1.. SPDX-License-Identifier: GPL-2.0 2 3=================================================================== 4The Definitive KVM (Kernel-based Virtual Machine) API Documentation 5=================================================================== 6 71. General description 8====================== 9 10The kvm API is a set of ioctls that are issued to control various aspects 11of a virtual machine. The ioctls belong to the following classes: 12 13 - System ioctls: These query and set global attributes which affect the 14 whole kvm subsystem. In addition a system ioctl is used to create 15 virtual machines. 16 17 - VM ioctls: These query and set attributes that affect an entire virtual 18 machine, for example memory layout. In addition a VM ioctl is used to 19 create virtual cpus (vcpus) and devices. 20 21 VM ioctls must be issued from the same process (address space) that was 22 used to create the VM. 23 24 - vcpu ioctls: These query and set attributes that control the operation 25 of a single virtual cpu. 26 27 vcpu ioctls should be issued from the same thread that was used to create 28 the vcpu, except for asynchronous vcpu ioctl that are marked as such in 29 the documentation. Otherwise, the first ioctl after switching threads 30 could see a performance impact. 31 32 - device ioctls: These query and set attributes that control the operation 33 of a single device. 34 35 device ioctls must be issued from the same process (address space) that 36 was used to create the VM. 37 382. File descriptors 39=================== 40 41The kvm API is centered around file descriptors. An initial 42open("/dev/kvm") obtains a handle to the kvm subsystem; this handle 43can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this 44handle will create a VM file descriptor which can be used to issue VM 45ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVICE ioctl on a VM fd will 46create a virtual cpu or device and return a file descriptor pointing to 47the new resource. Finally, ioctls on a vcpu or device fd can be used 48to control the vcpu or device. For vcpus, this includes the important 49task of actually running guest code. 50 51In general file descriptors can be migrated among processes by means 52of fork() and the SCM_RIGHTS facility of unix domain socket. These 53kinds of tricks are explicitly not supported by kvm. While they will 54not cause harm to the host, their actual behavior is not guaranteed by 55the API. See "General description" for details on the ioctl usage 56model that is supported by KVM. 57 58It is important to note that although VM ioctls may only be issued from 59the process that created the VM, a VM's lifecycle is associated with its 60file descriptor, not its creator (process). In other words, the VM and 61its resources, *including the associated address space*, are not freed 62until the last reference to the VM's file descriptor has been released. 63For example, if fork() is issued after ioctl(KVM_CREATE_VM), the VM will 64not be freed until both the parent (original) process and its child have 65put their references to the VM's file descriptor. 66 67Because a VM's resources are not freed until the last reference to its 68file descriptor is released, creating additional references to a VM 69via fork(), dup(), etc... without careful consideration is strongly 70discouraged and may have unwanted side effects, e.g. memory allocated 71by and on behalf of the VM's process may not be freed/unaccounted when 72the VM is shut down. 73 74 753. Extensions 76============= 77 78As of Linux 2.6.22, the KVM ABI has been stabilized: no backward 79incompatible change are allowed. However, there is an extension 80facility that allows backward-compatible extensions to the API to be 81queried and used. 82 83The extension mechanism is not based on the Linux version number. 84Instead, kvm defines extension identifiers and a facility to query 85whether a particular extension identifier is available. If it is, a 86set of ioctls is available for application use. 87 88 894. API description 90================== 91 92This section describes ioctls that can be used to control kvm guests. 93For each ioctl, the following information is provided along with a 94description: 95 96 Capability: 97 which KVM extension provides this ioctl. Can be 'basic', 98 which means that is will be provided by any kernel that supports 99 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which 100 means availability needs to be checked with KVM_CHECK_EXTENSION 101 (see section 4.4), or 'none' which means that while not all kernels 102 support this ioctl, there's no capability bit to check its 103 availability: for kernels that don't support the ioctl, 104 the ioctl returns -ENOTTY. 105 106 Architectures: 107 which instruction set architectures provide this ioctl. 108 x86 includes both i386 and x86_64. 109 110 Type: 111 system, vm, or vcpu. 112 113 Parameters: 114 what parameters are accepted by the ioctl. 115 116 Returns: 117 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 118 are not detailed, but errors with specific meanings are. 119 120 1214.1 KVM_GET_API_VERSION 122----------------------- 123 124:Capability: basic 125:Architectures: all 126:Type: system ioctl 127:Parameters: none 128:Returns: the constant KVM_API_VERSION (=12) 129 130This identifies the API version as the stable kvm API. It is not 131expected that this number will change. However, Linux 2.6.20 and 1322.6.21 report earlier versions; these are not documented and not 133supported. Applications should refuse to run if KVM_GET_API_VERSION 134returns a value other than 12. If this check passes, all ioctls 135described as 'basic' will be available. 136 137 1384.2 KVM_CREATE_VM 139----------------- 140 141:Capability: basic 142:Architectures: all 143:Type: system ioctl 144:Parameters: machine type identifier (KVM_VM_*) 145:Returns: a VM fd that can be used to control the new virtual machine. 146 147The new VM has no virtual cpus and no memory. 148You probably want to use 0 as machine type. 149 150In order to create user controlled virtual machines on S390, check 151KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as 152privileged user (CAP_SYS_ADMIN). 153 154On arm64, the physical address size for a VM (IPA Size limit) is limited 155to 40bits by default. The limit can be configured if the host supports the 156extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use 157KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type 158identifier, where IPA_Bits is the maximum width of any physical 159address used by the VM. The IPA_Bits is encoded in bits[7-0] of the 160machine type identifier. 161 162e.g, to configure a guest to use 48bit physical address size:: 163 164 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48)); 165 166The requested size (IPA_Bits) must be: 167 168 == ========================================================= 169 0 Implies default size, 40bits (for backward compatibility) 170 N Implies N bits, where N is a positive integer such that, 171 32 <= N <= Host_IPA_Limit 172 == ========================================================= 173 174Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and 175is dependent on the CPU capability and the kernel configuration. The limit can 176be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION 177ioctl() at run-time. 178 179Creation of the VM will fail if the requested IPA size (whether it is 180implicit or explicit) is unsupported on the host. 181 182Please note that configuring the IPA size does not affect the capability 183exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects 184size of the address translated by the stage2 level (guest physical to 185host physical address translations). 186 187 1884.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST 189---------------------------------------------------------- 190 191:Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST 192:Architectures: x86 193:Type: system ioctl 194:Parameters: struct kvm_msr_list (in/out) 195:Returns: 0 on success; -1 on error 196 197Errors: 198 199 ====== ============================================================ 200 EFAULT the msr index list cannot be read from or written to 201 E2BIG the msr index list is too big to fit in the array specified by 202 the user. 203 ====== ============================================================ 204 205:: 206 207 struct kvm_msr_list { 208 __u32 nmsrs; /* number of msrs in entries */ 209 __u32 indices[0]; 210 }; 211 212The user fills in the size of the indices array in nmsrs, and in return 213kvm adjusts nmsrs to reflect the actual number of msrs and fills in the 214indices array with their numbers. 215 216KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list 217varies by kvm version and host processor, but does not change otherwise. 218 219Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 220not returned in the MSR list, as different vcpus can have a different number 221of banks, as set via the KVM_X86_SETUP_MCE ioctl. 222 223KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed 224to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities 225and processor features that are exposed via MSRs (e.g., VMX capabilities). 226This list also varies by kvm version and host processor, but does not change 227otherwise. 228 229 2304.4 KVM_CHECK_EXTENSION 231----------------------- 232 233:Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl 234:Architectures: all 235:Type: system ioctl, vm ioctl 236:Parameters: extension identifier (KVM_CAP_*) 237:Returns: 0 if unsupported; 1 (or some other positive integer) if supported 238 239The API allows the application to query about extensions to the core 240kvm API. Userspace passes an extension identifier (an integer) and 241receives an integer that describes the extension availability. 242Generally 0 means no and 1 means yes, but some extensions may report 243additional information in the integer return value. 244 245Based on their initialization different VMs may have different capabilities. 246It is thus encouraged to use the vm ioctl to query for capabilities (available 247with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 248 2494.5 KVM_GET_VCPU_MMAP_SIZE 250-------------------------- 251 252:Capability: basic 253:Architectures: all 254:Type: system ioctl 255:Parameters: none 256:Returns: size of vcpu mmap area, in bytes 257 258The KVM_RUN ioctl (cf.) communicates with userspace via a shared 259memory region. This ioctl returns the size of that region. See the 260KVM_RUN documentation for details. 261 262Besides the size of the KVM_RUN communication region, other areas of 263the VCPU file descriptor can be mmap-ed, including: 264 265- if KVM_CAP_COALESCED_MMIO is available, a page at 266 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons, 267 this page is included in the result of KVM_GET_VCPU_MMAP_SIZE. 268 KVM_CAP_COALESCED_MMIO is not documented yet. 269 270- if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at 271 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For more information on 272 KVM_CAP_DIRTY_LOG_RING, see section 8.3. 273 274 2754.6 KVM_SET_MEMORY_REGION 276------------------------- 277 278:Capability: basic 279:Architectures: all 280:Type: vm ioctl 281:Parameters: struct kvm_memory_region (in) 282:Returns: 0 on success, -1 on error 283 284This ioctl is obsolete and has been removed. 285 286 2874.7 KVM_CREATE_VCPU 288------------------- 289 290:Capability: basic 291:Architectures: all 292:Type: vm ioctl 293:Parameters: vcpu id (apic id on x86) 294:Returns: vcpu fd on success, -1 on error 295 296This API adds a vcpu to a virtual machine. No more than max_vcpus may be added. 297The vcpu id is an integer in the range [0, max_vcpu_id). 298 299The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 300the KVM_CHECK_EXTENSION ioctl() at run-time. 301The maximum possible value for max_vcpus can be retrieved using the 302KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 303 304If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 305cpus max. 306If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 307same as the value returned from KVM_CAP_NR_VCPUS. 308 309The maximum possible value for max_vcpu_id can be retrieved using the 310KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time. 311 312If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id 313is the same as the value returned from KVM_CAP_MAX_VCPUS. 314 315On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 316threads in one or more virtual CPU cores. (This is because the 317hardware requires all the hardware threads in a CPU core to be in the 318same partition.) The KVM_CAP_PPC_SMT capability indicates the number 319of vcpus per virtual core (vcore). The vcore id is obtained by 320dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 321given vcore will always be in the same physical core as each other 322(though that might be a different physical core from time to time). 323Userspace can control the threading (SMT) mode of the guest by its 324allocation of vcpu ids. For example, if userspace wants 325single-threaded guest vcpus, it should make all vcpu ids be a multiple 326of the number of vcpus per vcore. 327 328For virtual cpus that have been created with S390 user controlled virtual 329machines, the resulting vcpu fd can be memory mapped at page offset 330KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 331cpu's hardware control block. 332 333 3344.8 KVM_GET_DIRTY_LOG (vm ioctl) 335-------------------------------- 336 337:Capability: basic 338:Architectures: all 339:Type: vm ioctl 340:Parameters: struct kvm_dirty_log (in/out) 341:Returns: 0 on success, -1 on error 342 343:: 344 345 /* for KVM_GET_DIRTY_LOG */ 346 struct kvm_dirty_log { 347 __u32 slot; 348 __u32 padding; 349 union { 350 void __user *dirty_bitmap; /* one bit per page */ 351 __u64 padding; 352 }; 353 }; 354 355Given a memory slot, return a bitmap containing any pages dirtied 356since the last call to this ioctl. Bit 0 is the first page in the 357memory slot. Ensure the entire structure is cleared to avoid padding 358issues. 359 360If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 361the address space for which you want to return the dirty bitmap. See 362KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 363 364The bits in the dirty bitmap are cleared before the ioctl returns, unless 365KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information, 366see the description of the capability. 367 368Note that the Xen shared info page, if configured, shall always be assumed 369to be dirty. KVM will not explicitly mark it such. 370 3714.9 KVM_SET_MEMORY_ALIAS 372------------------------ 373 374:Capability: basic 375:Architectures: x86 376:Type: vm ioctl 377:Parameters: struct kvm_memory_alias (in) 378:Returns: 0 (success), -1 (error) 379 380This ioctl is obsolete and has been removed. 381 382 3834.10 KVM_RUN 384------------ 385 386:Capability: basic 387:Architectures: all 388:Type: vcpu ioctl 389:Parameters: none 390:Returns: 0 on success, -1 on error 391 392Errors: 393 394 ======= ============================================================== 395 EINTR an unmasked signal is pending 396 ENOEXEC the vcpu hasn't been initialized or the guest tried to execute 397 instructions from device memory (arm64) 398 ENOSYS data abort outside memslots with no syndrome info and 399 KVM_CAP_ARM_NISV_TO_USER not enabled (arm64) 400 EPERM SVE feature set but not finalized (arm64) 401 ======= ============================================================== 402 403This ioctl is used to run a guest virtual cpu. While there are no 404explicit parameters, there is an implicit parameter block that can be 405obtained by mmap()ing the vcpu fd at offset 0, with the size given by 406KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct 407kvm_run' (see below). 408 409 4104.11 KVM_GET_REGS 411----------------- 412 413:Capability: basic 414:Architectures: all except arm64 415:Type: vcpu ioctl 416:Parameters: struct kvm_regs (out) 417:Returns: 0 on success, -1 on error 418 419Reads the general purpose registers from the vcpu. 420 421:: 422 423 /* x86 */ 424 struct kvm_regs { 425 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 426 __u64 rax, rbx, rcx, rdx; 427 __u64 rsi, rdi, rsp, rbp; 428 __u64 r8, r9, r10, r11; 429 __u64 r12, r13, r14, r15; 430 __u64 rip, rflags; 431 }; 432 433 /* mips */ 434 struct kvm_regs { 435 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 436 __u64 gpr[32]; 437 __u64 hi; 438 __u64 lo; 439 __u64 pc; 440 }; 441 442 4434.12 KVM_SET_REGS 444----------------- 445 446:Capability: basic 447:Architectures: all except arm64 448:Type: vcpu ioctl 449:Parameters: struct kvm_regs (in) 450:Returns: 0 on success, -1 on error 451 452Writes the general purpose registers into the vcpu. 453 454See KVM_GET_REGS for the data structure. 455 456 4574.13 KVM_GET_SREGS 458------------------ 459 460:Capability: basic 461:Architectures: x86, ppc 462:Type: vcpu ioctl 463:Parameters: struct kvm_sregs (out) 464:Returns: 0 on success, -1 on error 465 466Reads special registers from the vcpu. 467 468:: 469 470 /* x86 */ 471 struct kvm_sregs { 472 struct kvm_segment cs, ds, es, fs, gs, ss; 473 struct kvm_segment tr, ldt; 474 struct kvm_dtable gdt, idt; 475 __u64 cr0, cr2, cr3, cr4, cr8; 476 __u64 efer; 477 __u64 apic_base; 478 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; 479 }; 480 481 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ 482 483interrupt_bitmap is a bitmap of pending external interrupts. At most 484one bit may be set. This interrupt has been acknowledged by the APIC 485but not yet injected into the cpu core. 486 487 4884.14 KVM_SET_SREGS 489------------------ 490 491:Capability: basic 492:Architectures: x86, ppc 493:Type: vcpu ioctl 494:Parameters: struct kvm_sregs (in) 495:Returns: 0 on success, -1 on error 496 497Writes special registers into the vcpu. See KVM_GET_SREGS for the 498data structures. 499 500 5014.15 KVM_TRANSLATE 502------------------ 503 504:Capability: basic 505:Architectures: x86 506:Type: vcpu ioctl 507:Parameters: struct kvm_translation (in/out) 508:Returns: 0 on success, -1 on error 509 510Translates a virtual address according to the vcpu's current address 511translation mode. 512 513:: 514 515 struct kvm_translation { 516 /* in */ 517 __u64 linear_address; 518 519 /* out */ 520 __u64 physical_address; 521 __u8 valid; 522 __u8 writeable; 523 __u8 usermode; 524 __u8 pad[5]; 525 }; 526 527 5284.16 KVM_INTERRUPT 529------------------ 530 531:Capability: basic 532:Architectures: x86, ppc, mips, riscv 533:Type: vcpu ioctl 534:Parameters: struct kvm_interrupt (in) 535:Returns: 0 on success, negative on failure. 536 537Queues a hardware interrupt vector to be injected. 538 539:: 540 541 /* for KVM_INTERRUPT */ 542 struct kvm_interrupt { 543 /* in */ 544 __u32 irq; 545 }; 546 547X86: 548^^^^ 549 550:Returns: 551 552 ========= =================================== 553 0 on success, 554 -EEXIST if an interrupt is already enqueued 555 -EINVAL the irq number is invalid 556 -ENXIO if the PIC is in the kernel 557 -EFAULT if the pointer is invalid 558 ========= =================================== 559 560Note 'irq' is an interrupt vector, not an interrupt pin or line. This 561ioctl is useful if the in-kernel PIC is not used. 562 563PPC: 564^^^^ 565 566Queues an external interrupt to be injected. This ioctl is overleaded 567with 3 different irq values: 568 569a) KVM_INTERRUPT_SET 570 571 This injects an edge type external interrupt into the guest once it's ready 572 to receive interrupts. When injected, the interrupt is done. 573 574b) KVM_INTERRUPT_UNSET 575 576 This unsets any pending interrupt. 577 578 Only available with KVM_CAP_PPC_UNSET_IRQ. 579 580c) KVM_INTERRUPT_SET_LEVEL 581 582 This injects a level type external interrupt into the guest context. The 583 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET 584 is triggered. 585 586 Only available with KVM_CAP_PPC_IRQ_LEVEL. 587 588Note that any value for 'irq' other than the ones stated above is invalid 589and incurs unexpected behavior. 590 591This is an asynchronous vcpu ioctl and can be invoked from any thread. 592 593MIPS: 594^^^^^ 595 596Queues an external interrupt to be injected into the virtual CPU. A negative 597interrupt number dequeues the interrupt. 598 599This is an asynchronous vcpu ioctl and can be invoked from any thread. 600 601RISC-V: 602^^^^^^^ 603 604Queues an external interrupt to be injected into the virutal CPU. This ioctl 605is overloaded with 2 different irq values: 606 607a) KVM_INTERRUPT_SET 608 609 This sets external interrupt for a virtual CPU and it will receive 610 once it is ready. 611 612b) KVM_INTERRUPT_UNSET 613 614 This clears pending external interrupt for a virtual CPU. 615 616This is an asynchronous vcpu ioctl and can be invoked from any thread. 617 618 6194.17 KVM_DEBUG_GUEST 620-------------------- 621 622:Capability: basic 623:Architectures: none 624:Type: vcpu ioctl 625:Parameters: none) 626:Returns: -1 on error 627 628Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. 629 630 6314.18 KVM_GET_MSRS 632----------------- 633 634:Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system) 635:Architectures: x86 636:Type: system ioctl, vcpu ioctl 637:Parameters: struct kvm_msrs (in/out) 638:Returns: number of msrs successfully returned; 639 -1 on error 640 641When used as a system ioctl: 642Reads the values of MSR-based features that are available for the VM. This 643is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values. 644The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST 645in a system ioctl. 646 647When used as a vcpu ioctl: 648Reads model-specific registers from the vcpu. Supported msr indices can 649be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl. 650 651:: 652 653 struct kvm_msrs { 654 __u32 nmsrs; /* number of msrs in entries */ 655 __u32 pad; 656 657 struct kvm_msr_entry entries[0]; 658 }; 659 660 struct kvm_msr_entry { 661 __u32 index; 662 __u32 reserved; 663 __u64 data; 664 }; 665 666Application code should set the 'nmsrs' member (which indicates the 667size of the entries array) and the 'index' member of each array entry. 668kvm will fill in the 'data' member. 669 670 6714.19 KVM_SET_MSRS 672----------------- 673 674:Capability: basic 675:Architectures: x86 676:Type: vcpu ioctl 677:Parameters: struct kvm_msrs (in) 678:Returns: number of msrs successfully set (see below), -1 on error 679 680Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 681data structures. 682 683Application code should set the 'nmsrs' member (which indicates the 684size of the entries array), and the 'index' and 'data' members of each 685array entry. 686 687It tries to set the MSRs in array entries[] one by one. If setting an MSR 688fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated 689by KVM, etc..., it stops processing the MSR list and returns the number of 690MSRs that have been set successfully. 691 692 6934.20 KVM_SET_CPUID 694------------------ 695 696:Capability: basic 697:Architectures: x86 698:Type: vcpu ioctl 699:Parameters: struct kvm_cpuid (in) 700:Returns: 0 on success, -1 on error 701 702Defines the vcpu responses to the cpuid instruction. Applications 703should use the KVM_SET_CPUID2 ioctl if available. 704 705Caveat emptor: 706 - If this IOCTL fails, KVM gives no guarantees that previous valid CPUID 707 configuration (if there is) is not corrupted. Userspace can get a copy 708 of the resulting CPUID configuration through KVM_GET_CPUID2 in case. 709 - Using KVM_SET_CPUID{,2} after KVM_RUN, i.e. changing the guest vCPU model 710 after running the guest, may cause guest instability. 711 - Using heterogeneous CPUID configurations, modulo APIC IDs, topology, etc... 712 may cause guest instability. 713 714:: 715 716 struct kvm_cpuid_entry { 717 __u32 function; 718 __u32 eax; 719 __u32 ebx; 720 __u32 ecx; 721 __u32 edx; 722 __u32 padding; 723 }; 724 725 /* for KVM_SET_CPUID */ 726 struct kvm_cpuid { 727 __u32 nent; 728 __u32 padding; 729 struct kvm_cpuid_entry entries[0]; 730 }; 731 732 7334.21 KVM_SET_SIGNAL_MASK 734------------------------ 735 736:Capability: basic 737:Architectures: all 738:Type: vcpu ioctl 739:Parameters: struct kvm_signal_mask (in) 740:Returns: 0 on success, -1 on error 741 742Defines which signals are blocked during execution of KVM_RUN. This 743signal mask temporarily overrides the threads signal mask. Any 744unblocked signal received (except SIGKILL and SIGSTOP, which retain 745their traditional behaviour) will cause KVM_RUN to return with -EINTR. 746 747Note the signal will only be delivered if not blocked by the original 748signal mask. 749 750:: 751 752 /* for KVM_SET_SIGNAL_MASK */ 753 struct kvm_signal_mask { 754 __u32 len; 755 __u8 sigset[0]; 756 }; 757 758 7594.22 KVM_GET_FPU 760---------------- 761 762:Capability: basic 763:Architectures: x86 764:Type: vcpu ioctl 765:Parameters: struct kvm_fpu (out) 766:Returns: 0 on success, -1 on error 767 768Reads the floating point state from the vcpu. 769 770:: 771 772 /* for KVM_GET_FPU and KVM_SET_FPU */ 773 struct kvm_fpu { 774 __u8 fpr[8][16]; 775 __u16 fcw; 776 __u16 fsw; 777 __u8 ftwx; /* in fxsave format */ 778 __u8 pad1; 779 __u16 last_opcode; 780 __u64 last_ip; 781 __u64 last_dp; 782 __u8 xmm[16][16]; 783 __u32 mxcsr; 784 __u32 pad2; 785 }; 786 787 7884.23 KVM_SET_FPU 789---------------- 790 791:Capability: basic 792:Architectures: x86 793:Type: vcpu ioctl 794:Parameters: struct kvm_fpu (in) 795:Returns: 0 on success, -1 on error 796 797Writes the floating point state to the vcpu. 798 799:: 800 801 /* for KVM_GET_FPU and KVM_SET_FPU */ 802 struct kvm_fpu { 803 __u8 fpr[8][16]; 804 __u16 fcw; 805 __u16 fsw; 806 __u8 ftwx; /* in fxsave format */ 807 __u8 pad1; 808 __u16 last_opcode; 809 __u64 last_ip; 810 __u64 last_dp; 811 __u8 xmm[16][16]; 812 __u32 mxcsr; 813 __u32 pad2; 814 }; 815 816 8174.24 KVM_CREATE_IRQCHIP 818----------------------- 819 820:Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) 821:Architectures: x86, arm64, s390 822:Type: vm ioctl 823:Parameters: none 824:Returns: 0 on success, -1 on error 825 826Creates an interrupt controller model in the kernel. 827On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up 828future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both 829PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. 830On arm64, a GICv2 is created. Any other GIC versions require the usage of 831KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 832KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 833On s390, a dummy irq routing table is created. 834 835Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 836before KVM_CREATE_IRQCHIP can be used. 837 838 8394.25 KVM_IRQ_LINE 840----------------- 841 842:Capability: KVM_CAP_IRQCHIP 843:Architectures: x86, arm64 844:Type: vm ioctl 845:Parameters: struct kvm_irq_level 846:Returns: 0 on success, -1 on error 847 848Sets the level of a GSI input to the interrupt controller model in the kernel. 849On some architectures it is required that an interrupt controller model has 850been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 851interrupts require the level to be set to 1 and then back to 0. 852 853On real hardware, interrupt pins can be active-low or active-high. This 854does not matter for the level field of struct kvm_irq_level: 1 always 855means active (asserted), 0 means inactive (deasserted). 856 857x86 allows the operating system to program the interrupt polarity 858(active-low/active-high) for level-triggered interrupts, and KVM used 859to consider the polarity. However, due to bitrot in the handling of 860active-low interrupts, the above convention is now valid on x86 too. 861This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 862should not present interrupts to the guest as active-low unless this 863capability is present (or unless it is not using the in-kernel irqchip, 864of course). 865 866 867arm64 can signal an interrupt either at the CPU level, or at the 868in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 869use PPIs designated for specific cpus. The irq field is interpreted 870like this:: 871 872 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 | 873 field: | vcpu2_index | irq_type | vcpu_index | irq_id | 874 875The irq_type field has the following values: 876 877- irq_type[0]: 878 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 879- irq_type[1]: 880 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 881 (the vcpu_index field is ignored) 882- irq_type[2]: 883 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 884 885(The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 886 887In both cases, level is used to assert/deassert the line. 888 889When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is 890identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index 891must be zero. 892 893Note that on arm64, the KVM_CAP_IRQCHIP capability only conditions 894injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always 895be used for a userspace interrupt controller. 896 897:: 898 899 struct kvm_irq_level { 900 union { 901 __u32 irq; /* GSI */ 902 __s32 status; /* not used for KVM_IRQ_LEVEL */ 903 }; 904 __u32 level; /* 0 or 1 */ 905 }; 906 907 9084.26 KVM_GET_IRQCHIP 909-------------------- 910 911:Capability: KVM_CAP_IRQCHIP 912:Architectures: x86 913:Type: vm ioctl 914:Parameters: struct kvm_irqchip (in/out) 915:Returns: 0 on success, -1 on error 916 917Reads the state of a kernel interrupt controller created with 918KVM_CREATE_IRQCHIP into a buffer provided by the caller. 919 920:: 921 922 struct kvm_irqchip { 923 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 924 __u32 pad; 925 union { 926 char dummy[512]; /* reserving space */ 927 struct kvm_pic_state pic; 928 struct kvm_ioapic_state ioapic; 929 } chip; 930 }; 931 932 9334.27 KVM_SET_IRQCHIP 934-------------------- 935 936:Capability: KVM_CAP_IRQCHIP 937:Architectures: x86 938:Type: vm ioctl 939:Parameters: struct kvm_irqchip (in) 940:Returns: 0 on success, -1 on error 941 942Sets the state of a kernel interrupt controller created with 943KVM_CREATE_IRQCHIP from a buffer provided by the caller. 944 945:: 946 947 struct kvm_irqchip { 948 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 949 __u32 pad; 950 union { 951 char dummy[512]; /* reserving space */ 952 struct kvm_pic_state pic; 953 struct kvm_ioapic_state ioapic; 954 } chip; 955 }; 956 957 9584.28 KVM_XEN_HVM_CONFIG 959----------------------- 960 961:Capability: KVM_CAP_XEN_HVM 962:Architectures: x86 963:Type: vm ioctl 964:Parameters: struct kvm_xen_hvm_config (in) 965:Returns: 0 on success, -1 on error 966 967Sets the MSR that the Xen HVM guest uses to initialize its hypercall 968page, and provides the starting address and size of the hypercall 969blobs in userspace. When the guest writes the MSR, kvm copies one 970page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 971memory. 972 973:: 974 975 struct kvm_xen_hvm_config { 976 __u32 flags; 977 __u32 msr; 978 __u64 blob_addr_32; 979 __u64 blob_addr_64; 980 __u8 blob_size_32; 981 __u8 blob_size_64; 982 __u8 pad2[30]; 983 }; 984 985If certain flags are returned from the KVM_CAP_XEN_HVM check, they may 986be set in the flags field of this ioctl: 987 988The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag requests KVM to generate 989the contents of the hypercall page automatically; hypercalls will be 990intercepted and passed to userspace through KVM_EXIT_XEN. In this 991ase, all of the blob size and address fields must be zero. 992 993The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates to KVM that userspace 994will always use the KVM_XEN_HVM_EVTCHN_SEND ioctl to deliver event 995channel interrupts rather than manipulating the guest's shared_info 996structures directly. This, in turn, may allow KVM to enable features 997such as intercepting the SCHEDOP_poll hypercall to accelerate PV 998spinlock operation for the guest. Userspace may still use the ioctl 999to deliver events if it was advertised, even if userspace does not 1000send this indication that it will always do so 1001 1002No other flags are currently valid in the struct kvm_xen_hvm_config. 1003 10044.29 KVM_GET_CLOCK 1005------------------ 1006 1007:Capability: KVM_CAP_ADJUST_CLOCK 1008:Architectures: x86 1009:Type: vm ioctl 1010:Parameters: struct kvm_clock_data (out) 1011:Returns: 0 on success, -1 on error 1012 1013Gets the current timestamp of kvmclock as seen by the current guest. In 1014conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 1015such as migration. 1016 1017When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 1018set of bits that KVM can return in struct kvm_clock_data's flag member. 1019 1020The following flags are defined: 1021 1022KVM_CLOCK_TSC_STABLE 1023 If set, the returned value is the exact kvmclock 1024 value seen by all VCPUs at the instant when KVM_GET_CLOCK was called. 1025 If clear, the returned value is simply CLOCK_MONOTONIC plus a constant 1026 offset; the offset can be modified with KVM_SET_CLOCK. KVM will try 1027 to make all VCPUs follow this clock, but the exact value read by each 1028 VCPU could differ, because the host TSC is not stable. 1029 1030KVM_CLOCK_REALTIME 1031 If set, the `realtime` field in the kvm_clock_data 1032 structure is populated with the value of the host's real time 1033 clocksource at the instant when KVM_GET_CLOCK was called. If clear, 1034 the `realtime` field does not contain a value. 1035 1036KVM_CLOCK_HOST_TSC 1037 If set, the `host_tsc` field in the kvm_clock_data 1038 structure is populated with the value of the host's timestamp counter (TSC) 1039 at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field 1040 does not contain a value. 1041 1042:: 1043 1044 struct kvm_clock_data { 1045 __u64 clock; /* kvmclock current value */ 1046 __u32 flags; 1047 __u32 pad0; 1048 __u64 realtime; 1049 __u64 host_tsc; 1050 __u32 pad[4]; 1051 }; 1052 1053 10544.30 KVM_SET_CLOCK 1055------------------ 1056 1057:Capability: KVM_CAP_ADJUST_CLOCK 1058:Architectures: x86 1059:Type: vm ioctl 1060:Parameters: struct kvm_clock_data (in) 1061:Returns: 0 on success, -1 on error 1062 1063Sets the current timestamp of kvmclock to the value specified in its parameter. 1064In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 1065such as migration. 1066 1067The following flags can be passed: 1068 1069KVM_CLOCK_REALTIME 1070 If set, KVM will compare the value of the `realtime` field 1071 with the value of the host's real time clocksource at the instant when 1072 KVM_SET_CLOCK was called. The difference in elapsed time is added to the final 1073 kvmclock value that will be provided to guests. 1074 1075Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored. 1076 1077:: 1078 1079 struct kvm_clock_data { 1080 __u64 clock; /* kvmclock current value */ 1081 __u32 flags; 1082 __u32 pad0; 1083 __u64 realtime; 1084 __u64 host_tsc; 1085 __u32 pad[4]; 1086 }; 1087 1088 10894.31 KVM_GET_VCPU_EVENTS 1090------------------------ 1091 1092:Capability: KVM_CAP_VCPU_EVENTS 1093:Extended by: KVM_CAP_INTR_SHADOW 1094:Architectures: x86, arm64 1095:Type: vcpu ioctl 1096:Parameters: struct kvm_vcpu_event (out) 1097:Returns: 0 on success, -1 on error 1098 1099X86: 1100^^^^ 1101 1102Gets currently pending exceptions, interrupts, and NMIs as well as related 1103states of the vcpu. 1104 1105:: 1106 1107 struct kvm_vcpu_events { 1108 struct { 1109 __u8 injected; 1110 __u8 nr; 1111 __u8 has_error_code; 1112 __u8 pending; 1113 __u32 error_code; 1114 } exception; 1115 struct { 1116 __u8 injected; 1117 __u8 nr; 1118 __u8 soft; 1119 __u8 shadow; 1120 } interrupt; 1121 struct { 1122 __u8 injected; 1123 __u8 pending; 1124 __u8 masked; 1125 __u8 pad; 1126 } nmi; 1127 __u32 sipi_vector; 1128 __u32 flags; 1129 struct { 1130 __u8 smm; 1131 __u8 pending; 1132 __u8 smm_inside_nmi; 1133 __u8 latched_init; 1134 } smi; 1135 __u8 reserved[27]; 1136 __u8 exception_has_payload; 1137 __u64 exception_payload; 1138 }; 1139 1140The following bits are defined in the flags field: 1141 1142- KVM_VCPUEVENT_VALID_SHADOW may be set to signal that 1143 interrupt.shadow contains a valid state. 1144 1145- KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a 1146 valid state. 1147 1148- KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the 1149 exception_has_payload, exception_payload, and exception.pending 1150 fields contain a valid state. This bit will be set whenever 1151 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1152 1153ARM64: 1154^^^^^^ 1155 1156If the guest accesses a device that is being emulated by the host kernel in 1157such a way that a real device would generate a physical SError, KVM may make 1158a virtual SError pending for that VCPU. This system error interrupt remains 1159pending until the guest takes the exception by unmasking PSTATE.A. 1160 1161Running the VCPU may cause it to take a pending SError, or make an access that 1162causes an SError to become pending. The event's description is only valid while 1163the VPCU is not running. 1164 1165This API provides a way to read and write the pending 'event' state that is not 1166visible to the guest. To save, restore or migrate a VCPU the struct representing 1167the state can be read then written using this GET/SET API, along with the other 1168guest-visible registers. It is not possible to 'cancel' an SError that has been 1169made pending. 1170 1171A device being emulated in user-space may also wish to generate an SError. To do 1172this the events structure can be populated by user-space. The current state 1173should be read first, to ensure no existing SError is pending. If an existing 1174SError is pending, the architecture's 'Multiple SError interrupts' rules should 1175be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and 1176Serviceability (RAS) Specification"). 1177 1178SError exceptions always have an ESR value. Some CPUs have the ability to 1179specify what the virtual SError's ESR value should be. These systems will 1180advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will 1181always have a non-zero value when read, and the agent making an SError pending 1182should specify the ISS field in the lower 24 bits of exception.serror_esr. If 1183the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events 1184with exception.has_esr as zero, KVM will choose an ESR. 1185 1186Specifying exception.has_esr on a system that does not support it will return 1187-EINVAL. Setting anything other than the lower 24bits of exception.serror_esr 1188will return -EINVAL. 1189 1190It is not possible to read back a pending external abort (injected via 1191KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered 1192directly to the virtual CPU). 1193 1194:: 1195 1196 struct kvm_vcpu_events { 1197 struct { 1198 __u8 serror_pending; 1199 __u8 serror_has_esr; 1200 __u8 ext_dabt_pending; 1201 /* Align it to 8 bytes */ 1202 __u8 pad[5]; 1203 __u64 serror_esr; 1204 } exception; 1205 __u32 reserved[12]; 1206 }; 1207 12084.32 KVM_SET_VCPU_EVENTS 1209------------------------ 1210 1211:Capability: KVM_CAP_VCPU_EVENTS 1212:Extended by: KVM_CAP_INTR_SHADOW 1213:Architectures: x86, arm64 1214:Type: vcpu ioctl 1215:Parameters: struct kvm_vcpu_event (in) 1216:Returns: 0 on success, -1 on error 1217 1218X86: 1219^^^^ 1220 1221Set pending exceptions, interrupts, and NMIs as well as related states of the 1222vcpu. 1223 1224See KVM_GET_VCPU_EVENTS for the data structure. 1225 1226Fields that may be modified asynchronously by running VCPUs can be excluded 1227from the update. These fields are nmi.pending, sipi_vector, smi.smm, 1228smi.pending. Keep the corresponding bits in the flags field cleared to 1229suppress overwriting the current in-kernel state. The bits are: 1230 1231=============================== ================================== 1232KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel 1233KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector 1234KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct. 1235=============================== ================================== 1236 1237If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 1238the flags field to signal that interrupt.shadow contains a valid state and 1239shall be written into the VCPU. 1240 1241KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 1242 1243If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD 1244can be set in the flags field to signal that the 1245exception_has_payload, exception_payload, and exception.pending fields 1246contain a valid state and shall be written into the VCPU. 1247 1248ARM64: 1249^^^^^^ 1250 1251User space may need to inject several types of events to the guest. 1252 1253Set the pending SError exception state for this VCPU. It is not possible to 1254'cancel' an Serror that has been made pending. 1255 1256If the guest performed an access to I/O memory which could not be handled by 1257userspace, for example because of missing instruction syndrome decode 1258information or because there is no device mapped at the accessed IPA, then 1259userspace can ask the kernel to inject an external abort using the address 1260from the exiting fault on the VCPU. It is a programming error to set 1261ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or 1262KVM_EXIT_ARM_NISV. This feature is only available if the system supports 1263KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in 1264how userspace reports accesses for the above cases to guests, across different 1265userspace implementations. Nevertheless, userspace can still emulate all Arm 1266exceptions by manipulating individual registers using the KVM_SET_ONE_REG API. 1267 1268See KVM_GET_VCPU_EVENTS for the data structure. 1269 1270 12714.33 KVM_GET_DEBUGREGS 1272---------------------- 1273 1274:Capability: KVM_CAP_DEBUGREGS 1275:Architectures: x86 1276:Type: vm ioctl 1277:Parameters: struct kvm_debugregs (out) 1278:Returns: 0 on success, -1 on error 1279 1280Reads debug registers from the vcpu. 1281 1282:: 1283 1284 struct kvm_debugregs { 1285 __u64 db[4]; 1286 __u64 dr6; 1287 __u64 dr7; 1288 __u64 flags; 1289 __u64 reserved[9]; 1290 }; 1291 1292 12934.34 KVM_SET_DEBUGREGS 1294---------------------- 1295 1296:Capability: KVM_CAP_DEBUGREGS 1297:Architectures: x86 1298:Type: vm ioctl 1299:Parameters: struct kvm_debugregs (in) 1300:Returns: 0 on success, -1 on error 1301 1302Writes debug registers into the vcpu. 1303 1304See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 1305yet and must be cleared on entry. 1306 1307 13084.35 KVM_SET_USER_MEMORY_REGION 1309------------------------------- 1310 1311:Capability: KVM_CAP_USER_MEMORY 1312:Architectures: all 1313:Type: vm ioctl 1314:Parameters: struct kvm_userspace_memory_region (in) 1315:Returns: 0 on success, -1 on error 1316 1317:: 1318 1319 struct kvm_userspace_memory_region { 1320 __u32 slot; 1321 __u32 flags; 1322 __u64 guest_phys_addr; 1323 __u64 memory_size; /* bytes */ 1324 __u64 userspace_addr; /* start of the userspace allocated memory */ 1325 }; 1326 1327 /* for kvm_memory_region::flags */ 1328 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 1329 #define KVM_MEM_READONLY (1UL << 1) 1330 1331This ioctl allows the user to create, modify or delete a guest physical 1332memory slot. Bits 0-15 of "slot" specify the slot id and this value 1333should be less than the maximum number of user memory slots supported per 1334VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS. 1335Slots may not overlap in guest physical address space. 1336 1337If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 1338specifies the address space which is being modified. They must be 1339less than the value that KVM_CHECK_EXTENSION returns for the 1340KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 1341are unrelated; the restriction on overlapping slots only applies within 1342each address space. 1343 1344Deleting a slot is done by passing zero for memory_size. When changing 1345an existing slot, it may be moved in the guest physical memory space, 1346or its flags may be modified, but it may not be resized. 1347 1348Memory for the region is taken starting at the address denoted by the 1349field userspace_addr, which must point at user addressable memory for 1350the entire memory slot size. Any object may back this memory, including 1351anonymous memory, ordinary files, and hugetlbfs. 1352 1353On architectures that support a form of address tagging, userspace_addr must 1354be an untagged address. 1355 1356It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 1357be identical. This allows large pages in the guest to be backed by large 1358pages in the host. 1359 1360The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 1361KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 1362writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 1363use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 1364to make a new slot read-only. In this case, writes to this memory will be 1365posted to userspace as KVM_EXIT_MMIO exits. 1366 1367When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 1368the memory region are automatically reflected into the guest. For example, an 1369mmap() that affects the region will be made visible immediately. Another 1370example is madvise(MADV_DROP). 1371 1372It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. 1373The KVM_SET_MEMORY_REGION does not allow fine grained control over memory 1374allocation and is deprecated. 1375 1376 13774.36 KVM_SET_TSS_ADDR 1378--------------------- 1379 1380:Capability: KVM_CAP_SET_TSS_ADDR 1381:Architectures: x86 1382:Type: vm ioctl 1383:Parameters: unsigned long tss_address (in) 1384:Returns: 0 on success, -1 on error 1385 1386This ioctl defines the physical address of a three-page region in the guest 1387physical address space. The region must be within the first 4GB of the 1388guest physical address space and must not conflict with any memory slot 1389or any mmio address. The guest may malfunction if it accesses this memory 1390region. 1391 1392This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1393because of a quirk in the virtualization implementation (see the internals 1394documentation when it pops into existence). 1395 1396 13974.37 KVM_ENABLE_CAP 1398------------------- 1399 1400:Capability: KVM_CAP_ENABLE_CAP 1401:Architectures: mips, ppc, s390, x86 1402:Type: vcpu ioctl 1403:Parameters: struct kvm_enable_cap (in) 1404:Returns: 0 on success; -1 on error 1405 1406:Capability: KVM_CAP_ENABLE_CAP_VM 1407:Architectures: all 1408:Type: vm ioctl 1409:Parameters: struct kvm_enable_cap (in) 1410:Returns: 0 on success; -1 on error 1411 1412.. note:: 1413 1414 Not all extensions are enabled by default. Using this ioctl the application 1415 can enable an extension, making it available to the guest. 1416 1417On systems that do not support this ioctl, it always fails. On systems that 1418do support it, it only works for extensions that are supported for enablement. 1419 1420To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1421be used. 1422 1423:: 1424 1425 struct kvm_enable_cap { 1426 /* in */ 1427 __u32 cap; 1428 1429The capability that is supposed to get enabled. 1430 1431:: 1432 1433 __u32 flags; 1434 1435A bitfield indicating future enhancements. Has to be 0 for now. 1436 1437:: 1438 1439 __u64 args[4]; 1440 1441Arguments for enabling a feature. If a feature needs initial values to 1442function properly, this is the place to put them. 1443 1444:: 1445 1446 __u8 pad[64]; 1447 }; 1448 1449The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1450for vm-wide capabilities. 1451 14524.38 KVM_GET_MP_STATE 1453--------------------- 1454 1455:Capability: KVM_CAP_MP_STATE 1456:Architectures: x86, s390, arm64, riscv 1457:Type: vcpu ioctl 1458:Parameters: struct kvm_mp_state (out) 1459:Returns: 0 on success; -1 on error 1460 1461:: 1462 1463 struct kvm_mp_state { 1464 __u32 mp_state; 1465 }; 1466 1467Returns the vcpu's current "multiprocessing state" (though also valid on 1468uniprocessor guests). 1469 1470Possible values are: 1471 1472 ========================== =============================================== 1473 KVM_MP_STATE_RUNNABLE the vcpu is currently running 1474 [x86,arm64,riscv] 1475 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP) 1476 which has not yet received an INIT signal [x86] 1477 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is 1478 now ready for a SIPI [x86] 1479 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and 1480 is waiting for an interrupt [x86] 1481 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector 1482 accessible via KVM_GET_VCPU_EVENTS) [x86] 1483 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm64,riscv] 1484 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390] 1485 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted) 1486 [s390] 1487 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state 1488 [s390] 1489 KVM_MP_STATE_SUSPENDED the vcpu is in a suspend state and is waiting 1490 for a wakeup event [arm64] 1491 ========================== =============================================== 1492 1493On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1494in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1495these architectures. 1496 1497For arm64: 1498^^^^^^^^^^ 1499 1500If a vCPU is in the KVM_MP_STATE_SUSPENDED state, KVM will emulate the 1501architectural execution of a WFI instruction. 1502 1503If a wakeup event is recognized, KVM will exit to userspace with a 1504KVM_SYSTEM_EVENT exit, where the event type is KVM_SYSTEM_EVENT_WAKEUP. If 1505userspace wants to honor the wakeup, it must set the vCPU's MP state to 1506KVM_MP_STATE_RUNNABLE. If it does not, KVM will continue to await a wakeup 1507event in subsequent calls to KVM_RUN. 1508 1509.. warning:: 1510 1511 If userspace intends to keep the vCPU in a SUSPENDED state, it is 1512 strongly recommended that userspace take action to suppress the 1513 wakeup event (such as masking an interrupt). Otherwise, subsequent 1514 calls to KVM_RUN will immediately exit with a KVM_SYSTEM_EVENT_WAKEUP 1515 event and inadvertently waste CPU cycles. 1516 1517 Additionally, if userspace takes action to suppress a wakeup event, 1518 it is strongly recommended that it also restores the vCPU to its 1519 original state when the vCPU is made RUNNABLE again. For example, 1520 if userspace masked a pending interrupt to suppress the wakeup, 1521 the interrupt should be unmasked before returning control to the 1522 guest. 1523 1524For riscv: 1525^^^^^^^^^^ 1526 1527The only states that are valid are KVM_MP_STATE_STOPPED and 1528KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1529 15304.39 KVM_SET_MP_STATE 1531--------------------- 1532 1533:Capability: KVM_CAP_MP_STATE 1534:Architectures: x86, s390, arm64, riscv 1535:Type: vcpu ioctl 1536:Parameters: struct kvm_mp_state (in) 1537:Returns: 0 on success; -1 on error 1538 1539Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1540arguments. 1541 1542On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1543in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1544these architectures. 1545 1546For arm64/riscv: 1547^^^^^^^^^^^^^^^^ 1548 1549The only states that are valid are KVM_MP_STATE_STOPPED and 1550KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1551 15524.40 KVM_SET_IDENTITY_MAP_ADDR 1553------------------------------ 1554 1555:Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1556:Architectures: x86 1557:Type: vm ioctl 1558:Parameters: unsigned long identity (in) 1559:Returns: 0 on success, -1 on error 1560 1561This ioctl defines the physical address of a one-page region in the guest 1562physical address space. The region must be within the first 4GB of the 1563guest physical address space and must not conflict with any memory slot 1564or any mmio address. The guest may malfunction if it accesses this memory 1565region. 1566 1567Setting the address to 0 will result in resetting the address to its default 1568(0xfffbc000). 1569 1570This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1571because of a quirk in the virtualization implementation (see the internals 1572documentation when it pops into existence). 1573 1574Fails if any VCPU has already been created. 1575 15764.41 KVM_SET_BOOT_CPU_ID 1577------------------------ 1578 1579:Capability: KVM_CAP_SET_BOOT_CPU_ID 1580:Architectures: x86 1581:Type: vm ioctl 1582:Parameters: unsigned long vcpu_id 1583:Returns: 0 on success, -1 on error 1584 1585Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1586as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1587is vcpu 0. This ioctl has to be called before vcpu creation, 1588otherwise it will return EBUSY error. 1589 1590 15914.42 KVM_GET_XSAVE 1592------------------ 1593 1594:Capability: KVM_CAP_XSAVE 1595:Architectures: x86 1596:Type: vcpu ioctl 1597:Parameters: struct kvm_xsave (out) 1598:Returns: 0 on success, -1 on error 1599 1600 1601:: 1602 1603 struct kvm_xsave { 1604 __u32 region[1024]; 1605 __u32 extra[0]; 1606 }; 1607 1608This ioctl would copy current vcpu's xsave struct to the userspace. 1609 1610 16114.43 KVM_SET_XSAVE 1612------------------ 1613 1614:Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2 1615:Architectures: x86 1616:Type: vcpu ioctl 1617:Parameters: struct kvm_xsave (in) 1618:Returns: 0 on success, -1 on error 1619 1620:: 1621 1622 1623 struct kvm_xsave { 1624 __u32 region[1024]; 1625 __u32 extra[0]; 1626 }; 1627 1628This ioctl would copy userspace's xsave struct to the kernel. It copies 1629as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2), 1630when invoked on the vm file descriptor. The size value returned by 1631KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 1632Currently, it is only greater than 4096 if a dynamic feature has been 1633enabled with ``arch_prctl()``, but this may change in the future. 1634 1635The offsets of the state save areas in struct kvm_xsave follow the 1636contents of CPUID leaf 0xD on the host. 1637 1638 16394.44 KVM_GET_XCRS 1640----------------- 1641 1642:Capability: KVM_CAP_XCRS 1643:Architectures: x86 1644:Type: vcpu ioctl 1645:Parameters: struct kvm_xcrs (out) 1646:Returns: 0 on success, -1 on error 1647 1648:: 1649 1650 struct kvm_xcr { 1651 __u32 xcr; 1652 __u32 reserved; 1653 __u64 value; 1654 }; 1655 1656 struct kvm_xcrs { 1657 __u32 nr_xcrs; 1658 __u32 flags; 1659 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1660 __u64 padding[16]; 1661 }; 1662 1663This ioctl would copy current vcpu's xcrs to the userspace. 1664 1665 16664.45 KVM_SET_XCRS 1667----------------- 1668 1669:Capability: KVM_CAP_XCRS 1670:Architectures: x86 1671:Type: vcpu ioctl 1672:Parameters: struct kvm_xcrs (in) 1673:Returns: 0 on success, -1 on error 1674 1675:: 1676 1677 struct kvm_xcr { 1678 __u32 xcr; 1679 __u32 reserved; 1680 __u64 value; 1681 }; 1682 1683 struct kvm_xcrs { 1684 __u32 nr_xcrs; 1685 __u32 flags; 1686 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1687 __u64 padding[16]; 1688 }; 1689 1690This ioctl would set vcpu's xcr to the value userspace specified. 1691 1692 16934.46 KVM_GET_SUPPORTED_CPUID 1694---------------------------- 1695 1696:Capability: KVM_CAP_EXT_CPUID 1697:Architectures: x86 1698:Type: system ioctl 1699:Parameters: struct kvm_cpuid2 (in/out) 1700:Returns: 0 on success, -1 on error 1701 1702:: 1703 1704 struct kvm_cpuid2 { 1705 __u32 nent; 1706 __u32 padding; 1707 struct kvm_cpuid_entry2 entries[0]; 1708 }; 1709 1710 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1711 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 1712 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 1713 1714 struct kvm_cpuid_entry2 { 1715 __u32 function; 1716 __u32 index; 1717 __u32 flags; 1718 __u32 eax; 1719 __u32 ebx; 1720 __u32 ecx; 1721 __u32 edx; 1722 __u32 padding[3]; 1723 }; 1724 1725This ioctl returns x86 cpuid features which are supported by both the 1726hardware and kvm in its default configuration. Userspace can use the 1727information returned by this ioctl to construct cpuid information (for 1728KVM_SET_CPUID2) that is consistent with hardware, kernel, and 1729userspace capabilities, and with user requirements (for example, the 1730user may wish to constrain cpuid to emulate older hardware, or for 1731feature consistency across a cluster). 1732 1733Dynamically-enabled feature bits need to be requested with 1734``arch_prctl()`` before calling this ioctl. Feature bits that have not 1735been requested are excluded from the result. 1736 1737Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may 1738expose cpuid features (e.g. MONITOR) which are not supported by kvm in 1739its default configuration. If userspace enables such capabilities, it 1740is responsible for modifying the results of this ioctl appropriately. 1741 1742Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1743with the 'nent' field indicating the number of entries in the variable-size 1744array 'entries'. If the number of entries is too low to describe the cpu 1745capabilities, an error (E2BIG) is returned. If the number is too high, 1746the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1747number is just right, the 'nent' field is adjusted to the number of valid 1748entries in the 'entries' array, which is then filled. 1749 1750The entries returned are the host cpuid as returned by the cpuid instruction, 1751with unknown or unsupported features masked out. Some features (for example, 1752x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1753emulate them efficiently. The fields in each entry are defined as follows: 1754 1755 function: 1756 the eax value used to obtain the entry 1757 1758 index: 1759 the ecx value used to obtain the entry (for entries that are 1760 affected by ecx) 1761 1762 flags: 1763 an OR of zero or more of the following: 1764 1765 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1766 if the index field is valid 1767 1768 eax, ebx, ecx, edx: 1769 the values returned by the cpuid instruction for 1770 this function/index combination 1771 1772The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned 1773as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC 1774support. Instead it is reported via:: 1775 1776 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1777 1778if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1779feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1780 1781 17824.47 KVM_PPC_GET_PVINFO 1783----------------------- 1784 1785:Capability: KVM_CAP_PPC_GET_PVINFO 1786:Architectures: ppc 1787:Type: vm ioctl 1788:Parameters: struct kvm_ppc_pvinfo (out) 1789:Returns: 0 on success, !0 on error 1790 1791:: 1792 1793 struct kvm_ppc_pvinfo { 1794 __u32 flags; 1795 __u32 hcall[4]; 1796 __u8 pad[108]; 1797 }; 1798 1799This ioctl fetches PV specific information that need to be passed to the guest 1800using the device tree or other means from vm context. 1801 1802The hcall array defines 4 instructions that make up a hypercall. 1803 1804If any additional field gets added to this structure later on, a bit for that 1805additional piece of information will be set in the flags bitmap. 1806 1807The flags bitmap is defined as:: 1808 1809 /* the host supports the ePAPR idle hcall 1810 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1811 18124.52 KVM_SET_GSI_ROUTING 1813------------------------ 1814 1815:Capability: KVM_CAP_IRQ_ROUTING 1816:Architectures: x86 s390 arm64 1817:Type: vm ioctl 1818:Parameters: struct kvm_irq_routing (in) 1819:Returns: 0 on success, -1 on error 1820 1821Sets the GSI routing table entries, overwriting any previously set entries. 1822 1823On arm64, GSI routing has the following limitation: 1824 1825- GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1826 1827:: 1828 1829 struct kvm_irq_routing { 1830 __u32 nr; 1831 __u32 flags; 1832 struct kvm_irq_routing_entry entries[0]; 1833 }; 1834 1835No flags are specified so far, the corresponding field must be set to zero. 1836 1837:: 1838 1839 struct kvm_irq_routing_entry { 1840 __u32 gsi; 1841 __u32 type; 1842 __u32 flags; 1843 __u32 pad; 1844 union { 1845 struct kvm_irq_routing_irqchip irqchip; 1846 struct kvm_irq_routing_msi msi; 1847 struct kvm_irq_routing_s390_adapter adapter; 1848 struct kvm_irq_routing_hv_sint hv_sint; 1849 struct kvm_irq_routing_xen_evtchn xen_evtchn; 1850 __u32 pad[8]; 1851 } u; 1852 }; 1853 1854 /* gsi routing entry types */ 1855 #define KVM_IRQ_ROUTING_IRQCHIP 1 1856 #define KVM_IRQ_ROUTING_MSI 2 1857 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1858 #define KVM_IRQ_ROUTING_HV_SINT 4 1859 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 1860 1861flags: 1862 1863- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1864 type, specifies that the devid field contains a valid value. The per-VM 1865 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1866 the device ID. If this capability is not available, userspace should 1867 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1868- zero otherwise 1869 1870:: 1871 1872 struct kvm_irq_routing_irqchip { 1873 __u32 irqchip; 1874 __u32 pin; 1875 }; 1876 1877 struct kvm_irq_routing_msi { 1878 __u32 address_lo; 1879 __u32 address_hi; 1880 __u32 data; 1881 union { 1882 __u32 pad; 1883 __u32 devid; 1884 }; 1885 }; 1886 1887If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1888for the device that wrote the MSI message. For PCI, this is usually a 1889BFD identifier in the lower 16 bits. 1890 1891On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1892feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1893address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1894address_hi must be zero. 1895 1896:: 1897 1898 struct kvm_irq_routing_s390_adapter { 1899 __u64 ind_addr; 1900 __u64 summary_addr; 1901 __u64 ind_offset; 1902 __u32 summary_offset; 1903 __u32 adapter_id; 1904 }; 1905 1906 struct kvm_irq_routing_hv_sint { 1907 __u32 vcpu; 1908 __u32 sint; 1909 }; 1910 1911 struct kvm_irq_routing_xen_evtchn { 1912 __u32 port; 1913 __u32 vcpu; 1914 __u32 priority; 1915 }; 1916 1917 1918When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL bit 1919in its indication of supported features, routing to Xen event channels 1920is supported. Although the priority field is present, only the value 1921KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported, which means delivery by 19222 level event channels. FIFO event channel support may be added in 1923the future. 1924 1925 19264.55 KVM_SET_TSC_KHZ 1927-------------------- 1928 1929:Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM_TSC_CONTROL 1930:Architectures: x86 1931:Type: vcpu ioctl / vm ioctl 1932:Parameters: virtual tsc_khz 1933:Returns: 0 on success, -1 on error 1934 1935Specifies the tsc frequency for the virtual machine. The unit of the 1936frequency is KHz. 1937 1938If the KVM_CAP_VM_TSC_CONTROL capability is advertised, this can also 1939be used as a vm ioctl to set the initial tsc frequency of subsequently 1940created vCPUs. 1941 19424.56 KVM_GET_TSC_KHZ 1943-------------------- 1944 1945:Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM_TSC_CONTROL 1946:Architectures: x86 1947:Type: vcpu ioctl / vm ioctl 1948:Parameters: none 1949:Returns: virtual tsc-khz on success, negative value on error 1950 1951Returns the tsc frequency of the guest. The unit of the return value is 1952KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 1953error. 1954 1955 19564.57 KVM_GET_LAPIC 1957------------------ 1958 1959:Capability: KVM_CAP_IRQCHIP 1960:Architectures: x86 1961:Type: vcpu ioctl 1962:Parameters: struct kvm_lapic_state (out) 1963:Returns: 0 on success, -1 on error 1964 1965:: 1966 1967 #define KVM_APIC_REG_SIZE 0x400 1968 struct kvm_lapic_state { 1969 char regs[KVM_APIC_REG_SIZE]; 1970 }; 1971 1972Reads the Local APIC registers and copies them into the input argument. The 1973data format and layout are the same as documented in the architecture manual. 1974 1975If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 1976enabled, then the format of APIC_ID register depends on the APIC mode 1977(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 1978the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 1979which is stored in bits 31-24 of the APIC register, or equivalently in 1980byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 1981be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 1982 1983If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 1984always uses xAPIC format. 1985 1986 19874.58 KVM_SET_LAPIC 1988------------------ 1989 1990:Capability: KVM_CAP_IRQCHIP 1991:Architectures: x86 1992:Type: vcpu ioctl 1993:Parameters: struct kvm_lapic_state (in) 1994:Returns: 0 on success, -1 on error 1995 1996:: 1997 1998 #define KVM_APIC_REG_SIZE 0x400 1999 struct kvm_lapic_state { 2000 char regs[KVM_APIC_REG_SIZE]; 2001 }; 2002 2003Copies the input argument into the Local APIC registers. The data format 2004and layout are the same as documented in the architecture manual. 2005 2006The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 2007regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 2008See the note in KVM_GET_LAPIC. 2009 2010 20114.59 KVM_IOEVENTFD 2012------------------ 2013 2014:Capability: KVM_CAP_IOEVENTFD 2015:Architectures: all 2016:Type: vm ioctl 2017:Parameters: struct kvm_ioeventfd (in) 2018:Returns: 0 on success, !0 on error 2019 2020This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 2021within the guest. A guest write in the registered address will signal the 2022provided event instead of triggering an exit. 2023 2024:: 2025 2026 struct kvm_ioeventfd { 2027 __u64 datamatch; 2028 __u64 addr; /* legal pio/mmio address */ 2029 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 2030 __s32 fd; 2031 __u32 flags; 2032 __u8 pad[36]; 2033 }; 2034 2035For the special case of virtio-ccw devices on s390, the ioevent is matched 2036to a subchannel/virtqueue tuple instead. 2037 2038The following flags are defined:: 2039 2040 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 2041 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 2042 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 2043 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 2044 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 2045 2046If datamatch flag is set, the event will be signaled only if the written value 2047to the registered address is equal to datamatch in struct kvm_ioeventfd. 2048 2049For virtio-ccw devices, addr contains the subchannel id and datamatch the 2050virtqueue index. 2051 2052With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 2053the kernel will ignore the length of guest write and may get a faster vmexit. 2054The speedup may only apply to specific architectures, but the ioeventfd will 2055work anyway. 2056 20574.60 KVM_DIRTY_TLB 2058------------------ 2059 2060:Capability: KVM_CAP_SW_TLB 2061:Architectures: ppc 2062:Type: vcpu ioctl 2063:Parameters: struct kvm_dirty_tlb (in) 2064:Returns: 0 on success, -1 on error 2065 2066:: 2067 2068 struct kvm_dirty_tlb { 2069 __u64 bitmap; 2070 __u32 num_dirty; 2071 }; 2072 2073This must be called whenever userspace has changed an entry in the shared 2074TLB, prior to calling KVM_RUN on the associated vcpu. 2075 2076The "bitmap" field is the userspace address of an array. This array 2077consists of a number of bits, equal to the total number of TLB entries as 2078determined by the last successful call to KVM_CONFIG_TLB, rounded up to the 2079nearest multiple of 64. 2080 2081Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 2082array. 2083 2084The array is little-endian: the bit 0 is the least significant bit of the 2085first byte, bit 8 is the least significant bit of the second byte, etc. 2086This avoids any complications with differing word sizes. 2087 2088The "num_dirty" field is a performance hint for KVM to determine whether it 2089should skip processing the bitmap and just invalidate everything. It must 2090be set to the number of set bits in the bitmap. 2091 2092 20934.62 KVM_CREATE_SPAPR_TCE 2094------------------------- 2095 2096:Capability: KVM_CAP_SPAPR_TCE 2097:Architectures: powerpc 2098:Type: vm ioctl 2099:Parameters: struct kvm_create_spapr_tce (in) 2100:Returns: file descriptor for manipulating the created TCE table 2101 2102This creates a virtual TCE (translation control entry) table, which 2103is an IOMMU for PAPR-style virtual I/O. It is used to translate 2104logical addresses used in virtual I/O into guest physical addresses, 2105and provides a scatter/gather capability for PAPR virtual I/O. 2106 2107:: 2108 2109 /* for KVM_CAP_SPAPR_TCE */ 2110 struct kvm_create_spapr_tce { 2111 __u64 liobn; 2112 __u32 window_size; 2113 }; 2114 2115The liobn field gives the logical IO bus number for which to create a 2116TCE table. The window_size field specifies the size of the DMA window 2117which this TCE table will translate - the table will contain one 64 2118bit TCE entry for every 4kiB of the DMA window. 2119 2120When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 2121table has been created using this ioctl(), the kernel will handle it 2122in real mode, updating the TCE table. H_PUT_TCE calls for other 2123liobns will cause a vm exit and must be handled by userspace. 2124 2125The return value is a file descriptor which can be passed to mmap(2) 2126to map the created TCE table into userspace. This lets userspace read 2127the entries written by kernel-handled H_PUT_TCE calls, and also lets 2128userspace update the TCE table directly which is useful in some 2129circumstances. 2130 2131 21324.63 KVM_ALLOCATE_RMA 2133--------------------- 2134 2135:Capability: KVM_CAP_PPC_RMA 2136:Architectures: powerpc 2137:Type: vm ioctl 2138:Parameters: struct kvm_allocate_rma (out) 2139:Returns: file descriptor for mapping the allocated RMA 2140 2141This allocates a Real Mode Area (RMA) from the pool allocated at boot 2142time by the kernel. An RMA is a physically-contiguous, aligned region 2143of memory used on older POWER processors to provide the memory which 2144will be accessed by real-mode (MMU off) accesses in a KVM guest. 2145POWER processors support a set of sizes for the RMA that usually 2146includes 64MB, 128MB, 256MB and some larger powers of two. 2147 2148:: 2149 2150 /* for KVM_ALLOCATE_RMA */ 2151 struct kvm_allocate_rma { 2152 __u64 rma_size; 2153 }; 2154 2155The return value is a file descriptor which can be passed to mmap(2) 2156to map the allocated RMA into userspace. The mapped area can then be 2157passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the 2158RMA for a virtual machine. The size of the RMA in bytes (which is 2159fixed at host kernel boot time) is returned in the rma_size field of 2160the argument structure. 2161 2162The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl 2163is supported; 2 if the processor requires all virtual machines to have 2164an RMA, or 1 if the processor can use an RMA but doesn't require it, 2165because it supports the Virtual RMA (VRMA) facility. 2166 2167 21684.64 KVM_NMI 2169------------ 2170 2171:Capability: KVM_CAP_USER_NMI 2172:Architectures: x86 2173:Type: vcpu ioctl 2174:Parameters: none 2175:Returns: 0 on success, -1 on error 2176 2177Queues an NMI on the thread's vcpu. Note this is well defined only 2178when KVM_CREATE_IRQCHIP has not been called, since this is an interface 2179between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 2180has been called, this interface is completely emulated within the kernel. 2181 2182To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 2183following algorithm: 2184 2185 - pause the vcpu 2186 - read the local APIC's state (KVM_GET_LAPIC) 2187 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 2188 - if so, issue KVM_NMI 2189 - resume the vcpu 2190 2191Some guests configure the LINT1 NMI input to cause a panic, aiding in 2192debugging. 2193 2194 21954.65 KVM_S390_UCAS_MAP 2196---------------------- 2197 2198:Capability: KVM_CAP_S390_UCONTROL 2199:Architectures: s390 2200:Type: vcpu ioctl 2201:Parameters: struct kvm_s390_ucas_mapping (in) 2202:Returns: 0 in case of success 2203 2204The parameter is defined like this:: 2205 2206 struct kvm_s390_ucas_mapping { 2207 __u64 user_addr; 2208 __u64 vcpu_addr; 2209 __u64 length; 2210 }; 2211 2212This ioctl maps the memory at "user_addr" with the length "length" to 2213the vcpu's address space starting at "vcpu_addr". All parameters need to 2214be aligned by 1 megabyte. 2215 2216 22174.66 KVM_S390_UCAS_UNMAP 2218------------------------ 2219 2220:Capability: KVM_CAP_S390_UCONTROL 2221:Architectures: s390 2222:Type: vcpu ioctl 2223:Parameters: struct kvm_s390_ucas_mapping (in) 2224:Returns: 0 in case of success 2225 2226The parameter is defined like this:: 2227 2228 struct kvm_s390_ucas_mapping { 2229 __u64 user_addr; 2230 __u64 vcpu_addr; 2231 __u64 length; 2232 }; 2233 2234This ioctl unmaps the memory in the vcpu's address space starting at 2235"vcpu_addr" with the length "length". The field "user_addr" is ignored. 2236All parameters need to be aligned by 1 megabyte. 2237 2238 22394.67 KVM_S390_VCPU_FAULT 2240------------------------ 2241 2242:Capability: KVM_CAP_S390_UCONTROL 2243:Architectures: s390 2244:Type: vcpu ioctl 2245:Parameters: vcpu absolute address (in) 2246:Returns: 0 in case of success 2247 2248This call creates a page table entry on the virtual cpu's address space 2249(for user controlled virtual machines) or the virtual machine's address 2250space (for regular virtual machines). This only works for minor faults, 2251thus it's recommended to access subject memory page via the user page 2252table upfront. This is useful to handle validity intercepts for user 2253controlled virtual machines to fault in the virtual cpu's lowcore pages 2254prior to calling the KVM_RUN ioctl. 2255 2256 22574.68 KVM_SET_ONE_REG 2258-------------------- 2259 2260:Capability: KVM_CAP_ONE_REG 2261:Architectures: all 2262:Type: vcpu ioctl 2263:Parameters: struct kvm_one_reg (in) 2264:Returns: 0 on success, negative value on failure 2265 2266Errors: 2267 2268 ====== ============================================================ 2269 ENOENT no such register 2270 EINVAL invalid register ID, or no such register or used with VMs in 2271 protected virtualization mode on s390 2272 EPERM (arm64) register access not allowed before vcpu finalization 2273 ====== ============================================================ 2274 2275(These error codes are indicative only: do not rely on a specific error 2276code being returned in a specific situation.) 2277 2278:: 2279 2280 struct kvm_one_reg { 2281 __u64 id; 2282 __u64 addr; 2283 }; 2284 2285Using this ioctl, a single vcpu register can be set to a specific value 2286defined by user space with the passed in struct kvm_one_reg, where id 2287refers to the register identifier as described below and addr is a pointer 2288to a variable with the respective size. There can be architecture agnostic 2289and architecture specific registers. Each have their own range of operation 2290and their own constants and width. To keep track of the implemented 2291registers, find a list below: 2292 2293 ======= =============================== ============ 2294 Arch Register Width (bits) 2295 ======= =============================== ============ 2296 PPC KVM_REG_PPC_HIOR 64 2297 PPC KVM_REG_PPC_IAC1 64 2298 PPC KVM_REG_PPC_IAC2 64 2299 PPC KVM_REG_PPC_IAC3 64 2300 PPC KVM_REG_PPC_IAC4 64 2301 PPC KVM_REG_PPC_DAC1 64 2302 PPC KVM_REG_PPC_DAC2 64 2303 PPC KVM_REG_PPC_DABR 64 2304 PPC KVM_REG_PPC_DSCR 64 2305 PPC KVM_REG_PPC_PURR 64 2306 PPC KVM_REG_PPC_SPURR 64 2307 PPC KVM_REG_PPC_DAR 64 2308 PPC KVM_REG_PPC_DSISR 32 2309 PPC KVM_REG_PPC_AMR 64 2310 PPC KVM_REG_PPC_UAMOR 64 2311 PPC KVM_REG_PPC_MMCR0 64 2312 PPC KVM_REG_PPC_MMCR1 64 2313 PPC KVM_REG_PPC_MMCRA 64 2314 PPC KVM_REG_PPC_MMCR2 64 2315 PPC KVM_REG_PPC_MMCRS 64 2316 PPC KVM_REG_PPC_MMCR3 64 2317 PPC KVM_REG_PPC_SIAR 64 2318 PPC KVM_REG_PPC_SDAR 64 2319 PPC KVM_REG_PPC_SIER 64 2320 PPC KVM_REG_PPC_SIER2 64 2321 PPC KVM_REG_PPC_SIER3 64 2322 PPC KVM_REG_PPC_PMC1 32 2323 PPC KVM_REG_PPC_PMC2 32 2324 PPC KVM_REG_PPC_PMC3 32 2325 PPC KVM_REG_PPC_PMC4 32 2326 PPC KVM_REG_PPC_PMC5 32 2327 PPC KVM_REG_PPC_PMC6 32 2328 PPC KVM_REG_PPC_PMC7 32 2329 PPC KVM_REG_PPC_PMC8 32 2330 PPC KVM_REG_PPC_FPR0 64 2331 ... 2332 PPC KVM_REG_PPC_FPR31 64 2333 PPC KVM_REG_PPC_VR0 128 2334 ... 2335 PPC KVM_REG_PPC_VR31 128 2336 PPC KVM_REG_PPC_VSR0 128 2337 ... 2338 PPC KVM_REG_PPC_VSR31 128 2339 PPC KVM_REG_PPC_FPSCR 64 2340 PPC KVM_REG_PPC_VSCR 32 2341 PPC KVM_REG_PPC_VPA_ADDR 64 2342 PPC KVM_REG_PPC_VPA_SLB 128 2343 PPC KVM_REG_PPC_VPA_DTL 128 2344 PPC KVM_REG_PPC_EPCR 32 2345 PPC KVM_REG_PPC_EPR 32 2346 PPC KVM_REG_PPC_TCR 32 2347 PPC KVM_REG_PPC_TSR 32 2348 PPC KVM_REG_PPC_OR_TSR 32 2349 PPC KVM_REG_PPC_CLEAR_TSR 32 2350 PPC KVM_REG_PPC_MAS0 32 2351 PPC KVM_REG_PPC_MAS1 32 2352 PPC KVM_REG_PPC_MAS2 64 2353 PPC KVM_REG_PPC_MAS7_3 64 2354 PPC KVM_REG_PPC_MAS4 32 2355 PPC KVM_REG_PPC_MAS6 32 2356 PPC KVM_REG_PPC_MMUCFG 32 2357 PPC KVM_REG_PPC_TLB0CFG 32 2358 PPC KVM_REG_PPC_TLB1CFG 32 2359 PPC KVM_REG_PPC_TLB2CFG 32 2360 PPC KVM_REG_PPC_TLB3CFG 32 2361 PPC KVM_REG_PPC_TLB0PS 32 2362 PPC KVM_REG_PPC_TLB1PS 32 2363 PPC KVM_REG_PPC_TLB2PS 32 2364 PPC KVM_REG_PPC_TLB3PS 32 2365 PPC KVM_REG_PPC_EPTCFG 32 2366 PPC KVM_REG_PPC_ICP_STATE 64 2367 PPC KVM_REG_PPC_VP_STATE 128 2368 PPC KVM_REG_PPC_TB_OFFSET 64 2369 PPC KVM_REG_PPC_SPMC1 32 2370 PPC KVM_REG_PPC_SPMC2 32 2371 PPC KVM_REG_PPC_IAMR 64 2372 PPC KVM_REG_PPC_TFHAR 64 2373 PPC KVM_REG_PPC_TFIAR 64 2374 PPC KVM_REG_PPC_TEXASR 64 2375 PPC KVM_REG_PPC_FSCR 64 2376 PPC KVM_REG_PPC_PSPB 32 2377 PPC KVM_REG_PPC_EBBHR 64 2378 PPC KVM_REG_PPC_EBBRR 64 2379 PPC KVM_REG_PPC_BESCR 64 2380 PPC KVM_REG_PPC_TAR 64 2381 PPC KVM_REG_PPC_DPDES 64 2382 PPC KVM_REG_PPC_DAWR 64 2383 PPC KVM_REG_PPC_DAWRX 64 2384 PPC KVM_REG_PPC_CIABR 64 2385 PPC KVM_REG_PPC_IC 64 2386 PPC KVM_REG_PPC_VTB 64 2387 PPC KVM_REG_PPC_CSIGR 64 2388 PPC KVM_REG_PPC_TACR 64 2389 PPC KVM_REG_PPC_TCSCR 64 2390 PPC KVM_REG_PPC_PID 64 2391 PPC KVM_REG_PPC_ACOP 64 2392 PPC KVM_REG_PPC_VRSAVE 32 2393 PPC KVM_REG_PPC_LPCR 32 2394 PPC KVM_REG_PPC_LPCR_64 64 2395 PPC KVM_REG_PPC_PPR 64 2396 PPC KVM_REG_PPC_ARCH_COMPAT 32 2397 PPC KVM_REG_PPC_DABRX 32 2398 PPC KVM_REG_PPC_WORT 64 2399 PPC KVM_REG_PPC_SPRG9 64 2400 PPC KVM_REG_PPC_DBSR 32 2401 PPC KVM_REG_PPC_TIDR 64 2402 PPC KVM_REG_PPC_PSSCR 64 2403 PPC KVM_REG_PPC_DEC_EXPIRY 64 2404 PPC KVM_REG_PPC_PTCR 64 2405 PPC KVM_REG_PPC_DAWR1 64 2406 PPC KVM_REG_PPC_DAWRX1 64 2407 PPC KVM_REG_PPC_TM_GPR0 64 2408 ... 2409 PPC KVM_REG_PPC_TM_GPR31 64 2410 PPC KVM_REG_PPC_TM_VSR0 128 2411 ... 2412 PPC KVM_REG_PPC_TM_VSR63 128 2413 PPC KVM_REG_PPC_TM_CR 64 2414 PPC KVM_REG_PPC_TM_LR 64 2415 PPC KVM_REG_PPC_TM_CTR 64 2416 PPC KVM_REG_PPC_TM_FPSCR 64 2417 PPC KVM_REG_PPC_TM_AMR 64 2418 PPC KVM_REG_PPC_TM_PPR 64 2419 PPC KVM_REG_PPC_TM_VRSAVE 64 2420 PPC KVM_REG_PPC_TM_VSCR 32 2421 PPC KVM_REG_PPC_TM_DSCR 64 2422 PPC KVM_REG_PPC_TM_TAR 64 2423 PPC KVM_REG_PPC_TM_XER 64 2424 2425 MIPS KVM_REG_MIPS_R0 64 2426 ... 2427 MIPS KVM_REG_MIPS_R31 64 2428 MIPS KVM_REG_MIPS_HI 64 2429 MIPS KVM_REG_MIPS_LO 64 2430 MIPS KVM_REG_MIPS_PC 64 2431 MIPS KVM_REG_MIPS_CP0_INDEX 32 2432 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2433 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2434 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2435 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2436 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2437 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2438 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2439 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2440 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2441 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2442 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2443 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2444 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2445 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2446 MIPS KVM_REG_MIPS_CP0_WIRED 32 2447 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2448 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2449 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2450 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2451 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2452 MIPS KVM_REG_MIPS_CP0_COUNT 32 2453 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2454 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2455 MIPS KVM_REG_MIPS_CP0_STATUS 32 2456 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2457 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2458 MIPS KVM_REG_MIPS_CP0_EPC 64 2459 MIPS KVM_REG_MIPS_CP0_PRID 32 2460 MIPS KVM_REG_MIPS_CP0_EBASE 64 2461 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2462 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2463 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2464 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2465 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2466 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2467 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2468 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2469 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2470 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2471 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2472 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2473 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2474 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2475 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2476 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2477 MIPS KVM_REG_MIPS_COUNT_CTL 64 2478 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2479 MIPS KVM_REG_MIPS_COUNT_HZ 64 2480 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2481 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2482 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2483 MIPS KVM_REG_MIPS_FCR_IR 32 2484 MIPS KVM_REG_MIPS_FCR_CSR 32 2485 MIPS KVM_REG_MIPS_MSA_IR 32 2486 MIPS KVM_REG_MIPS_MSA_CSR 32 2487 ======= =============================== ============ 2488 2489ARM registers are mapped using the lower 32 bits. The upper 16 of that 2490is the register group type, or coprocessor number: 2491 2492ARM core registers have the following id bit patterns:: 2493 2494 0x4020 0000 0010 <index into the kvm_regs struct:16> 2495 2496ARM 32-bit CP15 registers have the following id bit patterns:: 2497 2498 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 2499 2500ARM 64-bit CP15 registers have the following id bit patterns:: 2501 2502 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 2503 2504ARM CCSIDR registers are demultiplexed by CSSELR value:: 2505 2506 0x4020 0000 0011 00 <csselr:8> 2507 2508ARM 32-bit VFP control registers have the following id bit patterns:: 2509 2510 0x4020 0000 0012 1 <regno:12> 2511 2512ARM 64-bit FP registers have the following id bit patterns:: 2513 2514 0x4030 0000 0012 0 <regno:12> 2515 2516ARM firmware pseudo-registers have the following bit pattern:: 2517 2518 0x4030 0000 0014 <regno:16> 2519 2520 2521arm64 registers are mapped using the lower 32 bits. The upper 16 of 2522that is the register group type, or coprocessor number: 2523 2524arm64 core/FP-SIMD registers have the following id bit patterns. Note 2525that the size of the access is variable, as the kvm_regs structure 2526contains elements ranging from 32 to 128 bits. The index is a 32bit 2527value in the kvm_regs structure seen as a 32bit array:: 2528 2529 0x60x0 0000 0010 <index into the kvm_regs struct:16> 2530 2531Specifically: 2532 2533======================= ========= ===== ======================================= 2534 Encoding Register Bits kvm_regs member 2535======================= ========= ===== ======================================= 2536 0x6030 0000 0010 0000 X0 64 regs.regs[0] 2537 0x6030 0000 0010 0002 X1 64 regs.regs[1] 2538 ... 2539 0x6030 0000 0010 003c X30 64 regs.regs[30] 2540 0x6030 0000 0010 003e SP 64 regs.sp 2541 0x6030 0000 0010 0040 PC 64 regs.pc 2542 0x6030 0000 0010 0042 PSTATE 64 regs.pstate 2543 0x6030 0000 0010 0044 SP_EL1 64 sp_el1 2544 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1 2545 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC) 2546 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT] 2547 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND] 2548 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ] 2549 0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ] 2550 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_ 2551 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_ 2552 ... 2553 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_ 2554 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr 2555 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr 2556======================= ========= ===== ======================================= 2557 2558.. [1] These encodings are not accepted for SVE-enabled vcpus. See 2559 KVM_ARM_VCPU_INIT. 2560 2561 The equivalent register content can be accessed via bits [127:0] of 2562 the corresponding SVE Zn registers instead for vcpus that have SVE 2563 enabled (see below). 2564 2565arm64 CCSIDR registers are demultiplexed by CSSELR value:: 2566 2567 0x6020 0000 0011 00 <csselr:8> 2568 2569arm64 system registers have the following id bit patterns:: 2570 2571 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 2572 2573.. warning:: 2574 2575 Two system register IDs do not follow the specified pattern. These 2576 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to 2577 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These 2578 two had their values accidentally swapped, which means TIMER_CVAL is 2579 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is 2580 derived from the register encoding for CNTV_CVAL_EL0. As this is 2581 API, it must remain this way. 2582 2583arm64 firmware pseudo-registers have the following bit pattern:: 2584 2585 0x6030 0000 0014 <regno:16> 2586 2587arm64 SVE registers have the following bit patterns:: 2588 2589 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice] 2590 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice] 2591 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice] 2592 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register 2593 2594Access to register IDs where 2048 * slice >= 128 * max_vq will fail with 2595ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit 2596quadwords: see [2]_ below. 2597 2598These registers are only accessible on vcpus for which SVE is enabled. 2599See KVM_ARM_VCPU_INIT for details. 2600 2601In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not 2602accessible until the vcpu's SVE configuration has been finalized 2603using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT 2604and KVM_ARM_VCPU_FINALIZE for more information about this procedure. 2605 2606KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector 2607lengths supported by the vcpu to be discovered and configured by 2608userspace. When transferred to or from user memory via KVM_GET_ONE_REG 2609or KVM_SET_ONE_REG, the value of this register is of type 2610__u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as 2611follows:: 2612 2613 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS]; 2614 2615 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2616 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >> 2617 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1)) 2618 /* Vector length vq * 16 bytes supported */ 2619 else 2620 /* Vector length vq * 16 bytes not supported */ 2621 2622.. [2] The maximum value vq for which the above condition is true is 2623 max_vq. This is the maximum vector length available to the guest on 2624 this vcpu, and determines which register slices are visible through 2625 this ioctl interface. 2626 2627(See Documentation/arm64/sve.rst for an explanation of the "vq" 2628nomenclature.) 2629 2630KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT. 2631KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that 2632the host supports. 2633 2634Userspace may subsequently modify it if desired until the vcpu's SVE 2635configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). 2636 2637Apart from simply removing all vector lengths from the host set that 2638exceed some value, support for arbitrarily chosen sets of vector lengths 2639is hardware-dependent and may not be available. Attempting to configure 2640an invalid set of vector lengths via KVM_SET_ONE_REG will fail with 2641EINVAL. 2642 2643After the vcpu's SVE configuration is finalized, further attempts to 2644write this register will fail with EPERM. 2645 2646arm64 bitmap feature firmware pseudo-registers have the following bit pattern:: 2647 2648 0x6030 0000 0016 <regno:16> 2649 2650The bitmap feature firmware registers exposes the hypercall services that 2651are available for userspace to configure. The set bits corresponds to the 2652services that are available for the guests to access. By default, KVM 2653sets all the supported bits during VM initialization. The userspace can 2654discover the available services via KVM_GET_ONE_REG, and write back the 2655bitmap corresponding to the features that it wishes guests to see via 2656KVM_SET_ONE_REG. 2657 2658Note: These registers are immutable once any of the vCPUs of the VM has 2659run at least once. A KVM_SET_ONE_REG in such a scenario will return 2660a -EBUSY to userspace. 2661 2662(See Documentation/virt/kvm/arm/hypercalls.rst for more details.) 2663 2664 2665MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 2666the register group type: 2667 2668MIPS core registers (see above) have the following id bit patterns:: 2669 2670 0x7030 0000 0000 <reg:16> 2671 2672MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 2673patterns depending on whether they're 32-bit or 64-bit registers:: 2674 2675 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 2676 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 2677 2678Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 2679versions of the EntryLo registers regardless of the word size of the host 2680hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 2681with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 2682the PFNX field starting at bit 30. 2683 2684MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 2685patterns:: 2686 2687 0x7030 0000 0001 01 <reg:8> 2688 2689MIPS KVM control registers (see above) have the following id bit patterns:: 2690 2691 0x7030 0000 0002 <reg:16> 2692 2693MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2694id bit patterns depending on the size of the register being accessed. They are 2695always accessed according to the current guest FPU mode (Status.FR and 2696Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2697if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2698registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2699overlap the FPU registers:: 2700 2701 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2702 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2703 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2704 2705MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2706following id bit patterns:: 2707 2708 0x7020 0000 0003 01 <0:3> <reg:5> 2709 2710MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2711following id bit patterns:: 2712 2713 0x7020 0000 0003 02 <0:3> <reg:5> 2714 2715RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of 2716that is the register group type. 2717 2718RISC-V config registers are meant for configuring a Guest VCPU and it has 2719the following id bit patterns:: 2720 2721 0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host) 2722 0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host) 2723 2724Following are the RISC-V config registers: 2725 2726======================= ========= ============================================= 2727 Encoding Register Description 2728======================= ========= ============================================= 2729 0x80x0 0000 0100 0000 isa ISA feature bitmap of Guest VCPU 2730======================= ========= ============================================= 2731 2732The isa config register can be read anytime but can only be written before 2733a Guest VCPU runs. It will have ISA feature bits matching underlying host 2734set by default. 2735 2736RISC-V core registers represent the general excution state of a Guest VCPU 2737and it has the following id bit patterns:: 2738 2739 0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host) 2740 0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host) 2741 2742Following are the RISC-V core registers: 2743 2744======================= ========= ============================================= 2745 Encoding Register Description 2746======================= ========= ============================================= 2747 0x80x0 0000 0200 0000 regs.pc Program counter 2748 0x80x0 0000 0200 0001 regs.ra Return address 2749 0x80x0 0000 0200 0002 regs.sp Stack pointer 2750 0x80x0 0000 0200 0003 regs.gp Global pointer 2751 0x80x0 0000 0200 0004 regs.tp Task pointer 2752 0x80x0 0000 0200 0005 regs.t0 Caller saved register 0 2753 0x80x0 0000 0200 0006 regs.t1 Caller saved register 1 2754 0x80x0 0000 0200 0007 regs.t2 Caller saved register 2 2755 0x80x0 0000 0200 0008 regs.s0 Callee saved register 0 2756 0x80x0 0000 0200 0009 regs.s1 Callee saved register 1 2757 0x80x0 0000 0200 000a regs.a0 Function argument (or return value) 0 2758 0x80x0 0000 0200 000b regs.a1 Function argument (or return value) 1 2759 0x80x0 0000 0200 000c regs.a2 Function argument 2 2760 0x80x0 0000 0200 000d regs.a3 Function argument 3 2761 0x80x0 0000 0200 000e regs.a4 Function argument 4 2762 0x80x0 0000 0200 000f regs.a5 Function argument 5 2763 0x80x0 0000 0200 0010 regs.a6 Function argument 6 2764 0x80x0 0000 0200 0011 regs.a7 Function argument 7 2765 0x80x0 0000 0200 0012 regs.s2 Callee saved register 2 2766 0x80x0 0000 0200 0013 regs.s3 Callee saved register 3 2767 0x80x0 0000 0200 0014 regs.s4 Callee saved register 4 2768 0x80x0 0000 0200 0015 regs.s5 Callee saved register 5 2769 0x80x0 0000 0200 0016 regs.s6 Callee saved register 6 2770 0x80x0 0000 0200 0017 regs.s7 Callee saved register 7 2771 0x80x0 0000 0200 0018 regs.s8 Callee saved register 8 2772 0x80x0 0000 0200 0019 regs.s9 Callee saved register 9 2773 0x80x0 0000 0200 001a regs.s10 Callee saved register 10 2774 0x80x0 0000 0200 001b regs.s11 Callee saved register 11 2775 0x80x0 0000 0200 001c regs.t3 Caller saved register 3 2776 0x80x0 0000 0200 001d regs.t4 Caller saved register 4 2777 0x80x0 0000 0200 001e regs.t5 Caller saved register 5 2778 0x80x0 0000 0200 001f regs.t6 Caller saved register 6 2779 0x80x0 0000 0200 0020 mode Privilege mode (1 = S-mode or 0 = U-mode) 2780======================= ========= ============================================= 2781 2782RISC-V csr registers represent the supervisor mode control/status registers 2783of a Guest VCPU and it has the following id bit patterns:: 2784 2785 0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host) 2786 0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host) 2787 2788Following are the RISC-V csr registers: 2789 2790======================= ========= ============================================= 2791 Encoding Register Description 2792======================= ========= ============================================= 2793 0x80x0 0000 0300 0000 sstatus Supervisor status 2794 0x80x0 0000 0300 0001 sie Supervisor interrupt enable 2795 0x80x0 0000 0300 0002 stvec Supervisor trap vector base 2796 0x80x0 0000 0300 0003 sscratch Supervisor scratch register 2797 0x80x0 0000 0300 0004 sepc Supervisor exception program counter 2798 0x80x0 0000 0300 0005 scause Supervisor trap cause 2799 0x80x0 0000 0300 0006 stval Supervisor bad address or instruction 2800 0x80x0 0000 0300 0007 sip Supervisor interrupt pending 2801 0x80x0 0000 0300 0008 satp Supervisor address translation and protection 2802======================= ========= ============================================= 2803 2804RISC-V timer registers represent the timer state of a Guest VCPU and it has 2805the following id bit patterns:: 2806 2807 0x8030 0000 04 <index into the kvm_riscv_timer struct:24> 2808 2809Following are the RISC-V timer registers: 2810 2811======================= ========= ============================================= 2812 Encoding Register Description 2813======================= ========= ============================================= 2814 0x8030 0000 0400 0000 frequency Time base frequency (read-only) 2815 0x8030 0000 0400 0001 time Time value visible to Guest 2816 0x8030 0000 0400 0002 compare Time compare programmed by Guest 2817 0x8030 0000 0400 0003 state Time compare state (1 = ON or 0 = OFF) 2818======================= ========= ============================================= 2819 2820RISC-V F-extension registers represent the single precision floating point 2821state of a Guest VCPU and it has the following id bit patterns:: 2822 2823 0x8020 0000 05 <index into the __riscv_f_ext_state struct:24> 2824 2825Following are the RISC-V F-extension registers: 2826 2827======================= ========= ============================================= 2828 Encoding Register Description 2829======================= ========= ============================================= 2830 0x8020 0000 0500 0000 f[0] Floating point register 0 2831 ... 2832 0x8020 0000 0500 001f f[31] Floating point register 31 2833 0x8020 0000 0500 0020 fcsr Floating point control and status register 2834======================= ========= ============================================= 2835 2836RISC-V D-extension registers represent the double precision floating point 2837state of a Guest VCPU and it has the following id bit patterns:: 2838 2839 0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr) 2840 0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr) 2841 2842Following are the RISC-V D-extension registers: 2843 2844======================= ========= ============================================= 2845 Encoding Register Description 2846======================= ========= ============================================= 2847 0x8030 0000 0600 0000 f[0] Floating point register 0 2848 ... 2849 0x8030 0000 0600 001f f[31] Floating point register 31 2850 0x8020 0000 0600 0020 fcsr Floating point control and status register 2851======================= ========= ============================================= 2852 2853 28544.69 KVM_GET_ONE_REG 2855-------------------- 2856 2857:Capability: KVM_CAP_ONE_REG 2858:Architectures: all 2859:Type: vcpu ioctl 2860:Parameters: struct kvm_one_reg (in and out) 2861:Returns: 0 on success, negative value on failure 2862 2863Errors include: 2864 2865 ======== ============================================================ 2866 ENOENT no such register 2867 EINVAL invalid register ID, or no such register or used with VMs in 2868 protected virtualization mode on s390 2869 EPERM (arm64) register access not allowed before vcpu finalization 2870 ======== ============================================================ 2871 2872(These error codes are indicative only: do not rely on a specific error 2873code being returned in a specific situation.) 2874 2875This ioctl allows to receive the value of a single register implemented 2876in a vcpu. The register to read is indicated by the "id" field of the 2877kvm_one_reg struct passed in. On success, the register value can be found 2878at the memory location pointed to by "addr". 2879 2880The list of registers accessible using this interface is identical to the 2881list in 4.68. 2882 2883 28844.70 KVM_KVMCLOCK_CTRL 2885---------------------- 2886 2887:Capability: KVM_CAP_KVMCLOCK_CTRL 2888:Architectures: Any that implement pvclocks (currently x86 only) 2889:Type: vcpu ioctl 2890:Parameters: None 2891:Returns: 0 on success, -1 on error 2892 2893This ioctl sets a flag accessible to the guest indicating that the specified 2894vCPU has been paused by the host userspace. 2895 2896The host will set a flag in the pvclock structure that is checked from the 2897soft lockup watchdog. The flag is part of the pvclock structure that is 2898shared between guest and host, specifically the second bit of the flags 2899field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2900the host and read/cleared exclusively by the guest. The guest operation of 2901checking and clearing the flag must be an atomic operation so 2902load-link/store-conditional, or equivalent must be used. There are two cases 2903where the guest will clear the flag: when the soft lockup watchdog timer resets 2904itself or when a soft lockup is detected. This ioctl can be called any time 2905after pausing the vcpu, but before it is resumed. 2906 2907 29084.71 KVM_SIGNAL_MSI 2909------------------- 2910 2911:Capability: KVM_CAP_SIGNAL_MSI 2912:Architectures: x86 arm64 2913:Type: vm ioctl 2914:Parameters: struct kvm_msi (in) 2915:Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2916 2917Directly inject a MSI message. Only valid with in-kernel irqchip that handles 2918MSI messages. 2919 2920:: 2921 2922 struct kvm_msi { 2923 __u32 address_lo; 2924 __u32 address_hi; 2925 __u32 data; 2926 __u32 flags; 2927 __u32 devid; 2928 __u8 pad[12]; 2929 }; 2930 2931flags: 2932 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 2933 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 2934 the device ID. If this capability is not available, userspace 2935 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 2936 2937If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 2938for the device that wrote the MSI message. For PCI, this is usually a 2939BFD identifier in the lower 16 bits. 2940 2941On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 2942feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 2943address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 2944address_hi must be zero. 2945 2946 29474.71 KVM_CREATE_PIT2 2948-------------------- 2949 2950:Capability: KVM_CAP_PIT2 2951:Architectures: x86 2952:Type: vm ioctl 2953:Parameters: struct kvm_pit_config (in) 2954:Returns: 0 on success, -1 on error 2955 2956Creates an in-kernel device model for the i8254 PIT. This call is only valid 2957after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 2958parameters have to be passed:: 2959 2960 struct kvm_pit_config { 2961 __u32 flags; 2962 __u32 pad[15]; 2963 }; 2964 2965Valid flags are:: 2966 2967 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 2968 2969PIT timer interrupts may use a per-VM kernel thread for injection. If it 2970exists, this thread will have a name of the following pattern:: 2971 2972 kvm-pit/<owner-process-pid> 2973 2974When running a guest with elevated priorities, the scheduling parameters of 2975this thread may have to be adjusted accordingly. 2976 2977This IOCTL replaces the obsolete KVM_CREATE_PIT. 2978 2979 29804.72 KVM_GET_PIT2 2981----------------- 2982 2983:Capability: KVM_CAP_PIT_STATE2 2984:Architectures: x86 2985:Type: vm ioctl 2986:Parameters: struct kvm_pit_state2 (out) 2987:Returns: 0 on success, -1 on error 2988 2989Retrieves the state of the in-kernel PIT model. Only valid after 2990KVM_CREATE_PIT2. The state is returned in the following structure:: 2991 2992 struct kvm_pit_state2 { 2993 struct kvm_pit_channel_state channels[3]; 2994 __u32 flags; 2995 __u32 reserved[9]; 2996 }; 2997 2998Valid flags are:: 2999 3000 /* disable PIT in HPET legacy mode */ 3001 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 3002 3003This IOCTL replaces the obsolete KVM_GET_PIT. 3004 3005 30064.73 KVM_SET_PIT2 3007----------------- 3008 3009:Capability: KVM_CAP_PIT_STATE2 3010:Architectures: x86 3011:Type: vm ioctl 3012:Parameters: struct kvm_pit_state2 (in) 3013:Returns: 0 on success, -1 on error 3014 3015Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 3016See KVM_GET_PIT2 for details on struct kvm_pit_state2. 3017 3018This IOCTL replaces the obsolete KVM_SET_PIT. 3019 3020 30214.74 KVM_PPC_GET_SMMU_INFO 3022-------------------------- 3023 3024:Capability: KVM_CAP_PPC_GET_SMMU_INFO 3025:Architectures: powerpc 3026:Type: vm ioctl 3027:Parameters: None 3028:Returns: 0 on success, -1 on error 3029 3030This populates and returns a structure describing the features of 3031the "Server" class MMU emulation supported by KVM. 3032This can in turn be used by userspace to generate the appropriate 3033device-tree properties for the guest operating system. 3034 3035The structure contains some global information, followed by an 3036array of supported segment page sizes:: 3037 3038 struct kvm_ppc_smmu_info { 3039 __u64 flags; 3040 __u32 slb_size; 3041 __u32 pad; 3042 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3043 }; 3044 3045The supported flags are: 3046 3047 - KVM_PPC_PAGE_SIZES_REAL: 3048 When that flag is set, guest page sizes must "fit" the backing 3049 store page sizes. When not set, any page size in the list can 3050 be used regardless of how they are backed by userspace. 3051 3052 - KVM_PPC_1T_SEGMENTS 3053 The emulated MMU supports 1T segments in addition to the 3054 standard 256M ones. 3055 3056 - KVM_PPC_NO_HASH 3057 This flag indicates that HPT guests are not supported by KVM, 3058 thus all guests must use radix MMU mode. 3059 3060The "slb_size" field indicates how many SLB entries are supported 3061 3062The "sps" array contains 8 entries indicating the supported base 3063page sizes for a segment in increasing order. Each entry is defined 3064as follow:: 3065 3066 struct kvm_ppc_one_seg_page_size { 3067 __u32 page_shift; /* Base page shift of segment (or 0) */ 3068 __u32 slb_enc; /* SLB encoding for BookS */ 3069 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 3070 }; 3071 3072An entry with a "page_shift" of 0 is unused. Because the array is 3073organized in increasing order, a lookup can stop when encoutering 3074such an entry. 3075 3076The "slb_enc" field provides the encoding to use in the SLB for the 3077page size. The bits are in positions such as the value can directly 3078be OR'ed into the "vsid" argument of the slbmte instruction. 3079 3080The "enc" array is a list which for each of those segment base page 3081size provides the list of supported actual page sizes (which can be 3082only larger or equal to the base page size), along with the 3083corresponding encoding in the hash PTE. Similarly, the array is 30848 entries sorted by increasing sizes and an entry with a "0" shift 3085is an empty entry and a terminator:: 3086 3087 struct kvm_ppc_one_page_size { 3088 __u32 page_shift; /* Page shift (or 0) */ 3089 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 3090 }; 3091 3092The "pte_enc" field provides a value that can OR'ed into the hash 3093PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 3094into the hash PTE second double word). 3095 30964.75 KVM_IRQFD 3097-------------- 3098 3099:Capability: KVM_CAP_IRQFD 3100:Architectures: x86 s390 arm64 3101:Type: vm ioctl 3102:Parameters: struct kvm_irqfd (in) 3103:Returns: 0 on success, -1 on error 3104 3105Allows setting an eventfd to directly trigger a guest interrupt. 3106kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 3107kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 3108an event is triggered on the eventfd, an interrupt is injected into 3109the guest using the specified gsi pin. The irqfd is removed using 3110the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 3111and kvm_irqfd.gsi. 3112 3113With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 3114mechanism allowing emulation of level-triggered, irqfd-based 3115interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 3116additional eventfd in the kvm_irqfd.resamplefd field. When operating 3117in resample mode, posting of an interrupt through kvm_irq.fd asserts 3118the specified gsi in the irqchip. When the irqchip is resampled, such 3119as from an EOI, the gsi is de-asserted and the user is notified via 3120kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 3121the interrupt if the device making use of it still requires service. 3122Note that closing the resamplefd is not sufficient to disable the 3123irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 3124and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 3125 3126On arm64, gsi routing being supported, the following can happen: 3127 3128- in case no routing entry is associated to this gsi, injection fails 3129- in case the gsi is associated to an irqchip routing entry, 3130 irqchip.pin + 32 corresponds to the injected SPI ID. 3131- in case the gsi is associated to an MSI routing entry, the MSI 3132 message and device ID are translated into an LPI (support restricted 3133 to GICv3 ITS in-kernel emulation). 3134 31354.76 KVM_PPC_ALLOCATE_HTAB 3136-------------------------- 3137 3138:Capability: KVM_CAP_PPC_ALLOC_HTAB 3139:Architectures: powerpc 3140:Type: vm ioctl 3141:Parameters: Pointer to u32 containing hash table order (in/out) 3142:Returns: 0 on success, -1 on error 3143 3144This requests the host kernel to allocate an MMU hash table for a 3145guest using the PAPR paravirtualization interface. This only does 3146anything if the kernel is configured to use the Book 3S HV style of 3147virtualization. Otherwise the capability doesn't exist and the ioctl 3148returns an ENOTTY error. The rest of this description assumes Book 3S 3149HV. 3150 3151There must be no vcpus running when this ioctl is called; if there 3152are, it will do nothing and return an EBUSY error. 3153 3154The parameter is a pointer to a 32-bit unsigned integer variable 3155containing the order (log base 2) of the desired size of the hash 3156table, which must be between 18 and 46. On successful return from the 3157ioctl, the value will not be changed by the kernel. 3158 3159If no hash table has been allocated when any vcpu is asked to run 3160(with the KVM_RUN ioctl), the host kernel will allocate a 3161default-sized hash table (16 MB). 3162 3163If this ioctl is called when a hash table has already been allocated, 3164with a different order from the existing hash table, the existing hash 3165table will be freed and a new one allocated. If this is ioctl is 3166called when a hash table has already been allocated of the same order 3167as specified, the kernel will clear out the existing hash table (zero 3168all HPTEs). In either case, if the guest is using the virtualized 3169real-mode area (VRMA) facility, the kernel will re-create the VMRA 3170HPTEs on the next KVM_RUN of any vcpu. 3171 31724.77 KVM_S390_INTERRUPT 3173----------------------- 3174 3175:Capability: basic 3176:Architectures: s390 3177:Type: vm ioctl, vcpu ioctl 3178:Parameters: struct kvm_s390_interrupt (in) 3179:Returns: 0 on success, -1 on error 3180 3181Allows to inject an interrupt to the guest. Interrupts can be floating 3182(vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 3183 3184Interrupt parameters are passed via kvm_s390_interrupt:: 3185 3186 struct kvm_s390_interrupt { 3187 __u32 type; 3188 __u32 parm; 3189 __u64 parm64; 3190 }; 3191 3192type can be one of the following: 3193 3194KVM_S390_SIGP_STOP (vcpu) 3195 - sigp stop; optional flags in parm 3196KVM_S390_PROGRAM_INT (vcpu) 3197 - program check; code in parm 3198KVM_S390_SIGP_SET_PREFIX (vcpu) 3199 - sigp set prefix; prefix address in parm 3200KVM_S390_RESTART (vcpu) 3201 - restart 3202KVM_S390_INT_CLOCK_COMP (vcpu) 3203 - clock comparator interrupt 3204KVM_S390_INT_CPU_TIMER (vcpu) 3205 - CPU timer interrupt 3206KVM_S390_INT_VIRTIO (vm) 3207 - virtio external interrupt; external interrupt 3208 parameters in parm and parm64 3209KVM_S390_INT_SERVICE (vm) 3210 - sclp external interrupt; sclp parameter in parm 3211KVM_S390_INT_EMERGENCY (vcpu) 3212 - sigp emergency; source cpu in parm 3213KVM_S390_INT_EXTERNAL_CALL (vcpu) 3214 - sigp external call; source cpu in parm 3215KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 3216 - compound value to indicate an 3217 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 3218 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 3219 interruption subclass) 3220KVM_S390_MCHK (vm, vcpu) 3221 - machine check interrupt; cr 14 bits in parm, machine check interrupt 3222 code in parm64 (note that machine checks needing further payload are not 3223 supported by this ioctl) 3224 3225This is an asynchronous vcpu ioctl and can be invoked from any thread. 3226 32274.78 KVM_PPC_GET_HTAB_FD 3228------------------------ 3229 3230:Capability: KVM_CAP_PPC_HTAB_FD 3231:Architectures: powerpc 3232:Type: vm ioctl 3233:Parameters: Pointer to struct kvm_get_htab_fd (in) 3234:Returns: file descriptor number (>= 0) on success, -1 on error 3235 3236This returns a file descriptor that can be used either to read out the 3237entries in the guest's hashed page table (HPT), or to write entries to 3238initialize the HPT. The returned fd can only be written to if the 3239KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 3240can only be read if that bit is clear. The argument struct looks like 3241this:: 3242 3243 /* For KVM_PPC_GET_HTAB_FD */ 3244 struct kvm_get_htab_fd { 3245 __u64 flags; 3246 __u64 start_index; 3247 __u64 reserved[2]; 3248 }; 3249 3250 /* Values for kvm_get_htab_fd.flags */ 3251 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 3252 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 3253 3254The 'start_index' field gives the index in the HPT of the entry at 3255which to start reading. It is ignored when writing. 3256 3257Reads on the fd will initially supply information about all 3258"interesting" HPT entries. Interesting entries are those with the 3259bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 3260all entries. When the end of the HPT is reached, the read() will 3261return. If read() is called again on the fd, it will start again from 3262the beginning of the HPT, but will only return HPT entries that have 3263changed since they were last read. 3264 3265Data read or written is structured as a header (8 bytes) followed by a 3266series of valid HPT entries (16 bytes) each. The header indicates how 3267many valid HPT entries there are and how many invalid entries follow 3268the valid entries. The invalid entries are not represented explicitly 3269in the stream. The header format is:: 3270 3271 struct kvm_get_htab_header { 3272 __u32 index; 3273 __u16 n_valid; 3274 __u16 n_invalid; 3275 }; 3276 3277Writes to the fd create HPT entries starting at the index given in the 3278header; first 'n_valid' valid entries with contents from the data 3279written, then 'n_invalid' invalid entries, invalidating any previously 3280valid entries found. 3281 32824.79 KVM_CREATE_DEVICE 3283---------------------- 3284 3285:Capability: KVM_CAP_DEVICE_CTRL 3286:Type: vm ioctl 3287:Parameters: struct kvm_create_device (in/out) 3288:Returns: 0 on success, -1 on error 3289 3290Errors: 3291 3292 ====== ======================================================= 3293 ENODEV The device type is unknown or unsupported 3294 EEXIST Device already created, and this type of device may not 3295 be instantiated multiple times 3296 ====== ======================================================= 3297 3298 Other error conditions may be defined by individual device types or 3299 have their standard meanings. 3300 3301Creates an emulated device in the kernel. The file descriptor returned 3302in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 3303 3304If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 3305device type is supported (not necessarily whether it can be created 3306in the current vm). 3307 3308Individual devices should not define flags. Attributes should be used 3309for specifying any behavior that is not implied by the device type 3310number. 3311 3312:: 3313 3314 struct kvm_create_device { 3315 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 3316 __u32 fd; /* out: device handle */ 3317 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 3318 }; 3319 33204.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3321-------------------------------------------- 3322 3323:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3324 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3325 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device (no set) 3326:Type: device ioctl, vm ioctl, vcpu ioctl 3327:Parameters: struct kvm_device_attr 3328:Returns: 0 on success, -1 on error 3329 3330Errors: 3331 3332 ===== ============================================================= 3333 ENXIO The group or attribute is unknown/unsupported for this device 3334 or hardware support is missing. 3335 EPERM The attribute cannot (currently) be accessed this way 3336 (e.g. read-only attribute, or attribute that only makes 3337 sense when the device is in a different state) 3338 ===== ============================================================= 3339 3340 Other error conditions may be defined by individual device types. 3341 3342Gets/sets a specified piece of device configuration and/or state. The 3343semantics are device-specific. See individual device documentation in 3344the "devices" directory. As with ONE_REG, the size of the data 3345transferred is defined by the particular attribute. 3346 3347:: 3348 3349 struct kvm_device_attr { 3350 __u32 flags; /* no flags currently defined */ 3351 __u32 group; /* device-defined */ 3352 __u64 attr; /* group-defined */ 3353 __u64 addr; /* userspace address of attr data */ 3354 }; 3355 33564.81 KVM_HAS_DEVICE_ATTR 3357------------------------ 3358 3359:Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 3360 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 3361 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device 3362:Type: device ioctl, vm ioctl, vcpu ioctl 3363:Parameters: struct kvm_device_attr 3364:Returns: 0 on success, -1 on error 3365 3366Errors: 3367 3368 ===== ============================================================= 3369 ENXIO The group or attribute is unknown/unsupported for this device 3370 or hardware support is missing. 3371 ===== ============================================================= 3372 3373Tests whether a device supports a particular attribute. A successful 3374return indicates the attribute is implemented. It does not necessarily 3375indicate that the attribute can be read or written in the device's 3376current state. "addr" is ignored. 3377 33784.82 KVM_ARM_VCPU_INIT 3379---------------------- 3380 3381:Capability: basic 3382:Architectures: arm64 3383:Type: vcpu ioctl 3384:Parameters: struct kvm_vcpu_init (in) 3385:Returns: 0 on success; -1 on error 3386 3387Errors: 3388 3389 ====== ================================================================= 3390 EINVAL the target is unknown, or the combination of features is invalid. 3391 ENOENT a features bit specified is unknown. 3392 ====== ================================================================= 3393 3394This tells KVM what type of CPU to present to the guest, and what 3395optional features it should have. This will cause a reset of the cpu 3396registers to their initial values. If this is not called, KVM_RUN will 3397return ENOEXEC for that vcpu. 3398 3399The initial values are defined as: 3400 - Processor state: 3401 * AArch64: EL1h, D, A, I and F bits set. All other bits 3402 are cleared. 3403 * AArch32: SVC, A, I and F bits set. All other bits are 3404 cleared. 3405 - General Purpose registers, including PC and SP: set to 0 3406 - FPSIMD/NEON registers: set to 0 3407 - SVE registers: set to 0 3408 - System registers: Reset to their architecturally defined 3409 values as for a warm reset to EL1 (resp. SVC) 3410 3411Note that because some registers reflect machine topology, all vcpus 3412should be created before this ioctl is invoked. 3413 3414Userspace can call this function multiple times for a given vcpu, including 3415after the vcpu has been run. This will reset the vcpu to its initial 3416state. All calls to this function after the initial call must use the same 3417target and same set of feature flags, otherwise EINVAL will be returned. 3418 3419Possible features: 3420 3421 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 3422 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 3423 and execute guest code when KVM_RUN is called. 3424 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 3425 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 3426 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision 3427 backward compatible with v0.2) for the CPU. 3428 Depends on KVM_CAP_ARM_PSCI_0_2. 3429 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 3430 Depends on KVM_CAP_ARM_PMU_V3. 3431 3432 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication 3433 for arm64 only. 3434 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS. 3435 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3436 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3437 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3438 requested. 3439 3440 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication 3441 for arm64 only. 3442 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC. 3443 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are 3444 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and 3445 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be 3446 requested. 3447 3448 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only). 3449 Depends on KVM_CAP_ARM_SVE. 3450 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3451 3452 * After KVM_ARM_VCPU_INIT: 3453 3454 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the 3455 initial value of this pseudo-register indicates the best set of 3456 vector lengths possible for a vcpu on this host. 3457 3458 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3459 3460 - KVM_RUN and KVM_GET_REG_LIST are not available; 3461 3462 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access 3463 the scalable archietctural SVE registers 3464 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or 3465 KVM_REG_ARM64_SVE_FFR; 3466 3467 - KVM_REG_ARM64_SVE_VLS may optionally be written using 3468 KVM_SET_ONE_REG, to modify the set of vector lengths available 3469 for the vcpu. 3470 3471 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE): 3472 3473 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can 3474 no longer be written using KVM_SET_ONE_REG. 3475 34764.83 KVM_ARM_PREFERRED_TARGET 3477----------------------------- 3478 3479:Capability: basic 3480:Architectures: arm64 3481:Type: vm ioctl 3482:Parameters: struct kvm_vcpu_init (out) 3483:Returns: 0 on success; -1 on error 3484 3485Errors: 3486 3487 ====== ========================================== 3488 ENODEV no preferred target available for the host 3489 ====== ========================================== 3490 3491This queries KVM for preferred CPU target type which can be emulated 3492by KVM on underlying host. 3493 3494The ioctl returns struct kvm_vcpu_init instance containing information 3495about preferred CPU target type and recommended features for it. The 3496kvm_vcpu_init->features bitmap returned will have feature bits set if 3497the preferred target recommends setting these features, but this is 3498not mandatory. 3499 3500The information returned by this ioctl can be used to prepare an instance 3501of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 3502VCPU matching underlying host. 3503 3504 35054.84 KVM_GET_REG_LIST 3506--------------------- 3507 3508:Capability: basic 3509:Architectures: arm64, mips 3510:Type: vcpu ioctl 3511:Parameters: struct kvm_reg_list (in/out) 3512:Returns: 0 on success; -1 on error 3513 3514Errors: 3515 3516 ===== ============================================================== 3517 E2BIG the reg index list is too big to fit in the array specified by 3518 the user (the number required will be written into n). 3519 ===== ============================================================== 3520 3521:: 3522 3523 struct kvm_reg_list { 3524 __u64 n; /* number of registers in reg[] */ 3525 __u64 reg[0]; 3526 }; 3527 3528This ioctl returns the guest registers that are supported for the 3529KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3530 3531 35324.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3533----------------------------------------- 3534 3535:Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3536:Architectures: arm64 3537:Type: vm ioctl 3538:Parameters: struct kvm_arm_device_address (in) 3539:Returns: 0 on success, -1 on error 3540 3541Errors: 3542 3543 ====== ============================================ 3544 ENODEV The device id is unknown 3545 ENXIO Device not supported on current system 3546 EEXIST Address already set 3547 E2BIG Address outside guest physical address space 3548 EBUSY Address overlaps with other device range 3549 ====== ============================================ 3550 3551:: 3552 3553 struct kvm_arm_device_addr { 3554 __u64 id; 3555 __u64 addr; 3556 }; 3557 3558Specify a device address in the guest's physical address space where guests 3559can access emulated or directly exposed devices, which the host kernel needs 3560to know about. The id field is an architecture specific identifier for a 3561specific device. 3562 3563arm64 divides the id field into two parts, a device id and an 3564address type id specific to the individual device:: 3565 3566 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 3567 field: | 0x00000000 | device id | addr type id | 3568 3569arm64 currently only require this when using the in-kernel GIC 3570support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 3571as the device id. When setting the base address for the guest's 3572mapping of the VGIC virtual CPU and distributor interface, the ioctl 3573must be called after calling KVM_CREATE_IRQCHIP, but before calling 3574KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 3575base addresses will return -EEXIST. 3576 3577Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 3578should be used instead. 3579 3580 35814.86 KVM_PPC_RTAS_DEFINE_TOKEN 3582------------------------------ 3583 3584:Capability: KVM_CAP_PPC_RTAS 3585:Architectures: ppc 3586:Type: vm ioctl 3587:Parameters: struct kvm_rtas_token_args 3588:Returns: 0 on success, -1 on error 3589 3590Defines a token value for a RTAS (Run Time Abstraction Services) 3591service in order to allow it to be handled in the kernel. The 3592argument struct gives the name of the service, which must be the name 3593of a service that has a kernel-side implementation. If the token 3594value is non-zero, it will be associated with that service, and 3595subsequent RTAS calls by the guest specifying that token will be 3596handled by the kernel. If the token value is 0, then any token 3597associated with the service will be forgotten, and subsequent RTAS 3598calls by the guest for that service will be passed to userspace to be 3599handled. 3600 36014.87 KVM_SET_GUEST_DEBUG 3602------------------------ 3603 3604:Capability: KVM_CAP_SET_GUEST_DEBUG 3605:Architectures: x86, s390, ppc, arm64 3606:Type: vcpu ioctl 3607:Parameters: struct kvm_guest_debug (in) 3608:Returns: 0 on success; -1 on error 3609 3610:: 3611 3612 struct kvm_guest_debug { 3613 __u32 control; 3614 __u32 pad; 3615 struct kvm_guest_debug_arch arch; 3616 }; 3617 3618Set up the processor specific debug registers and configure vcpu for 3619handling guest debug events. There are two parts to the structure, the 3620first a control bitfield indicates the type of debug events to handle 3621when running. Common control bits are: 3622 3623 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 3624 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 3625 3626The top 16 bits of the control field are architecture specific control 3627flags which can include the following: 3628 3629 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 3630 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390] 3631 - KVM_GUESTDBG_USE_HW: using hardware debug events [arm64] 3632 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 3633 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 3634 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 3635 - KVM_GUESTDBG_BLOCKIRQ: avoid injecting interrupts/NMI/SMI [x86] 3636 3637For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 3638are enabled in memory so we need to ensure breakpoint exceptions are 3639correctly trapped and the KVM run loop exits at the breakpoint and not 3640running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 3641we need to ensure the guest vCPUs architecture specific registers are 3642updated to the correct (supplied) values. 3643 3644The second part of the structure is architecture specific and 3645typically contains a set of debug registers. 3646 3647For arm64 the number of debug registers is implementation defined and 3648can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 3649KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 3650indicating the number of supported registers. 3651 3652For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether 3653the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported. 3654 3655Also when supported, KVM_CAP_SET_GUEST_DEBUG2 capability indicates the 3656supported KVM_GUESTDBG_* bits in the control field. 3657 3658When debug events exit the main run loop with the reason 3659KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 3660structure containing architecture specific debug information. 3661 36624.88 KVM_GET_EMULATED_CPUID 3663--------------------------- 3664 3665:Capability: KVM_CAP_EXT_EMUL_CPUID 3666:Architectures: x86 3667:Type: system ioctl 3668:Parameters: struct kvm_cpuid2 (in/out) 3669:Returns: 0 on success, -1 on error 3670 3671:: 3672 3673 struct kvm_cpuid2 { 3674 __u32 nent; 3675 __u32 flags; 3676 struct kvm_cpuid_entry2 entries[0]; 3677 }; 3678 3679The member 'flags' is used for passing flags from userspace. 3680 3681:: 3682 3683 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 3684 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */ 3685 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */ 3686 3687 struct kvm_cpuid_entry2 { 3688 __u32 function; 3689 __u32 index; 3690 __u32 flags; 3691 __u32 eax; 3692 __u32 ebx; 3693 __u32 ecx; 3694 __u32 edx; 3695 __u32 padding[3]; 3696 }; 3697 3698This ioctl returns x86 cpuid features which are emulated by 3699kvm.Userspace can use the information returned by this ioctl to query 3700which features are emulated by kvm instead of being present natively. 3701 3702Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 3703structure with the 'nent' field indicating the number of entries in 3704the variable-size array 'entries'. If the number of entries is too low 3705to describe the cpu capabilities, an error (E2BIG) is returned. If the 3706number is too high, the 'nent' field is adjusted and an error (ENOMEM) 3707is returned. If the number is just right, the 'nent' field is adjusted 3708to the number of valid entries in the 'entries' array, which is then 3709filled. 3710 3711The entries returned are the set CPUID bits of the respective features 3712which kvm emulates, as returned by the CPUID instruction, with unknown 3713or unsupported feature bits cleared. 3714 3715Features like x2apic, for example, may not be present in the host cpu 3716but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 3717emulated efficiently and thus not included here. 3718 3719The fields in each entry are defined as follows: 3720 3721 function: 3722 the eax value used to obtain the entry 3723 index: 3724 the ecx value used to obtain the entry (for entries that are 3725 affected by ecx) 3726 flags: 3727 an OR of zero or more of the following: 3728 3729 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3730 if the index field is valid 3731 3732 eax, ebx, ecx, edx: 3733 3734 the values returned by the cpuid instruction for 3735 this function/index combination 3736 37374.89 KVM_S390_MEM_OP 3738-------------------- 3739 3740:Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S390_PROTECTED, KVM_CAP_S390_MEM_OP_EXTENSION 3741:Architectures: s390 3742:Type: vm ioctl, vcpu ioctl 3743:Parameters: struct kvm_s390_mem_op (in) 3744:Returns: = 0 on success, 3745 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 3746 > 0 if an exception occurred while walking the page tables 3747 3748Read or write data from/to the VM's memory. 3749The KVM_CAP_S390_MEM_OP_EXTENSION capability specifies what functionality is 3750supported. 3751 3752Parameters are specified via the following structure:: 3753 3754 struct kvm_s390_mem_op { 3755 __u64 gaddr; /* the guest address */ 3756 __u64 flags; /* flags */ 3757 __u32 size; /* amount of bytes */ 3758 __u32 op; /* type of operation */ 3759 __u64 buf; /* buffer in userspace */ 3760 union { 3761 struct { 3762 __u8 ar; /* the access register number */ 3763 __u8 key; /* access key, ignored if flag unset */ 3764 }; 3765 __u32 sida_offset; /* offset into the sida */ 3766 __u8 reserved[32]; /* ignored */ 3767 }; 3768 }; 3769 3770The start address of the memory region has to be specified in the "gaddr" 3771field, and the length of the region in the "size" field (which must not 3772be 0). The maximum value for "size" can be obtained by checking the 3773KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the 3774userspace application where the read data should be written to for 3775a read access, or where the data that should be written is stored for 3776a write access. The "reserved" field is meant for future extensions. 3777Reserved and unused values are ignored. Future extension that add members must 3778introduce new flags. 3779 3780The type of operation is specified in the "op" field. Flags modifying 3781their behavior can be set in the "flags" field. Undefined flag bits must 3782be set to 0. 3783 3784Possible operations are: 3785 * ``KVM_S390_MEMOP_LOGICAL_READ`` 3786 * ``KVM_S390_MEMOP_LOGICAL_WRITE`` 3787 * ``KVM_S390_MEMOP_ABSOLUTE_READ`` 3788 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE`` 3789 * ``KVM_S390_MEMOP_SIDA_READ`` 3790 * ``KVM_S390_MEMOP_SIDA_WRITE`` 3791 3792Logical read/write: 3793^^^^^^^^^^^^^^^^^^^ 3794 3795Access logical memory, i.e. translate the given guest address to an absolute 3796address given the state of the VCPU and use the absolute address as target of 3797the access. "ar" designates the access register number to be used; the valid 3798range is 0..15. 3799Logical accesses are permitted for the VCPU ioctl only. 3800Logical accesses are permitted for non-protected guests only. 3801 3802Supported flags: 3803 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3804 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION`` 3805 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3806 3807The KVM_S390_MEMOP_F_CHECK_ONLY flag can be set to check whether the 3808corresponding memory access would cause an access exception; however, 3809no actual access to the data in memory at the destination is performed. 3810In this case, "buf" is unused and can be NULL. 3811 3812In case an access exception occurred during the access (or would occur 3813in case of KVM_S390_MEMOP_F_CHECK_ONLY), the ioctl returns a positive 3814error number indicating the type of exception. This exception is also 3815raised directly at the corresponding VCPU if the flag 3816KVM_S390_MEMOP_F_INJECT_EXCEPTION is set. 3817On protection exceptions, unless specified otherwise, the injected 3818translation-exception identifier (TEID) indicates suppression. 3819 3820If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key 3821protection is also in effect and may cause exceptions if accesses are 3822prohibited given the access key designated by "key"; the valid range is 0..15. 3823KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION 3824is > 0. 3825Since the accessed memory may span multiple pages and those pages might have 3826different storage keys, it is possible that a protection exception occurs 3827after memory has been modified. In this case, if the exception is injected, 3828the TEID does not indicate suppression. 3829 3830Absolute read/write: 3831^^^^^^^^^^^^^^^^^^^^ 3832 3833Access absolute memory. This operation is intended to be used with the 3834KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to allow accessing memory and performing 3835the checks required for storage key protection as one operation (as opposed to 3836user space getting the storage keys, performing the checks, and accessing 3837memory thereafter, which could lead to a delay between check and access). 3838Absolute accesses are permitted for the VM ioctl if KVM_CAP_S390_MEM_OP_EXTENSION 3839is > 0. 3840Currently absolute accesses are not permitted for VCPU ioctls. 3841Absolute accesses are permitted for non-protected guests only. 3842 3843Supported flags: 3844 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3845 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3846 3847The semantics of the flags are as for logical accesses. 3848 3849SIDA read/write: 3850^^^^^^^^^^^^^^^^ 3851 3852Access the secure instruction data area which contains memory operands necessary 3853for instruction emulation for protected guests. 3854SIDA accesses are available if the KVM_CAP_S390_PROTECTED capability is available. 3855SIDA accesses are permitted for the VCPU ioctl only. 3856SIDA accesses are permitted for protected guests only. 3857 3858No flags are supported. 3859 38604.90 KVM_S390_GET_SKEYS 3861----------------------- 3862 3863:Capability: KVM_CAP_S390_SKEYS 3864:Architectures: s390 3865:Type: vm ioctl 3866:Parameters: struct kvm_s390_skeys 3867:Returns: 0 on success, KVM_S390_GET_SKEYS_NONE if guest is not using storage 3868 keys, negative value on error 3869 3870This ioctl is used to get guest storage key values on the s390 3871architecture. The ioctl takes parameters via the kvm_s390_skeys struct:: 3872 3873 struct kvm_s390_skeys { 3874 __u64 start_gfn; 3875 __u64 count; 3876 __u64 skeydata_addr; 3877 __u32 flags; 3878 __u32 reserved[9]; 3879 }; 3880 3881The start_gfn field is the number of the first guest frame whose storage keys 3882you want to get. 3883 3884The count field is the number of consecutive frames (starting from start_gfn) 3885whose storage keys to get. The count field must be at least 1 and the maximum 3886allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 3887will cause the ioctl to return -EINVAL. 3888 3889The skeydata_addr field is the address to a buffer large enough to hold count 3890bytes. This buffer will be filled with storage key data by the ioctl. 3891 38924.91 KVM_S390_SET_SKEYS 3893----------------------- 3894 3895:Capability: KVM_CAP_S390_SKEYS 3896:Architectures: s390 3897:Type: vm ioctl 3898:Parameters: struct kvm_s390_skeys 3899:Returns: 0 on success, negative value on error 3900 3901This ioctl is used to set guest storage key values on the s390 3902architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 3903See section on KVM_S390_GET_SKEYS for struct definition. 3904 3905The start_gfn field is the number of the first guest frame whose storage keys 3906you want to set. 3907 3908The count field is the number of consecutive frames (starting from start_gfn) 3909whose storage keys to get. The count field must be at least 1 and the maximum 3910allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range 3911will cause the ioctl to return -EINVAL. 3912 3913The skeydata_addr field is the address to a buffer containing count bytes of 3914storage keys. Each byte in the buffer will be set as the storage key for a 3915single frame starting at start_gfn for count frames. 3916 3917Note: If any architecturally invalid key value is found in the given data then 3918the ioctl will return -EINVAL. 3919 39204.92 KVM_S390_IRQ 3921----------------- 3922 3923:Capability: KVM_CAP_S390_INJECT_IRQ 3924:Architectures: s390 3925:Type: vcpu ioctl 3926:Parameters: struct kvm_s390_irq (in) 3927:Returns: 0 on success, -1 on error 3928 3929Errors: 3930 3931 3932 ====== ================================================================= 3933 EINVAL interrupt type is invalid 3934 type is KVM_S390_SIGP_STOP and flag parameter is invalid value, 3935 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 3936 than the maximum of VCPUs 3937 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped, 3938 type is KVM_S390_SIGP_STOP and a stop irq is already pending, 3939 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 3940 is already pending 3941 ====== ================================================================= 3942 3943Allows to inject an interrupt to the guest. 3944 3945Using struct kvm_s390_irq as a parameter allows 3946to inject additional payload which is not 3947possible via KVM_S390_INTERRUPT. 3948 3949Interrupt parameters are passed via kvm_s390_irq:: 3950 3951 struct kvm_s390_irq { 3952 __u64 type; 3953 union { 3954 struct kvm_s390_io_info io; 3955 struct kvm_s390_ext_info ext; 3956 struct kvm_s390_pgm_info pgm; 3957 struct kvm_s390_emerg_info emerg; 3958 struct kvm_s390_extcall_info extcall; 3959 struct kvm_s390_prefix_info prefix; 3960 struct kvm_s390_stop_info stop; 3961 struct kvm_s390_mchk_info mchk; 3962 char reserved[64]; 3963 } u; 3964 }; 3965 3966type can be one of the following: 3967 3968- KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 3969- KVM_S390_PROGRAM_INT - program check; parameters in .pgm 3970- KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 3971- KVM_S390_RESTART - restart; no parameters 3972- KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 3973- KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 3974- KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 3975- KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 3976- KVM_S390_MCHK - machine check interrupt; parameters in .mchk 3977 3978This is an asynchronous vcpu ioctl and can be invoked from any thread. 3979 39804.94 KVM_S390_GET_IRQ_STATE 3981--------------------------- 3982 3983:Capability: KVM_CAP_S390_IRQ_STATE 3984:Architectures: s390 3985:Type: vcpu ioctl 3986:Parameters: struct kvm_s390_irq_state (out) 3987:Returns: >= number of bytes copied into buffer, 3988 -EINVAL if buffer size is 0, 3989 -ENOBUFS if buffer size is too small to fit all pending interrupts, 3990 -EFAULT if the buffer address was invalid 3991 3992This ioctl allows userspace to retrieve the complete state of all currently 3993pending interrupts in a single buffer. Use cases include migration 3994and introspection. The parameter structure contains the address of a 3995userspace buffer and its length:: 3996 3997 struct kvm_s390_irq_state { 3998 __u64 buf; 3999 __u32 flags; /* will stay unused for compatibility reasons */ 4000 __u32 len; 4001 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4002 }; 4003 4004Userspace passes in the above struct and for each pending interrupt a 4005struct kvm_s390_irq is copied to the provided buffer. 4006 4007The structure contains a flags and a reserved field for future extensions. As 4008the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 4009reserved, these fields can not be used in the future without breaking 4010compatibility. 4011 4012If -ENOBUFS is returned the buffer provided was too small and userspace 4013may retry with a bigger buffer. 4014 40154.95 KVM_S390_SET_IRQ_STATE 4016--------------------------- 4017 4018:Capability: KVM_CAP_S390_IRQ_STATE 4019:Architectures: s390 4020:Type: vcpu ioctl 4021:Parameters: struct kvm_s390_irq_state (in) 4022:Returns: 0 on success, 4023 -EFAULT if the buffer address was invalid, 4024 -EINVAL for an invalid buffer length (see below), 4025 -EBUSY if there were already interrupts pending, 4026 errors occurring when actually injecting the 4027 interrupt. See KVM_S390_IRQ. 4028 4029This ioctl allows userspace to set the complete state of all cpu-local 4030interrupts currently pending for the vcpu. It is intended for restoring 4031interrupt state after a migration. The input parameter is a userspace buffer 4032containing a struct kvm_s390_irq_state:: 4033 4034 struct kvm_s390_irq_state { 4035 __u64 buf; 4036 __u32 flags; /* will stay unused for compatibility reasons */ 4037 __u32 len; 4038 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 4039 }; 4040 4041The restrictions for flags and reserved apply as well. 4042(see KVM_S390_GET_IRQ_STATE) 4043 4044The userspace memory referenced by buf contains a struct kvm_s390_irq 4045for each interrupt to be injected into the guest. 4046If one of the interrupts could not be injected for some reason the 4047ioctl aborts. 4048 4049len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 4050and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 4051which is the maximum number of possibly pending cpu-local interrupts. 4052 40534.96 KVM_SMI 4054------------ 4055 4056:Capability: KVM_CAP_X86_SMM 4057:Architectures: x86 4058:Type: vcpu ioctl 4059:Parameters: none 4060:Returns: 0 on success, -1 on error 4061 4062Queues an SMI on the thread's vcpu. 4063 40644.97 KVM_X86_SET_MSR_FILTER 4065---------------------------- 4066 4067:Capability: KVM_X86_SET_MSR_FILTER 4068:Architectures: x86 4069:Type: vm ioctl 4070:Parameters: struct kvm_msr_filter 4071:Returns: 0 on success, < 0 on error 4072 4073:: 4074 4075 struct kvm_msr_filter_range { 4076 #define KVM_MSR_FILTER_READ (1 << 0) 4077 #define KVM_MSR_FILTER_WRITE (1 << 1) 4078 __u32 flags; 4079 __u32 nmsrs; /* number of msrs in bitmap */ 4080 __u32 base; /* MSR index the bitmap starts at */ 4081 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 4082 }; 4083 4084 #define KVM_MSR_FILTER_MAX_RANGES 16 4085 struct kvm_msr_filter { 4086 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 4087 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 4088 __u32 flags; 4089 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 4090 }; 4091 4092flags values for ``struct kvm_msr_filter_range``: 4093 4094``KVM_MSR_FILTER_READ`` 4095 4096 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 4097 indicates that a read should immediately fail, while a 1 indicates that 4098 a read for a particular MSR should be handled regardless of the default 4099 filter action. 4100 4101``KVM_MSR_FILTER_WRITE`` 4102 4103 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 4104 indicates that a write should immediately fail, while a 1 indicates that 4105 a write for a particular MSR should be handled regardless of the default 4106 filter action. 4107 4108``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE`` 4109 4110 Filter both read and write accesses to MSRs using the given bitmap. A 0 4111 in the bitmap indicates that both reads and writes should immediately fail, 4112 while a 1 indicates that reads and writes for a particular MSR are not 4113 filtered by this range. 4114 4115flags values for ``struct kvm_msr_filter``: 4116 4117``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4118 4119 If no filter range matches an MSR index that is getting accessed, KVM will 4120 fall back to allowing access to the MSR. 4121 4122``KVM_MSR_FILTER_DEFAULT_DENY`` 4123 4124 If no filter range matches an MSR index that is getting accessed, KVM will 4125 fall back to rejecting access to the MSR. In this mode, all MSRs that should 4126 be processed by KVM need to explicitly be marked as allowed in the bitmaps. 4127 4128This ioctl allows user space to define up to 16 bitmaps of MSR ranges to 4129specify whether a certain MSR access should be explicitly filtered for or not. 4130 4131If this ioctl has never been invoked, MSR accesses are not guarded and the 4132default KVM in-kernel emulation behavior is fully preserved. 4133 4134Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 4135filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 4136an error. 4137 4138As soon as the filtering is in place, every MSR access is processed through 4139the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff); 4140x2APIC MSRs are always allowed, independent of the ``default_allow`` setting, 4141and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base 4142register. 4143 4144.. warning:: 4145 MSR accesses coming from nested vmentry/vmexit are not filtered. 4146 This includes both writes to individual VMCS fields and reads/writes 4147 through the MSR lists pointed to by the VMCS. 4148 4149If a bit is within one of the defined ranges, read and write accesses are 4150guarded by the bitmap's value for the MSR index if the kind of access 4151is included in the ``struct kvm_msr_filter_range`` flags. If no range 4152cover this particular access, the behavior is determined by the flags 4153field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4154and ``KVM_MSR_FILTER_DEFAULT_DENY``. 4155 4156Each bitmap range specifies a range of MSRs to potentially allow access on. 4157The range goes from MSR index [base .. base+nmsrs]. The flags field 4158indicates whether reads, writes or both reads and writes are filtered 4159by setting a 1 bit in the bitmap for the corresponding MSR index. 4160 4161If an MSR access is not permitted through the filtering, it generates a 4162#GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that 4163allows user space to deflect and potentially handle various MSR accesses 4164into user space. 4165 4166If a vCPU is in running state while this ioctl is invoked, the vCPU may 4167experience inconsistent filtering behavior on MSR accesses. 4168 41694.98 KVM_CREATE_SPAPR_TCE_64 4170---------------------------- 4171 4172:Capability: KVM_CAP_SPAPR_TCE_64 4173:Architectures: powerpc 4174:Type: vm ioctl 4175:Parameters: struct kvm_create_spapr_tce_64 (in) 4176:Returns: file descriptor for manipulating the created TCE table 4177 4178This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 4179windows, described in 4.62 KVM_CREATE_SPAPR_TCE 4180 4181This capability uses extended struct in ioctl interface:: 4182 4183 /* for KVM_CAP_SPAPR_TCE_64 */ 4184 struct kvm_create_spapr_tce_64 { 4185 __u64 liobn; 4186 __u32 page_shift; 4187 __u32 flags; 4188 __u64 offset; /* in pages */ 4189 __u64 size; /* in pages */ 4190 }; 4191 4192The aim of extension is to support an additional bigger DMA window with 4193a variable page size. 4194KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 4195a bus offset of the corresponding DMA window, @size and @offset are numbers 4196of IOMMU pages. 4197 4198@flags are not used at the moment. 4199 4200The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 4201 42024.99 KVM_REINJECT_CONTROL 4203------------------------- 4204 4205:Capability: KVM_CAP_REINJECT_CONTROL 4206:Architectures: x86 4207:Type: vm ioctl 4208:Parameters: struct kvm_reinject_control (in) 4209:Returns: 0 on success, 4210 -EFAULT if struct kvm_reinject_control cannot be read, 4211 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 4212 4213i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 4214where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 4215vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 4216interrupt whenever there isn't a pending interrupt from i8254. 4217!reinject mode injects an interrupt as soon as a tick arrives. 4218 4219:: 4220 4221 struct kvm_reinject_control { 4222 __u8 pit_reinject; 4223 __u8 reserved[31]; 4224 }; 4225 4226pit_reinject = 0 (!reinject mode) is recommended, unless running an old 4227operating system that uses the PIT for timing (e.g. Linux 2.4.x). 4228 42294.100 KVM_PPC_CONFIGURE_V3_MMU 4230------------------------------ 4231 4232:Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 4233:Architectures: ppc 4234:Type: vm ioctl 4235:Parameters: struct kvm_ppc_mmuv3_cfg (in) 4236:Returns: 0 on success, 4237 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 4238 -EINVAL if the configuration is invalid 4239 4240This ioctl controls whether the guest will use radix or HPT (hashed 4241page table) translation, and sets the pointer to the process table for 4242the guest. 4243 4244:: 4245 4246 struct kvm_ppc_mmuv3_cfg { 4247 __u64 flags; 4248 __u64 process_table; 4249 }; 4250 4251There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 4252KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 4253to use radix tree translation, and if clear, to use HPT translation. 4254KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 4255to be able to use the global TLB and SLB invalidation instructions; 4256if clear, the guest may not use these instructions. 4257 4258The process_table field specifies the address and size of the guest 4259process table, which is in the guest's space. This field is formatted 4260as the second doubleword of the partition table entry, as defined in 4261the Power ISA V3.00, Book III section 5.7.6.1. 4262 42634.101 KVM_PPC_GET_RMMU_INFO 4264--------------------------- 4265 4266:Capability: KVM_CAP_PPC_RADIX_MMU 4267:Architectures: ppc 4268:Type: vm ioctl 4269:Parameters: struct kvm_ppc_rmmu_info (out) 4270:Returns: 0 on success, 4271 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 4272 -EINVAL if no useful information can be returned 4273 4274This ioctl returns a structure containing two things: (a) a list 4275containing supported radix tree geometries, and (b) a list that maps 4276page sizes to put in the "AP" (actual page size) field for the tlbie 4277(TLB invalidate entry) instruction. 4278 4279:: 4280 4281 struct kvm_ppc_rmmu_info { 4282 struct kvm_ppc_radix_geom { 4283 __u8 page_shift; 4284 __u8 level_bits[4]; 4285 __u8 pad[3]; 4286 } geometries[8]; 4287 __u32 ap_encodings[8]; 4288 }; 4289 4290The geometries[] field gives up to 8 supported geometries for the 4291radix page table, in terms of the log base 2 of the smallest page 4292size, and the number of bits indexed at each level of the tree, from 4293the PTE level up to the PGD level in that order. Any unused entries 4294will have 0 in the page_shift field. 4295 4296The ap_encodings gives the supported page sizes and their AP field 4297encodings, encoded with the AP value in the top 3 bits and the log 4298base 2 of the page size in the bottom 6 bits. 4299 43004.102 KVM_PPC_RESIZE_HPT_PREPARE 4301-------------------------------- 4302 4303:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4304:Architectures: powerpc 4305:Type: vm ioctl 4306:Parameters: struct kvm_ppc_resize_hpt (in) 4307:Returns: 0 on successful completion, 4308 >0 if a new HPT is being prepared, the value is an estimated 4309 number of milliseconds until preparation is complete, 4310 -EFAULT if struct kvm_reinject_control cannot be read, 4311 -EINVAL if the supplied shift or flags are invalid, 4312 -ENOMEM if unable to allocate the new HPT, 4313 4314Used to implement the PAPR extension for runtime resizing of a guest's 4315Hashed Page Table (HPT). Specifically this starts, stops or monitors 4316the preparation of a new potential HPT for the guest, essentially 4317implementing the H_RESIZE_HPT_PREPARE hypercall. 4318 4319:: 4320 4321 struct kvm_ppc_resize_hpt { 4322 __u64 flags; 4323 __u32 shift; 4324 __u32 pad; 4325 }; 4326 4327If called with shift > 0 when there is no pending HPT for the guest, 4328this begins preparation of a new pending HPT of size 2^(shift) bytes. 4329It then returns a positive integer with the estimated number of 4330milliseconds until preparation is complete. 4331 4332If called when there is a pending HPT whose size does not match that 4333requested in the parameters, discards the existing pending HPT and 4334creates a new one as above. 4335 4336If called when there is a pending HPT of the size requested, will: 4337 4338 * If preparation of the pending HPT is already complete, return 0 4339 * If preparation of the pending HPT has failed, return an error 4340 code, then discard the pending HPT. 4341 * If preparation of the pending HPT is still in progress, return an 4342 estimated number of milliseconds until preparation is complete. 4343 4344If called with shift == 0, discards any currently pending HPT and 4345returns 0 (i.e. cancels any in-progress preparation). 4346 4347flags is reserved for future expansion, currently setting any bits in 4348flags will result in an -EINVAL. 4349 4350Normally this will be called repeatedly with the same parameters until 4351it returns <= 0. The first call will initiate preparation, subsequent 4352ones will monitor preparation until it completes or fails. 4353 43544.103 KVM_PPC_RESIZE_HPT_COMMIT 4355------------------------------- 4356 4357:Capability: KVM_CAP_SPAPR_RESIZE_HPT 4358:Architectures: powerpc 4359:Type: vm ioctl 4360:Parameters: struct kvm_ppc_resize_hpt (in) 4361:Returns: 0 on successful completion, 4362 -EFAULT if struct kvm_reinject_control cannot be read, 4363 -EINVAL if the supplied shift or flags are invalid, 4364 -ENXIO is there is no pending HPT, or the pending HPT doesn't 4365 have the requested size, 4366 -EBUSY if the pending HPT is not fully prepared, 4367 -ENOSPC if there was a hash collision when moving existing 4368 HPT entries to the new HPT, 4369 -EIO on other error conditions 4370 4371Used to implement the PAPR extension for runtime resizing of a guest's 4372Hashed Page Table (HPT). Specifically this requests that the guest be 4373transferred to working with the new HPT, essentially implementing the 4374H_RESIZE_HPT_COMMIT hypercall. 4375 4376:: 4377 4378 struct kvm_ppc_resize_hpt { 4379 __u64 flags; 4380 __u32 shift; 4381 __u32 pad; 4382 }; 4383 4384This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 4385returned 0 with the same parameters. In other cases 4386KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 4387-EBUSY, though others may be possible if the preparation was started, 4388but failed). 4389 4390This will have undefined effects on the guest if it has not already 4391placed itself in a quiescent state where no vcpu will make MMU enabled 4392memory accesses. 4393 4394On succsful completion, the pending HPT will become the guest's active 4395HPT and the previous HPT will be discarded. 4396 4397On failure, the guest will still be operating on its previous HPT. 4398 43994.104 KVM_X86_GET_MCE_CAP_SUPPORTED 4400----------------------------------- 4401 4402:Capability: KVM_CAP_MCE 4403:Architectures: x86 4404:Type: system ioctl 4405:Parameters: u64 mce_cap (out) 4406:Returns: 0 on success, -1 on error 4407 4408Returns supported MCE capabilities. The u64 mce_cap parameter 4409has the same format as the MSR_IA32_MCG_CAP register. Supported 4410capabilities will have the corresponding bits set. 4411 44124.105 KVM_X86_SETUP_MCE 4413----------------------- 4414 4415:Capability: KVM_CAP_MCE 4416:Architectures: x86 4417:Type: vcpu ioctl 4418:Parameters: u64 mcg_cap (in) 4419:Returns: 0 on success, 4420 -EFAULT if u64 mcg_cap cannot be read, 4421 -EINVAL if the requested number of banks is invalid, 4422 -EINVAL if requested MCE capability is not supported. 4423 4424Initializes MCE support for use. The u64 mcg_cap parameter 4425has the same format as the MSR_IA32_MCG_CAP register and 4426specifies which capabilities should be enabled. The maximum 4427supported number of error-reporting banks can be retrieved when 4428checking for KVM_CAP_MCE. The supported capabilities can be 4429retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 4430 44314.106 KVM_X86_SET_MCE 4432--------------------- 4433 4434:Capability: KVM_CAP_MCE 4435:Architectures: x86 4436:Type: vcpu ioctl 4437:Parameters: struct kvm_x86_mce (in) 4438:Returns: 0 on success, 4439 -EFAULT if struct kvm_x86_mce cannot be read, 4440 -EINVAL if the bank number is invalid, 4441 -EINVAL if VAL bit is not set in status field. 4442 4443Inject a machine check error (MCE) into the guest. The input 4444parameter is:: 4445 4446 struct kvm_x86_mce { 4447 __u64 status; 4448 __u64 addr; 4449 __u64 misc; 4450 __u64 mcg_status; 4451 __u8 bank; 4452 __u8 pad1[7]; 4453 __u64 pad2[3]; 4454 }; 4455 4456If the MCE being reported is an uncorrected error, KVM will 4457inject it as an MCE exception into the guest. If the guest 4458MCG_STATUS register reports that an MCE is in progress, KVM 4459causes an KVM_EXIT_SHUTDOWN vmexit. 4460 4461Otherwise, if the MCE is a corrected error, KVM will just 4462store it in the corresponding bank (provided this bank is 4463not holding a previously reported uncorrected error). 4464 44654.107 KVM_S390_GET_CMMA_BITS 4466---------------------------- 4467 4468:Capability: KVM_CAP_S390_CMMA_MIGRATION 4469:Architectures: s390 4470:Type: vm ioctl 4471:Parameters: struct kvm_s390_cmma_log (in, out) 4472:Returns: 0 on success, a negative value on error 4473 4474This ioctl is used to get the values of the CMMA bits on the s390 4475architecture. It is meant to be used in two scenarios: 4476 4477- During live migration to save the CMMA values. Live migration needs 4478 to be enabled via the KVM_REQ_START_MIGRATION VM property. 4479- To non-destructively peek at the CMMA values, with the flag 4480 KVM_S390_CMMA_PEEK set. 4481 4482The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 4483values are written to a buffer whose location is indicated via the "values" 4484member in the kvm_s390_cmma_log struct. The values in the input struct are 4485also updated as needed. 4486 4487Each CMMA value takes up one byte. 4488 4489:: 4490 4491 struct kvm_s390_cmma_log { 4492 __u64 start_gfn; 4493 __u32 count; 4494 __u32 flags; 4495 union { 4496 __u64 remaining; 4497 __u64 mask; 4498 }; 4499 __u64 values; 4500 }; 4501 4502start_gfn is the number of the first guest frame whose CMMA values are 4503to be retrieved, 4504 4505count is the length of the buffer in bytes, 4506 4507values points to the buffer where the result will be written to. 4508 4509If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 4510KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 4511other ioctls. 4512 4513The result is written in the buffer pointed to by the field values, and 4514the values of the input parameter are updated as follows. 4515 4516Depending on the flags, different actions are performed. The only 4517supported flag so far is KVM_S390_CMMA_PEEK. 4518 4519The default behaviour if KVM_S390_CMMA_PEEK is not set is: 4520start_gfn will indicate the first page frame whose CMMA bits were dirty. 4521It is not necessarily the same as the one passed as input, as clean pages 4522are skipped. 4523 4524count will indicate the number of bytes actually written in the buffer. 4525It can (and very often will) be smaller than the input value, since the 4526buffer is only filled until 16 bytes of clean values are found (which 4527are then not copied in the buffer). Since a CMMA migration block needs 4528the base address and the length, for a total of 16 bytes, we will send 4529back some clean data if there is some dirty data afterwards, as long as 4530the size of the clean data does not exceed the size of the header. This 4531allows to minimize the amount of data to be saved or transferred over 4532the network at the expense of more roundtrips to userspace. The next 4533invocation of the ioctl will skip over all the clean values, saving 4534potentially more than just the 16 bytes we found. 4535 4536If KVM_S390_CMMA_PEEK is set: 4537the existing storage attributes are read even when not in migration 4538mode, and no other action is performed; 4539 4540the output start_gfn will be equal to the input start_gfn, 4541 4542the output count will be equal to the input count, except if the end of 4543memory has been reached. 4544 4545In both cases: 4546the field "remaining" will indicate the total number of dirty CMMA values 4547still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 4548not enabled. 4549 4550mask is unused. 4551 4552values points to the userspace buffer where the result will be stored. 4553 4554This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4555complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4556KVM_S390_CMMA_PEEK is not set but migration mode was not enabled, with 4557-EFAULT if the userspace address is invalid or if no page table is 4558present for the addresses (e.g. when using hugepages). 4559 45604.108 KVM_S390_SET_CMMA_BITS 4561---------------------------- 4562 4563:Capability: KVM_CAP_S390_CMMA_MIGRATION 4564:Architectures: s390 4565:Type: vm ioctl 4566:Parameters: struct kvm_s390_cmma_log (in) 4567:Returns: 0 on success, a negative value on error 4568 4569This ioctl is used to set the values of the CMMA bits on the s390 4570architecture. It is meant to be used during live migration to restore 4571the CMMA values, but there are no restrictions on its use. 4572The ioctl takes parameters via the kvm_s390_cmma_values struct. 4573Each CMMA value takes up one byte. 4574 4575:: 4576 4577 struct kvm_s390_cmma_log { 4578 __u64 start_gfn; 4579 __u32 count; 4580 __u32 flags; 4581 union { 4582 __u64 remaining; 4583 __u64 mask; 4584 }; 4585 __u64 values; 4586 }; 4587 4588start_gfn indicates the starting guest frame number, 4589 4590count indicates how many values are to be considered in the buffer, 4591 4592flags is not used and must be 0. 4593 4594mask indicates which PGSTE bits are to be considered. 4595 4596remaining is not used. 4597 4598values points to the buffer in userspace where to store the values. 4599 4600This ioctl can fail with -ENOMEM if not enough memory can be allocated to 4601complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 4602the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 4603if the flags field was not 0, with -EFAULT if the userspace address is 4604invalid, if invalid pages are written to (e.g. after the end of memory) 4605or if no page table is present for the addresses (e.g. when using 4606hugepages). 4607 46084.109 KVM_PPC_GET_CPU_CHAR 4609-------------------------- 4610 4611:Capability: KVM_CAP_PPC_GET_CPU_CHAR 4612:Architectures: powerpc 4613:Type: vm ioctl 4614:Parameters: struct kvm_ppc_cpu_char (out) 4615:Returns: 0 on successful completion, 4616 -EFAULT if struct kvm_ppc_cpu_char cannot be written 4617 4618This ioctl gives userspace information about certain characteristics 4619of the CPU relating to speculative execution of instructions and 4620possible information leakage resulting from speculative execution (see 4621CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 4622returned in struct kvm_ppc_cpu_char, which looks like this:: 4623 4624 struct kvm_ppc_cpu_char { 4625 __u64 character; /* characteristics of the CPU */ 4626 __u64 behaviour; /* recommended software behaviour */ 4627 __u64 character_mask; /* valid bits in character */ 4628 __u64 behaviour_mask; /* valid bits in behaviour */ 4629 }; 4630 4631For extensibility, the character_mask and behaviour_mask fields 4632indicate which bits of character and behaviour have been filled in by 4633the kernel. If the set of defined bits is extended in future then 4634userspace will be able to tell whether it is running on a kernel that 4635knows about the new bits. 4636 4637The character field describes attributes of the CPU which can help 4638with preventing inadvertent information disclosure - specifically, 4639whether there is an instruction to flash-invalidate the L1 data cache 4640(ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 4641to a mode where entries can only be used by the thread that created 4642them, whether the bcctr[l] instruction prevents speculation, and 4643whether a speculation barrier instruction (ori 31,31,0) is provided. 4644 4645The behaviour field describes actions that software should take to 4646prevent inadvertent information disclosure, and thus describes which 4647vulnerabilities the hardware is subject to; specifically whether the 4648L1 data cache should be flushed when returning to user mode from the 4649kernel, and whether a speculation barrier should be placed between an 4650array bounds check and the array access. 4651 4652These fields use the same bit definitions as the new 4653H_GET_CPU_CHARACTERISTICS hypercall. 4654 46554.110 KVM_MEMORY_ENCRYPT_OP 4656--------------------------- 4657 4658:Capability: basic 4659:Architectures: x86 4660:Type: vm 4661:Parameters: an opaque platform specific structure (in/out) 4662:Returns: 0 on success; -1 on error 4663 4664If the platform supports creating encrypted VMs then this ioctl can be used 4665for issuing platform-specific memory encryption commands to manage those 4666encrypted VMs. 4667 4668Currently, this ioctl is used for issuing Secure Encrypted Virtualization 4669(SEV) commands on AMD Processors. The SEV commands are defined in 4670Documentation/virt/kvm/amd-memory-encryption.rst. 4671 46724.111 KVM_MEMORY_ENCRYPT_REG_REGION 4673----------------------------------- 4674 4675:Capability: basic 4676:Architectures: x86 4677:Type: system 4678:Parameters: struct kvm_enc_region (in) 4679:Returns: 0 on success; -1 on error 4680 4681This ioctl can be used to register a guest memory region which may 4682contain encrypted data (e.g. guest RAM, SMRAM etc). 4683 4684It is used in the SEV-enabled guest. When encryption is enabled, a guest 4685memory region may contain encrypted data. The SEV memory encryption 4686engine uses a tweak such that two identical plaintext pages, each at 4687different locations will have differing ciphertexts. So swapping or 4688moving ciphertext of those pages will not result in plaintext being 4689swapped. So relocating (or migrating) physical backing pages for the SEV 4690guest will require some additional steps. 4691 4692Note: The current SEV key management spec does not provide commands to 4693swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 4694memory region registered with the ioctl. 4695 46964.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4697------------------------------------- 4698 4699:Capability: basic 4700:Architectures: x86 4701:Type: system 4702:Parameters: struct kvm_enc_region (in) 4703:Returns: 0 on success; -1 on error 4704 4705This ioctl can be used to unregister the guest memory region registered 4706with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 4707 47084.113 KVM_HYPERV_EVENTFD 4709------------------------ 4710 4711:Capability: KVM_CAP_HYPERV_EVENTFD 4712:Architectures: x86 4713:Type: vm ioctl 4714:Parameters: struct kvm_hyperv_eventfd (in) 4715 4716This ioctl (un)registers an eventfd to receive notifications from the guest on 4717the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without 4718causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number 4719(bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit. 4720 4721:: 4722 4723 struct kvm_hyperv_eventfd { 4724 __u32 conn_id; 4725 __s32 fd; 4726 __u32 flags; 4727 __u32 padding[3]; 4728 }; 4729 4730The conn_id field should fit within 24 bits:: 4731 4732 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff 4733 4734The acceptable values for the flags field are:: 4735 4736 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0) 4737 4738:Returns: 0 on success, 4739 -EINVAL if conn_id or flags is outside the allowed range, 4740 -ENOENT on deassign if the conn_id isn't registered, 4741 -EEXIST on assign if the conn_id is already registered 4742 47434.114 KVM_GET_NESTED_STATE 4744-------------------------- 4745 4746:Capability: KVM_CAP_NESTED_STATE 4747:Architectures: x86 4748:Type: vcpu ioctl 4749:Parameters: struct kvm_nested_state (in/out) 4750:Returns: 0 on success, -1 on error 4751 4752Errors: 4753 4754 ===== ============================================================= 4755 E2BIG the total state size exceeds the value of 'size' specified by 4756 the user; the size required will be written into size. 4757 ===== ============================================================= 4758 4759:: 4760 4761 struct kvm_nested_state { 4762 __u16 flags; 4763 __u16 format; 4764 __u32 size; 4765 4766 union { 4767 struct kvm_vmx_nested_state_hdr vmx; 4768 struct kvm_svm_nested_state_hdr svm; 4769 4770 /* Pad the header to 128 bytes. */ 4771 __u8 pad[120]; 4772 } hdr; 4773 4774 union { 4775 struct kvm_vmx_nested_state_data vmx[0]; 4776 struct kvm_svm_nested_state_data svm[0]; 4777 } data; 4778 }; 4779 4780 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001 4781 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002 4782 #define KVM_STATE_NESTED_EVMCS 0x00000004 4783 4784 #define KVM_STATE_NESTED_FORMAT_VMX 0 4785 #define KVM_STATE_NESTED_FORMAT_SVM 1 4786 4787 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 4788 4789 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001 4790 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002 4791 4792 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001 4793 4794 struct kvm_vmx_nested_state_hdr { 4795 __u64 vmxon_pa; 4796 __u64 vmcs12_pa; 4797 4798 struct { 4799 __u16 flags; 4800 } smm; 4801 4802 __u32 flags; 4803 __u64 preemption_timer_deadline; 4804 }; 4805 4806 struct kvm_vmx_nested_state_data { 4807 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4808 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE]; 4809 }; 4810 4811This ioctl copies the vcpu's nested virtualization state from the kernel to 4812userspace. 4813 4814The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE 4815to the KVM_CHECK_EXTENSION ioctl(). 4816 48174.115 KVM_SET_NESTED_STATE 4818-------------------------- 4819 4820:Capability: KVM_CAP_NESTED_STATE 4821:Architectures: x86 4822:Type: vcpu ioctl 4823:Parameters: struct kvm_nested_state (in) 4824:Returns: 0 on success, -1 on error 4825 4826This copies the vcpu's kvm_nested_state struct from userspace to the kernel. 4827For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE. 4828 48294.116 KVM_(UN)REGISTER_COALESCED_MMIO 4830------------------------------------- 4831 4832:Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio) 4833 KVM_CAP_COALESCED_PIO (for coalesced pio) 4834:Architectures: all 4835:Type: vm ioctl 4836:Parameters: struct kvm_coalesced_mmio_zone 4837:Returns: 0 on success, < 0 on error 4838 4839Coalesced I/O is a performance optimization that defers hardware 4840register write emulation so that userspace exits are avoided. It is 4841typically used to reduce the overhead of emulating frequently accessed 4842hardware registers. 4843 4844When a hardware register is configured for coalesced I/O, write accesses 4845do not exit to userspace and their value is recorded in a ring buffer 4846that is shared between kernel and userspace. 4847 4848Coalesced I/O is used if one or more write accesses to a hardware 4849register can be deferred until a read or a write to another hardware 4850register on the same device. This last access will cause a vmexit and 4851userspace will process accesses from the ring buffer before emulating 4852it. That will avoid exiting to userspace on repeated writes. 4853 4854Coalesced pio is based on coalesced mmio. There is little difference 4855between coalesced mmio and pio except that coalesced pio records accesses 4856to I/O ports. 4857 48584.117 KVM_CLEAR_DIRTY_LOG (vm ioctl) 4859------------------------------------ 4860 4861:Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4862:Architectures: x86, arm64, mips 4863:Type: vm ioctl 4864:Parameters: struct kvm_clear_dirty_log (in) 4865:Returns: 0 on success, -1 on error 4866 4867:: 4868 4869 /* for KVM_CLEAR_DIRTY_LOG */ 4870 struct kvm_clear_dirty_log { 4871 __u32 slot; 4872 __u32 num_pages; 4873 __u64 first_page; 4874 union { 4875 void __user *dirty_bitmap; /* one bit per page */ 4876 __u64 padding; 4877 }; 4878 }; 4879 4880The ioctl clears the dirty status of pages in a memory slot, according to 4881the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap 4882field. Bit 0 of the bitmap corresponds to page "first_page" in the 4883memory slot, and num_pages is the size in bits of the input bitmap. 4884first_page must be a multiple of 64; num_pages must also be a multiple of 488564 unless first_page + num_pages is the size of the memory slot. For each 4886bit that is set in the input bitmap, the corresponding page is marked "clean" 4887in KVM's dirty bitmap, and dirty tracking is re-enabled for that page 4888(for example via write-protection, or by clearing the dirty bit in 4889a page table entry). 4890 4891If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies 4892the address space for which you want to clear the dirty status. See 4893KVM_SET_USER_MEMORY_REGION for details on the usage of slot field. 4894 4895This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 4896is enabled; for more information, see the description of the capability. 4897However, it can always be used as long as KVM_CHECK_EXTENSION confirms 4898that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present. 4899 49004.118 KVM_GET_SUPPORTED_HV_CPUID 4901-------------------------------- 4902 4903:Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system) 4904:Architectures: x86 4905:Type: system ioctl, vcpu ioctl 4906:Parameters: struct kvm_cpuid2 (in/out) 4907:Returns: 0 on success, -1 on error 4908 4909:: 4910 4911 struct kvm_cpuid2 { 4912 __u32 nent; 4913 __u32 padding; 4914 struct kvm_cpuid_entry2 entries[0]; 4915 }; 4916 4917 struct kvm_cpuid_entry2 { 4918 __u32 function; 4919 __u32 index; 4920 __u32 flags; 4921 __u32 eax; 4922 __u32 ebx; 4923 __u32 ecx; 4924 __u32 edx; 4925 __u32 padding[3]; 4926 }; 4927 4928This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in 4929KVM. Userspace can use the information returned by this ioctl to construct 4930cpuid information presented to guests consuming Hyper-V enlightenments (e.g. 4931Windows or Hyper-V guests). 4932 4933CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level 4934Functional Specification (TLFS). These leaves can't be obtained with 4935KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature 4936leaves (0x40000000, 0x40000001). 4937 4938Currently, the following list of CPUID leaves are returned: 4939 4940 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 4941 - HYPERV_CPUID_INTERFACE 4942 - HYPERV_CPUID_VERSION 4943 - HYPERV_CPUID_FEATURES 4944 - HYPERV_CPUID_ENLIGHTMENT_INFO 4945 - HYPERV_CPUID_IMPLEMENT_LIMITS 4946 - HYPERV_CPUID_NESTED_FEATURES 4947 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS 4948 - HYPERV_CPUID_SYNDBG_INTERFACE 4949 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 4950 4951Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure 4952with the 'nent' field indicating the number of entries in the variable-size 4953array 'entries'. If the number of entries is too low to describe all Hyper-V 4954feature leaves, an error (E2BIG) is returned. If the number is more or equal 4955to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the 4956number of valid entries in the 'entries' array, which is then filled. 4957 4958'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved, 4959userspace should not expect to get any particular value there. 4960 4961Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike 4962system ioctl which exposes all supported feature bits unconditionally, vcpu 4963version has the following quirks: 4964 4965- HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED 4966 feature bit are only exposed when Enlightened VMCS was previously enabled 4967 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS). 4968- HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC. 4969 (presumes KVM_CREATE_IRQCHIP has already been called). 4970 49714.119 KVM_ARM_VCPU_FINALIZE 4972--------------------------- 4973 4974:Architectures: arm64 4975:Type: vcpu ioctl 4976:Parameters: int feature (in) 4977:Returns: 0 on success, -1 on error 4978 4979Errors: 4980 4981 ====== ============================================================== 4982 EPERM feature not enabled, needs configuration, or already finalized 4983 EINVAL feature unknown or not present 4984 ====== ============================================================== 4985 4986Recognised values for feature: 4987 4988 ===== =========================================== 4989 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE) 4990 ===== =========================================== 4991 4992Finalizes the configuration of the specified vcpu feature. 4993 4994The vcpu must already have been initialised, enabling the affected feature, by 4995means of a successful KVM_ARM_VCPU_INIT call with the appropriate flag set in 4996features[]. 4997 4998For affected vcpu features, this is a mandatory step that must be performed 4999before the vcpu is fully usable. 5000 5001Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be 5002configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration 5003that should be performaned and how to do it are feature-dependent. 5004 5005Other calls that depend on a particular feature being finalized, such as 5006KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with 5007-EPERM unless the feature has already been finalized by means of a 5008KVM_ARM_VCPU_FINALIZE call. 5009 5010See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization 5011using this ioctl. 5012 50134.120 KVM_SET_PMU_EVENT_FILTER 5014------------------------------ 5015 5016:Capability: KVM_CAP_PMU_EVENT_FILTER 5017:Architectures: x86 5018:Type: vm ioctl 5019:Parameters: struct kvm_pmu_event_filter (in) 5020:Returns: 0 on success, -1 on error 5021 5022:: 5023 5024 struct kvm_pmu_event_filter { 5025 __u32 action; 5026 __u32 nevents; 5027 __u32 fixed_counter_bitmap; 5028 __u32 flags; 5029 __u32 pad[4]; 5030 __u64 events[0]; 5031 }; 5032 5033This ioctl restricts the set of PMU events that the guest can program. 5034The argument holds a list of events which will be allowed or denied. 5035The eventsel+umask of each event the guest attempts to program is compared 5036against the events field to determine whether the guest should have access. 5037The events field only controls general purpose counters; fixed purpose 5038counters are controlled by the fixed_counter_bitmap. 5039 5040No flags are defined yet, the field must be zero. 5041 5042Valid values for 'action':: 5043 5044 #define KVM_PMU_EVENT_ALLOW 0 5045 #define KVM_PMU_EVENT_DENY 1 5046 50474.121 KVM_PPC_SVM_OFF 5048--------------------- 5049 5050:Capability: basic 5051:Architectures: powerpc 5052:Type: vm ioctl 5053:Parameters: none 5054:Returns: 0 on successful completion, 5055 5056Errors: 5057 5058 ====== ================================================================ 5059 EINVAL if ultravisor failed to terminate the secure guest 5060 ENOMEM if hypervisor failed to allocate new radix page tables for guest 5061 ====== ================================================================ 5062 5063This ioctl is used to turn off the secure mode of the guest or transition 5064the guest from secure mode to normal mode. This is invoked when the guest 5065is reset. This has no effect if called for a normal guest. 5066 5067This ioctl issues an ultravisor call to terminate the secure guest, 5068unpins the VPA pages and releases all the device pages that are used to 5069track the secure pages by hypervisor. 5070 50714.122 KVM_S390_NORMAL_RESET 5072--------------------------- 5073 5074:Capability: KVM_CAP_S390_VCPU_RESETS 5075:Architectures: s390 5076:Type: vcpu ioctl 5077:Parameters: none 5078:Returns: 0 5079 5080This ioctl resets VCPU registers and control structures according to 5081the cpu reset definition in the POP (Principles Of Operation). 5082 50834.123 KVM_S390_INITIAL_RESET 5084---------------------------- 5085 5086:Capability: none 5087:Architectures: s390 5088:Type: vcpu ioctl 5089:Parameters: none 5090:Returns: 0 5091 5092This ioctl resets VCPU registers and control structures according to 5093the initial cpu reset definition in the POP. However, the cpu is not 5094put into ESA mode. This reset is a superset of the normal reset. 5095 50964.124 KVM_S390_CLEAR_RESET 5097-------------------------- 5098 5099:Capability: KVM_CAP_S390_VCPU_RESETS 5100:Architectures: s390 5101:Type: vcpu ioctl 5102:Parameters: none 5103:Returns: 0 5104 5105This ioctl resets VCPU registers and control structures according to 5106the clear cpu reset definition in the POP. However, the cpu is not put 5107into ESA mode. This reset is a superset of the initial reset. 5108 5109 51104.125 KVM_S390_PV_COMMAND 5111------------------------- 5112 5113:Capability: KVM_CAP_S390_PROTECTED 5114:Architectures: s390 5115:Type: vm ioctl 5116:Parameters: struct kvm_pv_cmd 5117:Returns: 0 on success, < 0 on error 5118 5119:: 5120 5121 struct kvm_pv_cmd { 5122 __u32 cmd; /* Command to be executed */ 5123 __u16 rc; /* Ultravisor return code */ 5124 __u16 rrc; /* Ultravisor return reason code */ 5125 __u64 data; /* Data or address */ 5126 __u32 flags; /* flags for future extensions. Must be 0 for now */ 5127 __u32 reserved[3]; 5128 }; 5129 5130cmd values: 5131 5132KVM_PV_ENABLE 5133 Allocate memory and register the VM with the Ultravisor, thereby 5134 donating memory to the Ultravisor that will become inaccessible to 5135 KVM. All existing CPUs are converted to protected ones. After this 5136 command has succeeded, any CPU added via hotplug will become 5137 protected during its creation as well. 5138 5139 Errors: 5140 5141 ===== ============================= 5142 EINTR an unmasked signal is pending 5143 ===== ============================= 5144 5145KVM_PV_DISABLE 5146 5147 Deregister the VM from the Ultravisor and reclaim the memory that 5148 had been donated to the Ultravisor, making it usable by the kernel 5149 again. All registered VCPUs are converted back to non-protected 5150 ones. 5151 5152KVM_PV_VM_SET_SEC_PARMS 5153 Pass the image header from VM memory to the Ultravisor in 5154 preparation of image unpacking and verification. 5155 5156KVM_PV_VM_UNPACK 5157 Unpack (protect and decrypt) a page of the encrypted boot image. 5158 5159KVM_PV_VM_VERIFY 5160 Verify the integrity of the unpacked image. Only if this succeeds, 5161 KVM is allowed to start protected VCPUs. 5162 51634.126 KVM_X86_SET_MSR_FILTER 5164---------------------------- 5165 5166:Capability: KVM_CAP_X86_MSR_FILTER 5167:Architectures: x86 5168:Type: vm ioctl 5169:Parameters: struct kvm_msr_filter 5170:Returns: 0 on success, < 0 on error 5171 5172:: 5173 5174 struct kvm_msr_filter_range { 5175 #define KVM_MSR_FILTER_READ (1 << 0) 5176 #define KVM_MSR_FILTER_WRITE (1 << 1) 5177 __u32 flags; 5178 __u32 nmsrs; /* number of msrs in bitmap */ 5179 __u32 base; /* MSR index the bitmap starts at */ 5180 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */ 5181 }; 5182 5183 #define KVM_MSR_FILTER_MAX_RANGES 16 5184 struct kvm_msr_filter { 5185 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0) 5186 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0) 5187 __u32 flags; 5188 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES]; 5189 }; 5190 5191flags values for ``struct kvm_msr_filter_range``: 5192 5193``KVM_MSR_FILTER_READ`` 5194 5195 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap 5196 indicates that a read should immediately fail, while a 1 indicates that 5197 a read for a particular MSR should be handled regardless of the default 5198 filter action. 5199 5200``KVM_MSR_FILTER_WRITE`` 5201 5202 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap 5203 indicates that a write should immediately fail, while a 1 indicates that 5204 a write for a particular MSR should be handled regardless of the default 5205 filter action. 5206 5207``KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE`` 5208 5209 Filter both read and write accesses to MSRs using the given bitmap. A 0 5210 in the bitmap indicates that both reads and writes should immediately fail, 5211 while a 1 indicates that reads and writes for a particular MSR are not 5212 filtered by this range. 5213 5214flags values for ``struct kvm_msr_filter``: 5215 5216``KVM_MSR_FILTER_DEFAULT_ALLOW`` 5217 5218 If no filter range matches an MSR index that is getting accessed, KVM will 5219 fall back to allowing access to the MSR. 5220 5221``KVM_MSR_FILTER_DEFAULT_DENY`` 5222 5223 If no filter range matches an MSR index that is getting accessed, KVM will 5224 fall back to rejecting access to the MSR. In this mode, all MSRs that should 5225 be processed by KVM need to explicitly be marked as allowed in the bitmaps. 5226 5227This ioctl allows user space to define up to 16 bitmaps of MSR ranges to 5228specify whether a certain MSR access should be explicitly filtered for or not. 5229 5230If this ioctl has never been invoked, MSR accesses are not guarded and the 5231default KVM in-kernel emulation behavior is fully preserved. 5232 5233Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR 5234filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes 5235an error. 5236 5237As soon as the filtering is in place, every MSR access is processed through 5238the filtering except for accesses to the x2APIC MSRs (from 0x800 to 0x8ff); 5239x2APIC MSRs are always allowed, independent of the ``default_allow`` setting, 5240and their behavior depends on the ``X2APIC_ENABLE`` bit of the APIC base 5241register. 5242 5243If a bit is within one of the defined ranges, read and write accesses are 5244guarded by the bitmap's value for the MSR index if the kind of access 5245is included in the ``struct kvm_msr_filter_range`` flags. If no range 5246cover this particular access, the behavior is determined by the flags 5247field in the kvm_msr_filter struct: ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 5248and ``KVM_MSR_FILTER_DEFAULT_DENY``. 5249 5250Each bitmap range specifies a range of MSRs to potentially allow access on. 5251The range goes from MSR index [base .. base+nmsrs]. The flags field 5252indicates whether reads, writes or both reads and writes are filtered 5253by setting a 1 bit in the bitmap for the corresponding MSR index. 5254 5255If an MSR access is not permitted through the filtering, it generates a 5256#GP inside the guest. When combined with KVM_CAP_X86_USER_SPACE_MSR, that 5257allows user space to deflect and potentially handle various MSR accesses 5258into user space. 5259 5260Note, invoking this ioctl with a vCPU is running is inherently racy. However, 5261KVM does guarantee that vCPUs will see either the previous filter or the new 5262filter, e.g. MSRs with identical settings in both the old and new filter will 5263have deterministic behavior. 5264 52654.127 KVM_XEN_HVM_SET_ATTR 5266-------------------------- 5267 5268:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5269:Architectures: x86 5270:Type: vm ioctl 5271:Parameters: struct kvm_xen_hvm_attr 5272:Returns: 0 on success, < 0 on error 5273 5274:: 5275 5276 struct kvm_xen_hvm_attr { 5277 __u16 type; 5278 __u16 pad[3]; 5279 union { 5280 __u8 long_mode; 5281 __u8 vector; 5282 struct { 5283 __u64 gfn; 5284 } shared_info; 5285 struct { 5286 __u32 send_port; 5287 __u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */ 5288 __u32 flags; 5289 union { 5290 struct { 5291 __u32 port; 5292 __u32 vcpu; 5293 __u32 priority; 5294 } port; 5295 struct { 5296 __u32 port; /* Zero for eventfd */ 5297 __s32 fd; 5298 } eventfd; 5299 __u32 padding[4]; 5300 } deliver; 5301 } evtchn; 5302 __u32 xen_version; 5303 __u64 pad[8]; 5304 } u; 5305 }; 5306 5307type values: 5308 5309KVM_XEN_ATTR_TYPE_LONG_MODE 5310 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This 5311 determines the layout of the shared info pages exposed to the VM. 5312 5313KVM_XEN_ATTR_TYPE_SHARED_INFO 5314 Sets the guest physical frame number at which the Xen "shared info" 5315 page resides. Note that although Xen places vcpu_info for the first 5316 32 vCPUs in the shared_info page, KVM does not automatically do so 5317 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO be used 5318 explicitly even when the vcpu_info for a given vCPU resides at the 5319 "default" location in the shared_info page. This is because KVM is 5320 not aware of the Xen CPU id which is used as the index into the 5321 vcpu_info[] array, so cannot know the correct default location. 5322 5323 Note that the shared info page may be constantly written to by KVM; 5324 it contains the event channel bitmap used to deliver interrupts to 5325 a Xen guest, amongst other things. It is exempt from dirty tracking 5326 mechanisms — KVM will not explicitly mark the page as dirty each 5327 time an event channel interrupt is delivered to the guest! Thus, 5328 userspace should always assume that the designated GFN is dirty if 5329 any vCPU has been running or any event channel interrupts can be 5330 routed to the guest. 5331 5332KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 5333 Sets the exception vector used to deliver Xen event channel upcalls. 5334 This is the HVM-wide vector injected directly by the hypervisor 5335 (not through the local APIC), typically configured by a guest via 5336 HVM_PARAM_CALLBACK_IRQ. 5337 5338KVM_XEN_ATTR_TYPE_EVTCHN 5339 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5340 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5341 an outbound port number for interception of EVTCHNOP_send requests 5342 from the guest. A given sending port number may be directed back 5343 to a specified vCPU (by APIC ID) / port / priority on the guest, 5344 or to trigger events on an eventfd. The vCPU and priority can be 5345 changed by setting KVM_XEN_EVTCHN_UPDATE in a subsequent call, 5346 but other fields cannot change for a given sending port. A port 5347 mapping is removed by using KVM_XEN_EVTCHN_DEASSIGN in the flags 5348 field. 5349 5350KVM_XEN_ATTR_TYPE_XEN_VERSION 5351 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5352 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures 5353 the 32-bit version code returned to the guest when it invokes the 5354 XENVER_version call; typically (XEN_MAJOR << 16 | XEN_MINOR). PV 5355 Xen guests will often use this to as a dummy hypercall to trigger 5356 event channel delivery, so responding within the kernel without 5357 exiting to userspace is beneficial. 5358 53594.127 KVM_XEN_HVM_GET_ATTR 5360-------------------------- 5361 5362:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5363:Architectures: x86 5364:Type: vm ioctl 5365:Parameters: struct kvm_xen_hvm_attr 5366:Returns: 0 on success, < 0 on error 5367 5368Allows Xen VM attributes to be read. For the structure and types, 5369see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_ATTR_TYPE_EVTCHN 5370attribute cannot be read. 5371 53724.128 KVM_XEN_VCPU_SET_ATTR 5373--------------------------- 5374 5375:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5376:Architectures: x86 5377:Type: vcpu ioctl 5378:Parameters: struct kvm_xen_vcpu_attr 5379:Returns: 0 on success, < 0 on error 5380 5381:: 5382 5383 struct kvm_xen_vcpu_attr { 5384 __u16 type; 5385 __u16 pad[3]; 5386 union { 5387 __u64 gpa; 5388 __u64 pad[4]; 5389 struct { 5390 __u64 state; 5391 __u64 state_entry_time; 5392 __u64 time_running; 5393 __u64 time_runnable; 5394 __u64 time_blocked; 5395 __u64 time_offline; 5396 } runstate; 5397 __u32 vcpu_id; 5398 struct { 5399 __u32 port; 5400 __u32 priority; 5401 __u64 expires_ns; 5402 } timer; 5403 __u8 vector; 5404 } u; 5405 }; 5406 5407type values: 5408 5409KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 5410 Sets the guest physical address of the vcpu_info for a given vCPU. 5411 As with the shared_info page for the VM, the corresponding page may be 5412 dirtied at any time if event channel interrupt delivery is enabled, so 5413 userspace should always assume that the page is dirty without relying 5414 on dirty logging. 5415 5416KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 5417 Sets the guest physical address of an additional pvclock structure 5418 for a given vCPU. This is typically used for guest vsyscall support. 5419 5420KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 5421 Sets the guest physical address of the vcpu_runstate_info for a given 5422 vCPU. This is how a Xen guest tracks CPU state such as steal time. 5423 5424KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 5425 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of 5426 the given vCPU from the .u.runstate.state member of the structure. 5427 KVM automatically accounts running and runnable time but blocked 5428 and offline states are only entered explicitly. 5429 5430KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 5431 Sets all fields of the vCPU runstate data from the .u.runstate member 5432 of the structure, including the current runstate. The state_entry_time 5433 must equal the sum of the other four times. 5434 5435KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 5436 This *adds* the contents of the .u.runstate members of the structure 5437 to the corresponding members of the given vCPU's runstate data, thus 5438 permitting atomic adjustments to the runstate times. The adjustment 5439 to the state_entry_time must equal the sum of the adjustments to the 5440 other four times. The state field must be set to -1, or to a valid 5441 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked 5442 or RUNSTATE_offline) to set the current accounted state as of the 5443 adjusted state_entry_time. 5444 5445KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID 5446 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5447 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the Xen 5448 vCPU ID of the given vCPU, to allow timer-related VCPU operations to 5449 be intercepted by KVM. 5450 5451KVM_XEN_VCPU_ATTR_TYPE_TIMER 5452 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5453 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5454 event channel port/priority for the VIRQ_TIMER of the vCPU, as well 5455 as allowing a pending timer to be saved/restored. 5456 5457KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR 5458 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates 5459 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the 5460 per-vCPU local APIC upcall vector, configured by a Xen guest with 5461 the HVMOP_set_evtchn_upcall_vector hypercall. This is typically 5462 used by Windows guests, and is distinct from the HVM-wide upcall 5463 vector configured with HVM_PARAM_CALLBACK_IRQ. 5464 5465 54664.129 KVM_XEN_VCPU_GET_ATTR 5467--------------------------- 5468 5469:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO 5470:Architectures: x86 5471:Type: vcpu ioctl 5472:Parameters: struct kvm_xen_vcpu_attr 5473:Returns: 0 on success, < 0 on error 5474 5475Allows Xen vCPU attributes to be read. For the structure and types, 5476see KVM_XEN_VCPU_SET_ATTR above. 5477 5478The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used 5479with the KVM_XEN_VCPU_GET_ATTR ioctl. 5480 54814.130 KVM_ARM_MTE_COPY_TAGS 5482--------------------------- 5483 5484:Capability: KVM_CAP_ARM_MTE 5485:Architectures: arm64 5486:Type: vm ioctl 5487:Parameters: struct kvm_arm_copy_mte_tags 5488:Returns: number of bytes copied, < 0 on error (-EINVAL for incorrect 5489 arguments, -EFAULT if memory cannot be accessed). 5490 5491:: 5492 5493 struct kvm_arm_copy_mte_tags { 5494 __u64 guest_ipa; 5495 __u64 length; 5496 void __user *addr; 5497 __u64 flags; 5498 __u64 reserved[2]; 5499 }; 5500 5501Copies Memory Tagging Extension (MTE) tags to/from guest tag memory. The 5502``guest_ipa`` and ``length`` fields must be ``PAGE_SIZE`` aligned. The ``addr`` 5503field must point to a buffer which the tags will be copied to or from. 5504 5505``flags`` specifies the direction of copy, either ``KVM_ARM_TAGS_TO_GUEST`` or 5506``KVM_ARM_TAGS_FROM_GUEST``. 5507 5508The size of the buffer to store the tags is ``(length / 16)`` bytes 5509(granules in MTE are 16 bytes long). Each byte contains a single tag 5510value. This matches the format of ``PTRACE_PEEKMTETAGS`` and 5511``PTRACE_POKEMTETAGS``. 5512 5513If an error occurs before any data is copied then a negative error code is 5514returned. If some tags have been copied before an error occurs then the number 5515of bytes successfully copied is returned. If the call completes successfully 5516then ``length`` is returned. 5517 55184.131 KVM_GET_SREGS2 5519-------------------- 5520 5521:Capability: KVM_CAP_SREGS2 5522:Architectures: x86 5523:Type: vcpu ioctl 5524:Parameters: struct kvm_sregs2 (out) 5525:Returns: 0 on success, -1 on error 5526 5527Reads special registers from the vcpu. 5528This ioctl (when supported) replaces the KVM_GET_SREGS. 5529 5530:: 5531 5532 struct kvm_sregs2 { 5533 /* out (KVM_GET_SREGS2) / in (KVM_SET_SREGS2) */ 5534 struct kvm_segment cs, ds, es, fs, gs, ss; 5535 struct kvm_segment tr, ldt; 5536 struct kvm_dtable gdt, idt; 5537 __u64 cr0, cr2, cr3, cr4, cr8; 5538 __u64 efer; 5539 __u64 apic_base; 5540 __u64 flags; 5541 __u64 pdptrs[4]; 5542 }; 5543 5544flags values for ``kvm_sregs2``: 5545 5546``KVM_SREGS2_FLAGS_PDPTRS_VALID`` 5547 5548 Indicates thats the struct contain valid PDPTR values. 5549 5550 55514.132 KVM_SET_SREGS2 5552-------------------- 5553 5554:Capability: KVM_CAP_SREGS2 5555:Architectures: x86 5556:Type: vcpu ioctl 5557:Parameters: struct kvm_sregs2 (in) 5558:Returns: 0 on success, -1 on error 5559 5560Writes special registers into the vcpu. 5561See KVM_GET_SREGS2 for the data structures. 5562This ioctl (when supported) replaces the KVM_SET_SREGS. 5563 55644.133 KVM_GET_STATS_FD 5565---------------------- 5566 5567:Capability: KVM_CAP_STATS_BINARY_FD 5568:Architectures: all 5569:Type: vm ioctl, vcpu ioctl 5570:Parameters: none 5571:Returns: statistics file descriptor on success, < 0 on error 5572 5573Errors: 5574 5575 ====== ====================================================== 5576 ENOMEM if the fd could not be created due to lack of memory 5577 EMFILE if the number of opened files exceeds the limit 5578 ====== ====================================================== 5579 5580The returned file descriptor can be used to read VM/vCPU statistics data in 5581binary format. The data in the file descriptor consists of four blocks 5582organized as follows: 5583 5584+-------------+ 5585| Header | 5586+-------------+ 5587| id string | 5588+-------------+ 5589| Descriptors | 5590+-------------+ 5591| Stats Data | 5592+-------------+ 5593 5594Apart from the header starting at offset 0, please be aware that it is 5595not guaranteed that the four blocks are adjacent or in the above order; 5596the offsets of the id, descriptors and data blocks are found in the 5597header. However, all four blocks are aligned to 64 bit offsets in the 5598file and they do not overlap. 5599 5600All blocks except the data block are immutable. Userspace can read them 5601only one time after retrieving the file descriptor, and then use ``pread`` or 5602``lseek`` to read the statistics repeatedly. 5603 5604All data is in system endianness. 5605 5606The format of the header is as follows:: 5607 5608 struct kvm_stats_header { 5609 __u32 flags; 5610 __u32 name_size; 5611 __u32 num_desc; 5612 __u32 id_offset; 5613 __u32 desc_offset; 5614 __u32 data_offset; 5615 }; 5616 5617The ``flags`` field is not used at the moment. It is always read as 0. 5618 5619The ``name_size`` field is the size (in byte) of the statistics name string 5620(including trailing '\0') which is contained in the "id string" block and 5621appended at the end of every descriptor. 5622 5623The ``num_desc`` field is the number of descriptors that are included in the 5624descriptor block. (The actual number of values in the data block may be 5625larger, since each descriptor may comprise more than one value). 5626 5627The ``id_offset`` field is the offset of the id string from the start of the 5628file indicated by the file descriptor. It is a multiple of 8. 5629 5630The ``desc_offset`` field is the offset of the Descriptors block from the start 5631of the file indicated by the file descriptor. It is a multiple of 8. 5632 5633The ``data_offset`` field is the offset of the Stats Data block from the start 5634of the file indicated by the file descriptor. It is a multiple of 8. 5635 5636The id string block contains a string which identifies the file descriptor on 5637which KVM_GET_STATS_FD was invoked. The size of the block, including the 5638trailing ``'\0'``, is indicated by the ``name_size`` field in the header. 5639 5640The descriptors block is only needed to be read once for the lifetime of the 5641file descriptor contains a sequence of ``struct kvm_stats_desc``, each followed 5642by a string of size ``name_size``. 5643:: 5644 5645 #define KVM_STATS_TYPE_SHIFT 0 5646 #define KVM_STATS_TYPE_MASK (0xF << KVM_STATS_TYPE_SHIFT) 5647 #define KVM_STATS_TYPE_CUMULATIVE (0x0 << KVM_STATS_TYPE_SHIFT) 5648 #define KVM_STATS_TYPE_INSTANT (0x1 << KVM_STATS_TYPE_SHIFT) 5649 #define KVM_STATS_TYPE_PEAK (0x2 << KVM_STATS_TYPE_SHIFT) 5650 #define KVM_STATS_TYPE_LINEAR_HIST (0x3 << KVM_STATS_TYPE_SHIFT) 5651 #define KVM_STATS_TYPE_LOG_HIST (0x4 << KVM_STATS_TYPE_SHIFT) 5652 #define KVM_STATS_TYPE_MAX KVM_STATS_TYPE_LOG_HIST 5653 5654 #define KVM_STATS_UNIT_SHIFT 4 5655 #define KVM_STATS_UNIT_MASK (0xF << KVM_STATS_UNIT_SHIFT) 5656 #define KVM_STATS_UNIT_NONE (0x0 << KVM_STATS_UNIT_SHIFT) 5657 #define KVM_STATS_UNIT_BYTES (0x1 << KVM_STATS_UNIT_SHIFT) 5658 #define KVM_STATS_UNIT_SECONDS (0x2 << KVM_STATS_UNIT_SHIFT) 5659 #define KVM_STATS_UNIT_CYCLES (0x3 << KVM_STATS_UNIT_SHIFT) 5660 #define KVM_STATS_UNIT_MAX KVM_STATS_UNIT_CYCLES 5661 5662 #define KVM_STATS_BASE_SHIFT 8 5663 #define KVM_STATS_BASE_MASK (0xF << KVM_STATS_BASE_SHIFT) 5664 #define KVM_STATS_BASE_POW10 (0x0 << KVM_STATS_BASE_SHIFT) 5665 #define KVM_STATS_BASE_POW2 (0x1 << KVM_STATS_BASE_SHIFT) 5666 #define KVM_STATS_BASE_MAX KVM_STATS_BASE_POW2 5667 5668 struct kvm_stats_desc { 5669 __u32 flags; 5670 __s16 exponent; 5671 __u16 size; 5672 __u32 offset; 5673 __u32 bucket_size; 5674 char name[]; 5675 }; 5676 5677The ``flags`` field contains the type and unit of the statistics data described 5678by this descriptor. Its endianness is CPU native. 5679The following flags are supported: 5680 5681Bits 0-3 of ``flags`` encode the type: 5682 5683 * ``KVM_STATS_TYPE_CUMULATIVE`` 5684 The statistics reports a cumulative count. The value of data can only be increased. 5685 Most of the counters used in KVM are of this type. 5686 The corresponding ``size`` field for this type is always 1. 5687 All cumulative statistics data are read/write. 5688 * ``KVM_STATS_TYPE_INSTANT`` 5689 The statistics reports an instantaneous value. Its value can be increased or 5690 decreased. This type is usually used as a measurement of some resources, 5691 like the number of dirty pages, the number of large pages, etc. 5692 All instant statistics are read only. 5693 The corresponding ``size`` field for this type is always 1. 5694 * ``KVM_STATS_TYPE_PEAK`` 5695 The statistics data reports a peak value, for example the maximum number 5696 of items in a hash table bucket, the longest time waited and so on. 5697 The value of data can only be increased. 5698 The corresponding ``size`` field for this type is always 1. 5699 * ``KVM_STATS_TYPE_LINEAR_HIST`` 5700 The statistic is reported as a linear histogram. The number of 5701 buckets is specified by the ``size`` field. The size of buckets is specified 5702 by the ``hist_param`` field. The range of the Nth bucket (1 <= N < ``size``) 5703 is [``hist_param``*(N-1), ``hist_param``*N), while the range of the last 5704 bucket is [``hist_param``*(``size``-1), +INF). (+INF means positive infinity 5705 value.) The bucket value indicates how many samples fell in the bucket's range. 5706 * ``KVM_STATS_TYPE_LOG_HIST`` 5707 The statistic is reported as a logarithmic histogram. The number of 5708 buckets is specified by the ``size`` field. The range of the first bucket is 5709 [0, 1), while the range of the last bucket is [pow(2, ``size``-2), +INF). 5710 Otherwise, The Nth bucket (1 < N < ``size``) covers 5711 [pow(2, N-2), pow(2, N-1)). The bucket value indicates how many samples fell 5712 in the bucket's range. 5713 5714Bits 4-7 of ``flags`` encode the unit: 5715 5716 * ``KVM_STATS_UNIT_NONE`` 5717 There is no unit for the value of statistics data. This usually means that 5718 the value is a simple counter of an event. 5719 * ``KVM_STATS_UNIT_BYTES`` 5720 It indicates that the statistics data is used to measure memory size, in the 5721 unit of Byte, KiByte, MiByte, GiByte, etc. The unit of the data is 5722 determined by the ``exponent`` field in the descriptor. 5723 * ``KVM_STATS_UNIT_SECONDS`` 5724 It indicates that the statistics data is used to measure time or latency. 5725 * ``KVM_STATS_UNIT_CYCLES`` 5726 It indicates that the statistics data is used to measure CPU clock cycles. 5727 5728Bits 8-11 of ``flags``, together with ``exponent``, encode the scale of the 5729unit: 5730 5731 * ``KVM_STATS_BASE_POW10`` 5732 The scale is based on power of 10. It is used for measurement of time and 5733 CPU clock cycles. For example, an exponent of -9 can be used with 5734 ``KVM_STATS_UNIT_SECONDS`` to express that the unit is nanoseconds. 5735 * ``KVM_STATS_BASE_POW2`` 5736 The scale is based on power of 2. It is used for measurement of memory size. 5737 For example, an exponent of 20 can be used with ``KVM_STATS_UNIT_BYTES`` to 5738 express that the unit is MiB. 5739 5740The ``size`` field is the number of values of this statistics data. Its 5741value is usually 1 for most of simple statistics. 1 means it contains an 5742unsigned 64bit data. 5743 5744The ``offset`` field is the offset from the start of Data Block to the start of 5745the corresponding statistics data. 5746 5747The ``bucket_size`` field is used as a parameter for histogram statistics data. 5748It is only used by linear histogram statistics data, specifying the size of a 5749bucket. 5750 5751The ``name`` field is the name string of the statistics data. The name string 5752starts at the end of ``struct kvm_stats_desc``. The maximum length including 5753the trailing ``'\0'``, is indicated by ``name_size`` in the header. 5754 5755The Stats Data block contains an array of 64-bit values in the same order 5756as the descriptors in Descriptors block. 5757 57584.134 KVM_GET_XSAVE2 5759-------------------- 5760 5761:Capability: KVM_CAP_XSAVE2 5762:Architectures: x86 5763:Type: vcpu ioctl 5764:Parameters: struct kvm_xsave (out) 5765:Returns: 0 on success, -1 on error 5766 5767 5768:: 5769 5770 struct kvm_xsave { 5771 __u32 region[1024]; 5772 __u32 extra[0]; 5773 }; 5774 5775This ioctl would copy current vcpu's xsave struct to the userspace. It 5776copies as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) 5777when invoked on the vm file descriptor. The size value returned by 5778KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096. 5779Currently, it is only greater than 4096 if a dynamic feature has been 5780enabled with ``arch_prctl()``, but this may change in the future. 5781 5782The offsets of the state save areas in struct kvm_xsave follow the contents 5783of CPUID leaf 0xD on the host. 5784 57854.135 KVM_XEN_HVM_EVTCHN_SEND 5786----------------------------- 5787 5788:Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5789:Architectures: x86 5790:Type: vm ioctl 5791:Parameters: struct kvm_irq_routing_xen_evtchn 5792:Returns: 0 on success, < 0 on error 5793 5794 5795:: 5796 5797 struct kvm_irq_routing_xen_evtchn { 5798 __u32 port; 5799 __u32 vcpu; 5800 __u32 priority; 5801 }; 5802 5803This ioctl injects an event channel interrupt directly to the guest vCPU. 5804 58055. The kvm_run structure 5806======================== 5807 5808Application code obtains a pointer to the kvm_run structure by 5809mmap()ing a vcpu fd. From that point, application code can control 5810execution by changing fields in kvm_run prior to calling the KVM_RUN 5811ioctl, and obtain information about the reason KVM_RUN returned by 5812looking up structure members. 5813 5814:: 5815 5816 struct kvm_run { 5817 /* in */ 5818 __u8 request_interrupt_window; 5819 5820Request that KVM_RUN return when it becomes possible to inject external 5821interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 5822 5823:: 5824 5825 __u8 immediate_exit; 5826 5827This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 5828exits immediately, returning -EINTR. In the common scenario where a 5829signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 5830to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 5831Rather than blocking the signal outside KVM_RUN, userspace can set up 5832a signal handler that sets run->immediate_exit to a non-zero value. 5833 5834This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 5835 5836:: 5837 5838 __u8 padding1[6]; 5839 5840 /* out */ 5841 __u32 exit_reason; 5842 5843When KVM_RUN has returned successfully (return value 0), this informs 5844application code why KVM_RUN has returned. Allowable values for this 5845field are detailed below. 5846 5847:: 5848 5849 __u8 ready_for_interrupt_injection; 5850 5851If request_interrupt_window has been specified, this field indicates 5852an interrupt can be injected now with KVM_INTERRUPT. 5853 5854:: 5855 5856 __u8 if_flag; 5857 5858The value of the current interrupt flag. Only valid if in-kernel 5859local APIC is not used. 5860 5861:: 5862 5863 __u16 flags; 5864 5865More architecture-specific flags detailing state of the VCPU that may 5866affect the device's behavior. Current defined flags:: 5867 5868 /* x86, set if the VCPU is in system management mode */ 5869 #define KVM_RUN_X86_SMM (1 << 0) 5870 /* x86, set if bus lock detected in VM */ 5871 #define KVM_RUN_BUS_LOCK (1 << 1) 5872 /* arm64, set for KVM_EXIT_DEBUG */ 5873 #define KVM_DEBUG_ARCH_HSR_HIGH_VALID (1 << 0) 5874 5875:: 5876 5877 /* in (pre_kvm_run), out (post_kvm_run) */ 5878 __u64 cr8; 5879 5880The value of the cr8 register. Only valid if in-kernel local APIC is 5881not used. Both input and output. 5882 5883:: 5884 5885 __u64 apic_base; 5886 5887The value of the APIC BASE msr. Only valid if in-kernel local 5888APIC is not used. Both input and output. 5889 5890:: 5891 5892 union { 5893 /* KVM_EXIT_UNKNOWN */ 5894 struct { 5895 __u64 hardware_exit_reason; 5896 } hw; 5897 5898If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 5899reasons. Further architecture-specific information is available in 5900hardware_exit_reason. 5901 5902:: 5903 5904 /* KVM_EXIT_FAIL_ENTRY */ 5905 struct { 5906 __u64 hardware_entry_failure_reason; 5907 __u32 cpu; /* if KVM_LAST_CPU */ 5908 } fail_entry; 5909 5910If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 5911to unknown reasons. Further architecture-specific information is 5912available in hardware_entry_failure_reason. 5913 5914:: 5915 5916 /* KVM_EXIT_EXCEPTION */ 5917 struct { 5918 __u32 exception; 5919 __u32 error_code; 5920 } ex; 5921 5922Unused. 5923 5924:: 5925 5926 /* KVM_EXIT_IO */ 5927 struct { 5928 #define KVM_EXIT_IO_IN 0 5929 #define KVM_EXIT_IO_OUT 1 5930 __u8 direction; 5931 __u8 size; /* bytes */ 5932 __u16 port; 5933 __u32 count; 5934 __u64 data_offset; /* relative to kvm_run start */ 5935 } io; 5936 5937If exit_reason is KVM_EXIT_IO, then the vcpu has 5938executed a port I/O instruction which could not be satisfied by kvm. 5939data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 5940where kvm expects application code to place the data for the next 5941KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 5942 5943:: 5944 5945 /* KVM_EXIT_DEBUG */ 5946 struct { 5947 struct kvm_debug_exit_arch arch; 5948 } debug; 5949 5950If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 5951for which architecture specific information is returned. 5952 5953:: 5954 5955 /* KVM_EXIT_MMIO */ 5956 struct { 5957 __u64 phys_addr; 5958 __u8 data[8]; 5959 __u32 len; 5960 __u8 is_write; 5961 } mmio; 5962 5963If exit_reason is KVM_EXIT_MMIO, then the vcpu has 5964executed a memory-mapped I/O instruction which could not be satisfied 5965by kvm. The 'data' member contains the written data if 'is_write' is 5966true, and should be filled by application code otherwise. 5967 5968The 'data' member contains, in its first 'len' bytes, the value as it would 5969appear if the VCPU performed a load or store of the appropriate width directly 5970to the byte array. 5971 5972.. note:: 5973 5974 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN, 5975 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding 5976 operations are complete (and guest state is consistent) only after userspace 5977 has re-entered the kernel with KVM_RUN. The kernel side will first finish 5978 incomplete operations and then check for pending signals. 5979 5980 The pending state of the operation is not preserved in state which is 5981 visible to userspace, thus userspace should ensure that the operation is 5982 completed before performing a live migration. Userspace can re-enter the 5983 guest with an unmasked signal pending or with the immediate_exit field set 5984 to complete pending operations without allowing any further instructions 5985 to be executed. 5986 5987:: 5988 5989 /* KVM_EXIT_HYPERCALL */ 5990 struct { 5991 __u64 nr; 5992 __u64 args[6]; 5993 __u64 ret; 5994 __u32 longmode; 5995 __u32 pad; 5996 } hypercall; 5997 5998Unused. This was once used for 'hypercall to userspace'. To implement 5999such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). 6000 6001.. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 6002 6003:: 6004 6005 /* KVM_EXIT_TPR_ACCESS */ 6006 struct { 6007 __u64 rip; 6008 __u32 is_write; 6009 __u32 pad; 6010 } tpr_access; 6011 6012To be documented (KVM_TPR_ACCESS_REPORTING). 6013 6014:: 6015 6016 /* KVM_EXIT_S390_SIEIC */ 6017 struct { 6018 __u8 icptcode; 6019 __u64 mask; /* psw upper half */ 6020 __u64 addr; /* psw lower half */ 6021 __u16 ipa; 6022 __u32 ipb; 6023 } s390_sieic; 6024 6025s390 specific. 6026 6027:: 6028 6029 /* KVM_EXIT_S390_RESET */ 6030 #define KVM_S390_RESET_POR 1 6031 #define KVM_S390_RESET_CLEAR 2 6032 #define KVM_S390_RESET_SUBSYSTEM 4 6033 #define KVM_S390_RESET_CPU_INIT 8 6034 #define KVM_S390_RESET_IPL 16 6035 __u64 s390_reset_flags; 6036 6037s390 specific. 6038 6039:: 6040 6041 /* KVM_EXIT_S390_UCONTROL */ 6042 struct { 6043 __u64 trans_exc_code; 6044 __u32 pgm_code; 6045 } s390_ucontrol; 6046 6047s390 specific. A page fault has occurred for a user controlled virtual 6048machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be 6049resolved by the kernel. 6050The program code and the translation exception code that were placed 6051in the cpu's lowcore are presented here as defined by the z Architecture 6052Principles of Operation Book in the Chapter for Dynamic Address Translation 6053(DAT) 6054 6055:: 6056 6057 /* KVM_EXIT_DCR */ 6058 struct { 6059 __u32 dcrn; 6060 __u32 data; 6061 __u8 is_write; 6062 } dcr; 6063 6064Deprecated - was used for 440 KVM. 6065 6066:: 6067 6068 /* KVM_EXIT_OSI */ 6069 struct { 6070 __u64 gprs[32]; 6071 } osi; 6072 6073MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 6074hypercalls and exit with this exit struct that contains all the guest gprs. 6075 6076If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 6077Userspace can now handle the hypercall and when it's done modify the gprs as 6078necessary. Upon guest entry all guest GPRs will then be replaced by the values 6079in this struct. 6080 6081:: 6082 6083 /* KVM_EXIT_PAPR_HCALL */ 6084 struct { 6085 __u64 nr; 6086 __u64 ret; 6087 __u64 args[9]; 6088 } papr_hcall; 6089 6090This is used on 64-bit PowerPC when emulating a pSeries partition, 6091e.g. with the 'pseries' machine type in qemu. It occurs when the 6092guest does a hypercall using the 'sc 1' instruction. The 'nr' field 6093contains the hypercall number (from the guest R3), and 'args' contains 6094the arguments (from the guest R4 - R12). Userspace should put the 6095return code in 'ret' and any extra returned values in args[]. 6096The possible hypercalls are defined in the Power Architecture Platform 6097Requirements (PAPR) document available from www.power.org (free 6098developer registration required to access it). 6099 6100:: 6101 6102 /* KVM_EXIT_S390_TSCH */ 6103 struct { 6104 __u16 subchannel_id; 6105 __u16 subchannel_nr; 6106 __u32 io_int_parm; 6107 __u32 io_int_word; 6108 __u32 ipb; 6109 __u8 dequeued; 6110 } s390_tsch; 6111 6112s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 6113and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 6114interrupt for the target subchannel has been dequeued and subchannel_id, 6115subchannel_nr, io_int_parm and io_int_word contain the parameters for that 6116interrupt. ipb is needed for instruction parameter decoding. 6117 6118:: 6119 6120 /* KVM_EXIT_EPR */ 6121 struct { 6122 __u32 epr; 6123 } epr; 6124 6125On FSL BookE PowerPC chips, the interrupt controller has a fast patch 6126interrupt acknowledge path to the core. When the core successfully 6127delivers an interrupt, it automatically populates the EPR register with 6128the interrupt vector number and acknowledges the interrupt inside 6129the interrupt controller. 6130 6131In case the interrupt controller lives in user space, we need to do 6132the interrupt acknowledge cycle through it to fetch the next to be 6133delivered interrupt vector using this exit. 6134 6135It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 6136external interrupt has just been delivered into the guest. User space 6137should put the acknowledged interrupt vector into the 'epr' field. 6138 6139:: 6140 6141 /* KVM_EXIT_SYSTEM_EVENT */ 6142 struct { 6143 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 6144 #define KVM_SYSTEM_EVENT_RESET 2 6145 #define KVM_SYSTEM_EVENT_CRASH 3 6146 #define KVM_SYSTEM_EVENT_WAKEUP 4 6147 #define KVM_SYSTEM_EVENT_SUSPEND 5 6148 #define KVM_SYSTEM_EVENT_SEV_TERM 6 6149 __u32 type; 6150 __u32 ndata; 6151 __u64 data[16]; 6152 } system_event; 6153 6154If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 6155a system-level event using some architecture specific mechanism (hypercall 6156or some special instruction). In case of ARM64, this is triggered using 6157HVC instruction based PSCI call from the vcpu. 6158 6159The 'type' field describes the system-level event type. 6160Valid values for 'type' are: 6161 6162 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 6163 VM. Userspace is not obliged to honour this, and if it does honour 6164 this does not need to destroy the VM synchronously (ie it may call 6165 KVM_RUN again before shutdown finally occurs). 6166 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 6167 As with SHUTDOWN, userspace can choose to ignore the request, or 6168 to schedule the reset to occur in the future and may call KVM_RUN again. 6169 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 6170 has requested a crash condition maintenance. Userspace can choose 6171 to ignore the request, or to gather VM memory core dump and/or 6172 reset/shutdown of the VM. 6173 - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination. 6174 The guest physical address of the guest's GHCB is stored in `data[0]`. 6175 - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and 6176 KVM has recognized a wakeup event. Userspace may honor this event by 6177 marking the exiting vCPU as runnable, or deny it and call KVM_RUN again. 6178 - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested a suspension of 6179 the VM. 6180 6181If KVM_CAP_SYSTEM_EVENT_DATA is present, the 'data' field can contain 6182architecture specific information for the system-level event. Only 6183the first `ndata` items (possibly zero) of the data array are valid. 6184 6185 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2 if 6186 the guest issued a SYSTEM_RESET2 call according to v1.1 of the PSCI 6187 specification. 6188 6189 - for RISC-V, data[0] is set to the value of the second argument of the 6190 ``sbi_system_reset`` call. 6191 6192Previous versions of Linux defined a `flags` member in this struct. The 6193field is now aliased to `data[0]`. Userspace can assume that it is only 6194written if ndata is greater than 0. 6195 6196For arm/arm64: 6197-------------- 6198 6199KVM_SYSTEM_EVENT_SUSPEND exits are enabled with the 6200KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If a guest invokes the PSCI 6201SYSTEM_SUSPEND function, KVM will exit to userspace with this event 6202type. 6203 6204It is the sole responsibility of userspace to implement the PSCI 6205SYSTEM_SUSPEND call according to ARM DEN0022D.b 5.19 "SYSTEM_SUSPEND". 6206KVM does not change the vCPU's state before exiting to userspace, so 6207the call parameters are left in-place in the vCPU registers. 6208 6209Userspace is _required_ to take action for such an exit. It must 6210either: 6211 6212 - Honor the guest request to suspend the VM. Userspace can request 6213 in-kernel emulation of suspension by setting the calling vCPU's 6214 state to KVM_MP_STATE_SUSPENDED. Userspace must configure the vCPU's 6215 state according to the parameters passed to the PSCI function when 6216 the calling vCPU is resumed. See ARM DEN0022D.b 5.19.1 "Intended use" 6217 for details on the function parameters. 6218 6219 - Deny the guest request to suspend the VM. See ARM DEN0022D.b 5.19.2 6220 "Caller responsibilities" for possible return values. 6221 6222:: 6223 6224 /* KVM_EXIT_IOAPIC_EOI */ 6225 struct { 6226 __u8 vector; 6227 } eoi; 6228 6229Indicates that the VCPU's in-kernel local APIC received an EOI for a 6230level-triggered IOAPIC interrupt. This exit only triggers when the 6231IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 6232the userspace IOAPIC should process the EOI and retrigger the interrupt if 6233it is still asserted. Vector is the LAPIC interrupt vector for which the 6234EOI was received. 6235 6236:: 6237 6238 struct kvm_hyperv_exit { 6239 #define KVM_EXIT_HYPERV_SYNIC 1 6240 #define KVM_EXIT_HYPERV_HCALL 2 6241 #define KVM_EXIT_HYPERV_SYNDBG 3 6242 __u32 type; 6243 __u32 pad1; 6244 union { 6245 struct { 6246 __u32 msr; 6247 __u32 pad2; 6248 __u64 control; 6249 __u64 evt_page; 6250 __u64 msg_page; 6251 } synic; 6252 struct { 6253 __u64 input; 6254 __u64 result; 6255 __u64 params[2]; 6256 } hcall; 6257 struct { 6258 __u32 msr; 6259 __u32 pad2; 6260 __u64 control; 6261 __u64 status; 6262 __u64 send_page; 6263 __u64 recv_page; 6264 __u64 pending_page; 6265 } syndbg; 6266 } u; 6267 }; 6268 /* KVM_EXIT_HYPERV */ 6269 struct kvm_hyperv_exit hyperv; 6270 6271Indicates that the VCPU exits into userspace to process some tasks 6272related to Hyper-V emulation. 6273 6274Valid values for 'type' are: 6275 6276 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 6277 6278Hyper-V SynIC state change. Notification is used to remap SynIC 6279event/message pages and to enable/disable SynIC messages/events processing 6280in userspace. 6281 6282 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about 6283 6284Hyper-V Synthetic debugger state change. Notification is used to either update 6285the pending_page location or to send a control command (send the buffer located 6286in send_page or recv a buffer to recv_page). 6287 6288:: 6289 6290 /* KVM_EXIT_ARM_NISV */ 6291 struct { 6292 __u64 esr_iss; 6293 __u64 fault_ipa; 6294 } arm_nisv; 6295 6296Used on arm64 systems. If a guest accesses memory not in a memslot, 6297KVM will typically return to userspace and ask it to do MMIO emulation on its 6298behalf. However, for certain classes of instructions, no instruction decode 6299(direction, length of memory access) is provided, and fetching and decoding 6300the instruction from the VM is overly complicated to live in the kernel. 6301 6302Historically, when this situation occurred, KVM would print a warning and kill 6303the VM. KVM assumed that if the guest accessed non-memslot memory, it was 6304trying to do I/O, which just couldn't be emulated, and the warning message was 6305phrased accordingly. However, what happened more often was that a guest bug 6306caused access outside the guest memory areas which should lead to a more 6307meaningful warning message and an external abort in the guest, if the access 6308did not fall within an I/O window. 6309 6310Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable 6311this capability at VM creation. Once this is done, these types of errors will 6312instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from 6313the ESR_EL2 in the esr_iss field, and the faulting IPA in the fault_ipa field. 6314Userspace can either fix up the access if it's actually an I/O access by 6315decoding the instruction from guest memory (if it's very brave) and continue 6316executing the guest, or it can decide to suspend, dump, or restart the guest. 6317 6318Note that KVM does not skip the faulting instruction as it does for 6319KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state 6320if it decides to decode and emulate the instruction. 6321 6322:: 6323 6324 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */ 6325 struct { 6326 __u8 error; /* user -> kernel */ 6327 __u8 pad[7]; 6328 __u32 reason; /* kernel -> user */ 6329 __u32 index; /* kernel -> user */ 6330 __u64 data; /* kernel <-> user */ 6331 } msr; 6332 6333Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is 6334enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code 6335will instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR 6336exit for writes. 6337 6338The "reason" field specifies why the MSR trap occurred. User space will only 6339receive MSR exit traps when a particular reason was requested during through 6340ENABLE_CAP. Currently valid exit reasons are: 6341 6342 KVM_MSR_EXIT_REASON_UNKNOWN - access to MSR that is unknown to KVM 6343 KVM_MSR_EXIT_REASON_INVAL - access to invalid MSRs or reserved bits 6344 KVM_MSR_EXIT_REASON_FILTER - access blocked by KVM_X86_SET_MSR_FILTER 6345 6346For KVM_EXIT_X86_RDMSR, the "index" field tells user space which MSR the guest 6347wants to read. To respond to this request with a successful read, user space 6348writes the respective data into the "data" field and must continue guest 6349execution to ensure the read data is transferred into guest register state. 6350 6351If the RDMSR request was unsuccessful, user space indicates that with a "1" in 6352the "error" field. This will inject a #GP into the guest when the VCPU is 6353executed again. 6354 6355For KVM_EXIT_X86_WRMSR, the "index" field tells user space which MSR the guest 6356wants to write. Once finished processing the event, user space must continue 6357vCPU execution. If the MSR write was unsuccessful, user space also sets the 6358"error" field to "1". 6359 6360:: 6361 6362 6363 struct kvm_xen_exit { 6364 #define KVM_EXIT_XEN_HCALL 1 6365 __u32 type; 6366 union { 6367 struct { 6368 __u32 longmode; 6369 __u32 cpl; 6370 __u64 input; 6371 __u64 result; 6372 __u64 params[6]; 6373 } hcall; 6374 } u; 6375 }; 6376 /* KVM_EXIT_XEN */ 6377 struct kvm_hyperv_exit xen; 6378 6379Indicates that the VCPU exits into userspace to process some tasks 6380related to Xen emulation. 6381 6382Valid values for 'type' are: 6383 6384 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall. 6385 Userspace is expected to place the hypercall result into the appropriate 6386 field before invoking KVM_RUN again. 6387 6388:: 6389 6390 /* KVM_EXIT_RISCV_SBI */ 6391 struct { 6392 unsigned long extension_id; 6393 unsigned long function_id; 6394 unsigned long args[6]; 6395 unsigned long ret[2]; 6396 } riscv_sbi; 6397 6398If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has 6399done a SBI call which is not handled by KVM RISC-V kernel module. The details 6400of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The 6401'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the 6402'function_id' field represents function ID of given SBI extension. The 'args' 6403array field of 'riscv_sbi' represents parameters for the SBI call and 'ret' 6404array field represents return values. The userspace should update the return 6405values of SBI call before resuming the VCPU. For more details on RISC-V SBI 6406spec refer, https://github.com/riscv/riscv-sbi-doc. 6407 6408:: 6409 6410 /* Fix the size of the union. */ 6411 char padding[256]; 6412 }; 6413 6414 /* 6415 * shared registers between kvm and userspace. 6416 * kvm_valid_regs specifies the register classes set by the host 6417 * kvm_dirty_regs specified the register classes dirtied by userspace 6418 * struct kvm_sync_regs is architecture specific, as well as the 6419 * bits for kvm_valid_regs and kvm_dirty_regs 6420 */ 6421 __u64 kvm_valid_regs; 6422 __u64 kvm_dirty_regs; 6423 union { 6424 struct kvm_sync_regs regs; 6425 char padding[SYNC_REGS_SIZE_BYTES]; 6426 } s; 6427 6428If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 6429certain guest registers without having to call SET/GET_*REGS. Thus we can 6430avoid some system call overhead if userspace has to handle the exit. 6431Userspace can query the validity of the structure by checking 6432kvm_valid_regs for specific bits. These bits are architecture specific 6433and usually define the validity of a groups of registers. (e.g. one bit 6434for general purpose registers) 6435 6436Please note that the kernel is allowed to use the kvm_run structure as the 6437primary storage for certain register types. Therefore, the kernel may use the 6438values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 6439 6440:: 6441 6442 }; 6443 6444 6445 64466. Capabilities that can be enabled on vCPUs 6447============================================ 6448 6449There are certain capabilities that change the behavior of the virtual CPU or 6450the virtual machine when enabled. To enable them, please see section 4.37. 6451Below you can find a list of capabilities and what their effect on the vCPU or 6452the virtual machine is when enabling them. 6453 6454The following information is provided along with the description: 6455 6456 Architectures: 6457 which instruction set architectures provide this ioctl. 6458 x86 includes both i386 and x86_64. 6459 6460 Target: 6461 whether this is a per-vcpu or per-vm capability. 6462 6463 Parameters: 6464 what parameters are accepted by the capability. 6465 6466 Returns: 6467 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 6468 are not detailed, but errors with specific meanings are. 6469 6470 64716.1 KVM_CAP_PPC_OSI 6472------------------- 6473 6474:Architectures: ppc 6475:Target: vcpu 6476:Parameters: none 6477:Returns: 0 on success; -1 on error 6478 6479This capability enables interception of OSI hypercalls that otherwise would 6480be treated as normal system calls to be injected into the guest. OSI hypercalls 6481were invented by Mac-on-Linux to have a standardized communication mechanism 6482between the guest and the host. 6483 6484When this capability is enabled, KVM_EXIT_OSI can occur. 6485 6486 64876.2 KVM_CAP_PPC_PAPR 6488-------------------- 6489 6490:Architectures: ppc 6491:Target: vcpu 6492:Parameters: none 6493:Returns: 0 on success; -1 on error 6494 6495This capability enables interception of PAPR hypercalls. PAPR hypercalls are 6496done using the hypercall instruction "sc 1". 6497 6498It also sets the guest privilege level to "supervisor" mode. Usually the guest 6499runs in "hypervisor" privilege mode with a few missing features. 6500 6501In addition to the above, it changes the semantics of SDR1. In this mode, the 6502HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 6503HTAB invisible to the guest. 6504 6505When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 6506 6507 65086.3 KVM_CAP_SW_TLB 6509------------------ 6510 6511:Architectures: ppc 6512:Target: vcpu 6513:Parameters: args[0] is the address of a struct kvm_config_tlb 6514:Returns: 0 on success; -1 on error 6515 6516:: 6517 6518 struct kvm_config_tlb { 6519 __u64 params; 6520 __u64 array; 6521 __u32 mmu_type; 6522 __u32 array_len; 6523 }; 6524 6525Configures the virtual CPU's TLB array, establishing a shared memory area 6526between userspace and KVM. The "params" and "array" fields are userspace 6527addresses of mmu-type-specific data structures. The "array_len" field is an 6528safety mechanism, and should be set to the size in bytes of the memory that 6529userspace has reserved for the array. It must be at least the size dictated 6530by "mmu_type" and "params". 6531 6532While KVM_RUN is active, the shared region is under control of KVM. Its 6533contents are undefined, and any modification by userspace results in 6534boundedly undefined behavior. 6535 6536On return from KVM_RUN, the shared region will reflect the current state of 6537the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 6538to tell KVM which entries have been changed, prior to calling KVM_RUN again 6539on this vcpu. 6540 6541For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 6542 6543 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 6544 - The "array" field points to an array of type "struct 6545 kvm_book3e_206_tlb_entry". 6546 - The array consists of all entries in the first TLB, followed by all 6547 entries in the second TLB. 6548 - Within a TLB, entries are ordered first by increasing set number. Within a 6549 set, entries are ordered by way (increasing ESEL). 6550 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 6551 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 6552 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 6553 hardware ignores this value for TLB0. 6554 65556.4 KVM_CAP_S390_CSS_SUPPORT 6556---------------------------- 6557 6558:Architectures: s390 6559:Target: vcpu 6560:Parameters: none 6561:Returns: 0 on success; -1 on error 6562 6563This capability enables support for handling of channel I/O instructions. 6564 6565TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 6566handled in-kernel, while the other I/O instructions are passed to userspace. 6567 6568When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 6569SUBCHANNEL intercepts. 6570 6571Note that even though this capability is enabled per-vcpu, the complete 6572virtual machine is affected. 6573 65746.5 KVM_CAP_PPC_EPR 6575------------------- 6576 6577:Architectures: ppc 6578:Target: vcpu 6579:Parameters: args[0] defines whether the proxy facility is active 6580:Returns: 0 on success; -1 on error 6581 6582This capability enables or disables the delivery of interrupts through the 6583external proxy facility. 6584 6585When enabled (args[0] != 0), every time the guest gets an external interrupt 6586delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 6587to receive the topmost interrupt vector. 6588 6589When disabled (args[0] == 0), behavior is as if this facility is unsupported. 6590 6591When this capability is enabled, KVM_EXIT_EPR can occur. 6592 65936.6 KVM_CAP_IRQ_MPIC 6594-------------------- 6595 6596:Architectures: ppc 6597:Parameters: args[0] is the MPIC device fd; 6598 args[1] is the MPIC CPU number for this vcpu 6599 6600This capability connects the vcpu to an in-kernel MPIC device. 6601 66026.7 KVM_CAP_IRQ_XICS 6603-------------------- 6604 6605:Architectures: ppc 6606:Target: vcpu 6607:Parameters: args[0] is the XICS device fd; 6608 args[1] is the XICS CPU number (server ID) for this vcpu 6609 6610This capability connects the vcpu to an in-kernel XICS device. 6611 66126.8 KVM_CAP_S390_IRQCHIP 6613------------------------ 6614 6615:Architectures: s390 6616:Target: vm 6617:Parameters: none 6618 6619This capability enables the in-kernel irqchip for s390. Please refer to 6620"4.24 KVM_CREATE_IRQCHIP" for details. 6621 66226.9 KVM_CAP_MIPS_FPU 6623-------------------- 6624 6625:Architectures: mips 6626:Target: vcpu 6627:Parameters: args[0] is reserved for future use (should be 0). 6628 6629This capability allows the use of the host Floating Point Unit by the guest. It 6630allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 6631done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be 6632accessed (depending on the current guest FPU register mode), and the Status.FR, 6633Config5.FRE bits are accessible via the KVM API and also from the guest, 6634depending on them being supported by the FPU. 6635 66366.10 KVM_CAP_MIPS_MSA 6637--------------------- 6638 6639:Architectures: mips 6640:Target: vcpu 6641:Parameters: args[0] is reserved for future use (should be 0). 6642 6643This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 6644It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 6645Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*`` 6646registers can be accessed, and the Config5.MSAEn bit is accessible via the 6647KVM API and also from the guest. 6648 66496.74 KVM_CAP_SYNC_REGS 6650---------------------- 6651 6652:Architectures: s390, x86 6653:Target: s390: always enabled, x86: vcpu 6654:Parameters: none 6655:Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register 6656 sets are supported 6657 (bitfields defined in arch/x86/include/uapi/asm/kvm.h). 6658 6659As described above in the kvm_sync_regs struct info in section 5 (kvm_run): 6660KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers 6661without having to call SET/GET_*REGS". This reduces overhead by eliminating 6662repeated ioctl calls for setting and/or getting register values. This is 6663particularly important when userspace is making synchronous guest state 6664modifications, e.g. when emulating and/or intercepting instructions in 6665userspace. 6666 6667For s390 specifics, please refer to the source code. 6668 6669For x86: 6670 6671- the register sets to be copied out to kvm_run are selectable 6672 by userspace (rather that all sets being copied out for every exit). 6673- vcpu_events are available in addition to regs and sregs. 6674 6675For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to 6676function as an input bit-array field set by userspace to indicate the 6677specific register sets to be copied out on the next exit. 6678 6679To indicate when userspace has modified values that should be copied into 6680the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set. 6681This is done using the same bitflags as for the 'kvm_valid_regs' field. 6682If the dirty bit is not set, then the register set values will not be copied 6683into the vCPU even if they've been modified. 6684 6685Unused bitfields in the bitarrays must be set to zero. 6686 6687:: 6688 6689 struct kvm_sync_regs { 6690 struct kvm_regs regs; 6691 struct kvm_sregs sregs; 6692 struct kvm_vcpu_events events; 6693 }; 6694 66956.75 KVM_CAP_PPC_IRQ_XIVE 6696------------------------- 6697 6698:Architectures: ppc 6699:Target: vcpu 6700:Parameters: args[0] is the XIVE device fd; 6701 args[1] is the XIVE CPU number (server ID) for this vcpu 6702 6703This capability connects the vcpu to an in-kernel XIVE device. 6704 67057. Capabilities that can be enabled on VMs 6706========================================== 6707 6708There are certain capabilities that change the behavior of the virtual 6709machine when enabled. To enable them, please see section 4.37. Below 6710you can find a list of capabilities and what their effect on the VM 6711is when enabling them. 6712 6713The following information is provided along with the description: 6714 6715 Architectures: 6716 which instruction set architectures provide this ioctl. 6717 x86 includes both i386 and x86_64. 6718 6719 Parameters: 6720 what parameters are accepted by the capability. 6721 6722 Returns: 6723 the return value. General error numbers (EBADF, ENOMEM, EINVAL) 6724 are not detailed, but errors with specific meanings are. 6725 6726 67277.1 KVM_CAP_PPC_ENABLE_HCALL 6728---------------------------- 6729 6730:Architectures: ppc 6731:Parameters: args[0] is the sPAPR hcall number; 6732 args[1] is 0 to disable, 1 to enable in-kernel handling 6733 6734This capability controls whether individual sPAPR hypercalls (hcalls) 6735get handled by the kernel or not. Enabling or disabling in-kernel 6736handling of an hcall is effective across the VM. On creation, an 6737initial set of hcalls are enabled for in-kernel handling, which 6738consists of those hcalls for which in-kernel handlers were implemented 6739before this capability was implemented. If disabled, the kernel will 6740not to attempt to handle the hcall, but will always exit to userspace 6741to handle it. Note that it may not make sense to enable some and 6742disable others of a group of related hcalls, but KVM does not prevent 6743userspace from doing that. 6744 6745If the hcall number specified is not one that has an in-kernel 6746implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 6747error. 6748 67497.2 KVM_CAP_S390_USER_SIGP 6750-------------------------- 6751 6752:Architectures: s390 6753:Parameters: none 6754 6755This capability controls which SIGP orders will be handled completely in user 6756space. With this capability enabled, all fast orders will be handled completely 6757in the kernel: 6758 6759- SENSE 6760- SENSE RUNNING 6761- EXTERNAL CALL 6762- EMERGENCY SIGNAL 6763- CONDITIONAL EMERGENCY SIGNAL 6764 6765All other orders will be handled completely in user space. 6766 6767Only privileged operation exceptions will be checked for in the kernel (or even 6768in the hardware prior to interception). If this capability is not enabled, the 6769old way of handling SIGP orders is used (partially in kernel and user space). 6770 67717.3 KVM_CAP_S390_VECTOR_REGISTERS 6772--------------------------------- 6773 6774:Architectures: s390 6775:Parameters: none 6776:Returns: 0 on success, negative value on error 6777 6778Allows use of the vector registers introduced with z13 processor, and 6779provides for the synchronization between host and user space. Will 6780return -EINVAL if the machine does not support vectors. 6781 67827.4 KVM_CAP_S390_USER_STSI 6783-------------------------- 6784 6785:Architectures: s390 6786:Parameters: none 6787 6788This capability allows post-handlers for the STSI instruction. After 6789initial handling in the kernel, KVM exits to user space with 6790KVM_EXIT_S390_STSI to allow user space to insert further data. 6791 6792Before exiting to userspace, kvm handlers should fill in s390_stsi field of 6793vcpu->run:: 6794 6795 struct { 6796 __u64 addr; 6797 __u8 ar; 6798 __u8 reserved; 6799 __u8 fc; 6800 __u8 sel1; 6801 __u16 sel2; 6802 } s390_stsi; 6803 6804 @addr - guest address of STSI SYSIB 6805 @fc - function code 6806 @sel1 - selector 1 6807 @sel2 - selector 2 6808 @ar - access register number 6809 6810KVM handlers should exit to userspace with rc = -EREMOTE. 6811 68127.5 KVM_CAP_SPLIT_IRQCHIP 6813------------------------- 6814 6815:Architectures: x86 6816:Parameters: args[0] - number of routes reserved for userspace IOAPICs 6817:Returns: 0 on success, -1 on error 6818 6819Create a local apic for each processor in the kernel. This can be used 6820instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 6821IOAPIC and PIC (and also the PIT, even though this has to be enabled 6822separately). 6823 6824This capability also enables in kernel routing of interrupt requests; 6825when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 6826used in the IRQ routing table. The first args[0] MSI routes are reserved 6827for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 6828a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 6829 6830Fails if VCPU has already been created, or if the irqchip is already in the 6831kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 6832 68337.6 KVM_CAP_S390_RI 6834------------------- 6835 6836:Architectures: s390 6837:Parameters: none 6838 6839Allows use of runtime-instrumentation introduced with zEC12 processor. 6840Will return -EINVAL if the machine does not support runtime-instrumentation. 6841Will return -EBUSY if a VCPU has already been created. 6842 68437.7 KVM_CAP_X2APIC_API 6844---------------------- 6845 6846:Architectures: x86 6847:Parameters: args[0] - features that should be enabled 6848:Returns: 0 on success, -EINVAL when args[0] contains invalid features 6849 6850Valid feature flags in args[0] are:: 6851 6852 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 6853 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 6854 6855Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 6856KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 6857allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 6858respective sections. 6859 6860KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 6861in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 6862as a broadcast even in x2APIC mode in order to support physical x2APIC 6863without interrupt remapping. This is undesirable in logical mode, 6864where 0xff represents CPUs 0-7 in cluster 0. 6865 68667.8 KVM_CAP_S390_USER_INSTR0 6867---------------------------- 6868 6869:Architectures: s390 6870:Parameters: none 6871 6872With this capability enabled, all illegal instructions 0x0000 (2 bytes) will 6873be intercepted and forwarded to user space. User space can use this 6874mechanism e.g. to realize 2-byte software breakpoints. The kernel will 6875not inject an operating exception for these instructions, user space has 6876to take care of that. 6877 6878This capability can be enabled dynamically even if VCPUs were already 6879created and are running. 6880 68817.9 KVM_CAP_S390_GS 6882------------------- 6883 6884:Architectures: s390 6885:Parameters: none 6886:Returns: 0 on success; -EINVAL if the machine does not support 6887 guarded storage; -EBUSY if a VCPU has already been created. 6888 6889Allows use of guarded storage for the KVM guest. 6890 68917.10 KVM_CAP_S390_AIS 6892--------------------- 6893 6894:Architectures: s390 6895:Parameters: none 6896 6897Allow use of adapter-interruption suppression. 6898:Returns: 0 on success; -EBUSY if a VCPU has already been created. 6899 69007.11 KVM_CAP_PPC_SMT 6901-------------------- 6902 6903:Architectures: ppc 6904:Parameters: vsmt_mode, flags 6905 6906Enabling this capability on a VM provides userspace with a way to set 6907the desired virtual SMT mode (i.e. the number of virtual CPUs per 6908virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 6909between 1 and 8. On POWER8, vsmt_mode must also be no greater than 6910the number of threads per subcore for the host. Currently flags must 6911be 0. A successful call to enable this capability will result in 6912vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 6913subsequently queried for the VM. This capability is only supported by 6914HV KVM, and can only be set before any VCPUs have been created. 6915The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 6916modes are available. 6917 69187.12 KVM_CAP_PPC_FWNMI 6919---------------------- 6920 6921:Architectures: ppc 6922:Parameters: none 6923 6924With this capability a machine check exception in the guest address 6925space will cause KVM to exit the guest with NMI exit reason. This 6926enables QEMU to build error log and branch to guest kernel registered 6927machine check handling routine. Without this capability KVM will 6928branch to guests' 0x200 interrupt vector. 6929 69307.13 KVM_CAP_X86_DISABLE_EXITS 6931------------------------------ 6932 6933:Architectures: x86 6934:Parameters: args[0] defines which exits are disabled 6935:Returns: 0 on success, -EINVAL when args[0] contains invalid exits 6936 6937Valid bits in args[0] are:: 6938 6939 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0) 6940 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1) 6941 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2) 6942 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3) 6943 6944Enabling this capability on a VM provides userspace with a way to no 6945longer intercept some instructions for improved latency in some 6946workloads, and is suggested when vCPUs are associated to dedicated 6947physical CPUs. More bits can be added in the future; userspace can 6948just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable 6949all such vmexits. 6950 6951Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits. 6952 69537.14 KVM_CAP_S390_HPAGE_1M 6954-------------------------- 6955 6956:Architectures: s390 6957:Parameters: none 6958:Returns: 0 on success, -EINVAL if hpage module parameter was not set 6959 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL 6960 flag set 6961 6962With this capability the KVM support for memory backing with 1m pages 6963through hugetlbfs can be enabled for a VM. After the capability is 6964enabled, cmma can't be enabled anymore and pfmfi and the storage key 6965interpretation are disabled. If cmma has already been enabled or the 6966hpage module parameter is not set to 1, -EINVAL is returned. 6967 6968While it is generally possible to create a huge page backed VM without 6969this capability, the VM will not be able to run. 6970 69717.15 KVM_CAP_MSR_PLATFORM_INFO 6972------------------------------ 6973 6974:Architectures: x86 6975:Parameters: args[0] whether feature should be enabled or not 6976 6977With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise, 6978a #GP would be raised when the guest tries to access. Currently, this 6979capability does not enable write permissions of this MSR for the guest. 6980 69817.16 KVM_CAP_PPC_NESTED_HV 6982-------------------------- 6983 6984:Architectures: ppc 6985:Parameters: none 6986:Returns: 0 on success, -EINVAL when the implementation doesn't support 6987 nested-HV virtualization. 6988 6989HV-KVM on POWER9 and later systems allows for "nested-HV" 6990virtualization, which provides a way for a guest VM to run guests that 6991can run using the CPU's supervisor mode (privileged non-hypervisor 6992state). Enabling this capability on a VM depends on the CPU having 6993the necessary functionality and on the facility being enabled with a 6994kvm-hv module parameter. 6995 69967.17 KVM_CAP_EXCEPTION_PAYLOAD 6997------------------------------ 6998 6999:Architectures: x86 7000:Parameters: args[0] whether feature should be enabled or not 7001 7002With this capability enabled, CR2 will not be modified prior to the 7003emulated VM-exit when L1 intercepts a #PF exception that occurs in 7004L2. Similarly, for kvm-intel only, DR6 will not be modified prior to 7005the emulated VM-exit when L1 intercepts a #DB exception that occurs in 7006L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or 7007#DB) exception for L2, exception.has_payload will be set and the 7008faulting address (or the new DR6 bits*) will be reported in the 7009exception_payload field. Similarly, when userspace injects a #PF (or 7010#DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set 7011exception.has_payload and to put the faulting address - or the new DR6 7012bits\ [#]_ - in the exception_payload field. 7013 7014This capability also enables exception.pending in struct 7015kvm_vcpu_events, which allows userspace to distinguish between pending 7016and injected exceptions. 7017 7018 7019.. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception 7020 will clear DR6.RTM. 7021 70227.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 7023 7024:Architectures: x86, arm64, mips 7025:Parameters: args[0] whether feature should be enabled or not 7026 7027Valid flags are:: 7028 7029 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0) 7030 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1) 7031 7032With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not 7033automatically clear and write-protect all pages that are returned as dirty. 7034Rather, userspace will have to do this operation separately using 7035KVM_CLEAR_DIRTY_LOG. 7036 7037At the cost of a slightly more complicated operation, this provides better 7038scalability and responsiveness for two reasons. First, 7039KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather 7040than requiring to sync a full memslot; this ensures that KVM does not 7041take spinlocks for an extended period of time. Second, in some cases a 7042large amount of time can pass between a call to KVM_GET_DIRTY_LOG and 7043userspace actually using the data in the page. Pages can be modified 7044during this time, which is inefficient for both the guest and userspace: 7045the guest will incur a higher penalty due to write protection faults, 7046while userspace can see false reports of dirty pages. Manual reprotection 7047helps reducing this time, improving guest performance and reducing the 7048number of dirty log false positives. 7049 7050With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap 7051will be initialized to 1 when created. This also improves performance because 7052dirty logging can be enabled gradually in small chunks on the first call 7053to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on 7054KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on 7055x86 and arm64 for now). 7056 7057KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name 7058KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make 7059it hard or impossible to use it correctly. The availability of 7060KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed. 7061Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT. 7062 70637.19 KVM_CAP_PPC_SECURE_GUEST 7064------------------------------ 7065 7066:Architectures: ppc 7067 7068This capability indicates that KVM is running on a host that has 7069ultravisor firmware and thus can support a secure guest. On such a 7070system, a guest can ask the ultravisor to make it a secure guest, 7071one whose memory is inaccessible to the host except for pages which 7072are explicitly requested to be shared with the host. The ultravisor 7073notifies KVM when a guest requests to become a secure guest, and KVM 7074has the opportunity to veto the transition. 7075 7076If present, this capability can be enabled for a VM, meaning that KVM 7077will allow the transition to secure guest mode. Otherwise KVM will 7078veto the transition. 7079 70807.20 KVM_CAP_HALT_POLL 7081---------------------- 7082 7083:Architectures: all 7084:Target: VM 7085:Parameters: args[0] is the maximum poll time in nanoseconds 7086:Returns: 0 on success; -1 on error 7087 7088This capability overrides the kvm module parameter halt_poll_ns for the 7089target VM. 7090 7091VCPU polling allows a VCPU to poll for wakeup events instead of immediately 7092scheduling during guest halts. The maximum time a VCPU can spend polling is 7093controlled by the kvm module parameter halt_poll_ns. This capability allows 7094the maximum halt time to specified on a per-VM basis, effectively overriding 7095the module parameter for the target VM. 7096 70977.21 KVM_CAP_X86_USER_SPACE_MSR 7098------------------------------- 7099 7100:Architectures: x86 7101:Target: VM 7102:Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report 7103:Returns: 0 on success; -1 on error 7104 7105This capability enables trapping of #GP invoking RDMSR and WRMSR instructions 7106into user space. 7107 7108When a guest requests to read or write an MSR, KVM may not implement all MSRs 7109that are relevant to a respective system. It also does not differentiate by 7110CPU type. 7111 7112To allow more fine grained control over MSR handling, user space may enable 7113this capability. With it enabled, MSR accesses that match the mask specified in 7114args[0] and trigger a #GP event inside the guest by KVM will instead trigger 7115KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications which user space 7116can then handle to implement model specific MSR handling and/or user notifications 7117to inform a user that an MSR was not handled. 7118 71197.22 KVM_CAP_X86_BUS_LOCK_EXIT 7120------------------------------- 7121 7122:Architectures: x86 7123:Target: VM 7124:Parameters: args[0] defines the policy used when bus locks detected in guest 7125:Returns: 0 on success, -EINVAL when args[0] contains invalid bits 7126 7127Valid bits in args[0] are:: 7128 7129 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0) 7130 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1) 7131 7132Enabling this capability on a VM provides userspace with a way to select 7133a policy to handle the bus locks detected in guest. Userspace can obtain 7134the supported modes from the result of KVM_CHECK_EXTENSION and define it 7135through the KVM_ENABLE_CAP. 7136 7137KVM_BUS_LOCK_DETECTION_OFF and KVM_BUS_LOCK_DETECTION_EXIT are supported 7138currently and mutually exclusive with each other. More bits can be added in 7139the future. 7140 7141With KVM_BUS_LOCK_DETECTION_OFF set, bus locks in guest will not cause vm exits 7142so that no additional actions are needed. This is the default mode. 7143 7144With KVM_BUS_LOCK_DETECTION_EXIT set, vm exits happen when bus lock detected 7145in VM. KVM just exits to userspace when handling them. Userspace can enforce 7146its own throttling or other policy based mitigations. 7147 7148This capability is aimed to address the thread that VM can exploit bus locks to 7149degree the performance of the whole system. Once the userspace enable this 7150capability and select the KVM_BUS_LOCK_DETECTION_EXIT mode, KVM will set the 7151KVM_RUN_BUS_LOCK flag in vcpu-run->flags field and exit to userspace. Concerning 7152the bus lock vm exit can be preempted by a higher priority VM exit, the exit 7153notifications to userspace can be KVM_EXIT_BUS_LOCK or other reasons. 7154KVM_RUN_BUS_LOCK flag is used to distinguish between them. 7155 71567.23 KVM_CAP_PPC_DAWR1 7157---------------------- 7158 7159:Architectures: ppc 7160:Parameters: none 7161:Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR 7162 7163This capability can be used to check / enable 2nd DAWR feature provided 7164by POWER10 processor. 7165 7166 71677.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM 7168------------------------------------- 7169 7170Architectures: x86 SEV enabled 7171Type: vm 7172Parameters: args[0] is the fd of the source vm 7173Returns: 0 on success; ENOTTY on error 7174 7175This capability enables userspace to copy encryption context from the vm 7176indicated by the fd to the vm this is called on. 7177 7178This is intended to support in-guest workloads scheduled by the host. This 7179allows the in-guest workload to maintain its own NPTs and keeps the two vms 7180from accidentally clobbering each other with interrupts and the like (separate 7181APIC/MSRs/etc). 7182 71837.25 KVM_CAP_SGX_ATTRIBUTE 7184-------------------------- 7185 7186:Architectures: x86 7187:Target: VM 7188:Parameters: args[0] is a file handle of a SGX attribute file in securityfs 7189:Returns: 0 on success, -EINVAL if the file handle is invalid or if a requested 7190 attribute is not supported by KVM. 7191 7192KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM to grant a VM access to one or 7193more priveleged enclave attributes. args[0] must hold a file handle to a valid 7194SGX attribute file corresponding to an attribute that is supported/restricted 7195by KVM (currently only PROVISIONKEY). 7196 7197The SGX subsystem restricts access to a subset of enclave attributes to provide 7198additional security for an uncompromised kernel, e.g. use of the PROVISIONKEY 7199is restricted to deter malware from using the PROVISIONKEY to obtain a stable 7200system fingerprint. To prevent userspace from circumventing such restrictions 7201by running an enclave in a VM, KVM prevents access to privileged attributes by 7202default. 7203 7204See Documentation/x86/sgx.rst for more details. 7205 72067.26 KVM_CAP_PPC_RPT_INVALIDATE 7207------------------------------- 7208 7209:Capability: KVM_CAP_PPC_RPT_INVALIDATE 7210:Architectures: ppc 7211:Type: vm 7212 7213This capability indicates that the kernel is capable of handling 7214H_RPT_INVALIDATE hcall. 7215 7216In order to enable the use of H_RPT_INVALIDATE in the guest, 7217user space might have to advertise it for the guest. For example, 7218IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is 7219present in the "ibm,hypertas-functions" device-tree property. 7220 7221This capability is enabled for hypervisors on platforms like POWER9 7222that support radix MMU. 7223 72247.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE 7225-------------------------------------- 7226 7227:Architectures: x86 7228:Parameters: args[0] whether the feature should be enabled or not 7229 7230When this capability is enabled, an emulation failure will result in an exit 7231to userspace with KVM_INTERNAL_ERROR (except when the emulator was invoked 7232to handle a VMware backdoor instruction). Furthermore, KVM will now provide up 7233to 15 instruction bytes for any exit to userspace resulting from an emulation 7234failure. When these exits to userspace occur use the emulation_failure struct 7235instead of the internal struct. They both have the same layout, but the 7236emulation_failure struct matches the content better. It also explicitly 7237defines the 'flags' field which is used to describe the fields in the struct 7238that are valid (ie: if KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES is 7239set in the 'flags' field then both 'insn_size' and 'insn_bytes' have valid data 7240in them.) 7241 72427.28 KVM_CAP_ARM_MTE 7243-------------------- 7244 7245:Architectures: arm64 7246:Parameters: none 7247 7248This capability indicates that KVM (and the hardware) supports exposing the 7249Memory Tagging Extensions (MTE) to the guest. It must also be enabled by the 7250VMM before creating any VCPUs to allow the guest access. Note that MTE is only 7251available to a guest running in AArch64 mode and enabling this capability will 7252cause attempts to create AArch32 VCPUs to fail. 7253 7254When enabled the guest is able to access tags associated with any memory given 7255to the guest. KVM will ensure that the tags are maintained during swap or 7256hibernation of the host; however the VMM needs to manually save/restore the 7257tags as appropriate if the VM is migrated. 7258 7259When this capability is enabled all memory in memslots must be mapped as 7260not-shareable (no MAP_SHARED), attempts to create a memslot with a 7261MAP_SHARED mmap will result in an -EINVAL return. 7262 7263When enabled the VMM may make use of the ``KVM_ARM_MTE_COPY_TAGS`` ioctl to 7264perform a bulk copy of tags to/from the guest. 7265 72667.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 7267------------------------------------- 7268 7269Architectures: x86 SEV enabled 7270Type: vm 7271Parameters: args[0] is the fd of the source vm 7272Returns: 0 on success 7273 7274This capability enables userspace to migrate the encryption context from the VM 7275indicated by the fd to the VM this is called on. 7276 7277This is intended to support intra-host migration of VMs between userspace VMMs, 7278upgrading the VMM process without interrupting the guest. 7279 72807.30 KVM_CAP_PPC_AIL_MODE_3 7281------------------------------- 7282 7283:Capability: KVM_CAP_PPC_AIL_MODE_3 7284:Architectures: ppc 7285:Type: vm 7286 7287This capability indicates that the kernel supports the mode 3 setting for the 7288"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location" 7289resource that is controlled with the H_SET_MODE hypercall. 7290 7291This capability allows a guest kernel to use a better-performance mode for 7292handling interrupts and system calls. 7293 72947.31 KVM_CAP_DISABLE_QUIRKS2 7295---------------------------- 7296 7297:Capability: KVM_CAP_DISABLE_QUIRKS2 7298:Parameters: args[0] - set of KVM quirks to disable 7299:Architectures: x86 7300:Type: vm 7301 7302This capability, if enabled, will cause KVM to disable some behavior 7303quirks. 7304 7305Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 7306quirks that can be disabled in KVM. 7307 7308The argument to KVM_ENABLE_CAP for this capability is a bitmask of 7309quirks to disable, and must be a subset of the bitmask returned by 7310KVM_CHECK_EXTENSION. 7311 7312The valid bits in cap.args[0] are: 7313 7314=================================== ============================================ 7315 KVM_X86_QUIRK_LINT0_REENABLED By default, the reset value for the LVT 7316 LINT0 register is 0x700 (APIC_MODE_EXTINT). 7317 When this quirk is disabled, the reset value 7318 is 0x10000 (APIC_LVT_MASKED). 7319 7320 KVM_X86_QUIRK_CD_NW_CLEARED By default, KVM clears CR0.CD and CR0.NW. 7321 When this quirk is disabled, KVM does not 7322 change the value of CR0.CD and CR0.NW. 7323 7324 KVM_X86_QUIRK_LAPIC_MMIO_HOLE By default, the MMIO LAPIC interface is 7325 available even when configured for x2APIC 7326 mode. When this quirk is disabled, KVM 7327 disables the MMIO LAPIC interface if the 7328 LAPIC is in x2APIC mode. 7329 7330 KVM_X86_QUIRK_OUT_7E_INC_RIP By default, KVM pre-increments %rip before 7331 exiting to userspace for an OUT instruction 7332 to port 0x7e. When this quirk is disabled, 7333 KVM does not pre-increment %rip before 7334 exiting to userspace. 7335 7336 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this quirk is disabled, KVM sets 7337 CPUID.01H:ECX[bit 3] (MONITOR/MWAIT) if 7338 IA32_MISC_ENABLE[bit 18] (MWAIT) is set. 7339 Additionally, when this quirk is disabled, 7340 KVM clears CPUID.01H:ECX[bit 3] if 7341 IA32_MISC_ENABLE[bit 18] is cleared. 7342 7343 KVM_X86_QUIRK_FIX_HYPERCALL_INSN By default, KVM rewrites guest 7344 VMMCALL/VMCALL instructions to match the 7345 vendor's hypercall instruction for the 7346 system. When this quirk is disabled, KVM 7347 will no longer rewrite invalid guest 7348 hypercall instructions. Executing the 7349 incorrect hypercall instruction will 7350 generate a #UD within the guest. 7351=================================== ============================================ 7352 73538. Other capabilities. 7354====================== 7355 7356This section lists capabilities that give information about other 7357features of the KVM implementation. 7358 73598.1 KVM_CAP_PPC_HWRNG 7360--------------------- 7361 7362:Architectures: ppc 7363 7364This capability, if KVM_CHECK_EXTENSION indicates that it is 7365available, means that the kernel has an implementation of the 7366H_RANDOM hypercall backed by a hardware random-number generator. 7367If present, the kernel H_RANDOM handler can be enabled for guest use 7368with the KVM_CAP_PPC_ENABLE_HCALL capability. 7369 73708.2 KVM_CAP_HYPERV_SYNIC 7371------------------------ 7372 7373:Architectures: x86 7374 7375This capability, if KVM_CHECK_EXTENSION indicates that it is 7376available, means that the kernel has an implementation of the 7377Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 7378used to support Windows Hyper-V based guest paravirt drivers(VMBus). 7379 7380In order to use SynIC, it has to be activated by setting this 7381capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 7382will disable the use of APIC hardware virtualization even if supported 7383by the CPU, as it's incompatible with SynIC auto-EOI behavior. 7384 73858.3 KVM_CAP_PPC_RADIX_MMU 7386------------------------- 7387 7388:Architectures: ppc 7389 7390This capability, if KVM_CHECK_EXTENSION indicates that it is 7391available, means that the kernel can support guests using the 7392radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 7393processor). 7394 73958.4 KVM_CAP_PPC_HASH_MMU_V3 7396--------------------------- 7397 7398:Architectures: ppc 7399 7400This capability, if KVM_CHECK_EXTENSION indicates that it is 7401available, means that the kernel can support guests using the 7402hashed page table MMU defined in Power ISA V3.00 (as implemented in 7403the POWER9 processor), including in-memory segment tables. 7404 74058.5 KVM_CAP_MIPS_VZ 7406------------------- 7407 7408:Architectures: mips 7409 7410This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 7411it is available, means that full hardware assisted virtualization capabilities 7412of the hardware are available for use through KVM. An appropriate 7413KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 7414utilises it. 7415 7416If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 7417available, it means that the VM is using full hardware assisted virtualization 7418capabilities of the hardware. This is useful to check after creating a VM with 7419KVM_VM_MIPS_DEFAULT. 7420 7421The value returned by KVM_CHECK_EXTENSION should be compared against known 7422values (see below). All other values are reserved. This is to allow for the 7423possibility of other hardware assisted virtualization implementations which 7424may be incompatible with the MIPS VZ ASE. 7425 7426== ========================================================================== 7427 0 The trap & emulate implementation is in use to run guest code in user 7428 mode. Guest virtual memory segments are rearranged to fit the guest in the 7429 user mode address space. 7430 7431 1 The MIPS VZ ASE is in use, providing full hardware assisted 7432 virtualization, including standard guest virtual memory segments. 7433== ========================================================================== 7434 74358.6 KVM_CAP_MIPS_TE 7436------------------- 7437 7438:Architectures: mips 7439 7440This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 7441it is available, means that the trap & emulate implementation is available to 7442run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware 7443assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed 7444to KVM_CREATE_VM to create a VM which utilises it. 7445 7446If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 7447available, it means that the VM is using trap & emulate. 7448 74498.7 KVM_CAP_MIPS_64BIT 7450---------------------- 7451 7452:Architectures: mips 7453 7454This capability indicates the supported architecture type of the guest, i.e. the 7455supported register and address width. 7456 7457The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 7458kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 7459be checked specifically against known values (see below). All other values are 7460reserved. 7461 7462== ======================================================================== 7463 0 MIPS32 or microMIPS32. 7464 Both registers and addresses are 32-bits wide. 7465 It will only be possible to run 32-bit guest code. 7466 7467 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 7468 Registers are 64-bits wide, but addresses are 32-bits wide. 7469 64-bit guest code may run but cannot access MIPS64 memory segments. 7470 It will also be possible to run 32-bit guest code. 7471 7472 2 MIPS64 or microMIPS64 with access to all address segments. 7473 Both registers and addresses are 64-bits wide. 7474 It will be possible to run 64-bit or 32-bit guest code. 7475== ======================================================================== 7476 74778.9 KVM_CAP_ARM_USER_IRQ 7478------------------------ 7479 7480:Architectures: arm64 7481 7482This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 7483that if userspace creates a VM without an in-kernel interrupt controller, it 7484will be notified of changes to the output level of in-kernel emulated devices, 7485which can generate virtual interrupts, presented to the VM. 7486For such VMs, on every return to userspace, the kernel 7487updates the vcpu's run->s.regs.device_irq_level field to represent the actual 7488output level of the device. 7489 7490Whenever kvm detects a change in the device output level, kvm guarantees at 7491least one return to userspace before running the VM. This exit could either 7492be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 7493userspace can always sample the device output level and re-compute the state of 7494the userspace interrupt controller. Userspace should always check the state 7495of run->s.regs.device_irq_level on every kvm exit. 7496The value in run->s.regs.device_irq_level can represent both level and edge 7497triggered interrupt signals, depending on the device. Edge triggered interrupt 7498signals will exit to userspace with the bit in run->s.regs.device_irq_level 7499set exactly once per edge signal. 7500 7501The field run->s.regs.device_irq_level is available independent of 7502run->kvm_valid_regs or run->kvm_dirty_regs bits. 7503 7504If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 7505number larger than 0 indicating the version of this capability is implemented 7506and thereby which bits in run->s.regs.device_irq_level can signal values. 7507 7508Currently the following bits are defined for the device_irq_level bitmap:: 7509 7510 KVM_CAP_ARM_USER_IRQ >= 1: 7511 7512 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 7513 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 7514 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 7515 7516Future versions of kvm may implement additional events. These will get 7517indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 7518listed above. 7519 75208.10 KVM_CAP_PPC_SMT_POSSIBLE 7521----------------------------- 7522 7523:Architectures: ppc 7524 7525Querying this capability returns a bitmap indicating the possible 7526virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 7527(counting from the right) is set, then a virtual SMT mode of 2^N is 7528available. 7529 75308.11 KVM_CAP_HYPERV_SYNIC2 7531-------------------------- 7532 7533:Architectures: x86 7534 7535This capability enables a newer version of Hyper-V Synthetic interrupt 7536controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 7537doesn't clear SynIC message and event flags pages when they are enabled by 7538writing to the respective MSRs. 7539 75408.12 KVM_CAP_HYPERV_VP_INDEX 7541---------------------------- 7542 7543:Architectures: x86 7544 7545This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 7546value is used to denote the target vcpu for a SynIC interrupt. For 7547compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this 7548capability is absent, userspace can still query this msr's value. 7549 75508.13 KVM_CAP_S390_AIS_MIGRATION 7551------------------------------- 7552 7553:Architectures: s390 7554:Parameters: none 7555 7556This capability indicates if the flic device will be able to get/set the 7557AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 7558to discover this without having to create a flic device. 7559 75608.14 KVM_CAP_S390_PSW 7561--------------------- 7562 7563:Architectures: s390 7564 7565This capability indicates that the PSW is exposed via the kvm_run structure. 7566 75678.15 KVM_CAP_S390_GMAP 7568---------------------- 7569 7570:Architectures: s390 7571 7572This capability indicates that the user space memory used as guest mapping can 7573be anywhere in the user memory address space, as long as the memory slots are 7574aligned and sized to a segment (1MB) boundary. 7575 75768.16 KVM_CAP_S390_COW 7577--------------------- 7578 7579:Architectures: s390 7580 7581This capability indicates that the user space memory used as guest mapping can 7582use copy-on-write semantics as well as dirty pages tracking via read-only page 7583tables. 7584 75858.17 KVM_CAP_S390_BPB 7586--------------------- 7587 7588:Architectures: s390 7589 7590This capability indicates that kvm will implement the interfaces to handle 7591reset, migration and nested KVM for branch prediction blocking. The stfle 7592facility 82 should not be provided to the guest without this capability. 7593 75948.18 KVM_CAP_HYPERV_TLBFLUSH 7595---------------------------- 7596 7597:Architectures: x86 7598 7599This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush 7600hypercalls: 7601HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx, 7602HvFlushVirtualAddressList, HvFlushVirtualAddressListEx. 7603 76048.19 KVM_CAP_ARM_INJECT_SERROR_ESR 7605---------------------------------- 7606 7607:Architectures: arm64 7608 7609This capability indicates that userspace can specify (via the 7610KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it 7611takes a virtual SError interrupt exception. 7612If KVM advertises this capability, userspace can only specify the ISS field for 7613the ESR syndrome. Other parts of the ESR, such as the EC are generated by the 7614CPU when the exception is taken. If this virtual SError is taken to EL1 using 7615AArch64, this value will be reported in the ISS field of ESR_ELx. 7616 7617See KVM_CAP_VCPU_EVENTS for more details. 7618 76198.20 KVM_CAP_HYPERV_SEND_IPI 7620---------------------------- 7621 7622:Architectures: x86 7623 7624This capability indicates that KVM supports paravirtualized Hyper-V IPI send 7625hypercalls: 7626HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx. 7627 76288.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 7629----------------------------------- 7630 7631:Architectures: x86 7632 7633This capability indicates that KVM running on top of Hyper-V hypervisor 7634enables Direct TLB flush for its guests meaning that TLB flush 7635hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM. 7636Due to the different ABI for hypercall parameters between Hyper-V and 7637KVM, enabling this capability effectively disables all hypercall 7638handling by KVM (as some KVM hypercall may be mistakenly treated as TLB 7639flush hypercalls by Hyper-V) so userspace should disable KVM identification 7640in CPUID and only exposes Hyper-V identification. In this case, guest 7641thinks it's running on Hyper-V and only use Hyper-V hypercalls. 7642 76438.22 KVM_CAP_S390_VCPU_RESETS 7644----------------------------- 7645 7646:Architectures: s390 7647 7648This capability indicates that the KVM_S390_NORMAL_RESET and 7649KVM_S390_CLEAR_RESET ioctls are available. 7650 76518.23 KVM_CAP_S390_PROTECTED 7652--------------------------- 7653 7654:Architectures: s390 7655 7656This capability indicates that the Ultravisor has been initialized and 7657KVM can therefore start protected VMs. 7658This capability governs the KVM_S390_PV_COMMAND ioctl and the 7659KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected 7660guests when the state change is invalid. 7661 76628.24 KVM_CAP_STEAL_TIME 7663----------------------- 7664 7665:Architectures: arm64, x86 7666 7667This capability indicates that KVM supports steal time accounting. 7668When steal time accounting is supported it may be enabled with 7669architecture-specific interfaces. This capability and the architecture- 7670specific interfaces must be consistent, i.e. if one says the feature 7671is supported, than the other should as well and vice versa. For arm64 7672see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL". 7673For x86 see Documentation/virt/kvm/msr.rst "MSR_KVM_STEAL_TIME". 7674 76758.25 KVM_CAP_S390_DIAG318 7676------------------------- 7677 7678:Architectures: s390 7679 7680This capability enables a guest to set information about its control program 7681(i.e. guest kernel type and version). The information is helpful during 7682system/firmware service events, providing additional data about the guest 7683environments running on the machine. 7684 7685The information is associated with the DIAGNOSE 0x318 instruction, which sets 7686an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and 7687a 7-byte Control Program Version Code (CPVC). The CPNC determines what 7688environment the control program is running in (e.g. Linux, z/VM...), and the 7689CPVC is used for information specific to OS (e.g. Linux version, Linux 7690distribution...) 7691 7692If this capability is available, then the CPNC and CPVC can be synchronized 7693between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318). 7694 76958.26 KVM_CAP_X86_USER_SPACE_MSR 7696------------------------------- 7697 7698:Architectures: x86 7699 7700This capability indicates that KVM supports deflection of MSR reads and 7701writes to user space. It can be enabled on a VM level. If enabled, MSR 7702accesses that would usually trigger a #GP by KVM into the guest will 7703instead get bounced to user space through the KVM_EXIT_X86_RDMSR and 7704KVM_EXIT_X86_WRMSR exit notifications. 7705 77068.27 KVM_CAP_X86_MSR_FILTER 7707--------------------------- 7708 7709:Architectures: x86 7710 7711This capability indicates that KVM supports that accesses to user defined MSRs 7712may be rejected. With this capability exposed, KVM exports new VM ioctl 7713KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR 7714ranges that KVM should reject access to. 7715 7716In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to 7717trap and emulate MSRs that are outside of the scope of KVM as well as 7718limit the attack surface on KVM's MSR emulation code. 7719 77208.28 KVM_CAP_ENFORCE_PV_FEATURE_CPUID 7721------------------------------------- 7722 7723Architectures: x86 7724 7725When enabled, KVM will disable paravirtual features provided to the 7726guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf 7727(0x40000001). Otherwise, a guest may use the paravirtual features 7728regardless of what has actually been exposed through the CPUID leaf. 7729 77308.29 KVM_CAP_DIRTY_LOG_RING 7731--------------------------- 7732 7733:Architectures: x86 7734:Parameters: args[0] - size of the dirty log ring 7735 7736KVM is capable of tracking dirty memory using ring buffers that are 7737mmaped into userspace; there is one dirty ring per vcpu. 7738 7739The dirty ring is available to userspace as an array of 7740``struct kvm_dirty_gfn``. Each dirty entry it's defined as:: 7741 7742 struct kvm_dirty_gfn { 7743 __u32 flags; 7744 __u32 slot; /* as_id | slot_id */ 7745 __u64 offset; 7746 }; 7747 7748The following values are defined for the flags field to define the 7749current state of the entry:: 7750 7751 #define KVM_DIRTY_GFN_F_DIRTY BIT(0) 7752 #define KVM_DIRTY_GFN_F_RESET BIT(1) 7753 #define KVM_DIRTY_GFN_F_MASK 0x3 7754 7755Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM 7756ioctl to enable this capability for the new guest and set the size of 7757the rings. Enabling the capability is only allowed before creating any 7758vCPU, and the size of the ring must be a power of two. The larger the 7759ring buffer, the less likely the ring is full and the VM is forced to 7760exit to userspace. The optimal size depends on the workload, but it is 7761recommended that it be at least 64 KiB (4096 entries). 7762 7763Just like for dirty page bitmaps, the buffer tracks writes to 7764all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was 7765set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered 7766with the flag set, userspace can start harvesting dirty pages from the 7767ring buffer. 7768 7769An entry in the ring buffer can be unused (flag bits ``00``), 7770dirty (flag bits ``01``) or harvested (flag bits ``1X``). The 7771state machine for the entry is as follows:: 7772 7773 dirtied harvested reset 7774 00 -----------> 01 -------------> 1X -------+ 7775 ^ | 7776 | | 7777 +------------------------------------------+ 7778 7779To harvest the dirty pages, userspace accesses the mmaped ring buffer 7780to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage 7781the RESET bit must be cleared), then it means this GFN is a dirty GFN. 7782The userspace should harvest this GFN and mark the flags from state 7783``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set 7784to show that this GFN is harvested and waiting for a reset), and move 7785on to the next GFN. The userspace should continue to do this until the 7786flags of a GFN have the DIRTY bit cleared, meaning that it has harvested 7787all the dirty GFNs that were available. 7788 7789It's not necessary for userspace to harvest the all dirty GFNs at once. 7790However it must collect the dirty GFNs in sequence, i.e., the userspace 7791program cannot skip one dirty GFN to collect the one next to it. 7792 7793After processing one or more entries in the ring buffer, userspace 7794calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about 7795it, so that the kernel will reprotect those collected GFNs. 7796Therefore, the ioctl must be called *before* reading the content of 7797the dirty pages. 7798 7799The dirty ring can get full. When it happens, the KVM_RUN of the 7800vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL. 7801 7802The dirty ring interface has a major difference comparing to the 7803KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from 7804userspace, it's still possible that the kernel has not yet flushed the 7805processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the 7806flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one 7807needs to kick the vcpu out of KVM_RUN using a signal. The resulting 7808vmexit ensures that all dirty GFNs are flushed to the dirty rings. 7809 7810NOTE: the capability KVM_CAP_DIRTY_LOG_RING and the corresponding 7811ioctl KVM_RESET_DIRTY_RINGS are mutual exclusive to the existing ioctls 7812KVM_GET_DIRTY_LOG and KVM_CLEAR_DIRTY_LOG. After enabling 7813KVM_CAP_DIRTY_LOG_RING with an acceptable dirty ring size, the virtual 7814machine will switch to ring-buffer dirty page tracking and further 7815KVM_GET_DIRTY_LOG or KVM_CLEAR_DIRTY_LOG ioctls will fail. 7816 78178.30 KVM_CAP_XEN_HVM 7818-------------------- 7819 7820:Architectures: x86 7821 7822This capability indicates the features that Xen supports for hosting Xen 7823PVHVM guests. Valid flags are:: 7824 7825 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0) 7826 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1) 7827 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2) 7828 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 3) 7829 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL (1 << 4) 7830 #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND (1 << 5) 7831 7832The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG 7833ioctl is available, for the guest to set its hypercall page. 7834 7835If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be 7836provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page 7837contents, to request that KVM generate hypercall page content automatically 7838and also enable interception of guest hypercalls with KVM_EXIT_XEN. 7839 7840The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the 7841KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and 7842KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors 7843for event channel upcalls when the evtchn_upcall_pending field of a vcpu's 7844vcpu_info is set. 7845 7846The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related 7847features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are 7848supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls. 7849 7850The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries 7851of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority 7852field set to indicate 2 level event channel delivery. 7853 7854The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates that KVM supports 7855injecting event channel events directly into the guest with the 7856KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indicates support for the 7857KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attributes and the 7858KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_VECTOR vCPU attributes. 7859related to event channel delivery, timers, and the XENVER_version 7860interception. 7861 78628.31 KVM_CAP_PPC_MULTITCE 7863------------------------- 7864 7865:Capability: KVM_CAP_PPC_MULTITCE 7866:Architectures: ppc 7867:Type: vm 7868 7869This capability means the kernel is capable of handling hypercalls 7870H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 7871space. This significantly accelerates DMA operations for PPC KVM guests. 7872User space should expect that its handlers for these hypercalls 7873are not going to be called if user space previously registered LIOBN 7874in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 7875 7876In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 7877user space might have to advertise it for the guest. For example, 7878IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 7879present in the "ibm,hypertas-functions" device-tree property. 7880 7881The hypercalls mentioned above may or may not be processed successfully 7882in the kernel based fast path. If they can not be handled by the kernel, 7883they will get passed on to user space. So user space still has to have 7884an implementation for these despite the in kernel acceleration. 7885 7886This capability is always enabled. 7887 78888.32 KVM_CAP_PTP_KVM 7889-------------------- 7890 7891:Architectures: arm64 7892 7893This capability indicates that the KVM virtual PTP service is 7894supported in the host. A VMM can check whether the service is 7895available to the guest on migration. 7896 78978.33 KVM_CAP_HYPERV_ENFORCE_CPUID 7898--------------------------------- 7899 7900Architectures: x86 7901 7902When enabled, KVM will disable emulated Hyper-V features provided to the 7903guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all 7904currently implmented Hyper-V features are provided unconditionally when 7905Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001) 7906leaf. 7907 79088.34 KVM_CAP_EXIT_HYPERCALL 7909--------------------------- 7910 7911:Capability: KVM_CAP_EXIT_HYPERCALL 7912:Architectures: x86 7913:Type: vm 7914 7915This capability, if enabled, will cause KVM to exit to userspace 7916with KVM_EXIT_HYPERCALL exit reason to process some hypercalls. 7917 7918Calling KVM_CHECK_EXTENSION for this capability will return a bitmask 7919of hypercalls that can be configured to exit to userspace. 7920Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE. 7921 7922The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset 7923of the result of KVM_CHECK_EXTENSION. KVM will forward to userspace 7924the hypercalls whose corresponding bit is in the argument, and return 7925ENOSYS for the others. 7926 79278.35 KVM_CAP_PMU_CAPABILITY 7928--------------------------- 7929 7930:Capability KVM_CAP_PMU_CAPABILITY 7931:Architectures: x86 7932:Type: vm 7933:Parameters: arg[0] is bitmask of PMU virtualization capabilities. 7934:Returns 0 on success, -EINVAL when arg[0] contains invalid bits 7935 7936This capability alters PMU virtualization in KVM. 7937 7938Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of 7939PMU virtualization capabilities that can be adjusted on a VM. 7940 7941The argument to KVM_ENABLE_CAP is also a bitmask and selects specific 7942PMU virtualization capabilities to be applied to the VM. This can 7943only be invoked on a VM prior to the creation of VCPUs. 7944 7945At this time, KVM_PMU_CAP_DISABLE is the only capability. Setting 7946this capability will disable PMU virtualization for that VM. Usermode 7947should adjust CPUID leaf 0xA to reflect that the PMU is disabled. 7948 79498.36 KVM_CAP_ARM_SYSTEM_SUSPEND 7950------------------------------- 7951 7952:Capability: KVM_CAP_ARM_SYSTEM_SUSPEND 7953:Architectures: arm64 7954:Type: vm 7955 7956When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of 7957type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request. 7958 79599. Known KVM API problems 7960========================= 7961 7962In some cases, KVM's API has some inconsistencies or common pitfalls 7963that userspace need to be aware of. This section details some of 7964these issues. 7965 7966Most of them are architecture specific, so the section is split by 7967architecture. 7968 79699.1. x86 7970-------- 7971 7972``KVM_GET_SUPPORTED_CPUID`` issues 7973^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 7974 7975In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible 7976to take its result and pass it directly to ``KVM_SET_CPUID2``. This section 7977documents some cases in which that requires some care. 7978 7979Local APIC features 7980~~~~~~~~~~~~~~~~~~~ 7981 7982CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``, 7983but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or 7984``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of 7985the local APIC. 7986 7987The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature. 7988 7989CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by ``KVM_GET_SUPPORTED_CPUID``. 7990It can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` is present and the kernel 7991has enabled in-kernel emulation of the local APIC. 7992 7993Obsolete ioctls and capabilities 7994^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 7995 7996KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually 7997available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if 7998available. 7999 8000Ordering of KVM_GET_*/KVM_SET_* ioctls 8001^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 8002 8003TBD 8004