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. This ioctl can only be 860called before creating any vcpus. 861On arm64, a GICv2 is created. Any other GIC versions require the usage of 862KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 863KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 864On s390, a dummy irq routing table is created. 865 866On x86, subsequent vcpu creation may install a private 4 KiB memory slot at the 867default APIC base address (0xfee00000). User memory regions must not overlap 868this address; doing so will cause vcpu creation to fail with ``EEXIST``, or the 869memory region to be rejected if created after the vcpu. This occurs when 870APIC access acceleration is enabled (APICv on Intel, AVIC on AMD), which is 871the default on supported hardware. The same constraint applies when using 872``KVM_CAP_SPLIT_IRQCHIP``. 873 874Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 875before KVM_CREATE_IRQCHIP can be used. 876 877 8784.25 KVM_IRQ_LINE 879----------------- 880 881:Capability: KVM_CAP_IRQCHIP 882:Architectures: x86, arm64 883:Type: vm ioctl 884:Parameters: struct kvm_irq_level 885:Returns: 0 on success, -1 on error 886 887Sets the level of a GSI input to the interrupt controller model in the kernel. 888On some architectures it is required that an interrupt controller model has 889been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 890interrupts require the level to be set to 1 and then back to 0. 891 892On real hardware, interrupt pins can be active-low or active-high. This 893does not matter for the level field of struct kvm_irq_level: 1 always 894means active (asserted), 0 means inactive (deasserted). 895 896x86 allows the operating system to program the interrupt polarity 897(active-low/active-high) for level-triggered interrupts, and KVM used 898to consider the polarity. However, due to bitrot in the handling of 899active-low interrupts, the above convention is now valid on x86 too. 900This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 901should not present interrupts to the guest as active-low unless this 902capability is present (or unless it is not using the in-kernel irqchip, 903of course). 904 905 906arm64 can signal an interrupt either at the CPU level, or at the 907in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 908use PPIs designated for specific cpus. The irq field is interpreted 909like this:: 910 911 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 | 912 field: | vcpu2_index | irq_type | vcpu_index | irq_id | 913 914The irq_type field has the following values: 915 916- KVM_ARM_IRQ_TYPE_CPU: 917 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 918- KVM_ARM_IRQ_TYPE_SPI: 919 in-kernel GICv2/GICv3: SPI, irq_id between 32 and 1019 (incl.) 920 (the vcpu_index field is ignored) 921 in-kernel GICv5: SPI, irq_id between 0 and 65535 (incl.) 922- KVM_ARM_IRQ_TYPE_PPI: 923 in-kernel GICv2/GICv3: PPI, irq_id between 16 and 31 (incl.) 924 in-kernel GICv5: PPI, irq_id between 0 and 127 (incl.) 925 926(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 927 928In both cases, level is used to assert/deassert the line. 929 930When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is 931identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index 932must be zero. 933 934Note that on arm64, the KVM_CAP_IRQCHIP capability only conditions 935injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always 936be used for a userspace interrupt controller. 937 938:: 939 940 struct kvm_irq_level { 941 union { 942 __u32 irq; /* GSI */ 943 __s32 status; /* not used for KVM_IRQ_LEVEL */ 944 }; 945 __u32 level; /* 0 or 1 */ 946 }; 947 948 9494.26 KVM_GET_IRQCHIP 950-------------------- 951 952:Capability: KVM_CAP_IRQCHIP 953:Architectures: x86 954:Type: vm ioctl 955:Parameters: struct kvm_irqchip (in/out) 956:Returns: 0 on success, -1 on error 957 958Reads the state of a kernel interrupt controller created with 959KVM_CREATE_IRQCHIP into a buffer provided by the caller. 960 961:: 962 963 struct kvm_irqchip { 964 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 965 __u32 pad; 966 union { 967 char dummy[512]; /* reserving space */ 968 struct kvm_pic_state pic; 969 struct kvm_ioapic_state ioapic; 970 } chip; 971 }; 972 973 9744.27 KVM_SET_IRQCHIP 975-------------------- 976 977:Capability: KVM_CAP_IRQCHIP 978:Architectures: x86 979:Type: vm ioctl 980:Parameters: struct kvm_irqchip (in) 981:Returns: 0 on success, -1 on error 982 983Sets the state of a kernel interrupt controller created with 984KVM_CREATE_IRQCHIP from a buffer provided by the caller. 985 986:: 987 988 struct kvm_irqchip { 989 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 990 __u32 pad; 991 union { 992 char dummy[512]; /* reserving space */ 993 struct kvm_pic_state pic; 994 struct kvm_ioapic_state ioapic; 995 } chip; 996 }; 997 998 9994.28 KVM_XEN_HVM_CONFIG 1000----------------------- 1001 1002:Capability: KVM_CAP_XEN_HVM 1003:Architectures: x86 1004:Type: vm ioctl 1005:Parameters: struct kvm_xen_hvm_config (in) 1006:Returns: 0 on success, -1 on error 1007 1008Sets the MSR that the Xen HVM guest uses to initialize its hypercall 1009page, and provides the starting address and size of the hypercall 1010blobs in userspace. When the guest writes the MSR, kvm copies one 1011page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 1012memory. 1013 1014The MSR index must be in the range [0x40000000, 0x4fffffff], i.e. must reside 1015in the range that is unofficially reserved for use by hypervisors. The min/max 1016values are enumerated via KVM_XEN_MSR_MIN_INDEX and KVM_XEN_MSR_MAX_INDEX. 1017 1018:: 1019 1020 struct kvm_xen_hvm_config { 1021 __u32 flags; 1022 __u32 msr; 1023 __u64 blob_addr_32; 1024 __u64 blob_addr_64; 1025 __u8 blob_size_32; 1026 __u8 blob_size_64; 1027 __u8 pad2[30]; 1028 }; 1029 1030If certain flags are returned from the KVM_CAP_XEN_HVM check, they may 1031be set in the flags field of this ioctl: 1032 1033The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag requests KVM to generate 1034the contents of the hypercall page automatically; hypercalls will be 1035intercepted and passed to userspace through KVM_EXIT_XEN. In this 1036case, all of the blob size and address fields must be zero. 1037 1038The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates to KVM that userspace 1039will always use the KVM_XEN_HVM_EVTCHN_SEND ioctl to deliver event 1040channel interrupts rather than manipulating the guest's shared_info 1041structures directly. This, in turn, may allow KVM to enable features 1042such as intercepting the SCHEDOP_poll hypercall to accelerate PV 1043spinlock operation for the guest. Userspace may still use the ioctl 1044to deliver events if it was advertised, even if userspace does not 1045send this indication that it will always do so 1046 1047No other flags are currently valid in the struct kvm_xen_hvm_config. 1048 10494.29 KVM_GET_CLOCK 1050------------------ 1051 1052:Capability: KVM_CAP_ADJUST_CLOCK 1053:Architectures: x86 1054:Type: vm ioctl 1055:Parameters: struct kvm_clock_data (out) 1056:Returns: 0 on success, -1 on error 1057 1058Gets the current timestamp of kvmclock as seen by the current guest. In 1059conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 1060such as migration. 1061 1062When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 1063set of bits that KVM can return in struct kvm_clock_data's flag member. 1064 1065The following flags are defined: 1066 1067KVM_CLOCK_TSC_STABLE 1068 If set, the returned value is the exact kvmclock 1069 value seen by all VCPUs at the instant when KVM_GET_CLOCK was called. 1070 If clear, the returned value is simply CLOCK_MONOTONIC plus a constant 1071 offset; the offset can be modified with KVM_SET_CLOCK. KVM will try 1072 to make all VCPUs follow this clock, but the exact value read by each 1073 VCPU could differ, because the host TSC is not stable. 1074 1075KVM_CLOCK_REALTIME 1076 If set, the `realtime` field in the kvm_clock_data 1077 structure is populated with the value of the host's real time 1078 clocksource at the instant when KVM_GET_CLOCK was called. If clear, 1079 the `realtime` field does not contain a value. 1080 1081KVM_CLOCK_HOST_TSC 1082 If set, the `host_tsc` field in the kvm_clock_data 1083 structure is populated with the value of the host's timestamp counter (TSC) 1084 at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field 1085 does not contain a value. 1086 1087:: 1088 1089 struct kvm_clock_data { 1090 __u64 clock; /* kvmclock current value */ 1091 __u32 flags; 1092 __u32 pad0; 1093 __u64 realtime; 1094 __u64 host_tsc; 1095 __u32 pad[4]; 1096 }; 1097 1098 10994.30 KVM_SET_CLOCK 1100------------------ 1101 1102:Capability: KVM_CAP_ADJUST_CLOCK 1103:Architectures: x86 1104:Type: vm ioctl 1105:Parameters: struct kvm_clock_data (in) 1106:Returns: 0 on success, -1 on error 1107 1108Sets the current timestamp of kvmclock to the value specified in its parameter. 1109In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 1110such as migration. 1111 1112The following flags can be passed: 1113 1114KVM_CLOCK_REALTIME 1115 If set, KVM will compare the value of the `realtime` field 1116 with the value of the host's real time clocksource at the instant when 1117 KVM_SET_CLOCK was called. The difference in elapsed time is added to the final 1118 kvmclock value that will be provided to guests. 1119 1120Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored. 1121 1122:: 1123 1124 struct kvm_clock_data { 1125 __u64 clock; /* kvmclock current value */ 1126 __u32 flags; 1127 __u32 pad0; 1128 __u64 realtime; 1129 __u64 host_tsc; 1130 __u32 pad[4]; 1131 }; 1132 1133 11344.31 KVM_GET_VCPU_EVENTS 1135------------------------ 1136 1137:Capability: KVM_CAP_VCPU_EVENTS 1138:Extended by: KVM_CAP_INTR_SHADOW 1139:Architectures: x86, arm64 1140:Type: vcpu ioctl 1141:Parameters: struct kvm_vcpu_events (out) 1142:Returns: 0 on success, -1 on error 1143 1144X86: 1145^^^^ 1146 1147Gets currently pending exceptions, interrupts, and NMIs as well as related 1148states of the vcpu. 1149 1150:: 1151 1152 struct kvm_vcpu_events { 1153 struct { 1154 __u8 injected; 1155 __u8 nr; 1156 __u8 has_error_code; 1157 __u8 pending; 1158 __u32 error_code; 1159 } exception; 1160 struct { 1161 __u8 injected; 1162 __u8 nr; 1163 __u8 soft; 1164 __u8 shadow; 1165 } interrupt; 1166 struct { 1167 __u8 injected; 1168 __u8 pending; 1169 __u8 masked; 1170 __u8 pad; 1171 } nmi; 1172 __u32 sipi_vector; 1173 __u32 flags; 1174 struct { 1175 __u8 smm; 1176 __u8 pending; 1177 __u8 smm_inside_nmi; 1178 __u8 latched_init; 1179 } smi; 1180 __u8 reserved[27]; 1181 __u8 exception_has_payload; 1182 __u64 exception_payload; 1183 }; 1184 1185The following bits are defined in the flags field: 1186 1187- KVM_VCPUEVENT_VALID_SHADOW may be set to signal that 1188 interrupt.shadow contains a valid state. 1189 1190- KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a 1191 valid state. 1192 1193- KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the 1194 exception_has_payload, exception_payload, and exception.pending 1195 fields contain a valid state. This bit will be set whenever 1196 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1197 1198- KVM_VCPUEVENT_VALID_TRIPLE_FAULT may be set to signal that the 1199 triple_fault_pending field contains a valid state. This bit will 1200 be set whenever KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled. 1201 1202ARM64: 1203^^^^^^ 1204 1205If the guest accesses a device that is being emulated by the host kernel in 1206such a way that a real device would generate a physical SError, KVM may make 1207a virtual SError pending for that VCPU. This system error interrupt remains 1208pending until the guest takes the exception by unmasking PSTATE.A. 1209 1210Running the VCPU may cause it to take a pending SError, or make an access that 1211causes an SError to become pending. The event's description is only valid while 1212the VPCU is not running. 1213 1214This API provides a way to read and write the pending 'event' state that is not 1215visible to the guest. To save, restore or migrate a VCPU the struct representing 1216the state can be read then written using this GET/SET API, along with the other 1217guest-visible registers. It is not possible to 'cancel' an SError that has been 1218made pending. 1219 1220A device being emulated in user-space may also wish to generate an SError. To do 1221this the events structure can be populated by user-space. The current state 1222should be read first, to ensure no existing SError is pending. If an existing 1223SError is pending, the architecture's 'Multiple SError interrupts' rules should 1224be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and 1225Serviceability (RAS) Specification"). 1226 1227SError exceptions always have an ESR value. Some CPUs have the ability to 1228specify what the virtual SError's ESR value should be. These systems will 1229advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will 1230always have a non-zero value when read, and the agent making an SError pending 1231should specify the ISS field in the lower 24 bits of exception.serror_esr. If 1232the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events 1233with exception.has_esr as zero, KVM will choose an ESR. 1234 1235Specifying exception.has_esr on a system that does not support it will return 1236-EINVAL. Setting anything other than the lower 24bits of exception.serror_esr 1237will return -EINVAL. 1238 1239It is not possible to read back a pending external abort (injected via 1240KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered 1241directly to the virtual CPU). 1242 1243Calling this ioctl on a vCPU that hasn't been initialized will return 1244-ENOEXEC. 1245 1246:: 1247 1248 struct kvm_vcpu_events { 1249 struct { 1250 __u8 serror_pending; 1251 __u8 serror_has_esr; 1252 __u8 ext_dabt_pending; 1253 /* Align it to 8 bytes */ 1254 __u8 pad[5]; 1255 __u64 serror_esr; 1256 } exception; 1257 __u32 reserved[12]; 1258 }; 1259 12604.32 KVM_SET_VCPU_EVENTS 1261------------------------ 1262 1263:Capability: KVM_CAP_VCPU_EVENTS 1264:Extended by: KVM_CAP_INTR_SHADOW 1265:Architectures: x86, arm64 1266:Type: vcpu ioctl 1267:Parameters: struct kvm_vcpu_events (in) 1268:Returns: 0 on success, -1 on error 1269 1270X86: 1271^^^^ 1272 1273Set pending exceptions, interrupts, and NMIs as well as related states of the 1274vcpu. 1275 1276See KVM_GET_VCPU_EVENTS for the data structure. 1277 1278Fields that may be modified asynchronously by running VCPUs can be excluded 1279from the update. These fields are nmi.pending, sipi_vector, smi.smm, 1280smi.pending. Keep the corresponding bits in the flags field cleared to 1281suppress overwriting the current in-kernel state. The bits are: 1282 1283=============================== ================================== 1284KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel 1285KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector 1286KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct. 1287=============================== ================================== 1288 1289If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 1290the flags field to signal that interrupt.shadow contains a valid state and 1291shall be written into the VCPU. 1292 1293KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 1294 1295If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD 1296can be set in the flags field to signal that the 1297exception_has_payload, exception_payload, and exception.pending fields 1298contain a valid state and shall be written into the VCPU. 1299 1300If KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled, KVM_VCPUEVENT_VALID_TRIPLE_FAULT 1301can be set in flags field to signal that the triple_fault field contains 1302a valid state and shall be written into the VCPU. 1303 1304ARM64: 1305^^^^^^ 1306 1307User space may need to inject several types of events to the guest. 1308 1309Set the pending SError exception state for this VCPU. It is not possible to 1310'cancel' an Serror that has been made pending. 1311 1312If the guest performed an access to I/O memory which could not be handled by 1313userspace, for example because of missing instruction syndrome decode 1314information or because there is no device mapped at the accessed IPA, then 1315userspace can ask the kernel to inject an external abort using the address 1316from the exiting fault on the VCPU. It is a programming error to set 1317ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO, 1318KVM_EXIT_ARM_NISV, or KVM_EXIT_ARM_LDST64B. This feature is only available if 1319the system supports KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which 1320provides commonality in how userspace reports accesses for the above cases to 1321guests, across different userspace implementations. Nevertheless, userspace 1322can still emulate all Arm exceptions by manipulating individual registers 1323using the KVM_SET_ONE_REG API. 1324 1325See KVM_GET_VCPU_EVENTS for the data structure. 1326 1327Calling this ioctl on a vCPU that hasn't been initialized will return 1328-ENOEXEC. 1329 13304.33 KVM_GET_DEBUGREGS 1331---------------------- 1332 1333:Capability: KVM_CAP_DEBUGREGS 1334:Architectures: x86 1335:Type: vcpu ioctl 1336:Parameters: struct kvm_debugregs (out) 1337:Returns: 0 on success, -1 on error 1338 1339Reads debug registers from the vcpu. 1340 1341:: 1342 1343 struct kvm_debugregs { 1344 __u64 db[4]; 1345 __u64 dr6; 1346 __u64 dr7; 1347 __u64 flags; 1348 __u64 reserved[9]; 1349 }; 1350 1351 13524.34 KVM_SET_DEBUGREGS 1353---------------------- 1354 1355:Capability: KVM_CAP_DEBUGREGS 1356:Architectures: x86 1357:Type: vcpu ioctl 1358:Parameters: struct kvm_debugregs (in) 1359:Returns: 0 on success, -1 on error 1360 1361Writes debug registers into the vcpu. 1362 1363See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 1364yet and must be cleared on entry. 1365 1366 13674.35 KVM_SET_USER_MEMORY_REGION 1368------------------------------- 1369 1370:Capability: KVM_CAP_USER_MEMORY 1371:Architectures: all 1372:Type: vm ioctl 1373:Parameters: struct kvm_userspace_memory_region (in) 1374:Returns: 0 on success, -1 on error 1375 1376:: 1377 1378 struct kvm_userspace_memory_region { 1379 __u32 slot; 1380 __u32 flags; 1381 __u64 guest_phys_addr; 1382 __u64 memory_size; /* bytes */ 1383 __u64 userspace_addr; /* start of the userspace allocated memory */ 1384 }; 1385 1386 /* for kvm_userspace_memory_region::flags */ 1387 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 1388 #define KVM_MEM_READONLY (1UL << 1) 1389 1390This ioctl allows the user to create, modify or delete a guest physical 1391memory slot. Bits 0-15 of "slot" specify the slot id and this value 1392should be less than the maximum number of user memory slots supported per 1393VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS. 1394Slots may not overlap in guest physical address space. 1395 1396If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 1397specifies the address space which is being modified. They must be 1398less than the value that KVM_CHECK_EXTENSION returns for the 1399KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 1400are unrelated; the restriction on overlapping slots only applies within 1401each address space. 1402 1403Deleting a slot is done by passing zero for memory_size. When changing 1404an existing slot, it may be moved in the guest physical memory space, 1405or its flags may be modified, but it may not be resized. 1406 1407Memory for the region is taken starting at the address denoted by the 1408field userspace_addr, which must point at user addressable memory for 1409the entire memory slot size. Any object may back this memory, including 1410anonymous memory, ordinary files, and hugetlbfs. Changes in the backing 1411of the memory region are automatically reflected into the guest. 1412For example, an mmap() that affects the region will be made visible 1413immediately. Another example is madvise(MADV_DROP). 1414 1415On architectures that support a form of address tagging, userspace_addr must 1416be an untagged address. 1417 1418It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 1419be identical. This allows large pages in the guest to be backed by large 1420pages in the host. 1421 1422The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 1423KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 1424writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 1425use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 1426to make a new slot read-only. In this case, writes to this memory will be 1427posted to userspace as KVM_EXIT_MMIO exits. 1428 1429For TDX guest, deleting/moving memory region loses guest memory contents. 1430Read only region isn't supported. Only as-id 0 is supported. 1431 1432Note: On arm64, a write generated by the page-table walker (to update 1433the Access and Dirty flags, for example) never results in a 1434KVM_EXIT_MMIO exit when the slot has the KVM_MEM_READONLY flag. This 1435is because KVM cannot provide the data that would be written by the 1436page-table walker, making it impossible to emulate the access. 1437Instead, an abort (data abort if the cause of the page-table update 1438was a load or a store, instruction abort if it was an instruction 1439fetch) is injected in the guest. 1440 1441S390: 1442^^^^^ 1443 1444Returns -EINVAL or -EEXIST if the VM has the KVM_VM_S390_UCONTROL flag set. 1445Returns -EINVAL if called on a protected VM. 1446 14474.36 KVM_SET_TSS_ADDR 1448--------------------- 1449 1450:Capability: KVM_CAP_SET_TSS_ADDR 1451:Architectures: x86 1452:Type: vm ioctl 1453:Parameters: unsigned long tss_address (in) 1454:Returns: 0 on success, -1 on error 1455 1456This ioctl defines the physical address of a three-page region in the guest 1457physical address space. The region must be within the first 4GB of the 1458guest physical address space and must not conflict with any memory slot 1459or any mmio address. The guest may malfunction if it accesses this memory 1460region. 1461 1462This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1463because of a quirk in the virtualization implementation (see the internals 1464documentation when it pops into existence). 1465 1466 1467.. _KVM_ENABLE_CAP: 1468 14694.37 KVM_ENABLE_CAP 1470------------------- 1471 1472:Capability: KVM_CAP_ENABLE_CAP 1473:Architectures: mips, ppc, s390, x86, loongarch 1474:Type: vcpu ioctl 1475:Parameters: struct kvm_enable_cap (in) 1476:Returns: 0 on success; -1 on error 1477 1478:Capability: KVM_CAP_ENABLE_CAP_VM 1479:Architectures: all 1480:Type: vm ioctl 1481:Parameters: struct kvm_enable_cap (in) 1482:Returns: 0 on success; -1 on error 1483 1484.. note:: 1485 1486 Not all extensions are enabled by default. Using this ioctl the application 1487 can enable an extension, making it available to the guest. 1488 1489On systems that do not support this ioctl, it always fails. On systems that 1490do support it, it only works for extensions that are supported for enablement. 1491 1492To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1493be used. 1494 1495:: 1496 1497 struct kvm_enable_cap { 1498 /* in */ 1499 __u32 cap; 1500 1501The capability that is supposed to get enabled. 1502 1503:: 1504 1505 __u32 flags; 1506 1507A bitfield indicating future enhancements. Has to be 0 for now. 1508 1509:: 1510 1511 __u64 args[4]; 1512 1513Arguments for enabling a feature. If a feature needs initial values to 1514function properly, this is the place to put them. 1515 1516:: 1517 1518 __u8 pad[64]; 1519 }; 1520 1521The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1522for vm-wide capabilities. 1523 15244.38 KVM_GET_MP_STATE 1525--------------------- 1526 1527:Capability: KVM_CAP_MP_STATE 1528:Architectures: x86, s390, arm64, riscv, loongarch 1529:Type: vcpu ioctl 1530:Parameters: struct kvm_mp_state (out) 1531:Returns: 0 on success; -1 on error 1532 1533:: 1534 1535 struct kvm_mp_state { 1536 __u32 mp_state; 1537 }; 1538 1539Returns the vcpu's current "multiprocessing state" (though also valid on 1540uniprocessor guests). 1541 1542Possible values are: 1543 1544 ========================== =============================================== 1545 KVM_MP_STATE_RUNNABLE the vcpu is currently running 1546 [x86,arm64,riscv,loongarch] 1547 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP) 1548 which has not yet received an INIT signal [x86] 1549 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is 1550 now ready for a SIPI [x86] 1551 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and 1552 is waiting for an interrupt [x86] 1553 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector 1554 accessible via KVM_GET_VCPU_EVENTS) [x86] 1555 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm64,riscv] 1556 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390] 1557 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted) 1558 [s390] 1559 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state 1560 [s390] 1561 KVM_MP_STATE_SUSPENDED the vcpu is in a suspend state and is waiting 1562 for a wakeup event [arm64] 1563 ========================== =============================================== 1564 1565On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1566in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1567these architectures. 1568 1569For arm64: 1570^^^^^^^^^^ 1571 1572If a vCPU is in the KVM_MP_STATE_SUSPENDED state, KVM will emulate the 1573architectural execution of a WFI instruction. 1574 1575If a wakeup event is recognized, KVM will exit to userspace with a 1576KVM_SYSTEM_EVENT exit, where the event type is KVM_SYSTEM_EVENT_WAKEUP. If 1577userspace wants to honor the wakeup, it must set the vCPU's MP state to 1578KVM_MP_STATE_RUNNABLE. If it does not, KVM will continue to await a wakeup 1579event in subsequent calls to KVM_RUN. 1580 1581.. warning:: 1582 1583 If userspace intends to keep the vCPU in a SUSPENDED state, it is 1584 strongly recommended that userspace take action to suppress the 1585 wakeup event (such as masking an interrupt). Otherwise, subsequent 1586 calls to KVM_RUN will immediately exit with a KVM_SYSTEM_EVENT_WAKEUP 1587 event and inadvertently waste CPU cycles. 1588 1589 Additionally, if userspace takes action to suppress a wakeup event, 1590 it is strongly recommended that it also restores the vCPU to its 1591 original state when the vCPU is made RUNNABLE again. For example, 1592 if userspace masked a pending interrupt to suppress the wakeup, 1593 the interrupt should be unmasked before returning control to the 1594 guest. 1595 1596For riscv: 1597^^^^^^^^^^ 1598 1599The only states that are valid are KVM_MP_STATE_STOPPED and 1600KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1601 1602On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect 1603whether the vcpu is runnable. 1604 16054.39 KVM_SET_MP_STATE 1606--------------------- 1607 1608:Capability: KVM_CAP_MP_STATE 1609:Architectures: x86, s390, arm64, riscv, loongarch 1610:Type: vcpu ioctl 1611:Parameters: struct kvm_mp_state (in) 1612:Returns: 0 on success; -1 on error 1613 1614Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1615arguments. 1616 1617On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1618in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1619these architectures. 1620 1621For arm64/riscv: 1622^^^^^^^^^^^^^^^^ 1623 1624The only states that are valid are KVM_MP_STATE_STOPPED and 1625KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1626 1627On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect 1628whether the vcpu is runnable. 1629 16304.40 KVM_SET_IDENTITY_MAP_ADDR 1631------------------------------ 1632 1633:Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1634:Architectures: x86 1635:Type: vm ioctl 1636:Parameters: unsigned long identity (in) 1637:Returns: 0 on success, -1 on error 1638 1639This ioctl defines the physical address of a one-page region in the guest 1640physical address space. The region must be within the first 4GB of the 1641guest physical address space and must not conflict with any memory slot 1642or any mmio address. The guest may malfunction if it accesses this memory 1643region. 1644 1645Setting the address to 0 will result in resetting the address to its default 1646(0xfffbc000). 1647 1648This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1649because of a quirk in the virtualization implementation (see the internals 1650documentation when it pops into existence). 1651 1652Fails if any VCPU has already been created. 1653 16544.41 KVM_SET_BOOT_CPU_ID 1655------------------------ 1656 1657:Capability: KVM_CAP_SET_BOOT_CPU_ID 1658:Architectures: x86 1659:Type: vm ioctl 1660:Parameters: unsigned long vcpu_id 1661:Returns: 0 on success, -1 on error 1662 1663Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1664as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1665is vcpu 0. This ioctl has to be called before vcpu creation, 1666otherwise it will return EBUSY error. 1667 1668 16694.42 KVM_GET_XSAVE 1670------------------ 1671 1672:Capability: KVM_CAP_XSAVE 1673:Architectures: x86 1674:Type: vcpu ioctl 1675:Parameters: struct kvm_xsave (out) 1676:Returns: 0 on success, -1 on error 1677 1678 1679:: 1680 1681 struct kvm_xsave { 1682 __u32 region[1024]; 1683 __u32 extra[0]; 1684 }; 1685 1686This ioctl would copy current vcpu's xsave struct to the userspace. 1687 1688 16894.43 KVM_SET_XSAVE 1690------------------ 1691 1692:Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2 1693:Architectures: x86 1694:Type: vcpu ioctl 1695:Parameters: struct kvm_xsave (in) 1696:Returns: 0 on success, -1 on error 1697 1698:: 1699 1700 1701 struct kvm_xsave { 1702 __u32 region[1024]; 1703 __u32 extra[0]; 1704 }; 1705 1706This ioctl would copy userspace's xsave struct to the kernel. It copies 1707as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2), 1708when invoked on the vm file descriptor. The size value returned by 1709KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 1710Currently, it is only greater than 4096 if a dynamic feature has been 1711enabled with ``arch_prctl()``, but this may change in the future. 1712 1713The offsets of the state save areas in struct kvm_xsave follow the 1714contents of CPUID leaf 0xD on the host. 1715 1716 17174.44 KVM_GET_XCRS 1718----------------- 1719 1720:Capability: KVM_CAP_XCRS 1721:Architectures: x86 1722:Type: vcpu ioctl 1723:Parameters: struct kvm_xcrs (out) 1724:Returns: 0 on success, -1 on error 1725 1726:: 1727 1728 struct kvm_xcr { 1729 __u32 xcr; 1730 __u32 reserved; 1731 __u64 value; 1732 }; 1733 1734 struct kvm_xcrs { 1735 __u32 nr_xcrs; 1736 __u32 flags; 1737 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1738 __u64 padding[16]; 1739 }; 1740 1741This ioctl would copy current vcpu's xcrs to the userspace. 1742 1743 17444.45 KVM_SET_XCRS 1745----------------- 1746 1747:Capability: KVM_CAP_XCRS 1748:Architectures: x86 1749:Type: vcpu ioctl 1750:Parameters: struct kvm_xcrs (in) 1751:Returns: 0 on success, -1 on error 1752 1753:: 1754 1755 struct kvm_xcr { 1756 __u32 xcr; 1757 __u32 reserved; 1758 __u64 value; 1759 }; 1760 1761 struct kvm_xcrs { 1762 __u32 nr_xcrs; 1763 __u32 flags; 1764 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1765 __u64 padding[16]; 1766 }; 1767 1768This ioctl would set vcpu's xcr to the value userspace specified. 1769 1770 17714.46 KVM_GET_SUPPORTED_CPUID 1772---------------------------- 1773 1774:Capability: KVM_CAP_EXT_CPUID 1775:Architectures: x86 1776:Type: system ioctl 1777:Parameters: struct kvm_cpuid2 (in/out) 1778:Returns: 0 on success, -1 on error 1779 1780:: 1781 1782 struct kvm_cpuid2 { 1783 __u32 nent; 1784 __u32 padding; 1785 struct kvm_cpuid_entry2 entries[0]; 1786 }; 1787 1788 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1789 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 1790 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 1791 1792 struct kvm_cpuid_entry2 { 1793 __u32 function; 1794 __u32 index; 1795 __u32 flags; 1796 __u32 eax; 1797 __u32 ebx; 1798 __u32 ecx; 1799 __u32 edx; 1800 __u32 padding[3]; 1801 }; 1802 1803This ioctl returns x86 cpuid features which are supported by both the 1804hardware and kvm in its default configuration. Userspace can use the 1805information returned by this ioctl to construct cpuid information (for 1806KVM_SET_CPUID2) that is consistent with hardware, kernel, and 1807userspace capabilities, and with user requirements (for example, the 1808user may wish to constrain cpuid to emulate older hardware, or for 1809feature consistency across a cluster). 1810 1811Dynamically-enabled feature bits need to be requested with 1812``arch_prctl()`` before calling this ioctl. Feature bits that have not 1813been requested are excluded from the result. 1814 1815Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may 1816expose cpuid features (e.g. MONITOR) which are not supported by kvm in 1817its default configuration. If userspace enables such capabilities, it 1818is responsible for modifying the results of this ioctl appropriately. 1819 1820Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1821with the 'nent' field indicating the number of entries in the variable-size 1822array 'entries'. If the number of entries is too low to describe the cpu 1823capabilities, an error (E2BIG) is returned. If the number is too high, 1824the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1825number is just right, the 'nent' field is adjusted to the number of valid 1826entries in the 'entries' array, which is then filled. 1827 1828The entries returned are the host cpuid as returned by the cpuid instruction, 1829with unknown or unsupported features masked out. Some features (for example, 1830x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1831emulate them efficiently. The fields in each entry are defined as follows: 1832 1833 function: 1834 the eax value used to obtain the entry 1835 1836 index: 1837 the ecx value used to obtain the entry (for entries that are 1838 affected by ecx) 1839 1840 flags: 1841 an OR of zero or more of the following: 1842 1843 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1844 if the index field is valid 1845 1846 eax, ebx, ecx, edx: 1847 the values returned by the cpuid instruction for 1848 this function/index combination 1849 1850x2APIC (CPUID leaf 1, ecx[21) and TSC deadline timer (CPUID leaf 1, ecx[24]) 1851may be returned as true, but they depend on KVM_CREATE_IRQCHIP for in-kernel 1852emulation of the local APIC. TSC deadline timer support is also reported via:: 1853 1854 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1855 1856if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1857feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1858 1859Enabling x2APIC in KVM_SET_CPUID2 requires KVM_CREATE_IRQCHIP as KVM doesn't 1860support forwarding x2APIC MSR accesses to userspace, i.e. KVM does not support 1861emulating x2APIC in userspace. 1862 18634.47 KVM_PPC_GET_PVINFO 1864----------------------- 1865 1866:Capability: KVM_CAP_PPC_GET_PVINFO 1867:Architectures: ppc 1868:Type: vm ioctl 1869:Parameters: struct kvm_ppc_pvinfo (out) 1870:Returns: 0 on success, !0 on error 1871 1872:: 1873 1874 struct kvm_ppc_pvinfo { 1875 __u32 flags; 1876 __u32 hcall[4]; 1877 __u8 pad[108]; 1878 }; 1879 1880This ioctl fetches PV specific information that need to be passed to the guest 1881using the device tree or other means from vm context. 1882 1883The hcall array defines 4 instructions that make up a hypercall. 1884 1885If any additional field gets added to this structure later on, a bit for that 1886additional piece of information will be set in the flags bitmap. 1887 1888The flags bitmap is defined as:: 1889 1890 /* the host supports the ePAPR idle hcall 1891 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1892 18934.52 KVM_SET_GSI_ROUTING 1894------------------------ 1895 1896:Capability: KVM_CAP_IRQ_ROUTING 1897:Architectures: x86 s390 arm64 1898:Type: vm ioctl 1899:Parameters: struct kvm_irq_routing (in) 1900:Returns: 0 on success, -1 on error 1901 1902Sets the GSI routing table entries, overwriting any previously set entries. 1903 1904On arm64, GSI routing has the following limitation: 1905 1906- GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1907 1908:: 1909 1910 struct kvm_irq_routing { 1911 __u32 nr; 1912 __u32 flags; 1913 struct kvm_irq_routing_entry entries[0]; 1914 }; 1915 1916No flags are specified so far, the corresponding field must be set to zero. 1917 1918:: 1919 1920 struct kvm_irq_routing_entry { 1921 __u32 gsi; 1922 __u32 type; 1923 __u32 flags; 1924 __u32 pad; 1925 union { 1926 struct kvm_irq_routing_irqchip irqchip; 1927 struct kvm_irq_routing_msi msi; 1928 struct kvm_irq_routing_s390_adapter adapter; 1929 struct kvm_irq_routing_hv_sint hv_sint; 1930 struct kvm_irq_routing_xen_evtchn xen_evtchn; 1931 __u32 pad[8]; 1932 } u; 1933 }; 1934 1935 /* gsi routing entry types */ 1936 #define KVM_IRQ_ROUTING_IRQCHIP 1 1937 #define KVM_IRQ_ROUTING_MSI 2 1938 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1939 #define KVM_IRQ_ROUTING_HV_SINT 4 1940 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 1941 1942On s390, adding a KVM_IRQ_ROUTING_S390_ADAPTER is rejected on ucontrol VMs with 1943error -EINVAL. 1944 1945flags: 1946 1947- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1948 type, specifies that the devid field contains a valid value. The per-VM 1949 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1950 the device ID. If this capability is not available, userspace should 1951 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1952- zero otherwise 1953 1954:: 1955 1956 struct kvm_irq_routing_irqchip { 1957 __u32 irqchip; 1958 __u32 pin; 1959 }; 1960 1961 struct kvm_irq_routing_msi { 1962 __u32 address_lo; 1963 __u32 address_hi; 1964 __u32 data; 1965 union { 1966 __u32 pad; 1967 __u32 devid; 1968 }; 1969 }; 1970 1971If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1972for the device that wrote the MSI message. For PCI, this is usually a 1973BDF identifier in the lower 16 bits. 1974 1975On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1976feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1977address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1978address_hi must be zero. 1979 1980:: 1981 1982 struct kvm_irq_routing_s390_adapter { 1983 __u64 ind_addr; 1984 __u64 summary_addr; 1985 __u64 ind_offset; 1986 __u32 summary_offset; 1987 __u32 adapter_id; 1988 }; 1989 1990 struct kvm_irq_routing_hv_sint { 1991 __u32 vcpu; 1992 __u32 sint; 1993 }; 1994 1995 struct kvm_irq_routing_xen_evtchn { 1996 __u32 port; 1997 __u32 vcpu; 1998 __u32 priority; 1999 }; 2000 2001 2002When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL bit 2003in its indication of supported features, routing to Xen event channels 2004is supported. Although the priority field is present, only the value 2005KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported, which means delivery by 20062 level event channels. FIFO event channel support may be added in 2007the future. 2008 2009 20104.55 KVM_SET_TSC_KHZ 2011-------------------- 2012 2013:Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM_TSC_CONTROL 2014:Architectures: x86 2015:Type: vcpu ioctl / vm ioctl 2016:Parameters: virtual tsc_khz 2017:Returns: 0 on success, -1 on error 2018 2019Specifies the tsc frequency for the virtual machine. The unit of the 2020frequency is KHz. 2021 2022If the KVM_CAP_VM_TSC_CONTROL capability is advertised, this can also 2023be used as a vm ioctl to set the initial tsc frequency of subsequently 2024created vCPUs. Note, the vm ioctl is only allowed prior to creating vCPUs. 2025 2026For TSC protected Confidential Computing (CoCo) VMs where TSC frequency 2027is configured once at VM scope and remains unchanged during VM's 2028lifetime, the vm ioctl should be used to configure the TSC frequency 2029and the vcpu ioctl is not supported. 2030 2031Example of such CoCo VMs: TDX guests. 2032 20334.56 KVM_GET_TSC_KHZ 2034-------------------- 2035 2036:Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM_TSC_CONTROL 2037:Architectures: x86 2038:Type: vcpu ioctl / vm ioctl 2039:Parameters: none 2040:Returns: virtual tsc-khz on success, negative value on error 2041 2042Returns the tsc frequency of the guest. The unit of the return value is 2043KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 2044error. 2045 2046 20474.57 KVM_GET_LAPIC 2048------------------ 2049 2050:Capability: KVM_CAP_IRQCHIP 2051:Architectures: x86 2052:Type: vcpu ioctl 2053:Parameters: struct kvm_lapic_state (out) 2054:Returns: 0 on success, -1 on error 2055 2056:: 2057 2058 #define KVM_APIC_REG_SIZE 0x400 2059 struct kvm_lapic_state { 2060 char regs[KVM_APIC_REG_SIZE]; 2061 }; 2062 2063Reads the Local APIC registers and copies them into the input argument. The 2064data format and layout are the same as documented in the architecture manual. 2065 2066If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 2067enabled, then the format of APIC_ID register depends on the APIC mode 2068(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 2069the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 2070which is stored in bits 31-24 of the APIC register, or equivalently in 2071byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 2072be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 2073 2074If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 2075always uses xAPIC format. 2076 2077 20784.58 KVM_SET_LAPIC 2079------------------ 2080 2081:Capability: KVM_CAP_IRQCHIP 2082:Architectures: x86 2083:Type: vcpu ioctl 2084:Parameters: struct kvm_lapic_state (in) 2085:Returns: 0 on success, -1 on error 2086 2087:: 2088 2089 #define KVM_APIC_REG_SIZE 0x400 2090 struct kvm_lapic_state { 2091 char regs[KVM_APIC_REG_SIZE]; 2092 }; 2093 2094Copies the input argument into the Local APIC registers. The data format 2095and layout are the same as documented in the architecture manual. 2096 2097The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 2098regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 2099See the note in KVM_GET_LAPIC. 2100 2101 21024.59 KVM_IOEVENTFD 2103------------------ 2104 2105:Capability: KVM_CAP_IOEVENTFD 2106:Architectures: all 2107:Type: vm ioctl 2108:Parameters: struct kvm_ioeventfd (in) 2109:Returns: 0 on success, !0 on error 2110 2111This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 2112within the guest. A guest write in the registered address will signal the 2113provided event instead of triggering an exit. 2114 2115:: 2116 2117 struct kvm_ioeventfd { 2118 __u64 datamatch; 2119 __u64 addr; /* legal pio/mmio address */ 2120 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 2121 __s32 fd; 2122 __u32 flags; 2123 __u8 pad[36]; 2124 }; 2125 2126For the special case of virtio-ccw devices on s390, the ioevent is matched 2127to a subchannel/virtqueue tuple instead. 2128 2129The following flags are defined:: 2130 2131 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 2132 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 2133 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 2134 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 2135 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 2136 2137If datamatch flag is set, the event will be signaled only if the written value 2138to the registered address is equal to datamatch in struct kvm_ioeventfd. 2139 2140For virtio-ccw devices, addr contains the subchannel id and datamatch the 2141virtqueue index. 2142 2143With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 2144the kernel will ignore the length of guest write and may get a faster vmexit. 2145The speedup may only apply to specific architectures, but the ioeventfd will 2146work anyway. 2147 21484.60 KVM_DIRTY_TLB 2149------------------ 2150 2151:Capability: KVM_CAP_SW_TLB 2152:Architectures: ppc 2153:Type: vcpu ioctl 2154:Parameters: struct kvm_dirty_tlb (in) 2155:Returns: 0 on success, -1 on error 2156 2157:: 2158 2159 struct kvm_dirty_tlb { 2160 __u64 bitmap; 2161 __u32 num_dirty; 2162 }; 2163 2164This must be called whenever userspace has changed an entry in the shared 2165TLB, prior to calling KVM_RUN on the associated vcpu. 2166 2167The "bitmap" field is the userspace address of an array. This array 2168consists of a number of bits, equal to the total number of TLB entries as 2169determined by the last successful call to ``KVM_ENABLE_CAP(KVM_CAP_SW_TLB)``, 2170rounded up to the nearest multiple of 64. 2171 2172Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 2173array. 2174 2175The array is little-endian: the bit 0 is the least significant bit of the 2176first byte, bit 8 is the least significant bit of the second byte, etc. 2177This avoids any complications with differing word sizes. 2178 2179The "num_dirty" field is a performance hint for KVM to determine whether it 2180should skip processing the bitmap and just invalidate everything. It must 2181be set to the number of set bits in the bitmap. 2182 2183 21844.62 KVM_CREATE_SPAPR_TCE 2185------------------------- 2186 2187:Capability: KVM_CAP_SPAPR_TCE 2188:Architectures: powerpc 2189:Type: vm ioctl 2190:Parameters: struct kvm_create_spapr_tce (in) 2191:Returns: file descriptor for manipulating the created TCE table 2192 2193This creates a virtual TCE (translation control entry) table, which 2194is an IOMMU for PAPR-style virtual I/O. It is used to translate 2195logical addresses used in virtual I/O into guest physical addresses, 2196and provides a scatter/gather capability for PAPR virtual I/O. 2197 2198:: 2199 2200 /* for KVM_CAP_SPAPR_TCE */ 2201 struct kvm_create_spapr_tce { 2202 __u64 liobn; 2203 __u32 window_size; 2204 }; 2205 2206The liobn field gives the logical IO bus number for which to create a 2207TCE table. The window_size field specifies the size of the DMA window 2208which this TCE table will translate - the table will contain one 64 2209bit TCE entry for every 4kiB of the DMA window. 2210 2211When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 2212table has been created using this ioctl(), the kernel will handle it 2213in real mode, updating the TCE table. H_PUT_TCE calls for other 2214liobns will cause a vm exit and must be handled by userspace. 2215 2216The return value is a file descriptor which can be passed to mmap(2) 2217to map the created TCE table into userspace. This lets userspace read 2218the entries written by kernel-handled H_PUT_TCE calls, and also lets 2219userspace update the TCE table directly which is useful in some 2220circumstances. 2221 2222 22234.64 KVM_NMI 2224------------ 2225 2226:Capability: KVM_CAP_USER_NMI 2227:Architectures: x86 2228:Type: vcpu ioctl 2229:Parameters: none 2230:Returns: 0 on success, -1 on error 2231 2232Queues an NMI on the thread's vcpu. Note this is well defined only 2233when KVM_CREATE_IRQCHIP has not been called, since this is an interface 2234between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 2235has been called, this interface is completely emulated within the kernel. 2236 2237To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 2238following algorithm: 2239 2240 - pause the vcpu 2241 - read the local APIC's state (KVM_GET_LAPIC) 2242 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 2243 - if so, issue KVM_NMI 2244 - resume the vcpu 2245 2246Some guests configure the LINT1 NMI input to cause a panic, aiding in 2247debugging. 2248 2249 22504.65 KVM_S390_UCAS_MAP 2251---------------------- 2252 2253:Capability: KVM_CAP_S390_UCONTROL 2254:Architectures: s390 2255:Type: vcpu ioctl 2256:Parameters: struct kvm_s390_ucas_mapping (in) 2257:Returns: 0 in case of success 2258 2259The parameter is defined like this:: 2260 2261 struct kvm_s390_ucas_mapping { 2262 __u64 user_addr; 2263 __u64 vcpu_addr; 2264 __u64 length; 2265 }; 2266 2267This ioctl maps the memory at "user_addr" with the length "length" to 2268the vcpu's address space starting at "vcpu_addr". All parameters need to 2269be aligned by 1 megabyte. 2270 2271 22724.66 KVM_S390_UCAS_UNMAP 2273------------------------ 2274 2275:Capability: KVM_CAP_S390_UCONTROL 2276:Architectures: s390 2277:Type: vcpu ioctl 2278:Parameters: struct kvm_s390_ucas_mapping (in) 2279:Returns: 0 in case of success 2280 2281The parameter is defined like this:: 2282 2283 struct kvm_s390_ucas_mapping { 2284 __u64 user_addr; 2285 __u64 vcpu_addr; 2286 __u64 length; 2287 }; 2288 2289This ioctl unmaps the memory in the vcpu's address space starting at 2290"vcpu_addr" with the length "length". The field "user_addr" is ignored. 2291All parameters need to be aligned by 1 megabyte. 2292 2293 22944.67 KVM_S390_VCPU_FAULT 2295------------------------ 2296 2297:Capability: KVM_CAP_S390_UCONTROL 2298:Architectures: s390 2299:Type: vcpu ioctl 2300:Parameters: vcpu absolute address (in) 2301:Returns: 0 in case of success 2302 2303This call creates a page table entry on the virtual cpu's address space 2304(for user controlled virtual machines) or the virtual machine's address 2305space (for regular virtual machines). This only works for minor faults, 2306thus it's recommended to access subject memory page via the user page 2307table upfront. This is useful to handle validity intercepts for user 2308controlled virtual machines to fault in the virtual cpu's lowcore pages 2309prior to calling the KVM_RUN ioctl. 2310 2311 23124.68 KVM_SET_ONE_REG 2313-------------------- 2314 2315:Capability: KVM_CAP_ONE_REG 2316:Architectures: all 2317:Type: vcpu ioctl 2318:Parameters: struct kvm_one_reg (in) 2319:Returns: 0 on success, negative value on failure 2320 2321Errors: 2322 2323 ====== ============================================================ 2324 ENOENT no such register 2325 EINVAL invalid register ID, or no such register or used with VMs in 2326 protected virtualization mode on s390 2327 EPERM (arm64) register access not allowed before vcpu finalization 2328 EBUSY (riscv) changing register value not allowed after the vcpu 2329 has run at least once 2330 ====== ============================================================ 2331 2332(These error codes are indicative only: do not rely on a specific error 2333code being returned in a specific situation.) 2334 2335:: 2336 2337 struct kvm_one_reg { 2338 __u64 id; 2339 __u64 addr; 2340 }; 2341 2342Using this ioctl, a single vcpu register can be set to a specific value 2343defined by user space with the passed in struct kvm_one_reg, where id 2344refers to the register identifier as described below and addr is a pointer 2345to a variable with the respective size. There can be architecture agnostic 2346and architecture specific registers. Each have their own range of operation 2347and their own constants and width. To keep track of the implemented 2348registers, find a list below: 2349 2350 ======= =============================== ============ 2351 Arch Register Width (bits) 2352 ======= =============================== ============ 2353 PPC KVM_REG_PPC_HIOR 64 2354 PPC KVM_REG_PPC_IAC1 64 2355 PPC KVM_REG_PPC_IAC2 64 2356 PPC KVM_REG_PPC_IAC3 64 2357 PPC KVM_REG_PPC_IAC4 64 2358 PPC KVM_REG_PPC_DAC1 64 2359 PPC KVM_REG_PPC_DAC2 64 2360 PPC KVM_REG_PPC_DABR 64 2361 PPC KVM_REG_PPC_DSCR 64 2362 PPC KVM_REG_PPC_PURR 64 2363 PPC KVM_REG_PPC_SPURR 64 2364 PPC KVM_REG_PPC_DAR 64 2365 PPC KVM_REG_PPC_DSISR 32 2366 PPC KVM_REG_PPC_AMR 64 2367 PPC KVM_REG_PPC_UAMOR 64 2368 PPC KVM_REG_PPC_MMCR0 64 2369 PPC KVM_REG_PPC_MMCR1 64 2370 PPC KVM_REG_PPC_MMCRA 64 2371 PPC KVM_REG_PPC_MMCR2 64 2372 PPC KVM_REG_PPC_MMCRS 64 2373 PPC KVM_REG_PPC_MMCR3 64 2374 PPC KVM_REG_PPC_SIAR 64 2375 PPC KVM_REG_PPC_SDAR 64 2376 PPC KVM_REG_PPC_SIER 64 2377 PPC KVM_REG_PPC_SIER2 64 2378 PPC KVM_REG_PPC_SIER3 64 2379 PPC KVM_REG_PPC_PMC1 32 2380 PPC KVM_REG_PPC_PMC2 32 2381 PPC KVM_REG_PPC_PMC3 32 2382 PPC KVM_REG_PPC_PMC4 32 2383 PPC KVM_REG_PPC_PMC5 32 2384 PPC KVM_REG_PPC_PMC6 32 2385 PPC KVM_REG_PPC_PMC7 32 2386 PPC KVM_REG_PPC_PMC8 32 2387 PPC KVM_REG_PPC_FPR0 64 2388 ... 2389 PPC KVM_REG_PPC_FPR31 64 2390 PPC KVM_REG_PPC_VR0 128 2391 ... 2392 PPC KVM_REG_PPC_VR31 128 2393 PPC KVM_REG_PPC_VSR0 128 2394 ... 2395 PPC KVM_REG_PPC_VSR31 128 2396 PPC KVM_REG_PPC_FPSCR 64 2397 PPC KVM_REG_PPC_VSCR 32 2398 PPC KVM_REG_PPC_VPA_ADDR 64 2399 PPC KVM_REG_PPC_VPA_SLB 128 2400 PPC KVM_REG_PPC_VPA_DTL 128 2401 PPC KVM_REG_PPC_EPCR 32 2402 PPC KVM_REG_PPC_EPR 32 2403 PPC KVM_REG_PPC_TCR 32 2404 PPC KVM_REG_PPC_TSR 32 2405 PPC KVM_REG_PPC_OR_TSR 32 2406 PPC KVM_REG_PPC_CLEAR_TSR 32 2407 PPC KVM_REG_PPC_MAS0 32 2408 PPC KVM_REG_PPC_MAS1 32 2409 PPC KVM_REG_PPC_MAS2 64 2410 PPC KVM_REG_PPC_MAS7_3 64 2411 PPC KVM_REG_PPC_MAS4 32 2412 PPC KVM_REG_PPC_MAS6 32 2413 PPC KVM_REG_PPC_MMUCFG 32 2414 PPC KVM_REG_PPC_TLB0CFG 32 2415 PPC KVM_REG_PPC_TLB1CFG 32 2416 PPC KVM_REG_PPC_TLB2CFG 32 2417 PPC KVM_REG_PPC_TLB3CFG 32 2418 PPC KVM_REG_PPC_TLB0PS 32 2419 PPC KVM_REG_PPC_TLB1PS 32 2420 PPC KVM_REG_PPC_TLB2PS 32 2421 PPC KVM_REG_PPC_TLB3PS 32 2422 PPC KVM_REG_PPC_EPTCFG 32 2423 PPC KVM_REG_PPC_ICP_STATE 64 2424 PPC KVM_REG_PPC_VP_STATE 128 2425 PPC KVM_REG_PPC_TB_OFFSET 64 2426 PPC KVM_REG_PPC_SPMC1 32 2427 PPC KVM_REG_PPC_SPMC2 32 2428 PPC KVM_REG_PPC_IAMR 64 2429 PPC KVM_REG_PPC_TFHAR 64 2430 PPC KVM_REG_PPC_TFIAR 64 2431 PPC KVM_REG_PPC_TEXASR 64 2432 PPC KVM_REG_PPC_FSCR 64 2433 PPC KVM_REG_PPC_PSPB 32 2434 PPC KVM_REG_PPC_EBBHR 64 2435 PPC KVM_REG_PPC_EBBRR 64 2436 PPC KVM_REG_PPC_BESCR 64 2437 PPC KVM_REG_PPC_TAR 64 2438 PPC KVM_REG_PPC_DPDES 64 2439 PPC KVM_REG_PPC_DAWR 64 2440 PPC KVM_REG_PPC_DAWRX 64 2441 PPC KVM_REG_PPC_CIABR 64 2442 PPC KVM_REG_PPC_IC 64 2443 PPC KVM_REG_PPC_VTB 64 2444 PPC KVM_REG_PPC_CSIGR 64 2445 PPC KVM_REG_PPC_TACR 64 2446 PPC KVM_REG_PPC_TCSCR 64 2447 PPC KVM_REG_PPC_PID 64 2448 PPC KVM_REG_PPC_ACOP 64 2449 PPC KVM_REG_PPC_VRSAVE 32 2450 PPC KVM_REG_PPC_LPCR 32 2451 PPC KVM_REG_PPC_LPCR_64 64 2452 PPC KVM_REG_PPC_PPR 64 2453 PPC KVM_REG_PPC_ARCH_COMPAT 32 2454 PPC KVM_REG_PPC_DABRX 32 2455 PPC KVM_REG_PPC_WORT 64 2456 PPC KVM_REG_PPC_SPRG9 64 2457 PPC KVM_REG_PPC_DBSR 32 2458 PPC KVM_REG_PPC_TIDR 64 2459 PPC KVM_REG_PPC_PSSCR 64 2460 PPC KVM_REG_PPC_DEC_EXPIRY 64 2461 PPC KVM_REG_PPC_PTCR 64 2462 PPC KVM_REG_PPC_HASHKEYR 64 2463 PPC KVM_REG_PPC_HASHPKEYR 64 2464 PPC KVM_REG_PPC_DAWR1 64 2465 PPC KVM_REG_PPC_DAWRX1 64 2466 PPC KVM_REG_PPC_DEXCR 64 2467 PPC KVM_REG_PPC_TM_GPR0 64 2468 ... 2469 PPC KVM_REG_PPC_TM_GPR31 64 2470 PPC KVM_REG_PPC_TM_VSR0 128 2471 ... 2472 PPC KVM_REG_PPC_TM_VSR63 128 2473 PPC KVM_REG_PPC_TM_CR 64 2474 PPC KVM_REG_PPC_TM_LR 64 2475 PPC KVM_REG_PPC_TM_CTR 64 2476 PPC KVM_REG_PPC_TM_FPSCR 64 2477 PPC KVM_REG_PPC_TM_AMR 64 2478 PPC KVM_REG_PPC_TM_PPR 64 2479 PPC KVM_REG_PPC_TM_VRSAVE 64 2480 PPC KVM_REG_PPC_TM_VSCR 32 2481 PPC KVM_REG_PPC_TM_DSCR 64 2482 PPC KVM_REG_PPC_TM_TAR 64 2483 PPC KVM_REG_PPC_TM_XER 64 2484 2485 MIPS KVM_REG_MIPS_R0 64 2486 ... 2487 MIPS KVM_REG_MIPS_R31 64 2488 MIPS KVM_REG_MIPS_HI 64 2489 MIPS KVM_REG_MIPS_LO 64 2490 MIPS KVM_REG_MIPS_PC 64 2491 MIPS KVM_REG_MIPS_CP0_INDEX 32 2492 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2493 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2494 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2495 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2496 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2497 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2498 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2499 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2500 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2501 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2502 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2503 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2504 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2505 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2506 MIPS KVM_REG_MIPS_CP0_WIRED 32 2507 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2508 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2509 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2510 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2511 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2512 MIPS KVM_REG_MIPS_CP0_COUNT 32 2513 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2514 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2515 MIPS KVM_REG_MIPS_CP0_STATUS 32 2516 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2517 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2518 MIPS KVM_REG_MIPS_CP0_EPC 64 2519 MIPS KVM_REG_MIPS_CP0_PRID 32 2520 MIPS KVM_REG_MIPS_CP0_EBASE 64 2521 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2522 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2523 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2524 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2525 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2526 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2527 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2528 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2529 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2530 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2531 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2532 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2533 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2534 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2535 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2536 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2537 MIPS KVM_REG_MIPS_COUNT_CTL 64 2538 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2539 MIPS KVM_REG_MIPS_COUNT_HZ 64 2540 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2541 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2542 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2543 MIPS KVM_REG_MIPS_FCR_IR 32 2544 MIPS KVM_REG_MIPS_FCR_CSR 32 2545 MIPS KVM_REG_MIPS_MSA_IR 32 2546 MIPS KVM_REG_MIPS_MSA_CSR 32 2547 ======= =============================== ============ 2548 2549ARM registers are mapped using the lower 32 bits. The upper 16 of that 2550is the register group type, or coprocessor number: 2551 2552ARM core registers have the following id bit patterns:: 2553 2554 0x4020 0000 0010 <index into the kvm_regs struct:16> 2555 2556ARM 32-bit CP15 registers have the following id bit patterns:: 2557 2558 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 2559 2560ARM 64-bit CP15 registers have the following id bit patterns:: 2561 2562 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 2563 2564ARM CCSIDR registers are demultiplexed by CSSELR value:: 2565 2566 0x4020 0000 0011 00 <csselr:8> 2567 2568ARM 32-bit VFP control registers have the following id bit patterns:: 2569 2570 0x4020 0000 0012 1 <regno:12> 2571 2572ARM 64-bit FP registers have the following id bit patterns:: 2573 2574 0x4030 0000 0012 0 <regno:12> 2575 2576ARM firmware pseudo-registers have the following bit pattern:: 2577 2578 0x4030 0000 0014 <regno:16> 2579 2580 2581arm64 registers are mapped using the lower 32 bits. The upper 16 of 2582that is the register group type, or coprocessor number: 2583 2584arm64 core/FP-SIMD registers have the following id bit patterns. Note 2585that the size of the access is variable, as the kvm_regs structure 2586contains elements ranging from 32 to 128 bits. The index is a 32bit 2587value in the kvm_regs structure seen as a 32bit array:: 2588 2589 0x60x0 0000 0010 <index into the kvm_regs struct:16> 2590 2591Specifically: 2592 2593======================= ========= ===== ======================================= 2594 Encoding Register Bits kvm_regs member 2595======================= ========= ===== ======================================= 2596 0x6030 0000 0010 0000 X0 64 regs.regs[0] 2597 0x6030 0000 0010 0002 X1 64 regs.regs[1] 2598 ... 2599 0x6030 0000 0010 003c X30 64 regs.regs[30] 2600 0x6030 0000 0010 003e SP 64 regs.sp 2601 0x6030 0000 0010 0040 PC 64 regs.pc 2602 0x6030 0000 0010 0042 PSTATE 64 regs.pstate 2603 0x6030 0000 0010 0044 SP_EL1 64 sp_el1 2604 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1 2605 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC) 2606 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT] 2607 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND] 2608 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ] 2609 0x6030 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ] 2610 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_ 2611 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_ 2612 ... 2613 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_ 2614 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr 2615 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr 2616======================= ========= ===== ======================================= 2617 2618.. [1] These encodings are not accepted for SVE-enabled vcpus. See 2619 :ref:`KVM_ARM_VCPU_INIT`. 2620 2621 The equivalent register content can be accessed via bits [127:0] of 2622 the corresponding SVE Zn registers instead for vcpus that have SVE 2623 enabled (see below). 2624 2625arm64 CCSIDR registers are demultiplexed by CSSELR value:: 2626 2627 0x6020 0000 0011 00 <csselr:8> 2628 2629arm64 system registers have the following id bit patterns:: 2630 2631 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 2632 2633.. warning:: 2634 2635 Two system register IDs do not follow the specified pattern. These 2636 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to 2637 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These 2638 two had their values accidentally swapped, which means TIMER_CVAL is 2639 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is 2640 derived from the register encoding for CNTV_CVAL_EL0. As this is 2641 API, it must remain this way. 2642 2643arm64 firmware pseudo-registers have the following bit pattern:: 2644 2645 0x6030 0000 0014 <regno:16> 2646 2647arm64 SVE registers have the following bit patterns:: 2648 2649 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice] 2650 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice] 2651 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice] 2652 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register 2653 2654Access to register IDs where 2048 * slice >= 128 * max_vq will fail with 2655ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit 2656quadwords: see [2]_ below. 2657 2658These registers are only accessible on vcpus for which SVE is enabled. 2659See KVM_ARM_VCPU_INIT for details. 2660 2661In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not 2662accessible until the vcpu's SVE configuration has been finalized 2663using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT 2664and KVM_ARM_VCPU_FINALIZE for more information about this procedure. 2665 2666KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector 2667lengths supported by the vcpu to be discovered and configured by 2668userspace. When transferred to or from user memory via KVM_GET_ONE_REG 2669or KVM_SET_ONE_REG, the value of this register is of type 2670__u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as 2671follows:: 2672 2673 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS]; 2674 2675 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2676 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >> 2677 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1)) 2678 /* Vector length vq * 16 bytes supported */ 2679 else 2680 /* Vector length vq * 16 bytes not supported */ 2681 2682.. [2] The maximum value vq for which the above condition is true is 2683 max_vq. This is the maximum vector length available to the guest on 2684 this vcpu, and determines which register slices are visible through 2685 this ioctl interface. 2686 2687(See Documentation/arch/arm64/sve.rst for an explanation of the "vq" 2688nomenclature.) 2689 2690KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT. 2691KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that 2692the host supports. 2693 2694Userspace may subsequently modify it if desired until the vcpu's SVE 2695configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). 2696 2697Apart from simply removing all vector lengths from the host set that 2698exceed some value, support for arbitrarily chosen sets of vector lengths 2699is hardware-dependent and may not be available. Attempting to configure 2700an invalid set of vector lengths via KVM_SET_ONE_REG will fail with 2701EINVAL. 2702 2703After the vcpu's SVE configuration is finalized, further attempts to 2704write this register will fail with EPERM. 2705 2706arm64 bitmap feature firmware pseudo-registers have the following bit pattern:: 2707 2708 0x6030 0000 0016 <regno:16> 2709 2710The bitmap feature firmware registers exposes the hypercall services that 2711are available for userspace to configure. The set bits corresponds to the 2712services that are available for the guests to access. By default, KVM 2713sets all the supported bits during VM initialization. The userspace can 2714discover the available services via KVM_GET_ONE_REG, and write back the 2715bitmap corresponding to the features that it wishes guests to see via 2716KVM_SET_ONE_REG. 2717 2718Note: These registers are immutable once any of the vCPUs of the VM has 2719run at least once. A KVM_SET_ONE_REG in such a scenario will return 2720a -EBUSY to userspace. 2721 2722(See Documentation/virt/kvm/arm/hypercalls.rst for more details.) 2723 2724 2725MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 2726the register group type: 2727 2728MIPS core registers (see above) have the following id bit patterns:: 2729 2730 0x7030 0000 0000 <reg:16> 2731 2732MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 2733patterns depending on whether they're 32-bit or 64-bit registers:: 2734 2735 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 2736 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2737 2738Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 2739versions of the EntryLo registers regardless of the word size of the host 2740hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 2741with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 2742the PFNX field starting at bit 30. 2743 2744MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 2745patterns:: 2746 2747 0x7030 0000 0001 01 <reg:8> 2748 2749MIPS KVM control registers (see above) have the following id bit patterns:: 2750 2751 0x7030 0000 0002 <reg:16> 2752 2753MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2754id bit patterns depending on the size of the register being accessed. They are 2755always accessed according to the current guest FPU mode (Status.FR and 2756Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2757if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2758registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2759overlap the FPU registers:: 2760 2761 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2762 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2763 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2764 2765MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2766following id bit patterns:: 2767 2768 0x7020 0000 0003 01 <0:3> <reg:5> 2769 2770MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2771following id bit patterns:: 2772 2773 0x7020 0000 0003 02 <0:3> <reg:5> 2774 2775RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of 2776that is the register group type. 2777 2778RISC-V config registers are meant for configuring a Guest VCPU and it has 2779the following id bit patterns:: 2780 2781 0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host) 2782 0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host) 2783 2784Following are the RISC-V config registers: 2785 2786======================= ========= ============================================= 2787 Encoding Register Description 2788======================= ========= ============================================= 2789 0x80x0 0000 0100 0000 isa ISA feature bitmap of Guest VCPU 2790======================= ========= ============================================= 2791 2792The isa config register can be read anytime but can only be written before 2793a Guest VCPU runs. It will have ISA feature bits matching underlying host 2794set by default. 2795 2796RISC-V core registers represent the general execution state of a Guest VCPU 2797and it has the following id bit patterns:: 2798 2799 0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host) 2800 0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host) 2801 2802Following are the RISC-V core registers: 2803 2804======================= ========= ============================================= 2805 Encoding Register Description 2806======================= ========= ============================================= 2807 0x80x0 0000 0200 0000 regs.pc Program counter 2808 0x80x0 0000 0200 0001 regs.ra Return address 2809 0x80x0 0000 0200 0002 regs.sp Stack pointer 2810 0x80x0 0000 0200 0003 regs.gp Global pointer 2811 0x80x0 0000 0200 0004 regs.tp Task pointer 2812 0x80x0 0000 0200 0005 regs.t0 Caller saved register 0 2813 0x80x0 0000 0200 0006 regs.t1 Caller saved register 1 2814 0x80x0 0000 0200 0007 regs.t2 Caller saved register 2 2815 0x80x0 0000 0200 0008 regs.s0 Callee saved register 0 2816 0x80x0 0000 0200 0009 regs.s1 Callee saved register 1 2817 0x80x0 0000 0200 000a regs.a0 Function argument (or return value) 0 2818 0x80x0 0000 0200 000b regs.a1 Function argument (or return value) 1 2819 0x80x0 0000 0200 000c regs.a2 Function argument 2 2820 0x80x0 0000 0200 000d regs.a3 Function argument 3 2821 0x80x0 0000 0200 000e regs.a4 Function argument 4 2822 0x80x0 0000 0200 000f regs.a5 Function argument 5 2823 0x80x0 0000 0200 0010 regs.a6 Function argument 6 2824 0x80x0 0000 0200 0011 regs.a7 Function argument 7 2825 0x80x0 0000 0200 0012 regs.s2 Callee saved register 2 2826 0x80x0 0000 0200 0013 regs.s3 Callee saved register 3 2827 0x80x0 0000 0200 0014 regs.s4 Callee saved register 4 2828 0x80x0 0000 0200 0015 regs.s5 Callee saved register 5 2829 0x80x0 0000 0200 0016 regs.s6 Callee saved register 6 2830 0x80x0 0000 0200 0017 regs.s7 Callee saved register 7 2831 0x80x0 0000 0200 0018 regs.s8 Callee saved register 8 2832 0x80x0 0000 0200 0019 regs.s9 Callee saved register 9 2833 0x80x0 0000 0200 001a regs.s10 Callee saved register 10 2834 0x80x0 0000 0200 001b regs.s11 Callee saved register 11 2835 0x80x0 0000 0200 001c regs.t3 Caller saved register 3 2836 0x80x0 0000 0200 001d regs.t4 Caller saved register 4 2837 0x80x0 0000 0200 001e regs.t5 Caller saved register 5 2838 0x80x0 0000 0200 001f regs.t6 Caller saved register 6 2839 0x80x0 0000 0200 0020 mode Privilege mode (1 = S-mode or 0 = U-mode) 2840======================= ========= ============================================= 2841 2842RISC-V csr registers represent the supervisor mode control/status registers 2843of a Guest VCPU and it has the following id bit patterns:: 2844 2845 0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host) 2846 0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host) 2847 2848Following are the RISC-V csr registers: 2849 2850======================= ========= ============================================= 2851 Encoding Register Description 2852======================= ========= ============================================= 2853 0x80x0 0000 0300 0000 sstatus Supervisor status 2854 0x80x0 0000 0300 0001 sie Supervisor interrupt enable 2855 0x80x0 0000 0300 0002 stvec Supervisor trap vector base 2856 0x80x0 0000 0300 0003 sscratch Supervisor scratch register 2857 0x80x0 0000 0300 0004 sepc Supervisor exception program counter 2858 0x80x0 0000 0300 0005 scause Supervisor trap cause 2859 0x80x0 0000 0300 0006 stval Supervisor bad address or instruction 2860 0x80x0 0000 0300 0007 sip Supervisor interrupt pending 2861 0x80x0 0000 0300 0008 satp Supervisor address translation and protection 2862======================= ========= ============================================= 2863 2864RISC-V timer registers represent the timer state of a Guest VCPU and it has 2865the following id bit patterns:: 2866 2867 0x8030 0000 04 <index into the kvm_riscv_timer struct:24> 2868 2869Following are the RISC-V timer registers: 2870 2871======================= ========= ============================================= 2872 Encoding Register Description 2873======================= ========= ============================================= 2874 0x8030 0000 0400 0000 frequency Time base frequency (read-only) 2875 0x8030 0000 0400 0001 time Time value visible to Guest 2876 0x8030 0000 0400 0002 compare Time compare programmed by Guest 2877 0x8030 0000 0400 0003 state Time compare state (1 = ON or 0 = OFF) 2878======================= ========= ============================================= 2879 2880RISC-V F-extension registers represent the single precision floating point 2881state of a Guest VCPU and it has the following id bit patterns:: 2882 2883 0x8020 0000 05 <index into the __riscv_f_ext_state struct:24> 2884 2885Following are the RISC-V F-extension registers: 2886 2887======================= ========= ============================================= 2888 Encoding Register Description 2889======================= ========= ============================================= 2890 0x8020 0000 0500 0000 f[0] Floating point register 0 2891 ... 2892 0x8020 0000 0500 001f f[31] Floating point register 31 2893 0x8020 0000 0500 0020 fcsr Floating point control and status register 2894======================= ========= ============================================= 2895 2896RISC-V D-extension registers represent the double precision floating point 2897state of a Guest VCPU and it has the following id bit patterns:: 2898 2899 0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr) 2900 0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr) 2901 2902Following are the RISC-V D-extension registers: 2903 2904======================= ========= ============================================= 2905 Encoding Register Description 2906======================= ========= ============================================= 2907 0x8030 0000 0600 0000 f[0] Floating point register 0 2908 ... 2909 0x8030 0000 0600 001f f[31] Floating point register 31 2910 0x8020 0000 0600 0020 fcsr Floating point control and status register 2911======================= ========= ============================================= 2912 2913LoongArch registers are mapped using the lower 32 bits. The upper 16 bits of 2914that is the register group type. 2915 2916LoongArch csr registers are used to control guest cpu or get status of guest 2917cpu, and they have the following id bit patterns:: 2918 2919 0x9030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2920 2921LoongArch KVM control registers are used to implement some new defined functions 2922such as set vcpu counter or reset vcpu, and they have the following id bit patterns:: 2923 2924 0x9030 0000 0002 <reg:16> 2925 2926x86 MSR registers have the following id bit patterns:: 2927 0x2030 0002 <msr number:32> 2928 2929Following are the KVM-defined registers for x86: 2930 2931======================= ========= ============================================= 2932 Encoding Register Description 2933======================= ========= ============================================= 2934 0x2030 0003 0000 0000 SSP Shadow Stack Pointer 2935======================= ========= ============================================= 2936 29374.69 KVM_GET_ONE_REG 2938-------------------- 2939 2940:Capability: KVM_CAP_ONE_REG 2941:Architectures: all 2942:Type: vcpu ioctl 2943:Parameters: struct kvm_one_reg (in and out) 2944:Returns: 0 on success, negative value on failure 2945 2946Errors include: 2947 2948 ======== ============================================================ 2949 ENOENT no such register 2950 EINVAL invalid register ID, or no such register or used with VMs in 2951 protected virtualization mode on s390 2952 EPERM (arm64) register access not allowed before vcpu finalization 2953 ======== ============================================================ 2954 2955(These error codes are indicative only: do not rely on a specific error 2956code being returned in a specific situation.) 2957 2958This ioctl allows to receive the value of a single register implemented 2959in a vcpu. The register to read is indicated by the "id" field of the 2960kvm_one_reg struct passed in. On success, the register value can be found 2961at the memory location pointed to by "addr". 2962 2963The list of registers accessible using this interface is identical to the 2964list in 4.68. 2965 2966 29674.70 KVM_KVMCLOCK_CTRL 2968---------------------- 2969 2970:Capability: KVM_CAP_KVMCLOCK_CTRL 2971:Architectures: Any that implement pvclocks (currently x86 only) 2972:Type: vcpu ioctl 2973:Parameters: None 2974:Returns: 0 on success, -1 on error 2975 2976This ioctl sets a flag accessible to the guest indicating that the specified 2977vCPU has been paused by the host userspace. 2978 2979The host will set a flag in the pvclock structure that is checked from the 2980soft lockup watchdog. The flag is part of the pvclock structure that is 2981shared between guest and host, specifically the second bit of the flags 2982field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2983the host and read/cleared exclusively by the guest. The guest operation of 2984checking and clearing the flag must be an atomic operation so 2985load-link/store-conditional, or equivalent must be used. There are two cases 2986where the guest will clear the flag: when the soft lockup watchdog timer resets 2987itself or when a soft lockup is detected. This ioctl can be called any time 2988after pausing the vcpu, but before it is resumed. 2989 2990 29914.71 KVM_SIGNAL_MSI 2992------------------- 2993 2994:Capability: KVM_CAP_SIGNAL_MSI 2995:Architectures: x86 arm64 2996:Type: vm ioctl 2997:Parameters: struct kvm_msi (in) 2998:Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2999 3000Directly inject a MSI message. Only valid with in-kernel irqchip that handles 3001MSI messages. 3002 3003:: 3004 3005 struct kvm_msi { 3006 __u32 address_lo; 3007 __u32 address_hi; 3008 __u32 data; 3009 __u32 flags; 3010 __u32 devid; 3011 __u8 pad[12]; 3012 }; 3013 3014flags: 3015 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 3016 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 3017 the device ID. If this capability is not available, userspace 3018 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 3019 3020If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 3021for the device that wrote the MSI message. For PCI, this is usually a 3022BDF identifier in the lower 16 bits. 3023 3024On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 3025feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 3026address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 3027address_hi must be zero. 3028 3029 30304.71 KVM_CREATE_PIT2 3031-------------------- 3032 3033:Capability: KVM_CAP_PIT2 3034:Architectures: x86 3035:Type: vm ioctl 3036:Parameters: struct kvm_pit_config (in) 3037:Returns: 0 on success, -1 on error 3038 3039Creates an in-kernel device model for the i8254 PIT. This call is only valid 3040after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 3041parameters have to be passed:: 3042 3043 struct kvm_pit_config { 3044 __u32 flags; 3045 __u32 pad[15]; 3046 }; 3047 3048Valid flags are:: 3049 3050 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 3051 3052PIT timer interrupts may use a per-VM kernel thread for injection. If it 3053exists, this thread will have a name of the following pattern:: 3054 3055 kvm-pit/<owner-process-pid> 3056 3057When running a guest with elevated priorities, the scheduling parameters of 3058this thread may have to be adjusted accordingly. 3059 3060This IOCTL replaces the obsolete KVM_CREATE_PIT. 3061 3062 30634.72 KVM_GET_PIT2 3064----------------- 3065 3066:Capability: KVM_CAP_PIT_STATE2 3067:Architectures: x86 3068:Type: vm ioctl 3069:Parameters: struct kvm_pit_state2 (out) 3070:Returns: 0 on success, -1 on error 3071 3072Retrieves the state of the in-kernel PIT model. Only valid after 3073KVM_CREATE_PIT2. The state is returned in the following structure:: 3074 3075 struct kvm_pit_state2 { 3076 struct kvm_pit_channel_state channels[3]; 3077 __u32 flags; 3078 __u32 reserved[9]; 3079 }; 3080 3081Valid flags are:: 3082 3083 /* disable PIT in HPET legacy mode */ 3084 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 3085 /* speaker port data bit enabled */ 3086 #define KVM_PIT_FLAGS_SPEAKER_DATA_ON 0x00000002 3087 3088This IOCTL replaces the obsolete KVM_GET_PIT. 3089 3090 30914.73 KVM_SET_PIT2 3092----------------- 3093 3094:Capability: KVM_CAP_PIT_STATE2 3095:Architectures: x86 3096:Type: vm ioctl 3097:Parameters: struct kvm_pit_state2 (in) 3098:Returns: 0 on success, -1 on error 3099 3100Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 3101See KVM_GET_PIT2 for details on struct kvm_pit_state2. 3102 3103.. Tip:: 3104 ``KVM_SET_PIT2`` strictly adheres to the spec of Intel 8254 PIT. For example, 3105 a ``count`` value of 0 in ``struct kvm_pit_channel_state`` is interpreted as 3106 65536, which is the maximum count value. Refer to `Intel 8254 programmable 3107 interval timer <https://www.scs.stanford.edu/10wi-cs140/pintos/specs/8254.pdf>`_. 3108 3109This IOCTL replaces the obsolete KVM_SET_PIT. 3110 3111 31124.74 KVM_PPC_GET_SMMU_INFO 3113-------------------------- 3114 3115:Capability: KVM_CAP_PPC_GET_SMMU_INFO 3116:Architectures: powerpc 3117:Type: vm ioctl 3118:Parameters: None 3119:Returns: 0 on success, -1 on error 3120 3121This populates and returns a structure describing the features of 3122the "Server" class MMU emulation supported by KVM. 3123This can in turn be used by userspace to generate the appropriate 3124device-tree properties for the guest operating system. 3125 3126The structure contains some global information, followed by an 3127array of supported segment page sizes:: 3128 3129 struct kvm_ppc_smmu_info { 3130 __u64 flags; 3131 __u32 slb_size; 3132 __u32 pad; 3133 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3134 }; 3135 3136The supported flags are: 3137 3138 - KVM_PPC_PAGE_SIZES_REAL: 3139 When that flag is set, guest page sizes must "fit" the backing 3140 store page sizes. When not set, any page size in the list can 3141 be used regardless of how they are backed by userspace. 3142 3143 - KVM_PPC_1T_SEGMENTS 3144 The emulated MMU supports 1T segments in addition to the 3145 standard 256M ones. 3146 3147 - KVM_PPC_NO_HASH 3148 This flag indicates that HPT guests are not supported by KVM, 3149 thus all guests must use radix MMU mode. 3150 3151The "slb_size" field indicates how many SLB entries are supported 3152 3153The "sps" array contains 8 entries indicating the supported base 3154page sizes for a segment in increasing order. Each entry is defined 3155as follow:: 3156 3157 struct kvm_ppc_one_seg_page_size { 3158 __u32 page_shift; /* Base page shift of segment (or 0) */ 3159 __u32 slb_enc; /* SLB encoding for BookS */ 3160 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3161 }; 3162 3163An entry with a "page_shift" of 0 is unused. Because the array is 3164organized in increasing order, a lookup can stop when encountering 3165such an entry. 3166 3167The "slb_enc" field provides the encoding to use in the SLB for the 3168page size. The bits are in positions such as the value can directly 3169be OR'ed into the "vsid" argument of the slbmte instruction. 3170 3171The "enc" array is a list which for each of those segment base page 3172size provides the list of supported actual page sizes (which can be 3173only larger or equal to the base page size), along with the 3174corresponding encoding in the hash PTE. Similarly, the array is 31758 entries sorted by increasing sizes and an entry with a "0" shift 3176is an empty entry and a terminator:: 3177 3178 struct kvm_ppc_one_page_size { 3179 __u32 page_shift; /* Page shift (or 0) */ 3180 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 3181 }; 3182 3183The "pte_enc" field provides a value that can OR'ed into the hash 3184PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 3185into the hash PTE second double word). 3186 31874.75 KVM_IRQFD 3188-------------- 3189 3190:Capability: KVM_CAP_IRQFD 3191:Architectures: x86 s390 arm64 3192:Type: vm ioctl 3193:Parameters: struct kvm_irqfd (in) 3194:Returns: 0 on success, -1 on error 3195 3196Allows setting an eventfd to directly trigger a guest interrupt. 3197kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 3198kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 3199an event is triggered on the eventfd, an interrupt is injected into 3200the guest using the specified gsi pin. The irqfd is removed using 3201the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 3202and kvm_irqfd.gsi. 3203 3204With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 3205mechanism allowing emulation of level-triggered, irqfd-based 3206interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 3207additional eventfd in the kvm_irqfd.resamplefd field. When operating 3208in resample mode, posting of an interrupt through kvm_irq.fd asserts 3209the specified gsi in the irqchip. When the irqchip is resampled, such 3210as from an EOI, the gsi is de-asserted and the user is notified via 3211kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 3212the interrupt if the device making use of it still requires service. 3213Note that closing the resamplefd is not sufficient to disable the 3214irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 3215and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 3216 3217On arm64, gsi routing being supported, the following can happen: 3218 3219- in case no routing entry is associated to this gsi, injection fails 3220- in case the gsi is associated to an irqchip routing entry, 3221 irqchip.pin + 32 corresponds to the injected SPI ID. 3222- in case the gsi is associated to an MSI routing entry, the MSI 3223 message and device ID are translated into an LPI (support restricted 3224 to GICv3 ITS in-kernel emulation). 3225 32264.76 KVM_PPC_ALLOCATE_HTAB 3227-------------------------- 3228 3229:Capability: KVM_CAP_PPC_ALLOC_HTAB 3230:Architectures: powerpc 3231:Type: vm ioctl 3232:Parameters: Pointer to u32 containing hash table order (in/out) 3233:Returns: 0 on success, -1 on error 3234 3235This requests the host kernel to allocate an MMU hash table for a 3236guest using the PAPR paravirtualization interface. This only does 3237anything if the kernel is configured to use the Book 3S HV style of 3238virtualization. Otherwise the capability doesn't exist and the ioctl 3239returns an ENOTTY error. The rest of this description assumes Book 3S 3240HV. 3241 3242There must be no vcpus running when this ioctl is called; if there 3243are, it will do nothing and return an EBUSY error. 3244 3245The parameter is a pointer to a 32-bit unsigned integer variable 3246containing the order (log base 2) of the desired size of the hash 3247table, which must be between 18 and 46. On successful return from the 3248ioctl, the value will not be changed by the kernel. 3249 3250If no hash table has been allocated when any vcpu is asked to run 3251(with the KVM_RUN ioctl), the host kernel will allocate a 3252default-sized hash table (16 MB). 3253 3254If this ioctl is called when a hash table has already been allocated, 3255with a different order from the existing hash table, the existing hash 3256table will be freed and a new one allocated. If this is ioctl is 3257called when a hash table has already been allocated of the same order 3258as specified, the kernel will clear out the existing hash table (zero 3259all HPTEs). In either case, if the guest is using the virtualized 3260real-mode area (VRMA) facility, the kernel will re-create the VMRA 3261HPTEs on the next KVM_RUN of any vcpu. 3262 32634.77 KVM_S390_INTERRUPT 3264----------------------- 3265 3266:Capability: basic 3267:Architectures: s390 3268:Type: vm ioctl, vcpu ioctl 3269:Parameters: struct kvm_s390_interrupt (in) 3270:Returns: 0 on success, -1 on error 3271 3272Allows to inject an interrupt to the guest. Interrupts can be floating 3273(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 3274 3275Interrupt parameters are passed via kvm_s390_interrupt:: 3276 3277 struct kvm_s390_interrupt { 3278 __u32 type; 3279 __u32 parm; 3280 __u64 parm64; 3281 }; 3282 3283type can be one of the following: 3284 3285KVM_S390_SIGP_STOP (vcpu) 3286 - sigp stop; optional flags in parm 3287KVM_S390_PROGRAM_INT (vcpu) 3288 - program check; code in parm 3289KVM_S390_SIGP_SET_PREFIX (vcpu) 3290 - sigp set prefix; prefix address in parm 3291KVM_S390_RESTART (vcpu) 3292 - restart 3293KVM_S390_INT_CLOCK_COMP (vcpu) 3294 - clock comparator interrupt 3295KVM_S390_INT_CPU_TIMER (vcpu) 3296 - CPU timer interrupt 3297KVM_S390_INT_VIRTIO (vm) 3298 - virtio external interrupt; external interrupt 3299 parameters in parm and parm64 3300KVM_S390_INT_SERVICE (vm) 3301 - sclp external interrupt; sclp parameter in parm 3302KVM_S390_INT_EMERGENCY (vcpu) 3303 - sigp emergency; source cpu in parm 3304KVM_S390_INT_EXTERNAL_CALL (vcpu) 3305 - sigp external call; source cpu in parm 3306KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 3307 - compound value to indicate an 3308 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 3309 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 3310 interruption subclass) 3311KVM_S390_MCHK (vm, vcpu) 3312 - machine check interrupt; cr 14 bits in parm, machine check interrupt 3313 code in parm64 (note that machine checks needing further payload are not 3314 supported by this ioctl) 3315 3316This is an asynchronous vcpu ioctl and can be invoked from any thread. 3317 33184.78 KVM_PPC_GET_HTAB_FD 3319------------------------ 3320 3321:Capability: KVM_CAP_PPC_HTAB_FD 3322:Architectures: powerpc 3323:Type: vm ioctl 3324:Parameters: Pointer to struct kvm_get_htab_fd (in) 3325:Returns: file descriptor number (>= 0) on success, -1 on error 3326 3327This returns a file descriptor that can be used either to read out the 3328entries in the guest's hashed page table (HPT), or to write entries to 3329initialize the HPT. The returned fd can only be written to if the 3330KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 3331can only be read if that bit is clear. The argument struct looks like 3332this:: 3333 3334 /* For KVM_PPC_GET_HTAB_FD */ 3335 struct kvm_get_htab_fd { 3336 __u64 flags; 3337 __u64 start_index; 3338 __u64 reserved[2]; 3339 }; 3340 3341 /* Values for kvm_get_htab_fd.flags */ 3342 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 3343 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 3344 3345The 'start_index' field gives the index in the HPT of the entry at 3346which to start reading. It is ignored when writing. 3347 3348Reads on the fd will initially supply information about all 3349"interesting" HPT entries. Interesting entries are those with the 3350bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 3351all entries. When the end of the HPT is reached, the read() will 3352return. If read() is called again on the fd, it will start again from 3353the beginning of the HPT, but will only return HPT entries that have 3354changed since they were last read. 3355 3356Data read or written is structured as a header (8 bytes) followed by a 3357series of valid HPT entries (16 bytes) each. The header indicates how 3358many valid HPT entries there are and how many invalid entries follow 3359the valid entries. The invalid entries are not represented explicitly 3360in the stream. The header format is:: 3361 3362 struct kvm_get_htab_header { 3363 __u32 index; 3364 __u16 n_valid; 3365 __u16 n_invalid; 3366 }; 3367 3368Writes to the fd create HPT entries starting at the index given in the 3369header; first 'n_valid' valid entries with contents from the data 3370written, then 'n_invalid' invalid entries, invalidating any previously 3371valid entries found. 3372 33734.79 KVM_CREATE_DEVICE 3374---------------------- 3375 3376:Capability: KVM_CAP_DEVICE_CTRL 3377:Architectures: all 3378:Type: vm ioctl 3379:Parameters: struct kvm_create_device (in/out) 3380:Returns: 0 on success, -1 on error 3381 3382Errors: 3383 3384 ====== ======================================================= 3385 ENODEV The device type is unknown or unsupported 3386 EEXIST Device already created, and this type of device may not 3387 be instantiated multiple times 3388 ====== ======================================================= 3389 3390 Other error conditions may be defined by individual device types or 3391 have their standard meanings. 3392 3393Creates an emulated device in the kernel. The file descriptor returned 3394in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 3395 3396If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 3397device type is supported (not necessarily whether it can be created 3398in the current vm). 3399 3400Individual devices should not define flags. Attributes should be used 3401for specifying any behavior that is not implied by the device type 3402number. 3403 3404:: 3405 3406 struct kvm_create_device { 3407 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 3408 __u32 fd; /* out: device handle */ 3409 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 3410 }; 3411 34124.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3413-------------------------------------------- 3414 3415:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3416 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3417 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device (no set) 3418:Architectures: x86, arm64, s390 3419:Type: device ioctl, vm ioctl, vcpu ioctl 3420:Parameters: struct kvm_device_attr 3421:Returns: 0 on success, -1 on error 3422 3423Errors: 3424 3425 ===== ============================================================= 3426 ENXIO The group or attribute is unknown/unsupported for this device 3427 or hardware support is missing. 3428 EPERM The attribute cannot (currently) be accessed this way 3429 (e.g. read-only attribute, or attribute that only makes 3430 sense when the device is in a different state) 3431 ===== ============================================================= 3432 3433 Other error conditions may be defined by individual device types. 3434 3435Gets/sets a specified piece of device configuration and/or state. The 3436semantics are device-specific. See individual device documentation in 3437the "devices" directory. As with ONE_REG, the size of the data 3438transferred is defined by the particular attribute. 3439 3440:: 3441 3442 struct kvm_device_attr { 3443 __u32 flags; /* no flags currently defined */ 3444 __u32 group; /* device-defined */ 3445 __u64 attr; /* group-defined */ 3446 __u64 addr; /* userspace address of attr data */ 3447 }; 3448 34494.81 KVM_HAS_DEVICE_ATTR 3450------------------------ 3451 3452:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3453 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3454 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device 3455:Type: device ioctl, vm ioctl, vcpu ioctl 3456:Parameters: struct kvm_device_attr 3457:Returns: 0 on success, -1 on error 3458 3459Errors: 3460 3461 ===== ============================================================= 3462 ENXIO The group or attribute is unknown/unsupported for this device 3463 or hardware support is missing. 3464 ===== ============================================================= 3465 3466Tests whether a device supports a particular attribute. A successful 3467return indicates the attribute is implemented. It does not necessarily 3468indicate that the attribute can be read or written in the device's 3469current state. "addr" is ignored. 3470 3471.. _KVM_ARM_VCPU_INIT: 3472 34734.82 KVM_ARM_VCPU_INIT 3474---------------------- 3475 3476:Capability: basic 3477:Architectures: arm64 3478:Type: vcpu ioctl 3479:Parameters: struct kvm_vcpu_init (in) 3480:Returns: 0 on success; -1 on error 3481 3482Errors: 3483 3484 ====== ================================================================= 3485 EINVAL the target is unknown, or the combination of features is invalid. 3486 ENOENT a features bit specified is unknown. 3487 ====== ================================================================= 3488 3489This tells KVM what type of CPU to present to the guest, and what 3490optional features it should have. This will cause a reset of the cpu 3491registers to their initial values. If this is not called, KVM_RUN will 3492return ENOEXEC for that vcpu. 3493 3494The initial values are defined as: 3495 - Processor state: 3496 * AArch64: EL1h, D, A, I and F bits set. All other bits 3497 are cleared. 3498 * AArch32: SVC, A, I and F bits set. All other bits are 3499 cleared. 3500 - General Purpose registers, including PC and SP: set to 0 3501 - FPSIMD/NEON registers: set to 0 3502 - SVE registers: set to 0 3503 - System registers: Reset to their architecturally defined 3504 values as for a warm reset to EL1 (resp. SVC) or EL2 (in the 3505 case of EL2 being enabled). 3506 3507Note that because some registers reflect machine topology, all vcpus 3508should be created before this ioctl is invoked. 3509 3510Userspace can call this function multiple times for a given vcpu, including 3511after the vcpu has been run. This will reset the vcpu to its initial 3512state. All calls to this function after the initial call must use the same 3513target and same set of feature flags, otherwise EINVAL will be returned. 3514 3515Possible features: 3516 3517 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 3518 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 3519 and execute guest code when KVM_RUN is called. 3520 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 3521 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 3522 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision 3523 backward compatible with v0.2) for the CPU. 3524 Depends on KVM_CAP_ARM_PSCI_0_2. 3525 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 3526 Depends on KVM_CAP_ARM_PMU_V3. 3527 3528 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication 3529 for arm64 only. 3530 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS. 3531 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3532 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3533 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3534 requested. 3535 3536 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication 3537 for arm64 only. 3538 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC. 3539 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3540 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3541 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3542 requested. 3543 3544 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only). 3545 Depends on KVM_CAP_ARM_SVE. 3546 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3547 3548 * After KVM_ARM_VCPU_INIT: 3549 3550 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the 3551 initial value of this pseudo-register indicates the best set of 3552 vector lengths possible for a vcpu on this host. 3553 3554 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3555 3556 - KVM_RUN and KVM_GET_REG_LIST are not available; 3557 3558 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access 3559 the scalable architectural SVE registers 3560 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or 3561 KVM_REG_ARM64_SVE_FFR; 3562 3563 - KVM_REG_ARM64_SVE_VLS may optionally be written using 3564 KVM_SET_ONE_REG, to modify the set of vector lengths available 3565 for the vcpu. 3566 3567 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3568 3569 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can 3570 no longer be written using KVM_SET_ONE_REG. 3571 3572 - KVM_ARM_VCPU_HAS_EL2: Enable Nested Virtualisation support, 3573 booting the guest from EL2 instead of EL1. 3574 Depends on KVM_CAP_ARM_EL2. 3575 The VM is running with HCR_EL2.E2H being RES1 (VHE) unless 3576 KVM_ARM_VCPU_HAS_EL2_E2H0 is also set. 3577 3578 - KVM_ARM_VCPU_HAS_EL2_E2H0: Restrict Nested Virtualisation 3579 support to HCR_EL2.E2H being RES0 (non-VHE). 3580 Depends on KVM_CAP_ARM_EL2_E2H0. 3581 KVM_ARM_VCPU_HAS_EL2 must also be set. 3582 35834.83 KVM_ARM_PREFERRED_TARGET 3584----------------------------- 3585 3586:Capability: basic 3587:Architectures: arm64 3588:Type: vm ioctl 3589:Parameters: struct kvm_vcpu_init (out) 3590:Returns: 0 on success; -1 on error 3591 3592Errors: 3593 3594 ====== ========================================== 3595 ENODEV no preferred target available for the host 3596 ====== ========================================== 3597 3598This queries KVM for preferred CPU target type which can be emulated 3599by KVM on underlying host. 3600 3601The ioctl returns struct kvm_vcpu_init instance containing information 3602about preferred CPU target type and recommended features for it. The 3603kvm_vcpu_init->features bitmap returned will have feature bits set if 3604the preferred target recommends setting these features, but this is 3605not mandatory. 3606 3607The information returned by this ioctl can be used to prepare an instance 3608of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 3609VCPU matching underlying host. 3610 3611 36124.84 KVM_GET_REG_LIST 3613--------------------- 3614 3615:Capability: basic 3616:Architectures: arm64, mips, riscv, x86 (if KVM_CAP_ONE_REG) 3617:Type: vcpu ioctl 3618:Parameters: struct kvm_reg_list (in/out) 3619:Returns: 0 on success; -1 on error 3620 3621Errors: 3622 3623 ===== ============================================================== 3624 E2BIG the reg index list is too big to fit in the array specified by 3625 the user (the number required will be written into n). 3626 ===== ============================================================== 3627 3628:: 3629 3630 struct kvm_reg_list { 3631 __u64 n; /* number of registers in reg[] */ 3632 __u64 reg[0]; 3633 }; 3634 3635This ioctl returns the guest registers that are supported for the 3636KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3637 3638Note that s390 does not support KVM_GET_REG_LIST for historical reasons 3639(read: nobody cared). The set of registers in kernels 4.x and newer is: 3640 3641- KVM_REG_S390_TODPR 3642 3643- KVM_REG_S390_EPOCHDIFF 3644 3645- KVM_REG_S390_CPU_TIMER 3646 3647- KVM_REG_S390_CLOCK_COMP 3648 3649- KVM_REG_S390_PFTOKEN 3650 3651- KVM_REG_S390_PFCOMPARE 3652 3653- KVM_REG_S390_PFSELECT 3654 3655- KVM_REG_S390_PP 3656 3657- KVM_REG_S390_GBEA 3658 3659Note, for x86, all MSRs enumerated by KVM_GET_MSR_INDEX_LIST are supported as 3660type KVM_X86_REG_TYPE_MSR, but are NOT enumerated via KVM_GET_REG_LIST. 3661 36624.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3663----------------------------------------- 3664 3665:Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3666:Architectures: arm64 3667:Type: vm ioctl 3668:Parameters: struct kvm_arm_device_address (in) 3669:Returns: 0 on success, -1 on error 3670 3671Errors: 3672 3673 ====== ============================================ 3674 ENODEV The device id is unknown 3675 ENXIO Device not supported on current system 3676 EEXIST Address already set 3677 E2BIG Address outside guest physical address space 3678 EBUSY Address overlaps with other device range 3679 ====== ============================================ 3680 3681:: 3682 3683 struct kvm_arm_device_addr { 3684 __u64 id; 3685 __u64 addr; 3686 }; 3687 3688Specify a device address in the guest's physical address space where guests 3689can access emulated or directly exposed devices, which the host kernel needs 3690to know about. The id field is an architecture specific identifier for a 3691specific device. 3692 3693arm64 divides the id field into two parts, a device id and an 3694address type id specific to the individual device:: 3695 3696 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 3697 field: | 0x00000000 | device id | addr type id | 3698 3699arm64 currently only require this when using the in-kernel GIC 3700support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 3701as the device id. When setting the base address for the guest's 3702mapping of the VGIC virtual CPU and distributor interface, the ioctl 3703must be called after calling KVM_CREATE_IRQCHIP, but before calling 3704KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 3705base addresses will return -EEXIST. 3706 3707Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 3708should be used instead. 3709 3710 37114.86 KVM_PPC_RTAS_DEFINE_TOKEN 3712------------------------------ 3713 3714:Capability: KVM_CAP_PPC_RTAS 3715:Architectures: ppc 3716:Type: vm ioctl 3717:Parameters: struct kvm_rtas_token_args 3718:Returns: 0 on success, -1 on error 3719 3720Defines a token value for a RTAS (Run Time Abstraction Services) 3721service in order to allow it to be handled in the kernel. The 3722argument struct gives the name of the service, which must be the name 3723of a service that has a kernel-side implementation. If the token 3724value is non-zero, it will be associated with that service, and 3725subsequent RTAS calls by the guest specifying that token will be 3726handled by the kernel. If the token value is 0, then any token 3727associated with the service will be forgotten, and subsequent RTAS 3728calls by the guest for that service will be passed to userspace to be 3729handled. 3730 37314.87 KVM_SET_GUEST_DEBUG 3732------------------------ 3733 3734:Capability: KVM_CAP_SET_GUEST_DEBUG 3735:Architectures: x86, s390, ppc, arm64 3736:Type: vcpu ioctl 3737:Parameters: struct kvm_guest_debug (in) 3738:Returns: 0 on success; -1 on error 3739 3740:: 3741 3742 struct kvm_guest_debug { 3743 __u32 control; 3744 __u32 pad; 3745 struct kvm_guest_debug_arch arch; 3746 }; 3747 3748Set up the processor specific debug registers and configure vcpu for 3749handling guest debug events. There are two parts to the structure, the 3750first a control bitfield indicates the type of debug events to handle 3751when running. Common control bits are: 3752 3753 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 3754 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 3755 3756The top 16 bits of the control field are architecture specific control 3757flags which can include the following: 3758 3759 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 3760 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390] 3761 - KVM_GUESTDBG_USE_HW: using hardware debug events [arm64] 3762 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 3763 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 3764 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 3765 - KVM_GUESTDBG_BLOCKIRQ: avoid injecting interrupts/NMI/SMI [x86] 3766 3767For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 3768are enabled in memory so we need to ensure breakpoint exceptions are 3769correctly trapped and the KVM run loop exits at the breakpoint and not 3770running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 3771we need to ensure the guest vCPUs architecture specific registers are 3772updated to the correct (supplied) values. 3773 3774The second part of the structure is architecture specific and 3775typically contains a set of debug registers. 3776 3777For arm64 the number of debug registers is implementation defined and 3778can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 3779KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 3780indicating the number of supported registers. 3781 3782For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether 3783the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported. 3784 3785Also when supported, KVM_CAP_SET_GUEST_DEBUG2 capability indicates the 3786supported KVM_GUESTDBG_* bits in the control field. 3787 3788When debug events exit the main run loop with the reason 3789KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 3790structure containing architecture specific debug information. 3791 37924.88 KVM_GET_EMULATED_CPUID 3793--------------------------- 3794 3795:Capability: KVM_CAP_EXT_EMUL_CPUID 3796:Architectures: x86 3797:Type: system ioctl 3798:Parameters: struct kvm_cpuid2 (in/out) 3799:Returns: 0 on success, -1 on error 3800 3801:: 3802 3803 struct kvm_cpuid2 { 3804 __u32 nent; 3805 __u32 flags; 3806 struct kvm_cpuid_entry2 entries[0]; 3807 }; 3808 3809The member 'flags' is used for passing flags from userspace. 3810 3811:: 3812 3813 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 3814 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 3815 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 3816 3817 struct kvm_cpuid_entry2 { 3818 __u32 function; 3819 __u32 index; 3820 __u32 flags; 3821 __u32 eax; 3822 __u32 ebx; 3823 __u32 ecx; 3824 __u32 edx; 3825 __u32 padding[3]; 3826 }; 3827 3828This ioctl returns x86 cpuid features which are emulated by 3829kvm.Userspace can use the information returned by this ioctl to query 3830which features are emulated by kvm instead of being present natively. 3831 3832Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 3833structure with the 'nent' field indicating the number of entries in 3834the variable-size array 'entries'. If the number of entries is too low 3835to describe the cpu capabilities, an error (E2BIG) is returned. If the 3836number is too high, the 'nent' field is adjusted and an error (ENOMEM) 3837is returned. If the number is just right, the 'nent' field is adjusted 3838to the number of valid entries in the 'entries' array, which is then 3839filled. 3840 3841The entries returned are the set CPUID bits of the respective features 3842which kvm emulates, as returned by the CPUID instruction, with unknown 3843or unsupported feature bits cleared. 3844 3845Features like x2apic, for example, may not be present in the host cpu 3846but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 3847emulated efficiently and thus not included here. 3848 3849The fields in each entry are defined as follows: 3850 3851 function: 3852 the eax value used to obtain the entry 3853 index: 3854 the ecx value used to obtain the entry (for entries that are 3855 affected by ecx) 3856 flags: 3857 an OR of zero or more of the following: 3858 3859 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3860 if the index field is valid 3861 3862 eax, ebx, ecx, edx: 3863 3864 the values returned by the cpuid instruction for 3865 this function/index combination 3866 38674.89 KVM_S390_MEM_OP 3868-------------------- 3869 3870:Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S390_PROTECTED, KVM_CAP_S390_MEM_OP_EXTENSION 3871:Architectures: s390 3872:Type: vm ioctl, vcpu ioctl 3873:Parameters: struct kvm_s390_mem_op (in) 3874:Returns: = 0 on success, 3875 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 3876 16 bit program exception code if the access causes such an exception 3877 3878Read or write data from/to the VM's memory. 3879The KVM_CAP_S390_MEM_OP_EXTENSION capability specifies what functionality is 3880supported. 3881 3882Parameters are specified via the following structure:: 3883 3884 struct kvm_s390_mem_op { 3885 __u64 gaddr; /* the guest address */ 3886 __u64 flags; /* flags */ 3887 __u32 size; /* amount of bytes */ 3888 __u32 op; /* type of operation */ 3889 __u64 buf; /* buffer in userspace */ 3890 union { 3891 struct { 3892 __u8 ar; /* the access register number */ 3893 __u8 key; /* access key, ignored if flag unset */ 3894 __u8 pad1[6]; /* ignored */ 3895 __u64 old_addr; /* ignored if flag unset */ 3896 }; 3897 __u32 sida_offset; /* offset into the sida */ 3898 __u8 reserved[32]; /* ignored */ 3899 }; 3900 }; 3901 3902The start address of the memory region has to be specified in the "gaddr" 3903field, and the length of the region in the "size" field (which must not 3904be 0). The maximum value for "size" can be obtained by checking the 3905KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the 3906userspace application where the read data should be written to for 3907a read access, or where the data that should be written is stored for 3908a write access. The "reserved" field is meant for future extensions. 3909Reserved and unused values are ignored. Future extension that add members must 3910introduce new flags. 3911 3912The type of operation is specified in the "op" field. Flags modifying 3913their behavior can be set in the "flags" field. Undefined flag bits must 3914be set to 0. 3915 3916Possible operations are: 3917 * ``KVM_S390_MEMOP_LOGICAL_READ`` 3918 * ``KVM_S390_MEMOP_LOGICAL_WRITE`` 3919 * ``KVM_S390_MEMOP_ABSOLUTE_READ`` 3920 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE`` 3921 * ``KVM_S390_MEMOP_SIDA_READ`` 3922 * ``KVM_S390_MEMOP_SIDA_WRITE`` 3923 * ``KVM_S390_MEMOP_ABSOLUTE_CMPXCHG`` 3924 3925Logical read/write: 3926^^^^^^^^^^^^^^^^^^^ 3927 3928Access logical memory, i.e. translate the given guest address to an absolute 3929address given the state of the VCPU and use the absolute address as target of 3930the access. "ar" designates the access register number to be used; the valid 3931range is 0..15. 3932Logical accesses are permitted for the VCPU ioctl only. 3933Logical accesses are permitted for non-protected guests only. 3934 3935Supported flags: 3936 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3937 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION`` 3938 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3939 3940The KVM_S390_MEMOP_F_CHECK_ONLY flag can be set to check whether the 3941corresponding memory access would cause an access exception; however, 3942no actual access to the data in memory at the destination is performed. 3943In this case, "buf" is unused and can be NULL. 3944 3945In case an access exception occurred during the access (or would occur 3946in case of KVM_S390_MEMOP_F_CHECK_ONLY), the ioctl returns a positive 3947error number indicating the type of exception. This exception is also 3948raised directly at the corresponding VCPU if the flag 3949KVM_S390_MEMOP_F_INJECT_EXCEPTION is set. 3950On protection exceptions, unless specified otherwise, the injected 3951translation-exception identifier (TEID) indicates suppression. 3952 3953If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key 3954protection is also in effect and may cause exceptions if accesses are 3955prohibited given the access key designated by "key"; the valid range is 0..15. 3956KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION 3957is > 0. 3958Since the accessed memory may span multiple pages and those pages might have 3959different storage keys, it is possible that a protection exception occurs 3960after memory has been modified. In this case, if the exception is injected, 3961the TEID does not indicate suppression. 3962 3963Absolute read/write: 3964^^^^^^^^^^^^^^^^^^^^ 3965 3966Access absolute memory. This operation is intended to be used with the 3967KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to allow accessing memory and performing 3968the checks required for storage key protection as one operation (as opposed to 3969user space getting the storage keys, performing the checks, and accessing 3970memory thereafter, which could lead to a delay between check and access). 3971Absolute accesses are permitted for the VM ioctl if KVM_CAP_S390_MEM_OP_EXTENSION 3972has the KVM_S390_MEMOP_EXTENSION_CAP_BASE bit set. 3973Currently absolute accesses are not permitted for VCPU ioctls. 3974Absolute accesses are permitted for non-protected guests only. 3975 3976Supported flags: 3977 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3978 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3979 3980The semantics of the flags common with logical accesses are as for logical 3981accesses. 3982 3983Absolute cmpxchg: 3984^^^^^^^^^^^^^^^^^ 3985 3986Perform cmpxchg on absolute guest memory. Intended for use with the 3987KVM_S390_MEMOP_F_SKEY_PROTECTION flag. 3988Instead of doing an unconditional write, the access occurs only if the target 3989location contains the value pointed to by "old_addr". 3990This is performed as an atomic cmpxchg with the length specified by the "size" 3991parameter. "size" must be a power of two up to and including 16. 3992If the exchange did not take place because the target value doesn't match the 3993old value, the value "old_addr" points to is replaced by the target value. 3994User space can tell if an exchange took place by checking if this replacement 3995occurred. The cmpxchg op is permitted for the VM ioctl if 3996KVM_CAP_S390_MEM_OP_EXTENSION has flag KVM_S390_MEMOP_EXTENSION_CAP_CMPXCHG set. 3997 3998Supported flags: 3999 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 4000 4001SIDA read/write: 4002^^^^^^^^^^^^^^^^ 4003 4004Access the secure instruction data area which contains memory operands necessary 4005for instruction emulation for protected guests. 4006SIDA accesses are available if the KVM_CAP_S390_PROTECTED capability is available. 4007SIDA accesses are permitted for the VCPU ioctl only. 4008SIDA accesses are permitted for protected guests only. 4009 4010No flags are supported. 4011 40124.90 KVM_S390_GET_SKEYS 4013----------------------- 4014 4015:Capability: KVM_CAP_S390_SKEYS 4016:Architectures: s390 4017:Type: vm ioctl 4018:Parameters: struct kvm_s390_skeys 4019:Returns: 0 on success, KVM_S390_GET_SKEYS_NONE if guest is not using storage 4020 keys, negative value on error 4021 4022This ioctl is used to get guest storage key values on the s390 4023architecture. The ioctl takes parameters via the kvm_s390_skeys struct:: 4024 4025 struct kvm_s390_skeys { 4026 __u64 start_gfn; 4027 __u64 count; 4028 __u64 skeydata_addr; 4029 __u32 flags; 4030 __u32 reserved[9]; 4031 }; 4032 4033The start_gfn field is the number of the first guest frame whose storage keys 4034you want to get. 4035 4036The count field is the number of consecutive frames (starting from start_gfn) 4037whose storage keys to get. The count field must be at least 1 and the maximum 4038allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 4039will cause the ioctl to return -EINVAL. 4040 4041The skeydata_addr field is the address to a buffer large enough to hold count 4042bytes. This buffer will be filled with storage key data by the ioctl. 4043 40444.91 KVM_S390_SET_SKEYS 4045----------------------- 4046 4047:Capability: KVM_CAP_S390_SKEYS 4048:Architectures: s390 4049:Type: vm ioctl 4050:Parameters: struct kvm_s390_skeys 4051:Returns: 0 on success, negative value on error 4052 4053This ioctl is used to set guest storage key values on the s390 4054architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 4055See section on KVM_S390_GET_SKEYS for struct definition. 4056 4057The start_gfn field is the number of the first guest frame whose storage keys 4058you want to set. 4059 4060The count field is the number of consecutive frames (starting from start_gfn) 4061whose storage keys to get. The count field must be at least 1 and the maximum 4062allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 4063will cause the ioctl to return -EINVAL. 4064 4065The skeydata_addr field is the address to a buffer containing count bytes of 4066storage keys. Each byte in the buffer will be set as the storage key for a 4067single frame starting at start_gfn for count frames. 4068 4069Note: If any architecturally invalid key value is found in the given data then 4070the ioctl will return -EINVAL. 4071 40724.92 KVM_S390_IRQ 4073----------------- 4074 4075:Capability: KVM_CAP_S390_INJECT_IRQ 4076:Architectures: s390 4077:Type: vcpu ioctl 4078:Parameters: struct kvm_s390_irq (in) 4079:Returns: 0 on success, -1 on error 4080 4081Errors: 4082 4083 4084 ====== ================================================================= 4085 EINVAL interrupt type is invalid 4086 type is KVM_S390_SIGP_STOP and flag parameter is invalid value, 4087 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 4088 than the maximum of VCPUs 4089 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped, 4090 type is KVM_S390_SIGP_STOP and a stop irq is already pending, 4091 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 4092 is already pending 4093 ====== ================================================================= 4094 4095Allows to inject an interrupt to the guest. 4096 4097Using struct kvm_s390_irq as a parameter allows 4098to inject additional payload which is not 4099possible via KVM_S390_INTERRUPT. 4100 4101Interrupt parameters are passed via kvm_s390_irq:: 4102 4103 struct kvm_s390_irq { 4104 __u64 type; 4105 union { 4106 struct kvm_s390_io_info io; 4107 struct kvm_s390_ext_info ext; 4108 struct kvm_s390_pgm_info pgm; 4109 struct kvm_s390_emerg_info emerg; 4110 struct kvm_s390_extcall_info extcall; 4111 struct kvm_s390_prefix_info prefix; 4112 struct kvm_s390_stop_info stop; 4113 struct kvm_s390_mchk_info mchk; 4114 char reserved[64]; 4115 } u; 4116 }; 4117 4118type can be one of the following: 4119 4120- KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 4121- KVM_S390_PROGRAM_INT - program check; parameters in .pgm 4122- KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 4123- KVM_S390_RESTART - restart; no parameters 4124- KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 4125- KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 4126- KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 4127- KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 4128- KVM_S390_MCHK - machine check interrupt; parameters in .mchk 4129 4130This is an asynchronous vcpu ioctl and can be invoked from any thread. 4131 41324.94 KVM_S390_GET_IRQ_STATE 4133--------------------------- 4134 4135:Capability: KVM_CAP_S390_IRQ_STATE 4136:Architectures: s390 4137:Type: vcpu ioctl 4138:Parameters: struct kvm_s390_irq_state (out) 4139:Returns: >= number of bytes copied into buffer, 4140 -EINVAL if buffer size is 0, 4141 -ENOBUFS if buffer size is too small to fit all pending interrupts, 4142 -EFAULT if the buffer address was invalid 4143 4144This ioctl allows userspace to retrieve the complete state of all currently 4145pending interrupts in a single buffer. Use cases include migration 4146and introspection. The parameter structure contains the address of a 4147userspace buffer and its length:: 4148 4149 struct kvm_s390_irq_state { 4150 __u64 buf; 4151 __u32 flags; /* will stay unused for compatibility reasons */ 4152 __u32 len; 4153 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4154 }; 4155 4156Userspace passes in the above struct and for each pending interrupt a 4157struct kvm_s390_irq is copied to the provided buffer. 4158 4159The structure contains a flags and a reserved field for future extensions. As 4160the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 4161reserved, these fields can not be used in the future without breaking 4162compatibility. 4163 4164If -ENOBUFS is returned the buffer provided was too small and userspace 4165may retry with a bigger buffer. 4166 41674.95 KVM_S390_SET_IRQ_STATE 4168--------------------------- 4169 4170:Capability: KVM_CAP_S390_IRQ_STATE 4171:Architectures: s390 4172:Type: vcpu ioctl 4173:Parameters: struct kvm_s390_irq_state (in) 4174:Returns: 0 on success, 4175 -EFAULT if the buffer address was invalid, 4176 -EINVAL for an invalid buffer length (see below), 4177 -EBUSY if there were already interrupts pending, 4178 errors occurring when actually injecting the 4179 interrupt. See KVM_S390_IRQ. 4180 4181This ioctl allows userspace to set the complete state of all cpu-local 4182interrupts currently pending for the vcpu. It is intended for restoring 4183interrupt state after a migration. The input parameter is a userspace buffer 4184containing a struct kvm_s390_irq_state:: 4185 4186 struct kvm_s390_irq_state { 4187 __u64 buf; 4188 __u32 flags; /* will stay unused for compatibility reasons */ 4189 __u32 len; 4190 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4191 }; 4192 4193The restrictions for flags and reserved apply as well. 4194(see KVM_S390_GET_IRQ_STATE) 4195 4196The userspace memory referenced by buf contains a struct kvm_s390_irq 4197for each interrupt to be injected into the guest. 4198If one of the interrupts could not be injected for some reason the 4199ioctl aborts. 4200 4201len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 4202and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 4203which is the maximum number of possibly pending cpu-local interrupts. 4204 42054.96 KVM_SMI 4206------------ 4207 4208:Capability: KVM_CAP_X86_SMM 4209:Architectures: x86 4210:Type: vcpu ioctl 4211:Parameters: none 4212:Returns: 0 on success, -1 on error 4213 4214Queues an SMI on the thread's vcpu. 4215 42164.97 KVM_X86_SET_MSR_FILTER 4217---------------------------- 4218 4219:Capability: KVM_CAP_X86_MSR_FILTER 4220:Architectures: x86 4221:Type: vm ioctl 4222:Parameters: struct kvm_msr_filter 4223:Returns: 0 on success, < 0 on error 4224 4225:: 4226 4227 struct kvm_msr_filter_range { 4228 #define KVM_MSR_FILTER_READ (1 << 0) 4229 #define KVM_MSR_FILTER_WRITE (1 << 1) 4230 __u32 flags; 4231 __u32 nmsrs; /* number of msrs in bitmap */ 4232 __u32 base; /* MSR index the bitmap starts at */ 4233 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 4234 }; 4235 4236 #define KVM_MSR_FILTER_MAX_RANGES 16 4237 struct kvm_msr_filter { 4238 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 4239 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 4240 __u32 flags; 4241 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 4242 }; 4243 4244flags values for ``struct kvm_msr_filter_range``: 4245 4246``KVM_MSR_FILTER_READ`` 4247 4248 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 4249 indicates that read accesses should be denied, while a 1 indicates that 4250 a read for a particular MSR should be allowed regardless of the default 4251 filter action. 4252 4253``KVM_MSR_FILTER_WRITE`` 4254 4255 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 4256 indicates that write accesses should be denied, while a 1 indicates that 4257 a write for a particular MSR should be allowed regardless of the default 4258 filter action. 4259 4260flags values for ``struct kvm_msr_filter``: 4261 4262``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4263 4264 If no filter range matches an MSR index that is getting accessed, KVM will 4265 allow accesses to all MSRs by default. 4266 4267``KVM_MSR_FILTER_DEFAULT_DENY`` 4268 4269 If no filter range matches an MSR index that is getting accessed, KVM will 4270 deny accesses to all MSRs by default. 4271 4272This ioctl allows userspace to define up to 16 bitmaps of MSR ranges to deny 4273guest MSR accesses that would normally be allowed by KVM. If an MSR is not 4274covered by a specific range, the "default" filtering behavior applies. Each 4275bitmap range covers MSRs from [base .. base+nmsrs). 4276 4277If an MSR access is denied by userspace, the resulting KVM behavior depends on 4278whether or not KVM_CAP_X86_USER_SPACE_MSR's KVM_MSR_EXIT_REASON_FILTER is 4279enabled. If KVM_MSR_EXIT_REASON_FILTER is enabled, KVM will exit to userspace 4280on denied accesses, i.e. userspace effectively intercepts the MSR access. If 4281KVM_MSR_EXIT_REASON_FILTER is not enabled, KVM will inject a #GP into the guest 4282on denied accesses. Note, if an MSR access is denied during emulation of MSR 4283load/stores during VMX transitions, KVM ignores KVM_MSR_EXIT_REASON_FILTER. 4284See the below warning for full details. 4285 4286If an MSR access is allowed by userspace, KVM will emulate and/or virtualize 4287the access in accordance with the vCPU model. Note, KVM may still ultimately 4288inject a #GP if an access is allowed by userspace, e.g. if KVM doesn't support 4289the MSR, or to follow architectural behavior for the MSR. 4290 4291By default, KVM operates in KVM_MSR_FILTER_DEFAULT_ALLOW mode with no MSR range 4292filters. 4293 4294Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 4295filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 4296an error. 4297 4298.. warning:: 4299 MSR accesses that are side effects of instruction execution (emulated or 4300 native) are not filtered as hardware does not honor MSR bitmaps outside of 4301 RDMSR and WRMSR, and KVM mimics that behavior when emulating instructions 4302 to avoid pointless divergence from hardware. E.g. RDPID reads MSR_TSC_AUX, 4303 SYSENTER reads the SYSENTER MSRs, etc. 4304 4305 MSRs that are loaded/stored via dedicated VMCS fields are not filtered as 4306 part of VM-Enter/VM-Exit emulation. 4307 4308 MSRs that are loaded/store via VMX's load/store lists _are_ filtered as part 4309 of VM-Enter/VM-Exit emulation. If an MSR access is denied on VM-Enter, KVM 4310 synthesizes a consistency check VM-Exit(EXIT_REASON_MSR_LOAD_FAIL). If an 4311 MSR access is denied on VM-Exit, KVM synthesizes a VM-Abort. In short, KVM 4312 extends Intel's architectural list of MSRs that cannot be loaded/saved via 4313 the VM-Enter/VM-Exit MSR list. It is platform owner's responsibility to 4314 to communicate any such restrictions to their end users. 4315 4316 x2APIC MSR accesses cannot be filtered (KVM silently ignores filters that 4317 cover any x2APIC MSRs). 4318 4319Note, invoking this ioctl while a vCPU is running is inherently racy. However, 4320KVM does guarantee that vCPUs will see either the previous filter or the new 4321filter, e.g. MSRs with identical settings in both the old and new filter will 4322have deterministic behavior. 4323 4324Similarly, if userspace wishes to intercept on denied accesses, 4325KVM_MSR_EXIT_REASON_FILTER must be enabled before activating any filters, and 4326left enabled until after all filters are deactivated. Failure to do so may 4327result in KVM injecting a #GP instead of exiting to userspace. 4328 43294.98 KVM_CREATE_SPAPR_TCE_64 4330---------------------------- 4331 4332:Capability: KVM_CAP_SPAPR_TCE_64 4333:Architectures: powerpc 4334:Type: vm ioctl 4335:Parameters: struct kvm_create_spapr_tce_64 (in) 4336:Returns: file descriptor for manipulating the created TCE table 4337 4338This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 4339windows, described in 4.62 KVM_CREATE_SPAPR_TCE 4340 4341This capability uses extended struct in ioctl interface:: 4342 4343 /* for KVM_CAP_SPAPR_TCE_64 */ 4344 struct kvm_create_spapr_tce_64 { 4345 __u64 liobn; 4346 __u32 page_shift; 4347 __u32 flags; 4348 __u64 offset; /* in pages */ 4349 __u64 size; /* in pages */ 4350 }; 4351 4352The aim of extension is to support an additional bigger DMA window with 4353a variable page size. 4354KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 4355a bus offset of the corresponding DMA window, @size and @offset are numbers 4356of IOMMU pages. 4357 4358@flags are not used at the moment. 4359 4360The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 4361 43624.99 KVM_REINJECT_CONTROL 4363------------------------- 4364 4365:Capability: KVM_CAP_REINJECT_CONTROL 4366:Architectures: x86 4367:Type: vm ioctl 4368:Parameters: struct kvm_reinject_control (in) 4369:Returns: 0 on success, 4370 -EFAULT if struct kvm_reinject_control cannot be read, 4371 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 4372 4373i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 4374where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 4375vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 4376interrupt whenever there isn't a pending interrupt from i8254. 4377!reinject mode injects an interrupt as soon as a tick arrives. 4378 4379:: 4380 4381 struct kvm_reinject_control { 4382 __u8 pit_reinject; 4383 __u8 reserved[31]; 4384 }; 4385 4386pit_reinject = 0 (!reinject mode) is recommended, unless running an old 4387operating system that uses the PIT for timing (e.g. Linux 2.4.x). 4388 43894.100 KVM_PPC_CONFIGURE_V3_MMU 4390------------------------------ 4391 4392:Capability: KVM_CAP_PPC_MMU_RADIX or KVM_CAP_PPC_MMU_HASH_V3 4393:Architectures: ppc 4394:Type: vm ioctl 4395:Parameters: struct kvm_ppc_mmuv3_cfg (in) 4396:Returns: 0 on success, 4397 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 4398 -EINVAL if the configuration is invalid 4399 4400This ioctl controls whether the guest will use radix or HPT (hashed 4401page table) translation, and sets the pointer to the process table for 4402the guest. 4403 4404:: 4405 4406 struct kvm_ppc_mmuv3_cfg { 4407 __u64 flags; 4408 __u64 process_table; 4409 }; 4410 4411There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 4412KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 4413to use radix tree translation, and if clear, to use HPT translation. 4414KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 4415to be able to use the global TLB and SLB invalidation instructions; 4416if clear, the guest may not use these instructions. 4417 4418The process_table field specifies the address and size of the guest 4419process table, which is in the guest's space. This field is formatted 4420as the second doubleword of the partition table entry, as defined in 4421the Power ISA V3.00, Book III section 5.7.6.1. 4422 44234.101 KVM_PPC_GET_RMMU_INFO 4424--------------------------- 4425 4426:Capability: KVM_CAP_PPC_MMU_RADIX 4427:Architectures: ppc 4428:Type: vm ioctl 4429:Parameters: struct kvm_ppc_rmmu_info (out) 4430:Returns: 0 on success, 4431 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 4432 -EINVAL if no useful information can be returned 4433 4434This ioctl returns a structure containing two things: (a) a list 4435containing supported radix tree geometries, and (b) a list that maps 4436page sizes to put in the "AP" (actual page size) field for the tlbie 4437(TLB invalidate entry) instruction. 4438 4439:: 4440 4441 struct kvm_ppc_rmmu_info { 4442 struct kvm_ppc_radix_geom { 4443 __u8 page_shift; 4444 __u8 level_bits[4]; 4445 __u8 pad[3]; 4446 } geometries[8]; 4447 __u32 ap_encodings[8]; 4448 }; 4449 4450The geometries[] field gives up to 8 supported geometries for the 4451radix page table, in terms of the log base 2 of the smallest page 4452size, and the number of bits indexed at each level of the tree, from 4453the PTE level up to the PGD level in that order. Any unused entries 4454will have 0 in the page_shift field. 4455 4456The ap_encodings gives the supported page sizes and their AP field 4457encodings, encoded with the AP value in the top 3 bits and the log 4458base 2 of the page size in the bottom 6 bits. 4459 44604.102 KVM_PPC_RESIZE_HPT_PREPARE 4461-------------------------------- 4462 4463:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4464:Architectures: powerpc 4465:Type: vm ioctl 4466:Parameters: struct kvm_ppc_resize_hpt (in) 4467:Returns: 0 on successful completion, 4468 >0 if a new HPT is being prepared, the value is an estimated 4469 number of milliseconds until preparation is complete, 4470 -EFAULT if struct kvm_reinject_control cannot be read, 4471 -EINVAL if the supplied shift or flags are invalid, 4472 -ENOMEM if unable to allocate the new HPT, 4473 4474Used to implement the PAPR extension for runtime resizing of a guest's 4475Hashed Page Table (HPT). Specifically this starts, stops or monitors 4476the preparation of a new potential HPT for the guest, essentially 4477implementing the H_RESIZE_HPT_PREPARE hypercall. 4478 4479:: 4480 4481 struct kvm_ppc_resize_hpt { 4482 __u64 flags; 4483 __u32 shift; 4484 __u32 pad; 4485 }; 4486 4487If called with shift > 0 when there is no pending HPT for the guest, 4488this begins preparation of a new pending HPT of size 2^(shift) bytes. 4489It then returns a positive integer with the estimated number of 4490milliseconds until preparation is complete. 4491 4492If called when there is a pending HPT whose size does not match that 4493requested in the parameters, discards the existing pending HPT and 4494creates a new one as above. 4495 4496If called when there is a pending HPT of the size requested, will: 4497 4498 * If preparation of the pending HPT is already complete, return 0 4499 * If preparation of the pending HPT has failed, return an error 4500 code, then discard the pending HPT. 4501 * If preparation of the pending HPT is still in progress, return an 4502 estimated number of milliseconds until preparation is complete. 4503 4504If called with shift == 0, discards any currently pending HPT and 4505returns 0 (i.e. cancels any in-progress preparation). 4506 4507flags is reserved for future expansion, currently setting any bits in 4508flags will result in an -EINVAL. 4509 4510Normally this will be called repeatedly with the same parameters until 4511it returns <= 0. The first call will initiate preparation, subsequent 4512ones will monitor preparation until it completes or fails. 4513 45144.103 KVM_PPC_RESIZE_HPT_COMMIT 4515------------------------------- 4516 4517:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4518:Architectures: powerpc 4519:Type: vm ioctl 4520:Parameters: struct kvm_ppc_resize_hpt (in) 4521:Returns: 0 on successful completion, 4522 -EFAULT if struct kvm_reinject_control cannot be read, 4523 -EINVAL if the supplied shift or flags are invalid, 4524 -ENXIO is there is no pending HPT, or the pending HPT doesn't 4525 have the requested size, 4526 -EBUSY if the pending HPT is not fully prepared, 4527 -ENOSPC if there was a hash collision when moving existing 4528 HPT entries to the new HPT, 4529 -EIO on other error conditions 4530 4531Used to implement the PAPR extension for runtime resizing of a guest's 4532Hashed Page Table (HPT). Specifically this requests that the guest be 4533transferred to working with the new HPT, essentially implementing the 4534H_RESIZE_HPT_COMMIT hypercall. 4535 4536:: 4537 4538 struct kvm_ppc_resize_hpt { 4539 __u64 flags; 4540 __u32 shift; 4541 __u32 pad; 4542 }; 4543 4544This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 4545returned 0 with the same parameters. In other cases 4546KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 4547-EBUSY, though others may be possible if the preparation was started, 4548but failed). 4549 4550This will have undefined effects on the guest if it has not already 4551placed itself in a quiescent state where no vcpu will make MMU enabled 4552memory accesses. 4553 4554On successful completion, the pending HPT will become the guest's active 4555HPT and the previous HPT will be discarded. 4556 4557On failure, the guest will still be operating on its previous HPT. 4558 45594.104 KVM_X86_GET_MCE_CAP_SUPPORTED 4560----------------------------------- 4561 4562:Capability: KVM_CAP_MCE 4563:Architectures: x86 4564:Type: system ioctl 4565:Parameters: u64 mce_cap (out) 4566:Returns: 0 on success, -1 on error 4567 4568Returns supported MCE capabilities. The u64 mce_cap parameter 4569has the same format as the MSR_IA32_MCG_CAP register. Supported 4570capabilities will have the corresponding bits set. 4571 45724.105 KVM_X86_SETUP_MCE 4573----------------------- 4574 4575:Capability: KVM_CAP_MCE 4576:Architectures: x86 4577:Type: vcpu ioctl 4578:Parameters: u64 mcg_cap (in) 4579:Returns: 0 on success, 4580 -EFAULT if u64 mcg_cap cannot be read, 4581 -EINVAL if the requested number of banks is invalid, 4582 -EINVAL if requested MCE capability is not supported. 4583 4584Initializes MCE support for use. The u64 mcg_cap parameter 4585has the same format as the MSR_IA32_MCG_CAP register and 4586specifies which capabilities should be enabled. The maximum 4587supported number of error-reporting banks can be retrieved when 4588checking for KVM_CAP_MCE. The supported capabilities can be 4589retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 4590 45914.106 KVM_X86_SET_MCE 4592--------------------- 4593 4594:Capability: KVM_CAP_MCE 4595:Architectures: x86 4596:Type: vcpu ioctl 4597:Parameters: struct kvm_x86_mce (in) 4598:Returns: 0 on success, 4599 -EFAULT if struct kvm_x86_mce cannot be read, 4600 -EINVAL if the bank number is invalid, 4601 -EINVAL if VAL bit is not set in status field. 4602 4603Inject a machine check error (MCE) into the guest. The input 4604parameter is:: 4605 4606 struct kvm_x86_mce { 4607 __u64 status; 4608 __u64 addr; 4609 __u64 misc; 4610 __u64 mcg_status; 4611 __u8 bank; 4612 __u8 pad1[7]; 4613 __u64 pad2[3]; 4614 }; 4615 4616If the MCE being reported is an uncorrected error, KVM will 4617inject it as an MCE exception into the guest. If the guest 4618MCG_STATUS register reports that an MCE is in progress, KVM 4619causes an KVM_EXIT_SHUTDOWN vmexit. 4620 4621Otherwise, if the MCE is a corrected error, KVM will just 4622store it in the corresponding bank (provided this bank is 4623not holding a previously reported uncorrected error). 4624 46254.107 KVM_S390_GET_CMMA_BITS 4626---------------------------- 4627 4628:Capability: KVM_CAP_S390_CMMA_MIGRATION 4629:Architectures: s390 4630:Type: vm ioctl 4631:Parameters: struct kvm_s390_cmma_log (in, out) 4632:Returns: 0 on success, a negative value on error 4633 4634Errors: 4635 4636 ====== ============================================================= 4637 ENOMEM not enough memory can be allocated to complete the task 4638 ENXIO if CMMA is not enabled 4639 EINVAL if KVM_S390_CMMA_PEEK is not set but migration mode was not enabled 4640 EINVAL if KVM_S390_CMMA_PEEK is not set but dirty tracking has been 4641 disabled (and thus migration mode was automatically disabled) 4642 EFAULT if the userspace address is invalid or if no page table is 4643 present for the addresses (e.g. when using hugepages). 4644 ====== ============================================================= 4645 4646This ioctl is used to get the values of the CMMA bits on the s390 4647architecture. It is meant to be used in two scenarios: 4648 4649- During live migration to save the CMMA values. Live migration needs 4650 to be enabled via the KVM_REQ_START_MIGRATION VM property. 4651- To non-destructively peek at the CMMA values, with the flag 4652 KVM_S390_CMMA_PEEK set. 4653 4654The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 4655values are written to a buffer whose location is indicated via the "values" 4656member in the kvm_s390_cmma_log struct. The values in the input struct are 4657also updated as needed. 4658 4659Each CMMA value takes up one byte. 4660 4661:: 4662 4663 struct kvm_s390_cmma_log { 4664 __u64 start_gfn; 4665 __u32 count; 4666 __u32 flags; 4667 union { 4668 __u64 remaining; 4669 __u64 mask; 4670 }; 4671 __u64 values; 4672 }; 4673 4674start_gfn is the number of the first guest frame whose CMMA values are 4675to be retrieved, 4676 4677count is the length of the buffer in bytes, 4678 4679values points to the buffer where the result will be written to. 4680 4681If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 4682KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 4683other ioctls. 4684 4685The result is written in the buffer pointed to by the field values, and 4686the values of the input parameter are updated as follows. 4687 4688Depending on the flags, different actions are performed. The only 4689supported flag so far is KVM_S390_CMMA_PEEK. 4690 4691The default behaviour if KVM_S390_CMMA_PEEK is not set is: 4692start_gfn will indicate the first page frame whose CMMA bits were dirty. 4693It is not necessarily the same as the one passed as input, as clean pages 4694are skipped. 4695 4696count will indicate the number of bytes actually written in the buffer. 4697It can (and very often will) be smaller than the input value, since the 4698buffer is only filled until 16 bytes of clean values are found (which 4699are then not copied in the buffer). Since a CMMA migration block needs 4700the base address and the length, for a total of 16 bytes, we will send 4701back some clean data if there is some dirty data afterwards, as long as 4702the size of the clean data does not exceed the size of the header. This 4703allows to minimize the amount of data to be saved or transferred over 4704the network at the expense of more roundtrips to userspace. The next 4705invocation of the ioctl will skip over all the clean values, saving 4706potentially more than just the 16 bytes we found. 4707 4708If KVM_S390_CMMA_PEEK is set: 4709the existing storage attributes are read even when not in migration 4710mode, and no other action is performed; 4711 4712the output start_gfn will be equal to the input start_gfn, 4713 4714the output count will be equal to the input count, except if the end of 4715memory has been reached. 4716 4717In both cases: 4718the field "remaining" will indicate the total number of dirty CMMA values 4719still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 4720not enabled. 4721 4722mask is unused. 4723 4724values points to the userspace buffer where the result will be stored. 4725 47264.108 KVM_S390_SET_CMMA_BITS 4727---------------------------- 4728 4729:Capability: KVM_CAP_S390_CMMA_MIGRATION 4730:Architectures: s390 4731:Type: vm ioctl 4732:Parameters: struct kvm_s390_cmma_log (in) 4733:Returns: 0 on success, a negative value on error 4734 4735This ioctl is used to set the values of the CMMA bits on the s390 4736architecture. It is meant to be used during live migration to restore 4737the CMMA values, but there are no restrictions on its use. 4738The ioctl takes parameters via the kvm_s390_cmma_values struct. 4739Each CMMA value takes up one byte. 4740 4741:: 4742 4743 struct kvm_s390_cmma_log { 4744 __u64 start_gfn; 4745 __u32 count; 4746 __u32 flags; 4747 union { 4748 __u64 remaining; 4749 __u64 mask; 4750 }; 4751 __u64 values; 4752 }; 4753 4754start_gfn indicates the starting guest frame number, 4755 4756count indicates how many values are to be considered in the buffer, 4757 4758flags is not used and must be 0. 4759 4760mask indicates which PGSTE bits are to be considered. 4761 4762remaining is not used. 4763 4764values points to the buffer in userspace where to store the values. 4765 4766This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4767complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4768the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 4769if the flags field was not 0, with -EFAULT if the userspace address is 4770invalid, if invalid pages are written to (e.g. after the end of memory) 4771or if no page table is present for the addresses (e.g. when using 4772hugepages). 4773 47744.109 KVM_PPC_GET_CPU_CHAR 4775-------------------------- 4776 4777:Capability: KVM_CAP_PPC_GET_CPU_CHAR 4778:Architectures: powerpc 4779:Type: vm ioctl 4780:Parameters: struct kvm_ppc_cpu_char (out) 4781:Returns: 0 on successful completion, 4782 -EFAULT if struct kvm_ppc_cpu_char cannot be written 4783 4784This ioctl gives userspace information about certain characteristics 4785of the CPU relating to speculative execution of instructions and 4786possible information leakage resulting from speculative execution (see 4787CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 4788returned in struct kvm_ppc_cpu_char, which looks like this:: 4789 4790 struct kvm_ppc_cpu_char { 4791 __u64 character; /* characteristics of the CPU */ 4792 __u64 behaviour; /* recommended software behaviour */ 4793 __u64 character_mask; /* valid bits in character */ 4794 __u64 behaviour_mask; /* valid bits in behaviour */ 4795 }; 4796 4797For extensibility, the character_mask and behaviour_mask fields 4798indicate which bits of character and behaviour have been filled in by 4799the kernel. If the set of defined bits is extended in future then 4800userspace will be able to tell whether it is running on a kernel that 4801knows about the new bits. 4802 4803The character field describes attributes of the CPU which can help 4804with preventing inadvertent information disclosure - specifically, 4805whether there is an instruction to flash-invalidate the L1 data cache 4806(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 4807to a mode where entries can only be used by the thread that created 4808them, whether the bcctr[l] instruction prevents speculation, and 4809whether a speculation barrier instruction (ori 31,31,0) is provided. 4810 4811The behaviour field describes actions that software should take to 4812prevent inadvertent information disclosure, and thus describes which 4813vulnerabilities the hardware is subject to; specifically whether the 4814L1 data cache should be flushed when returning to user mode from the 4815kernel, and whether a speculation barrier should be placed between an 4816array bounds check and the array access. 4817 4818These fields use the same bit definitions as the new 4819H_GET_CPU_CHARACTERISTICS hypercall. 4820 48214.110 KVM_MEMORY_ENCRYPT_OP 4822--------------------------- 4823 4824:Capability: basic 4825:Architectures: x86 4826:Type: vm ioctl, vcpu ioctl 4827:Parameters: an opaque platform specific structure (in/out) 4828:Returns: 0 on success; -1 on error 4829 4830If the platform supports creating encrypted VMs then this ioctl can be used 4831for issuing platform-specific memory encryption commands to manage those 4832encrypted VMs. 4833 4834Currently, this ioctl is used for issuing both Secure Encrypted Virtualization 4835(SEV) commands on AMD Processors and Trusted Domain Extensions (TDX) commands 4836on Intel Processors. The detailed commands are defined in 4837Documentation/virt/kvm/x86/amd-memory-encryption.rst and 4838Documentation/virt/kvm/x86/intel-tdx.rst. 4839 48404.111 KVM_MEMORY_ENCRYPT_REG_REGION 4841----------------------------------- 4842 4843:Capability: basic 4844:Architectures: x86 4845:Type: system 4846:Parameters: struct kvm_enc_region (in) 4847:Returns: 0 on success; -1 on error 4848 4849This ioctl can be used to register a guest memory region which may 4850contain encrypted data (e.g. guest RAM, SMRAM etc). 4851 4852It is used in the SEV-enabled guest. When encryption is enabled, a guest 4853memory region may contain encrypted data. The SEV memory encryption 4854engine uses a tweak such that two identical plaintext pages, each at 4855different locations will have differing ciphertexts. So swapping or 4856moving ciphertext of those pages will not result in plaintext being 4857swapped. So relocating (or migrating) physical backing pages for the SEV 4858guest will require some additional steps. 4859 4860Note: The current SEV key management spec does not provide commands to 4861swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 4862memory region registered with the ioctl. 4863 48644.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4865------------------------------------- 4866 4867:Capability: basic 4868:Architectures: x86 4869:Type: system 4870:Parameters: struct kvm_enc_region (in) 4871:Returns: 0 on success; -1 on error 4872 4873This ioctl can be used to unregister the guest memory region registered 4874with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 4875 48764.113 KVM_HYPERV_EVENTFD 4877------------------------ 4878 4879:Capability: KVM_CAP_HYPERV_EVENTFD 4880:Architectures: x86 4881:Type: vm ioctl 4882:Parameters: struct kvm_hyperv_eventfd (in) 4883 4884This ioctl (un)registers an eventfd to receive notifications from the guest on 4885the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without 4886causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number 4887(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit. 4888 4889:: 4890 4891 struct kvm_hyperv_eventfd { 4892 __u32 conn_id; 4893 __s32 fd; 4894 __u32 flags; 4895 __u32 padding[3]; 4896 }; 4897 4898The conn_id field should fit within 24 bits:: 4899 4900 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff 4901 4902The acceptable values for the flags field are:: 4903 4904 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0) 4905 4906:Returns: 0 on success, 4907 -EINVAL if conn_id or flags is outside the allowed range, 4908 -ENOENT on deassign if the conn_id isn't registered, 4909 -EEXIST on assign if the conn_id is already registered 4910 49114.114 KVM_GET_NESTED_STATE 4912-------------------------- 4913 4914:Capability: KVM_CAP_NESTED_STATE 4915:Architectures: x86 4916:Type: vcpu ioctl 4917:Parameters: struct kvm_nested_state (in/out) 4918:Returns: 0 on success, -1 on error 4919 4920Errors: 4921 4922 ===== ============================================================= 4923 E2BIG the total state size exceeds the value of 'size' specified by 4924 the user; the size required will be written into size. 4925 ===== ============================================================= 4926 4927:: 4928 4929 struct kvm_nested_state { 4930 __u16 flags; 4931 __u16 format; 4932 __u32 size; 4933 4934 union { 4935 struct kvm_vmx_nested_state_hdr vmx; 4936 struct kvm_svm_nested_state_hdr svm; 4937 4938 /* Pad the header to 128 bytes. */ 4939 __u8 pad[120]; 4940 } hdr; 4941 4942 union { 4943 struct kvm_vmx_nested_state_data vmx[0]; 4944 struct kvm_svm_nested_state_data svm[0]; 4945 } data; 4946 }; 4947 4948 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001 4949 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002 4950 #define KVM_STATE_NESTED_EVMCS 0x00000004 4951 4952 #define KVM_STATE_NESTED_FORMAT_VMX 0 4953 #define KVM_STATE_NESTED_FORMAT_SVM 1 4954 4955 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 4956 #define KVM_STATE_NESTED_SVM_VMCB_SIZE 0x1000 4957 4958 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001 4959 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002 4960 4961 #define KVM_STATE_NESTED_GIF_SET 0x00000100 4962 4963 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001 4964 4965 struct kvm_vmx_nested_state_hdr { 4966 __u64 vmxon_pa; 4967 __u64 vmcs12_pa; 4968 4969 struct { 4970 __u16 flags; 4971 } smm; 4972 4973 __u32 flags; 4974 __u64 preemption_timer_deadline; 4975 }; 4976 4977 struct kvm_svm_nested_state_hdr { 4978 __u64 vmcb_pa; 4979 __u64 gpat; 4980 }; 4981 4982 struct kvm_vmx_nested_state_data { 4983 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4984 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4985 }; 4986 4987 struct kvm_svm_nested_state_data { 4988 __u8 vmcb12[KVM_STATE_NESTED_SVM_VMCB_SIZE]; 4989 }; 4990 4991This ioctl copies the vcpu's nested virtualization state from the kernel to 4992userspace. 4993 4994The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE 4995to the KVM_CHECK_EXTENSION ioctl(). 4996 49974.115 KVM_SET_NESTED_STATE 4998-------------------------- 4999 5000:Capability: KVM_CAP_NESTED_STATE 5001:Architectures: x86 5002:Type: vcpu ioctl 5003:Parameters: struct kvm_nested_state (in) 5004:Returns: 0 on success, -1 on error 5005 5006This copies the vcpu's kvm_nested_state struct from userspace to the kernel. 5007For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE. 5008 50094.116 KVM_(UN)REGISTER_COALESCED_MMIO 5010------------------------------------- 5011 5012:Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio) 5013 KVM_CAP_COALESCED_PIO (for coalesced pio) 5014:Architectures: all 5015:Type: vm ioctl 5016:Parameters: struct kvm_coalesced_mmio_zone 5017:Returns: 0 on success, < 0 on error 5018 5019Coalesced I/O is a performance optimization that defers hardware 5020register write emulation so that userspace exits are avoided. It is 5021typically used to reduce the overhead of emulating frequently accessed 5022hardware registers. 5023 5024When a hardware register is configured for coalesced I/O, write accesses 5025do not exit to userspace and their value is recorded in a ring buffer 5026that is shared between kernel and userspace. 5027 5028Coalesced I/O is used if one or more write accesses to a hardware 5029register can be deferred until a read or a write to another hardware 5030register on the same device. This last access will cause a vmexit and 5031userspace will process accesses from the ring buffer before emulating 5032it. That will avoid exiting to userspace on repeated writes. 5033 5034Coalesced pio is based on coalesced mmio. There is little difference 5035between coalesced mmio and pio except that coalesced pio records accesses 5036to I/O ports. 5037 50384.117 KVM_CLEAR_DIRTY_LOG 5039------------------------- 5040 5041:Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 5042:Architectures: x86, arm64, mips 5043:Type: vm ioctl 5044:Parameters: struct kvm_clear_dirty_log (in) 5045:Returns: 0 on success, -1 on error 5046 5047:: 5048 5049 /* for KVM_CLEAR_DIRTY_LOG */ 5050 struct kvm_clear_dirty_log { 5051 __u32 slot; 5052 __u32 num_pages; 5053 __u64 first_page; 5054 union { 5055 void __user *dirty_bitmap; /* one bit per page */ 5056 __u64 padding; 5057 }; 5058 }; 5059 5060The ioctl clears the dirty status of pages in a memory slot, according to 5061the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap 5062field. Bit 0 of the bitmap corresponds to page "first_page" in the 5063memory slot, and num_pages is the size in bits of the input bitmap. 5064first_page must be a multiple of 64; num_pages must also be a multiple of 506564 unless first_page + num_pages is the size of the memory slot. For each 5066bit that is set in the input bitmap, the corresponding page is marked "clean" 5067in KVM's dirty bitmap, and dirty tracking is re-enabled for that page 5068(for example via write-protection, or by clearing the dirty bit in 5069a page table entry). 5070 5071If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 5072the address space for which you want to clear the dirty status. See 5073KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 5074 5075This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 5076is enabled; for more information, see the description of the capability. 5077However, it can always be used as long as KVM_CHECK_EXTENSION confirms 5078that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present. 5079 50804.118 KVM_GET_SUPPORTED_HV_CPUID 5081-------------------------------- 5082 5083:Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system) 5084:Architectures: x86 5085:Type: system ioctl, vcpu ioctl 5086:Parameters: struct kvm_cpuid2 (in/out) 5087:Returns: 0 on success, -1 on error 5088 5089:: 5090 5091 struct kvm_cpuid2 { 5092 __u32 nent; 5093 __u32 padding; 5094 struct kvm_cpuid_entry2 entries[0]; 5095 }; 5096 5097 struct kvm_cpuid_entry2 { 5098 __u32 function; 5099 __u32 index; 5100 __u32 flags; 5101 __u32 eax; 5102 __u32 ebx; 5103 __u32 ecx; 5104 __u32 edx; 5105 __u32 padding[3]; 5106 }; 5107 5108This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in 5109KVM. Userspace can use the information returned by this ioctl to construct 5110cpuid information presented to guests consuming Hyper-V enlightenments (e.g. 5111Windows or Hyper-V guests). 5112 5113CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level 5114Functional Specification (TLFS). These leaves can't be obtained with 5115KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature 5116leaves (0x40000000, 0x40000001). 5117 5118Currently, the following list of CPUID leaves are returned: 5119 5120 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 5121 - HYPERV_CPUID_INTERFACE 5122 - HYPERV_CPUID_VERSION 5123 - HYPERV_CPUID_FEATURES 5124 - HYPERV_CPUID_ENLIGHTMENT_INFO 5125 - HYPERV_CPUID_IMPLEMENT_LIMITS 5126 - HYPERV_CPUID_NESTED_FEATURES 5127 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS 5128 - HYPERV_CPUID_SYNDBG_INTERFACE 5129 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 5130 5131Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure 5132with the 'nent' field indicating the number of entries in the variable-size 5133array 'entries'. If the number of entries is too low to describe all Hyper-V 5134feature leaves, an error (E2BIG) is returned. If the number is more or equal 5135to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the 5136number of valid entries in the 'entries' array, which is then filled. 5137 5138'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved, 5139userspace should not expect to get any particular value there. 5140 5141Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike 5142system ioctl which exposes all supported feature bits unconditionally, vcpu 5143version has the following quirks: 5144 5145- HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED 5146 feature bit are only exposed when Enlightened VMCS was previously enabled 5147 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS). 5148- HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC. 5149 (presumes KVM_CREATE_IRQCHIP has already been called). 5150 51514.119 KVM_ARM_VCPU_FINALIZE 5152--------------------------- 5153 5154:Architectures: arm64 5155:Type: vcpu ioctl 5156:Parameters: int feature (in) 5157:Returns: 0 on success, -1 on error 5158 5159Errors: 5160 5161 ====== ============================================================== 5162 EPERM feature not enabled, needs configuration, or already finalized 5163 EINVAL feature unknown or not present 5164 ====== ============================================================== 5165 5166Recognised values for feature: 5167 5168 ===== =========================================== 5169 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE) 5170 ===== =========================================== 5171 5172Finalizes the configuration of the specified vcpu feature. 5173 5174The vcpu must already have been initialised, enabling the affected feature, by 5175means of a successful :ref:`KVM_ARM_VCPU_INIT <KVM_ARM_VCPU_INIT>` call with the 5176appropriate flag set in features[]. 5177 5178For affected vcpu features, this is a mandatory step that must be performed 5179before the vcpu is fully usable. 5180 5181Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be 5182configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration 5183that should be performed and how to do it are feature-dependent. 5184 5185Other calls that depend on a particular feature being finalized, such as 5186KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with 5187-EPERM unless the feature has already been finalized by means of a 5188KVM_ARM_VCPU_FINALIZE call. 5189 5190See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization 5191using this ioctl. 5192 51934.120 KVM_SET_PMU_EVENT_FILTER 5194------------------------------ 5195 5196:Capability: KVM_CAP_PMU_EVENT_FILTER 5197:Architectures: x86 5198:Type: vm ioctl 5199:Parameters: struct kvm_pmu_event_filter (in) 5200:Returns: 0 on success, -1 on error 5201 5202Errors: 5203 5204 ====== ============================================================ 5205 EFAULT args[0] cannot be accessed 5206 EINVAL args[0] contains invalid data in the filter or filter events 5207 E2BIG nevents is too large 5208 EBUSY not enough memory to allocate the filter 5209 ====== ============================================================ 5210 5211:: 5212 5213 struct kvm_pmu_event_filter { 5214 __u32 action; 5215 __u32 nevents; 5216 __u32 fixed_counter_bitmap; 5217 __u32 flags; 5218 __u32 pad[4]; 5219 __u64 events[0]; 5220 }; 5221 5222This ioctl restricts the set of PMU events the guest can program by limiting 5223which event select and unit mask combinations are permitted. 5224 5225The argument holds a list of filter events which will be allowed or denied. 5226 5227Filter events only control general purpose counters; fixed purpose counters 5228are controlled by the fixed_counter_bitmap. 5229 5230Valid values for 'flags':: 5231 5232``0`` 5233 5234To use this mode, clear the 'flags' field. 5235 5236In this mode each event will contain an event select + unit mask. 5237 5238When the guest attempts to program the PMU the guest's event select + 5239unit mask is compared against the filter events to determine whether the 5240guest should have access. 5241 5242``KVM_PMU_EVENT_FLAG_MASKED_EVENTS`` 5243:Capability: KVM_CAP_PMU_EVENT_MASKED_EVENTS 5244 5245In this mode each filter event will contain an event select, mask, match, and 5246exclude value. To encode a masked event use:: 5247 5248 KVM_PMU_ENCODE_MASKED_ENTRY() 5249 5250An encoded event will follow this layout:: 5251 5252 Bits Description 5253 ---- ----------- 5254 7:0 event select (low bits) 5255 15:8 umask match 5256 31:16 unused 5257 35:32 event select (high bits) 5258 36:54 unused 5259 55 exclude bit 5260 63:56 umask mask 5261 5262When the guest attempts to program the PMU, these steps are followed in 5263determining if the guest should have access: 5264 5265 1. Match the event select from the guest against the filter events. 5266 2. If a match is found, match the guest's unit mask to the mask and match 5267 values of the included filter events. 5268 I.e. (unit mask & mask) == match && !exclude. 5269 3. If a match is found, match the guest's unit mask to the mask and match 5270 values of the excluded filter events. 5271 I.e. (unit mask & mask) == match && exclude. 5272 4. 5273 a. If an included match is found and an excluded match is not found, filter 5274 the event. 5275 b. For everything else, do not filter the event. 5276 5. 5277 a. If the event is filtered and it's an allow list, allow the guest to 5278 program the event. 5279 b. If the event is filtered and it's a deny list, do not allow the guest to 5280 program the event. 5281 5282When setting a new pmu event filter, -EINVAL will be returned if any of the 5283unused fields are set or if any of the high bits (35:32) in the event 5284select are set when called on Intel. 5285 5286Valid values for 'action':: 5287 5288 #define KVM_PMU_EVENT_ALLOW 0 5289 #define KVM_PMU_EVENT_DENY 1 5290 5291Via this API, KVM userspace can also control the behavior of the VM's fixed 5292counters (if any) by configuring the "action" and "fixed_counter_bitmap" fields. 5293 5294Specifically, KVM follows the following pseudo-code when determining whether to 5295allow the guest FixCtr[i] to count its pre-defined fixed event:: 5296 5297 FixCtr[i]_is_allowed = (action == ALLOW) && (bitmap & BIT(i)) || 5298 (action == DENY) && !(bitmap & BIT(i)); 5299 FixCtr[i]_is_denied = !FixCtr[i]_is_allowed; 5300 5301KVM always consumes fixed_counter_bitmap, it's userspace's responsibility to 5302ensure fixed_counter_bitmap is set correctly, e.g. if userspace wants to define 5303a filter that only affects general purpose counters. 5304 5305Note, the "events" field also applies to fixed counters' hardcoded event_select 5306and unit_mask values. "fixed_counter_bitmap" has higher priority than "events" 5307if there is a contradiction between the two. 5308 53094.121 KVM_PPC_SVM_OFF 5310--------------------- 5311 5312:Capability: basic 5313:Architectures: powerpc 5314:Type: vm ioctl 5315:Parameters: none 5316:Returns: 0 on successful completion, 5317 5318Errors: 5319 5320 ====== ================================================================ 5321 EINVAL if ultravisor failed to terminate the secure guest 5322 ENOMEM if hypervisor failed to allocate new radix page tables for guest 5323 ====== ================================================================ 5324 5325This ioctl is used to turn off the secure mode of the guest or transition 5326the guest from secure mode to normal mode. This is invoked when the guest 5327is reset. This has no effect if called for a normal guest. 5328 5329This ioctl issues an ultravisor call to terminate the secure guest, 5330unpins the VPA pages and releases all the device pages that are used to 5331track the secure pages by hypervisor. 5332 53334.122 KVM_S390_NORMAL_RESET 5334--------------------------- 5335 5336:Capability: KVM_CAP_S390_VCPU_RESETS 5337:Architectures: s390 5338:Type: vcpu ioctl 5339:Parameters: none 5340:Returns: 0 5341 5342This ioctl resets VCPU registers and control structures according to 5343the cpu reset definition in the POP (Principles Of Operation). 5344 53454.123 KVM_S390_INITIAL_RESET 5346---------------------------- 5347 5348:Capability: basic 5349:Architectures: s390 5350:Type: vcpu ioctl 5351:Parameters: none 5352:Returns: 0 5353 5354This ioctl resets VCPU registers and control structures according to 5355the initial cpu reset definition in the POP. However, the cpu is not 5356put into ESA mode. This reset is a superset of the normal reset. 5357 53584.124 KVM_S390_CLEAR_RESET 5359-------------------------- 5360 5361:Capability: KVM_CAP_S390_VCPU_RESETS 5362:Architectures: s390 5363:Type: vcpu ioctl 5364:Parameters: none 5365:Returns: 0 5366 5367This ioctl resets VCPU registers and control structures according to 5368the clear cpu reset definition in the POP. However, the cpu is not put 5369into ESA mode. This reset is a superset of the initial reset. 5370 5371 53724.125 KVM_S390_PV_COMMAND 5373------------------------- 5374 5375:Capability: KVM_CAP_S390_PROTECTED 5376:Architectures: s390 5377:Type: vm ioctl 5378:Parameters: struct kvm_pv_cmd 5379:Returns: 0 on success, < 0 on error 5380 5381:: 5382 5383 struct kvm_pv_cmd { 5384 __u32 cmd; /* Command to be executed */ 5385 __u16 rc; /* Ultravisor return code */ 5386 __u16 rrc; /* Ultravisor return reason code */ 5387 __u64 data; /* Data or address */ 5388 __u32 flags; /* flags for future extensions. Must be 0 for now */ 5389 __u32 reserved[3]; 5390 }; 5391 5392**Ultravisor return codes** 5393The Ultravisor return (reason) codes are provided by the kernel if a 5394Ultravisor call has been executed to achieve the results expected by 5395the command. Therefore they are independent of the IOCTL return 5396code. If KVM changes `rc`, its value will always be greater than 0 5397hence setting it to 0 before issuing a PV command is advised to be 5398able to detect a change of `rc`. 5399 5400**cmd values:** 5401 5402KVM_PV_ENABLE 5403 Allocate memory and register the VM with the Ultravisor, thereby 5404 donating memory to the Ultravisor that will become inaccessible to 5405 KVM. All existing CPUs are converted to protected ones. After this 5406 command has succeeded, any CPU added via hotplug will become 5407 protected during its creation as well. 5408 5409 Errors: 5410 5411 ===== ============================= 5412 EINTR an unmasked signal is pending 5413 ===== ============================= 5414 5415KVM_PV_DISABLE 5416 Deregister the VM from the Ultravisor and reclaim the memory that had 5417 been donated to the Ultravisor, making it usable by the kernel again. 5418 All registered VCPUs are converted back to non-protected ones. If a 5419 previous protected VM had been prepared for asynchronous teardown with 5420 KVM_PV_ASYNC_CLEANUP_PREPARE and not subsequently torn down with 5421 KVM_PV_ASYNC_CLEANUP_PERFORM, it will be torn down in this call 5422 together with the current protected VM. 5423 5424KVM_PV_VM_SET_SEC_PARMS 5425 Pass the image header from VM memory to the Ultravisor in 5426 preparation of image unpacking and verification. 5427 5428KVM_PV_VM_UNPACK 5429 Unpack (protect and decrypt) a page of the encrypted boot image. 5430 5431KVM_PV_VM_VERIFY 5432 Verify the integrity of the unpacked image. Only if this succeeds, 5433 KVM is allowed to start protected VCPUs. 5434 5435KVM_PV_INFO 5436 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5437 5438 Presents an API that provides Ultravisor related data to userspace 5439 via subcommands. len_max is the size of the user space buffer, 5440 len_written is KVM's indication of how much bytes of that buffer 5441 were actually written to. len_written can be used to determine the 5442 valid fields if more response fields are added in the future. 5443 5444 :: 5445 5446 enum pv_cmd_info_id { 5447 KVM_PV_INFO_VM, 5448 KVM_PV_INFO_DUMP, 5449 }; 5450 5451 struct kvm_s390_pv_info_header { 5452 __u32 id; 5453 __u32 len_max; 5454 __u32 len_written; 5455 __u32 reserved; 5456 }; 5457 5458 struct kvm_s390_pv_info { 5459 struct kvm_s390_pv_info_header header; 5460 struct kvm_s390_pv_info_dump dump; 5461 struct kvm_s390_pv_info_vm vm; 5462 }; 5463 5464**subcommands:** 5465 5466 KVM_PV_INFO_VM 5467 This subcommand provides basic Ultravisor information for PV 5468 hosts. These values are likely also exported as files in the sysfs 5469 firmware UV query interface but they are more easily available to 5470 programs in this API. 5471 5472 The installed calls and feature_indication members provide the 5473 installed UV calls and the UV's other feature indications. 5474 5475 The max_* members provide information about the maximum number of PV 5476 vcpus, PV guests and PV guest memory size. 5477 5478 :: 5479 5480 struct kvm_s390_pv_info_vm { 5481 __u64 inst_calls_list[4]; 5482 __u64 max_cpus; 5483 __u64 max_guests; 5484 __u64 max_guest_addr; 5485 __u64 feature_indication; 5486 }; 5487 5488 5489 KVM_PV_INFO_DUMP 5490 This subcommand provides information related to dumping PV guests. 5491 5492 :: 5493 5494 struct kvm_s390_pv_info_dump { 5495 __u64 dump_cpu_buffer_len; 5496 __u64 dump_config_mem_buffer_per_1m; 5497 __u64 dump_config_finalize_len; 5498 }; 5499 5500KVM_PV_DUMP 5501 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5502 5503 Presents an API that provides calls which facilitate dumping a 5504 protected VM. 5505 5506 :: 5507 5508 struct kvm_s390_pv_dmp { 5509 __u64 subcmd; 5510 __u64 buff_addr; 5511 __u64 buff_len; 5512 __u64 gaddr; /* For dump storage state */ 5513 }; 5514 5515 **subcommands:** 5516 5517 KVM_PV_DUMP_INIT 5518 Initializes the dump process of a protected VM. If this call does 5519 not succeed all other subcommands will fail with -EINVAL. This 5520 subcommand will return -EINVAL if a dump process has not yet been 5521 completed. 5522 5523 Not all PV vms can be dumped, the owner needs to set `dump 5524 allowed` PCF bit 34 in the SE header to allow dumping. 5525 5526 KVM_PV_DUMP_CONFIG_STOR_STATE 5527 Stores `buff_len` bytes of tweak component values starting with 5528 the 1MB block specified by the absolute guest address 5529 (`gaddr`). `buff_len` needs to be `conf_dump_storage_state_len` 5530 aligned and at least >= the `conf_dump_storage_state_len` value 5531 provided by the dump uv_info data. buff_user might be written to 5532 even if an error rc is returned. For instance if we encounter a 5533 fault after writing the first page of data. 5534 5535 KVM_PV_DUMP_COMPLETE 5536 If the subcommand succeeds it completes the dump process and lets 5537 KVM_PV_DUMP_INIT be called again. 5538 5539 On success `conf_dump_finalize_len` bytes of completion data will be 5540 stored to the `buff_addr`. The completion data contains a key 5541 derivation seed, IV, tweak nonce and encryption keys as well as an 5542 authentication tag all of which are needed to decrypt the dump at a 5543 later time. 5544 5545KVM_PV_ASYNC_CLEANUP_PREPARE 5546 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE 5547 5548 Prepare the current protected VM for asynchronous teardown. Most 5549 resources used by the current protected VM will be set aside for a 5550 subsequent asynchronous teardown. The current protected VM will then 5551 resume execution immediately as non-protected. There can be at most 5552 one protected VM prepared for asynchronous teardown at any time. If 5553 a protected VM had already been prepared for teardown without 5554 subsequently calling KVM_PV_ASYNC_CLEANUP_PERFORM, this call will 5555 fail. In that case, the userspace process should issue a normal 5556 KVM_PV_DISABLE. The resources set aside with this call will need to 5557 be cleaned up with a subsequent call to KVM_PV_ASYNC_CLEANUP_PERFORM 5558 or KVM_PV_DISABLE, otherwise they will be cleaned up when KVM 5559 terminates. KVM_PV_ASYNC_CLEANUP_PREPARE can be called again as soon 5560 as cleanup starts, i.e. before KVM_PV_ASYNC_CLEANUP_PERFORM finishes. 5561 5562KVM_PV_ASYNC_CLEANUP_PERFORM 5563 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE 5564 5565 Tear down the protected VM previously prepared for teardown with 5566 KVM_PV_ASYNC_CLEANUP_PREPARE. The resources that had been set aside 5567 will be freed during the execution of this command. This PV command 5568 should ideally be issued by userspace from a separate thread. If a 5569 fatal signal is received (or the process terminates naturally), the 5570 command will terminate immediately without completing, and the normal 5571 KVM shutdown procedure will take care of cleaning up all remaining 5572 protected VMs, including the ones whose teardown was interrupted by 5573 process termination. 5574 55754.126 KVM_XEN_HVM_SET_ATTR 5576-------------------------- 5577 5578:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5579:Architectures: x86 5580:Type: vm ioctl 5581:Parameters: struct kvm_xen_hvm_attr 5582:Returns: 0 on success, < 0 on error 5583 5584:: 5585 5586 struct kvm_xen_hvm_attr { 5587 __u16 type; 5588 __u16 pad[3]; 5589 union { 5590 __u8 long_mode; 5591 __u8 vector; 5592 __u8 runstate_update_flag; 5593 union { 5594 __u64 gfn; 5595 __u64 hva; 5596 } shared_info; 5597 struct { 5598 __u32 send_port; 5599 __u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */ 5600 __u32 flags; 5601 union { 5602 struct { 5603 __u32 port; 5604 __u32 vcpu; 5605 __u32 priority; 5606 } port; 5607 struct { 5608 __u32 port; /* Zero for eventfd */ 5609 __s32 fd; 5610 } eventfd; 5611 __u32 padding[4]; 5612 } deliver; 5613 } evtchn; 5614 __u32 xen_version; 5615 __u64 pad[8]; 5616 } u; 5617 }; 5618 5619type values: 5620 5621KVM_XEN_ATTR_TYPE_LONG_MODE 5622 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This 5623 determines the layout of the shared_info page exposed to the VM. 5624 5625KVM_XEN_ATTR_TYPE_SHARED_INFO 5626 Sets the guest physical frame number at which the Xen shared_info 5627 page resides. Note that although Xen places vcpu_info for the first 5628 32 vCPUs in the shared_info page, KVM does not automatically do so 5629 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO or 5630 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA be used explicitly even when 5631 the vcpu_info for a given vCPU resides at the "default" location 5632 in the shared_info page. This is because KVM may not be aware of 5633 the Xen CPU id which is used as the index into the vcpu_info[] 5634 array, so may know the correct default location. 5635 5636 Note that the shared_info page may be constantly written to by KVM; 5637 it contains the event channel bitmap used to deliver interrupts to 5638 a Xen guest, amongst other things. It is exempt from dirty tracking 5639 mechanisms — KVM will not explicitly mark the page as dirty each 5640 time an event channel interrupt is delivered to the guest! Thus, 5641 userspace should always assume that the designated GFN is dirty if 5642 any vCPU has been running or any event channel interrupts can be 5643 routed to the guest. 5644 5645 Setting the gfn to KVM_XEN_INVALID_GFN will disable the shared_info 5646 page. 5647 5648KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA 5649 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA flag is also set in the 5650 Xen capabilities, then this attribute may be used to set the 5651 userspace address at which the shared_info page resides, which 5652 will always be fixed in the VMM regardless of where it is mapped 5653 in guest physical address space. This attribute should be used in 5654 preference to KVM_XEN_ATTR_TYPE_SHARED_INFO as it avoids 5655 unnecessary invalidation of an internal cache when the page is 5656 re-mapped in guest physical address space. 5657 5658 Setting the hva to zero will disable the shared_info page. 5659 5660KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 5661 Sets the exception vector used to deliver Xen event channel upcalls. 5662 This is the HVM-wide vector injected directly by the hypervisor 5663 (not through the local APIC), typically configured by a guest via 5664 HVM_PARAM_CALLBACK_IRQ. This can be disabled again (e.g. for guest 5665 SHUTDOWN_soft_reset) by setting it to zero. 5666 5667KVM_XEN_ATTR_TYPE_EVTCHN 5668 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5669 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5670 an outbound port number for interception of EVTCHNOP_send requests 5671 from the guest. A given sending port number may be directed back to 5672 a specified vCPU (by APIC ID) / port / priority on the guest, or to 5673 trigger events on an eventfd. The vCPU and priority can be changed 5674 by setting KVM_XEN_EVTCHN_UPDATE in a subsequent call, but other 5675 fields cannot change for a given sending port. A port mapping is 5676 removed by using KVM_XEN_EVTCHN_DEASSIGN in the flags field. Passing 5677 KVM_XEN_EVTCHN_RESET in the flags field removes all interception of 5678 outbound event channels. The values of the flags field are mutually 5679 exclusive and cannot be combined as a bitmask. 5680 5681KVM_XEN_ATTR_TYPE_XEN_VERSION 5682 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5683 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5684 the 32-bit version code returned to the guest when it invokes the 5685 XENVER_version call; typically (XEN_MAJOR << 16 | XEN_MINOR). PV 5686 Xen guests will often use this to as a dummy hypercall to trigger 5687 event channel delivery, so responding within the kernel without 5688 exiting to userspace is beneficial. 5689 5690KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG 5691 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5692 support for KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG. It enables the 5693 XEN_RUNSTATE_UPDATE flag which allows guest vCPUs to safely read 5694 other vCPUs' vcpu_runstate_info. Xen guests enable this feature via 5695 the VMASST_TYPE_runstate_update_flag of the HYPERVISOR_vm_assist 5696 hypercall. 5697 56984.127 KVM_XEN_HVM_GET_ATTR 5699-------------------------- 5700 5701:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5702:Architectures: x86 5703:Type: vm ioctl 5704:Parameters: struct kvm_xen_hvm_attr 5705:Returns: 0 on success, < 0 on error 5706 5707Allows Xen VM attributes to be read. For the structure and types, 5708see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_ATTR_TYPE_EVTCHN 5709attribute cannot be read. 5710 57114.128 KVM_XEN_VCPU_SET_ATTR 5712--------------------------- 5713 5714:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5715:Architectures: x86 5716:Type: vcpu ioctl 5717:Parameters: struct kvm_xen_vcpu_attr 5718:Returns: 0 on success, < 0 on error 5719 5720:: 5721 5722 struct kvm_xen_vcpu_attr { 5723 __u16 type; 5724 __u16 pad[3]; 5725 union { 5726 __u64 gpa; 5727 __u64 pad[4]; 5728 struct { 5729 __u64 state; 5730 __u64 state_entry_time; 5731 __u64 time_running; 5732 __u64 time_runnable; 5733 __u64 time_blocked; 5734 __u64 time_offline; 5735 } runstate; 5736 __u32 vcpu_id; 5737 struct { 5738 __u32 port; 5739 __u32 priority; 5740 __u64 expires_ns; 5741 } timer; 5742 __u8 vector; 5743 } u; 5744 }; 5745 5746type values: 5747 5748KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 5749 Sets the guest physical address of the vcpu_info for a given vCPU. 5750 As with the shared_info page for the VM, the corresponding page may be 5751 dirtied at any time if event channel interrupt delivery is enabled, so 5752 userspace should always assume that the page is dirty without relying 5753 on dirty logging. Setting the gpa to KVM_XEN_INVALID_GPA will disable 5754 the vcpu_info. 5755 5756KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA 5757 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA flag is also set in the 5758 Xen capabilities, then this attribute may be used to set the 5759 userspace address of the vcpu_info for a given vCPU. It should 5760 only be used when the vcpu_info resides at the "default" location 5761 in the shared_info page. In this case it is safe to assume the 5762 userspace address will not change, because the shared_info page is 5763 an overlay on guest memory and remains at a fixed host address 5764 regardless of where it is mapped in guest physical address space 5765 and hence unnecessary invalidation of an internal cache may be 5766 avoided if the guest memory layout is modified. 5767 If the vcpu_info does not reside at the "default" location then 5768 it is not guaranteed to remain at the same host address and 5769 hence the aforementioned cache invalidation is required. 5770 5771KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 5772 Sets the guest physical address of an additional pvclock structure 5773 for a given vCPU. This is typically used for guest vsyscall support. 5774 Setting the gpa to KVM_XEN_INVALID_GPA will disable the structure. 5775 5776KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 5777 Sets the guest physical address of the vcpu_runstate_info for a given 5778 vCPU. This is how a Xen guest tracks CPU state such as steal time. 5779 Setting the gpa to KVM_XEN_INVALID_GPA will disable the runstate area. 5780 5781KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 5782 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of 5783 the given vCPU from the .u.runstate.state member of the structure. 5784 KVM automatically accounts running and runnable time but blocked 5785 and offline states are only entered explicitly. 5786 5787KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 5788 Sets all fields of the vCPU runstate data from the .u.runstate member 5789 of the structure, including the current runstate. The state_entry_time 5790 must equal the sum of the other four times. 5791 5792KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 5793 This *adds* the contents of the .u.runstate members of the structure 5794 to the corresponding members of the given vCPU's runstate data, thus 5795 permitting atomic adjustments to the runstate times. The adjustment 5796 to the state_entry_time must equal the sum of the adjustments to the 5797 other four times. The state field must be set to -1, or to a valid 5798 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked 5799 or RUNSTATE_offline) to set the current accounted state as of the 5800 adjusted state_entry_time. 5801 5802KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID 5803 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5804 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the Xen 5805 vCPU ID of the given vCPU, to allow timer-related VCPU operations to 5806 be intercepted by KVM. 5807 5808KVM_XEN_VCPU_ATTR_TYPE_TIMER 5809 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5810 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5811 event channel port/priority for the VIRQ_TIMER of the vCPU, as well 5812 as allowing a pending timer to be saved/restored. Setting the timer 5813 port to zero disables kernel handling of the singleshot timer. 5814 5815KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR 5816 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5817 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5818 per-vCPU local APIC upcall vector, configured by a Xen guest with 5819 the HVMOP_set_evtchn_upcall_vector hypercall. This is typically 5820 used by Windows guests, and is distinct from the HVM-wide upcall 5821 vector configured with HVM_PARAM_CALLBACK_IRQ. It is disabled by 5822 setting the vector to zero. 5823 5824 58254.129 KVM_XEN_VCPU_GET_ATTR 5826--------------------------- 5827 5828:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5829:Architectures: x86 5830:Type: vcpu ioctl 5831:Parameters: struct kvm_xen_vcpu_attr 5832:Returns: 0 on success, < 0 on error 5833 5834Allows Xen vCPU attributes to be read. For the structure and types, 5835see KVM_XEN_VCPU_SET_ATTR above. 5836 5837The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used 5838with the KVM_XEN_VCPU_GET_ATTR ioctl. 5839 58404.130 KVM_ARM_MTE_COPY_TAGS 5841--------------------------- 5842 5843:Capability: KVM_CAP_ARM_MTE 5844:Architectures: arm64 5845:Type: vm ioctl 5846:Parameters: struct kvm_arm_copy_mte_tags 5847:Returns: number of bytes copied, < 0 on error (-EINVAL for incorrect 5848 arguments, -EFAULT if memory cannot be accessed). 5849 5850:: 5851 5852 struct kvm_arm_copy_mte_tags { 5853 __u64 guest_ipa; 5854 __u64 length; 5855 void __user *addr; 5856 __u64 flags; 5857 __u64 reserved[2]; 5858 }; 5859 5860Copies Memory Tagging Extension (MTE) tags to/from guest tag memory. The 5861``guest_ipa`` and ``length`` fields must be ``PAGE_SIZE`` aligned. 5862``length`` must not be bigger than 2^31 - PAGE_SIZE bytes. The ``addr`` 5863field must point to a buffer which the tags will be copied to or from. 5864 5865``flags`` specifies the direction of copy, either ``KVM_ARM_TAGS_TO_GUEST`` or 5866``KVM_ARM_TAGS_FROM_GUEST``. 5867 5868The size of the buffer to store the tags is ``(length / 16)`` bytes 5869(granules in MTE are 16 bytes long). Each byte contains a single tag 5870value. This matches the format of ``PTRACE_PEEKMTETAGS`` and 5871``PTRACE_POKEMTETAGS``. 5872 5873If an error occurs before any data is copied then a negative error code is 5874returned. If some tags have been copied before an error occurs then the number 5875of bytes successfully copied is returned. If the call completes successfully 5876then ``length`` is returned. 5877 58784.131 KVM_GET_SREGS2 5879-------------------- 5880 5881:Capability: KVM_CAP_SREGS2 5882:Architectures: x86 5883:Type: vcpu ioctl 5884:Parameters: struct kvm_sregs2 (out) 5885:Returns: 0 on success, -1 on error 5886 5887Reads special registers from the vcpu. 5888This ioctl (when supported) replaces the KVM_GET_SREGS. 5889 5890:: 5891 5892 struct kvm_sregs2 { 5893 /* out (KVM_GET_SREGS2) / in (KVM_SET_SREGS2) */ 5894 struct kvm_segment cs, ds, es, fs, gs, ss; 5895 struct kvm_segment tr, ldt; 5896 struct kvm_dtable gdt, idt; 5897 __u64 cr0, cr2, cr3, cr4, cr8; 5898 __u64 efer; 5899 __u64 apic_base; 5900 __u64 flags; 5901 __u64 pdptrs[4]; 5902 }; 5903 5904flags values for ``kvm_sregs2``: 5905 5906``KVM_SREGS2_FLAGS_PDPTRS_VALID`` 5907 5908 Indicates that the struct contains valid PDPTR values. 5909 5910 59114.132 KVM_SET_SREGS2 5912-------------------- 5913 5914:Capability: KVM_CAP_SREGS2 5915:Architectures: x86 5916:Type: vcpu ioctl 5917:Parameters: struct kvm_sregs2 (in) 5918:Returns: 0 on success, -1 on error 5919 5920Writes special registers into the vcpu. 5921See KVM_GET_SREGS2 for the data structures. 5922This ioctl (when supported) replaces the KVM_SET_SREGS. 5923 59244.133 KVM_GET_STATS_FD 5925---------------------- 5926 5927:Capability: KVM_CAP_STATS_BINARY_FD 5928:Architectures: all 5929:Type: vm ioctl, vcpu ioctl 5930:Parameters: none 5931:Returns: statistics file descriptor on success, < 0 on error 5932 5933Errors: 5934 5935 ====== ====================================================== 5936 ENOMEM if the fd could not be created due to lack of memory 5937 EMFILE if the number of opened files exceeds the limit 5938 ====== ====================================================== 5939 5940The returned file descriptor can be used to read VM/vCPU statistics data in 5941binary format. The data in the file descriptor consists of four blocks 5942organized as follows: 5943 5944+-------------+ 5945| Header | 5946+-------------+ 5947| id string | 5948+-------------+ 5949| Descriptors | 5950+-------------+ 5951| Stats Data | 5952+-------------+ 5953 5954Apart from the header starting at offset 0, please be aware that it is 5955not guaranteed that the four blocks are adjacent or in the above order; 5956the offsets of the id, descriptors and data blocks are found in the 5957header. However, all four blocks are aligned to 64 bit offsets in the 5958file and they do not overlap. 5959 5960All blocks except the data block are immutable. Userspace can read them 5961only one time after retrieving the file descriptor, and then use ``pread`` or 5962``lseek`` to read the statistics repeatedly. 5963 5964All data is in system endianness. 5965 5966The format of the header is as follows:: 5967 5968 struct kvm_stats_header { 5969 __u32 flags; 5970 __u32 name_size; 5971 __u32 num_desc; 5972 __u32 id_offset; 5973 __u32 desc_offset; 5974 __u32 data_offset; 5975 }; 5976 5977The ``flags`` field is not used at the moment. It is always read as 0. 5978 5979The ``name_size`` field is the size (in byte) of the statistics name string 5980(including trailing '\0') which is contained in the "id string" block and 5981appended at the end of every descriptor. 5982 5983The ``num_desc`` field is the number of descriptors that are included in the 5984descriptor block. (The actual number of values in the data block may be 5985larger, since each descriptor may comprise more than one value). 5986 5987The ``id_offset`` field is the offset of the id string from the start of the 5988file indicated by the file descriptor. It is a multiple of 8. 5989 5990The ``desc_offset`` field is the offset of the Descriptors block from the start 5991of the file indicated by the file descriptor. It is a multiple of 8. 5992 5993The ``data_offset`` field is the offset of the Stats Data block from the start 5994of the file indicated by the file descriptor. It is a multiple of 8. 5995 5996The id string block contains a string which identifies the file descriptor on 5997which KVM_GET_STATS_FD was invoked. The size of the block, including the 5998trailing ``'\0'``, is indicated by the ``name_size`` field in the header. 5999 6000The descriptors block is only needed to be read once for the lifetime of the 6001file descriptor contains a sequence of ``struct kvm_stats_desc``, each followed 6002by a string of size ``name_size``. 6003:: 6004 6005 #define KVM_STATS_TYPE_SHIFT 0 6006 #define KVM_STATS_TYPE_MASK (0xF << KVM_STATS_TYPE_SHIFT) 6007 #define KVM_STATS_TYPE_CUMULATIVE (0x0 << KVM_STATS_TYPE_SHIFT) 6008 #define KVM_STATS_TYPE_INSTANT (0x1 << KVM_STATS_TYPE_SHIFT) 6009 #define KVM_STATS_TYPE_PEAK (0x2 << KVM_STATS_TYPE_SHIFT) 6010 #define KVM_STATS_TYPE_LINEAR_HIST (0x3 << KVM_STATS_TYPE_SHIFT) 6011 #define KVM_STATS_TYPE_LOG_HIST (0x4 << KVM_STATS_TYPE_SHIFT) 6012 #define KVM_STATS_TYPE_MAX KVM_STATS_TYPE_LOG_HIST 6013 6014 #define KVM_STATS_UNIT_SHIFT 4 6015 #define KVM_STATS_UNIT_MASK (0xF << KVM_STATS_UNIT_SHIFT) 6016 #define KVM_STATS_UNIT_NONE (0x0 << KVM_STATS_UNIT_SHIFT) 6017 #define KVM_STATS_UNIT_BYTES (0x1 << KVM_STATS_UNIT_SHIFT) 6018 #define KVM_STATS_UNIT_SECONDS (0x2 << KVM_STATS_UNIT_SHIFT) 6019 #define KVM_STATS_UNIT_CYCLES (0x3 << KVM_STATS_UNIT_SHIFT) 6020 #define KVM_STATS_UNIT_BOOLEAN (0x4 << KVM_STATS_UNIT_SHIFT) 6021 #define KVM_STATS_UNIT_MAX KVM_STATS_UNIT_BOOLEAN 6022 6023 #define KVM_STATS_BASE_SHIFT 8 6024 #define KVM_STATS_BASE_MASK (0xF << KVM_STATS_BASE_SHIFT) 6025 #define KVM_STATS_BASE_POW10 (0x0 << KVM_STATS_BASE_SHIFT) 6026 #define KVM_STATS_BASE_POW2 (0x1 << KVM_STATS_BASE_SHIFT) 6027 #define KVM_STATS_BASE_MAX KVM_STATS_BASE_POW2 6028 6029 struct kvm_stats_desc { 6030 __u32 flags; 6031 __s16 exponent; 6032 __u16 size; 6033 __u32 offset; 6034 __u32 bucket_size; 6035 char name[]; 6036 }; 6037 6038The ``flags`` field contains the type and unit of the statistics data described 6039by this descriptor. Its endianness is CPU native. 6040The following flags are supported: 6041 6042Bits 0-3 of ``flags`` encode the type: 6043 6044 * ``KVM_STATS_TYPE_CUMULATIVE`` 6045 The statistics reports a cumulative count. The value of data can only be increased. 6046 Most of the counters used in KVM are of this type. 6047 The corresponding ``size`` field for this type is always 1. 6048 All cumulative statistics data are read/write. 6049 * ``KVM_STATS_TYPE_INSTANT`` 6050 The statistics reports an instantaneous value. Its value can be increased or 6051 decreased. This type is usually used as a measurement of some resources, 6052 like the number of dirty pages, the number of large pages, etc. 6053 All instant statistics are read only. 6054 The corresponding ``size`` field for this type is always 1. 6055 * ``KVM_STATS_TYPE_PEAK`` 6056 The statistics data reports a peak value, for example the maximum number 6057 of items in a hash table bucket, the longest time waited and so on. 6058 The value of data can only be increased. 6059 The corresponding ``size`` field for this type is always 1. 6060 * ``KVM_STATS_TYPE_LINEAR_HIST`` 6061 The statistic is reported as a linear histogram. The number of 6062 buckets is specified by the ``size`` field. The size of buckets is specified 6063 by the ``hist_param`` field. The range of the Nth bucket (1 <= N < ``size``) 6064 is [``hist_param``*(N-1), ``hist_param``*N), while the range of the last 6065 bucket is [``hist_param``*(``size``-1), +INF). (+INF means positive infinity 6066 value.) 6067 * ``KVM_STATS_TYPE_LOG_HIST`` 6068 The statistic is reported as a logarithmic histogram. The number of 6069 buckets is specified by the ``size`` field. The range of the first bucket is 6070 [0, 1), while the range of the last bucket is [pow(2, ``size``-2), +INF). 6071 Otherwise, The Nth bucket (1 < N < ``size``) covers 6072 [pow(2, N-2), pow(2, N-1)). 6073 6074Bits 4-7 of ``flags`` encode the unit: 6075 6076 * ``KVM_STATS_UNIT_NONE`` 6077 There is no unit for the value of statistics data. This usually means that 6078 the value is a simple counter of an event. 6079 * ``KVM_STATS_UNIT_BYTES`` 6080 It indicates that the statistics data is used to measure memory size, in the 6081 unit of Byte, KiByte, MiByte, GiByte, etc. The unit of the data is 6082 determined by the ``exponent`` field in the descriptor. 6083 * ``KVM_STATS_UNIT_SECONDS`` 6084 It indicates that the statistics data is used to measure time or latency. 6085 * ``KVM_STATS_UNIT_CYCLES`` 6086 It indicates that the statistics data is used to measure CPU clock cycles. 6087 * ``KVM_STATS_UNIT_BOOLEAN`` 6088 It indicates that the statistic will always be either 0 or 1. Boolean 6089 statistics of "peak" type will never go back from 1 to 0. Boolean 6090 statistics can be linear histograms (with two buckets) but not logarithmic 6091 histograms. 6092 6093Note that, in the case of histograms, the unit applies to the bucket 6094ranges, while the bucket value indicates how many samples fell in the 6095bucket's range. 6096 6097Bits 8-11 of ``flags``, together with ``exponent``, encode the scale of the 6098unit: 6099 6100 * ``KVM_STATS_BASE_POW10`` 6101 The scale is based on power of 10. It is used for measurement of time and 6102 CPU clock cycles. For example, an exponent of -9 can be used with 6103 ``KVM_STATS_UNIT_SECONDS`` to express that the unit is nanoseconds. 6104 * ``KVM_STATS_BASE_POW2`` 6105 The scale is based on power of 2. It is used for measurement of memory size. 6106 For example, an exponent of 20 can be used with ``KVM_STATS_UNIT_BYTES`` to 6107 express that the unit is MiB. 6108 6109The ``size`` field is the number of values of this statistics data. Its 6110value is usually 1 for most of simple statistics. 1 means it contains an 6111unsigned 64bit data. 6112 6113The ``offset`` field is the offset from the start of Data Block to the start of 6114the corresponding statistics data. 6115 6116The ``bucket_size`` field is used as a parameter for histogram statistics data. 6117It is only used by linear histogram statistics data, specifying the size of a 6118bucket in the unit expressed by bits 4-11 of ``flags`` together with ``exponent``. 6119 6120The ``name`` field is the name string of the statistics data. The name string 6121starts at the end of ``struct kvm_stats_desc``. The maximum length including 6122the trailing ``'\0'``, is indicated by ``name_size`` in the header. 6123 6124The Stats Data block contains an array of 64-bit values in the same order 6125as the descriptors in Descriptors block. 6126 61274.134 KVM_GET_XSAVE2 6128-------------------- 6129 6130:Capability: KVM_CAP_XSAVE2 6131:Architectures: x86 6132:Type: vcpu ioctl 6133:Parameters: struct kvm_xsave (out) 6134:Returns: 0 on success, -1 on error 6135 6136 6137:: 6138 6139 struct kvm_xsave { 6140 __u32 region[1024]; 6141 __u32 extra[0]; 6142 }; 6143 6144This ioctl would copy current vcpu's xsave struct to the userspace. It 6145copies as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) 6146when invoked on the vm file descriptor. The size value returned by 6147KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 6148Currently, it is only greater than 4096 if a dynamic feature has been 6149enabled with ``arch_prctl()``, but this may change in the future. 6150 6151The offsets of the state save areas in struct kvm_xsave follow the contents 6152of CPUID leaf 0xD on the host. 6153 61544.135 KVM_XEN_HVM_EVTCHN_SEND 6155----------------------------- 6156 6157:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND 6158:Architectures: x86 6159:Type: vm ioctl 6160:Parameters: struct kvm_irq_routing_xen_evtchn 6161:Returns: 0 on success, < 0 on error 6162 6163 6164:: 6165 6166 struct kvm_irq_routing_xen_evtchn { 6167 __u32 port; 6168 __u32 vcpu; 6169 __u32 priority; 6170 }; 6171 6172This ioctl injects an event channel interrupt directly to the guest vCPU. 6173 61744.136 KVM_S390_PV_CPU_COMMAND 6175----------------------------- 6176 6177:Capability: KVM_CAP_S390_PROTECTED_DUMP 6178:Architectures: s390 6179:Type: vcpu ioctl 6180:Parameters: none 6181:Returns: 0 on success, < 0 on error 6182 6183This ioctl closely mirrors `KVM_S390_PV_COMMAND` but handles requests 6184for vcpus. It re-uses the kvm_s390_pv_dmp struct and hence also shares 6185the command ids. 6186 6187**command:** 6188 6189KVM_PV_DUMP 6190 Presents an API that provides calls which facilitate dumping a vcpu 6191 of a protected VM. 6192 6193**subcommand:** 6194 6195KVM_PV_DUMP_CPU 6196 Provides encrypted dump data like register values. 6197 The length of the returned data is provided by uv_info.guest_cpu_stor_len. 6198 61994.137 KVM_S390_ZPCI_OP 6200---------------------- 6201 6202:Capability: KVM_CAP_S390_ZPCI_OP 6203:Architectures: s390 6204:Type: vm ioctl 6205:Parameters: struct kvm_s390_zpci_op (in) 6206:Returns: 0 on success, <0 on error 6207 6208Used to manage hardware-assisted virtualization features for zPCI devices. 6209 6210Parameters are specified via the following structure:: 6211 6212 struct kvm_s390_zpci_op { 6213 /* in */ 6214 __u32 fh; /* target device */ 6215 __u8 op; /* operation to perform */ 6216 __u8 pad[3]; 6217 union { 6218 /* for KVM_S390_ZPCIOP_REG_AEN */ 6219 struct { 6220 __u64 ibv; /* Guest addr of interrupt bit vector */ 6221 __u64 sb; /* Guest addr of summary bit */ 6222 __u32 flags; 6223 __u32 noi; /* Number of interrupts */ 6224 __u8 isc; /* Guest interrupt subclass */ 6225 __u8 sbo; /* Offset of guest summary bit vector */ 6226 __u16 pad; 6227 } reg_aen; 6228 __u64 reserved[8]; 6229 } u; 6230 }; 6231 6232The type of operation is specified in the "op" field. 6233KVM_S390_ZPCIOP_REG_AEN is used to register the VM for adapter event 6234notification interpretation, which will allow firmware delivery of adapter 6235events directly to the vm, with KVM providing a backup delivery mechanism; 6236KVM_S390_ZPCIOP_DEREG_AEN is used to subsequently disable interpretation of 6237adapter event notifications. 6238 6239The target zPCI function must also be specified via the "fh" field. For the 6240KVM_S390_ZPCIOP_REG_AEN operation, additional information to establish firmware 6241delivery must be provided via the "reg_aen" struct. 6242 6243The "pad" and "reserved" fields may be used for future extensions and should be 6244set to 0s by userspace. 6245 62464.138 KVM_ARM_SET_COUNTER_OFFSET 6247-------------------------------- 6248 6249:Capability: KVM_CAP_COUNTER_OFFSET 6250:Architectures: arm64 6251:Type: vm ioctl 6252:Parameters: struct kvm_arm_counter_offset (in) 6253:Returns: 0 on success, < 0 on error 6254 6255This capability indicates that userspace is able to apply a single VM-wide 6256offset to both the virtual and physical counters as viewed by the guest 6257using the KVM_ARM_SET_CNT_OFFSET ioctl and the following data structure: 6258 6259:: 6260 6261 struct kvm_arm_counter_offset { 6262 __u64 counter_offset; 6263 __u64 reserved; 6264 }; 6265 6266The offset describes a number of counter cycles that are subtracted from 6267both virtual and physical counter views (similar to the effects of the 6268CNTVOFF_EL2 and CNTPOFF_EL2 system registers, but only global). The offset 6269always applies to all vcpus (already created or created after this ioctl) 6270for this VM. 6271 6272It is userspace's responsibility to compute the offset based, for example, 6273on previous values of the guest counters. 6274 6275Any value other than 0 for the "reserved" field may result in an error 6276(-EINVAL) being returned. This ioctl can also return -EBUSY if any vcpu 6277ioctl is issued concurrently. 6278 6279Note that using this ioctl results in KVM ignoring subsequent userspace 6280writes to the CNTVCT_EL0 and CNTPCT_EL0 registers using the SET_ONE_REG 6281interface. No error will be returned, but the resulting offset will not be 6282applied. 6283 6284.. _KVM_ARM_GET_REG_WRITABLE_MASKS: 6285 62864.139 KVM_ARM_GET_REG_WRITABLE_MASKS 6287------------------------------------ 6288 6289:Capability: KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES 6290:Architectures: arm64 6291:Type: vm ioctl 6292:Parameters: struct reg_mask_range (in/out) 6293:Returns: 0 on success, < 0 on error 6294 6295 6296:: 6297 6298 #define KVM_ARM_FEATURE_ID_RANGE 0 6299 #define KVM_ARM_FEATURE_ID_RANGE_SIZE (3 * 8 * 8) 6300 6301 struct reg_mask_range { 6302 __u64 addr; /* Pointer to mask array */ 6303 __u32 range; /* Requested range */ 6304 __u32 reserved[13]; 6305 }; 6306 6307This ioctl copies the writable masks for a selected range of registers to 6308userspace. 6309 6310The ``addr`` field is a pointer to the destination array where KVM copies 6311the writable masks. 6312 6313The ``range`` field indicates the requested range of registers. 6314``KVM_CHECK_EXTENSION`` for the ``KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES`` 6315capability returns the supported ranges, expressed as a set of flags. Each 6316flag's bit index represents a possible value for the ``range`` field. 6317All other values are reserved for future use and KVM may return an error. 6318 6319The ``reserved[13]`` array is reserved for future use and should be 0, or 6320KVM may return an error. 6321 6322KVM_ARM_FEATURE_ID_RANGE (0) 6323^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 6324 6325The Feature ID range is defined as the AArch64 System register space with 6326op0==3, op1=={0, 1, 3}, CRn==0, CRm=={0-7}, op2=={0-7}. 6327 6328The mask returned array pointed to by ``addr`` is indexed by the macro 6329``ARM64_FEATURE_ID_RANGE_IDX(op0, op1, crn, crm, op2)``, allowing userspace 6330to know what fields can be changed for the system register described by 6331``op0, op1, crn, crm, op2``. KVM rejects ID register values that describe a 6332superset of the features supported by the system. 6333 63344.140 KVM_SET_USER_MEMORY_REGION2 6335--------------------------------- 6336 6337:Capability: KVM_CAP_USER_MEMORY2 6338:Architectures: all 6339:Type: vm ioctl 6340:Parameters: struct kvm_userspace_memory_region2 (in) 6341:Returns: 0 on success, -1 on error 6342 6343KVM_SET_USER_MEMORY_REGION2 is an extension to KVM_SET_USER_MEMORY_REGION that 6344allows mapping guest_memfd memory into a guest. All fields shared with 6345KVM_SET_USER_MEMORY_REGION identically. Userspace can set KVM_MEM_GUEST_MEMFD 6346in flags to have KVM bind the memory region to a given guest_memfd range of 6347[guest_memfd_offset, guest_memfd_offset + memory_size]. The target guest_memfd 6348must point at a file created via KVM_CREATE_GUEST_MEMFD on the current VM, and 6349the target range must not be bound to any other memory region. All standard 6350bounds checks apply (use common sense). 6351 6352:: 6353 6354 struct kvm_userspace_memory_region2 { 6355 __u32 slot; 6356 __u32 flags; 6357 __u64 guest_phys_addr; 6358 __u64 memory_size; /* bytes */ 6359 __u64 userspace_addr; /* start of the userspace allocated memory */ 6360 __u64 guest_memfd_offset; 6361 __u32 guest_memfd; 6362 __u32 pad1; 6363 __u64 pad2[14]; 6364 }; 6365 6366A KVM_MEM_GUEST_MEMFD region _must_ have a valid guest_memfd (private memory) and 6367userspace_addr (shared memory). However, "valid" for userspace_addr simply 6368means that the address itself must be a legal userspace address. The backing 6369mapping for userspace_addr is not required to be valid/populated at the time of 6370KVM_SET_USER_MEMORY_REGION2, e.g. shared memory can be lazily mapped/allocated 6371on-demand. 6372 6373When mapping a gfn into the guest, KVM selects shared vs. private, i.e consumes 6374userspace_addr vs. guest_memfd, based on the gfn's KVM_MEMORY_ATTRIBUTE_PRIVATE 6375state. At VM creation time, all memory is shared, i.e. the PRIVATE attribute 6376is '0' for all gfns. Userspace can control whether memory is shared/private by 6377toggling KVM_MEMORY_ATTRIBUTE_PRIVATE via KVM_SET_MEMORY_ATTRIBUTES as needed. 6378 6379S390: 6380^^^^^ 6381 6382Returns -EINVAL if the VM has the KVM_VM_S390_UCONTROL flag set. 6383Returns -EINVAL if called on a protected VM. 6384 63854.141 KVM_SET_MEMORY_ATTRIBUTES 6386------------------------------- 6387 6388:Capability: KVM_CAP_MEMORY_ATTRIBUTES 6389:Architectures: x86 6390:Type: vm ioctl 6391:Parameters: struct kvm_memory_attributes (in) 6392:Returns: 0 on success, <0 on error 6393 6394KVM_SET_MEMORY_ATTRIBUTES allows userspace to set memory attributes for a range 6395of guest physical memory. 6396 6397:: 6398 6399 struct kvm_memory_attributes { 6400 __u64 address; 6401 __u64 size; 6402 __u64 attributes; 6403 __u64 flags; 6404 }; 6405 6406 #define KVM_MEMORY_ATTRIBUTE_PRIVATE (1ULL << 3) 6407 6408The address and size must be page aligned. The supported attributes can be 6409retrieved via ioctl(KVM_CHECK_EXTENSION) on KVM_CAP_MEMORY_ATTRIBUTES. If 6410executed on a VM, KVM_CAP_MEMORY_ATTRIBUTES precisely returns the attributes 6411supported by that VM. If executed at system scope, KVM_CAP_MEMORY_ATTRIBUTES 6412returns all attributes supported by KVM. The only attribute defined at this 6413time is KVM_MEMORY_ATTRIBUTE_PRIVATE, which marks the associated gfn as being 6414guest private memory. 6415 6416Note, there is no "get" API. Userspace is responsible for explicitly tracking 6417the state of a gfn/page as needed. 6418 6419The "flags" field is reserved for future extensions and must be '0'. 6420 64214.142 KVM_CREATE_GUEST_MEMFD 6422---------------------------- 6423 6424:Capability: KVM_CAP_GUEST_MEMFD 6425:Architectures: none 6426:Type: vm ioctl 6427:Parameters: struct kvm_create_guest_memfd(in) 6428:Returns: A file descriptor on success, <0 on error 6429 6430KVM_CREATE_GUEST_MEMFD creates an anonymous file and returns a file descriptor 6431that refers to it. guest_memfd files are roughly analogous to files created 6432via memfd_create(), e.g. guest_memfd files live in RAM, have volatile storage, 6433and are automatically released when the last reference is dropped. Unlike 6434"regular" memfd_create() files, guest_memfd files are bound to their owning 6435virtual machine (see below), cannot be mapped, read, or written by userspace, 6436and cannot be resized (guest_memfd files do however support PUNCH_HOLE). 6437 6438:: 6439 6440 struct kvm_create_guest_memfd { 6441 __u64 size; 6442 __u64 flags; 6443 __u64 reserved[6]; 6444 }; 6445 6446Conceptually, the inode backing a guest_memfd file represents physical memory, 6447i.e. is coupled to the virtual machine as a thing, not to a "struct kvm". The 6448file itself, which is bound to a "struct kvm", is that instance's view of the 6449underlying memory, e.g. effectively provides the translation of guest addresses 6450to host memory. This allows for use cases where multiple KVM structures are 6451used to manage a single virtual machine, e.g. when performing intrahost 6452migration of a virtual machine. 6453 6454KVM currently only supports mapping guest_memfd via KVM_SET_USER_MEMORY_REGION2, 6455and more specifically via the guest_memfd and guest_memfd_offset fields in 6456"struct kvm_userspace_memory_region2", where guest_memfd_offset is the offset 6457into the guest_memfd instance. For a given guest_memfd file, there can be at 6458most one mapping per page, i.e. binding multiple memory regions to a single 6459guest_memfd range is not allowed (any number of memory regions can be bound to 6460a single guest_memfd file, but the bound ranges must not overlap). 6461 6462The capability KVM_CAP_GUEST_MEMFD_FLAGS enumerates the `flags` that can be 6463specified via KVM_CREATE_GUEST_MEMFD. Currently defined flags: 6464 6465 ============================ ================================================ 6466 GUEST_MEMFD_FLAG_MMAP Enable using mmap() on the guest_memfd file 6467 descriptor. 6468 GUEST_MEMFD_FLAG_INIT_SHARED Make all memory in the file shared during 6469 KVM_CREATE_GUEST_MEMFD (memory files created 6470 without INIT_SHARED will be marked private). 6471 Shared memory can be faulted into host userspace 6472 page tables. Private memory cannot. 6473 ============================ ================================================ 6474 6475When the KVM MMU performs a PFN lookup to service a guest fault and the backing 6476guest_memfd has the GUEST_MEMFD_FLAG_MMAP set, then the fault will always be 6477consumed from guest_memfd, regardless of whether it is a shared or a private 6478fault. 6479 6480See KVM_SET_USER_MEMORY_REGION2 for additional details. 6481 64824.143 KVM_PRE_FAULT_MEMORY 6483--------------------------- 6484 6485:Capability: KVM_CAP_PRE_FAULT_MEMORY 6486:Architectures: none 6487:Type: vcpu ioctl 6488:Parameters: struct kvm_pre_fault_memory (in/out) 6489:Returns: 0 if at least one page is processed, < 0 on error 6490 6491Errors: 6492 6493 ========== =============================================================== 6494 EINVAL The specified `gpa` and `size` were invalid (e.g. not 6495 page aligned, causes an overflow, or size is zero), or the VM 6496 is UCONTROL (s390). 6497 ENOENT The specified `gpa` is outside defined memslots. 6498 EINTR An unmasked signal is pending and no page was processed. 6499 EFAULT The parameter address was invalid. 6500 EOPNOTSUPP Mapping memory for a GPA is unsupported by the 6501 hypervisor, and/or for the current vCPU state/mode. 6502 EIO unexpected error conditions (also causes a WARN) 6503 ========== =============================================================== 6504 6505:: 6506 6507 struct kvm_pre_fault_memory { 6508 /* in/out */ 6509 __u64 gpa; 6510 __u64 size; 6511 /* in */ 6512 __u64 flags; 6513 __u64 padding[5]; 6514 }; 6515 6516KVM_PRE_FAULT_MEMORY populates KVM's stage-2 page tables used to map memory 6517for the current vCPU state. KVM maps memory as if the vCPU generated a 6518stage-2 read page fault, e.g. faults in memory as needed, but doesn't break 6519CoW. On x86, KVM does not mark any newly created stage-2 PTE as Accessed. 6520 6521In the case of confidential VM types where there is an initial set up of 6522private guest memory before the guest is 'finalized'/measured, this ioctl 6523should only be issued after completing all the necessary setup to put the 6524guest into a 'finalized' state so that the above semantics can be reliably 6525ensured. 6526 6527In some cases, multiple vCPUs might share the page tables. In this 6528case, the ioctl can be called in parallel. 6529 6530When the ioctl returns, the input values are updated to point to the 6531remaining range. If `size` > 0 on return, the caller can just issue 6532the ioctl again with the same `struct kvm_map_memory` argument. 6533 6534Shadow page tables cannot support this ioctl because they 6535are indexed by virtual address or nested guest physical address. 6536Calling this ioctl when the guest is using shadow page tables (for 6537example because it is running a nested guest with nested page tables) 6538will fail with `EOPNOTSUPP` even if `KVM_CHECK_EXTENSION` reports 6539the capability to be present. 6540 6541`flags` must currently be zero. 6542 65434.144 KVM_S390_KEYOP 6544-------------------- 6545 6546:Capability: KVM_CAP_S390_KEYOP 6547:Architectures: s390 6548:Type: vm ioctl 6549:Parameters: struct kvm_s390_keyop (in/out) 6550:Returns: 0 in case of success, < 0 on error 6551 6552The specified key operation is performed on the given guest address. The 6553previous storage key (or the relevant part thereof) will be returned in 6554`key`. 6555 6556:: 6557 6558 struct kvm_s390_keyop { 6559 __u64 guest_addr; 6560 __u8 key; 6561 __u8 operation; 6562 }; 6563 6564Currently supported values for ``operation``: 6565 6566KVM_S390_KEYOP_ISKE 6567 Returns the storage key for the guest address ``guest_addr`` in ``key``. 6568 6569KVM_S390_KEYOP_RRBE 6570 Resets the reference bit for the guest address ``guest_addr``, returning the 6571 R and C bits of the old storage key in ``key``; the remaining fields of 6572 the storage key will be set to 0. 6573 6574KVM_S390_KEYOP_SSKE 6575 Sets the storage key for the guest address ``guest_addr`` to the key 6576 specified in ``key``, returning the previous value in ``key``. 6577 6578.. _kvm_run: 6579 65805. The kvm_run structure 6581======================== 6582 6583Application code obtains a pointer to the kvm_run structure by 6584mmap()ing a vcpu fd. From that point, application code can control 6585execution by changing fields in kvm_run prior to calling the KVM_RUN 6586ioctl, and obtain information about the reason KVM_RUN returned by 6587looking up structure members. 6588 6589:: 6590 6591 struct kvm_run { 6592 /* in */ 6593 __u8 request_interrupt_window; 6594 6595Request that KVM_RUN return when it becomes possible to inject external 6596interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 6597 6598:: 6599 6600 __u8 immediate_exit; 6601 6602This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 6603exits immediately, returning -EINTR. In the common scenario where a 6604signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 6605to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 6606Rather than blocking the signal outside KVM_RUN, userspace can set up 6607a signal handler that sets run->immediate_exit to a non-zero value. 6608 6609This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 6610 6611:: 6612 6613 __u8 padding1[6]; 6614 6615 /* out */ 6616 __u32 exit_reason; 6617 6618When KVM_RUN has returned successfully (return value 0), this informs 6619application code why KVM_RUN has returned. Allowable values for this 6620field are detailed below. 6621 6622:: 6623 6624 __u8 ready_for_interrupt_injection; 6625 6626If request_interrupt_window has been specified, this field indicates 6627an interrupt can be injected now with KVM_INTERRUPT. 6628 6629:: 6630 6631 __u8 if_flag; 6632 6633The value of the current interrupt flag. Only valid if in-kernel 6634local APIC is not used. 6635 6636:: 6637 6638 __u16 flags; 6639 6640More architecture-specific flags detailing state of the VCPU that may 6641affect the device's behavior. Current defined flags:: 6642 6643 /* x86, set if the VCPU is in system management mode */ 6644 #define KVM_RUN_X86_SMM (1 << 0) 6645 /* x86, set if bus lock detected in VM */ 6646 #define KVM_RUN_X86_BUS_LOCK (1 << 1) 6647 /* x86, set if the VCPU is executing a nested (L2) guest */ 6648 #define KVM_RUN_X86_GUEST_MODE (1 << 2) 6649 6650 /* arm64, set for KVM_EXIT_DEBUG */ 6651 #define KVM_DEBUG_ARCH_HSR_HIGH_VALID (1 << 0) 6652 6653:: 6654 6655 /* in (pre_kvm_run), out (post_kvm_run) */ 6656 __u64 cr8; 6657 6658The value of the cr8 register. Only valid if in-kernel local APIC is 6659not used. Both input and output. 6660 6661:: 6662 6663 __u64 apic_base; 6664 6665The value of the APIC BASE msr. Only valid if in-kernel local 6666APIC is not used. Both input and output. 6667 6668:: 6669 6670 union { 6671 /* KVM_EXIT_UNKNOWN */ 6672 struct { 6673 __u64 hardware_exit_reason; 6674 } hw; 6675 6676If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 6677reasons. Further architecture-specific information is available in 6678hardware_exit_reason. 6679 6680:: 6681 6682 /* KVM_EXIT_FAIL_ENTRY */ 6683 struct { 6684 __u64 hardware_entry_failure_reason; 6685 __u32 cpu; /* if KVM_LAST_CPU */ 6686 } fail_entry; 6687 6688If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 6689to unknown reasons. Further architecture-specific information is 6690available in hardware_entry_failure_reason. 6691 6692:: 6693 6694 /* KVM_EXIT_EXCEPTION */ 6695 struct { 6696 __u32 exception; 6697 __u32 error_code; 6698 } ex; 6699 6700Unused. 6701 6702:: 6703 6704 /* KVM_EXIT_IO */ 6705 struct { 6706 #define KVM_EXIT_IO_IN 0 6707 #define KVM_EXIT_IO_OUT 1 6708 __u8 direction; 6709 __u8 size; /* bytes */ 6710 __u16 port; 6711 __u32 count; 6712 __u64 data_offset; /* relative to kvm_run start */ 6713 } io; 6714 6715If exit_reason is KVM_EXIT_IO, then the vcpu has 6716executed a port I/O instruction which could not be satisfied by kvm. 6717data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 6718where kvm expects application code to place the data for the next 6719KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 6720 6721:: 6722 6723 /* KVM_EXIT_DEBUG */ 6724 struct { 6725 struct kvm_debug_exit_arch arch; 6726 } debug; 6727 6728If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 6729for which architecture specific information is returned. 6730 6731:: 6732 6733 /* KVM_EXIT_MMIO */ 6734 struct { 6735 __u64 phys_addr; 6736 __u8 data[8]; 6737 __u32 len; 6738 __u8 is_write; 6739 } mmio; 6740 6741If exit_reason is KVM_EXIT_MMIO, then the vcpu has 6742executed a memory-mapped I/O instruction which could not be satisfied 6743by kvm. The 'data' member contains the written data if 'is_write' is 6744true, and should be filled by application code otherwise. 6745 6746The 'data' member contains, in its first 'len' bytes, the value as it would 6747appear if the VCPU performed a load or store of the appropriate width directly 6748to the byte array. 6749 6750.. note:: 6751 6752 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN, 6753 KVM_EXIT_EPR, KVM_EXIT_HYPERCALL, KVM_EXIT_TDX, 6754 KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding 6755 operations are complete (and guest state is consistent) only after userspace 6756 has re-entered the kernel with KVM_RUN. The kernel side will first finish 6757 incomplete operations and then check for pending signals. 6758 6759 The pending state of the operation is not preserved in state which is 6760 visible to userspace, thus userspace should ensure that the operation is 6761 completed before performing a live migration. Userspace can re-enter the 6762 guest with an unmasked signal pending or with the immediate_exit field set 6763 to complete pending operations without allowing any further instructions 6764 to be executed. 6765 6766:: 6767 6768 /* KVM_EXIT_HYPERCALL */ 6769 struct { 6770 __u64 nr; 6771 __u64 args[6]; 6772 __u64 ret; 6773 __u64 flags; 6774 } hypercall; 6775 6776 6777It is strongly recommended that userspace use ``KVM_EXIT_IO`` (x86) or 6778``KVM_EXIT_MMIO`` (all except s390) to implement functionality that 6779requires a guest to interact with host userspace. 6780 6781.. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 6782 6783For arm64: 6784---------- 6785 6786SMCCC exits can be enabled depending on the configuration of the SMCCC 6787filter. See the Documentation/virt/kvm/devices/vm.rst 6788``KVM_ARM_SMCCC_FILTER`` for more details. 6789 6790``nr`` contains the function ID of the guest's SMCCC call. Userspace is 6791expected to use the ``KVM_GET_ONE_REG`` ioctl to retrieve the call 6792parameters from the vCPU's GPRs. 6793 6794Definition of ``flags``: 6795 - ``KVM_HYPERCALL_EXIT_SMC``: Indicates that the guest used the SMC 6796 conduit to initiate the SMCCC call. If this bit is 0 then the guest 6797 used the HVC conduit for the SMCCC call. 6798 6799 - ``KVM_HYPERCALL_EXIT_16BIT``: Indicates that the guest used a 16bit 6800 instruction to initiate the SMCCC call. If this bit is 0 then the 6801 guest used a 32bit instruction. An AArch64 guest always has this 6802 bit set to 0. 6803 6804At the point of exit, PC points to the instruction immediately following 6805the trapping instruction. 6806 6807:: 6808 6809 /* KVM_EXIT_TPR_ACCESS */ 6810 struct { 6811 __u64 rip; 6812 __u32 is_write; 6813 __u32 pad; 6814 } tpr_access; 6815 6816To be documented (KVM_TPR_ACCESS_REPORTING). 6817 6818:: 6819 6820 /* KVM_EXIT_S390_SIEIC */ 6821 struct { 6822 __u8 icptcode; 6823 __u64 mask; /* psw upper half */ 6824 __u64 addr; /* psw lower half */ 6825 __u16 ipa; 6826 __u32 ipb; 6827 } s390_sieic; 6828 6829s390 specific. 6830 6831:: 6832 6833 /* KVM_EXIT_S390_RESET */ 6834 #define KVM_S390_RESET_POR 1 6835 #define KVM_S390_RESET_CLEAR 2 6836 #define KVM_S390_RESET_SUBSYSTEM 4 6837 #define KVM_S390_RESET_CPU_INIT 8 6838 #define KVM_S390_RESET_IPL 16 6839 __u64 s390_reset_flags; 6840 6841s390 specific. 6842 6843:: 6844 6845 /* KVM_EXIT_S390_UCONTROL */ 6846 struct { 6847 __u64 trans_exc_code; 6848 __u32 pgm_code; 6849 } s390_ucontrol; 6850 6851s390 specific. A page fault has occurred for a user controlled virtual 6852machine (KVM_VM_S390_UCONTROL) on its host page table that cannot be 6853resolved by the kernel. 6854The program code and the translation exception code that were placed 6855in the cpu's lowcore are presented here as defined by the z Architecture 6856Principles of Operation Book in the Chapter for Dynamic Address Translation 6857(DAT) 6858 6859:: 6860 6861 /* KVM_EXIT_DCR */ 6862 struct { 6863 __u32 dcrn; 6864 __u32 data; 6865 __u8 is_write; 6866 } dcr; 6867 6868Deprecated - was used for 440 KVM. 6869 6870:: 6871 6872 /* KVM_EXIT_OSI */ 6873 struct { 6874 __u64 gprs[32]; 6875 } osi; 6876 6877MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 6878hypercalls and exit with this exit struct that contains all the guest gprs. 6879 6880If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 6881Userspace can now handle the hypercall and when it's done modify the gprs as 6882necessary. Upon guest entry all guest GPRs will then be replaced by the values 6883in this struct. 6884 6885:: 6886 6887 /* KVM_EXIT_PAPR_HCALL */ 6888 struct { 6889 __u64 nr; 6890 __u64 ret; 6891 __u64 args[9]; 6892 } papr_hcall; 6893 6894This is used on 64-bit PowerPC when emulating a pSeries partition, 6895e.g. with the 'pseries' machine type in qemu. It occurs when the 6896guest does a hypercall using the 'sc 1' instruction. The 'nr' field 6897contains the hypercall number (from the guest R3), and 'args' contains 6898the arguments (from the guest R4 - R12). Userspace should put the 6899return code in 'ret' and any extra returned values in args[]. 6900The possible hypercalls are defined in the Power Architecture Platform 6901Requirements (PAPR) document available from www.power.org (free 6902developer registration required to access it). 6903 6904:: 6905 6906 /* KVM_EXIT_S390_TSCH */ 6907 struct { 6908 __u16 subchannel_id; 6909 __u16 subchannel_nr; 6910 __u32 io_int_parm; 6911 __u32 io_int_word; 6912 __u32 ipb; 6913 __u8 dequeued; 6914 } s390_tsch; 6915 6916s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 6917and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 6918interrupt for the target subchannel has been dequeued and subchannel_id, 6919subchannel_nr, io_int_parm and io_int_word contain the parameters for that 6920interrupt. ipb is needed for instruction parameter decoding. 6921 6922:: 6923 6924 /* KVM_EXIT_EPR */ 6925 struct { 6926 __u32 epr; 6927 } epr; 6928 6929On FSL BookE PowerPC chips, the interrupt controller has a fast patch 6930interrupt acknowledge path to the core. When the core successfully 6931delivers an interrupt, it automatically populates the EPR register with 6932the interrupt vector number and acknowledges the interrupt inside 6933the interrupt controller. 6934 6935In case the interrupt controller lives in user space, we need to do 6936the interrupt acknowledge cycle through it to fetch the next to be 6937delivered interrupt vector using this exit. 6938 6939It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 6940external interrupt has just been delivered into the guest. User space 6941should put the acknowledged interrupt vector into the 'epr' field. 6942 6943:: 6944 6945 /* KVM_EXIT_SYSTEM_EVENT */ 6946 struct { 6947 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 6948 #define KVM_SYSTEM_EVENT_RESET 2 6949 #define KVM_SYSTEM_EVENT_CRASH 3 6950 #define KVM_SYSTEM_EVENT_WAKEUP 4 6951 #define KVM_SYSTEM_EVENT_SUSPEND 5 6952 #define KVM_SYSTEM_EVENT_SEV_TERM 6 6953 #define KVM_SYSTEM_EVENT_TDX_FATAL 7 6954 __u32 type; 6955 __u32 ndata; 6956 __u64 data[16]; 6957 } system_event; 6958 6959If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 6960a system-level event using some architecture specific mechanism (hypercall 6961or some special instruction). In case of ARM64, this is triggered using 6962HVC instruction based PSCI call from the vcpu. 6963 6964The 'type' field describes the system-level event type. 6965Valid values for 'type' are: 6966 6967 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 6968 VM. Userspace is not obliged to honour this, and if it does honour 6969 this does not need to destroy the VM synchronously (ie it may call 6970 KVM_RUN again before shutdown finally occurs). 6971 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 6972 As with SHUTDOWN, userspace can choose to ignore the request, or 6973 to schedule the reset to occur in the future and may call KVM_RUN again. 6974 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 6975 has requested a crash condition maintenance. Userspace can choose 6976 to ignore the request, or to gather VM memory core dump and/or 6977 reset/shutdown of the VM. 6978 - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination. 6979 The guest physical address of the guest's GHCB is stored in `data[0]`. 6980 - KVM_SYSTEM_EVENT_TDX_FATAL -- a TDX guest reported a fatal error state. 6981 KVM doesn't do any parsing or conversion, it just dumps 16 general-purpose 6982 registers to userspace, in ascending order of the 4-bit indices for x86-64 6983 general-purpose registers in instruction encoding, as defined in the Intel 6984 SDM. 6985 - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and 6986 KVM has recognized a wakeup event. Userspace may honor this event by 6987 marking the exiting vCPU as runnable, or deny it and call KVM_RUN again. 6988 - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested a suspension of 6989 the VM. 6990 6991If KVM_CAP_SYSTEM_EVENT_DATA is present, the 'data' field can contain 6992architecture specific information for the system-level event. Only 6993the first `ndata` items (possibly zero) of the data array are valid. 6994 6995 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2 if 6996 the guest issued a SYSTEM_RESET2 call according to v1.1 of the PSCI 6997 specification. 6998 6999 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2 7000 if the guest issued a SYSTEM_OFF2 call according to v1.3 of the PSCI 7001 specification. 7002 7003 - for RISC-V, data[0] is set to the value of the second argument of the 7004 ``sbi_system_reset`` call. 7005 7006Previous versions of Linux defined a `flags` member in this struct. The 7007field is now aliased to `data[0]`. Userspace can assume that it is only 7008written if ndata is greater than 0. 7009 7010For arm/arm64: 7011-------------- 7012 7013KVM_SYSTEM_EVENT_SUSPEND exits are enabled with the 7014KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If a guest invokes the PSCI 7015SYSTEM_SUSPEND function, KVM will exit to userspace with this event 7016type. 7017 7018It is the sole responsibility of userspace to implement the PSCI 7019SYSTEM_SUSPEND call according to ARM DEN0022D.b 5.19 "SYSTEM_SUSPEND". 7020KVM does not change the vCPU's state before exiting to userspace, so 7021the call parameters are left in-place in the vCPU registers. 7022 7023Userspace is _required_ to take action for such an exit. It must 7024either: 7025 7026 - Honor the guest request to suspend the VM. Userspace can request 7027 in-kernel emulation of suspension by setting the calling vCPU's 7028 state to KVM_MP_STATE_SUSPENDED. Userspace must configure the vCPU's 7029 state according to the parameters passed to the PSCI function when 7030 the calling vCPU is resumed. See ARM DEN0022D.b 5.19.1 "Intended use" 7031 for details on the function parameters. 7032 7033 - Deny the guest request to suspend the VM. See ARM DEN0022D.b 5.19.2 7034 "Caller responsibilities" for possible return values. 7035 7036Hibernation using the PSCI SYSTEM_OFF2 call is enabled when PSCI v1.3 7037is enabled. If a guest invokes the PSCI SYSTEM_OFF2 function, KVM will 7038exit to userspace with the KVM_SYSTEM_EVENT_SHUTDOWN event type and with 7039data[0] set to KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2. The only 7040supported hibernate type for the SYSTEM_OFF2 function is HIBERNATE_OFF. 7041 7042:: 7043 7044 /* KVM_EXIT_IOAPIC_EOI */ 7045 struct { 7046 __u8 vector; 7047 } eoi; 7048 7049Indicates that the VCPU's in-kernel local APIC received an EOI for a 7050level-triggered IOAPIC interrupt. This exit only triggers when the 7051IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 7052the userspace IOAPIC should process the EOI and retrigger the interrupt if 7053it is still asserted. Vector is the LAPIC interrupt vector for which the 7054EOI was received. 7055 7056:: 7057 7058 struct kvm_hyperv_exit { 7059 #define KVM_EXIT_HYPERV_SYNIC 1 7060 #define KVM_EXIT_HYPERV_HCALL 2 7061 #define KVM_EXIT_HYPERV_SYNDBG 3 7062 __u32 type; 7063 __u32 pad1; 7064 union { 7065 struct { 7066 __u32 msr; 7067 __u32 pad2; 7068 __u64 control; 7069 __u64 evt_page; 7070 __u64 msg_page; 7071 } synic; 7072 struct { 7073 __u64 input; 7074 __u64 result; 7075 __u64 params[2]; 7076 } hcall; 7077 struct { 7078 __u32 msr; 7079 __u32 pad2; 7080 __u64 control; 7081 __u64 status; 7082 __u64 send_page; 7083 __u64 recv_page; 7084 __u64 pending_page; 7085 } syndbg; 7086 } u; 7087 }; 7088 /* KVM_EXIT_HYPERV */ 7089 struct kvm_hyperv_exit hyperv; 7090 7091Indicates that the VCPU exits into userspace to process some tasks 7092related to Hyper-V emulation. 7093 7094Valid values for 'type' are: 7095 7096 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 7097 7098Hyper-V SynIC state change. Notification is used to remap SynIC 7099event/message pages and to enable/disable SynIC messages/events processing 7100in userspace. 7101 7102 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about 7103 7104Hyper-V Synthetic debugger state change. Notification is used to either update 7105the pending_page location or to send a control command (send the buffer located 7106in send_page or recv a buffer to recv_page). 7107 7108:: 7109 7110 /* KVM_EXIT_ARM_NISV / KVM_EXIT_ARM_LDST64B */ 7111 struct { 7112 __u64 esr_iss; 7113 __u64 fault_ipa; 7114 } arm_nisv; 7115 7116- KVM_EXIT_ARM_NISV: 7117 7118Used on arm64 systems. If a guest accesses memory not in a memslot, 7119KVM will typically return to userspace and ask it to do MMIO emulation on its 7120behalf. However, for certain classes of instructions, no instruction decode 7121(direction, length of memory access) is provided, and fetching and decoding 7122the instruction from the VM is overly complicated to live in the kernel. 7123 7124Historically, when this situation occurred, KVM would print a warning and kill 7125the VM. KVM assumed that if the guest accessed non-memslot memory, it was 7126trying to do I/O, which just couldn't be emulated, and the warning message was 7127phrased accordingly. However, what happened more often was that a guest bug 7128caused access outside the guest memory areas which should lead to a more 7129meaningful warning message and an external abort in the guest, if the access 7130did not fall within an I/O window. 7131 7132Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable 7133this capability at VM creation. Once this is done, these types of errors will 7134instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from 7135the ESR_EL2 in the esr_iss field, and the faulting IPA in the fault_ipa field. 7136Userspace can either fix up the access if it's actually an I/O access by 7137decoding the instruction from guest memory (if it's very brave) and continue 7138executing the guest, or it can decide to suspend, dump, or restart the guest. 7139 7140Note that KVM does not skip the faulting instruction as it does for 7141KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state 7142if it decides to decode and emulate the instruction. 7143 7144This feature isn't available to protected VMs, as userspace does not 7145have access to the state that is required to perform the emulation. 7146Instead, a data abort exception is directly injected in the guest. 7147Note that although KVM_CAP_ARM_NISV_TO_USER will be reported if 7148queried outside of a protected VM context, the feature will not be 7149exposed if queried on a protected VM file descriptor. 7150 7151- KVM_EXIT_ARM_LDST64B: 7152 7153Used on arm64 systems. When a guest using a LD64B, ST64B, ST64BV, ST64BV0, 7154outside of a memslot, KVM will return to userspace with KVM_EXIT_ARM_LDST64B, 7155exposing the relevant ESR_EL2 information and faulting IPA, similarly to 7156KVM_EXIT_ARM_NISV. 7157 7158Userspace is supposed to fully emulate the instructions, which includes: 7159 7160 - fetch of the operands for a store, including ACCDATA_EL1 in the case 7161 of a ST64BV0 instruction 7162 - deal with the endianness if the guest is big-endian 7163 - emulate the access, including the delivery of an exception if the 7164 access didn't succeed 7165 - provide a return value in the case of ST64BV/ST64BV0 7166 - return the data in the case of a load 7167 - increment PC if the instruction was successfully executed 7168 7169Note that there is no expectation of performance for this emulation, as it 7170involves a large number of interaction with the guest state. It is, however, 7171expected that the instruction's semantics are preserved, specially the 7172single-copy atomicity property of the 64 byte access. 7173 7174This exit reason must be handled if userspace sets ID_AA64ISAR1_EL1.LS64 to a 7175non-zero value, indicating that FEAT_LS64* is enabled. 7176 7177:: 7178 7179 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */ 7180 struct { 7181 __u8 error; /* user -> kernel */ 7182 __u8 pad[7]; 7183 __u32 reason; /* kernel -> user */ 7184 __u32 index; /* kernel -> user */ 7185 __u64 data; /* kernel <-> user */ 7186 } msr; 7187 7188Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is 7189enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code 7190may instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR 7191exit for writes. 7192 7193The "reason" field specifies why the MSR interception occurred. Userspace will 7194only receive MSR exits when a particular reason was requested during through 7195ENABLE_CAP. Currently valid exit reasons are: 7196 7197============================ ======================================== 7198 KVM_MSR_EXIT_REASON_UNKNOWN access to MSR that is unknown to KVM 7199 KVM_MSR_EXIT_REASON_INVAL access to invalid MSRs or reserved bits 7200 KVM_MSR_EXIT_REASON_FILTER access blocked by KVM_X86_SET_MSR_FILTER 7201============================ ======================================== 7202 7203For KVM_EXIT_X86_RDMSR, the "index" field tells userspace which MSR the guest 7204wants to read. To respond to this request with a successful read, userspace 7205writes the respective data into the "data" field and must continue guest 7206execution to ensure the read data is transferred into guest register state. 7207 7208If the RDMSR request was unsuccessful, userspace indicates that with a "1" in 7209the "error" field. This will inject a #GP into the guest when the VCPU is 7210executed again. 7211 7212For KVM_EXIT_X86_WRMSR, the "index" field tells userspace which MSR the guest 7213wants to write. Once finished processing the event, userspace must continue 7214vCPU execution. If the MSR write was unsuccessful, userspace also sets the 7215"error" field to "1". 7216 7217See KVM_X86_SET_MSR_FILTER for details on the interaction with MSR filtering. 7218 7219:: 7220 7221 7222 struct kvm_xen_exit { 7223 #define KVM_EXIT_XEN_HCALL 1 7224 __u32 type; 7225 union { 7226 struct { 7227 __u32 longmode; 7228 __u32 cpl; 7229 __u64 input; 7230 __u64 result; 7231 __u64 params[6]; 7232 } hcall; 7233 } u; 7234 }; 7235 /* KVM_EXIT_XEN */ 7236 struct kvm_hyperv_exit xen; 7237 7238Indicates that the VCPU exits into userspace to process some tasks 7239related to Xen emulation. 7240 7241Valid values for 'type' are: 7242 7243 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall. 7244 Userspace is expected to place the hypercall result into the appropriate 7245 field before invoking KVM_RUN again. 7246 7247:: 7248 7249 /* KVM_EXIT_RISCV_SBI */ 7250 struct { 7251 unsigned long extension_id; 7252 unsigned long function_id; 7253 unsigned long args[6]; 7254 unsigned long ret[2]; 7255 } riscv_sbi; 7256 7257If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has 7258done a SBI call which is not handled by KVM RISC-V kernel module. The details 7259of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The 7260'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the 7261'function_id' field represents function ID of given SBI extension. The 'args' 7262array field of 'riscv_sbi' represents parameters for the SBI call and 'ret' 7263array field represents return values. The userspace should update the return 7264values of SBI call before resuming the VCPU. For more details on RISC-V SBI 7265spec refer, https://github.com/riscv/riscv-sbi-doc. 7266 7267:: 7268 7269 /* KVM_EXIT_MEMORY_FAULT */ 7270 struct { 7271 #define KVM_MEMORY_EXIT_FLAG_PRIVATE (1ULL << 3) 7272 __u64 flags; 7273 __u64 gpa; 7274 __u64 size; 7275 } memory_fault; 7276 7277KVM_EXIT_MEMORY_FAULT indicates the vCPU has encountered a memory fault that 7278could not be resolved by KVM. The 'gpa' and 'size' (in bytes) describe the 7279guest physical address range [gpa, gpa + size) of the fault. The 'flags' field 7280describes properties of the faulting access that are likely pertinent: 7281 7282 - KVM_MEMORY_EXIT_FLAG_PRIVATE - When set, indicates the memory fault occurred 7283 on a private memory access. When clear, indicates the fault occurred on a 7284 shared access. 7285 7286Note! KVM_EXIT_MEMORY_FAULT is unique among all KVM exit reasons in that it 7287accompanies a return code of '-1', not '0'! errno will always be set to EFAULT 7288or EHWPOISON when KVM exits with KVM_EXIT_MEMORY_FAULT, userspace should assume 7289kvm_run.exit_reason is stale/undefined for all other error numbers. 7290 7291:: 7292 7293 /* KVM_EXIT_NOTIFY */ 7294 struct { 7295 #define KVM_NOTIFY_CONTEXT_INVALID (1 << 0) 7296 __u32 flags; 7297 } notify; 7298 7299Used on x86 systems. When the VM capability KVM_CAP_X86_NOTIFY_VMEXIT is 7300enabled, a VM exit generated if no event window occurs in VM non-root mode 7301for a specified amount of time. Once KVM_X86_NOTIFY_VMEXIT_USER is set when 7302enabling the cap, it would exit to userspace with the exit reason 7303KVM_EXIT_NOTIFY for further handling. The "flags" field contains more 7304detailed info. 7305 7306The valid value for 'flags' is: 7307 7308 - KVM_NOTIFY_CONTEXT_INVALID -- the VM context is corrupted and not valid 7309 in VMCS. It would run into unknown result if resume the target VM. 7310 7311:: 7312 7313 /* KVM_EXIT_TDX */ 7314 struct { 7315 __u64 flags; 7316 __u64 nr; 7317 union { 7318 struct { 7319 u64 ret; 7320 u64 data[5]; 7321 } unknown; 7322 struct { 7323 u64 ret; 7324 u64 gpa; 7325 u64 size; 7326 } get_quote; 7327 struct { 7328 u64 ret; 7329 u64 leaf; 7330 u64 r11, r12, r13, r14; 7331 } get_tdvmcall_info; 7332 struct { 7333 u64 ret; 7334 u64 vector; 7335 } setup_event_notify; 7336 }; 7337 } tdx; 7338 7339Process a TDVMCALL from the guest. KVM forwards select TDVMCALL based 7340on the Guest-Hypervisor Communication Interface (GHCI) specification; 7341KVM bridges these requests to the userspace VMM with minimal changes, 7342placing the inputs in the union and copying them back to the guest 7343on re-entry. 7344 7345Flags are currently always zero, whereas ``nr`` contains the TDVMCALL 7346number from register R11. The remaining field of the union provide the 7347inputs and outputs of the TDVMCALL. Currently the following values of 7348``nr`` are defined: 7349 7350 * ``TDVMCALL_GET_QUOTE``: the guest has requested to generate a TD-Quote 7351 signed by a service hosting TD-Quoting Enclave operating on the host. 7352 Parameters and return value are in the ``get_quote`` field of the union. 7353 The ``gpa`` field and ``size`` specify the guest physical address 7354 (without the shared bit set) and the size of a shared-memory buffer, in 7355 which the TDX guest passes a TD Report. The ``ret`` field represents 7356 the return value of the GetQuote request. When the request has been 7357 queued successfully, the TDX guest can poll the status field in the 7358 shared-memory area to check whether the Quote generation is completed or 7359 not. When completed, the generated Quote is returned via the same buffer. 7360 7361 * ``TDVMCALL_GET_TD_VM_CALL_INFO``: the guest has requested the support 7362 status of TDVMCALLs. The output values for the given leaf should be 7363 placed in fields from ``r11`` to ``r14`` of the ``get_tdvmcall_info`` 7364 field of the union. 7365 7366 * ``TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT``: the guest has requested to 7367 set up a notification interrupt for vector ``vector``. 7368 7369KVM may add support for more values in the future that may cause a userspace 7370exit, even without calls to ``KVM_ENABLE_CAP`` or similar. In this case, 7371it will enter with output fields already valid; in the common case, the 7372``unknown.ret`` field of the union will be ``TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED``. 7373Userspace need not do anything if it does not wish to support a TDVMCALL. 7374 7375:: 7376 7377 /* KVM_EXIT_ARM_SEA */ 7378 struct { 7379 #define KVM_EXIT_ARM_SEA_FLAG_GPA_VALID (1ULL << 0) 7380 __u64 flags; 7381 __u64 esr; 7382 __u64 gva; 7383 __u64 gpa; 7384 } arm_sea; 7385 7386Used on arm64 systems. When the VM capability ``KVM_CAP_ARM_SEA_TO_USER`` is 7387enabled, a KVM exits to userspace if a guest access causes a synchronous 7388external abort (SEA) and the host APEI fails to handle the SEA. 7389 7390``esr`` is set to a sanitized value of ESR_EL2 from the exception taken to KVM, 7391consisting of the following fields: 7392 7393 - ``ESR_EL2.EC`` 7394 - ``ESR_EL2.IL`` 7395 - ``ESR_EL2.FnV`` 7396 - ``ESR_EL2.EA`` 7397 - ``ESR_EL2.CM`` 7398 - ``ESR_EL2.WNR`` 7399 - ``ESR_EL2.FSC`` 7400 - ``ESR_EL2.SET`` (when FEAT_RAS is implemented for the VM) 7401 7402``gva`` is set to the value of FAR_EL2 from the exception taken to KVM when 7403``ESR_EL2.FnV == 0``. Otherwise, the value of ``gva`` is unknown. 7404 7405``gpa`` is set to the faulting IPA from the exception taken to KVM when 7406the ``KVM_EXIT_ARM_SEA_FLAG_GPA_VALID`` flag is set. Otherwise, the value of 7407``gpa`` is unknown. 7408 7409:: 7410 7411 /* Fix the size of the union. */ 7412 char padding[256]; 7413 }; 7414 7415 /* 7416 * shared registers between kvm and userspace. 7417 * kvm_valid_regs specifies the register classes set by the host 7418 * kvm_dirty_regs specified the register classes dirtied by userspace 7419 * struct kvm_sync_regs is architecture specific, as well as the 7420 * bits for kvm_valid_regs and kvm_dirty_regs 7421 */ 7422 __u64 kvm_valid_regs; 7423 __u64 kvm_dirty_regs; 7424 union { 7425 struct kvm_sync_regs regs; 7426 char padding[SYNC_REGS_SIZE_BYTES]; 7427 } s; 7428 7429If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 7430certain guest registers without having to call SET/GET_*REGS. Thus we can 7431avoid some system call overhead if userspace has to handle the exit. 7432Userspace can query the validity of the structure by checking 7433kvm_valid_regs for specific bits. These bits are architecture specific 7434and usually define the validity of a groups of registers. (e.g. one bit 7435for general purpose registers) 7436 7437Please note that the kernel is allowed to use the kvm_run structure as the 7438primary storage for certain register types. Therefore, the kernel may use the 7439values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 7440 7441:: 7442 7443 /* KVM_EXIT_SNP_REQ_CERTS */ 7444 struct kvm_exit_snp_req_certs { 7445 __u64 gpa; 7446 __u64 npages; 7447 __u64 ret; 7448 }; 7449 7450KVM_EXIT_SNP_REQ_CERTS indicates an SEV-SNP guest with certificate-fetching 7451enabled (see KVM_SEV_SNP_ENABLE_REQ_CERTS) has generated an Extended Guest 7452Request NAE #VMGEXIT (SNP_GUEST_REQUEST) with message type MSG_REPORT_REQ, 7453i.e. has requested an attestation report from firmware, and would like the 7454certificate data corresponding to the attestation report signature to be 7455provided by the hypervisor as part of the request. 7456 7457To allow for userspace to provide the certificate, the 'gpa' and 'npages' 7458are forwarded verbatim from the guest request (the RAX and RBX GHCB fields 7459respectively). 'ret' is not an "output" from KVM, and is always '0' on 7460exit. KVM verifies the 'gpa' is 4KiB aligned prior to exiting to userspace, 7461but otherwise the information from the guest isn't validated. 7462 7463Upon the next KVM_RUN, e.g. after userspace has serviced the request (or not), 7464KVM will complete the #VMGEXIT, using the 'ret' field to determine whether to 7465signal success or failure to the guest, and on failure, what reason code will 7466be communicated via SW_EXITINFO2. If 'ret' is set to an unsupported value (see 7467the table below), KVM_RUN will fail with -EINVAL. For a 'ret' of 'ENOSPC', KVM 7468also consumes the 'npages' field, i.e. userspace can use the field to inform 7469the guest of the number of pages needed to hold all the certificate data. 7470 7471The supported 'ret' values and their respective SW_EXITINFO2 encodings: 7472 7473 ====== ============================================================= 7474 0 0x0, i.e. success. KVM will emit an SNP_GUEST_REQUEST command 7475 to SNP firmware. 7476 ENOSPC 0x0000000100000000, i.e. not enough guest pages to hold the 7477 certificate table and certificate data. KVM will also set the 7478 RBX field in the GHBC to 'npages'. 7479 EAGAIN 0x0000000200000000, i.e. the host is busy and the guest should 7480 retry the request. 7481 EIO 0xffffffff00000000, for all other errors (this return code is 7482 a KVM-defined hypervisor value, as allowed by the GHCB) 7483 ====== ============================================================= 7484 7485 7486.. _cap_enable: 7487 74886. Capabilities that can be enabled on vCPUs 7489============================================ 7490 7491There are certain capabilities that change the behavior of the virtual CPU or 7492the virtual machine when enabled. To enable them, please see 7493:ref:`KVM_ENABLE_CAP`. 7494 7495Below you can find a list of capabilities and what their effect on the vCPU or 7496the virtual machine is when enabling them. 7497 7498The following information is provided along with the description: 7499 7500 Architectures: 7501 which instruction set architectures provide this ioctl. 7502 x86 includes both i386 and x86_64. 7503 7504 Target: 7505 whether this is a per-vcpu or per-vm capability. 7506 7507 Parameters: 7508 what parameters are accepted by the capability. 7509 7510 Returns: 7511 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 7512 are not detailed, but errors with specific meanings are. 7513 7514 75156.1 KVM_CAP_PPC_OSI 7516------------------- 7517 7518:Architectures: ppc 7519:Target: vcpu 7520:Parameters: none 7521:Returns: 0 on success; -1 on error 7522 7523This capability enables interception of OSI hypercalls that otherwise would 7524be treated as normal system calls to be injected into the guest. OSI hypercalls 7525were invented by Mac-on-Linux to have a standardized communication mechanism 7526between the guest and the host. 7527 7528When this capability is enabled, KVM_EXIT_OSI can occur. 7529 7530 75316.2 KVM_CAP_PPC_PAPR 7532-------------------- 7533 7534:Architectures: ppc 7535:Target: vcpu 7536:Parameters: none 7537:Returns: 0 on success; -1 on error 7538 7539This capability enables interception of PAPR hypercalls. PAPR hypercalls are 7540done using the hypercall instruction "sc 1". 7541 7542It also sets the guest privilege level to "supervisor" mode. Usually the guest 7543runs in "hypervisor" privilege mode with a few missing features. 7544 7545In addition to the above, it changes the semantics of SDR1. In this mode, the 7546HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 7547HTAB invisible to the guest. 7548 7549When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 7550 7551 75526.3 KVM_CAP_SW_TLB 7553------------------ 7554 7555:Architectures: ppc 7556:Target: vcpu 7557:Parameters: args[0] is the address of a struct kvm_config_tlb 7558:Returns: 0 on success; -1 on error 7559 7560:: 7561 7562 struct kvm_config_tlb { 7563 __u64 params; 7564 __u64 array; 7565 __u32 mmu_type; 7566 __u32 array_len; 7567 }; 7568 7569Configures the virtual CPU's TLB array, establishing a shared memory area 7570between userspace and KVM. The "params" and "array" fields are userspace 7571addresses of mmu-type-specific data structures. The "array_len" field is an 7572safety mechanism, and should be set to the size in bytes of the memory that 7573userspace has reserved for the array. It must be at least the size dictated 7574by "mmu_type" and "params". 7575 7576While KVM_RUN is active, the shared region is under control of KVM. Its 7577contents are undefined, and any modification by userspace results in 7578boundedly undefined behavior. 7579 7580On return from KVM_RUN, the shared region will reflect the current state of 7581the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 7582to tell KVM which entries have been changed, prior to calling KVM_RUN again 7583on this vcpu. 7584 7585For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 7586 7587 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 7588 - The "array" field points to an array of type "struct 7589 kvm_book3e_206_tlb_entry". 7590 - The array consists of all entries in the first TLB, followed by all 7591 entries in the second TLB. 7592 - Within a TLB, entries are ordered first by increasing set number. Within a 7593 set, entries are ordered by way (increasing ESEL). 7594 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 7595 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 7596 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 7597 hardware ignores this value for TLB0. 7598 75996.4 KVM_CAP_S390_CSS_SUPPORT 7600---------------------------- 7601 7602:Architectures: s390 7603:Target: vcpu 7604:Parameters: none 7605:Returns: 0 on success; -1 on error 7606 7607This capability enables support for handling of channel I/O instructions. 7608 7609TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 7610handled in-kernel, while the other I/O instructions are passed to userspace. 7611 7612When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 7613SUBCHANNEL intercepts. 7614 7615Note that even though this capability is enabled per-vcpu, the complete 7616virtual machine is affected. 7617 76186.5 KVM_CAP_PPC_EPR 7619------------------- 7620 7621:Architectures: ppc 7622:Target: vcpu 7623:Parameters: args[0] defines whether the proxy facility is active 7624:Returns: 0 on success; -1 on error 7625 7626This capability enables or disables the delivery of interrupts through the 7627external proxy facility. 7628 7629When enabled (args[0] != 0), every time the guest gets an external interrupt 7630delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 7631to receive the topmost interrupt vector. 7632 7633When disabled (args[0] == 0), behavior is as if this facility is unsupported. 7634 7635When this capability is enabled, KVM_EXIT_EPR can occur. 7636 76376.6 KVM_CAP_IRQ_MPIC 7638-------------------- 7639 7640:Architectures: ppc 7641:Parameters: args[0] is the MPIC device fd; 7642 args[1] is the MPIC CPU number for this vcpu 7643 7644This capability connects the vcpu to an in-kernel MPIC device. 7645 76466.7 KVM_CAP_IRQ_XICS 7647-------------------- 7648 7649:Architectures: ppc 7650:Target: vcpu 7651:Parameters: args[0] is the XICS device fd; 7652 args[1] is the XICS CPU number (server ID) for this vcpu 7653 7654This capability connects the vcpu to an in-kernel XICS device. 7655 76566.8 KVM_CAP_S390_IRQCHIP 7657------------------------ 7658 7659:Architectures: s390 7660:Target: vm 7661:Parameters: none 7662 7663This capability enables the in-kernel irqchip for s390. Please refer to 7664"4.24 KVM_CREATE_IRQCHIP" for details. 7665 76666.9 KVM_CAP_MIPS_FPU 7667-------------------- 7668 7669:Architectures: mips 7670:Target: vcpu 7671:Parameters: args[0] is reserved for future use (should be 0). 7672 7673This capability allows the use of the host Floating Point Unit by the guest. It 7674allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 7675done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be 7676accessed (depending on the current guest FPU register mode), and the Status.FR, 7677Config5.FRE bits are accessible via the KVM API and also from the guest, 7678depending on them being supported by the FPU. 7679 76806.10 KVM_CAP_MIPS_MSA 7681--------------------- 7682 7683:Architectures: mips 7684:Target: vcpu 7685:Parameters: args[0] is reserved for future use (should be 0). 7686 7687This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 7688It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 7689Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*`` 7690registers can be accessed, and the Config5.MSAEn bit is accessible via the 7691KVM API and also from the guest. 7692 76936.74 KVM_CAP_SYNC_REGS 7694---------------------- 7695 7696:Architectures: s390, x86 7697:Target: s390: always enabled, x86: vcpu 7698:Parameters: none 7699:Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register 7700 sets are supported 7701 (bitfields defined in arch/x86/include/uapi/asm/kvm.h). 7702 7703As described above in the kvm_sync_regs struct info in section :ref:`kvm_run`, 7704KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers 7705without having to call SET/GET_*REGS". This reduces overhead by eliminating 7706repeated ioctl calls for setting and/or getting register values. This is 7707particularly important when userspace is making synchronous guest state 7708modifications, e.g. when emulating and/or intercepting instructions in 7709userspace. 7710 7711For s390 specifics, please refer to the source code. 7712 7713For x86: 7714 7715- the register sets to be copied out to kvm_run are selectable 7716 by userspace (rather that all sets being copied out for every exit). 7717- vcpu_events are available in addition to regs and sregs. 7718 7719For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to 7720function as an input bit-array field set by userspace to indicate the 7721specific register sets to be copied out on the next exit. 7722 7723To indicate when userspace has modified values that should be copied into 7724the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set. 7725This is done using the same bitflags as for the 'kvm_valid_regs' field. 7726If the dirty bit is not set, then the register set values will not be copied 7727into the vCPU even if they've been modified. 7728 7729Unused bitfields in the bitarrays must be set to zero. 7730 7731:: 7732 7733 struct kvm_sync_regs { 7734 struct kvm_regs regs; 7735 struct kvm_sregs sregs; 7736 struct kvm_vcpu_events events; 7737 }; 7738 77396.75 KVM_CAP_PPC_IRQ_XIVE 7740------------------------- 7741 7742:Architectures: ppc 7743:Target: vcpu 7744:Parameters: args[0] is the XIVE device fd; 7745 args[1] is the XIVE CPU number (server ID) for this vcpu 7746 7747This capability connects the vcpu to an in-kernel XIVE device. 7748 77496.76 KVM_CAP_HYPERV_SYNIC 7750------------------------- 7751 7752:Architectures: x86 7753:Target: vcpu 7754 7755This capability, if KVM_CHECK_EXTENSION indicates that it is 7756available, means that the kernel has an implementation of the 7757Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 7758used to support Windows Hyper-V based guest paravirt drivers(VMBus). 7759 7760In order to use SynIC, it has to be activated by setting this 7761capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 7762will disable the use of APIC hardware virtualization even if supported 7763by the CPU, as it's incompatible with SynIC auto-EOI behavior. 7764 77656.77 KVM_CAP_HYPERV_SYNIC2 7766-------------------------- 7767 7768:Architectures: x86 7769:Target: vcpu 7770 7771This capability enables a newer version of Hyper-V Synthetic interrupt 7772controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 7773doesn't clear SynIC message and event flags pages when they are enabled by 7774writing to the respective MSRs. 7775 77766.78 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 7777----------------------------------- 7778 7779:Architectures: x86 7780:Target: vcpu 7781 7782This capability indicates that KVM running on top of Hyper-V hypervisor 7783enables Direct TLB flush for its guests meaning that TLB flush 7784hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM. 7785Due to the different ABI for hypercall parameters between Hyper-V and 7786KVM, enabling this capability effectively disables all hypercall 7787handling by KVM (as some KVM hypercall may be mistakenly treated as TLB 7788flush hypercalls by Hyper-V) so userspace should disable KVM identification 7789in CPUID and only exposes Hyper-V identification. In this case, guest 7790thinks it's running on Hyper-V and only use Hyper-V hypercalls. 7791 77926.79 KVM_CAP_HYPERV_ENFORCE_CPUID 7793--------------------------------- 7794 7795:Architectures: x86 7796:Target: vcpu 7797 7798When enabled, KVM will disable emulated Hyper-V features provided to the 7799guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all 7800currently implemented Hyper-V features are provided unconditionally when 7801Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001) 7802leaf. 7803 78046.80 KVM_CAP_ENFORCE_PV_FEATURE_CPUID 7805------------------------------------- 7806 7807:Architectures: x86 7808:Target: vcpu 7809 7810When enabled, KVM will disable paravirtual features provided to the 7811guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf 7812(0x40000001). Otherwise, a guest may use the paravirtual features 7813regardless of what has actually been exposed through the CPUID leaf. 7814 7815.. _KVM_CAP_DIRTY_LOG_RING: 7816 7817 7818.. _cap_enable_vm: 7819 78207. Capabilities that can be enabled on VMs 7821========================================== 7822 7823There are certain capabilities that change the behavior of the virtual 7824machine when enabled. To enable them, please see section 7825:ref:`KVM_ENABLE_CAP`. Below you can find a list of capabilities and 7826what their effect on the VM is when enabling them. 7827 7828The following information is provided along with the description: 7829 7830 Architectures: 7831 which instruction set architectures provide this ioctl. 7832 x86 includes both i386 and x86_64. 7833 7834 Parameters: 7835 what parameters are accepted by the capability. 7836 7837 Returns: 7838 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 7839 are not detailed, but errors with specific meanings are. 7840 7841 78427.1 KVM_CAP_PPC_ENABLE_HCALL 7843---------------------------- 7844 7845:Architectures: ppc 7846:Parameters: args[0] is the sPAPR hcall number; 7847 args[1] is 0 to disable, 1 to enable in-kernel handling 7848 7849This capability controls whether individual sPAPR hypercalls (hcalls) 7850get handled by the kernel or not. Enabling or disabling in-kernel 7851handling of an hcall is effective across the VM. On creation, an 7852initial set of hcalls are enabled for in-kernel handling, which 7853consists of those hcalls for which in-kernel handlers were implemented 7854before this capability was implemented. If disabled, the kernel will 7855not to attempt to handle the hcall, but will always exit to userspace 7856to handle it. Note that it may not make sense to enable some and 7857disable others of a group of related hcalls, but KVM does not prevent 7858userspace from doing that. 7859 7860If the hcall number specified is not one that has an in-kernel 7861implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 7862error. 7863 78647.2 KVM_CAP_S390_USER_SIGP 7865-------------------------- 7866 7867:Architectures: s390 7868:Parameters: none 7869 7870This capability controls which SIGP orders will be handled completely in user 7871space. With this capability enabled, all fast orders will be handled completely 7872in the kernel: 7873 7874- SENSE 7875- SENSE RUNNING 7876- EXTERNAL CALL 7877- EMERGENCY SIGNAL 7878- CONDITIONAL EMERGENCY SIGNAL 7879 7880All other orders will be handled completely in user space. 7881 7882Only privileged operation exceptions will be checked for in the kernel (or even 7883in the hardware prior to interception). If this capability is not enabled, the 7884old way of handling SIGP orders is used (partially in kernel and user space). 7885 78867.3 KVM_CAP_S390_VECTOR_REGISTERS 7887--------------------------------- 7888 7889:Architectures: s390 7890:Parameters: none 7891:Returns: 0 on success, negative value on error 7892 7893Allows use of the vector registers introduced with z13 processor, and 7894provides for the synchronization between host and user space. Will 7895return -EINVAL if the machine does not support vectors. 7896 78977.4 KVM_CAP_S390_USER_STSI 7898-------------------------- 7899 7900:Architectures: s390 7901:Parameters: none 7902 7903This capability allows post-handlers for the STSI instruction. After 7904initial handling in the kernel, KVM exits to user space with 7905KVM_EXIT_S390_STSI to allow user space to insert further data. 7906 7907Before exiting to userspace, kvm handlers should fill in s390_stsi field of 7908vcpu->run:: 7909 7910 struct { 7911 __u64 addr; 7912 __u8 ar; 7913 __u8 reserved; 7914 __u8 fc; 7915 __u8 sel1; 7916 __u16 sel2; 7917 } s390_stsi; 7918 7919 @addr - guest address of STSI SYSIB 7920 @fc - function code 7921 @sel1 - selector 1 7922 @sel2 - selector 2 7923 @ar - access register number 7924 7925KVM handlers should exit to userspace with rc = -EREMOTE. 7926 79277.5 KVM_CAP_SPLIT_IRQCHIP 7928------------------------- 7929 7930:Architectures: x86 7931:Parameters: args[0] - number of routes reserved for userspace IOAPICs 7932:Returns: 0 on success, -1 on error 7933 7934Create a local apic for each processor in the kernel. This can be used 7935instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 7936IOAPIC and PIC (and also the PIT, even though this has to be enabled 7937separately). 7938 7939This capability also enables in kernel routing of interrupt requests; 7940when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 7941used in the IRQ routing table. The first args[0] MSI routes are reserved 7942for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 7943a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 7944 7945As with ``KVM_CREATE_IRQCHIP``, subsequent vcpu creation may install a private 7946memory slot at the APIC base address (0xfee00000) that must not overlap user 7947memory regions. See ``KVM_CREATE_IRQCHIP`` for details. 7948 7949Fails if VCPU has already been created, or if the irqchip is already in the 7950kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 7951 79527.6 KVM_CAP_S390_RI 7953------------------- 7954 7955:Architectures: s390 7956:Parameters: none 7957 7958Allows use of runtime-instrumentation introduced with zEC12 processor. 7959Will return -EINVAL if the machine does not support runtime-instrumentation. 7960Will return -EBUSY if a VCPU has already been created. 7961 79627.7 KVM_CAP_X2APIC_API 7963---------------------- 7964 7965:Architectures: x86 7966:Parameters: args[0] - features that should be enabled 7967:Returns: 0 on success, -EINVAL when args[0] contains invalid features 7968 7969Valid feature flags in args[0] are:: 7970 7971 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 7972 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 7973 #define KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST (1ULL << 2) 7974 #define KVM_X2APIC_DISABLE_SUPPRESS_EOI_BROADCAST (1ULL << 3) 7975 7976Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 7977KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 7978allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 7979respective sections. 7980 7981KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 7982in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 7983as a broadcast even in x2APIC mode in order to support physical x2APIC 7984without interrupt remapping. This is undesirable in logical mode, 7985where 0xff represents CPUs 0-7 in cluster 0. 7986 7987Setting KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST instructs KVM to enable 7988Suppress EOI Broadcasts. KVM will advertise support for Suppress EOI 7989Broadcast to the guest and suppress LAPIC EOI broadcasts when the guest 7990sets the Suppress EOI Broadcast bit in the SPIV register. This flag is 7991supported only when using a split IRQCHIP. 7992 7993Setting KVM_X2APIC_DISABLE_SUPPRESS_EOI_BROADCAST disables support for 7994Suppress EOI Broadcasts entirely, i.e. instructs KVM to NOT advertise 7995support to the guest. 7996 7997Modern VMMs should either enable KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST 7998or KVM_X2APIC_DISABLE_SUPPRESS_EOI_BROADCAST. If not, legacy quirky 7999behavior will be used by KVM: in split IRQCHIP mode, KVM will advertise 8000support for Suppress EOI Broadcasts but not actually suppress EOI 8001broadcasts; for in-kernel IRQCHIP mode, KVM will not advertise support for 8002Suppress EOI Broadcasts. 8003 8004Setting both KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST and 8005KVM_X2APIC_DISABLE_SUPPRESS_EOI_BROADCAST will fail with an EINVAL error, 8006as will setting KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST without a split 8007IRCHIP. 8008 80097.8 KVM_CAP_S390_USER_INSTR0 8010---------------------------- 8011 8012:Architectures: s390 8013:Parameters: none 8014 8015With this capability enabled, the illegal instruction 0x0000 (2 bytes) will 8016be intercepted and forwarded to user space. User space can use this 8017mechanism e.g. to realize 2-byte software breakpoints. The kernel will 8018not inject an operating exception for these instructions, user space has 8019to take care of that. 8020 8021This capability can be enabled dynamically even if VCPUs were already 8022created and are running. 8023 80247.9 KVM_CAP_S390_GS 8025------------------- 8026 8027:Architectures: s390 8028:Parameters: none 8029:Returns: 0 on success; -EINVAL if the machine does not support 8030 guarded storage; -EBUSY if a VCPU has already been created. 8031 8032Allows use of guarded storage for the KVM guest. 8033 80347.10 KVM_CAP_S390_AIS 8035--------------------- 8036 8037:Architectures: s390 8038:Parameters: none 8039 8040Allow use of adapter-interruption suppression. 8041:Returns: 0 on success; -EBUSY if a VCPU has already been created. 8042 80437.11 KVM_CAP_PPC_SMT 8044-------------------- 8045 8046:Architectures: ppc 8047:Parameters: vsmt_mode, flags 8048 8049Enabling this capability on a VM provides userspace with a way to set 8050the desired virtual SMT mode (i.e. the number of virtual CPUs per 8051virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 8052between 1 and 8. On POWER8, vsmt_mode must also be no greater than 8053the number of threads per subcore for the host. Currently flags must 8054be 0. A successful call to enable this capability will result in 8055vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 8056subsequently queried for the VM. This capability is only supported by 8057HV KVM, and can only be set before any VCPUs have been created. 8058The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 8059modes are available. 8060 80617.12 KVM_CAP_PPC_FWNMI 8062---------------------- 8063 8064:Architectures: ppc 8065:Parameters: none 8066 8067With this capability a machine check exception in the guest address 8068space will cause KVM to exit the guest with NMI exit reason. This 8069enables QEMU to build error log and branch to guest kernel registered 8070machine check handling routine. Without this capability KVM will 8071branch to guests' 0x200 interrupt vector. 8072 80737.13 KVM_CAP_X86_DISABLE_EXITS 8074------------------------------ 8075 8076:Architectures: x86 8077:Parameters: args[0] defines which exits are disabled 8078:Returns: 0 on success, -EINVAL when args[0] contains invalid exits 8079 or if any vCPUs have already been created 8080 8081Valid bits in args[0] are:: 8082 8083 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) 8084 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1) 8085 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) 8086 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3) 8087 #define KVM_X86_DISABLE_EXITS_APERFMPERF (1 << 4) 8088 8089Enabling this capability on a VM provides userspace with a way to no 8090longer intercept some instructions for improved latency in some 8091workloads, and is suggested when vCPUs are associated to dedicated 8092physical CPUs. More bits can be added in the future; userspace can 8093just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable 8094all such vmexits. 8095 8096Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits. 8097 8098Virtualizing the ``IA32_APERF`` and ``IA32_MPERF`` MSRs requires more 8099than just disabling APERF/MPERF exits. While both Intel and AMD 8100document strict usage conditions for these MSRs--emphasizing that only 8101the ratio of their deltas over a time interval (T0 to T1) is 8102architecturally defined--simply passing through the MSRs can still 8103produce an incorrect ratio. 8104 8105This erroneous ratio can occur if, between T0 and T1: 8106 81071. The vCPU thread migrates between logical processors. 81082. Live migration or suspend/resume operations take place. 81093. Another task shares the vCPU's logical processor. 81104. C-states lower than C0 are emulated (e.g., via HLT interception). 81115. The guest TSC frequency doesn't match the host TSC frequency. 8112 8113Due to these complexities, KVM does not automatically associate this 8114passthrough capability with the guest CPUID bit, 8115``CPUID.6:ECX.APERFMPERF[bit 0]``. Userspace VMMs that deem this 8116mechanism adequate for virtualizing the ``IA32_APERF`` and 8117``IA32_MPERF`` MSRs must set the guest CPUID bit explicitly. 8118 8119 81207.14 KVM_CAP_S390_HPAGE_1M 8121-------------------------- 8122 8123:Architectures: s390 8124:Parameters: none 8125:Returns: 0 on success, -EINVAL if hpage module parameter was not set 8126 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 8127 flag set 8128 8129With this capability the KVM support for memory backing with 1m pages 8130through hugetlbfs can be enabled for a VM. After the capability is 8131enabled, cmma can't be enabled anymore and pfmfi and the storage key 8132interpretation are disabled. If cmma has already been enabled or the 8133hpage module parameter is not set to 1, -EINVAL is returned. 8134 8135While it is generally possible to create a huge page backed VM without 8136this capability, the VM will not be able to run. 8137 81387.15 KVM_CAP_MSR_PLATFORM_INFO 8139------------------------------ 8140 8141:Architectures: x86 8142:Parameters: args[0] whether feature should be enabled or not 8143 8144With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise, 8145a #GP would be raised when the guest tries to access. Currently, this 8146capability does not enable write permissions of this MSR for the guest. 8147 81487.16 KVM_CAP_PPC_NESTED_HV 8149-------------------------- 8150 8151:Architectures: ppc 8152:Parameters: none 8153:Returns: 0 on success, -EINVAL when the implementation doesn't support 8154 nested-HV virtualization. 8155 8156HV-KVM on POWER9 and later systems allows for "nested-HV" 8157virtualization, which provides a way for a guest VM to run guests that 8158can run using the CPU's supervisor mode (privileged non-hypervisor 8159state). Enabling this capability on a VM depends on the CPU having 8160the necessary functionality and on the facility being enabled with a 8161kvm-hv module parameter. 8162 81637.17 KVM_CAP_EXCEPTION_PAYLOAD 8164------------------------------ 8165 8166:Architectures: x86 8167:Parameters: args[0] whether feature should be enabled or not 8168 8169With this capability enabled, CR2 will not be modified prior to the 8170emulated VM-exit when L1 intercepts a #PF exception that occurs in 8171L2. Similarly, for kvm-intel only, DR6 will not be modified prior to 8172the emulated VM-exit when L1 intercepts a #DB exception that occurs in 8173L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or 8174#DB) exception for L2, exception.has_payload will be set and the 8175faulting address (or the new DR6 bits*) will be reported in the 8176exception_payload field. Similarly, when userspace injects a #PF (or 8177#DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set 8178exception.has_payload and to put the faulting address - or the new DR6 8179bits\ [#]_ - in the exception_payload field. 8180 8181This capability also enables exception.pending in struct 8182kvm_vcpu_events, which allows userspace to distinguish between pending 8183and injected exceptions. 8184 8185 8186.. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception 8187 will clear DR6.RTM. 8188 81897.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 8190-------------------------------------- 8191 8192:Architectures: x86, arm64, mips 8193:Parameters: args[0] whether feature should be enabled or not 8194 8195Valid flags are:: 8196 8197 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0) 8198 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1) 8199 8200With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not 8201automatically clear and write-protect all pages that are returned as dirty. 8202Rather, userspace will have to do this operation separately using 8203KVM_CLEAR_DIRTY_LOG. 8204 8205At the cost of a slightly more complicated operation, this provides better 8206scalability and responsiveness for two reasons. First, 8207KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather 8208than requiring to sync a full memslot; this ensures that KVM does not 8209take spinlocks for an extended period of time. Second, in some cases a 8210large amount of time can pass between a call to KVM_GET_DIRTY_LOG and 8211userspace actually using the data in the page. Pages can be modified 8212during this time, which is inefficient for both the guest and userspace: 8213the guest will incur a higher penalty due to write protection faults, 8214while userspace can see false reports of dirty pages. Manual reprotection 8215helps reducing this time, improving guest performance and reducing the 8216number of dirty log false positives. 8217 8218With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap 8219will be initialized to 1 when created. This also improves performance because 8220dirty logging can be enabled gradually in small chunks on the first call 8221to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on 8222KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on 8223x86, arm64 and riscv for now). 8224 8225KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name 8226KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make 8227it hard or impossible to use it correctly. The availability of 8228KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed. 8229Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT. 8230 82317.19 KVM_CAP_PPC_SECURE_GUEST 8232------------------------------ 8233 8234:Architectures: ppc 8235 8236This capability indicates that KVM is running on a host that has 8237ultravisor firmware and thus can support a secure guest. On such a 8238system, a guest can ask the ultravisor to make it a secure guest, 8239one whose memory is inaccessible to the host except for pages which 8240are explicitly requested to be shared with the host. The ultravisor 8241notifies KVM when a guest requests to become a secure guest, and KVM 8242has the opportunity to veto the transition. 8243 8244If present, this capability can be enabled for a VM, meaning that KVM 8245will allow the transition to secure guest mode. Otherwise KVM will 8246veto the transition. 8247 82487.20 KVM_CAP_HALT_POLL 8249---------------------- 8250 8251:Architectures: all 8252:Target: VM 8253:Parameters: args[0] is the maximum poll time in nanoseconds 8254:Returns: 0 on success; -1 on error 8255 8256KVM_CAP_HALT_POLL overrides the kvm.halt_poll_ns module parameter to set the 8257maximum halt-polling time for all vCPUs in the target VM. This capability can 8258be invoked at any time and any number of times to dynamically change the 8259maximum halt-polling time. 8260 8261See Documentation/virt/kvm/halt-polling.rst for more information on halt 8262polling. 8263 82647.21 KVM_CAP_X86_USER_SPACE_MSR 8265------------------------------- 8266 8267:Architectures: x86 8268:Target: VM 8269:Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report 8270:Returns: 0 on success; -1 on error 8271 8272This capability allows userspace to intercept RDMSR and WRMSR instructions if 8273access to an MSR is denied. By default, KVM injects #GP on denied accesses. 8274 8275When a guest requests to read or write an MSR, KVM may not implement all MSRs 8276that are relevant to a respective system. It also does not differentiate by 8277CPU type. 8278 8279To allow more fine grained control over MSR handling, userspace may enable 8280this capability. With it enabled, MSR accesses that match the mask specified in 8281args[0] and would trigger a #GP inside the guest will instead trigger 8282KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications. Userspace 8283can then implement model specific MSR handling and/or user notifications 8284to inform a user that an MSR was not emulated/virtualized by KVM. 8285 8286The valid mask flags are: 8287 8288============================ =============================================== 8289 KVM_MSR_EXIT_REASON_UNKNOWN intercept accesses to unknown (to KVM) MSRs 8290 KVM_MSR_EXIT_REASON_INVAL intercept accesses that are architecturally 8291 invalid according to the vCPU model and/or mode 8292 KVM_MSR_EXIT_REASON_FILTER intercept accesses that are denied by userspace 8293 via KVM_X86_SET_MSR_FILTER 8294============================ =============================================== 8295 82967.22 KVM_CAP_X86_BUS_LOCK_EXIT 8297------------------------------- 8298 8299:Architectures: x86 8300:Target: VM 8301:Parameters: args[0] defines the policy used when bus locks detected in guest 8302:Returns: 0 on success, -EINVAL when args[0] contains invalid bits 8303 8304Valid bits in args[0] are:: 8305 8306 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0) 8307 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1) 8308 8309Enabling this capability on a VM provides userspace with a way to select a 8310policy to handle the bus locks detected in guest. Userspace can obtain the 8311supported modes from the result of KVM_CHECK_EXTENSION and define it through 8312the KVM_ENABLE_CAP. The supported modes are mutually-exclusive. 8313 8314This capability allows userspace to force VM exits on bus locks detected in the 8315guest, irrespective whether or not the host has enabled split-lock detection 8316(which triggers an #AC exception that KVM intercepts). This capability is 8317intended to mitigate attacks where a malicious/buggy guest can exploit bus 8318locks to degrade the performance of the whole system. 8319 8320If KVM_BUS_LOCK_DETECTION_OFF is set, KVM doesn't force guest bus locks to VM 8321exit, although the host kernel's split-lock #AC detection still applies, if 8322enabled. 8323 8324If KVM_BUS_LOCK_DETECTION_EXIT is set, KVM enables a CPU feature that ensures 8325bus locks in the guest trigger a VM exit, and KVM exits to userspace for all 8326such VM exits, e.g. to allow userspace to throttle the offending guest and/or 8327apply some other policy-based mitigation. When exiting to userspace, KVM sets 8328KVM_RUN_X86_BUS_LOCK in vcpu-run->flags, and conditionally sets the exit_reason 8329to KVM_EXIT_X86_BUS_LOCK. 8330 8331Due to differences in the underlying hardware implementation, the vCPU's RIP at 8332the time of exit diverges between Intel and AMD. On Intel hosts, RIP points at 8333the next instruction, i.e. the exit is trap-like. On AMD hosts, RIP points at 8334the offending instruction, i.e. the exit is fault-like. 8335 8336Note! Detected bus locks may be coincident with other exits to userspace, i.e. 8337KVM_RUN_X86_BUS_LOCK should be checked regardless of the primary exit reason if 8338userspace wants to take action on all detected bus locks. 8339 83407.23 KVM_CAP_PPC_DAWR1 8341---------------------- 8342 8343:Architectures: ppc 8344:Parameters: none 8345:Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR 8346 8347This capability can be used to check / enable 2nd DAWR feature provided 8348by POWER10 processor. 8349 8350 83517.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM 8352------------------------------------- 8353 8354:Architectures: x86 SEV enabled 8355:Type: vm 8356:Parameters: args[0] is the fd of the source vm 8357:Returns: 0 on success; ENOTTY on error 8358 8359This capability enables userspace to copy encryption context from the vm 8360indicated by the fd to the vm this is called on. 8361 8362This is intended to support in-guest workloads scheduled by the host. This 8363allows the in-guest workload to maintain its own NPTs and keeps the two vms 8364from accidentally clobbering each other with interrupts and the like (separate 8365APIC/MSRs/etc). 8366 83677.25 KVM_CAP_SGX_ATTRIBUTE 8368-------------------------- 8369 8370:Architectures: x86 8371:Target: VM 8372:Parameters: args[0] is a file handle of a SGX attribute file in securityfs 8373:Returns: 0 on success, -EINVAL if the file handle is invalid or if a requested 8374 attribute is not supported by KVM. 8375 8376KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM to grant a VM access to one or 8377more privileged enclave attributes. args[0] must hold a file handle to a valid 8378SGX attribute file corresponding to an attribute that is supported/restricted 8379by KVM (currently only PROVISIONKEY). 8380 8381The SGX subsystem restricts access to a subset of enclave attributes to provide 8382additional security for an uncompromised kernel, e.g. use of the PROVISIONKEY 8383is restricted to deter malware from using the PROVISIONKEY to obtain a stable 8384system fingerprint. To prevent userspace from circumventing such restrictions 8385by running an enclave in a VM, KVM prevents access to privileged attributes by 8386default. 8387 8388See Documentation/arch/x86/sgx.rst for more details. 8389 83907.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE 8391-------------------------------------- 8392 8393:Architectures: x86 8394:Parameters: args[0] whether the feature should be enabled or not 8395 8396When this capability is enabled, an emulation failure will result in an exit 8397to userspace with KVM_INTERNAL_ERROR (except when the emulator was invoked 8398to handle a VMware backdoor instruction). Furthermore, KVM will now provide up 8399to 15 instruction bytes for any exit to userspace resulting from an emulation 8400failure. When these exits to userspace occur use the emulation_failure struct 8401instead of the internal struct. They both have the same layout, but the 8402emulation_failure struct matches the content better. It also explicitly 8403defines the 'flags' field which is used to describe the fields in the struct 8404that are valid (ie: if KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES is 8405set in the 'flags' field then both 'insn_size' and 'insn_bytes' have valid data 8406in them.) 8407 84087.28 KVM_CAP_ARM_MTE 8409-------------------- 8410 8411:Architectures: arm64 8412:Parameters: none 8413 8414This capability indicates that KVM (and the hardware) supports exposing the 8415Memory Tagging Extensions (MTE) to the guest. It must also be enabled by the 8416VMM before creating any VCPUs to allow the guest access. Note that MTE is only 8417available to a guest running in AArch64 mode and enabling this capability will 8418cause attempts to create AArch32 VCPUs to fail. 8419 8420When enabled the guest is able to access tags associated with any memory given 8421to the guest. KVM will ensure that the tags are maintained during swap or 8422hibernation of the host; however the VMM needs to manually save/restore the 8423tags as appropriate if the VM is migrated. 8424 8425When this capability is enabled all memory in memslots must be mapped as 8426``MAP_ANONYMOUS`` or with a RAM-based file mapping (``tmpfs``, ``memfd``), 8427attempts to create a memslot with an invalid mmap will result in an 8428-EINVAL return. 8429 8430``guest_memfd``, even though it is an anonymous file, is not supported with MTE. 8431Attempting to create a memslot backed by ``guest_memfd`` when the MTE capability 8432is enabled, or attempting to enable the MTE capability after 8433``guest_memfd``-backed memslots have been created, will result in an -EINVAL 8434return. 8435 8436When enabled the VMM may make use of the ``KVM_ARM_MTE_COPY_TAGS`` ioctl to 8437perform a bulk copy of tags to/from the guest. 8438 84397.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 8440------------------------------------- 8441 8442:Architectures: x86 SEV enabled 8443:Type: vm 8444:Parameters: args[0] is the fd of the source vm 8445:Returns: 0 on success 8446 8447This capability enables userspace to migrate the encryption context from the VM 8448indicated by the fd to the VM this is called on. 8449 8450This is intended to support intra-host migration of VMs between userspace VMMs, 8451upgrading the VMM process without interrupting the guest. 8452 84537.31 KVM_CAP_DISABLE_QUIRKS2 8454---------------------------- 8455 8456:Parameters: args[0] - set of KVM quirks to disable 8457:Architectures: x86 8458:Type: vm 8459 8460This capability, if enabled, will cause KVM to disable some behavior 8461quirks. 8462 8463Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 8464quirks that can be disabled in KVM. 8465 8466The argument to KVM_ENABLE_CAP for this capability is a bitmask of 8467quirks to disable, and must be a subset of the bitmask returned by 8468KVM_CHECK_EXTENSION. 8469 8470The valid bits in cap.args[0] are: 8471 8472======================================== ================================================ 8473KVM_X86_QUIRK_LINT0_REENABLED By default, the reset value for the LVT 8474 LINT0 register is 0x700 (APIC_MODE_EXTINT). 8475 When this quirk is disabled, the reset value 8476 is 0x10000 (APIC_LVT_MASKED). 8477 8478KVM_X86_QUIRK_CD_NW_CLEARED By default, KVM clears CR0.CD and CR0.NW on 8479 AMD CPUs to workaround buggy guest firmware 8480 that runs in perpetuity with CR0.CD, i.e. 8481 with caches in "no fill" mode. 8482 8483 When this quirk is disabled, KVM does not 8484 change the value of CR0.CD and CR0.NW. 8485 8486KVM_X86_QUIRK_LAPIC_MMIO_HOLE By default, the MMIO LAPIC interface is 8487 available even when configured for x2APIC 8488 mode. When this quirk is disabled, KVM 8489 disables the MMIO LAPIC interface if the 8490 LAPIC is in x2APIC mode. 8491 8492KVM_X86_QUIRK_OUT_7E_INC_RIP By default, KVM pre-increments %rip before 8493 exiting to userspace for an OUT instruction 8494 to port 0x7e. When this quirk is disabled, 8495 KVM does not pre-increment %rip before 8496 exiting to userspace. 8497 8498KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this quirk is disabled, KVM sets 8499 CPUID.01H:ECX[bit 3] (MONITOR/MWAIT) if 8500 IA32_MISC_ENABLE[bit 18] (MWAIT) is set. 8501 Additionally, when this quirk is disabled, 8502 KVM clears CPUID.01H:ECX[bit 3] if 8503 IA32_MISC_ENABLE[bit 18] is cleared. 8504 8505KVM_X86_QUIRK_FIX_HYPERCALL_INSN By default, KVM rewrites guest 8506 VMMCALL/VMCALL instructions to match the 8507 vendor's hypercall instruction for the 8508 system. When this quirk is disabled, KVM 8509 will no longer rewrite invalid guest 8510 hypercall instructions. Executing the 8511 incorrect hypercall instruction will 8512 generate a #UD within the guest. 8513 8514KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By default, KVM emulates MONITOR/MWAIT (if 8515 they are intercepted) as NOPs regardless of 8516 whether or not MONITOR/MWAIT are supported 8517 according to guest CPUID. When this quirk 8518 is disabled and KVM_X86_DISABLE_EXITS_MWAIT 8519 is not set (MONITOR/MWAIT are intercepted), 8520 KVM will inject a #UD on MONITOR/MWAIT if 8521 they're unsupported per guest CPUID. Note, 8522 KVM will modify MONITOR/MWAIT support in 8523 guest CPUID on writes to MISC_ENABLE if 8524 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT is 8525 disabled. 8526 8527KVM_X86_QUIRK_SLOT_ZAP_ALL By default, for KVM_X86_DEFAULT_VM VMs, KVM 8528 invalidates all SPTEs in all memslots and 8529 address spaces when a memslot is deleted or 8530 moved. When this quirk is disabled (or the 8531 VM type isn't KVM_X86_DEFAULT_VM), KVM only 8532 ensures the backing memory of the deleted 8533 or moved memslot isn't reachable, i.e KVM 8534 _may_ invalidate only SPTEs related to the 8535 memslot. 8536 8537KVM_X86_QUIRK_STUFF_FEATURE_MSRS By default, at vCPU creation, KVM sets the 8538 vCPU's MSR_IA32_PERF_CAPABILITIES (0x345), 8539 MSR_IA32_ARCH_CAPABILITIES (0x10a), 8540 MSR_PLATFORM_INFO (0xce), and all VMX MSRs 8541 (0x480..0x492) to the maximal capabilities 8542 supported by KVM. KVM also sets 8543 MSR_IA32_UCODE_REV (0x8b) to an arbitrary 8544 value (which is different for Intel vs. 8545 AMD). Lastly, when guest CPUID is set (by 8546 userspace), KVM modifies select VMX MSR 8547 fields to force consistency between guest 8548 CPUID and L2's effective ISA. When this 8549 quirk is disabled, KVM zeroes the vCPU's MSR 8550 values (with two exceptions, see below), 8551 i.e. treats the feature MSRs like CPUID 8552 leaves and gives userspace full control of 8553 the vCPU model definition. This quirk does 8554 not affect VMX MSRs CR0/CR4_FIXED1 (0x487 8555 and 0x489), as KVM does now allow them to 8556 be set by userspace (KVM sets them based on 8557 guest CPUID, for safety purposes). 8558 8559KVM_X86_QUIRK_IGNORE_GUEST_PAT By default, on Intel platforms, KVM ignores 8560 guest PAT and forces the effective memory 8561 type to WB in EPT. The quirk is not available 8562 on Intel platforms which are incapable of 8563 safely honoring guest PAT (i.e., without CPU 8564 self-snoop, KVM always ignores guest PAT and 8565 forces effective memory type to WB). It is 8566 also ignored on AMD platforms or, on Intel, 8567 when a VM has non-coherent DMA devices 8568 assigned; KVM always honors guest PAT in 8569 such case. The quirk is needed to avoid 8570 slowdowns on certain Intel Xeon platforms 8571 (e.g. ICX, SPR) where self-snoop feature is 8572 supported but UC is slow enough to cause 8573 issues with some older guests that use 8574 UC instead of WC to map the video RAM. 8575 Userspace can disable the quirk to honor 8576 guest PAT if it knows that there is no such 8577 guest software, for example if it does not 8578 expose a bochs graphics device (which is 8579 known to have had a buggy driver). 8580 8581KVM_X86_QUIRK_VMCS12_ALLOW_FREEZE_IN_SMM By default, KVM relaxes the consistency 8582 check for GUEST_IA32_DEBUGCTL in vmcs12 8583 to allow FREEZE_IN_SMM to be set. When 8584 this quirk is disabled, KVM requires this 8585 bit to be cleared. Note that the vmcs02 8586 bit is still completely controlled by the 8587 host, regardless of the quirk setting. 8588 8589KVM_X86_QUIRK_NESTED_SVM_SHARED_PAT By default, KVM for nested SVM guests 8590 shares the IA32_PAT MSR between L1 and 8591 L2. This is legacy behavior and does 8592 not match the AMD architecture 8593 specification. When this quirk is 8594 disabled and nested paging (NPT) is 8595 enabled for L2, KVM correctly 8596 virtualizes a separate guest PAT 8597 register for L2, using the g_pat 8598 field in the VMCB. When NPT is 8599 disabled for L2, L1 and L2 continue 8600 to share the IA32_PAT MSR regardless 8601 of the quirk setting. 8602======================================== ================================================ 8603 86047.32 KVM_CAP_MAX_VCPU_ID 8605------------------------ 8606 8607:Architectures: x86 8608:Target: VM 8609:Parameters: args[0] - maximum APIC ID value set for current VM 8610:Returns: 0 on success, -EINVAL if args[0] is beyond KVM_MAX_VCPU_IDS 8611 supported in KVM or if it has been set. 8612 8613This capability allows userspace to specify maximum possible APIC ID 8614assigned for current VM session prior to the creation of vCPUs, saving 8615memory for data structures indexed by the APIC ID. Userspace is able 8616to calculate the limit to APIC ID values from designated 8617CPU topology. 8618 8619The value can be changed only until KVM_ENABLE_CAP is set to a nonzero 8620value or until a vCPU is created. Upon creation of the first vCPU, 8621if the value was set to zero or KVM_ENABLE_CAP was not invoked, KVM 8622uses the return value of KVM_CHECK_EXTENSION(KVM_CAP_MAX_VCPU_ID) as 8623the maximum APIC ID. 8624 86257.33 KVM_CAP_X86_NOTIFY_VMEXIT 8626------------------------------ 8627 8628:Architectures: x86 8629:Target: VM 8630:Parameters: args[0] is the value of notify window as well as some flags 8631:Returns: 0 on success, -EINVAL if args[0] contains invalid flags or notify 8632 VM exit is unsupported. 8633 8634Bits 63:32 of args[0] are used for notify window. 8635Bits 31:0 of args[0] are for some flags. Valid bits are:: 8636 8637 #define KVM_X86_NOTIFY_VMEXIT_ENABLED (1 << 0) 8638 #define KVM_X86_NOTIFY_VMEXIT_USER (1 << 1) 8639 8640This capability allows userspace to configure the notify VM exit on/off 8641in per-VM scope during VM creation. Notify VM exit is disabled by default. 8642When userspace sets KVM_X86_NOTIFY_VMEXIT_ENABLED bit in args[0], VMM will 8643enable this feature with the notify window provided, which will generate 8644a VM exit if no event window occurs in VM non-root mode for a specified of 8645time (notify window). 8646 8647If KVM_X86_NOTIFY_VMEXIT_USER is set in args[0], upon notify VM exits happen, 8648KVM would exit to userspace for handling. 8649 8650This capability is aimed to mitigate the threat that malicious VMs can 8651cause CPU stuck (due to event windows don't open up) and make the CPU 8652unavailable to host or other VMs. 8653 86547.35 KVM_CAP_X86_APIC_BUS_CYCLES_NS 8655----------------------------------- 8656 8657:Architectures: x86 8658:Target: VM 8659:Parameters: args[0] is the desired APIC bus clock rate, in nanoseconds 8660:Returns: 0 on success, -EINVAL if args[0] contains an invalid value for the 8661 frequency or if any vCPUs have been created, -ENXIO if a virtual 8662 local APIC has not been created using KVM_CREATE_IRQCHIP. 8663 8664This capability sets the VM's APIC bus clock frequency, used by KVM's in-kernel 8665virtual APIC when emulating APIC timers. KVM's default value can be retrieved 8666by KVM_CHECK_EXTENSION. 8667 8668Note: Userspace is responsible for correctly configuring CPUID 0x15, a.k.a. the 8669core crystal clock frequency, if a non-zero CPUID 0x15 is exposed to the guest. 8670 86717.36 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8672---------------------------------------------------------- 8673 8674:Architectures: x86, arm64, riscv 8675:Type: vm 8676:Parameters: args[0] - size of the dirty log ring 8677 8678KVM is capable of tracking dirty memory using ring buffers that are 8679mmapped into userspace; there is one dirty ring per vcpu. 8680 8681The dirty ring is available to userspace as an array of 8682``struct kvm_dirty_gfn``. Each dirty entry is defined as:: 8683 8684 struct kvm_dirty_gfn { 8685 __u32 flags; 8686 __u32 slot; /* as_id | slot_id */ 8687 __u64 offset; 8688 }; 8689 8690The following values are defined for the flags field to define the 8691current state of the entry:: 8692 8693 #define KVM_DIRTY_GFN_F_DIRTY BIT(0) 8694 #define KVM_DIRTY_GFN_F_RESET BIT(1) 8695 #define KVM_DIRTY_GFN_F_MASK 0x3 8696 8697Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM 8698ioctl to enable this capability for the new guest and set the size of 8699the rings. Enabling the capability is only allowed before creating any 8700vCPU, and the size of the ring must be a power of two. The larger the 8701ring buffer, the less likely the ring is full and the VM is forced to 8702exit to userspace. The optimal size depends on the workload, but it is 8703recommended that it be at least 64 KiB (4096 entries). 8704 8705Just like for dirty page bitmaps, the buffer tracks writes to 8706all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was 8707set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered 8708with the flag set, userspace can start harvesting dirty pages from the 8709ring buffer. 8710 8711An entry in the ring buffer can be unused (flag bits ``00``), 8712dirty (flag bits ``01``) or harvested (flag bits ``1X``). The 8713state machine for the entry is as follows:: 8714 8715 dirtied harvested reset 8716 00 -----------> 01 -------------> 1X -------+ 8717 ^ | 8718 | | 8719 +------------------------------------------+ 8720 8721To harvest the dirty pages, userspace accesses the mmapped ring buffer 8722to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage 8723the RESET bit must be cleared), then it means this GFN is a dirty GFN. 8724The userspace should harvest this GFN and mark the flags from state 8725``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set 8726to show that this GFN is harvested and waiting for a reset), and move 8727on to the next GFN. The userspace should continue to do this until the 8728flags of a GFN have the DIRTY bit cleared, meaning that it has harvested 8729all the dirty GFNs that were available. 8730 8731Note that on weakly ordered architectures, userspace accesses to the 8732ring buffer (and more specifically the 'flags' field) must be ordered, 8733using load-acquire/store-release accessors when available, or any 8734other memory barrier that will ensure this ordering. 8735 8736It's not necessary for userspace to harvest the all dirty GFNs at once. 8737However it must collect the dirty GFNs in sequence, i.e., the userspace 8738program cannot skip one dirty GFN to collect the one next to it. 8739 8740After processing one or more entries in the ring buffer, userspace 8741calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about 8742it, so that the kernel will reprotect those collected GFNs. 8743Therefore, the ioctl must be called *before* reading the content of 8744the dirty pages. 8745 8746The dirty ring can get full. When it happens, the KVM_RUN of the 8747vcpu will return with exit reason KVM_EXIT_DIRTY_RING_FULL. 8748 8749The dirty ring interface has a major difference comparing to the 8750KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from 8751userspace, it's still possible that the kernel has not yet flushed the 8752processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the 8753flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one 8754needs to kick the vcpu out of KVM_RUN using a signal. The resulting 8755vmexit ensures that all dirty GFNs are flushed to the dirty rings. 8756 8757NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the only capability that 8758should be exposed by weakly ordered architecture, in order to indicate 8759the additional memory ordering requirements imposed on userspace when 8760reading the state of an entry and mutating it from DIRTY to HARVESTED. 8761Architecture with TSO-like ordering (such as x86) are allowed to 8762expose both KVM_CAP_DIRTY_LOG_RING and KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8763to userspace. 8764 8765After enabling the dirty rings, the userspace needs to detect the 8766capability of KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP to see whether the 8767ring structures can be backed by per-slot bitmaps. With this capability 8768advertised, it means the architecture can dirty guest pages without 8769vcpu/ring context, so that some of the dirty information will still be 8770maintained in the bitmap structure. KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP 8771can't be enabled if the capability of KVM_CAP_DIRTY_LOG_RING_ACQ_REL 8772hasn't been enabled, or any memslot has been existing. 8773 8774Note that the bitmap here is only a backup of the ring structure. The 8775use of the ring and bitmap combination is only beneficial if there is 8776only a very small amount of memory that is dirtied out of vcpu/ring 8777context. Otherwise, the stand-alone per-slot bitmap mechanism needs to 8778be considered. 8779 8780To collect dirty bits in the backup bitmap, userspace can use the same 8781KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG isn't needed as long as all 8782the generation of the dirty bits is done in a single pass. Collecting 8783the dirty bitmap should be the very last thing that the VMM does before 8784considering the state as complete. VMM needs to ensure that the dirty 8785state is final and avoid missing dirty pages from another ioctl ordered 8786after the bitmap collection. 8787 8788NOTE: Multiple examples of using the backup bitmap: (1) save vgic/its 8789tables through command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} on 8790KVM device "kvm-arm-vgic-its". (2) restore vgic/its tables through 8791command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} on KVM device 8792"kvm-arm-vgic-its". VGICv3 LPI pending status is restored. (3) save 8793vgic3 pending table through KVM_DEV_ARM_VGIC_{GRP_CTRL, SAVE_PENDING_TABLES} 8794command on KVM device "kvm-arm-vgic-v3". 8795 87967.37 KVM_CAP_PMU_CAPABILITY 8797--------------------------- 8798 8799:Architectures: x86 8800:Type: vm 8801:Parameters: arg[0] is bitmask of PMU virtualization capabilities. 8802:Returns: 0 on success, -EINVAL when arg[0] contains invalid bits 8803 8804This capability alters PMU virtualization in KVM. 8805 8806Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 8807PMU virtualization capabilities that can be adjusted on a VM. 8808 8809The argument to KVM_ENABLE_CAP is also a bitmask and selects specific 8810PMU virtualization capabilities to be applied to the VM. This can 8811only be invoked on a VM prior to the creation of VCPUs. 8812 8813At this time, KVM_PMU_CAP_DISABLE is the only capability. Setting 8814this capability will disable PMU virtualization for that VM. Usermode 8815should adjust CPUID leaf 0xA to reflect that the PMU is disabled. 8816 88177.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES 8818------------------------------------- 8819 8820:Architectures: x86 8821:Type: vm 8822:Parameters: arg[0] must be 0. 8823:Returns: 0 on success, -EPERM if the userspace process does not 8824 have CAP_SYS_BOOT, -EINVAL if args[0] is not 0 or any vCPUs have been 8825 created. 8826 8827This capability disables the NX huge pages mitigation for iTLB MULTIHIT. 8828 8829The capability has no effect if the nx_huge_pages module parameter is not set. 8830 8831This capability may only be set before any vCPUs are created. 8832 88337.39 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE 8834--------------------------------------- 8835 8836:Architectures: arm64 8837:Type: vm 8838:Parameters: arg[0] is the new split chunk size. 8839:Returns: 0 on success, -EINVAL if any memslot was already created. 8840 8841This capability sets the chunk size used in Eager Page Splitting. 8842 8843Eager Page Splitting improves the performance of dirty-logging (used 8844in live migrations) when guest memory is backed by huge-pages. It 8845avoids splitting huge-pages (into PAGE_SIZE pages) on fault, by doing 8846it eagerly when enabling dirty logging (with the 8847KVM_MEM_LOG_DIRTY_PAGES flag for a memory region), or when using 8848KVM_CLEAR_DIRTY_LOG. 8849 8850The chunk size specifies how many pages to break at a time, using a 8851single allocation for each chunk. Bigger the chunk size, more pages 8852need to be allocated ahead of time. 8853 8854The chunk size needs to be a valid block size. The list of acceptable 8855block sizes is exposed in KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES as a 885664-bit bitmap (each bit describing a block size). The default value is 88570, to disable the eager page splitting. 8858 88597.40 KVM_CAP_EXIT_HYPERCALL 8860--------------------------- 8861 8862:Architectures: x86 8863:Type: vm 8864 8865This capability, if enabled, will cause KVM to exit to userspace 8866with KVM_EXIT_HYPERCALL exit reason to process some hypercalls. 8867Userspace may fail the hypercall by setting hypercall.ret to EINVAL 8868or may request the hypercall to be retried the next time the guest run 8869by setting hypercall.ret to EAGAIN. 8870 8871Calling KVM_CHECK_EXTENSION for this capability will return a bitmask 8872of hypercalls that can be configured to exit to userspace. 8873Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE. 8874 8875The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset 8876of the result of KVM_CHECK_EXTENSION. KVM will forward to userspace 8877the hypercalls whose corresponding bit is in the argument, and return 8878ENOSYS for the others. 8879 88807.41 KVM_CAP_ARM_SYSTEM_SUSPEND 8881------------------------------- 8882 8883:Architectures: arm64 8884:Type: vm 8885 8886When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of 8887type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request. 8888 88897.42 KVM_CAP_ARM_WRITABLE_IMP_ID_REGS 8890------------------------------------- 8891 8892:Architectures: arm64 8893:Target: VM 8894:Parameters: None 8895:Returns: 0 on success, -EINVAL if vCPUs have been created before enabling this 8896 capability. 8897 8898This capability changes the behavior of the registers that identify a PE 8899implementation of the Arm architecture: MIDR_EL1, REVIDR_EL1, and AIDR_EL1. 8900By default, these registers are visible to userspace but treated as invariant. 8901 8902When this capability is enabled, KVM allows userspace to change the 8903aforementioned registers before the first KVM_RUN. These registers are VM 8904scoped, meaning that the same set of values are presented on all vCPUs in a 8905given VM. 8906 89077.43 KVM_CAP_RISCV_MP_STATE_RESET 8908--------------------------------- 8909 8910:Architectures: riscv 8911:Type: VM 8912:Parameters: None 8913:Returns: 0 on success, -EINVAL if arg[0] is not zero 8914 8915When this capability is enabled, KVM resets the VCPU when setting 8916MP_STATE_INIT_RECEIVED through IOCTL. The original MP_STATE is preserved. 8917 89187.44 KVM_CAP_ARM_CACHEABLE_PFNMAP_SUPPORTED 8919------------------------------------------- 8920 8921:Architectures: arm64 8922:Target: VM 8923:Parameters: None 8924 8925This capability indicate to the userspace whether a PFNMAP memory region 8926can be safely mapped as cacheable. This relies on the presence of 8927force write back (FWB) feature support on the hardware. 8928 89297.45 KVM_CAP_ARM_SEA_TO_USER 8930---------------------------- 8931 8932:Architecture: arm64 8933:Target: VM 8934:Parameters: none 8935:Returns: 0 on success, -EINVAL if unsupported. 8936 8937When this capability is enabled, KVM may exit to userspace for SEAs taken to 8938EL2 resulting from a guest access. See ``KVM_EXIT_ARM_SEA`` for more 8939information. 8940 89417.46 KVM_CAP_S390_USER_OPEREXEC 8942------------------------------- 8943 8944:Architectures: s390 8945:Parameters: none 8946 8947When this capability is enabled KVM forwards all operation exceptions 8948that it doesn't handle itself to user space. This also includes the 89490x0000 instructions managed by KVM_CAP_S390_USER_INSTR0. This is 8950helpful if user space wants to emulate instructions which are not 8951(yet) implemented in hardware. 8952 8953This capability can be enabled dynamically even if VCPUs were already 8954created and are running. 8955 89567.47 KVM_CAP_S390_HPAGE_2G 8957-------------------------- 8958 8959:Architectures: s390 8960:Parameters: none 8961:Returns: 0 on success; -EINVAL if hpage_2g module parameter was not set, 8962 cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 8963 flag set; -EBUSY if vCPUs were already created for the VM. 8964 8965With this capability the KVM support for memory backing with 2g pages 8966through hugetlbfs can be enabled for a VM. After the capability is 8967enabled, cmma can't be enabled anymore and pfmfi and the storage key 8968interpretation are disabled. If cmma has already been enabled or the 8969hpage_2g module parameter is not set to 1, -EINVAL is returned. 8970 89718. Other capabilities. 8972====================== 8973 8974This section lists capabilities that give information about other 8975features of the KVM implementation. 8976 89778.1 KVM_CAP_PPC_HWRNG 8978--------------------- 8979 8980:Architectures: ppc 8981 8982This capability, if KVM_CHECK_EXTENSION indicates that it is 8983available, means that the kernel has an implementation of the 8984H_RANDOM hypercall backed by a hardware random-number generator. 8985If present, the kernel H_RANDOM handler can be enabled for guest use 8986with the KVM_CAP_PPC_ENABLE_HCALL capability. 8987 89888.3 KVM_CAP_PPC_MMU_RADIX 8989------------------------- 8990 8991:Architectures: ppc 8992 8993This capability, if KVM_CHECK_EXTENSION indicates that it is 8994available, means that the kernel can support guests using the 8995radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 8996processor). 8997 89988.4 KVM_CAP_PPC_MMU_HASH_V3 8999--------------------------- 9000 9001:Architectures: ppc 9002 9003This capability, if KVM_CHECK_EXTENSION indicates that it is 9004available, means that the kernel can support guests using the 9005hashed page table MMU defined in Power ISA V3.00 (as implemented in 9006the POWER9 processor), including in-memory segment tables. 9007 90088.5 KVM_CAP_MIPS_VZ 9009------------------- 9010 9011:Architectures: mips 9012 9013This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 9014it is available, means that full hardware assisted virtualization capabilities 9015of the hardware are available for use through KVM. An appropriate 9016KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 9017utilises it. 9018 9019If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 9020available, it means that the VM is using full hardware assisted virtualization 9021capabilities of the hardware. This is useful to check after creating a VM with 9022KVM_VM_MIPS_DEFAULT. 9023 9024The value returned by KVM_CHECK_EXTENSION should be compared against known 9025values (see below). All other values are reserved. This is to allow for the 9026possibility of other hardware assisted virtualization implementations which 9027may be incompatible with the MIPS VZ ASE. 9028 9029== ========================================================================== 9030 0 The trap & emulate implementation is in use to run guest code in user 9031 mode. Guest virtual memory segments are rearranged to fit the guest in the 9032 user mode address space. 9033 9034 1 The MIPS VZ ASE is in use, providing full hardware assisted 9035 virtualization, including standard guest virtual memory segments. 9036== ========================================================================== 9037 90388.7 KVM_CAP_MIPS_64BIT 9039---------------------- 9040 9041:Architectures: mips 9042 9043This capability indicates the supported architecture type of the guest, i.e. the 9044supported register and address width. 9045 9046The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 9047kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 9048be checked specifically against known values (see below). All other values are 9049reserved. 9050 9051== ======================================================================== 9052 0 MIPS32 or microMIPS32. 9053 Both registers and addresses are 32-bits wide. 9054 It will only be possible to run 32-bit guest code. 9055 9056 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 9057 Registers are 64-bits wide, but addresses are 32-bits wide. 9058 64-bit guest code may run but cannot access MIPS64 memory segments. 9059 It will also be possible to run 32-bit guest code. 9060 9061 2 MIPS64 or microMIPS64 with access to all address segments. 9062 Both registers and addresses are 64-bits wide. 9063 It will be possible to run 64-bit or 32-bit guest code. 9064== ======================================================================== 9065 90668.9 KVM_CAP_ARM_USER_IRQ 9067------------------------ 9068 9069:Architectures: arm64 9070 9071This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 9072that if userspace creates a VM without an in-kernel interrupt controller, it 9073will be notified of changes to the output level of in-kernel emulated devices, 9074which can generate virtual interrupts, presented to the VM. 9075For such VMs, on every return to userspace, the kernel 9076updates the vcpu's run->s.regs.device_irq_level field to represent the actual 9077output level of the device. 9078 9079Whenever kvm detects a change in the device output level, kvm guarantees at 9080least one return to userspace before running the VM. This exit could either 9081be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 9082userspace can always sample the device output level and re-compute the state of 9083the userspace interrupt controller. Userspace should always check the state 9084of run->s.regs.device_irq_level on every kvm exit. 9085The value in run->s.regs.device_irq_level can represent both level and edge 9086triggered interrupt signals, depending on the device. Edge triggered interrupt 9087signals will exit to userspace with the bit in run->s.regs.device_irq_level 9088set exactly once per edge signal. 9089 9090The field run->s.regs.device_irq_level is available independent of 9091run->kvm_valid_regs or run->kvm_dirty_regs bits. 9092 9093If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 9094number larger than 0 indicating the version of this capability is implemented 9095and thereby which bits in run->s.regs.device_irq_level can signal values. 9096 9097Currently the following bits are defined for the device_irq_level bitmap:: 9098 9099 KVM_CAP_ARM_USER_IRQ >= 1: 9100 9101 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 9102 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 9103 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 9104 9105Future versions of kvm may implement additional events. These will get 9106indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 9107listed above. 9108 91098.10 KVM_CAP_PPC_SMT_POSSIBLE 9110----------------------------- 9111 9112:Architectures: ppc 9113 9114Querying this capability returns a bitmap indicating the possible 9115virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 9116(counting from the right) is set, then a virtual SMT mode of 2^N is 9117available. 9118 91198.12 KVM_CAP_HYPERV_VP_INDEX 9120---------------------------- 9121 9122:Architectures: x86 9123 9124This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 9125value is used to denote the target vcpu for a SynIC interrupt. For 9126compatibility, KVM initializes this msr to KVM's internal vcpu index. When this 9127capability is absent, userspace can still query this msr's value. 9128 91298.13 KVM_CAP_S390_AIS_MIGRATION 9130------------------------------- 9131 9132:Architectures: s390 9133 9134This capability indicates if the flic device will be able to get/set the 9135AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 9136to discover this without having to create a flic device. 9137 91388.14 KVM_CAP_S390_PSW 9139--------------------- 9140 9141:Architectures: s390 9142 9143This capability indicates that the PSW is exposed via the kvm_run structure. 9144 91458.15 KVM_CAP_S390_GMAP 9146---------------------- 9147 9148:Architectures: s390 9149 9150This capability indicates that the user space memory used as guest mapping can 9151be anywhere in the user memory address space, as long as the memory slots are 9152aligned and sized to a segment (1MB) boundary. 9153 91548.16 KVM_CAP_S390_COW 9155--------------------- 9156 9157:Architectures: s390 9158 9159This capability indicates that the user space memory used as guest mapping can 9160use copy-on-write semantics as well as dirty pages tracking via read-only page 9161tables. 9162 91638.17 KVM_CAP_S390_BPB 9164--------------------- 9165 9166:Architectures: s390 9167 9168This capability indicates that kvm will implement the interfaces to handle 9169reset, migration and nested KVM for branch prediction blocking. The stfle 9170facility 82 should not be provided to the guest without this capability. 9171 91728.18 KVM_CAP_HYPERV_TLBFLUSH 9173---------------------------- 9174 9175:Architectures: x86 9176 9177This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush 9178hypercalls: 9179HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx, 9180HvFlushVirtualAddressList, HvFlushVirtualAddressListEx. 9181 91828.19 KVM_CAP_ARM_INJECT_SERROR_ESR 9183---------------------------------- 9184 9185:Architectures: arm64 9186 9187This capability indicates that userspace can specify (via the 9188KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it 9189takes a virtual SError interrupt exception. 9190If KVM advertises this capability, userspace can only specify the ISS field for 9191the ESR syndrome. Other parts of the ESR, such as the EC are generated by the 9192CPU when the exception is taken. If this virtual SError is taken to EL1 using 9193AArch64, this value will be reported in the ISS field of ESR_ELx. 9194 9195See KVM_CAP_VCPU_EVENTS for more details. 9196 91978.20 KVM_CAP_HYPERV_SEND_IPI 9198---------------------------- 9199 9200:Architectures: x86 9201 9202This capability indicates that KVM supports paravirtualized Hyper-V IPI send 9203hypercalls: 9204HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx. 9205 92068.22 KVM_CAP_S390_VCPU_RESETS 9207----------------------------- 9208 9209:Architectures: s390 9210 9211This capability indicates that the KVM_S390_NORMAL_RESET and 9212KVM_S390_CLEAR_RESET ioctls are available. 9213 92148.23 KVM_CAP_S390_PROTECTED 9215--------------------------- 9216 9217:Architectures: s390 9218 9219This capability indicates that the Ultravisor has been initialized and 9220KVM can therefore start protected VMs. 9221This capability governs the KVM_S390_PV_COMMAND ioctl and the 9222KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected 9223guests when the state change is invalid. 9224 92258.24 KVM_CAP_STEAL_TIME 9226----------------------- 9227 9228:Architectures: arm64, x86 9229 9230This capability indicates that KVM supports steal time accounting. 9231When steal time accounting is supported it may be enabled with 9232architecture-specific interfaces. This capability and the architecture- 9233specific interfaces must be consistent, i.e. if one says the feature 9234is supported, than the other should as well and vice versa. For arm64 9235see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL". 9236For x86 see Documentation/virt/kvm/x86/msr.rst "MSR_KVM_STEAL_TIME". 9237 92388.25 KVM_CAP_S390_DIAG318 9239------------------------- 9240 9241:Architectures: s390 9242 9243This capability enables a guest to set information about its control program 9244(i.e. guest kernel type and version). The information is helpful during 9245system/firmware service events, providing additional data about the guest 9246environments running on the machine. 9247 9248The information is associated with the DIAGNOSE 0x318 instruction, which sets 9249an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and 9250a 7-byte Control Program Version Code (CPVC). The CPNC determines what 9251environment the control program is running in (e.g. Linux, z/VM...), and the 9252CPVC is used for information specific to OS (e.g. Linux version, Linux 9253distribution...) 9254 9255If this capability is available, then the CPNC and CPVC can be synchronized 9256between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318). 9257 92588.26 KVM_CAP_X86_USER_SPACE_MSR 9259------------------------------- 9260 9261:Architectures: x86 9262 9263This capability indicates that KVM supports deflection of MSR reads and 9264writes to user space. It can be enabled on a VM level. If enabled, MSR 9265accesses that would usually trigger a #GP by KVM into the guest will 9266instead get bounced to user space through the KVM_EXIT_X86_RDMSR and 9267KVM_EXIT_X86_WRMSR exit notifications. 9268 92698.27 KVM_CAP_X86_MSR_FILTER 9270--------------------------- 9271 9272:Architectures: x86 9273 9274This capability indicates that KVM supports that accesses to user defined MSRs 9275may be rejected. With this capability exposed, KVM exports new VM ioctl 9276KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR 9277ranges that KVM should deny access to. 9278 9279In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to 9280trap and emulate MSRs that are outside of the scope of KVM as well as 9281limit the attack surface on KVM's MSR emulation code. 9282 92838.30 KVM_CAP_XEN_HVM 9284-------------------- 9285 9286:Architectures: x86 9287 9288This capability indicates the features that Xen supports for hosting Xen 9289PVHVM guests. Valid flags are:: 9290 9291 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0) 9292 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1) 9293 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2) 9294 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 3) 9295 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL (1 << 4) 9296 #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND (1 << 5) 9297 #define KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG (1 << 6) 9298 #define KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE (1 << 7) 9299 9300The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG 9301ioctl is available, for the guest to set its hypercall page. 9302 9303If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be 9304provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page 9305contents, to request that KVM generate hypercall page content automatically 9306and also enable interception of guest hypercalls with KVM_EXIT_XEN. 9307 9308The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the 9309KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and 9310KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors 9311for event channel upcalls when the evtchn_upcall_pending field of a vcpu's 9312vcpu_info is set. 9313 9314The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related 9315features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are 9316supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls. 9317 9318The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries 9319of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority 9320field set to indicate 2 level event channel delivery. 9321 9322The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates that KVM supports 9323injecting event channel events directly into the guest with the 9324KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indicates support for the 9325KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attributes and the 9326KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_VECTOR vCPU attributes. 9327related to event channel delivery, timers, and the XENVER_version 9328interception. 9329 9330The KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG flag indicates that KVM supports 9331the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute in the KVM_XEN_SET_ATTR 9332and KVM_XEN_GET_ATTR ioctls. This controls whether KVM will set the 9333XEN_RUNSTATE_UPDATE flag in guest memory mapped vcpu_runstate_info during 9334updates of the runstate information. Note that versions of KVM which support 9335the RUNSTATE feature above, but not the RUNSTATE_UPDATE_FLAG feature, will 9336always set the XEN_RUNSTATE_UPDATE flag when updating the guest structure, 9337which is perhaps counterintuitive. When this flag is advertised, KVM will 9338behave more correctly, not using the XEN_RUNSTATE_UPDATE flag until/unless 9339specifically enabled (by the guest making the hypercall, causing the VMM 9340to enable the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute). 9341 9342The KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag indicates that KVM supports 9343clearing the PVCLOCK_TSC_STABLE_BIT flag in Xen pvclock sources. This will be 9344done when the KVM_CAP_XEN_HVM ioctl sets the 9345KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag. 9346 93478.31 KVM_CAP_SPAPR_MULTITCE 9348--------------------------- 9349 9350:Architectures: ppc 9351:Type: vm 9352 9353This capability means the kernel is capable of handling hypercalls 9354H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 9355space. This significantly accelerates DMA operations for PPC KVM guests. 9356User space should expect that its handlers for these hypercalls 9357are not going to be called if user space previously registered LIOBN 9358in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 9359 9360In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 9361user space might have to advertise it for the guest. For example, 9362IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 9363present in the "ibm,hypertas-functions" device-tree property. 9364 9365The hypercalls mentioned above may or may not be processed successfully 9366in the kernel based fast path. If they can not be handled by the kernel, 9367they will get passed on to user space. So user space still has to have 9368an implementation for these despite the in kernel acceleration. 9369 9370This capability is always enabled. 9371 93728.32 KVM_CAP_PTP_KVM 9373-------------------- 9374 9375:Architectures: arm64 9376 9377This capability indicates that the KVM virtual PTP service is 9378supported in the host. A VMM can check whether the service is 9379available to the guest on migration. 9380 93818.37 KVM_CAP_S390_PROTECTED_DUMP 9382-------------------------------- 9383 9384:Architectures: s390 9385:Type: vm 9386 9387This capability indicates that KVM and the Ultravisor support dumping 9388PV guests. The `KVM_PV_DUMP` command is available for the 9389`KVM_S390_PV_COMMAND` ioctl and the `KVM_PV_INFO` command provides 9390dump related UV data. Also the vcpu ioctl `KVM_S390_PV_CPU_COMMAND` is 9391available and supports the `KVM_PV_DUMP_CPU` subcommand. 9392 93938.39 KVM_CAP_S390_CPU_TOPOLOGY 9394------------------------------ 9395 9396:Architectures: s390 9397:Type: vm 9398 9399This capability indicates that KVM will provide the S390 CPU Topology 9400facility which consist of the interpretation of the PTF instruction for 9401the function code 2 along with interception and forwarding of both the 9402PTF instruction with function codes 0 or 1 and the STSI(15,1,x) 9403instruction to the userland hypervisor. 9404 9405The stfle facility 11, CPU Topology facility, should not be indicated 9406to the guest without this capability. 9407 9408When this capability is present, KVM provides a new attribute group 9409on vm fd, KVM_S390_VM_CPU_TOPOLOGY. 9410This new attribute allows to get, set or clear the Modified Change 9411Topology Report (MTCR) bit of the SCA through the kvm_device_attr 9412structure. 9413 9414When getting the Modified Change Topology Report value, the attr->addr 9415must point to a byte where the value will be stored or retrieved from. 9416 94178.41 KVM_CAP_VM_TYPES 9418--------------------- 9419 9420:Architectures: x86 9421:Type: system ioctl 9422 9423This capability returns a bitmap of support VM types. The 1-setting of bit @n 9424means the VM type with value @n is supported. Possible values of @n are:: 9425 9426 #define KVM_X86_DEFAULT_VM 0 9427 #define KVM_X86_SW_PROTECTED_VM 1 9428 #define KVM_X86_SEV_VM 2 9429 #define KVM_X86_SEV_ES_VM 3 9430 #define KVM_X86_SNP_VM 4 9431 #define KVM_X86_TDX_VM 5 9432 9433Note, KVM_X86_SW_PROTECTED_VM is currently only for development and testing. 9434Do not use KVM_X86_SW_PROTECTED_VM for "real" VMs, and especially not in 9435production. The behavior and effective ABI for software-protected VMs is 9436unstable. 9437 94388.42 KVM_CAP_PPC_RPT_INVALIDATE 9439------------------------------- 9440 9441:Architectures: ppc 9442 9443This capability indicates that the kernel is capable of handling 9444H_RPT_INVALIDATE hcall. 9445 9446In order to enable the use of H_RPT_INVALIDATE in the guest, 9447user space might have to advertise it for the guest. For example, 9448IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is 9449present in the "ibm,hypertas-functions" device-tree property. 9450 9451This capability is enabled for hypervisors on platforms like POWER9 9452that support radix MMU. 9453 94548.43 KVM_CAP_PPC_AIL_MODE_3 9455--------------------------- 9456 9457:Architectures: ppc 9458 9459This capability indicates that the kernel supports the mode 3 setting for the 9460"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location" 9461resource that is controlled with the H_SET_MODE hypercall. 9462 9463This capability allows a guest kernel to use a better-performance mode for 9464handling interrupts and system calls. 9465 94668.44 KVM_CAP_MEMORY_FAULT_INFO 9467------------------------------ 9468 9469:Architectures: x86 9470 9471The presence of this capability indicates that KVM_RUN will fill 9472kvm_run.memory_fault if KVM cannot resolve a guest page fault VM-Exit, e.g. if 9473there is a valid memslot but no backing VMA for the corresponding host virtual 9474address. 9475 9476The information in kvm_run.memory_fault is valid if and only if KVM_RUN returns 9477an error with errno=EFAULT or errno=EHWPOISON *and* kvm_run.exit_reason is set 9478to KVM_EXIT_MEMORY_FAULT. 9479 9480Note: Userspaces which attempt to resolve memory faults so that they can retry 9481KVM_RUN are encouraged to guard against repeatedly receiving the same 9482error/annotated fault. 9483 9484See KVM_EXIT_MEMORY_FAULT for more information. 9485 94868.45 KVM_CAP_X86_GUEST_MODE 9487--------------------------- 9488 9489:Architectures: x86 9490 9491The presence of this capability indicates that KVM_RUN will update the 9492KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags to indicate whether the 9493vCPU was executing nested guest code when it exited. 9494 94958.46 KVM_CAP_S390_KEYOP 9496----------------------- 9497 9498:Architectures: s390 9499 9500The presence of this capability indicates that the KVM_S390_KEYOP ioctl is 9501available. 9502 9503KVM exits with the register state of either the L1 or L2 guest 9504depending on which executed at the time of an exit. Userspace must 9505take care to differentiate between these cases. 9506 95078.47 KVM_CAP_S390_VSIE_ESAMODE 9508------------------------------ 9509 9510:Architectures: s390 9511 9512The presence of this capability indicates that the nested KVM guest can 9513start in ESA mode. 9514 95159. Known KVM API problems 9516========================= 9517 9518In some cases, KVM's API has some inconsistencies or common pitfalls 9519that userspace need to be aware of. This section details some of 9520these issues. 9521 9522Most of them are architecture specific, so the section is split by 9523architecture. 9524 95259.1. x86 9526-------- 9527 9528``KVM_GET_SUPPORTED_CPUID`` issues 9529^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9530 9531In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible 9532to take its result and pass it directly to ``KVM_SET_CPUID2``. This section 9533documents some cases in which that requires some care. 9534 9535Local APIC features 9536~~~~~~~~~~~~~~~~~~~ 9537 9538CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``, 9539but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or 9540``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of 9541the local APIC. 9542 9543The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature. 9544 9545On older versions of Linux, CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by 9546``KVM_GET_SUPPORTED_CPUID``, but it can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` 9547is present and the kernel has enabled in-kernel emulation of the local APIC. 9548On newer versions, ``KVM_GET_SUPPORTED_CPUID`` does report the bit as available. 9549 9550CPU topology 9551~~~~~~~~~~~~ 9552 9553Several CPUID values include topology information for the host CPU: 95540x0b and 0x1f for Intel systems, 0x8000001e for AMD systems. Different 9555versions of KVM return different values for this information and userspace 9556should not rely on it. Currently they return all zeroes. 9557 9558If userspace wishes to set up a guest topology, it should be careful that 9559the values of these three leaves differ for each CPU. In particular, 9560the APIC ID is found in EDX for all subleaves of 0x0b and 0x1f, and in EAX 9561for 0x8000001e; the latter also encodes the core id and node id in bits 95627:0 of EBX and ECX respectively. 9563 9564Obsolete ioctls and capabilities 9565^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9566 9567KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually 9568available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if 9569available. 9570 9571Ordering of KVM_GET_*/KVM_SET_* ioctls 9572^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 9573 9574TBD 9575