387 __u32 mmu_type; 7388 __u32 array_len; 7389 }; 7390 7391Configures the virtual CPU's TLB array, establishing a shared memory area 7392between userspace and KVM. The "params" and "array" fields are userspace 7393addresses of mmu-type-specific data structures. The "array_len" field is an 7394safety mechanism, and should be set to the size in bytes of the memory that 7395userspace has reserved for the array. It must be at least the size dictated 7396by "mmu_type" and "params". 7397 7398While KVM_RUN is active, the shared region is under control of KVM. Its 7399contents are undefined, and any modification by userspace results in 7400boundedly undefined behavior. 7401 7402On return from KVM_RUN, the shared region will reflect the current state of 7403the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 7404to tell KVM which entries have been changed, prior to calling KVM_RUN again 7405on this vcpu. 7406 7407For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 7408 7409 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 7410 - The "array" field points to an array of type "struct 7411 kvm_book3e_206_tlb_entry". 7412 - The array consists of all entries in the first TLB, followed by all 7413 entries in the second TLB. 7414 - Within a TLB, entries are ordered first by increasing set number. Within a 7415 set, entries are ordered by way (increasing ESEL). 7416 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 7417 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 7418 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 7419 hardware ignores this value for TLB0. 7420 74216.4 KVM_CAP_S390_CSS_SUPPORT 7422---------------------------- 7423 7424:Architectures: s390 7425:Target: vcpu 7426:Parameters: none 7427:Returns: 0 on success; -1 on error 7428 7429This capability enables support for handling of channel I/O instructions. 7430 7431TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 7432handled in-kernel, while the other I/O instructions are passed to userspace. 7433 7434When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 7435SUBCHANNEL intercepts. 7436 7437Note that even though this capability is enabled per-vcpu, the complete 7438virtual machine is affected. 7439 74406.5 KVM_CAP_PPC_EPR 7441------------------- 7442 7443:Architectures: ppc 7444:Target: vcpu 7445:Parameters: args[0] defines whether the proxy facility is active 7446:Returns: 0 on success; -1 on error 7447 7448This capability enables or disables the delivery of interrupts through the 7449external proxy facility. 7450 7451When enabled (args[0] != 0), every time the guest gets an external interrupt 7452delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 7453to receive the topmost interrupt vector. 7454 7455When disabled (args[0] == 0), behavior is as if this facility is unsupported. 7456 7457When this capability is enabled, KVM_EXIT_EPR can occur. 7458 74596.6 KVM_CAP_IRQ_MPIC 7460-------------------- 7461 7462:Architectures: ppc 7463:Parameters: args[0] is the MPIC device fd; 7464 args[1] is the MPIC CPU number for this vcpu 7465 7466This capability connects the vcpu to an in-kernel MPIC device. 7467 74686.7 KVM_CAP_IRQ_XICS 7469-------------------- 7470 7471:Architectures: ppc 7472:Target: vcpu 7473:Parameters: args[0] is the XICS device fd; 7474 args[1] is the XICS CPU number (server ID) for this vcpu 7475 7476This capability connects the vcpu to an in-kernel XICS device. 7477 74786.8 KVM_CAP_S390_IRQCHIP 7479------------------------ 7480 7481:Architectures: s390 7482:Target: vm 7483:Parameters: none 7484 7485This capability enables the in-kernel irqchip for s390. Please refer to 7486"4.24 KVM_CREATE_IRQCHIP" for details. 7487 74886.9 KVM_CAP_MIPS_FPU 7489-------------------- 7490 7491:Architectures: mips 7492:Target: vcpu 7493:Parameters: args[0] is reserved for future use (should be 0). 7494 7495This capability allows the use of the host Floating Point Unit by the guest. It 7496allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 7497done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be 7498accessed (depending on the current guest FPU register mode), and the Status.FR, 7499Config5.FRE bits are accessible via the KVM API and also from the guest, 7500depending on them being supported by the FPU. 7501 75026.10 KVM_CAP_MIPS_MSA 7503--------------------- 7504 7505:Architectures: mips 7506:Target: vcpu 7507:Parameters: args[0] is reserved for future use (should be 0). 7508 7509This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 7510It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 7511Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*`` 7512registers can be accessed, and the Config5.MSAEn bit is accessible via the 7513KVM API and also from the guest. 7514 75156.74 KVM_CAP_SYNC_REGS 7516---------------------- 7517 7518:Architectures: s390, x86 7519:Target: s390: always enabled, x86: vcpu 7520:Parameters: none 7521:Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register 7522 sets are supported 7523 (bitfields defined in arch/x86/include/uapi/asm/kvm.h). 7524 7525As described above in the kvm_sync_regs struct info in section :ref:`kvm_run`, 7526KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers 7527without having to call SET/GET_*REGS". This reduces overhead by eliminating 7528repeated ioctl calls for setting and/or getting register values. This is 7529particularly important when userspace is making synchronous guest state 7530modifications, e.g. when emulating and/or intercepting instructions in 7531userspace. 7532 7533For s390 specifics, please refer to the source code. 7534 7535For x86: 7536 7537- the register sets to be copied out to kvm_run are selectable 7538 by userspace (rather that all sets being copied out for every exit). 7539- vcpu_events are available in addition to regs and sregs. 7540 7541For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to 7542function as an input bit-array field set by userspace to indicate the 7543specific register sets to be copied out on the next exit. 7544 7545To indicate when userspace has modified values that should be copied into 7546the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set. 7547This is done using the same bitflags as for the 'kvm_valid_regs' field. 7548If the dirty bit is not set, then the register set values will not be copied 7549into the vCPU even if they've been modified. 7550 7551Unused bitfields in the bitarrays must be set to zero. 7552 7553:: 7554 7555 struct kvm_sync_regs { 7556 struct kvm_regs regs; 7557 struct kvm_sregs sregs; 7558 struct kvm_vcpu_events events; 7559 }; 7560 75616.75 KVM_CAP_PPC_IRQ_XIVE 7562------------------------- 7563 7564:Architectures: ppc 7565:Target: vcpu 7566:Parameters: args[0] is the XIVE device fd; 7567 args[1] is the XIVE CPU number (server ID) for this vcpu 7568 7569This capability connects the vcpu to an in-kernel XIVE device. 7570 75716.76 KVM_CAP_HYPERV_SYNIC 7572------------------------- 7573 7574:Architectures: x86 7575:Target: vcpu 7576 7577This capability, if KVM_CHECK_EXTENSION indicates that it is 7578available, means that the kernel has an implementation of the 7579Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 7580used to support Windows Hyper-V based guest paravirt drivers(VMBus). 7581 7582In order to use SynIC, it has to be activated by setting this 7583capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 7584will disable the use of APIC hardware virtualization even if supported 7585by the CPU, as it's incompatible with SynIC auto-EOI behavior. 7586 75876.77 KVM_CAP_HYPERV_SYNIC2 7588-------------------------- 7589 7590:Architectures: x86 7591:Target: vcpu 7592 7593This capability enables a newer version of Hyper-V Synthetic interrupt 7594controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 7595doesn't clear SynIC message and event flags pages when they are enabled by 7596writing to the respective MSRs. 7597 75986.78 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 7599----------------------------------- 7600 7601:Architectures: x86 7602:Target: vcpu 7603 7604This capability indicates that KVM running on top of Hyper-V hypervisor 7605enables Direct TLB flush for its guests meaning that TLB flush 7606hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM. 7607Due to the different ABI for hypercall parameters between Hyper-V and 7608KVM, enabling this capability effectively disables all hypercall 7609handling by KVM (as some KVM hypercall may be mistakenly treated as TLB 7610flush hypercalls by Hyper-V) so userspace should disable KVM identification 7611in CPUID and only exposes Hyper-V identification. In this case, guest 7612thinks it's running on Hyper-V and only use Hyper-V hypercalls. 7613 76146.79 KVM_CAP_HYPERV_ENFORCE_CPUID 7615--------------------------------- 7616 7617:Architectures: x86 7618:Target: vcpu 7619 7620When enabled, KVM will disable emulated Hyper-V features provided to the 7621guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all 7622currently implemented Hyper-V features are provided unconditionally when 7623Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001) 7624leaf. 7625 76266.80 KVM_CAP_ENFORCE_PV_FEATURE_CPUID 7627------------------------------------- 7628 7629:Architectures: x86 7630:Target: vcpu 7631 7632When enabled, KVM will disable paravirtual features provided to the 7633guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf 7634(0x40000001). Otherwise, a guest may use the paravirtual features 7635regardless of what has actually been exposed through the CPUID leaf. 7636 7637.. _KVM_CAP_DIRTY_LOG_RING: 7638 7639 7640.. _cap_enable_vm: 7641 76427. Capabilities that can be enabled on VMs 7643========================================== 7644 7645There are certain capabilities that change the behavior of the virtual 7646machine when enabled. To enable them, please see section 7647:ref:`KVM_ENABLE_CAP`. Below you can find a list of capabilities and 7648what their effect on the VM is when enabling them. 7649 7650The following information is provided along with the description: 7651 7652 Architectures: 7653 which instruction set architectures provide this ioctl. 7654 x86 includes both i386 and x86_64. 7655 7656 Parameters: 7657 what parameters are accepted by the capability. 7658 7659 Returns: 7660 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 7661 are not detailed, but errors with specific meanings are. 7662 7663 76647.1 KVM_CAP_PPC_ENABLE_HCALL 7665---------------------------- 7666 7667:Architectures: ppc 7668:Parameters: args[0] is the sPAPR hcall number; 7669 args[1] is 0 to disable, 1 to enable in-kernel handling 7670 7671This capability controls whether individual sPAPR hypercalls (hcalls) 7672get handled by the kernel or not. Enabling or disabling in-kernel 7673handling of an hcall is effective across the VM. On creation, an 7674initial set of hcalls are enabled for in-kernel handling, which 7675consists of those hcalls for which in-kernel handlers were implemented 7676before this capability was implemented. If disabled, the kernel will 7677not to attempt to handle the hcall, but will always exit to userspace 7678to handle it. Note that it may not make sense to enable some and 7679disable others of a group of related hcalls, but KVM does not prevent 7680userspace from doing that. 7681 7682If the hcall number specified is not one that has an in-kernel 7683implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 7684error. 7685 76867.2 KVM_CAP_S390_USER_SIGP 7687-------------------------- 7688 7689:Architectures: s390 7690:Parameters: none 7691 7692This capability controls which SIGP orders will be handled completely in user 7693space. With this capability enabled, all fast orders will be handled completely 7694in the kernel: 7695 7696- SENSE 7697- SENSE RUNNING 7698- EXTERNAL CALL 7699- EMERGENCY SIGNAL 7700- CONDITIONAL EMERGENCY SIGNAL 7701 7702All other orders will be handled completely in user space. 7703 7704Only privileged operation exceptions will be checked for in the kernel (or even 7705in the hardware prior to interception). If this capability is not enabled, the 7706old way of handling SIGP orders is used (partially in kernel and user space). 7707 77087.3 KVM_CAP_S390_VECTOR_REGISTERS 7709--------------------------------- 7710 7711:Architectures: s390 7712:Parameters: none 7713:Returns: 0 on success, negative value on error 7714 7715Allows use of the vector registers introduced with z13 processor, and 7716provides for the synchronization between host and user space. Will 7717return -EINVAL if the machine does not support vectors. 7718 77197.4 KVM_CAP_S390_USER_STSI 7720-------------------------- 7721 7722:Architectures: s390 7723:Parameters: none 7724 7725This capability allows post-handlers for the STSI instruction. After 7726initial handling in the kernel, KVM exits to user space with 7727KVM_EXIT_S390_STSI to allow user space to insert further data. 7728 7729Before exiting to userspace, kvm handlers should fill in s390_stsi field of 7730vcpu->run:: 7731 7732 struct { 7733 __u64 addr; 7734 __u8 ar; 7735 __u8 reserved; 7736 __u8 fc; 7737 __u8 sel1; 7738 __u16 sel2; 7739 } s390_stsi; 7740 7741 @addr - guest address of STSI SYSIB 7742 @fc - function code 7743 @sel1 - selector 1 7744 @sel2 - selector 2 7745 @ar - access register number 7746 7747KVM handlers should exit to userspace with rc = -EREMOTE. 7748 77497.5 KVM_CAP_SPLIT_IRQCHIP 7750------------------------- 7751 7752:Architectures: x86 7753:Parameters: args[0] - number of routes reserved for userspace IOAPICs 7754:Returns: 0 on success, -1 on error 7755 7756Create a local apic for each processor in the kernel. This can be used 7757instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 7758IOAPIC and PIC (and also the PIT, even though this has to be enabled 7759separately). 7760 7761This capability also enables in kernel routing of interrupt requests; 7762when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 7763used in the IRQ routing table. The first args[0] MSI routes are reserved 7764for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 7765a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 7766 7767Fails if VCPU has already been created, or if the irqchip is already in the 7768kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 7769 77707.6 KVM_CAP_S390_RI 7771------------------- 7772 7773:Architectures: s390 7774:Parameters: none 7775 7776Allows use of runtime-instrumentation introduced with zEC12 processor. 7777Will return -EINVAL if the machine does not support runtime-instrumentation. 7778Will return -EBUSY if a VCPU has already been created. 7779 77807.7 KVM_CAP_X2APIC_API 7781---------------------- 7782 7783:Architectures: x86 7784:Parameters: args[0] - features that should be enabled 7785:Returns: 0 on success, -EINVAL when args[0] contains invalid features 7786 7787Valid feature flags in args[0] are:: 7788 7789 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 7790 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 7791 7792Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 7793KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 7794allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 7795respective sections. 7796 7797KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 7798in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 7799as a broadcast even in x2APIC mode in order to support physical x2APIC 7800without interrupt remapping. This is undesirable in logical mode, 7801where 0xff represents CPUs 0-7 in cluster 0. 7802 78037.8 KVM_CAP_S390_USER_INSTR0 7804---------------------------- 7805 7806:Architectures: s390 7807:Parameters: none 7808 7809With this capability enabled, all illegal instructions 0x0000 (2 bytes) will 7810be intercepted and forwarded to user space. User space can use this 7811mechanism e.g. to realize 2-byte software breakpoints. The kernel will 7812not inject an operating exception for these instructions, user space has 7813to take care of that. 7814 7815This capability can be enabled dynamically even if VCPUs were already 7816created and are running. 7817 78187.9 KVM_CAP_S390_GS 7819------------------- 7820 7821:Architectures: s390 7822:Parameters: none 7823:Returns: 0 on success; -EINVAL if the machine does not support 7824 guarded storage; -EBUSY if a VCPU has already been created. 7825 7826Allows use of guarded storage for the KVM guest. 7827 78287.10 KVM_CAP_S390_AIS 7829--------------------- 7830 7831:Architectures: s390 7832:Parameters: none 7833 7834Allow use of adapter-interruption suppression. 7835:Returns: 0 on success; -EBUSY if a VCPU has already been created. 7836 78377.11 KVM_CAP_PPC_SMT 7838-------------------- 7839 7840:Architectures: ppc 7841:Parameters: vsmt_mode, flags 7842 7843Enabling this capability on a VM provides userspace with a way to set 7844the desired virtual SMT mode (i.e. the number of virtual CPUs per 7845virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 7846between 1 and 8. On POWER8, vsmt_mode must also be no greater than 7847the number of threads per subcore for the host. Currently flags must 7848be 0. A successful call to enable this capability will result in 7849vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 7850subsequently queried for the VM. This capability is only supported by 7851HV KVM, and can only be set before any VCPUs have been created. 7852The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 7853modes are available. 7854 78557.12 KVM_CAP_PPC_FWNMI 7856---------------------- 7857 7858:Architectures: ppc 7859:Parameters: none 7860 7861With this capability a machine check exception in the guest address 7862space will cause KVM to exit the guest with NMI exit reason. This 7863enables QEMU to build error log and branch to guest kernel registered 7864machine check handling routine. Without this capability KVM will 7865branch to guests' 0x200 interrupt vector. 7866 78677.13 KVM_CAP_X86_DISABLE_EXITS 7868------------------------------ 7869 7870:Architectures: x86 7871:Parameters: args[0] defines which exits are disabled 7872:Returns: 0 on success, -EINVAL when args[0] contains invalid exits 7873 or if any vCPUs have already been created 7874 7875Valid bits in args[0] are:: 7876 7877 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) 7878 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1) 7879 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) 7880 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3) 7881 #define KVM_X86_DISABLE_EXITS_APERFMPERF (1 << 4) 7882 7883Enabling this capability on a VM provides userspace with a way to no 7884longer intercept some instructions for improved latency in some 7885workloads, and is suggested when vCPUs are associated to dedicated 7886physical CPUs. More bits can be added in the future; userspace can 7887just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable 7888all such vmexits. 7889 7890Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits. 7891 7892Virtualizing the ``IA32_APERF`` and ``IA32_MPERF`` MSRs requires more 7893than just disabling APERF/MPERF exits. While both Intel and AMD 7894document strict usage conditions for these MSRs--emphasizing that only 7895the ratio of their deltas over a time interval (T0 to T1) is 7896architecturally defined--simply passing through the MSRs can still 7897produce an incorrect ratio. 7898 7899This erroneous ratio can occur if, between T0 and T1: 7900 79011. The vCPU thread migrates between logical processors. 79022. Live migration or suspend/resume operations take place. 79033. Another task shares the vCPU's logical processor. 79044. C-states lower than C0 are emulated (e.g., via HLT interception). 79055. The guest TSC frequency doesn't match the host TSC frequency. 7906 7907Due to these complexities, KVM does not automatically associate this 7908passthrough capability with the guest CPUID bit, 7909``CPUID.6:ECX.APERFMPERF[bit 0]``. Userspace VMMs that deem this 7910mechanism adequate for virtualizing the ``IA32_APERF`` and 7911``IA32_MPERF`` MSRs must set the guest CPUID bit explicitly. 7912 7913 79147.14 KVM_CAP_S390_HPAGE_1M 7915-------------------------- 7916 7917:Architectures: s390 7918:Parameters: none 7919:Returns: 0 on success, -EINVAL if hpage module parameter was not set 7920 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 7921 flag set 7922 7923With this capability the KVM support for memory backing with 1m pages 7924through hugetlbfs can be enabled for a VM. After the capability is 7925enabled, cmma can't be enabled anymore and pfmfi and the storage key 7926interpretation are disabled. If cmma has already been enabled or the 7927hpage module parameter is not set to 1, -EINVAL is returned. 7928 7929While it is generally possible to create a huge page backed VM without 7930this capability, the VM will not be able to run. 7931 79327.15 KVM_CAP_MSR_PLATFORM_INFO 7933------------------------------ 7934 7935:Architectures: x86 7936:Parameters: args[0] whether feature should be enabled or not 7937 7938With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise, 7939a #GP would be raised when the guest tries to access. Currently, this 7940capability does not enable write permissions of this MSR for the guest. 7941 79427.16 KVM_CAP_PPC_NESTED_HV 7943-------------------------- 7944 7945:Architectures: ppc 7946:Parameters: none 7947:Returns: 0 on success, -EINVAL when the implementation doesn't support 7948 nested-HV virtualization. 7949 7950HV-KVM on POWER9 and later systems allows for "nested-HV" 7951virtualization, which provides a way for a guest VM to run guests that 7952can run using the CPU's supervisor mode (privileged non-hypervisor 7953state). Enabling this capability on a VM depends on the CPU having 7954the necessary functionality and on the facility being enabled with a 7955kvm-hv module parameter. 7956 79577.17 KVM_CAP_EXCEPTION_PAYLOAD 7958------------------------------ 7959 7960:Architectures: x86 7961:Parameters: args[0] whether feature should be enabled or not 7962 7963With this capability enabled, CR2 will not be modified prior to the 7964emulated VM-exit when L1 intercepts a #PF exception that occurs in 7965L2. Similarly, for kvm-intel only, DR6 will not be modified prior to 7966the emulated VM-exit when L1 intercepts a #DB exception that occurs in 7967L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or 7968#DB) exception for L2, exception.has_payload will be set and the 7969faulting address (or the new DR6 bits*) will be reported in the 7970exception_payload field. Similarly, when userspace injects a #PF (or 7971#DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set 7972exception.has_payload and to put the faulting address - or the new DR6 7973bits\ [#]_ - in the exception_payload field. 7974 7975This capability also enables exception.pending in struct 7976kvm_vcpu_events, which allows userspace to distinguish between pending 7977and injected exceptions. 7978 7979 7980.. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception 7981 will clear DR6.RTM. 7982 79837.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 7984-------------------------------------- 7985 7986:Architectures: x86, arm64, mips 7987:Parameters: args[0] whether feature should be enabled or not 7988 7989Valid flags are:: 7990 7991 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0) 7992 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1) 7993 7994With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not 7995automatically clear and write-protect all pages that are returned as dirty. 7996Rather, userspace will have to do this operation separately using 7997KVM_CLEAR_DIRTY_LOG. 7998 7999At the cost of a slightly more complicated operation, this provides better 8000scalability and responsiveness for two reasons. First, 8001KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather 8002than requiring to sync a full memslot; this ensures that KVM does not 8003take spinlocks for an extended period of time. Second, in some cases a 8004large amount of time can pass between a call to KVM_GET_DIRTY_LOG and 8005userspace actually using the data in the page. Pages can be modified 8006during this time, which is inefficient for both the guest and userspace: 8007the guest will incur a higher penalty due to write protection faults, 8008while userspace can see false reports of dirty pages. Manual reprotection 8009helps reducing this time, improving guest performance and reducing the 8010number of dirty log false positives. 8011 8012With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap 8013will be initialized to 1 when created. This also improves performance because 8014dirty logging can be enabled gradually in small chunks on the first call 8015to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on 8016KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on 8017x86 and arm64 for now). 8018 8019KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name 8020KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make 8021it hard or impossible to use it correctly. The availability of 8022KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed. 8023Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT. 8024 80257.19 KVM_CAP_PPC_SECURE_GUEST 8026------------------------------ 8027 8028:Architectures: ppc 8029 8030This capability indicates that KVM is running on a host that has 8031ultravisor firmware and thus can support a secure guest. On such a 8032system, a guest can ask the ultravisor to make it a secure guest, 8033one whose memory is inaccessible to the host except for pages which 8034are explicitly requested to be shared with the host. The ultravisor 8035notifies KVM when a guest requests to become a secure guest, and KVM 8036has the opportunity to veto the transition. 8037 8038If present, this capability can be enabled for a VM, meaning that KVM 8039will allow the transition to secure guest mode. Otherwise KVM will 8040veto the transition. 8041 80427.20 KVM_CAP_HALT_POLL 8043---------------------- 8044 8045:Architectures: all 8046:Target: VM 8047:Parameters: args[0] is the maximum poll time in nanoseconds 8048:Returns: 0 on success; -1 on error 8049 8050KVM_CAP_HALT_POLL overrides the kvm.halt_poll_ns module parameter to set the 8051maximum halt-polling time for all vCPUs in the target VM. This capability can 8052be invoked at any time and any number of times to dynamically change the 8053maximum halt-polling time. 8054 8055See Documentation/virt/kvm/halt-polling.rst for more information on halt 8056polling. 8057 80587.21 KVM_CAP_X86_USER_SPACE_MSR 8059------------------------------- 8060 8061:Architectures: x86 8062:Target: VM 8063:Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report 8064:Returns: 0 on success; -1 on error 8065 8066This capability allows userspace to intercept RDMSR and WRMSR instructions if 8067access to an MSR is denied. By default, KVM injects #GP on denied accesses. 8068 8069When a guest requests to read or write an MSR, KVM may not implement all MSRs 8070that are relevant to a respective system. It also does not differentiate by 8071CPU type. 8072 8073To allow more fine grained control over MSR handling, userspace may enable 8074this capability. With it enabled, MSR accesses that match the mask specified in 8075args[0] and would trigger a #GP inside the guest will instead trigger 8076KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications. Userspace 8077can then implement model specific MSR handling and/or user notifications 8078to inform a user that an MSR was not emulated/virtualized by KVM. 8079 8080The valid mask flags are: 8081 8082============================ =============================================== 8083 KVM_MSR_EXIT_REASON_UNKNOWN intercept accesses to unknown (to KVM) MSRs 8084 KVM_MSR_EXIT_REASON_INVAL intercept accesses that are architecturally 8085 invalid according to the vCPU model and/or mode 8086 KVM_MSR_EXIT_REASON_FILTER intercept accesses that are denied by userspace 8087 via KVM_X86_SET_MSR_FILTER 8088============================ =============================================== 8089 80907.22 KVM_CAP_X86_BUS_LOCK_EXIT 8091------------------------------- 8092 8093:Architectures: x86 8094:Target: VM 8095:Parameters: args[0] defines the policy used when bus locks detected in guest 8096:Returns: 0 on success, -EINVAL when args[0] contains invalid bits 8097 8098Valid bits in args[0] are:: 8099 8100 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0) 8101 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1) 8102 8103Enabling this capability on a VM provides userspace with a way to select a 8104policy to handle the bus locks detected in guest. Userspace can obtain the 8105supported modes from the result of KVM_CHECK_EXTENSION and define it through 8106the KVM_ENABLE_CAP. The supported modes are mutually-exclusive. 8107 8108This capability allows userspace to force VM exits on bus locks detected in the 8109guest, irrespective whether or not the host has enabled split-lock detection 8110(which triggers an #AC exception that KVM intercepts). This capability is 8111intended to mitigate attacks where a malicious/buggy guest can exploit bus 8112locks to degrade the performance of the whole system. 8113 8114If KVM_BUS_LOCK_DETECTION_OFF is set, KVM doesn't force guest bus locks to VM 8115exit, although the host kernel's split-lock #AC detection still applies, if 8116enabled. 8117 8118If KVM_BUS_LOCK_DETECTION_EXIT is set, KVM enables a CPU feature that ensures 8119bus locks in the guest trigger a VM exit, and KVM exits to userspace for all 8120such VM exits, e.g. to allow userspace to throttle the offending guest and/or 8121apply some other policy-based mitigation. When exiting to userspace, KVM sets 8122KVM_RUN_X86_BUS_LOCK in vcpu-run->flags, and conditionally sets the exit_reason 8123to KVM_EXIT_X86_BUS_LOCK. 8124 8125Due to differences in the underlying hardware implementation, the vCPU's RIP at 8126the time of exit diverges between Intel and AMD. On Intel hosts, RIP points at 8127the next instruction, i.e. the exit is trap-like. On AMD hosts, RIP points at 8128the offending instruction, i.e. the exit is fault-like. 8129 8130Note! Detected bus locks may be coincident with other exits to userspace, i.e. 8131KVM_RUN_X86_BUS_LOCK should be checked regardless of the primary exit reason if 8132userspace wants to take action on all detected bus locks. 8133 81347.23 KVM_CAP_PPC_DAWR1 8135---------------------- 8136 8137:Architectures: ppc 8138:Parameters: none 8139:Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR 8140 8141This capability can be used to check / enable 2nd DAWR feature provided 8142by POWER10 processor. 8143 8144 81457.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM 8146------------------------------------- 8147 8148:Architectures: x86 SEV enabled 8149:Type: vm 8150:Parameters: args[0] is the fd of the source vm 8151:Returns: 0 on success; ENOTTY on error 8152 8153This capability enables userspace to copy encryption context from the vm 8154indicated by the fd to the vm this is called on. 8155 8156This is intended to support in-guest workloads scheduled by the host. This 8157allows the in-guest workload to maintain its own NPTs and keeps the two vms 8158from accidentally clobbering each other with interrupts and the like (separate 8159APIC/MSRs/etc). 8160 81617.25 KVM_CAP_SGX_ATTRIBUTE 8162-------------------------- 8163 8164:Architectures: x86 8165:Target: VM 8166:Parameters: args[0] is a file handle of a SGX attribute file in securityfs 8167:Returns: 0 on success, -EINVAL if the file handle is invalid or if a requested 8168 attribute is not supported by KVM. 8169 8170KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM to grant a VM access to one or 8171more privileged enclave attributes. args[0] must hold a file handle to a valid 8172SGX attribute file corresponding to an attribute that is supported/restricted 8173by KVM (currently only PROVISIONKEY). 8174 8175The SGX subsystem restricts access to a subset of enclave attributes to provide 8176additional security for an uncompromised kernel, e.g. use of the PROVISIONKEY 8177is restricted to deter malware from using the PROVISIONKEY to obtain a stable 8178system fingerprint. To prevent userspace from circumventing such restrictions 8179by running an enclave in a VM, KVM prevents access to privileged attributes by 8180default. 8181 8182See Documentation/arch/x86/sgx.rst for more details. 8183 81847.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE 8185-------------------------------------- 8186 8187:Architectures: x86 8188:Parameters: args[0] whether the feature should be enabled or not 8189 8190When this capability is enabled, an emulation failure will result in an exit 8191to userspace with KVM_INTERNAL_ERROR (except when the emulator was invoked 8192to handle a VMware backdoor instruction). Furthermore, KVM will now provide up 8193to 15 instruction bytes for any exit to userspace resulting from an emulation 8194failure. When these exits to userspace occur use the emulation_failure struct 8195instead of the internal struct. They both have the same layout, but the 8196emulation_failure struct matches the content better. It also explicitly 8197defines the 'flags' field which is used to describe the fields in the struct 8198that are valid (ie: if KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES is 8199set in the 'flags' field then both 'insn_size' and 'insn_bytes' have valid data 8200in them.) 8201 82027.28 KVM_CAP_ARM_MTE 8203-------------------- 8204 8205:Architectures: arm64 8206:Parameters: none 8207 8208This capability indicates that KVM (and the hardware) supports exposing the 8209Memory Tagging Extensions (MTE) to the guest. It must also be enabled by the 8210VMM before creating any VCPUs to allow the guest access. Note that MTE is only 8211available to a guest running in AArch64 mode and enabling this capability will 8212cause attempts to create AArch32 VCPUs to fail. 8213 8214When enabled the guest is able to access tags associated with any memory given 8215to the guest. KVM will ensure that the tags are maintained during swap or 8216hibernation of the host; however the VMM needs to manually save/restore the 8217tags as appropriate if the VM is migrated. 8218 8219When this capability is enabled all memory in memslots must be mapped as 8220``MAP_ANONYMOUS`` or with a RAM-based file mapping (``tmpfs``, ``memfd``), 8221attempts to create a memslot with an invalid mmap will result in an 8222-EINVAL return. 8223 8224When enabled the VMM may make use of the ``KVM_ARM_MTE_COPY_TAGS`` ioctl to 8225perform a bulk copy of tags to/from the guest. 8226 82277.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 8228------------------------------------- 8229 8230:Architectures: x86 SEV enabled 8231:Type: vm 8232:Parameters: args[0] is the fd of the source vm 8233:Returns: 0 on success 8234 8235This capability enables userspace to migrate the encryption context from the VM 8236indicated by the fd to the VM this is called on. 8237 8238This is intended to support intra-host migration of VMs between userspace VMMs, 8239upgrading the VMM process without interrupting the guest. 8240 82417.31 KVM_CAP_DISABLE_QUIRKS2 8242---------------------------- 8243 8244:Parameters: args[0] - set of KVM quirks to disable 8245:Architectures: x86 8246:Type: vm 8247 8248This capability, if enabled, will cause KVM to disable some behavior 8249quirks. 8250 8251Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 8252quirks that can be disabled in KVM. 8253 8254The argument to KVM_ENABLE_CAP for this capability is a bitmask of 8255quirks to disable, and must be a subset of the bitmask returned by 8256KVM_CHECK_EXTENSION. 8257 8258The valid bits in cap.args[0] are: 8259 8260=================================== ============================================ 8261 KVM_X86_QUIRK_LINT0_REENABLED By default, the reset value for the LVT 8262 LINT0 register is 0x700 (APIC_MODE_EXTINT). 8263 When this quirk is disabled, the reset value 8264 is 0x10000 (APIC_LVT_MASKED). 8265 8266 KVM_X86_QUIRK_CD_NW_CLEARED By default, KVM clears CR0.CD and CR0.NW on 8267 AMD CPUs to workaround buggy guest firmware 8268 that runs in perpetuity with CR0.CD, i.e. 8269 with caches in "no fill" mode. 8270 8271 When this quirk is disabled, KVM does not 8272 change the value of CR0.CD and CR0.NW. 8273 8274 KVM_X86_QUIRK_LAPIC_MMIO_HOLE By default, the MMIO LAPIC interface is 8275 available even when configured for x2APIC 8276 mode. When this quirk is disabled, KVM 8277 disables the MMIO LAPIC interface if the 8278 LAPIC is in x2APIC mode. 8279 8280 KVM_X86_QUIRK_OUT_7E_INC_RIP By default, KVM pre-increments %rip before 8281 exiting to userspace for an OUT instruction 8282 to port 0x7e. When this quirk is disabled, 8283 KVM does not pre-increment %rip before 8284 exiting to userspace. 8285 8286 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this quirk is disabled, KVM sets 8287 CPUID.01H:ECX[bit 3] (MONITOR/MWAIT) if 8288 IA32_MISC_ENABLE[bit 18] (MWAIT) is set. 8289 Additionally, when this quirk is disabled, 8290 KVM clears CPUID.01H:ECX[bit 3] if 8291 IA32_MISC_ENABLE[bit 18] is cleared. 8292 8293 KVM_X86_QUIRK_FIX_HYPERCALL_INSN By default, KVM rewrites guest 8294 VMMCALL/VMCALL instructions to match the 8295 vendor's hypercall instruction for the 8296 system. When this quirk is disabled, KVM 8297 will no longer rewrite invalid guest 8298 hypercall instructions. Executing the 8299 incorrect hypercall instruction will 8300 generate a #UD within the guest. 8301 8302KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By default, KVM emulates MONITOR/MWAIT (if 8303 they are intercepted) as NOPs regardless of 8304 whether or not MONITOR/MWAIT are supported 8305 according to guest CPUID. When this quirk 8306 is disabled and KVM_X86_DISABLE_EXITS_MWAIT 8307 is not set (MONITOR/MWAIT are intercepted), 8308 KVM will inject a #UD on MONITOR/MWAIT if 8309 they're unsupported per guest CPUID. Note, 8310 KVM will modify MONITOR/MWAIT support in 8311 guest CPUID on writes to MISC_ENABLE if 8312 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT is 8313 disabled. 8314 8315KVM_X86_QUIRK_SLOT_ZAP_ALL By default, for KVM_X86_DEFAULT_VM VMs, KVM 8316 invalidates all SPTEs in all memslots and 8317 address spaces when a memslot is deleted or 8318 moved. When this quirk is disabled (or the 8319 VM type isn't KVM_X86_DEFAULT_VM), KVM only 8320 ensures the backing memory of the deleted 8321 or moved memslot isn't reachable, i.e KVM 8322 _may_ invalidate only SPTEs related to the 8323 memslot. 8324 8325KVM_X86_QUIRK_STUFF_FEATURE_MSRS By default, at vCPU creation, KVM sets the 8326 vCPU's MSR_IA32_PERF_CAPABILITIES (0x345), 8327 MSR_IA32_ARCH_CAPABILITIES (0x10a), 8328 MSR_PLATFORM_INFO (0xce), and all VMX MSRs 8329 (0x480..0x492) to the maximal capabilities 8330 supported by KVM. KVM also sets 8331 MSR_IA32_UCODE_REV (0x8b) to an arbitrary 8332 value (which is different for Intel vs. 8333 AMD). Lastly, when guest CPUID is set (by 8334 userspace), KVM modifies select VMX MSR 8335 fields to force consistency between guest 8336 CPUID and L2's effective ISA. When this 8337 quirk is disabled, KVM zeroes the vCPU's MSR 8338 values (with two exceptions, see below), 8339 i.e. treats the feature MSRs like CPUID 8340 leaves and gives userspace full control of 8341 the vCPU model definition. This quirk does 8342 not affect VMX MSRs CR0/CR4_FIXED1 (0x487 8343 and 0x489), as KVM does now allow them to 8344 be set by userspace (KVM sets them based on 8345 guest CPUID, for safety purposes). 8346 8347KVM_X86_QUIRK_IGNORE_GUEST_PAT By default, on Intel platforms, KVM ignores 8348 guest PAT and forces the effective memory 8349 type to WB in EPT. The quirk is not available 8350 on Intel platforms which are incapable of 8351 safely honoring guest PAT (i.e., without CPU 8352 self-snoop, KVM always ignores guest PAT and 8353 forces effective memory type to WB). It is 8354 also ignored on AMD platforms or, on Intel, 8355 when a VM has non-coherent DMA devices 8356 assigned; KVM always honors guest PAT in 8357 such case. The quirk is needed to avoid 8358 slowdowns on certain Intel Xeon platforms 8359 (e.g. ICX, SPR) where self-snoop feature is 8360 supported but UC is slow enough to cause 8361 issues with some older guests that use 8362 UC instead of WC to map the video RAM. 8363 Userspace can disable the quirk to honor 8364 guest PAT if it knows that there is no such 8365 guest software, for example if it does not 8366 expose a bochs graphics device (which is 8367 known to have had a buggy driver). 8368=================================== ============================================ 8369 83707.32 KVM_CAP_MAX_VCPU_ID 8371------------------------ 8372 8373:Architectures: x86 8374:Target: VM 8375:Parameters: args[0] - maximum APIC ID value set for current VM 8376:Returns: 0 on success, -EINVAL if args[0] is beyond KVM_MAX_VCPU_IDS 8377 supported in KVM or if it has been set. 8378 8379This capability allows userspace to specify maximum possible APIC ID 8380assigned for current VM session prior to the creation of vCPUs, saving 8381memory for data structures indexed by the APIC ID. Userspace is able 8382to calculate the limit to APIC ID values from designated 8383CPU topology. 8384 8385The value can be changed only until KVM_ENABLE_CAP is set to a nonzero 8386value or until a vCPU is created. Upon creation of the first vCPU, 8387if the value was set to zero or KVM_ENABLE_CAP was not invoked, KVM 8388uses the return value of KVM_CHECK_EXTENSION(KVM_CAP_MAX_VCPU_ID) as 8389the maximum APIC ID. 8390 83917.33 KVM_CAP_X86_NOTIFY_VMEXIT 8392------------------------------ 8393 8394:Architectures: x86 8395:Target: VM 8396:Parameters: args[0] is the value of notify window as well as some flags 8397:Returns: 0 on success, -EINVAL if args[0] contains invalid flags or notify 8398 VM exit is unsupported. 8399 8400Bits 63:32 of args[0] are used for notify window. 8401Bits 31:0 of args[0] are for some flags. Valid bits are:: 8402 8403 #define KVM_X86_NOTIFY_VMEXIT_ENABLED (1 << 0) 8404 #define KVM_X86_NOTIFY_VMEXIT_USER (1 << 1) 8405 8406This capability allows userspace to configure the notify VM exit on/off 8407in per-VM scope during VM creation. Notify VM exit is disabled by default. 8408When userspace sets KVM_X86_NOTIFY_VMEXIT_ENABLED bit in args[0], VMM will 8409enable this feature with the notify window provided, which will generate 8410a VM exit if no event window occurs in VM non-root mode for a specified of 8411time (notify window). 8412 8413If KVM_X86_NOTIFY_VMEXIT_USER is set in args[0], upon notify VM exits happen, 8414KVM would exit to userspace for handling. 8415 8416This capability is aimed to mitigate the threat that malicious VMs can 8417cause CPU stuck (due to event windows don't open up) and make the CPU 8418unavailable to host or other VMs. 8419 84207.35 KVM_CAP_X86_APIC_BUS_CYCLES_NS 8421----------------------------------- 8422 8423:Architectures: x86 8424:Target: VM 8425:Parameters: args[0] is the desired APIC bus clock rate, in nanoseconds 8426:Returns: 0 on success, -EINVAL if args[0] contains an invalid value for the 8427 frequency or if any vCPUs have been created, -ENXIO if a virtual 8428 local APIC has not been created using KVM_CREATE_IRQCHIP. 8429 8430This capability sets the VM's APIC bus clock frequency, used by KVM's in-kernel 8431virtual APIC when emulating APIC timers. KVM's default value can be retrieved 8432by KVM_CHECK_EXTENSION. 8433 8434Note: Userspace is responsible for correctly configuring CPUID 0x15, a.k.a. the 8435core crystal clock frequency, if a non-zero CPUID 0x15 is exposed to the guest. 8436 84377.36 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8438---------------------------------------------------------- 8439 8440:Architectures: x86, arm64, riscv 8441:Type: vm 8442:Parameters: args[0] - size of the dirty log ring 8443 8444KVM is capable of tracking dirty memory using ring buffers that are 8445mmapped into userspace; there is one dirty ring per vcpu. 8446 8447The dirty ring is available to userspace as an array of 8448``struct kvm_dirty_gfn``. Each dirty entry is defined as:: 8449 8450 struct kvm_dirty_gfn { 8451 __u32 flags; 8452 __u32 slot; /* as_id | slot_id */ 8453 __u64 offset; 8454 }; 8455 8456The following values are defined for the flags field to define the 8457current state of the entry:: 8458 8459 #define KVM_DIRTY_GFN_F_DIRTY BIT(0) 8460 #define KVM_DIRTY_GFN_F_RESET BIT(1) 8461 #define KVM_DIRTY_GFN_F_MASK 0x3 8462 8463Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM 8464ioctl to enable this capability for the new guest and set the size of 8465the rings. Enabling the capability is only allowed before creating any 8466vCPU, and the size of the ring must be a power of two. The larger the 8467ring buffer, the less likely the ring is full and the VM is forced to 8468exit to userspace. The optimal size depends on the workload, but it is 8469recommended that it be at least 64 KiB (4096 entries). 8470 8471Just like for dirty page bitmaps, the buffer tracks writes to 8472all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was 8473set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered 8474with the flag set, userspace can start harvesting dirty pages from the 8475ring buffer. 8476 8477An entry in the ring buffer can be unused (flag bits ``00``), 8478dirty (flag bits ``01``) or harvested (flag bits ``1X``). The 8479state machine for the entry is as follows:: 8480 8481 dirtied harvested reset 8482 00 -----------> 01 -------------> 1X -------+ 8483 ^ | 8484 | | 8485 +------------------------------------------+ 8486 8487To harvest the dirty pages, userspace accesses the mmapped ring buffer 8488to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage 8489the RESET bit must be cleared), then it means this GFN is a dirty GFN. 8490The userspace should harvest this GFN and mark the flags from state 8491``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set 8492to show that this GFN is harvested and waiting for a reset), and move 8493on to the next GFN. The userspace should continue to do this until the 8494flags of a GFN have the DIRTY bit cleared, meaning that it has harvested 8495all the dirty GFNs that were available. 8496 8497Note that on weakly ordered architectures, userspace accesses to the 8498ring buffer (and more specifically the 'flags' field) must be ordered, 8499using load-acquire/store-release accessors when available, or any 8500other memory barrier that will ensure this ordering. 8501 8502It's not necessary for userspace to harvest the all dirty GFNs at once. 8503However it must collect the dirty GFNs in sequence, i.e., the userspace 8504program cannot skip one dirty GFN to collect the one next to it. 8505 8506After processing one or more entries in the ring buffer, userspace 8507calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about 8508it, so that the kernel will reprotect those collected GFNs. 8509Therefore, the ioctl must be called *before* reading the content of 8510the dirty pages. 8511 8512The dirty ring can get full. When it happens, the KVM_RUN of the 8513vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL. 8514 8515The dirty ring interface has a major difference comparing to the 8516KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from 8517userspace, it's still possible that the kernel has not yet flushed the 8518processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the 8519flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one 8520needs to kick the vcpu out of KVM_RUN using a signal. The resulting 8521vmexit ensures that all dirty GFNs are flushed to the dirty rings. 8522 8523NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the only capability that 8524should be exposed by weakly ordered architecture, in order to indicate 8525the additional memory ordering requirements imposed on userspace when 8526reading the state of an entry and mutating it from DIRTY to HARVESTED. 8527Architecture with TSO-like ordering (such as x86) are allowed to 8528expose both KVM_CAP_DIRTY_LOG_RING and KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8529to userspace. 8530 8531After enabling the dirty rings, the userspace needs to detect the 8532capability of KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP to see whether the 8533ring structures can be backed by per-slot bitmaps. With this capability 8534advertised, it means the architecture can dirty guest pages without 8535vcpu/ring context, so that some of the dirty information will still be 8536maintained in the bitmap structure. KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP 8537can't be enabled if the capability of KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8538hasn't been enabled, or any memslot has been existing. 8539 8540Note that the bitmap here is only a backup of the ring structure. The 8541use of the ring and bitmap combination is only beneficial if there is 8542only a very small amount of memory that is dirtied out of vcpu/ring 8543context. Otherwise, the stand-alone per-slot bitmap mechanism needs to 8544be considered. 8545 8546To collect dirty bits in the backup bitmap, userspace can use the same 8547KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG isn't needed as long as all 8548the generation of the dirty bits is done in a single pass. Collecting 8549the dirty bitmap should be the very last thing that the VMM does before 8550considering the state as complete. VMM needs to ensure that the dirty 8551state is final and avoid missing dirty pages from another ioctl ordered 8552after the bitmap collection. 8553 8554NOTE: Multiple examples of using the backup bitmap: (1) save vgic/its 8555tables through command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} on 8556KVM device "kvm-arm-vgic-its". (2) restore vgic/its tables through 8557command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} on KVM device 8558"kvm-arm-vgic-its". VGICv3 LPI pending status is restored. (3) save 8559vgic3 pending table through KVM_DEV_ARM_VGIC_{GRP_CTRL, SAVE_PENDING_TABLES} 8560command on KVM device "kvm-arm-vgic-v3". 8561 85627.37 KVM_CAP_PMU_CAPABILITY 8563--------------------------- 8564 8565:Architectures: x86 8566:Type: vm 8567:Parameters: arg[0] is bitmask of PMU virtualization capabilities. 8568:Returns: 0 on success, -EINVAL when arg[0] contains invalid bits 8569 8570This capability alters PMU virtualization in KVM. 8571 8572Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 8573PMU virtualization capabilities that can be adjusted on a VM. 8574 8575The argument to KVM_ENABLE_CAP is also a bitmask and selects specific 8576PMU virtualization capabilities to be applied to the VM. This can 8577only be invoked on a VM prior to the creation of VCPUs. 8578 8579At this time, KVM_PMU_CAP_DISABLE is the only capability. Setting 8580this capability will disable PMU virtualization for that VM. Usermode 8581should adjust CPUID leaf 0xA to reflect that the PMU is disabled. 8582 85837.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES 8584------------------------------------- 8585 8586:Architectures: x86 8587:Type: vm 8588:Parameters: arg[0] must be 0. 8589:Returns: 0 on success, -EPERM if the userspace process does not 8590 have CAP_SYS_BOOT, -EINVAL if args[0] is not 0 or any vCPUs have been 8591 created. 8592 8593This capability disables the NX huge pages mitigation for iTLB MULTIHIT. 8594 8595The capability has no effect if the nx_huge_pages module parameter is not set. 8596 8597This capability may only be set before any vCPUs are created. 8598 85997.39 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE 8600--------------------------------------- 8601 8602:Architectures: arm64 8603:Type: vm 8604:Parameters: arg[0] is the new split chunk size. 8605:Returns: 0 on success, -EINVAL if any memslot was already created. 8606 8607This capability sets the chunk size used in Eager Page Splitting. 8608 8609Eager Page Splitting improves the performance of dirty-logging (used 8610in live migrations) when guest memory is backed by huge-pages. It 8611avoids splitting huge-pages (into PAGE_SIZE pages) on fault, by doing 8612it eagerly when enabling dirty logging (with the 8613KVM_MEM_LOG_DIRTY_PAGES flag for a memory region), or when using 8614KVM_CLEAR_DIRTY_LOG. 8615 8616The chunk size specifies how many pages to break at a time, using a 8617single allocation for each chunk. Bigger the chunk size, more pages 8618need to be allocated ahead of time. 8619 8620The chunk size needs to be a valid block size. The list of acceptable 8621block sizes is exposed in KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES as a 862264-bit bitmap (each bit describing a block size). The default value is 86230, to disable the eager page splitting. 8624 86257.40 KVM_CAP_EXIT_HYPERCALL 8626--------------------------- 8627 8628:Architectures: x86 8629:Type: vm 8630 8631This capability, if enabled, will cause KVM to exit to userspace 8632with KVM_EXIT_HYPERCALL exit reason to process some hypercalls. 8633 8634Calling KVM_CHECK_EXTENSION for this capability will return a bitmask 8635of hypercalls that can be configured to exit to userspace. 8636Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE. 8637 8638The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset 8639of the result of KVM_CHECK_EXTENSION. KVM will forward to userspace 8640the hypercalls whose corresponding bit is in the argument, and return 8641ENOSYS for the others. 8642 86437.41 KVM_CAP_ARM_SYSTEM_SUSPEND 8644------------------------------- 8645 8646:Architectures: arm64 8647:Type: vm 8648 8649When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of 8650type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request. 8651 86527.42 KVM_CAP_ARM_WRITABLE_IMP_ID_REGS 8653------------------------------------- 8654 8655:Architectures: arm64 8656:Target: VM 8657:Parameters: None 8658:Returns: 0 on success, -EINVAL if vCPUs have been created before enabling this 8659 capability. 8660 8661This capability changes the behavior of the registers that identify a PE 8662implementation of the Arm architecture: MIDR_EL1, REVIDR_EL1, and AIDR_EL1. 8663By default, these registers are visible to userspace but treated as invariant. 8664 8665When this capability is enabled, KVM allows userspace to change the 8666aforementioned registers before the first KVM_RUN. These registers are VM 8667scoped, meaning that the same set of values are presented on all vCPUs in a 8668given VM. 8669 86707.43 KVM_CAP_RISCV_MP_STATE_RESET 8671--------------------------------- 8672 8673:Architectures: riscv 8674:Type: VM 8675:Parameters: None 8676:Returns: 0 on success, -EINVAL if arg[0] is not zero 8677 8678When this capability is enabled, KVM resets the VCPU when setting 8679MP_STATE_INIT_RECEIVED through IOCTL. The original MP_STATE is preserved. 8680 86817.43 KVM_CAP_ARM_CACHEABLE_PFNMAP_SUPPORTED 8682------------------------------------------- 8683 8684:Architectures: arm64 8685:Target: VM 8686:Parameters: None 8687 8688This capability indicate to the userspace whether a PFNMAP memory region 8689can be safely mapped as cacheable. This relies on the presence of 8690force write back (FWB) feature support on the hardware. 8691 86928. Other capabilities. 8693====================== 8694 8695This section lists capabilities that give information about other 8696features of the KVM implementation. 8697 86988.1 KVM_CAP_PPC_HWRNG 8699--------------------- 8700 8701:Architectures: ppc 8702 8703This capability, if KVM_CHECK_EXTENSION indicates that it is 8704available, means that the kernel has an implementation of the 8705H_RANDOM hypercall backed by a hardware random-number generator. 8706If present, the kernel H_RANDOM handler can be enabled for guest use 8707with the KVM_CAP_PPC_ENABLE_HCALL capability. 8708 87098.3 KVM_CAP_PPC_MMU_RADIX 8710------------------------- 8711 8712:Architectures: ppc 8713 8714This capability, if KVM_CHECK_EXTENSION indicates that it is 8715available, means that the kernel can support guests using the 8716radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 8717processor). 8718 87198.4 KVM_CAP_PPC_MMU_HASH_V3 8720--------------------------- 8721 8722:Architectures: ppc 8723 8724This capability, if KVM_CHECK_EXTENSION indicates that it is 8725available, means that the kernel can support guests using the 8726hashed page table MMU defined in Power ISA V3.00 (as implemented in 8727the POWER9 processor), including in-memory segment tables. 8728 87298.5 KVM_CAP_MIPS_VZ 8730------------------- 8731 8732:Architectures: mips 8733 8734This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 8735it is available, means that full hardware assisted virtualization capabilities 8736of the hardware are available for use through KVM. An appropriate 8737KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 8738utilises it. 8739 8740If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 8741available, it means that the VM is using full hardware assisted virtualization 8742capabilities of the hardware. This is useful to check after creating a VM with 8743KVM_VM_MIPS_DEFAULT. 8744 8745The value returned by KVM_CHECK_EXTENSION should be compared against known 8746values (see below). All other values are reserved. This is to allow for the 8747possibility of other hardware assisted virtualization implementations which 8748may be incompatible with the MIPS VZ ASE. 8749 8750== ========================================================================== 8751 0 The trap & emulate implementation is in use to run guest code in user 8752 mode. Guest virtual memory segments are rearranged to fit the guest in the 8753 user mode address space. 8754 8755 1 The MIPS VZ ASE is in use, providing full hardware assisted 8756 virtualization, including standard guest virtual memory segments. 8757== ========================================================================== 8758 87598.7 KVM_CAP_MIPS_64BIT 8760---------------------- 8761 8762:Architectures: mips 8763 8764This capability indicates the supported architecture type of the guest, i.e. the 8765supported register and address width. 8766 8767The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 8768kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 8769be checked specifically against known values (see below). All other values are 8770reserved. 8771 8772== ======================================================================== 8773 0 MIPS32 or microMIPS32. 8774 Both registers and addresses are 32-bits wide. 8775 It will only be possible to run 32-bit guest code. 8776 8777 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 8778 Registers are 64-bits wide, but addresses are 32-bits wide. 8779 64-bit guest code may run but cannot access MIPS64 memory segments. 8780 It will also be possible to run 32-bit guest code. 8781 8782 2 MIPS64 or microMIPS64 with access to all address segments. 8783 Both registers and addresses are 64-bits wide. 8784 It will be possible to run 64-bit or 32-bit guest code. 8785== ======================================================================== 8786 87878.9 KVM_CAP_ARM_USER_IRQ 8788------------------------ 8789 8790:Architectures: arm64 8791 8792This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 8793that if userspace creates a VM without an in-kernel interrupt controller, it 8794will be notified of changes to the output level of in-kernel emulated devices, 8795which can generate virtual interrupts, presented to the VM. 8796For such VMs, on every return to userspace, the kernel 8797updates the vcpu's run->s.regs.device_irq_level field to represent the actual 8798output level of the device. 8799 8800Whenever kvm detects a change in the device output level, kvm guarantees at 8801least one return to userspace before running the VM. This exit could either 8802be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 8803userspace can always sample the device output level and re-compute the state of 8804the userspace interrupt controller. Userspace should always check the state 8805of run->s.regs.device_irq_level on every kvm exit. 8806The value in run->s.regs.device_irq_level can represent both level and edge 8807triggered interrupt signals, depending on the device. Edge triggered interrupt 8808signals will exit to userspace with the bit in run->s.regs.device_irq_level 8809set exactly once per edge signal. 8810 8811The field run->s.regs.device_irq_level is available independent of 8812run->kvm_valid_regs or run->kvm_dirty_regs bits. 8813 8814If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 8815number larger than 0 indicating the version of this capability is implemented 8816and thereby which bits in run->s.regs.device_irq_level can signal values. 8817 8818Currently the following bits are defined for the device_irq_level bitmap:: 8819 8820 KVM_CAP_ARM_USER_IRQ >= 1: 8821 8822 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 8823 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 8824 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 8825 8826Future versions of kvm may implement additional events. These will get 8827indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 8828listed above. 8829 88308.10 KVM_CAP_PPC_SMT_POSSIBLE 8831----------------------------- 8832 8833:Architectures: ppc 8834 8835Querying this capability returns a bitmap indicating the possible 8836virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 8837(counting from the right) is set, then a virtual SMT mode of 2^N is 8838available. 8839 88408.12 KVM_CAP_HYPERV_VP_INDEX 8841---------------------------- 8842 8843:Architectures: x86 8844 8845This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 8846value is used to denote the target vcpu for a SynIC interrupt. For 8847compatibility, KVM initializes this msr to KVM's internal vcpu index. When this 8848capability is absent, userspace can still query this msr's value. 8849 88508.13 KVM_CAP_S390_AIS_MIGRATION 8851------------------------------- 8852 8853:Architectures: s390 8854 8855This capability indicates if the flic device will be able to get/set the 8856AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 8857to discover this without having to create a flic device. 8858 88598.14 KVM_CAP_S390_PSW 8860--------------------- 8861 8862:Architectures: s390 8863 8864This capability indicates that the PSW is exposed via the kvm_run structure. 8865 88668.15 KVM_CAP_S390_GMAP 8867---------------------- 8868 8869:Architectures: s390 8870 8871This capability indicates that the user space memory used as guest mapping can 8872be anywhere in the user memory address space, as long as the memory slots are 8873aligned and sized to a segment (1MB) boundary. 8874 88758.16 KVM_CAP_S390_COW 8876--------------------- 8877 8878:Architectures: s390 8879 8880This capability indicates that the user space memory used as guest mapping can 8881use copy-on-write semantics as well as dirty pages tracking via read-only page 8882tables. 8883 88848.17 KVM_CAP_S390_BPB 8885--------------------- 8886 8887:Architectures: s390 8888 8889This capability indicates that kvm will implement the interfaces to handle 8890reset, migration and nested KVM for branch prediction blocking. The stfle 8891facility 82 should not be provided to the guest without this capability. 8892 88938.18 KVM_CAP_HYPERV_TLBFLUSH 8894---------------------------- 8895 8896:Architectures: x86 8897 8898This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush 8899hypercalls: 8900HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx, 8901HvFlushVirtualAddressList, HvFlushVirtualAddressListEx. 8902 89038.19 KVM_CAP_ARM_INJECT_SERROR_ESR 8904---------------------------------- 8905 8906:Architectures: arm64 8907 8908This capability indicates that userspace can specify (via the 8909KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it 8910takes a virtual SError interrupt exception. 8911If KVM advertises this capability, userspace can only specify the ISS field for 8912the ESR syndrome. Other parts of the ESR, such as the EC are generated by the 8913CPU when the exception is taken. If this virtual SError is taken to EL1 using 8914AArch64, this value will be reported in the ISS field of ESR_ELx. 8915 8916See KVM_CAP_VCPU_EVENTS for more details. 8917 89188.20 KVM_CAP_HYPERV_SEND_IPI 8919---------------------------- 8920 8921:Architectures: x86 8922 8923This capability indicates that KVM supports paravirtualized Hyper-V IPI send 8924hypercalls: 8925HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx. 8926 89278.22 KVM_CAP_S390_VCPU_RESETS 8928----------------------------- 8929 8930:Architectures: s390 8931 8932This capability indicates that the KVM_S390_NORMAL_RESET and 8933KVM_S390_CLEAR_RESET ioctls are available. 8934 89358.23 KVM_CAP_S390_PROTECTED 8936--------------------------- 8937 8938:Architectures: s390 8939 8940This capability indicates that the Ultravisor has been initialized and 8941KVM can therefore start protected VMs. 8942This capability governs the KVM_S390_PV_COMMAND ioctl and the 8943KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected 8944guests when the state change is invalid. 8945 89468.24 KVM_CAP_STEAL_TIME 8947----------------------- 8948 8949:Architectures: arm64, x86 8950 8951This capability indicates that KVM supports steal time accounting. 8952When steal time accounting is supported it may be enabled with 8953architecture-specific interfaces. This capability and the architecture- 8954specific interfaces must be consistent, i.e. if one says the feature 8955is supported, than the other should as well and vice versa. For arm64 8956see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL". 8957For x86 see Documentation/virt/kvm/x86/msr.rst "MSR_KVM_STEAL_TIME". 8958 89598.25 KVM_CAP_S390_DIAG318 8960------------------------- 8961 8962:Architectures: s390 8963 8964This capability enables a guest to set information about its control program 8965(i.e. guest kernel type and version). The information is helpful during 8966system/firmware service events, providing additional data about the guest 8967environments running on the machine. 8968 8969The information is associated with the DIAGNOSE 0x318 instruction, which sets 8970an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and 8971a 7-byte Control Program Version Code (CPVC). The CPNC determines what 8972environment the control program is running in (e.g. Linux, z/VM...), and the 8973CPVC is used for information specific to OS (e.g. Linux version, Linux 8974distribution...) 8975 8976If this capability is available, then the CPNC and CPVC can be synchronized 8977between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318). 8978 89798.26 KVM_CAP_X86_USER_SPACE_MSR 8980------------------------------- 8981 8982:Architectures: x86 8983 8984This capability indicates that KVM supports deflection of MSR reads and 8985writes to user space. It can be enabled on a VM level. If enabled, MSR 8986accesses that would usually trigger a #GP by KVM into the guest will 8987instead get bounced to user space through the KVM_EXIT_X86_RDMSR and 8988KVM_EXIT_X86_WRMSR exit notifications. 8989 89908.27 KVM_CAP_X86_MSR_FILTER 8991--------------------------- 8992 8993:Architectures: x86 8994 8995This capability indicates that KVM supports that accesses to user defined MSRs 8996may be rejected. With this capability exposed, KVM exports new VM ioctl 8997KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR 8998ranges that KVM should deny access to. 8999 9000In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to 9001trap and emulate MSRs that are outside of the scope of KVM as well as 9002limit the attack surface on KVM's MSR emulation code. 9003 90048.30 KVM_CAP_XEN_HVM 9005-------------------- 9006 9007:Architectures: x86 9008 9009This capability indicates the features that Xen supports for hosting Xen 9010PVHVM guests. Valid flags are:: 9011 9012 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0) 9013 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1) 9014 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2) 9015 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 3) 9016 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL (1 << 4) 9017 #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND (1 << 5) 9018 #define KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG (1 << 6) 9019 #define KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE (1 << 7) 9020 9021The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG 9022ioctl is available, for the guest to set its hypercall page. 9023 9024If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be 9025provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page 9026contents, to request that KVM generate hypercall page content automatically 9027and also enable interception of guest hypercalls with KVM_EXIT_XEN. 9028 9029The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the 9030KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and 9031KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors 9032for event channel upcalls when the evtchn_upcall_pending field of a vcpu's 9033vcpu_info is set. 9034 9035The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related 9036features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are 9037supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls. 9038 9039The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries 9040of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority 9041field set to indicate 2 level event channel delivery. 9042 9043The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates that KVM supports 9044injecting event channel events directly into the guest with the 9045KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indicates support for the 9046KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attributes and the 9047KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_VECTOR vCPU attributes. 9048related to event channel delivery, timers, and the XENVER_version 9049interception. 9050 9051The KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG flag indicates that KVM supports 9052the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute in the KVM_XEN_SET_ATTR 9053and KVM_XEN_GET_ATTR ioctls. This controls whether KVM will set the 9054XEN_RUNSTATE_UPDATE flag in guest memory mapped vcpu_runstate_info during 9055updates of the runstate information. Note that versions of KVM which support 9056the RUNSTATE feature above, but not the RUNSTATE_UPDATE_FLAG feature, will 9057always set the XEN_RUNSTATE_UPDATE flag when updating the guest structure, 9058which is perhaps counterintuitive. When this flag is advertised, KVM will 9059behave more correctly, not using the XEN_RUNSTATE_UPDATE flag until/unless 9060specifically enabled (by the guest making the hypercall, causing the VMM 9061to enable the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute). 9062 9063The KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag indicates that KVM supports 9064clearing the PVCLOCK_TSC_STABLE_BIT flag in Xen pvclock sources. This will be 9065done when the KVM_CAP_XEN_HVM ioctl sets the 9066KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag. 9067 90688.31 KVM_CAP_SPAPR_MULTITCE 9069--------------------------- 9070 9071:Architectures: ppc 9072:Type: vm 9073 9074This capability means the kernel is capable of handling hypercalls 9075H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 9076space. This significantly accelerates DMA operations for PPC KVM guests. 9077User space should expect that its handlers for these hypercalls 9078are not going to be called if user space previously registered LIOBN 9079in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 9080 9081In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 9082user space might have to advertise it for the guest. For example, 9083IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 9084present in the "ibm,hypertas-functions" device-tree property. 9085 9086The hypercalls mentioned above may or may not be processed successfully 9087in the kernel based fast path. If they can not be handled by the kernel, 9088they will get passed on to user space. So user space still has to have 9089an implementation for these despite the in kernel acceleration. 9090 9091This capability is always enabled. 9092 90938.32 KVM_CAP_PTP_KVM 9094-------------------- 9095 9096:Architectures: arm64 9097 9098This capability indicates that the KVM virtual PTP service is 9099supported in the host. A VMM can check whether the service is 9100available to the guest on migration. 9101 91028.37 KVM_CAP_S390_PROTECTED_DUMP 9103-------------------------------- 9104 9105:Architectures: s390 9106:Type: vm 9107 9108This capability indicates that KVM and the Ultravisor support dumping 9109PV guests. The `KVM_PV_DUMP` command is available for the 9110`KVM_S390_PV_COMMAND` ioctl and the `KVM_PV_INFO` command provides 9111dump related UV data. Also the vcpu ioctl `KVM_S390_PV_CPU_COMMAND` is 9112available and supports the `KVM_PV_DUMP_CPU` subcommand. 9113 91148.39 KVM_CAP_S390_CPU_TOPOLOGY 9115------------------------------ 9116 9117:Architectures: s390 9118:Type: vm 9119 9120This capability indicates that KVM will provide the S390 CPU Topology 9121facility which consist of the interpretation of the PTF instruction for 9122the function code 2 along with interception and forwarding of both the 9123PTF instruction with function codes 0 or 1 and the STSI(15,1,x) 9124instruction to the userland hypervisor. 9125 9126The stfle facility 11, CPU Topology facility, should not be indicated 9127to the guest without this capability. 9128 9129When this capability is present, KVM provides a new attribute group 9130on vm fd, KVM_S390_VM_CPU_TOPOLOGY. 9131This new attribute allows to get, set or clear the Modified Change 9132Topology Report (MTCR) bit of the SCA through the kvm_device_attr 9133structure. 9134 9135When getting the Modified Change Topology Report value, the attr->addr 9136must point to a byte where the value will be stored or retrieved from. 9137 91388.41 KVM_CAP_VM_TYPES 9139--------------------- 9140 9141:Architectures: x86 9142:Type: system ioctl 9143 9144This capability returns a bitmap of support VM types. The 1-setting of bit @n 9145means the VM type with value @n is supported. Possible values of @n are:: 9146 9147 #define KVM_X86_DEFAULT_VM 0 9148 #define KVM_X86_SW_PROTECTED_VM 1 9149 #define KVM_X86_SEV_VM 2 9150 #define KVM_X86_SEV_ES_VM 3 9151 9152Note, KVM_X86_SW_PROTECTED_VM is currently only for development and testing. 9153Do not use KVM_X86_SW_PROTECTED_VM for "real" VMs, and especially not in 9154production. The behavior and effective ABI for software-protected VMs is 9155unstable. 9156 91578.42 KVM_CAP_PPC_RPT_INVALIDATE 9158------------------------------- 9159 9160:Architectures: ppc 9161 9162This capability indicates that the kernel is capable of handling 9163H_RPT_INVALIDATE hcall. 9164 9165In order to enable the use of H_RPT_INVALIDATE in the guest, 9166user space might have to advertise it for the guest. For example, 9167IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is 9168present in the "ibm,hypertas-functions" device-tree property. 9169 9170This capability is enabled for hypervisors on platforms like POWER9 9171that support radix MMU. 9172 91738.43 KVM_CAP_PPC_AIL_MODE_3 9174--------------------------- 9175 9176:Architectures: ppc 9177 9178This capability indicates that the kernel supports the mode 3 setting for the 9179"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location" 9180resource that is controlled with the H_SET_MODE hypercall. 9181 9182This capability allows a guest kernel to use a better-performance mode for 9183handling interrupts and system calls. 9184 91858.44 KVM_CAP_MEMORY_FAULT_INFO 9186------------------------------ 9187 9188:Architectures: x86 9189 9190The presence of this capability indicates that KVM_RUN will fill 9191kvm_run.memory_fault if KVM cannot resolve a guest page fault VM-Exit, e.g. if 9192there is a valid memslot but no backing VMA for the corresponding host virtual 9193address. 9194 9195The information in kvm_run.memory_fault is valid if and only if KVM_RUN returns 9196an error with errno=EFAULT or errno=EHWPOISON *and* kvm_run.exit_reason is set 9197to KVM_EXIT_MEMORY_FAULT. 9198 9199Note: Userspaces which attempt to resolve memory faults so that they can retry 9200KVM_RUN are encouraged to guard against repeatedly receiving the same 9201error/annotated fault. 9202 9203See KVM_EXIT_MEMORY_FAULT for more information. 9204 92058.45 KVM_CAP_X86_GUEST_MODE 9206--------------------------- 9207 9208:Architectures: x86 9209 9210The presence of this capability indicates that KVM_RUN will update the 9211KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags to indicate whether the 9212vCPU was executing nested guest code when it exited. 9213 9214KVM exits with the register state of either the L1 or L2 guest 9215depending on which executed at the time of an exit. Userspace must 9216take care to differentiate between these cases. 9217 92189. Known KVM API problems 9219========================= 9220 9221In some cases, KVM's API has some inconsistencies or common pitfalls 9222that userspace need to be aware of. This section details some of 9223these issues. 9224 9225Most of them are architecture specific, so the section is split by 9226architecture. 9227 92289.1. x86 9229-------- 9230 9231``KVM_GET_SUPPORTED_CPUID`` issues 9232^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9233 9234In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible 9235to take its result and pass it directly to ``KVM_SET_CPUID2``. This section 9236documents some cases in which that requires some care. 9237 9238Local APIC features 9239~~~~~~~~~~~~~~~~~~~ 9240 9241CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``, 9242but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or 9243``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of 9244the local APIC. 9245 9246The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature. 9247 9248On older versions of Linux, CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by 9249``KVM_GET_SUPPORTED_CPUID``, but it can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` 9250is present and the kernel has enabled in-kernel emulation of the local APIC. 9251On newer versions, ``KVM_GET_SUPPORTED_CPUID`` does report the bit as available. 9252 9253CPU topology 9254~~~~~~~~~~~~ 9255 9256Several CPUID values include topology information for the host CPU: 92570x0b and 0x1f for Intel systems, 0x8000001e for AMD systems. Different 9258versions of KVM return different values for this information and userspace 9259should not rely on it. Currently they return all zeroes. 9260 9261If userspace wishes to set up a guest topology, it should be careful that 9262the values of these three leaves differ for each CPU. In particular, 9263the APIC ID is found in EDX for all subleaves of 0x0b and 0x1f, and in EAX 9264for 0x8000001e; the latter also encodes the core id and node id in bits 92657:0 of EBX and ECX respectively. 9266 9267Obsolete ioctls and capabilities 9268^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9269 9270KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually 9271available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if 9272available. 9273 9274Ordering of KVM_GET_*/KVM_SET_* ioctls 9275^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9276 9277TBD 9278