Based on kernel version 4.16.1. Page generated on 2018-04-09 11:53 EST.
1 The Definitive KVM (Kernel-based Virtual Machine) API Documentation 2 =================================================================== 3 4 1. General description 5 ---------------------- 6 7 The kvm API is a set of ioctls that are issued to control various aspects 8 of a virtual machine. The ioctls belong to three classes 9 10 - System ioctls: These query and set global attributes which affect the 11 whole kvm subsystem. In addition a system ioctl is used to create 12 virtual machines 13 14 - VM ioctls: These query and set attributes that affect an entire virtual 15 machine, for example memory layout. In addition a VM ioctl is used to 16 create virtual cpus (vcpus). 17 18 Only run VM ioctls from the same process (address space) that was used 19 to create the VM. 20 21 - vcpu ioctls: These query and set attributes that control the operation 22 of a single virtual cpu. 23 24 Only run vcpu ioctls from the same thread that was used to create the 25 vcpu. 26 27 28 2. File descriptors 29 ------------------- 30 31 The kvm API is centered around file descriptors. An initial 32 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle 33 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this 34 handle will create a VM file descriptor which can be used to issue VM 35 ioctls. A KVM_CREATE_VCPU ioctl on a VM fd will create a virtual cpu 36 and return a file descriptor pointing to it. Finally, ioctls on a vcpu 37 fd can be used to control the vcpu, including the important task of 38 actually running guest code. 39 40 In general file descriptors can be migrated among processes by means 41 of fork() and the SCM_RIGHTS facility of unix domain socket. These 42 kinds of tricks are explicitly not supported by kvm. While they will 43 not cause harm to the host, their actual behavior is not guaranteed by 44 the API. The only supported use is one virtual machine per process, 45 and one vcpu per thread. 46 47 48 3. Extensions 49 ------------- 50 51 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward 52 incompatible change are allowed. However, there is an extension 53 facility that allows backward-compatible extensions to the API to be 54 queried and used. 55 56 The extension mechanism is not based on the Linux version number. 57 Instead, kvm defines extension identifiers and a facility to query 58 whether a particular extension identifier is available. If it is, a 59 set of ioctls is available for application use. 60 61 62 4. API description 63 ------------------ 64 65 This section describes ioctls that can be used to control kvm guests. 66 For each ioctl, the following information is provided along with a 67 description: 68 69 Capability: which KVM extension provides this ioctl. Can be 'basic', 70 which means that is will be provided by any kernel that supports 71 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which 72 means availability needs to be checked with KVM_CHECK_EXTENSION 73 (see section 4.4), or 'none' which means that while not all kernels 74 support this ioctl, there's no capability bit to check its 75 availability: for kernels that don't support the ioctl, 76 the ioctl returns -ENOTTY. 77 78 Architectures: which instruction set architectures provide this ioctl. 79 x86 includes both i386 and x86_64. 80 81 Type: system, vm, or vcpu. 82 83 Parameters: what parameters are accepted by the ioctl. 84 85 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) 86 are not detailed, but errors with specific meanings are. 87 88 89 4.1 KVM_GET_API_VERSION 90 91 Capability: basic 92 Architectures: all 93 Type: system ioctl 94 Parameters: none 95 Returns: the constant KVM_API_VERSION (=12) 96 97 This identifies the API version as the stable kvm API. It is not 98 expected that this number will change. However, Linux 2.6.20 and 99 2.6.21 report earlier versions; these are not documented and not 100 supported. Applications should refuse to run if KVM_GET_API_VERSION 101 returns a value other than 12. If this check passes, all ioctls 102 described as 'basic' will be available. 103 104 105 4.2 KVM_CREATE_VM 106 107 Capability: basic 108 Architectures: all 109 Type: system ioctl 110 Parameters: machine type identifier (KVM_VM_*) 111 Returns: a VM fd that can be used to control the new virtual machine. 112 113 The new VM has no virtual cpus and no memory. 114 You probably want to use 0 as machine type. 115 116 In order to create user controlled virtual machines on S390, check 117 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as 118 privileged user (CAP_SYS_ADMIN). 119 120 To use hardware assisted virtualization on MIPS (VZ ASE) rather than 121 the default trap & emulate implementation (which changes the virtual 122 memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the 123 flag KVM_VM_MIPS_VZ. 124 125 126 4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST 127 128 Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST 129 Architectures: x86 130 Type: system ioctl 131 Parameters: struct kvm_msr_list (in/out) 132 Returns: 0 on success; -1 on error 133 Errors: 134 EFAULT: the msr index list cannot be read from or written to 135 E2BIG: the msr index list is to be to fit in the array specified by 136 the user. 137 138 struct kvm_msr_list { 139 __u32 nmsrs; /* number of msrs in entries */ 140 __u32 indices[0]; 141 }; 142 143 The user fills in the size of the indices array in nmsrs, and in return 144 kvm adjusts nmsrs to reflect the actual number of msrs and fills in the 145 indices array with their numbers. 146 147 KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list 148 varies by kvm version and host processor, but does not change otherwise. 149 150 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are 151 not returned in the MSR list, as different vcpus can have a different number 152 of banks, as set via the KVM_X86_SETUP_MCE ioctl. 153 154 KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed 155 to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities 156 and processor features that are exposed via MSRs (e.g., VMX capabilities). 157 This list also varies by kvm version and host processor, but does not change 158 otherwise. 159 160 161 4.4 KVM_CHECK_EXTENSION 162 163 Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl 164 Architectures: all 165 Type: system ioctl, vm ioctl 166 Parameters: extension identifier (KVM_CAP_*) 167 Returns: 0 if unsupported; 1 (or some other positive integer) if supported 168 169 The API allows the application to query about extensions to the core 170 kvm API. Userspace passes an extension identifier (an integer) and 171 receives an integer that describes the extension availability. 172 Generally 0 means no and 1 means yes, but some extensions may report 173 additional information in the integer return value. 174 175 Based on their initialization different VMs may have different capabilities. 176 It is thus encouraged to use the vm ioctl to query for capabilities (available 177 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 178 179 4.5 KVM_GET_VCPU_MMAP_SIZE 180 181 Capability: basic 182 Architectures: all 183 Type: system ioctl 184 Parameters: none 185 Returns: size of vcpu mmap area, in bytes 186 187 The KVM_RUN ioctl (cf.) communicates with userspace via a shared 188 memory region. This ioctl returns the size of that region. See the 189 KVM_RUN documentation for details. 190 191 192 4.6 KVM_SET_MEMORY_REGION 193 194 Capability: basic 195 Architectures: all 196 Type: vm ioctl 197 Parameters: struct kvm_memory_region (in) 198 Returns: 0 on success, -1 on error 199 200 This ioctl is obsolete and has been removed. 201 202 203 4.7 KVM_CREATE_VCPU 204 205 Capability: basic 206 Architectures: all 207 Type: vm ioctl 208 Parameters: vcpu id (apic id on x86) 209 Returns: vcpu fd on success, -1 on error 210 211 This API adds a vcpu to a virtual machine. No more than max_vcpus may be added. 212 The vcpu id is an integer in the range [0, max_vcpu_id). 213 214 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of 215 the KVM_CHECK_EXTENSION ioctl() at run-time. 216 The maximum possible value for max_vcpus can be retrieved using the 217 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. 218 219 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 220 cpus max. 221 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is 222 same as the value returned from KVM_CAP_NR_VCPUS. 223 224 The maximum possible value for max_vcpu_id can be retrieved using the 225 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time. 226 227 If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id 228 is the same as the value returned from KVM_CAP_MAX_VCPUS. 229 230 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual 231 threads in one or more virtual CPU cores. (This is because the 232 hardware requires all the hardware threads in a CPU core to be in the 233 same partition.) The KVM_CAP_PPC_SMT capability indicates the number 234 of vcpus per virtual core (vcore). The vcore id is obtained by 235 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a 236 given vcore will always be in the same physical core as each other 237 (though that might be a different physical core from time to time). 238 Userspace can control the threading (SMT) mode of the guest by its 239 allocation of vcpu ids. For example, if userspace wants 240 single-threaded guest vcpus, it should make all vcpu ids be a multiple 241 of the number of vcpus per vcore. 242 243 For virtual cpus that have been created with S390 user controlled virtual 244 machines, the resulting vcpu fd can be memory mapped at page offset 245 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual 246 cpu's hardware control block. 247 248 249 4.8 KVM_GET_DIRTY_LOG (vm ioctl) 250 251 Capability: basic 252 Architectures: x86 253 Type: vm ioctl 254 Parameters: struct kvm_dirty_log (in/out) 255 Returns: 0 on success, -1 on error 256 257 /* for KVM_GET_DIRTY_LOG */ 258 struct kvm_dirty_log { 259 __u32 slot; 260 __u32 padding; 261 union { 262 void __user *dirty_bitmap; /* one bit per page */ 263 __u64 padding; 264 }; 265 }; 266 267 Given a memory slot, return a bitmap containing any pages dirtied 268 since the last call to this ioctl. Bit 0 is the first page in the 269 memory slot. Ensure the entire structure is cleared to avoid padding 270 issues. 271 272 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 specifies 273 the address space for which you want to return the dirty bitmap. 274 They must be less than the value that KVM_CHECK_EXTENSION returns for 275 the KVM_CAP_MULTI_ADDRESS_SPACE capability. 276 277 278 4.9 KVM_SET_MEMORY_ALIAS 279 280 Capability: basic 281 Architectures: x86 282 Type: vm ioctl 283 Parameters: struct kvm_memory_alias (in) 284 Returns: 0 (success), -1 (error) 285 286 This ioctl is obsolete and has been removed. 287 288 289 4.10 KVM_RUN 290 291 Capability: basic 292 Architectures: all 293 Type: vcpu ioctl 294 Parameters: none 295 Returns: 0 on success, -1 on error 296 Errors: 297 EINTR: an unmasked signal is pending 298 299 This ioctl is used to run a guest virtual cpu. While there are no 300 explicit parameters, there is an implicit parameter block that can be 301 obtained by mmap()ing the vcpu fd at offset 0, with the size given by 302 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct 303 kvm_run' (see below). 304 305 306 4.11 KVM_GET_REGS 307 308 Capability: basic 309 Architectures: all except ARM, arm64 310 Type: vcpu ioctl 311 Parameters: struct kvm_regs (out) 312 Returns: 0 on success, -1 on error 313 314 Reads the general purpose registers from the vcpu. 315 316 /* x86 */ 317 struct kvm_regs { 318 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 319 __u64 rax, rbx, rcx, rdx; 320 __u64 rsi, rdi, rsp, rbp; 321 __u64 r8, r9, r10, r11; 322 __u64 r12, r13, r14, r15; 323 __u64 rip, rflags; 324 }; 325 326 /* mips */ 327 struct kvm_regs { 328 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ 329 __u64 gpr[32]; 330 __u64 hi; 331 __u64 lo; 332 __u64 pc; 333 }; 334 335 336 4.12 KVM_SET_REGS 337 338 Capability: basic 339 Architectures: all except ARM, arm64 340 Type: vcpu ioctl 341 Parameters: struct kvm_regs (in) 342 Returns: 0 on success, -1 on error 343 344 Writes the general purpose registers into the vcpu. 345 346 See KVM_GET_REGS for the data structure. 347 348 349 4.13 KVM_GET_SREGS 350 351 Capability: basic 352 Architectures: x86, ppc 353 Type: vcpu ioctl 354 Parameters: struct kvm_sregs (out) 355 Returns: 0 on success, -1 on error 356 357 Reads special registers from the vcpu. 358 359 /* x86 */ 360 struct kvm_sregs { 361 struct kvm_segment cs, ds, es, fs, gs, ss; 362 struct kvm_segment tr, ldt; 363 struct kvm_dtable gdt, idt; 364 __u64 cr0, cr2, cr3, cr4, cr8; 365 __u64 efer; 366 __u64 apic_base; 367 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; 368 }; 369 370 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ 371 372 interrupt_bitmap is a bitmap of pending external interrupts. At most 373 one bit may be set. This interrupt has been acknowledged by the APIC 374 but not yet injected into the cpu core. 375 376 377 4.14 KVM_SET_SREGS 378 379 Capability: basic 380 Architectures: x86, ppc 381 Type: vcpu ioctl 382 Parameters: struct kvm_sregs (in) 383 Returns: 0 on success, -1 on error 384 385 Writes special registers into the vcpu. See KVM_GET_SREGS for the 386 data structures. 387 388 389 4.15 KVM_TRANSLATE 390 391 Capability: basic 392 Architectures: x86 393 Type: vcpu ioctl 394 Parameters: struct kvm_translation (in/out) 395 Returns: 0 on success, -1 on error 396 397 Translates a virtual address according to the vcpu's current address 398 translation mode. 399 400 struct kvm_translation { 401 /* in */ 402 __u64 linear_address; 403 404 /* out */ 405 __u64 physical_address; 406 __u8 valid; 407 __u8 writeable; 408 __u8 usermode; 409 __u8 pad[5]; 410 }; 411 412 413 4.16 KVM_INTERRUPT 414 415 Capability: basic 416 Architectures: x86, ppc, mips 417 Type: vcpu ioctl 418 Parameters: struct kvm_interrupt (in) 419 Returns: 0 on success, negative on failure. 420 421 Queues a hardware interrupt vector to be injected. 422 423 /* for KVM_INTERRUPT */ 424 struct kvm_interrupt { 425 /* in */ 426 __u32 irq; 427 }; 428 429 X86: 430 431 Returns: 0 on success, 432 -EEXIST if an interrupt is already enqueued 433 -EINVAL the the irq number is invalid 434 -ENXIO if the PIC is in the kernel 435 -EFAULT if the pointer is invalid 436 437 Note 'irq' is an interrupt vector, not an interrupt pin or line. This 438 ioctl is useful if the in-kernel PIC is not used. 439 440 PPC: 441 442 Queues an external interrupt to be injected. This ioctl is overleaded 443 with 3 different irq values: 444 445 a) KVM_INTERRUPT_SET 446 447 This injects an edge type external interrupt into the guest once it's ready 448 to receive interrupts. When injected, the interrupt is done. 449 450 b) KVM_INTERRUPT_UNSET 451 452 This unsets any pending interrupt. 453 454 Only available with KVM_CAP_PPC_UNSET_IRQ. 455 456 c) KVM_INTERRUPT_SET_LEVEL 457 458 This injects a level type external interrupt into the guest context. The 459 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET 460 is triggered. 461 462 Only available with KVM_CAP_PPC_IRQ_LEVEL. 463 464 Note that any value for 'irq' other than the ones stated above is invalid 465 and incurs unexpected behavior. 466 467 MIPS: 468 469 Queues an external interrupt to be injected into the virtual CPU. A negative 470 interrupt number dequeues the interrupt. 471 472 473 4.17 KVM_DEBUG_GUEST 474 475 Capability: basic 476 Architectures: none 477 Type: vcpu ioctl 478 Parameters: none) 479 Returns: -1 on error 480 481 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. 482 483 484 4.18 KVM_GET_MSRS 485 486 Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system) 487 Architectures: x86 488 Type: system ioctl, vcpu ioctl 489 Parameters: struct kvm_msrs (in/out) 490 Returns: number of msrs successfully returned; 491 -1 on error 492 493 When used as a system ioctl: 494 Reads the values of MSR-based features that are available for the VM. This 495 is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values. 496 The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST 497 in a system ioctl. 498 499 When used as a vcpu ioctl: 500 Reads model-specific registers from the vcpu. Supported msr indices can 501 be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl. 502 503 struct kvm_msrs { 504 __u32 nmsrs; /* number of msrs in entries */ 505 __u32 pad; 506 507 struct kvm_msr_entry entries[0]; 508 }; 509 510 struct kvm_msr_entry { 511 __u32 index; 512 __u32 reserved; 513 __u64 data; 514 }; 515 516 Application code should set the 'nmsrs' member (which indicates the 517 size of the entries array) and the 'index' member of each array entry. 518 kvm will fill in the 'data' member. 519 520 521 4.19 KVM_SET_MSRS 522 523 Capability: basic 524 Architectures: x86 525 Type: vcpu ioctl 526 Parameters: struct kvm_msrs (in) 527 Returns: 0 on success, -1 on error 528 529 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the 530 data structures. 531 532 Application code should set the 'nmsrs' member (which indicates the 533 size of the entries array), and the 'index' and 'data' members of each 534 array entry. 535 536 537 4.20 KVM_SET_CPUID 538 539 Capability: basic 540 Architectures: x86 541 Type: vcpu ioctl 542 Parameters: struct kvm_cpuid (in) 543 Returns: 0 on success, -1 on error 544 545 Defines the vcpu responses to the cpuid instruction. Applications 546 should use the KVM_SET_CPUID2 ioctl if available. 547 548 549 struct kvm_cpuid_entry { 550 __u32 function; 551 __u32 eax; 552 __u32 ebx; 553 __u32 ecx; 554 __u32 edx; 555 __u32 padding; 556 }; 557 558 /* for KVM_SET_CPUID */ 559 struct kvm_cpuid { 560 __u32 nent; 561 __u32 padding; 562 struct kvm_cpuid_entry entries[0]; 563 }; 564 565 566 4.21 KVM_SET_SIGNAL_MASK 567 568 Capability: basic 569 Architectures: all 570 Type: vcpu ioctl 571 Parameters: struct kvm_signal_mask (in) 572 Returns: 0 on success, -1 on error 573 574 Defines which signals are blocked during execution of KVM_RUN. This 575 signal mask temporarily overrides the threads signal mask. Any 576 unblocked signal received (except SIGKILL and SIGSTOP, which retain 577 their traditional behaviour) will cause KVM_RUN to return with -EINTR. 578 579 Note the signal will only be delivered if not blocked by the original 580 signal mask. 581 582 /* for KVM_SET_SIGNAL_MASK */ 583 struct kvm_signal_mask { 584 __u32 len; 585 __u8 sigset[0]; 586 }; 587 588 589 4.22 KVM_GET_FPU 590 591 Capability: basic 592 Architectures: x86 593 Type: vcpu ioctl 594 Parameters: struct kvm_fpu (out) 595 Returns: 0 on success, -1 on error 596 597 Reads the floating point state from the vcpu. 598 599 /* for KVM_GET_FPU and KVM_SET_FPU */ 600 struct kvm_fpu { 601 __u8 fpr[8][16]; 602 __u16 fcw; 603 __u16 fsw; 604 __u8 ftwx; /* in fxsave format */ 605 __u8 pad1; 606 __u16 last_opcode; 607 __u64 last_ip; 608 __u64 last_dp; 609 __u8 xmm[16][16]; 610 __u32 mxcsr; 611 __u32 pad2; 612 }; 613 614 615 4.23 KVM_SET_FPU 616 617 Capability: basic 618 Architectures: x86 619 Type: vcpu ioctl 620 Parameters: struct kvm_fpu (in) 621 Returns: 0 on success, -1 on error 622 623 Writes the floating point state to the vcpu. 624 625 /* for KVM_GET_FPU and KVM_SET_FPU */ 626 struct kvm_fpu { 627 __u8 fpr[8][16]; 628 __u16 fcw; 629 __u16 fsw; 630 __u8 ftwx; /* in fxsave format */ 631 __u8 pad1; 632 __u16 last_opcode; 633 __u64 last_ip; 634 __u64 last_dp; 635 __u8 xmm[16][16]; 636 __u32 mxcsr; 637 __u32 pad2; 638 }; 639 640 641 4.24 KVM_CREATE_IRQCHIP 642 643 Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) 644 Architectures: x86, ARM, arm64, s390 645 Type: vm ioctl 646 Parameters: none 647 Returns: 0 on success, -1 on error 648 649 Creates an interrupt controller model in the kernel. 650 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up 651 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both 652 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. 653 On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of 654 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using 655 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. 656 On s390, a dummy irq routing table is created. 657 658 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled 659 before KVM_CREATE_IRQCHIP can be used. 660 661 662 4.25 KVM_IRQ_LINE 663 664 Capability: KVM_CAP_IRQCHIP 665 Architectures: x86, arm, arm64 666 Type: vm ioctl 667 Parameters: struct kvm_irq_level 668 Returns: 0 on success, -1 on error 669 670 Sets the level of a GSI input to the interrupt controller model in the kernel. 671 On some architectures it is required that an interrupt controller model has 672 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered 673 interrupts require the level to be set to 1 and then back to 0. 674 675 On real hardware, interrupt pins can be active-low or active-high. This 676 does not matter for the level field of struct kvm_irq_level: 1 always 677 means active (asserted), 0 means inactive (deasserted). 678 679 x86 allows the operating system to program the interrupt polarity 680 (active-low/active-high) for level-triggered interrupts, and KVM used 681 to consider the polarity. However, due to bitrot in the handling of 682 active-low interrupts, the above convention is now valid on x86 too. 683 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace 684 should not present interrupts to the guest as active-low unless this 685 capability is present (or unless it is not using the in-kernel irqchip, 686 of course). 687 688 689 ARM/arm64 can signal an interrupt either at the CPU level, or at the 690 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to 691 use PPIs designated for specific cpus. The irq field is interpreted 692 like this: 693 694 bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 | 695 field: | irq_type | vcpu_index | irq_id | 696 697 The irq_type field has the following values: 698 - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ 699 - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) 700 (the vcpu_index field is ignored) 701 - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) 702 703 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) 704 705 In both cases, level is used to assert/deassert the line. 706 707 struct kvm_irq_level { 708 union { 709 __u32 irq; /* GSI */ 710 __s32 status; /* not used for KVM_IRQ_LEVEL */ 711 }; 712 __u32 level; /* 0 or 1 */ 713 }; 714 715 716 4.26 KVM_GET_IRQCHIP 717 718 Capability: KVM_CAP_IRQCHIP 719 Architectures: x86 720 Type: vm ioctl 721 Parameters: struct kvm_irqchip (in/out) 722 Returns: 0 on success, -1 on error 723 724 Reads the state of a kernel interrupt controller created with 725 KVM_CREATE_IRQCHIP into a buffer provided by the caller. 726 727 struct kvm_irqchip { 728 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 729 __u32 pad; 730 union { 731 char dummy[512]; /* reserving space */ 732 struct kvm_pic_state pic; 733 struct kvm_ioapic_state ioapic; 734 } chip; 735 }; 736 737 738 4.27 KVM_SET_IRQCHIP 739 740 Capability: KVM_CAP_IRQCHIP 741 Architectures: x86 742 Type: vm ioctl 743 Parameters: struct kvm_irqchip (in) 744 Returns: 0 on success, -1 on error 745 746 Sets the state of a kernel interrupt controller created with 747 KVM_CREATE_IRQCHIP from a buffer provided by the caller. 748 749 struct kvm_irqchip { 750 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ 751 __u32 pad; 752 union { 753 char dummy[512]; /* reserving space */ 754 struct kvm_pic_state pic; 755 struct kvm_ioapic_state ioapic; 756 } chip; 757 }; 758 759 760 4.28 KVM_XEN_HVM_CONFIG 761 762 Capability: KVM_CAP_XEN_HVM 763 Architectures: x86 764 Type: vm ioctl 765 Parameters: struct kvm_xen_hvm_config (in) 766 Returns: 0 on success, -1 on error 767 768 Sets the MSR that the Xen HVM guest uses to initialize its hypercall 769 page, and provides the starting address and size of the hypercall 770 blobs in userspace. When the guest writes the MSR, kvm copies one 771 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest 772 memory. 773 774 struct kvm_xen_hvm_config { 775 __u32 flags; 776 __u32 msr; 777 __u64 blob_addr_32; 778 __u64 blob_addr_64; 779 __u8 blob_size_32; 780 __u8 blob_size_64; 781 __u8 pad2[30]; 782 }; 783 784 785 4.29 KVM_GET_CLOCK 786 787 Capability: KVM_CAP_ADJUST_CLOCK 788 Architectures: x86 789 Type: vm ioctl 790 Parameters: struct kvm_clock_data (out) 791 Returns: 0 on success, -1 on error 792 793 Gets the current timestamp of kvmclock as seen by the current guest. In 794 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios 795 such as migration. 796 797 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the 798 set of bits that KVM can return in struct kvm_clock_data's flag member. 799 800 The only flag defined now is KVM_CLOCK_TSC_STABLE. If set, the returned 801 value is the exact kvmclock value seen by all VCPUs at the instant 802 when KVM_GET_CLOCK was called. If clear, the returned value is simply 803 CLOCK_MONOTONIC plus a constant offset; the offset can be modified 804 with KVM_SET_CLOCK. KVM will try to make all VCPUs follow this clock, 805 but the exact value read by each VCPU could differ, because the host 806 TSC is not stable. 807 808 struct kvm_clock_data { 809 __u64 clock; /* kvmclock current value */ 810 __u32 flags; 811 __u32 pad[9]; 812 }; 813 814 815 4.30 KVM_SET_CLOCK 816 817 Capability: KVM_CAP_ADJUST_CLOCK 818 Architectures: x86 819 Type: vm ioctl 820 Parameters: struct kvm_clock_data (in) 821 Returns: 0 on success, -1 on error 822 823 Sets the current timestamp of kvmclock to the value specified in its parameter. 824 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios 825 such as migration. 826 827 struct kvm_clock_data { 828 __u64 clock; /* kvmclock current value */ 829 __u32 flags; 830 __u32 pad[9]; 831 }; 832 833 834 4.31 KVM_GET_VCPU_EVENTS 835 836 Capability: KVM_CAP_VCPU_EVENTS 837 Extended by: KVM_CAP_INTR_SHADOW 838 Architectures: x86 839 Type: vm ioctl 840 Parameters: struct kvm_vcpu_event (out) 841 Returns: 0 on success, -1 on error 842 843 Gets currently pending exceptions, interrupts, and NMIs as well as related 844 states of the vcpu. 845 846 struct kvm_vcpu_events { 847 struct { 848 __u8 injected; 849 __u8 nr; 850 __u8 has_error_code; 851 __u8 pad; 852 __u32 error_code; 853 } exception; 854 struct { 855 __u8 injected; 856 __u8 nr; 857 __u8 soft; 858 __u8 shadow; 859 } interrupt; 860 struct { 861 __u8 injected; 862 __u8 pending; 863 __u8 masked; 864 __u8 pad; 865 } nmi; 866 __u32 sipi_vector; 867 __u32 flags; 868 struct { 869 __u8 smm; 870 __u8 pending; 871 __u8 smm_inside_nmi; 872 __u8 latched_init; 873 } smi; 874 }; 875 876 Only two fields are defined in the flags field: 877 878 - KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that 879 interrupt.shadow contains a valid state. 880 881 - KVM_VCPUEVENT_VALID_SMM may be set in the flags field to signal that 882 smi contains a valid state. 883 884 4.32 KVM_SET_VCPU_EVENTS 885 886 Capability: KVM_CAP_VCPU_EVENTS 887 Extended by: KVM_CAP_INTR_SHADOW 888 Architectures: x86 889 Type: vm ioctl 890 Parameters: struct kvm_vcpu_event (in) 891 Returns: 0 on success, -1 on error 892 893 Set pending exceptions, interrupts, and NMIs as well as related states of the 894 vcpu. 895 896 See KVM_GET_VCPU_EVENTS for the data structure. 897 898 Fields that may be modified asynchronously by running VCPUs can be excluded 899 from the update. These fields are nmi.pending, sipi_vector, smi.smm, 900 smi.pending. Keep the corresponding bits in the flags field cleared to 901 suppress overwriting the current in-kernel state. The bits are: 902 903 KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel 904 KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector 905 KVM_VCPUEVENT_VALID_SMM - transfer the smi sub-struct. 906 907 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in 908 the flags field to signal that interrupt.shadow contains a valid state and 909 shall be written into the VCPU. 910 911 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available. 912 913 914 4.33 KVM_GET_DEBUGREGS 915 916 Capability: KVM_CAP_DEBUGREGS 917 Architectures: x86 918 Type: vm ioctl 919 Parameters: struct kvm_debugregs (out) 920 Returns: 0 on success, -1 on error 921 922 Reads debug registers from the vcpu. 923 924 struct kvm_debugregs { 925 __u64 db[4]; 926 __u64 dr6; 927 __u64 dr7; 928 __u64 flags; 929 __u64 reserved[9]; 930 }; 931 932 933 4.34 KVM_SET_DEBUGREGS 934 935 Capability: KVM_CAP_DEBUGREGS 936 Architectures: x86 937 Type: vm ioctl 938 Parameters: struct kvm_debugregs (in) 939 Returns: 0 on success, -1 on error 940 941 Writes debug registers into the vcpu. 942 943 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused 944 yet and must be cleared on entry. 945 946 947 4.35 KVM_SET_USER_MEMORY_REGION 948 949 Capability: KVM_CAP_USER_MEM 950 Architectures: all 951 Type: vm ioctl 952 Parameters: struct kvm_userspace_memory_region (in) 953 Returns: 0 on success, -1 on error 954 955 struct kvm_userspace_memory_region { 956 __u32 slot; 957 __u32 flags; 958 __u64 guest_phys_addr; 959 __u64 memory_size; /* bytes */ 960 __u64 userspace_addr; /* start of the userspace allocated memory */ 961 }; 962 963 /* for kvm_memory_region::flags */ 964 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) 965 #define KVM_MEM_READONLY (1UL << 1) 966 967 This ioctl allows the user to create or modify a guest physical memory 968 slot. When changing an existing slot, it may be moved in the guest 969 physical memory space, or its flags may be modified. It may not be 970 resized. Slots may not overlap in guest physical address space. 971 Bits 0-15 of "slot" specifies the slot id and this value should be 972 less than the maximum number of user memory slots supported per VM. 973 The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS, 974 if this capability is supported by the architecture. 975 976 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot" 977 specifies the address space which is being modified. They must be 978 less than the value that KVM_CHECK_EXTENSION returns for the 979 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces 980 are unrelated; the restriction on overlapping slots only applies within 981 each address space. 982 983 Memory for the region is taken starting at the address denoted by the 984 field userspace_addr, which must point at user addressable memory for 985 the entire memory slot size. Any object may back this memory, including 986 anonymous memory, ordinary files, and hugetlbfs. 987 988 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr 989 be identical. This allows large pages in the guest to be backed by large 990 pages in the host. 991 992 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and 993 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of 994 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to 995 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, 996 to make a new slot read-only. In this case, writes to this memory will be 997 posted to userspace as KVM_EXIT_MMIO exits. 998 999 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of 1000 the memory region are automatically reflected into the guest. For example, an 1001 mmap() that affects the region will be made visible immediately. Another 1002 example is madvise(MADV_DROP). 1003 1004 It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. 1005 The KVM_SET_MEMORY_REGION does not allow fine grained control over memory 1006 allocation and is deprecated. 1007 1008 1009 4.36 KVM_SET_TSS_ADDR 1010 1011 Capability: KVM_CAP_SET_TSS_ADDR 1012 Architectures: x86 1013 Type: vm ioctl 1014 Parameters: unsigned long tss_address (in) 1015 Returns: 0 on success, -1 on error 1016 1017 This ioctl defines the physical address of a three-page region in the guest 1018 physical address space. The region must be within the first 4GB of the 1019 guest physical address space and must not conflict with any memory slot 1020 or any mmio address. The guest may malfunction if it accesses this memory 1021 region. 1022 1023 This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1024 because of a quirk in the virtualization implementation (see the internals 1025 documentation when it pops into existence). 1026 1027 1028 4.37 KVM_ENABLE_CAP 1029 1030 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM 1031 Architectures: x86 (only KVM_CAP_ENABLE_CAP_VM), 1032 mips (only KVM_CAP_ENABLE_CAP), ppc, s390 1033 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM) 1034 Parameters: struct kvm_enable_cap (in) 1035 Returns: 0 on success; -1 on error 1036 1037 +Not all extensions are enabled by default. Using this ioctl the application 1038 can enable an extension, making it available to the guest. 1039 1040 On systems that do not support this ioctl, it always fails. On systems that 1041 do support it, it only works for extensions that are supported for enablement. 1042 1043 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should 1044 be used. 1045 1046 struct kvm_enable_cap { 1047 /* in */ 1048 __u32 cap; 1049 1050 The capability that is supposed to get enabled. 1051 1052 __u32 flags; 1053 1054 A bitfield indicating future enhancements. Has to be 0 for now. 1055 1056 __u64 args[4]; 1057 1058 Arguments for enabling a feature. If a feature needs initial values to 1059 function properly, this is the place to put them. 1060 1061 __u8 pad[64]; 1062 }; 1063 1064 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl 1065 for vm-wide capabilities. 1066 1067 4.38 KVM_GET_MP_STATE 1068 1069 Capability: KVM_CAP_MP_STATE 1070 Architectures: x86, s390, arm, arm64 1071 Type: vcpu ioctl 1072 Parameters: struct kvm_mp_state (out) 1073 Returns: 0 on success; -1 on error 1074 1075 struct kvm_mp_state { 1076 __u32 mp_state; 1077 }; 1078 1079 Returns the vcpu's current "multiprocessing state" (though also valid on 1080 uniprocessor guests). 1081 1082 Possible values are: 1083 1084 - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86,arm/arm64] 1085 - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP) 1086 which has not yet received an INIT signal [x86] 1087 - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is 1088 now ready for a SIPI [x86] 1089 - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and 1090 is waiting for an interrupt [x86] 1091 - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector 1092 accessible via KVM_GET_VCPU_EVENTS) [x86] 1093 - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390,arm/arm64] 1094 - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390] 1095 - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted) 1096 [s390] 1097 - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state 1098 [s390] 1099 1100 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1101 in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1102 these architectures. 1103 1104 For arm/arm64: 1105 1106 The only states that are valid are KVM_MP_STATE_STOPPED and 1107 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not. 1108 1109 4.39 KVM_SET_MP_STATE 1110 1111 Capability: KVM_CAP_MP_STATE 1112 Architectures: x86, s390, arm, arm64 1113 Type: vcpu ioctl 1114 Parameters: struct kvm_mp_state (in) 1115 Returns: 0 on success; -1 on error 1116 1117 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for 1118 arguments. 1119 1120 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an 1121 in-kernel irqchip, the multiprocessing state must be maintained by userspace on 1122 these architectures. 1123 1124 For arm/arm64: 1125 1126 The only states that are valid are KVM_MP_STATE_STOPPED and 1127 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not. 1128 1129 4.40 KVM_SET_IDENTITY_MAP_ADDR 1130 1131 Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1132 Architectures: x86 1133 Type: vm ioctl 1134 Parameters: unsigned long identity (in) 1135 Returns: 0 on success, -1 on error 1136 1137 This ioctl defines the physical address of a one-page region in the guest 1138 physical address space. The region must be within the first 4GB of the 1139 guest physical address space and must not conflict with any memory slot 1140 or any mmio address. The guest may malfunction if it accesses this memory 1141 region. 1142 1143 Setting the address to 0 will result in resetting the address to its default 1144 (0xfffbc000). 1145 1146 This ioctl is required on Intel-based hosts. This is needed on Intel hardware 1147 because of a quirk in the virtualization implementation (see the internals 1148 documentation when it pops into existence). 1149 1150 Fails if any VCPU has already been created. 1151 1152 4.41 KVM_SET_BOOT_CPU_ID 1153 1154 Capability: KVM_CAP_SET_BOOT_CPU_ID 1155 Architectures: x86 1156 Type: vm ioctl 1157 Parameters: unsigned long vcpu_id 1158 Returns: 0 on success, -1 on error 1159 1160 Define which vcpu is the Bootstrap Processor (BSP). Values are the same 1161 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default 1162 is vcpu 0. 1163 1164 1165 4.42 KVM_GET_XSAVE 1166 1167 Capability: KVM_CAP_XSAVE 1168 Architectures: x86 1169 Type: vcpu ioctl 1170 Parameters: struct kvm_xsave (out) 1171 Returns: 0 on success, -1 on error 1172 1173 struct kvm_xsave { 1174 __u32 region[1024]; 1175 }; 1176 1177 This ioctl would copy current vcpu's xsave struct to the userspace. 1178 1179 1180 4.43 KVM_SET_XSAVE 1181 1182 Capability: KVM_CAP_XSAVE 1183 Architectures: x86 1184 Type: vcpu ioctl 1185 Parameters: struct kvm_xsave (in) 1186 Returns: 0 on success, -1 on error 1187 1188 struct kvm_xsave { 1189 __u32 region[1024]; 1190 }; 1191 1192 This ioctl would copy userspace's xsave struct to the kernel. 1193 1194 1195 4.44 KVM_GET_XCRS 1196 1197 Capability: KVM_CAP_XCRS 1198 Architectures: x86 1199 Type: vcpu ioctl 1200 Parameters: struct kvm_xcrs (out) 1201 Returns: 0 on success, -1 on error 1202 1203 struct kvm_xcr { 1204 __u32 xcr; 1205 __u32 reserved; 1206 __u64 value; 1207 }; 1208 1209 struct kvm_xcrs { 1210 __u32 nr_xcrs; 1211 __u32 flags; 1212 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1213 __u64 padding[16]; 1214 }; 1215 1216 This ioctl would copy current vcpu's xcrs to the userspace. 1217 1218 1219 4.45 KVM_SET_XCRS 1220 1221 Capability: KVM_CAP_XCRS 1222 Architectures: x86 1223 Type: vcpu ioctl 1224 Parameters: struct kvm_xcrs (in) 1225 Returns: 0 on success, -1 on error 1226 1227 struct kvm_xcr { 1228 __u32 xcr; 1229 __u32 reserved; 1230 __u64 value; 1231 }; 1232 1233 struct kvm_xcrs { 1234 __u32 nr_xcrs; 1235 __u32 flags; 1236 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1237 __u64 padding[16]; 1238 }; 1239 1240 This ioctl would set vcpu's xcr to the value userspace specified. 1241 1242 1243 4.46 KVM_GET_SUPPORTED_CPUID 1244 1245 Capability: KVM_CAP_EXT_CPUID 1246 Architectures: x86 1247 Type: system ioctl 1248 Parameters: struct kvm_cpuid2 (in/out) 1249 Returns: 0 on success, -1 on error 1250 1251 struct kvm_cpuid2 { 1252 __u32 nent; 1253 __u32 padding; 1254 struct kvm_cpuid_entry2 entries[0]; 1255 }; 1256 1257 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 1258 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) 1259 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) 1260 1261 struct kvm_cpuid_entry2 { 1262 __u32 function; 1263 __u32 index; 1264 __u32 flags; 1265 __u32 eax; 1266 __u32 ebx; 1267 __u32 ecx; 1268 __u32 edx; 1269 __u32 padding[3]; 1270 }; 1271 1272 This ioctl returns x86 cpuid features which are supported by both the hardware 1273 and kvm. Userspace can use the information returned by this ioctl to 1274 construct cpuid information (for KVM_SET_CPUID2) that is consistent with 1275 hardware, kernel, and userspace capabilities, and with user requirements (for 1276 example, the user may wish to constrain cpuid to emulate older hardware, 1277 or for feature consistency across a cluster). 1278 1279 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure 1280 with the 'nent' field indicating the number of entries in the variable-size 1281 array 'entries'. If the number of entries is too low to describe the cpu 1282 capabilities, an error (E2BIG) is returned. If the number is too high, 1283 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the 1284 number is just right, the 'nent' field is adjusted to the number of valid 1285 entries in the 'entries' array, which is then filled. 1286 1287 The entries returned are the host cpuid as returned by the cpuid instruction, 1288 with unknown or unsupported features masked out. Some features (for example, 1289 x2apic), may not be present in the host cpu, but are exposed by kvm if it can 1290 emulate them efficiently. The fields in each entry are defined as follows: 1291 1292 function: the eax value used to obtain the entry 1293 index: the ecx value used to obtain the entry (for entries that are 1294 affected by ecx) 1295 flags: an OR of zero or more of the following: 1296 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1297 if the index field is valid 1298 KVM_CPUID_FLAG_STATEFUL_FUNC: 1299 if cpuid for this function returns different values for successive 1300 invocations; there will be several entries with the same function, 1301 all with this flag set 1302 KVM_CPUID_FLAG_STATE_READ_NEXT: 1303 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is 1304 the first entry to be read by a cpu 1305 eax, ebx, ecx, edx: the values returned by the cpuid instruction for 1306 this function/index combination 1307 1308 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned 1309 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC 1310 support. Instead it is reported via 1311 1312 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) 1313 1314 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the 1315 feature in userspace, then you can enable the feature for KVM_SET_CPUID2. 1316 1317 1318 4.47 KVM_PPC_GET_PVINFO 1319 1320 Capability: KVM_CAP_PPC_GET_PVINFO 1321 Architectures: ppc 1322 Type: vm ioctl 1323 Parameters: struct kvm_ppc_pvinfo (out) 1324 Returns: 0 on success, !0 on error 1325 1326 struct kvm_ppc_pvinfo { 1327 __u32 flags; 1328 __u32 hcall[4]; 1329 __u8 pad[108]; 1330 }; 1331 1332 This ioctl fetches PV specific information that need to be passed to the guest 1333 using the device tree or other means from vm context. 1334 1335 The hcall array defines 4 instructions that make up a hypercall. 1336 1337 If any additional field gets added to this structure later on, a bit for that 1338 additional piece of information will be set in the flags bitmap. 1339 1340 The flags bitmap is defined as: 1341 1342 /* the host supports the ePAPR idle hcall 1343 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) 1344 1345 4.52 KVM_SET_GSI_ROUTING 1346 1347 Capability: KVM_CAP_IRQ_ROUTING 1348 Architectures: x86 s390 arm arm64 1349 Type: vm ioctl 1350 Parameters: struct kvm_irq_routing (in) 1351 Returns: 0 on success, -1 on error 1352 1353 Sets the GSI routing table entries, overwriting any previously set entries. 1354 1355 On arm/arm64, GSI routing has the following limitation: 1356 - GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD. 1357 1358 struct kvm_irq_routing { 1359 __u32 nr; 1360 __u32 flags; 1361 struct kvm_irq_routing_entry entries[0]; 1362 }; 1363 1364 No flags are specified so far, the corresponding field must be set to zero. 1365 1366 struct kvm_irq_routing_entry { 1367 __u32 gsi; 1368 __u32 type; 1369 __u32 flags; 1370 __u32 pad; 1371 union { 1372 struct kvm_irq_routing_irqchip irqchip; 1373 struct kvm_irq_routing_msi msi; 1374 struct kvm_irq_routing_s390_adapter adapter; 1375 struct kvm_irq_routing_hv_sint hv_sint; 1376 __u32 pad[8]; 1377 } u; 1378 }; 1379 1380 /* gsi routing entry types */ 1381 #define KVM_IRQ_ROUTING_IRQCHIP 1 1382 #define KVM_IRQ_ROUTING_MSI 2 1383 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1384 #define KVM_IRQ_ROUTING_HV_SINT 4 1385 1386 flags: 1387 - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry 1388 type, specifies that the devid field contains a valid value. The per-VM 1389 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 1390 the device ID. If this capability is not available, userspace should 1391 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 1392 - zero otherwise 1393 1394 struct kvm_irq_routing_irqchip { 1395 __u32 irqchip; 1396 __u32 pin; 1397 }; 1398 1399 struct kvm_irq_routing_msi { 1400 __u32 address_lo; 1401 __u32 address_hi; 1402 __u32 data; 1403 union { 1404 __u32 pad; 1405 __u32 devid; 1406 }; 1407 }; 1408 1409 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 1410 for the device that wrote the MSI message. For PCI, this is usually a 1411 BFD identifier in the lower 16 bits. 1412 1413 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 1414 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 1415 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 1416 address_hi must be zero. 1417 1418 struct kvm_irq_routing_s390_adapter { 1419 __u64 ind_addr; 1420 __u64 summary_addr; 1421 __u64 ind_offset; 1422 __u32 summary_offset; 1423 __u32 adapter_id; 1424 }; 1425 1426 struct kvm_irq_routing_hv_sint { 1427 __u32 vcpu; 1428 __u32 sint; 1429 }; 1430 1431 1432 4.55 KVM_SET_TSC_KHZ 1433 1434 Capability: KVM_CAP_TSC_CONTROL 1435 Architectures: x86 1436 Type: vcpu ioctl 1437 Parameters: virtual tsc_khz 1438 Returns: 0 on success, -1 on error 1439 1440 Specifies the tsc frequency for the virtual machine. The unit of the 1441 frequency is KHz. 1442 1443 1444 4.56 KVM_GET_TSC_KHZ 1445 1446 Capability: KVM_CAP_GET_TSC_KHZ 1447 Architectures: x86 1448 Type: vcpu ioctl 1449 Parameters: none 1450 Returns: virtual tsc-khz on success, negative value on error 1451 1452 Returns the tsc frequency of the guest. The unit of the return value is 1453 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an 1454 error. 1455 1456 1457 4.57 KVM_GET_LAPIC 1458 1459 Capability: KVM_CAP_IRQCHIP 1460 Architectures: x86 1461 Type: vcpu ioctl 1462 Parameters: struct kvm_lapic_state (out) 1463 Returns: 0 on success, -1 on error 1464 1465 #define KVM_APIC_REG_SIZE 0x400 1466 struct kvm_lapic_state { 1467 char regs[KVM_APIC_REG_SIZE]; 1468 }; 1469 1470 Reads the Local APIC registers and copies them into the input argument. The 1471 data format and layout are the same as documented in the architecture manual. 1472 1473 If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is 1474 enabled, then the format of APIC_ID register depends on the APIC mode 1475 (reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in 1476 the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID 1477 which is stored in bits 31-24 of the APIC register, or equivalently in 1478 byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then 1479 be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. 1480 1481 If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state 1482 always uses xAPIC format. 1483 1484 1485 4.58 KVM_SET_LAPIC 1486 1487 Capability: KVM_CAP_IRQCHIP 1488 Architectures: x86 1489 Type: vcpu ioctl 1490 Parameters: struct kvm_lapic_state (in) 1491 Returns: 0 on success, -1 on error 1492 1493 #define KVM_APIC_REG_SIZE 0x400 1494 struct kvm_lapic_state { 1495 char regs[KVM_APIC_REG_SIZE]; 1496 }; 1497 1498 Copies the input argument into the Local APIC registers. The data format 1499 and layout are the same as documented in the architecture manual. 1500 1501 The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's 1502 regs field) depends on the state of the KVM_CAP_X2APIC_API capability. 1503 See the note in KVM_GET_LAPIC. 1504 1505 1506 4.59 KVM_IOEVENTFD 1507 1508 Capability: KVM_CAP_IOEVENTFD 1509 Architectures: all 1510 Type: vm ioctl 1511 Parameters: struct kvm_ioeventfd (in) 1512 Returns: 0 on success, !0 on error 1513 1514 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address 1515 within the guest. A guest write in the registered address will signal the 1516 provided event instead of triggering an exit. 1517 1518 struct kvm_ioeventfd { 1519 __u64 datamatch; 1520 __u64 addr; /* legal pio/mmio address */ 1521 __u32 len; /* 0, 1, 2, 4, or 8 bytes */ 1522 __s32 fd; 1523 __u32 flags; 1524 __u8 pad[36]; 1525 }; 1526 1527 For the special case of virtio-ccw devices on s390, the ioevent is matched 1528 to a subchannel/virtqueue tuple instead. 1529 1530 The following flags are defined: 1531 1532 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) 1533 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) 1534 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) 1535 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ 1536 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) 1537 1538 If datamatch flag is set, the event will be signaled only if the written value 1539 to the registered address is equal to datamatch in struct kvm_ioeventfd. 1540 1541 For virtio-ccw devices, addr contains the subchannel id and datamatch the 1542 virtqueue index. 1543 1544 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and 1545 the kernel will ignore the length of guest write and may get a faster vmexit. 1546 The speedup may only apply to specific architectures, but the ioeventfd will 1547 work anyway. 1548 1549 4.60 KVM_DIRTY_TLB 1550 1551 Capability: KVM_CAP_SW_TLB 1552 Architectures: ppc 1553 Type: vcpu ioctl 1554 Parameters: struct kvm_dirty_tlb (in) 1555 Returns: 0 on success, -1 on error 1556 1557 struct kvm_dirty_tlb { 1558 __u64 bitmap; 1559 __u32 num_dirty; 1560 }; 1561 1562 This must be called whenever userspace has changed an entry in the shared 1563 TLB, prior to calling KVM_RUN on the associated vcpu. 1564 1565 The "bitmap" field is the userspace address of an array. This array 1566 consists of a number of bits, equal to the total number of TLB entries as 1567 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the 1568 nearest multiple of 64. 1569 1570 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB 1571 array. 1572 1573 The array is little-endian: the bit 0 is the least significant bit of the 1574 first byte, bit 8 is the least significant bit of the second byte, etc. 1575 This avoids any complications with differing word sizes. 1576 1577 The "num_dirty" field is a performance hint for KVM to determine whether it 1578 should skip processing the bitmap and just invalidate everything. It must 1579 be set to the number of set bits in the bitmap. 1580 1581 1582 4.62 KVM_CREATE_SPAPR_TCE 1583 1584 Capability: KVM_CAP_SPAPR_TCE 1585 Architectures: powerpc 1586 Type: vm ioctl 1587 Parameters: struct kvm_create_spapr_tce (in) 1588 Returns: file descriptor for manipulating the created TCE table 1589 1590 This creates a virtual TCE (translation control entry) table, which 1591 is an IOMMU for PAPR-style virtual I/O. It is used to translate 1592 logical addresses used in virtual I/O into guest physical addresses, 1593 and provides a scatter/gather capability for PAPR virtual I/O. 1594 1595 /* for KVM_CAP_SPAPR_TCE */ 1596 struct kvm_create_spapr_tce { 1597 __u64 liobn; 1598 __u32 window_size; 1599 }; 1600 1601 The liobn field gives the logical IO bus number for which to create a 1602 TCE table. The window_size field specifies the size of the DMA window 1603 which this TCE table will translate - the table will contain one 64 1604 bit TCE entry for every 4kiB of the DMA window. 1605 1606 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE 1607 table has been created using this ioctl(), the kernel will handle it 1608 in real mode, updating the TCE table. H_PUT_TCE calls for other 1609 liobns will cause a vm exit and must be handled by userspace. 1610 1611 The return value is a file descriptor which can be passed to mmap(2) 1612 to map the created TCE table into userspace. This lets userspace read 1613 the entries written by kernel-handled H_PUT_TCE calls, and also lets 1614 userspace update the TCE table directly which is useful in some 1615 circumstances. 1616 1617 1618 4.63 KVM_ALLOCATE_RMA 1619 1620 Capability: KVM_CAP_PPC_RMA 1621 Architectures: powerpc 1622 Type: vm ioctl 1623 Parameters: struct kvm_allocate_rma (out) 1624 Returns: file descriptor for mapping the allocated RMA 1625 1626 This allocates a Real Mode Area (RMA) from the pool allocated at boot 1627 time by the kernel. An RMA is a physically-contiguous, aligned region 1628 of memory used on older POWER processors to provide the memory which 1629 will be accessed by real-mode (MMU off) accesses in a KVM guest. 1630 POWER processors support a set of sizes for the RMA that usually 1631 includes 64MB, 128MB, 256MB and some larger powers of two. 1632 1633 /* for KVM_ALLOCATE_RMA */ 1634 struct kvm_allocate_rma { 1635 __u64 rma_size; 1636 }; 1637 1638 The return value is a file descriptor which can be passed to mmap(2) 1639 to map the allocated RMA into userspace. The mapped area can then be 1640 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the 1641 RMA for a virtual machine. The size of the RMA in bytes (which is 1642 fixed at host kernel boot time) is returned in the rma_size field of 1643 the argument structure. 1644 1645 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl 1646 is supported; 2 if the processor requires all virtual machines to have 1647 an RMA, or 1 if the processor can use an RMA but doesn't require it, 1648 because it supports the Virtual RMA (VRMA) facility. 1649 1650 1651 4.64 KVM_NMI 1652 1653 Capability: KVM_CAP_USER_NMI 1654 Architectures: x86 1655 Type: vcpu ioctl 1656 Parameters: none 1657 Returns: 0 on success, -1 on error 1658 1659 Queues an NMI on the thread's vcpu. Note this is well defined only 1660 when KVM_CREATE_IRQCHIP has not been called, since this is an interface 1661 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP 1662 has been called, this interface is completely emulated within the kernel. 1663 1664 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the 1665 following algorithm: 1666 1667 - pause the vcpu 1668 - read the local APIC's state (KVM_GET_LAPIC) 1669 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) 1670 - if so, issue KVM_NMI 1671 - resume the vcpu 1672 1673 Some guests configure the LINT1 NMI input to cause a panic, aiding in 1674 debugging. 1675 1676 1677 4.65 KVM_S390_UCAS_MAP 1678 1679 Capability: KVM_CAP_S390_UCONTROL 1680 Architectures: s390 1681 Type: vcpu ioctl 1682 Parameters: struct kvm_s390_ucas_mapping (in) 1683 Returns: 0 in case of success 1684 1685 The parameter is defined like this: 1686 struct kvm_s390_ucas_mapping { 1687 __u64 user_addr; 1688 __u64 vcpu_addr; 1689 __u64 length; 1690 }; 1691 1692 This ioctl maps the memory at "user_addr" with the length "length" to 1693 the vcpu's address space starting at "vcpu_addr". All parameters need to 1694 be aligned by 1 megabyte. 1695 1696 1697 4.66 KVM_S390_UCAS_UNMAP 1698 1699 Capability: KVM_CAP_S390_UCONTROL 1700 Architectures: s390 1701 Type: vcpu ioctl 1702 Parameters: struct kvm_s390_ucas_mapping (in) 1703 Returns: 0 in case of success 1704 1705 The parameter is defined like this: 1706 struct kvm_s390_ucas_mapping { 1707 __u64 user_addr; 1708 __u64 vcpu_addr; 1709 __u64 length; 1710 }; 1711 1712 This ioctl unmaps the memory in the vcpu's address space starting at 1713 "vcpu_addr" with the length "length". The field "user_addr" is ignored. 1714 All parameters need to be aligned by 1 megabyte. 1715 1716 1717 4.67 KVM_S390_VCPU_FAULT 1718 1719 Capability: KVM_CAP_S390_UCONTROL 1720 Architectures: s390 1721 Type: vcpu ioctl 1722 Parameters: vcpu absolute address (in) 1723 Returns: 0 in case of success 1724 1725 This call creates a page table entry on the virtual cpu's address space 1726 (for user controlled virtual machines) or the virtual machine's address 1727 space (for regular virtual machines). This only works for minor faults, 1728 thus it's recommended to access subject memory page via the user page 1729 table upfront. This is useful to handle validity intercepts for user 1730 controlled virtual machines to fault in the virtual cpu's lowcore pages 1731 prior to calling the KVM_RUN ioctl. 1732 1733 1734 4.68 KVM_SET_ONE_REG 1735 1736 Capability: KVM_CAP_ONE_REG 1737 Architectures: all 1738 Type: vcpu ioctl 1739 Parameters: struct kvm_one_reg (in) 1740 Returns: 0 on success, negative value on failure 1741 1742 struct kvm_one_reg { 1743 __u64 id; 1744 __u64 addr; 1745 }; 1746 1747 Using this ioctl, a single vcpu register can be set to a specific value 1748 defined by user space with the passed in struct kvm_one_reg, where id 1749 refers to the register identifier as described below and addr is a pointer 1750 to a variable with the respective size. There can be architecture agnostic 1751 and architecture specific registers. Each have their own range of operation 1752 and their own constants and width. To keep track of the implemented 1753 registers, find a list below: 1754 1755 Arch | Register | Width (bits) 1756 | | 1757 PPC | KVM_REG_PPC_HIOR | 64 1758 PPC | KVM_REG_PPC_IAC1 | 64 1759 PPC | KVM_REG_PPC_IAC2 | 64 1760 PPC | KVM_REG_PPC_IAC3 | 64 1761 PPC | KVM_REG_PPC_IAC4 | 64 1762 PPC | KVM_REG_PPC_DAC1 | 64 1763 PPC | KVM_REG_PPC_DAC2 | 64 1764 PPC | KVM_REG_PPC_DABR | 64 1765 PPC | KVM_REG_PPC_DSCR | 64 1766 PPC | KVM_REG_PPC_PURR | 64 1767 PPC | KVM_REG_PPC_SPURR | 64 1768 PPC | KVM_REG_PPC_DAR | 64 1769 PPC | KVM_REG_PPC_DSISR | 32 1770 PPC | KVM_REG_PPC_AMR | 64 1771 PPC | KVM_REG_PPC_UAMOR | 64 1772 PPC | KVM_REG_PPC_MMCR0 | 64 1773 PPC | KVM_REG_PPC_MMCR1 | 64 1774 PPC | KVM_REG_PPC_MMCRA | 64 1775 PPC | KVM_REG_PPC_MMCR2 | 64 1776 PPC | KVM_REG_PPC_MMCRS | 64 1777 PPC | KVM_REG_PPC_SIAR | 64 1778 PPC | KVM_REG_PPC_SDAR | 64 1779 PPC | KVM_REG_PPC_SIER | 64 1780 PPC | KVM_REG_PPC_PMC1 | 32 1781 PPC | KVM_REG_PPC_PMC2 | 32 1782 PPC | KVM_REG_PPC_PMC3 | 32 1783 PPC | KVM_REG_PPC_PMC4 | 32 1784 PPC | KVM_REG_PPC_PMC5 | 32 1785 PPC | KVM_REG_PPC_PMC6 | 32 1786 PPC | KVM_REG_PPC_PMC7 | 32 1787 PPC | KVM_REG_PPC_PMC8 | 32 1788 PPC | KVM_REG_PPC_FPR0 | 64 1789 ... 1790 PPC | KVM_REG_PPC_FPR31 | 64 1791 PPC | KVM_REG_PPC_VR0 | 128 1792 ... 1793 PPC | KVM_REG_PPC_VR31 | 128 1794 PPC | KVM_REG_PPC_VSR0 | 128 1795 ... 1796 PPC | KVM_REG_PPC_VSR31 | 128 1797 PPC | KVM_REG_PPC_FPSCR | 64 1798 PPC | KVM_REG_PPC_VSCR | 32 1799 PPC | KVM_REG_PPC_VPA_ADDR | 64 1800 PPC | KVM_REG_PPC_VPA_SLB | 128 1801 PPC | KVM_REG_PPC_VPA_DTL | 128 1802 PPC | KVM_REG_PPC_EPCR | 32 1803 PPC | KVM_REG_PPC_EPR | 32 1804 PPC | KVM_REG_PPC_TCR | 32 1805 PPC | KVM_REG_PPC_TSR | 32 1806 PPC | KVM_REG_PPC_OR_TSR | 32 1807 PPC | KVM_REG_PPC_CLEAR_TSR | 32 1808 PPC | KVM_REG_PPC_MAS0 | 32 1809 PPC | KVM_REG_PPC_MAS1 | 32 1810 PPC | KVM_REG_PPC_MAS2 | 64 1811 PPC | KVM_REG_PPC_MAS7_3 | 64 1812 PPC | KVM_REG_PPC_MAS4 | 32 1813 PPC | KVM_REG_PPC_MAS6 | 32 1814 PPC | KVM_REG_PPC_MMUCFG | 32 1815 PPC | KVM_REG_PPC_TLB0CFG | 32 1816 PPC | KVM_REG_PPC_TLB1CFG | 32 1817 PPC | KVM_REG_PPC_TLB2CFG | 32 1818 PPC | KVM_REG_PPC_TLB3CFG | 32 1819 PPC | KVM_REG_PPC_TLB0PS | 32 1820 PPC | KVM_REG_PPC_TLB1PS | 32 1821 PPC | KVM_REG_PPC_TLB2PS | 32 1822 PPC | KVM_REG_PPC_TLB3PS | 32 1823 PPC | KVM_REG_PPC_EPTCFG | 32 1824 PPC | KVM_REG_PPC_ICP_STATE | 64 1825 PPC | KVM_REG_PPC_TB_OFFSET | 64 1826 PPC | KVM_REG_PPC_SPMC1 | 32 1827 PPC | KVM_REG_PPC_SPMC2 | 32 1828 PPC | KVM_REG_PPC_IAMR | 64 1829 PPC | KVM_REG_PPC_TFHAR | 64 1830 PPC | KVM_REG_PPC_TFIAR | 64 1831 PPC | KVM_REG_PPC_TEXASR | 64 1832 PPC | KVM_REG_PPC_FSCR | 64 1833 PPC | KVM_REG_PPC_PSPB | 32 1834 PPC | KVM_REG_PPC_EBBHR | 64 1835 PPC | KVM_REG_PPC_EBBRR | 64 1836 PPC | KVM_REG_PPC_BESCR | 64 1837 PPC | KVM_REG_PPC_TAR | 64 1838 PPC | KVM_REG_PPC_DPDES | 64 1839 PPC | KVM_REG_PPC_DAWR | 64 1840 PPC | KVM_REG_PPC_DAWRX | 64 1841 PPC | KVM_REG_PPC_CIABR | 64 1842 PPC | KVM_REG_PPC_IC | 64 1843 PPC | KVM_REG_PPC_VTB | 64 1844 PPC | KVM_REG_PPC_CSIGR | 64 1845 PPC | KVM_REG_PPC_TACR | 64 1846 PPC | KVM_REG_PPC_TCSCR | 64 1847 PPC | KVM_REG_PPC_PID | 64 1848 PPC | KVM_REG_PPC_ACOP | 64 1849 PPC | KVM_REG_PPC_VRSAVE | 32 1850 PPC | KVM_REG_PPC_LPCR | 32 1851 PPC | KVM_REG_PPC_LPCR_64 | 64 1852 PPC | KVM_REG_PPC_PPR | 64 1853 PPC | KVM_REG_PPC_ARCH_COMPAT | 32 1854 PPC | KVM_REG_PPC_DABRX | 32 1855 PPC | KVM_REG_PPC_WORT | 64 1856 PPC | KVM_REG_PPC_SPRG9 | 64 1857 PPC | KVM_REG_PPC_DBSR | 32 1858 PPC | KVM_REG_PPC_TIDR | 64 1859 PPC | KVM_REG_PPC_PSSCR | 64 1860 PPC | KVM_REG_PPC_DEC_EXPIRY | 64 1861 PPC | KVM_REG_PPC_TM_GPR0 | 64 1862 ... 1863 PPC | KVM_REG_PPC_TM_GPR31 | 64 1864 PPC | KVM_REG_PPC_TM_VSR0 | 128 1865 ... 1866 PPC | KVM_REG_PPC_TM_VSR63 | 128 1867 PPC | KVM_REG_PPC_TM_CR | 64 1868 PPC | KVM_REG_PPC_TM_LR | 64 1869 PPC | KVM_REG_PPC_TM_CTR | 64 1870 PPC | KVM_REG_PPC_TM_FPSCR | 64 1871 PPC | KVM_REG_PPC_TM_AMR | 64 1872 PPC | KVM_REG_PPC_TM_PPR | 64 1873 PPC | KVM_REG_PPC_TM_VRSAVE | 64 1874 PPC | KVM_REG_PPC_TM_VSCR | 32 1875 PPC | KVM_REG_PPC_TM_DSCR | 64 1876 PPC | KVM_REG_PPC_TM_TAR | 64 1877 PPC | KVM_REG_PPC_TM_XER | 64 1878 | | 1879 MIPS | KVM_REG_MIPS_R0 | 64 1880 ... 1881 MIPS | KVM_REG_MIPS_R31 | 64 1882 MIPS | KVM_REG_MIPS_HI | 64 1883 MIPS | KVM_REG_MIPS_LO | 64 1884 MIPS | KVM_REG_MIPS_PC | 64 1885 MIPS | KVM_REG_MIPS_CP0_INDEX | 32 1886 MIPS | KVM_REG_MIPS_CP0_ENTRYLO0 | 64 1887 MIPS | KVM_REG_MIPS_CP0_ENTRYLO1 | 64 1888 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64 1889 MIPS | KVM_REG_MIPS_CP0_CONTEXTCONFIG| 32 1890 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64 1891 MIPS | KVM_REG_MIPS_CP0_XCONTEXTCONFIG| 64 1892 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32 1893 MIPS | KVM_REG_MIPS_CP0_PAGEGRAIN | 32 1894 MIPS | KVM_REG_MIPS_CP0_SEGCTL0 | 64 1895 MIPS | KVM_REG_MIPS_CP0_SEGCTL1 | 64 1896 MIPS | KVM_REG_MIPS_CP0_SEGCTL2 | 64 1897 MIPS | KVM_REG_MIPS_CP0_PWBASE | 64 1898 MIPS | KVM_REG_MIPS_CP0_PWFIELD | 64 1899 MIPS | KVM_REG_MIPS_CP0_PWSIZE | 64 1900 MIPS | KVM_REG_MIPS_CP0_WIRED | 32 1901 MIPS | KVM_REG_MIPS_CP0_PWCTL | 32 1902 MIPS | KVM_REG_MIPS_CP0_HWRENA | 32 1903 MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64 1904 MIPS | KVM_REG_MIPS_CP0_BADINSTR | 32 1905 MIPS | KVM_REG_MIPS_CP0_BADINSTRP | 32 1906 MIPS | KVM_REG_MIPS_CP0_COUNT | 32 1907 MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64 1908 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32 1909 MIPS | KVM_REG_MIPS_CP0_STATUS | 32 1910 MIPS | KVM_REG_MIPS_CP0_INTCTL | 32 1911 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32 1912 MIPS | KVM_REG_MIPS_CP0_EPC | 64 1913 MIPS | KVM_REG_MIPS_CP0_PRID | 32 1914 MIPS | KVM_REG_MIPS_CP0_EBASE | 64 1915 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32 1916 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32 1917 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32 1918 MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32 1919 MIPS | KVM_REG_MIPS_CP0_CONFIG4 | 32 1920 MIPS | KVM_REG_MIPS_CP0_CONFIG5 | 32 1921 MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32 1922 MIPS | KVM_REG_MIPS_CP0_XCONTEXT | 64 1923 MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64 1924 MIPS | KVM_REG_MIPS_CP0_KSCRATCH1 | 64 1925 MIPS | KVM_REG_MIPS_CP0_KSCRATCH2 | 64 1926 MIPS | KVM_REG_MIPS_CP0_KSCRATCH3 | 64 1927 MIPS | KVM_REG_MIPS_CP0_KSCRATCH4 | 64 1928 MIPS | KVM_REG_MIPS_CP0_KSCRATCH5 | 64 1929 MIPS | KVM_REG_MIPS_CP0_KSCRATCH6 | 64 1930 MIPS | KVM_REG_MIPS_CP0_MAAR(0..63) | 64 1931 MIPS | KVM_REG_MIPS_COUNT_CTL | 64 1932 MIPS | KVM_REG_MIPS_COUNT_RESUME | 64 1933 MIPS | KVM_REG_MIPS_COUNT_HZ | 64 1934 MIPS | KVM_REG_MIPS_FPR_32(0..31) | 32 1935 MIPS | KVM_REG_MIPS_FPR_64(0..31) | 64 1936 MIPS | KVM_REG_MIPS_VEC_128(0..31) | 128 1937 MIPS | KVM_REG_MIPS_FCR_IR | 32 1938 MIPS | KVM_REG_MIPS_FCR_CSR | 32 1939 MIPS | KVM_REG_MIPS_MSA_IR | 32 1940 MIPS | KVM_REG_MIPS_MSA_CSR | 32 1941 1942 ARM registers are mapped using the lower 32 bits. The upper 16 of that 1943 is the register group type, or coprocessor number: 1944 1945 ARM core registers have the following id bit patterns: 1946 0x4020 0000 0010 <index into the kvm_regs struct:16> 1947 1948 ARM 32-bit CP15 registers have the following id bit patterns: 1949 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> 1950 1951 ARM 64-bit CP15 registers have the following id bit patterns: 1952 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> 1953 1954 ARM CCSIDR registers are demultiplexed by CSSELR value: 1955 0x4020 0000 0011 00 <csselr:8> 1956 1957 ARM 32-bit VFP control registers have the following id bit patterns: 1958 0x4020 0000 0012 1 <regno:12> 1959 1960 ARM 64-bit FP registers have the following id bit patterns: 1961 0x4030 0000 0012 0 <regno:12> 1962 1963 1964 arm64 registers are mapped using the lower 32 bits. The upper 16 of 1965 that is the register group type, or coprocessor number: 1966 1967 arm64 core/FP-SIMD registers have the following id bit patterns. Note 1968 that the size of the access is variable, as the kvm_regs structure 1969 contains elements ranging from 32 to 128 bits. The index is a 32bit 1970 value in the kvm_regs structure seen as a 32bit array. 1971 0x60x0 0000 0010 <index into the kvm_regs struct:16> 1972 1973 arm64 CCSIDR registers are demultiplexed by CSSELR value: 1974 0x6020 0000 0011 00 <csselr:8> 1975 1976 arm64 system registers have the following id bit patterns: 1977 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> 1978 1979 1980 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is 1981 the register group type: 1982 1983 MIPS core registers (see above) have the following id bit patterns: 1984 0x7030 0000 0000 <reg:16> 1985 1986 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit 1987 patterns depending on whether they're 32-bit or 64-bit registers: 1988 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 1989 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) 1990 1991 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 1992 versions of the EntryLo registers regardless of the word size of the host 1993 hardware, host kernel, guest, and whether XPA is present in the guest, i.e. 1994 with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and 1995 the PFNX field starting at bit 30. 1996 1997 MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit 1998 patterns: 1999 0x7030 0000 0001 01 <reg:8> 2000 2001 MIPS KVM control registers (see above) have the following id bit patterns: 2002 0x7030 0000 0002 <reg:16> 2003 2004 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following 2005 id bit patterns depending on the size of the register being accessed. They are 2006 always accessed according to the current guest FPU mode (Status.FR and 2007 Config5.FRE), i.e. as the guest would see them, and they become unpredictable 2008 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector 2009 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they 2010 overlap the FPU registers: 2011 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers) 2012 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers) 2013 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers) 2014 2015 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the 2016 following id bit patterns: 2017 0x7020 0000 0003 01 <0:3> <reg:5> 2018 2019 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the 2020 following id bit patterns: 2021 0x7020 0000 0003 02 <0:3> <reg:5> 2022 2023 2024 4.69 KVM_GET_ONE_REG 2025 2026 Capability: KVM_CAP_ONE_REG 2027 Architectures: all 2028 Type: vcpu ioctl 2029 Parameters: struct kvm_one_reg (in and out) 2030 Returns: 0 on success, negative value on failure 2031 2032 This ioctl allows to receive the value of a single register implemented 2033 in a vcpu. The register to read is indicated by the "id" field of the 2034 kvm_one_reg struct passed in. On success, the register value can be found 2035 at the memory location pointed to by "addr". 2036 2037 The list of registers accessible using this interface is identical to the 2038 list in 4.68. 2039 2040 2041 4.70 KVM_KVMCLOCK_CTRL 2042 2043 Capability: KVM_CAP_KVMCLOCK_CTRL 2044 Architectures: Any that implement pvclocks (currently x86 only) 2045 Type: vcpu ioctl 2046 Parameters: None 2047 Returns: 0 on success, -1 on error 2048 2049 This signals to the host kernel that the specified guest is being paused by 2050 userspace. The host will set a flag in the pvclock structure that is checked 2051 from the soft lockup watchdog. The flag is part of the pvclock structure that 2052 is shared between guest and host, specifically the second bit of the flags 2053 field of the pvclock_vcpu_time_info structure. It will be set exclusively by 2054 the host and read/cleared exclusively by the guest. The guest operation of 2055 checking and clearing the flag must an atomic operation so 2056 load-link/store-conditional, or equivalent must be used. There are two cases 2057 where the guest will clear the flag: when the soft lockup watchdog timer resets 2058 itself or when a soft lockup is detected. This ioctl can be called any time 2059 after pausing the vcpu, but before it is resumed. 2060 2061 2062 4.71 KVM_SIGNAL_MSI 2063 2064 Capability: KVM_CAP_SIGNAL_MSI 2065 Architectures: x86 arm arm64 2066 Type: vm ioctl 2067 Parameters: struct kvm_msi (in) 2068 Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error 2069 2070 Directly inject a MSI message. Only valid with in-kernel irqchip that handles 2071 MSI messages. 2072 2073 struct kvm_msi { 2074 __u32 address_lo; 2075 __u32 address_hi; 2076 __u32 data; 2077 __u32 flags; 2078 __u32 devid; 2079 __u8 pad[12]; 2080 }; 2081 2082 flags: KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM 2083 KVM_CAP_MSI_DEVID capability advertises the requirement to provide 2084 the device ID. If this capability is not available, userspace 2085 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail. 2086 2087 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier 2088 for the device that wrote the MSI message. For PCI, this is usually a 2089 BFD identifier in the lower 16 bits. 2090 2091 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS 2092 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, 2093 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of 2094 address_hi must be zero. 2095 2096 2097 4.71 KVM_CREATE_PIT2 2098 2099 Capability: KVM_CAP_PIT2 2100 Architectures: x86 2101 Type: vm ioctl 2102 Parameters: struct kvm_pit_config (in) 2103 Returns: 0 on success, -1 on error 2104 2105 Creates an in-kernel device model for the i8254 PIT. This call is only valid 2106 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following 2107 parameters have to be passed: 2108 2109 struct kvm_pit_config { 2110 __u32 flags; 2111 __u32 pad[15]; 2112 }; 2113 2114 Valid flags are: 2115 2116 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ 2117 2118 PIT timer interrupts may use a per-VM kernel thread for injection. If it 2119 exists, this thread will have a name of the following pattern: 2120 2121 kvm-pit/<owner-process-pid> 2122 2123 When running a guest with elevated priorities, the scheduling parameters of 2124 this thread may have to be adjusted accordingly. 2125 2126 This IOCTL replaces the obsolete KVM_CREATE_PIT. 2127 2128 2129 4.72 KVM_GET_PIT2 2130 2131 Capability: KVM_CAP_PIT_STATE2 2132 Architectures: x86 2133 Type: vm ioctl 2134 Parameters: struct kvm_pit_state2 (out) 2135 Returns: 0 on success, -1 on error 2136 2137 Retrieves the state of the in-kernel PIT model. Only valid after 2138 KVM_CREATE_PIT2. The state is returned in the following structure: 2139 2140 struct kvm_pit_state2 { 2141 struct kvm_pit_channel_state channels[3]; 2142 __u32 flags; 2143 __u32 reserved[9]; 2144 }; 2145 2146 Valid flags are: 2147 2148 /* disable PIT in HPET legacy mode */ 2149 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 2150 2151 This IOCTL replaces the obsolete KVM_GET_PIT. 2152 2153 2154 4.73 KVM_SET_PIT2 2155 2156 Capability: KVM_CAP_PIT_STATE2 2157 Architectures: x86 2158 Type: vm ioctl 2159 Parameters: struct kvm_pit_state2 (in) 2160 Returns: 0 on success, -1 on error 2161 2162 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. 2163 See KVM_GET_PIT2 for details on struct kvm_pit_state2. 2164 2165 This IOCTL replaces the obsolete KVM_SET_PIT. 2166 2167 2168 4.74 KVM_PPC_GET_SMMU_INFO 2169 2170 Capability: KVM_CAP_PPC_GET_SMMU_INFO 2171 Architectures: powerpc 2172 Type: vm ioctl 2173 Parameters: None 2174 Returns: 0 on success, -1 on error 2175 2176 This populates and returns a structure describing the features of 2177 the "Server" class MMU emulation supported by KVM. 2178 This can in turn be used by userspace to generate the appropriate 2179 device-tree properties for the guest operating system. 2180 2181 The structure contains some global information, followed by an 2182 array of supported segment page sizes: 2183 2184 struct kvm_ppc_smmu_info { 2185 __u64 flags; 2186 __u32 slb_size; 2187 __u32 pad; 2188 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; 2189 }; 2190 2191 The supported flags are: 2192 2193 - KVM_PPC_PAGE_SIZES_REAL: 2194 When that flag is set, guest page sizes must "fit" the backing 2195 store page sizes. When not set, any page size in the list can 2196 be used regardless of how they are backed by userspace. 2197 2198 - KVM_PPC_1T_SEGMENTS 2199 The emulated MMU supports 1T segments in addition to the 2200 standard 256M ones. 2201 2202 The "slb_size" field indicates how many SLB entries are supported 2203 2204 The "sps" array contains 8 entries indicating the supported base 2205 page sizes for a segment in increasing order. Each entry is defined 2206 as follow: 2207 2208 struct kvm_ppc_one_seg_page_size { 2209 __u32 page_shift; /* Base page shift of segment (or 0) */ 2210 __u32 slb_enc; /* SLB encoding for BookS */ 2211 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; 2212 }; 2213 2214 An entry with a "page_shift" of 0 is unused. Because the array is 2215 organized in increasing order, a lookup can stop when encoutering 2216 such an entry. 2217 2218 The "slb_enc" field provides the encoding to use in the SLB for the 2219 page size. The bits are in positions such as the value can directly 2220 be OR'ed into the "vsid" argument of the slbmte instruction. 2221 2222 The "enc" array is a list which for each of those segment base page 2223 size provides the list of supported actual page sizes (which can be 2224 only larger or equal to the base page size), along with the 2225 corresponding encoding in the hash PTE. Similarly, the array is 2226 8 entries sorted by increasing sizes and an entry with a "0" shift 2227 is an empty entry and a terminator: 2228 2229 struct kvm_ppc_one_page_size { 2230 __u32 page_shift; /* Page shift (or 0) */ 2231 __u32 pte_enc; /* Encoding in the HPTE (>>12) */ 2232 }; 2233 2234 The "pte_enc" field provides a value that can OR'ed into the hash 2235 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it 2236 into the hash PTE second double word). 2237 2238 4.75 KVM_IRQFD 2239 2240 Capability: KVM_CAP_IRQFD 2241 Architectures: x86 s390 arm arm64 2242 Type: vm ioctl 2243 Parameters: struct kvm_irqfd (in) 2244 Returns: 0 on success, -1 on error 2245 2246 Allows setting an eventfd to directly trigger a guest interrupt. 2247 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and 2248 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When 2249 an event is triggered on the eventfd, an interrupt is injected into 2250 the guest using the specified gsi pin. The irqfd is removed using 2251 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd 2252 and kvm_irqfd.gsi. 2253 2254 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify 2255 mechanism allowing emulation of level-triggered, irqfd-based 2256 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an 2257 additional eventfd in the kvm_irqfd.resamplefd field. When operating 2258 in resample mode, posting of an interrupt through kvm_irq.fd asserts 2259 the specified gsi in the irqchip. When the irqchip is resampled, such 2260 as from an EOI, the gsi is de-asserted and the user is notified via 2261 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue 2262 the interrupt if the device making use of it still requires service. 2263 Note that closing the resamplefd is not sufficient to disable the 2264 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment 2265 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. 2266 2267 On arm/arm64, gsi routing being supported, the following can happen: 2268 - in case no routing entry is associated to this gsi, injection fails 2269 - in case the gsi is associated to an irqchip routing entry, 2270 irqchip.pin + 32 corresponds to the injected SPI ID. 2271 - in case the gsi is associated to an MSI routing entry, the MSI 2272 message and device ID are translated into an LPI (support restricted 2273 to GICv3 ITS in-kernel emulation). 2274 2275 4.76 KVM_PPC_ALLOCATE_HTAB 2276 2277 Capability: KVM_CAP_PPC_ALLOC_HTAB 2278 Architectures: powerpc 2279 Type: vm ioctl 2280 Parameters: Pointer to u32 containing hash table order (in/out) 2281 Returns: 0 on success, -1 on error 2282 2283 This requests the host kernel to allocate an MMU hash table for a 2284 guest using the PAPR paravirtualization interface. This only does 2285 anything if the kernel is configured to use the Book 3S HV style of 2286 virtualization. Otherwise the capability doesn't exist and the ioctl 2287 returns an ENOTTY error. The rest of this description assumes Book 3S 2288 HV. 2289 2290 There must be no vcpus running when this ioctl is called; if there 2291 are, it will do nothing and return an EBUSY error. 2292 2293 The parameter is a pointer to a 32-bit unsigned integer variable 2294 containing the order (log base 2) of the desired size of the hash 2295 table, which must be between 18 and 46. On successful return from the 2296 ioctl, the value will not be changed by the kernel. 2297 2298 If no hash table has been allocated when any vcpu is asked to run 2299 (with the KVM_RUN ioctl), the host kernel will allocate a 2300 default-sized hash table (16 MB). 2301 2302 If this ioctl is called when a hash table has already been allocated, 2303 with a different order from the existing hash table, the existing hash 2304 table will be freed and a new one allocated. If this is ioctl is 2305 called when a hash table has already been allocated of the same order 2306 as specified, the kernel will clear out the existing hash table (zero 2307 all HPTEs). In either case, if the guest is using the virtualized 2308 real-mode area (VRMA) facility, the kernel will re-create the VMRA 2309 HPTEs on the next KVM_RUN of any vcpu. 2310 2311 4.77 KVM_S390_INTERRUPT 2312 2313 Capability: basic 2314 Architectures: s390 2315 Type: vm ioctl, vcpu ioctl 2316 Parameters: struct kvm_s390_interrupt (in) 2317 Returns: 0 on success, -1 on error 2318 2319 Allows to inject an interrupt to the guest. Interrupts can be floating 2320 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. 2321 2322 Interrupt parameters are passed via kvm_s390_interrupt: 2323 2324 struct kvm_s390_interrupt { 2325 __u32 type; 2326 __u32 parm; 2327 __u64 parm64; 2328 }; 2329 2330 type can be one of the following: 2331 2332 KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm 2333 KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm 2334 KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm 2335 KVM_S390_RESTART (vcpu) - restart 2336 KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt 2337 KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt 2338 KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt 2339 parameters in parm and parm64 2340 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm 2341 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm 2342 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm 2343 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an 2344 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); 2345 I/O interruption parameters in parm (subchannel) and parm64 (intparm, 2346 interruption subclass) 2347 KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm, 2348 machine check interrupt code in parm64 (note that 2349 machine checks needing further payload are not 2350 supported by this ioctl) 2351 2352 Note that the vcpu ioctl is asynchronous to vcpu execution. 2353 2354 4.78 KVM_PPC_GET_HTAB_FD 2355 2356 Capability: KVM_CAP_PPC_HTAB_FD 2357 Architectures: powerpc 2358 Type: vm ioctl 2359 Parameters: Pointer to struct kvm_get_htab_fd (in) 2360 Returns: file descriptor number (>= 0) on success, -1 on error 2361 2362 This returns a file descriptor that can be used either to read out the 2363 entries in the guest's hashed page table (HPT), or to write entries to 2364 initialize the HPT. The returned fd can only be written to if the 2365 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and 2366 can only be read if that bit is clear. The argument struct looks like 2367 this: 2368 2369 /* For KVM_PPC_GET_HTAB_FD */ 2370 struct kvm_get_htab_fd { 2371 __u64 flags; 2372 __u64 start_index; 2373 __u64 reserved[2]; 2374 }; 2375 2376 /* Values for kvm_get_htab_fd.flags */ 2377 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) 2378 #define KVM_GET_HTAB_WRITE ((__u64)0x2) 2379 2380 The `start_index' field gives the index in the HPT of the entry at 2381 which to start reading. It is ignored when writing. 2382 2383 Reads on the fd will initially supply information about all 2384 "interesting" HPT entries. Interesting entries are those with the 2385 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise 2386 all entries. When the end of the HPT is reached, the read() will 2387 return. If read() is called again on the fd, it will start again from 2388 the beginning of the HPT, but will only return HPT entries that have 2389 changed since they were last read. 2390 2391 Data read or written is structured as a header (8 bytes) followed by a 2392 series of valid HPT entries (16 bytes) each. The header indicates how 2393 many valid HPT entries there are and how many invalid entries follow 2394 the valid entries. The invalid entries are not represented explicitly 2395 in the stream. The header format is: 2396 2397 struct kvm_get_htab_header { 2398 __u32 index; 2399 __u16 n_valid; 2400 __u16 n_invalid; 2401 }; 2402 2403 Writes to the fd create HPT entries starting at the index given in the 2404 header; first `n_valid' valid entries with contents from the data 2405 written, then `n_invalid' invalid entries, invalidating any previously 2406 valid entries found. 2407 2408 4.79 KVM_CREATE_DEVICE 2409 2410 Capability: KVM_CAP_DEVICE_CTRL 2411 Type: vm ioctl 2412 Parameters: struct kvm_create_device (in/out) 2413 Returns: 0 on success, -1 on error 2414 Errors: 2415 ENODEV: The device type is unknown or unsupported 2416 EEXIST: Device already created, and this type of device may not 2417 be instantiated multiple times 2418 2419 Other error conditions may be defined by individual device types or 2420 have their standard meanings. 2421 2422 Creates an emulated device in the kernel. The file descriptor returned 2423 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. 2424 2425 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the 2426 device type is supported (not necessarily whether it can be created 2427 in the current vm). 2428 2429 Individual devices should not define flags. Attributes should be used 2430 for specifying any behavior that is not implied by the device type 2431 number. 2432 2433 struct kvm_create_device { 2434 __u32 type; /* in: KVM_DEV_TYPE_xxx */ 2435 __u32 fd; /* out: device handle */ 2436 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ 2437 }; 2438 2439 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 2440 2441 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 2442 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 2443 Type: device ioctl, vm ioctl, vcpu ioctl 2444 Parameters: struct kvm_device_attr 2445 Returns: 0 on success, -1 on error 2446 Errors: 2447 ENXIO: The group or attribute is unknown/unsupported for this device 2448 or hardware support is missing. 2449 EPERM: The attribute cannot (currently) be accessed this way 2450 (e.g. read-only attribute, or attribute that only makes 2451 sense when the device is in a different state) 2452 2453 Other error conditions may be defined by individual device types. 2454 2455 Gets/sets a specified piece of device configuration and/or state. The 2456 semantics are device-specific. See individual device documentation in 2457 the "devices" directory. As with ONE_REG, the size of the data 2458 transferred is defined by the particular attribute. 2459 2460 struct kvm_device_attr { 2461 __u32 flags; /* no flags currently defined */ 2462 __u32 group; /* device-defined */ 2463 __u64 attr; /* group-defined */ 2464 __u64 addr; /* userspace address of attr data */ 2465 }; 2466 2467 4.81 KVM_HAS_DEVICE_ATTR 2468 2469 Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device, 2470 KVM_CAP_VCPU_ATTRIBUTES for vcpu device 2471 Type: device ioctl, vm ioctl, vcpu ioctl 2472 Parameters: struct kvm_device_attr 2473 Returns: 0 on success, -1 on error 2474 Errors: 2475 ENXIO: The group or attribute is unknown/unsupported for this device 2476 or hardware support is missing. 2477 2478 Tests whether a device supports a particular attribute. A successful 2479 return indicates the attribute is implemented. It does not necessarily 2480 indicate that the attribute can be read or written in the device's 2481 current state. "addr" is ignored. 2482 2483 4.82 KVM_ARM_VCPU_INIT 2484 2485 Capability: basic 2486 Architectures: arm, arm64 2487 Type: vcpu ioctl 2488 Parameters: struct kvm_vcpu_init (in) 2489 Returns: 0 on success; -1 on error 2490 Errors: 2491 EINVAL: the target is unknown, or the combination of features is invalid. 2492 ENOENT: a features bit specified is unknown. 2493 2494 This tells KVM what type of CPU to present to the guest, and what 2495 optional features it should have. This will cause a reset of the cpu 2496 registers to their initial values. If this is not called, KVM_RUN will 2497 return ENOEXEC for that vcpu. 2498 2499 Note that because some registers reflect machine topology, all vcpus 2500 should be created before this ioctl is invoked. 2501 2502 Userspace can call this function multiple times for a given vcpu, including 2503 after the vcpu has been run. This will reset the vcpu to its initial 2504 state. All calls to this function after the initial call must use the same 2505 target and same set of feature flags, otherwise EINVAL will be returned. 2506 2507 Possible features: 2508 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. 2509 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on 2510 and execute guest code when KVM_RUN is called. 2511 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. 2512 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). 2513 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU. 2514 Depends on KVM_CAP_ARM_PSCI_0_2. 2515 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU. 2516 Depends on KVM_CAP_ARM_PMU_V3. 2517 2518 2519 4.83 KVM_ARM_PREFERRED_TARGET 2520 2521 Capability: basic 2522 Architectures: arm, arm64 2523 Type: vm ioctl 2524 Parameters: struct struct kvm_vcpu_init (out) 2525 Returns: 0 on success; -1 on error 2526 Errors: 2527 ENODEV: no preferred target available for the host 2528 2529 This queries KVM for preferred CPU target type which can be emulated 2530 by KVM on underlying host. 2531 2532 The ioctl returns struct kvm_vcpu_init instance containing information 2533 about preferred CPU target type and recommended features for it. The 2534 kvm_vcpu_init->features bitmap returned will have feature bits set if 2535 the preferred target recommends setting these features, but this is 2536 not mandatory. 2537 2538 The information returned by this ioctl can be used to prepare an instance 2539 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in 2540 in VCPU matching underlying host. 2541 2542 2543 4.84 KVM_GET_REG_LIST 2544 2545 Capability: basic 2546 Architectures: arm, arm64, mips 2547 Type: vcpu ioctl 2548 Parameters: struct kvm_reg_list (in/out) 2549 Returns: 0 on success; -1 on error 2550 Errors: 2551 E2BIG: the reg index list is too big to fit in the array specified by 2552 the user (the number required will be written into n). 2553 2554 struct kvm_reg_list { 2555 __u64 n; /* number of registers in reg[] */ 2556 __u64 reg[0]; 2557 }; 2558 2559 This ioctl returns the guest registers that are supported for the 2560 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 2561 2562 2563 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 2564 2565 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 2566 Architectures: arm, arm64 2567 Type: vm ioctl 2568 Parameters: struct kvm_arm_device_address (in) 2569 Returns: 0 on success, -1 on error 2570 Errors: 2571 ENODEV: The device id is unknown 2572 ENXIO: Device not supported on current system 2573 EEXIST: Address already set 2574 E2BIG: Address outside guest physical address space 2575 EBUSY: Address overlaps with other device range 2576 2577 struct kvm_arm_device_addr { 2578 __u64 id; 2579 __u64 addr; 2580 }; 2581 2582 Specify a device address in the guest's physical address space where guests 2583 can access emulated or directly exposed devices, which the host kernel needs 2584 to know about. The id field is an architecture specific identifier for a 2585 specific device. 2586 2587 ARM/arm64 divides the id field into two parts, a device id and an 2588 address type id specific to the individual device. 2589 2590 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | 2591 field: | 0x00000000 | device id | addr type id | 2592 2593 ARM/arm64 currently only require this when using the in-kernel GIC 2594 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 2595 as the device id. When setting the base address for the guest's 2596 mapping of the VGIC virtual CPU and distributor interface, the ioctl 2597 must be called after calling KVM_CREATE_IRQCHIP, but before calling 2598 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the 2599 base addresses will return -EEXIST. 2600 2601 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API 2602 should be used instead. 2603 2604 2605 4.86 KVM_PPC_RTAS_DEFINE_TOKEN 2606 2607 Capability: KVM_CAP_PPC_RTAS 2608 Architectures: ppc 2609 Type: vm ioctl 2610 Parameters: struct kvm_rtas_token_args 2611 Returns: 0 on success, -1 on error 2612 2613 Defines a token value for a RTAS (Run Time Abstraction Services) 2614 service in order to allow it to be handled in the kernel. The 2615 argument struct gives the name of the service, which must be the name 2616 of a service that has a kernel-side implementation. If the token 2617 value is non-zero, it will be associated with that service, and 2618 subsequent RTAS calls by the guest specifying that token will be 2619 handled by the kernel. If the token value is 0, then any token 2620 associated with the service will be forgotten, and subsequent RTAS 2621 calls by the guest for that service will be passed to userspace to be 2622 handled. 2623 2624 4.87 KVM_SET_GUEST_DEBUG 2625 2626 Capability: KVM_CAP_SET_GUEST_DEBUG 2627 Architectures: x86, s390, ppc, arm64 2628 Type: vcpu ioctl 2629 Parameters: struct kvm_guest_debug (in) 2630 Returns: 0 on success; -1 on error 2631 2632 struct kvm_guest_debug { 2633 __u32 control; 2634 __u32 pad; 2635 struct kvm_guest_debug_arch arch; 2636 }; 2637 2638 Set up the processor specific debug registers and configure vcpu for 2639 handling guest debug events. There are two parts to the structure, the 2640 first a control bitfield indicates the type of debug events to handle 2641 when running. Common control bits are: 2642 2643 - KVM_GUESTDBG_ENABLE: guest debugging is enabled 2644 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step 2645 2646 The top 16 bits of the control field are architecture specific control 2647 flags which can include the following: 2648 2649 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64] 2650 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390, arm64] 2651 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] 2652 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] 2653 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] 2654 2655 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints 2656 are enabled in memory so we need to ensure breakpoint exceptions are 2657 correctly trapped and the KVM run loop exits at the breakpoint and not 2658 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP 2659 we need to ensure the guest vCPUs architecture specific registers are 2660 updated to the correct (supplied) values. 2661 2662 The second part of the structure is architecture specific and 2663 typically contains a set of debug registers. 2664 2665 For arm64 the number of debug registers is implementation defined and 2666 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and 2667 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number 2668 indicating the number of supported registers. 2669 2670 When debug events exit the main run loop with the reason 2671 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run 2672 structure containing architecture specific debug information. 2673 2674 4.88 KVM_GET_EMULATED_CPUID 2675 2676 Capability: KVM_CAP_EXT_EMUL_CPUID 2677 Architectures: x86 2678 Type: system ioctl 2679 Parameters: struct kvm_cpuid2 (in/out) 2680 Returns: 0 on success, -1 on error 2681 2682 struct kvm_cpuid2 { 2683 __u32 nent; 2684 __u32 flags; 2685 struct kvm_cpuid_entry2 entries[0]; 2686 }; 2687 2688 The member 'flags' is used for passing flags from userspace. 2689 2690 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) 2691 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) 2692 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) 2693 2694 struct kvm_cpuid_entry2 { 2695 __u32 function; 2696 __u32 index; 2697 __u32 flags; 2698 __u32 eax; 2699 __u32 ebx; 2700 __u32 ecx; 2701 __u32 edx; 2702 __u32 padding[3]; 2703 }; 2704 2705 This ioctl returns x86 cpuid features which are emulated by 2706 kvm.Userspace can use the information returned by this ioctl to query 2707 which features are emulated by kvm instead of being present natively. 2708 2709 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 2710 structure with the 'nent' field indicating the number of entries in 2711 the variable-size array 'entries'. If the number of entries is too low 2712 to describe the cpu capabilities, an error (E2BIG) is returned. If the 2713 number is too high, the 'nent' field is adjusted and an error (ENOMEM) 2714 is returned. If the number is just right, the 'nent' field is adjusted 2715 to the number of valid entries in the 'entries' array, which is then 2716 filled. 2717 2718 The entries returned are the set CPUID bits of the respective features 2719 which kvm emulates, as returned by the CPUID instruction, with unknown 2720 or unsupported feature bits cleared. 2721 2722 Features like x2apic, for example, may not be present in the host cpu 2723 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be 2724 emulated efficiently and thus not included here. 2725 2726 The fields in each entry are defined as follows: 2727 2728 function: the eax value used to obtain the entry 2729 index: the ecx value used to obtain the entry (for entries that are 2730 affected by ecx) 2731 flags: an OR of zero or more of the following: 2732 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 2733 if the index field is valid 2734 KVM_CPUID_FLAG_STATEFUL_FUNC: 2735 if cpuid for this function returns different values for successive 2736 invocations; there will be several entries with the same function, 2737 all with this flag set 2738 KVM_CPUID_FLAG_STATE_READ_NEXT: 2739 for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is 2740 the first entry to be read by a cpu 2741 eax, ebx, ecx, edx: the values returned by the cpuid instruction for 2742 this function/index combination 2743 2744 4.89 KVM_S390_MEM_OP 2745 2746 Capability: KVM_CAP_S390_MEM_OP 2747 Architectures: s390 2748 Type: vcpu ioctl 2749 Parameters: struct kvm_s390_mem_op (in) 2750 Returns: = 0 on success, 2751 < 0 on generic error (e.g. -EFAULT or -ENOMEM), 2752 > 0 if an exception occurred while walking the page tables 2753 2754 Read or write data from/to the logical (virtual) memory of a VCPU. 2755 2756 Parameters are specified via the following structure: 2757 2758 struct kvm_s390_mem_op { 2759 __u64 gaddr; /* the guest address */ 2760 __u64 flags; /* flags */ 2761 __u32 size; /* amount of bytes */ 2762 __u32 op; /* type of operation */ 2763 __u64 buf; /* buffer in userspace */ 2764 __u8 ar; /* the access register number */ 2765 __u8 reserved[31]; /* should be set to 0 */ 2766 }; 2767 2768 The type of operation is specified in the "op" field. It is either 2769 KVM_S390_MEMOP_LOGICAL_READ for reading from logical memory space or 2770 KVM_S390_MEMOP_LOGICAL_WRITE for writing to logical memory space. The 2771 KVM_S390_MEMOP_F_CHECK_ONLY flag can be set in the "flags" field to check 2772 whether the corresponding memory access would create an access exception 2773 (without touching the data in the memory at the destination). In case an 2774 access exception occurred while walking the MMU tables of the guest, the 2775 ioctl returns a positive error number to indicate the type of exception. 2776 This exception is also raised directly at the corresponding VCPU if the 2777 flag KVM_S390_MEMOP_F_INJECT_EXCEPTION is set in the "flags" field. 2778 2779 The start address of the memory region has to be specified in the "gaddr" 2780 field, and the length of the region in the "size" field. "buf" is the buffer 2781 supplied by the userspace application where the read data should be written 2782 to for KVM_S390_MEMOP_LOGICAL_READ, or where the data that should be written 2783 is stored for a KVM_S390_MEMOP_LOGICAL_WRITE. "buf" is unused and can be NULL 2784 when KVM_S390_MEMOP_F_CHECK_ONLY is specified. "ar" designates the access 2785 register number to be used. 2786 2787 The "reserved" field is meant for future extensions. It is not used by 2788 KVM with the currently defined set of flags. 2789 2790 4.90 KVM_S390_GET_SKEYS 2791 2792 Capability: KVM_CAP_S390_SKEYS 2793 Architectures: s390 2794 Type: vm ioctl 2795 Parameters: struct kvm_s390_skeys 2796 Returns: 0 on success, KVM_S390_GET_KEYS_NONE if guest is not using storage 2797 keys, negative value on error 2798 2799 This ioctl is used to get guest storage key values on the s390 2800 architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 2801 2802 struct kvm_s390_skeys { 2803 __u64 start_gfn; 2804 __u64 count; 2805 __u64 skeydata_addr; 2806 __u32 flags; 2807 __u32 reserved[9]; 2808 }; 2809 2810 The start_gfn field is the number of the first guest frame whose storage keys 2811 you want to get. 2812 2813 The count field is the number of consecutive frames (starting from start_gfn) 2814 whose storage keys to get. The count field must be at least 1 and the maximum 2815 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range 2816 will cause the ioctl to return -EINVAL. 2817 2818 The skeydata_addr field is the address to a buffer large enough to hold count 2819 bytes. This buffer will be filled with storage key data by the ioctl. 2820 2821 4.91 KVM_S390_SET_SKEYS 2822 2823 Capability: KVM_CAP_S390_SKEYS 2824 Architectures: s390 2825 Type: vm ioctl 2826 Parameters: struct kvm_s390_skeys 2827 Returns: 0 on success, negative value on error 2828 2829 This ioctl is used to set guest storage key values on the s390 2830 architecture. The ioctl takes parameters via the kvm_s390_skeys struct. 2831 See section on KVM_S390_GET_SKEYS for struct definition. 2832 2833 The start_gfn field is the number of the first guest frame whose storage keys 2834 you want to set. 2835 2836 The count field is the number of consecutive frames (starting from start_gfn) 2837 whose storage keys to get. The count field must be at least 1 and the maximum 2838 allowed value is defined as KVM_S390_SKEYS_ALLOC_MAX. Values outside this range 2839 will cause the ioctl to return -EINVAL. 2840 2841 The skeydata_addr field is the address to a buffer containing count bytes of 2842 storage keys. Each byte in the buffer will be set as the storage key for a 2843 single frame starting at start_gfn for count frames. 2844 2845 Note: If any architecturally invalid key value is found in the given data then 2846 the ioctl will return -EINVAL. 2847 2848 4.92 KVM_S390_IRQ 2849 2850 Capability: KVM_CAP_S390_INJECT_IRQ 2851 Architectures: s390 2852 Type: vcpu ioctl 2853 Parameters: struct kvm_s390_irq (in) 2854 Returns: 0 on success, -1 on error 2855 Errors: 2856 EINVAL: interrupt type is invalid 2857 type is KVM_S390_SIGP_STOP and flag parameter is invalid value 2858 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger 2859 than the maximum of VCPUs 2860 EBUSY: type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped 2861 type is KVM_S390_SIGP_STOP and a stop irq is already pending 2862 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt 2863 is already pending 2864 2865 Allows to inject an interrupt to the guest. 2866 2867 Using struct kvm_s390_irq as a parameter allows 2868 to inject additional payload which is not 2869 possible via KVM_S390_INTERRUPT. 2870 2871 Interrupt parameters are passed via kvm_s390_irq: 2872 2873 struct kvm_s390_irq { 2874 __u64 type; 2875 union { 2876 struct kvm_s390_io_info io; 2877 struct kvm_s390_ext_info ext; 2878 struct kvm_s390_pgm_info pgm; 2879 struct kvm_s390_emerg_info emerg; 2880 struct kvm_s390_extcall_info extcall; 2881 struct kvm_s390_prefix_info prefix; 2882 struct kvm_s390_stop_info stop; 2883 struct kvm_s390_mchk_info mchk; 2884 char reserved[64]; 2885 } u; 2886 }; 2887 2888 type can be one of the following: 2889 2890 KVM_S390_SIGP_STOP - sigp stop; parameter in .stop 2891 KVM_S390_PROGRAM_INT - program check; parameters in .pgm 2892 KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix 2893 KVM_S390_RESTART - restart; no parameters 2894 KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters 2895 KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters 2896 KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg 2897 KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall 2898 KVM_S390_MCHK - machine check interrupt; parameters in .mchk 2899 2900 2901 Note that the vcpu ioctl is asynchronous to vcpu execution. 2902 2903 4.94 KVM_S390_GET_IRQ_STATE 2904 2905 Capability: KVM_CAP_S390_IRQ_STATE 2906 Architectures: s390 2907 Type: vcpu ioctl 2908 Parameters: struct kvm_s390_irq_state (out) 2909 Returns: >= number of bytes copied into buffer, 2910 -EINVAL if buffer size is 0, 2911 -ENOBUFS if buffer size is too small to fit all pending interrupts, 2912 -EFAULT if the buffer address was invalid 2913 2914 This ioctl allows userspace to retrieve the complete state of all currently 2915 pending interrupts in a single buffer. Use cases include migration 2916 and introspection. The parameter structure contains the address of a 2917 userspace buffer and its length: 2918 2919 struct kvm_s390_irq_state { 2920 __u64 buf; 2921 __u32 flags; /* will stay unused for compatibility reasons */ 2922 __u32 len; 2923 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 2924 }; 2925 2926 Userspace passes in the above struct and for each pending interrupt a 2927 struct kvm_s390_irq is copied to the provided buffer. 2928 2929 The structure contains a flags and a reserved field for future extensions. As 2930 the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and 2931 reserved, these fields can not be used in the future without breaking 2932 compatibility. 2933 2934 If -ENOBUFS is returned the buffer provided was too small and userspace 2935 may retry with a bigger buffer. 2936 2937 4.95 KVM_S390_SET_IRQ_STATE 2938 2939 Capability: KVM_CAP_S390_IRQ_STATE 2940 Architectures: s390 2941 Type: vcpu ioctl 2942 Parameters: struct kvm_s390_irq_state (in) 2943 Returns: 0 on success, 2944 -EFAULT if the buffer address was invalid, 2945 -EINVAL for an invalid buffer length (see below), 2946 -EBUSY if there were already interrupts pending, 2947 errors occurring when actually injecting the 2948 interrupt. See KVM_S390_IRQ. 2949 2950 This ioctl allows userspace to set the complete state of all cpu-local 2951 interrupts currently pending for the vcpu. It is intended for restoring 2952 interrupt state after a migration. The input parameter is a userspace buffer 2953 containing a struct kvm_s390_irq_state: 2954 2955 struct kvm_s390_irq_state { 2956 __u64 buf; 2957 __u32 flags; /* will stay unused for compatibility reasons */ 2958 __u32 len; 2959 __u32 reserved[4]; /* will stay unused for compatibility reasons */ 2960 }; 2961 2962 The restrictions for flags and reserved apply as well. 2963 (see KVM_S390_GET_IRQ_STATE) 2964 2965 The userspace memory referenced by buf contains a struct kvm_s390_irq 2966 for each interrupt to be injected into the guest. 2967 If one of the interrupts could not be injected for some reason the 2968 ioctl aborts. 2969 2970 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0 2971 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq), 2972 which is the maximum number of possibly pending cpu-local interrupts. 2973 2974 4.96 KVM_SMI 2975 2976 Capability: KVM_CAP_X86_SMM 2977 Architectures: x86 2978 Type: vcpu ioctl 2979 Parameters: none 2980 Returns: 0 on success, -1 on error 2981 2982 Queues an SMI on the thread's vcpu. 2983 2984 4.97 KVM_CAP_PPC_MULTITCE 2985 2986 Capability: KVM_CAP_PPC_MULTITCE 2987 Architectures: ppc 2988 Type: vm 2989 2990 This capability means the kernel is capable of handling hypercalls 2991 H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user 2992 space. This significantly accelerates DMA operations for PPC KVM guests. 2993 User space should expect that its handlers for these hypercalls 2994 are not going to be called if user space previously registered LIOBN 2995 in KVM (via KVM_CREATE_SPAPR_TCE or similar calls). 2996 2997 In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest, 2998 user space might have to advertise it for the guest. For example, 2999 IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is 3000 present in the "ibm,hypertas-functions" device-tree property. 3001 3002 The hypercalls mentioned above may or may not be processed successfully 3003 in the kernel based fast path. If they can not be handled by the kernel, 3004 they will get passed on to user space. So user space still has to have 3005 an implementation for these despite the in kernel acceleration. 3006 3007 This capability is always enabled. 3008 3009 4.98 KVM_CREATE_SPAPR_TCE_64 3010 3011 Capability: KVM_CAP_SPAPR_TCE_64 3012 Architectures: powerpc 3013 Type: vm ioctl 3014 Parameters: struct kvm_create_spapr_tce_64 (in) 3015 Returns: file descriptor for manipulating the created TCE table 3016 3017 This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit 3018 windows, described in 4.62 KVM_CREATE_SPAPR_TCE 3019 3020 This capability uses extended struct in ioctl interface: 3021 3022 /* for KVM_CAP_SPAPR_TCE_64 */ 3023 struct kvm_create_spapr_tce_64 { 3024 __u64 liobn; 3025 __u32 page_shift; 3026 __u32 flags; 3027 __u64 offset; /* in pages */ 3028 __u64 size; /* in pages */ 3029 }; 3030 3031 The aim of extension is to support an additional bigger DMA window with 3032 a variable page size. 3033 KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and 3034 a bus offset of the corresponding DMA window, @size and @offset are numbers 3035 of IOMMU pages. 3036 3037 @flags are not used at the moment. 3038 3039 The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. 3040 3041 4.99 KVM_REINJECT_CONTROL 3042 3043 Capability: KVM_CAP_REINJECT_CONTROL 3044 Architectures: x86 3045 Type: vm ioctl 3046 Parameters: struct kvm_reinject_control (in) 3047 Returns: 0 on success, 3048 -EFAULT if struct kvm_reinject_control cannot be read, 3049 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier. 3050 3051 i8254 (PIT) has two modes, reinject and !reinject. The default is reinject, 3052 where KVM queues elapsed i8254 ticks and monitors completion of interrupt from 3053 vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its 3054 interrupt whenever there isn't a pending interrupt from i8254. 3055 !reinject mode injects an interrupt as soon as a tick arrives. 3056 3057 struct kvm_reinject_control { 3058 __u8 pit_reinject; 3059 __u8 reserved[31]; 3060 }; 3061 3062 pit_reinject = 0 (!reinject mode) is recommended, unless running an old 3063 operating system that uses the PIT for timing (e.g. Linux 2.4.x). 3064 3065 4.100 KVM_PPC_CONFIGURE_V3_MMU 3066 3067 Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 3068 Architectures: ppc 3069 Type: vm ioctl 3070 Parameters: struct kvm_ppc_mmuv3_cfg (in) 3071 Returns: 0 on success, 3072 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read, 3073 -EINVAL if the configuration is invalid 3074 3075 This ioctl controls whether the guest will use radix or HPT (hashed 3076 page table) translation, and sets the pointer to the process table for 3077 the guest. 3078 3079 struct kvm_ppc_mmuv3_cfg { 3080 __u64 flags; 3081 __u64 process_table; 3082 }; 3083 3084 There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and 3085 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest 3086 to use radix tree translation, and if clear, to use HPT translation. 3087 KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest 3088 to be able to use the global TLB and SLB invalidation instructions; 3089 if clear, the guest may not use these instructions. 3090 3091 The process_table field specifies the address and size of the guest 3092 process table, which is in the guest's space. This field is formatted 3093 as the second doubleword of the partition table entry, as defined in 3094 the Power ISA V3.00, Book III section 5.7.6.1. 3095 3096 4.101 KVM_PPC_GET_RMMU_INFO 3097 3098 Capability: KVM_CAP_PPC_RADIX_MMU 3099 Architectures: ppc 3100 Type: vm ioctl 3101 Parameters: struct kvm_ppc_rmmu_info (out) 3102 Returns: 0 on success, 3103 -EFAULT if struct kvm_ppc_rmmu_info cannot be written, 3104 -EINVAL if no useful information can be returned 3105 3106 This ioctl returns a structure containing two things: (a) a list 3107 containing supported radix tree geometries, and (b) a list that maps 3108 page sizes to put in the "AP" (actual page size) field for the tlbie 3109 (TLB invalidate entry) instruction. 3110 3111 struct kvm_ppc_rmmu_info { 3112 struct kvm_ppc_radix_geom { 3113 __u8 page_shift; 3114 __u8 level_bits[4]; 3115 __u8 pad[3]; 3116 } geometries[8]; 3117 __u32 ap_encodings[8]; 3118 }; 3119 3120 The geometries[] field gives up to 8 supported geometries for the 3121 radix page table, in terms of the log base 2 of the smallest page 3122 size, and the number of bits indexed at each level of the tree, from 3123 the PTE level up to the PGD level in that order. Any unused entries 3124 will have 0 in the page_shift field. 3125 3126 The ap_encodings gives the supported page sizes and their AP field 3127 encodings, encoded with the AP value in the top 3 bits and the log 3128 base 2 of the page size in the bottom 6 bits. 3129 3130 4.102 KVM_PPC_RESIZE_HPT_PREPARE 3131 3132 Capability: KVM_CAP_SPAPR_RESIZE_HPT 3133 Architectures: powerpc 3134 Type: vm ioctl 3135 Parameters: struct kvm_ppc_resize_hpt (in) 3136 Returns: 0 on successful completion, 3137 >0 if a new HPT is being prepared, the value is an estimated 3138 number of milliseconds until preparation is complete 3139 -EFAULT if struct kvm_reinject_control cannot be read, 3140 -EINVAL if the supplied shift or flags are invalid 3141 -ENOMEM if unable to allocate the new HPT 3142 -ENOSPC if there was a hash collision when moving existing 3143 HPT entries to the new HPT 3144 -EIO on other error conditions 3145 3146 Used to implement the PAPR extension for runtime resizing of a guest's 3147 Hashed Page Table (HPT). Specifically this starts, stops or monitors 3148 the preparation of a new potential HPT for the guest, essentially 3149 implementing the H_RESIZE_HPT_PREPARE hypercall. 3150 3151 If called with shift > 0 when there is no pending HPT for the guest, 3152 this begins preparation of a new pending HPT of size 2^(shift) bytes. 3153 It then returns a positive integer with the estimated number of 3154 milliseconds until preparation is complete. 3155 3156 If called when there is a pending HPT whose size does not match that 3157 requested in the parameters, discards the existing pending HPT and 3158 creates a new one as above. 3159 3160 If called when there is a pending HPT of the size requested, will: 3161 * If preparation of the pending HPT is already complete, return 0 3162 * If preparation of the pending HPT has failed, return an error 3163 code, then discard the pending HPT. 3164 * If preparation of the pending HPT is still in progress, return an 3165 estimated number of milliseconds until preparation is complete. 3166 3167 If called with shift == 0, discards any currently pending HPT and 3168 returns 0 (i.e. cancels any in-progress preparation). 3169 3170 flags is reserved for future expansion, currently setting any bits in 3171 flags will result in an -EINVAL. 3172 3173 Normally this will be called repeatedly with the same parameters until 3174 it returns <= 0. The first call will initiate preparation, subsequent 3175 ones will monitor preparation until it completes or fails. 3176 3177 struct kvm_ppc_resize_hpt { 3178 __u64 flags; 3179 __u32 shift; 3180 __u32 pad; 3181 }; 3182 3183 4.103 KVM_PPC_RESIZE_HPT_COMMIT 3184 3185 Capability: KVM_CAP_SPAPR_RESIZE_HPT 3186 Architectures: powerpc 3187 Type: vm ioctl 3188 Parameters: struct kvm_ppc_resize_hpt (in) 3189 Returns: 0 on successful completion, 3190 -EFAULT if struct kvm_reinject_control cannot be read, 3191 -EINVAL if the supplied shift or flags are invalid 3192 -ENXIO is there is no pending HPT, or the pending HPT doesn't 3193 have the requested size 3194 -EBUSY if the pending HPT is not fully prepared 3195 -ENOSPC if there was a hash collision when moving existing 3196 HPT entries to the new HPT 3197 -EIO on other error conditions 3198 3199 Used to implement the PAPR extension for runtime resizing of a guest's 3200 Hashed Page Table (HPT). Specifically this requests that the guest be 3201 transferred to working with the new HPT, essentially implementing the 3202 H_RESIZE_HPT_COMMIT hypercall. 3203 3204 This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has 3205 returned 0 with the same parameters. In other cases 3206 KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or 3207 -EBUSY, though others may be possible if the preparation was started, 3208 but failed). 3209 3210 This will have undefined effects on the guest if it has not already 3211 placed itself in a quiescent state where no vcpu will make MMU enabled 3212 memory accesses. 3213 3214 On succsful completion, the pending HPT will become the guest's active 3215 HPT and the previous HPT will be discarded. 3216 3217 On failure, the guest will still be operating on its previous HPT. 3218 3219 struct kvm_ppc_resize_hpt { 3220 __u64 flags; 3221 __u32 shift; 3222 __u32 pad; 3223 }; 3224 3225 4.104 KVM_X86_GET_MCE_CAP_SUPPORTED 3226 3227 Capability: KVM_CAP_MCE 3228 Architectures: x86 3229 Type: system ioctl 3230 Parameters: u64 mce_cap (out) 3231 Returns: 0 on success, -1 on error 3232 3233 Returns supported MCE capabilities. The u64 mce_cap parameter 3234 has the same format as the MSR_IA32_MCG_CAP register. Supported 3235 capabilities will have the corresponding bits set. 3236 3237 4.105 KVM_X86_SETUP_MCE 3238 3239 Capability: KVM_CAP_MCE 3240 Architectures: x86 3241 Type: vcpu ioctl 3242 Parameters: u64 mcg_cap (in) 3243 Returns: 0 on success, 3244 -EFAULT if u64 mcg_cap cannot be read, 3245 -EINVAL if the requested number of banks is invalid, 3246 -EINVAL if requested MCE capability is not supported. 3247 3248 Initializes MCE support for use. The u64 mcg_cap parameter 3249 has the same format as the MSR_IA32_MCG_CAP register and 3250 specifies which capabilities should be enabled. The maximum 3251 supported number of error-reporting banks can be retrieved when 3252 checking for KVM_CAP_MCE. The supported capabilities can be 3253 retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 3254 3255 4.106 KVM_X86_SET_MCE 3256 3257 Capability: KVM_CAP_MCE 3258 Architectures: x86 3259 Type: vcpu ioctl 3260 Parameters: struct kvm_x86_mce (in) 3261 Returns: 0 on success, 3262 -EFAULT if struct kvm_x86_mce cannot be read, 3263 -EINVAL if the bank number is invalid, 3264 -EINVAL if VAL bit is not set in status field. 3265 3266 Inject a machine check error (MCE) into the guest. The input 3267 parameter is: 3268 3269 struct kvm_x86_mce { 3270 __u64 status; 3271 __u64 addr; 3272 __u64 misc; 3273 __u64 mcg_status; 3274 __u8 bank; 3275 __u8 pad1[7]; 3276 __u64 pad2[3]; 3277 }; 3278 3279 If the MCE being reported is an uncorrected error, KVM will 3280 inject it as an MCE exception into the guest. If the guest 3281 MCG_STATUS register reports that an MCE is in progress, KVM 3282 causes an KVM_EXIT_SHUTDOWN vmexit. 3283 3284 Otherwise, if the MCE is a corrected error, KVM will just 3285 store it in the corresponding bank (provided this bank is 3286 not holding a previously reported uncorrected error). 3287 3288 4.107 KVM_S390_GET_CMMA_BITS 3289 3290 Capability: KVM_CAP_S390_CMMA_MIGRATION 3291 Architectures: s390 3292 Type: vm ioctl 3293 Parameters: struct kvm_s390_cmma_log (in, out) 3294 Returns: 0 on success, a negative value on error 3295 3296 This ioctl is used to get the values of the CMMA bits on the s390 3297 architecture. It is meant to be used in two scenarios: 3298 - During live migration to save the CMMA values. Live migration needs 3299 to be enabled via the KVM_REQ_START_MIGRATION VM property. 3300 - To non-destructively peek at the CMMA values, with the flag 3301 KVM_S390_CMMA_PEEK set. 3302 3303 The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired 3304 values are written to a buffer whose location is indicated via the "values" 3305 member in the kvm_s390_cmma_log struct. The values in the input struct are 3306 also updated as needed. 3307 Each CMMA value takes up one byte. 3308 3309 struct kvm_s390_cmma_log { 3310 __u64 start_gfn; 3311 __u32 count; 3312 __u32 flags; 3313 union { 3314 __u64 remaining; 3315 __u64 mask; 3316 }; 3317 __u64 values; 3318 }; 3319 3320 start_gfn is the number of the first guest frame whose CMMA values are 3321 to be retrieved, 3322 3323 count is the length of the buffer in bytes, 3324 3325 values points to the buffer where the result will be written to. 3326 3327 If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be 3328 KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with 3329 other ioctls. 3330 3331 The result is written in the buffer pointed to by the field values, and 3332 the values of the input parameter are updated as follows. 3333 3334 Depending on the flags, different actions are performed. The only 3335 supported flag so far is KVM_S390_CMMA_PEEK. 3336 3337 The default behaviour if KVM_S390_CMMA_PEEK is not set is: 3338 start_gfn will indicate the first page frame whose CMMA bits were dirty. 3339 It is not necessarily the same as the one passed as input, as clean pages 3340 are skipped. 3341 3342 count will indicate the number of bytes actually written in the buffer. 3343 It can (and very often will) be smaller than the input value, since the 3344 buffer is only filled until 16 bytes of clean values are found (which 3345 are then not copied in the buffer). Since a CMMA migration block needs 3346 the base address and the length, for a total of 16 bytes, we will send 3347 back some clean data if there is some dirty data afterwards, as long as 3348 the size of the clean data does not exceed the size of the header. This 3349 allows to minimize the amount of data to be saved or transferred over 3350 the network at the expense of more roundtrips to userspace. The next 3351 invocation of the ioctl will skip over all the clean values, saving 3352 potentially more than just the 16 bytes we found. 3353 3354 If KVM_S390_CMMA_PEEK is set: 3355 the existing storage attributes are read even when not in migration 3356 mode, and no other action is performed; 3357 3358 the output start_gfn will be equal to the input start_gfn, 3359 3360 the output count will be equal to the input count, except if the end of 3361 memory has been reached. 3362 3363 In both cases: 3364 the field "remaining" will indicate the total number of dirty CMMA values 3365 still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is 3366 not enabled. 3367 3368 mask is unused. 3369 3370 values points to the userspace buffer where the result will be stored. 3371 3372 This ioctl can fail with -ENOMEM if not enough memory can be allocated to 3373 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 3374 KVM_S390_CMMA_PEEK is not set but migration mode was not enabled, with 3375 -EFAULT if the userspace address is invalid or if no page table is 3376 present for the addresses (e.g. when using hugepages). 3377 3378 4.108 KVM_S390_SET_CMMA_BITS 3379 3380 Capability: KVM_CAP_S390_CMMA_MIGRATION 3381 Architectures: s390 3382 Type: vm ioctl 3383 Parameters: struct kvm_s390_cmma_log (in) 3384 Returns: 0 on success, a negative value on error 3385 3386 This ioctl is used to set the values of the CMMA bits on the s390 3387 architecture. It is meant to be used during live migration to restore 3388 the CMMA values, but there are no restrictions on its use. 3389 The ioctl takes parameters via the kvm_s390_cmma_values struct. 3390 Each CMMA value takes up one byte. 3391 3392 struct kvm_s390_cmma_log { 3393 __u64 start_gfn; 3394 __u32 count; 3395 __u32 flags; 3396 union { 3397 __u64 remaining; 3398 __u64 mask; 3399 }; 3400 __u64 values; 3401 }; 3402 3403 start_gfn indicates the starting guest frame number, 3404 3405 count indicates how many values are to be considered in the buffer, 3406 3407 flags is not used and must be 0. 3408 3409 mask indicates which PGSTE bits are to be considered. 3410 3411 remaining is not used. 3412 3413 values points to the buffer in userspace where to store the values. 3414 3415 This ioctl can fail with -ENOMEM if not enough memory can be allocated to 3416 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if 3417 the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or 3418 if the flags field was not 0, with -EFAULT if the userspace address is 3419 invalid, if invalid pages are written to (e.g. after the end of memory) 3420 or if no page table is present for the addresses (e.g. when using 3421 hugepages). 3422 3423 4.109 KVM_PPC_GET_CPU_CHAR 3424 3425 Capability: KVM_CAP_PPC_GET_CPU_CHAR 3426 Architectures: powerpc 3427 Type: vm ioctl 3428 Parameters: struct kvm_ppc_cpu_char (out) 3429 Returns: 0 on successful completion 3430 -EFAULT if struct kvm_ppc_cpu_char cannot be written 3431 3432 This ioctl gives userspace information about certain characteristics 3433 of the CPU relating to speculative execution of instructions and 3434 possible information leakage resulting from speculative execution (see 3435 CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is 3436 returned in struct kvm_ppc_cpu_char, which looks like this: 3437 3438 struct kvm_ppc_cpu_char { 3439 __u64 character; /* characteristics of the CPU */ 3440 __u64 behaviour; /* recommended software behaviour */ 3441 __u64 character_mask; /* valid bits in character */ 3442 __u64 behaviour_mask; /* valid bits in behaviour */ 3443 }; 3444 3445 For extensibility, the character_mask and behaviour_mask fields 3446 indicate which bits of character and behaviour have been filled in by 3447 the kernel. If the set of defined bits is extended in future then 3448 userspace will be able to tell whether it is running on a kernel that 3449 knows about the new bits. 3450 3451 The character field describes attributes of the CPU which can help 3452 with preventing inadvertent information disclosure - specifically, 3453 whether there is an instruction to flash-invalidate the L1 data cache 3454 (ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set 3455 to a mode where entries can only be used by the thread that created 3456 them, whether the bcctr[l] instruction prevents speculation, and 3457 whether a speculation barrier instruction (ori 31,31,0) is provided. 3458 3459 The behaviour field describes actions that software should take to 3460 prevent inadvertent information disclosure, and thus describes which 3461 vulnerabilities the hardware is subject to; specifically whether the 3462 L1 data cache should be flushed when returning to user mode from the 3463 kernel, and whether a speculation barrier should be placed between an 3464 array bounds check and the array access. 3465 3466 These fields use the same bit definitions as the new 3467 H_GET_CPU_CHARACTERISTICS hypercall. 3468 3469 4.110 KVM_MEMORY_ENCRYPT_OP 3470 3471 Capability: basic 3472 Architectures: x86 3473 Type: system 3474 Parameters: an opaque platform specific structure (in/out) 3475 Returns: 0 on success; -1 on error 3476 3477 If the platform supports creating encrypted VMs then this ioctl can be used 3478 for issuing platform-specific memory encryption commands to manage those 3479 encrypted VMs. 3480 3481 Currently, this ioctl is used for issuing Secure Encrypted Virtualization 3482 (SEV) commands on AMD Processors. The SEV commands are defined in 3483 Documentation/virtual/kvm/amd-memory-encryption.txt. 3484 3485 4.111 KVM_MEMORY_ENCRYPT_REG_REGION 3486 3487 Capability: basic 3488 Architectures: x86 3489 Type: system 3490 Parameters: struct kvm_enc_region (in) 3491 Returns: 0 on success; -1 on error 3492 3493 This ioctl can be used to register a guest memory region which may 3494 contain encrypted data (e.g. guest RAM, SMRAM etc). 3495 3496 It is used in the SEV-enabled guest. When encryption is enabled, a guest 3497 memory region may contain encrypted data. The SEV memory encryption 3498 engine uses a tweak such that two identical plaintext pages, each at 3499 different locations will have differing ciphertexts. So swapping or 3500 moving ciphertext of those pages will not result in plaintext being 3501 swapped. So relocating (or migrating) physical backing pages for the SEV 3502 guest will require some additional steps. 3503 3504 Note: The current SEV key management spec does not provide commands to 3505 swap or migrate (move) ciphertext pages. Hence, for now we pin the guest 3506 memory region registered with the ioctl. 3507 3508 4.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 3509 3510 Capability: basic 3511 Architectures: x86 3512 Type: system 3513 Parameters: struct kvm_enc_region (in) 3514 Returns: 0 on success; -1 on error 3515 3516 This ioctl can be used to unregister the guest memory region registered 3517 with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above. 3518 3519 3520 5. The kvm_run structure 3521 ------------------------ 3522 3523 Application code obtains a pointer to the kvm_run structure by 3524 mmap()ing a vcpu fd. From that point, application code can control 3525 execution by changing fields in kvm_run prior to calling the KVM_RUN 3526 ioctl, and obtain information about the reason KVM_RUN returned by 3527 looking up structure members. 3528 3529 struct kvm_run { 3530 /* in */ 3531 __u8 request_interrupt_window; 3532 3533 Request that KVM_RUN return when it becomes possible to inject external 3534 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. 3535 3536 __u8 immediate_exit; 3537 3538 This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN 3539 exits immediately, returning -EINTR. In the common scenario where a 3540 signal is used to "kick" a VCPU out of KVM_RUN, this field can be used 3541 to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. 3542 Rather than blocking the signal outside KVM_RUN, userspace can set up 3543 a signal handler that sets run->immediate_exit to a non-zero value. 3544 3545 This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. 3546 3547 __u8 padding1[6]; 3548 3549 /* out */ 3550 __u32 exit_reason; 3551 3552 When KVM_RUN has returned successfully (return value 0), this informs 3553 application code why KVM_RUN has returned. Allowable values for this 3554 field are detailed below. 3555 3556 __u8 ready_for_interrupt_injection; 3557 3558 If request_interrupt_window has been specified, this field indicates 3559 an interrupt can be injected now with KVM_INTERRUPT. 3560 3561 __u8 if_flag; 3562 3563 The value of the current interrupt flag. Only valid if in-kernel 3564 local APIC is not used. 3565 3566 __u16 flags; 3567 3568 More architecture-specific flags detailing state of the VCPU that may 3569 affect the device's behavior. The only currently defined flag is 3570 KVM_RUN_X86_SMM, which is valid on x86 machines and is set if the 3571 VCPU is in system management mode. 3572 3573 /* in (pre_kvm_run), out (post_kvm_run) */ 3574 __u64 cr8; 3575 3576 The value of the cr8 register. Only valid if in-kernel local APIC is 3577 not used. Both input and output. 3578 3579 __u64 apic_base; 3580 3581 The value of the APIC BASE msr. Only valid if in-kernel local 3582 APIC is not used. Both input and output. 3583 3584 union { 3585 /* KVM_EXIT_UNKNOWN */ 3586 struct { 3587 __u64 hardware_exit_reason; 3588 } hw; 3589 3590 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown 3591 reasons. Further architecture-specific information is available in 3592 hardware_exit_reason. 3593 3594 /* KVM_EXIT_FAIL_ENTRY */ 3595 struct { 3596 __u64 hardware_entry_failure_reason; 3597 } fail_entry; 3598 3599 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due 3600 to unknown reasons. Further architecture-specific information is 3601 available in hardware_entry_failure_reason. 3602 3603 /* KVM_EXIT_EXCEPTION */ 3604 struct { 3605 __u32 exception; 3606 __u32 error_code; 3607 } ex; 3608 3609 Unused. 3610 3611 /* KVM_EXIT_IO */ 3612 struct { 3613 #define KVM_EXIT_IO_IN 0 3614 #define KVM_EXIT_IO_OUT 1 3615 __u8 direction; 3616 __u8 size; /* bytes */ 3617 __u16 port; 3618 __u32 count; 3619 __u64 data_offset; /* relative to kvm_run start */ 3620 } io; 3621 3622 If exit_reason is KVM_EXIT_IO, then the vcpu has 3623 executed a port I/O instruction which could not be satisfied by kvm. 3624 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or 3625 where kvm expects application code to place the data for the next 3626 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. 3627 3628 /* KVM_EXIT_DEBUG */ 3629 struct { 3630 struct kvm_debug_exit_arch arch; 3631 } debug; 3632 3633 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event 3634 for which architecture specific information is returned. 3635 3636 /* KVM_EXIT_MMIO */ 3637 struct { 3638 __u64 phys_addr; 3639 __u8 data[8]; 3640 __u32 len; 3641 __u8 is_write; 3642 } mmio; 3643 3644 If exit_reason is KVM_EXIT_MMIO, then the vcpu has 3645 executed a memory-mapped I/O instruction which could not be satisfied 3646 by kvm. The 'data' member contains the written data if 'is_write' is 3647 true, and should be filled by application code otherwise. 3648 3649 The 'data' member contains, in its first 'len' bytes, the value as it would 3650 appear if the VCPU performed a load or store of the appropriate width directly 3651 to the byte array. 3652 3653 NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and 3654 KVM_EXIT_EPR the corresponding 3655 operations are complete (and guest state is consistent) only after userspace 3656 has re-entered the kernel with KVM_RUN. The kernel side will first finish 3657 incomplete operations and then check for pending signals. Userspace 3658 can re-enter the guest with an unmasked signal pending to complete 3659 pending operations. 3660 3661 /* KVM_EXIT_HYPERCALL */ 3662 struct { 3663 __u64 nr; 3664 __u64 args[6]; 3665 __u64 ret; 3666 __u32 longmode; 3667 __u32 pad; 3668 } hypercall; 3669 3670 Unused. This was once used for 'hypercall to userspace'. To implement 3671 such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). 3672 Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. 3673 3674 /* KVM_EXIT_TPR_ACCESS */ 3675 struct { 3676 __u64 rip; 3677 __u32 is_write; 3678 __u32 pad; 3679 } tpr_access; 3680 3681 To be documented (KVM_TPR_ACCESS_REPORTING). 3682 3683 /* KVM_EXIT_S390_SIEIC */ 3684 struct { 3685 __u8 icptcode; 3686 __u64 mask; /* psw upper half */ 3687 __u64 addr; /* psw lower half */ 3688 __u16 ipa; 3689 __u32 ipb; 3690 } s390_sieic; 3691 3692 s390 specific. 3693 3694 /* KVM_EXIT_S390_RESET */ 3695 #define KVM_S390_RESET_POR 1 3696 #define KVM_S390_RESET_CLEAR 2 3697 #define KVM_S390_RESET_SUBSYSTEM 4 3698 #define KVM_S390_RESET_CPU_INIT 8 3699 #define KVM_S390_RESET_IPL 16 3700 __u64 s390_reset_flags; 3701 3702 s390 specific. 3703 3704 /* KVM_EXIT_S390_UCONTROL */ 3705 struct { 3706 __u64 trans_exc_code; 3707 __u32 pgm_code; 3708 } s390_ucontrol; 3709 3710 s390 specific. A page fault has occurred for a user controlled virtual 3711 machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be 3712 resolved by the kernel. 3713 The program code and the translation exception code that were placed 3714 in the cpu's lowcore are presented here as defined by the z Architecture 3715 Principles of Operation Book in the Chapter for Dynamic Address Translation 3716 (DAT) 3717 3718 /* KVM_EXIT_DCR */ 3719 struct { 3720 __u32 dcrn; 3721 __u32 data; 3722 __u8 is_write; 3723 } dcr; 3724 3725 Deprecated - was used for 440 KVM. 3726 3727 /* KVM_EXIT_OSI */ 3728 struct { 3729 __u64 gprs[32]; 3730 } osi; 3731 3732 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch 3733 hypercalls and exit with this exit struct that contains all the guest gprs. 3734 3735 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. 3736 Userspace can now handle the hypercall and when it's done modify the gprs as 3737 necessary. Upon guest entry all guest GPRs will then be replaced by the values 3738 in this struct. 3739 3740 /* KVM_EXIT_PAPR_HCALL */ 3741 struct { 3742 __u64 nr; 3743 __u64 ret; 3744 __u64 args[9]; 3745 } papr_hcall; 3746 3747 This is used on 64-bit PowerPC when emulating a pSeries partition, 3748 e.g. with the 'pseries' machine type in qemu. It occurs when the 3749 guest does a hypercall using the 'sc 1' instruction. The 'nr' field 3750 contains the hypercall number (from the guest R3), and 'args' contains 3751 the arguments (from the guest R4 - R12). Userspace should put the 3752 return code in 'ret' and any extra returned values in args[]. 3753 The possible hypercalls are defined in the Power Architecture Platform 3754 Requirements (PAPR) document available from www.power.org (free 3755 developer registration required to access it). 3756 3757 /* KVM_EXIT_S390_TSCH */ 3758 struct { 3759 __u16 subchannel_id; 3760 __u16 subchannel_nr; 3761 __u32 io_int_parm; 3762 __u32 io_int_word; 3763 __u32 ipb; 3764 __u8 dequeued; 3765 } s390_tsch; 3766 3767 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled 3768 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O 3769 interrupt for the target subchannel has been dequeued and subchannel_id, 3770 subchannel_nr, io_int_parm and io_int_word contain the parameters for that 3771 interrupt. ipb is needed for instruction parameter decoding. 3772 3773 /* KVM_EXIT_EPR */ 3774 struct { 3775 __u32 epr; 3776 } epr; 3777 3778 On FSL BookE PowerPC chips, the interrupt controller has a fast patch 3779 interrupt acknowledge path to the core. When the core successfully 3780 delivers an interrupt, it automatically populates the EPR register with 3781 the interrupt vector number and acknowledges the interrupt inside 3782 the interrupt controller. 3783 3784 In case the interrupt controller lives in user space, we need to do 3785 the interrupt acknowledge cycle through it to fetch the next to be 3786 delivered interrupt vector using this exit. 3787 3788 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an 3789 external interrupt has just been delivered into the guest. User space 3790 should put the acknowledged interrupt vector into the 'epr' field. 3791 3792 /* KVM_EXIT_SYSTEM_EVENT */ 3793 struct { 3794 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 3795 #define KVM_SYSTEM_EVENT_RESET 2 3796 #define KVM_SYSTEM_EVENT_CRASH 3 3797 __u32 type; 3798 __u64 flags; 3799 } system_event; 3800 3801 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered 3802 a system-level event using some architecture specific mechanism (hypercall 3803 or some special instruction). In case of ARM/ARM64, this is triggered using 3804 HVC instruction based PSCI call from the vcpu. The 'type' field describes 3805 the system-level event type. The 'flags' field describes architecture 3806 specific flags for the system-level event. 3807 3808 Valid values for 'type' are: 3809 KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the 3810 VM. Userspace is not obliged to honour this, and if it does honour 3811 this does not need to destroy the VM synchronously (ie it may call 3812 KVM_RUN again before shutdown finally occurs). 3813 KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. 3814 As with SHUTDOWN, userspace can choose to ignore the request, or 3815 to schedule the reset to occur in the future and may call KVM_RUN again. 3816 KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest 3817 has requested a crash condition maintenance. Userspace can choose 3818 to ignore the request, or to gather VM memory core dump and/or 3819 reset/shutdown of the VM. 3820 3821 /* KVM_EXIT_IOAPIC_EOI */ 3822 struct { 3823 __u8 vector; 3824 } eoi; 3825 3826 Indicates that the VCPU's in-kernel local APIC received an EOI for a 3827 level-triggered IOAPIC interrupt. This exit only triggers when the 3828 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled); 3829 the userspace IOAPIC should process the EOI and retrigger the interrupt if 3830 it is still asserted. Vector is the LAPIC interrupt vector for which the 3831 EOI was received. 3832 3833 struct kvm_hyperv_exit { 3834 #define KVM_EXIT_HYPERV_SYNIC 1 3835 #define KVM_EXIT_HYPERV_HCALL 2 3836 __u32 type; 3837 union { 3838 struct { 3839 __u32 msr; 3840 __u64 control; 3841 __u64 evt_page; 3842 __u64 msg_page; 3843 } synic; 3844 struct { 3845 __u64 input; 3846 __u64 result; 3847 __u64 params[2]; 3848 } hcall; 3849 } u; 3850 }; 3851 /* KVM_EXIT_HYPERV */ 3852 struct kvm_hyperv_exit hyperv; 3853 Indicates that the VCPU exits into userspace to process some tasks 3854 related to Hyper-V emulation. 3855 Valid values for 'type' are: 3856 KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about 3857 Hyper-V SynIC state change. Notification is used to remap SynIC 3858 event/message pages and to enable/disable SynIC messages/events processing 3859 in userspace. 3860 3861 /* Fix the size of the union. */ 3862 char padding[256]; 3863 }; 3864 3865 /* 3866 * shared registers between kvm and userspace. 3867 * kvm_valid_regs specifies the register classes set by the host 3868 * kvm_dirty_regs specified the register classes dirtied by userspace 3869 * struct kvm_sync_regs is architecture specific, as well as the 3870 * bits for kvm_valid_regs and kvm_dirty_regs 3871 */ 3872 __u64 kvm_valid_regs; 3873 __u64 kvm_dirty_regs; 3874 union { 3875 struct kvm_sync_regs regs; 3876 char padding[1024]; 3877 } s; 3878 3879 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access 3880 certain guest registers without having to call SET/GET_*REGS. Thus we can 3881 avoid some system call overhead if userspace has to handle the exit. 3882 Userspace can query the validity of the structure by checking 3883 kvm_valid_regs for specific bits. These bits are architecture specific 3884 and usually define the validity of a groups of registers. (e.g. one bit 3885 for general purpose registers) 3886 3887 Please note that the kernel is allowed to use the kvm_run structure as the 3888 primary storage for certain register types. Therefore, the kernel may use the 3889 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. 3890 3891 }; 3892 3893 3894 3895 6. Capabilities that can be enabled on vCPUs 3896 -------------------------------------------- 3897 3898 There are certain capabilities that change the behavior of the virtual CPU or 3899 the virtual machine when enabled. To enable them, please see section 4.37. 3900 Below you can find a list of capabilities and what their effect on the vCPU or 3901 the virtual machine is when enabling them. 3902 3903 The following information is provided along with the description: 3904 3905 Architectures: which instruction set architectures provide this ioctl. 3906 x86 includes both i386 and x86_64. 3907 3908 Target: whether this is a per-vcpu or per-vm capability. 3909 3910 Parameters: what parameters are accepted by the capability. 3911 3912 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) 3913 are not detailed, but errors with specific meanings are. 3914 3915 3916 6.1 KVM_CAP_PPC_OSI 3917 3918 Architectures: ppc 3919 Target: vcpu 3920 Parameters: none 3921 Returns: 0 on success; -1 on error 3922 3923 This capability enables interception of OSI hypercalls that otherwise would 3924 be treated as normal system calls to be injected into the guest. OSI hypercalls 3925 were invented by Mac-on-Linux to have a standardized communication mechanism 3926 between the guest and the host. 3927 3928 When this capability is enabled, KVM_EXIT_OSI can occur. 3929 3930 3931 6.2 KVM_CAP_PPC_PAPR 3932 3933 Architectures: ppc 3934 Target: vcpu 3935 Parameters: none 3936 Returns: 0 on success; -1 on error 3937 3938 This capability enables interception of PAPR hypercalls. PAPR hypercalls are 3939 done using the hypercall instruction "sc 1". 3940 3941 It also sets the guest privilege level to "supervisor" mode. Usually the guest 3942 runs in "hypervisor" privilege mode with a few missing features. 3943 3944 In addition to the above, it changes the semantics of SDR1. In this mode, the 3945 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the 3946 HTAB invisible to the guest. 3947 3948 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. 3949 3950 3951 6.3 KVM_CAP_SW_TLB 3952 3953 Architectures: ppc 3954 Target: vcpu 3955 Parameters: args[0] is the address of a struct kvm_config_tlb 3956 Returns: 0 on success; -1 on error 3957 3958 struct kvm_config_tlb { 3959 __u64 params; 3960 __u64 array; 3961 __u32 mmu_type; 3962 __u32 array_len; 3963 }; 3964 3965 Configures the virtual CPU's TLB array, establishing a shared memory area 3966 between userspace and KVM. The "params" and "array" fields are userspace 3967 addresses of mmu-type-specific data structures. The "array_len" field is an 3968 safety mechanism, and should be set to the size in bytes of the memory that 3969 userspace has reserved for the array. It must be at least the size dictated 3970 by "mmu_type" and "params". 3971 3972 While KVM_RUN is active, the shared region is under control of KVM. Its 3973 contents are undefined, and any modification by userspace results in 3974 boundedly undefined behavior. 3975 3976 On return from KVM_RUN, the shared region will reflect the current state of 3977 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB 3978 to tell KVM which entries have been changed, prior to calling KVM_RUN again 3979 on this vcpu. 3980 3981 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: 3982 - The "params" field is of type "struct kvm_book3e_206_tlb_params". 3983 - The "array" field points to an array of type "struct 3984 kvm_book3e_206_tlb_entry". 3985 - The array consists of all entries in the first TLB, followed by all 3986 entries in the second TLB. 3987 - Within a TLB, entries are ordered first by increasing set number. Within a 3988 set, entries are ordered by way (increasing ESEL). 3989 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) 3990 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. 3991 - The tsize field of mas1 shall be set to 4K on TLB0, even though the 3992 hardware ignores this value for TLB0. 3993 3994 6.4 KVM_CAP_S390_CSS_SUPPORT 3995 3996 Architectures: s390 3997 Target: vcpu 3998 Parameters: none 3999 Returns: 0 on success; -1 on error 4000 4001 This capability enables support for handling of channel I/O instructions. 4002 4003 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are 4004 handled in-kernel, while the other I/O instructions are passed to userspace. 4005 4006 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST 4007 SUBCHANNEL intercepts. 4008 4009 Note that even though this capability is enabled per-vcpu, the complete 4010 virtual machine is affected. 4011 4012 6.5 KVM_CAP_PPC_EPR 4013 4014 Architectures: ppc 4015 Target: vcpu 4016 Parameters: args[0] defines whether the proxy facility is active 4017 Returns: 0 on success; -1 on error 4018 4019 This capability enables or disables the delivery of interrupts through the 4020 external proxy facility. 4021 4022 When enabled (args[0] != 0), every time the guest gets an external interrupt 4023 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit 4024 to receive the topmost interrupt vector. 4025 4026 When disabled (args[0] == 0), behavior is as if this facility is unsupported. 4027 4028 When this capability is enabled, KVM_EXIT_EPR can occur. 4029 4030 6.6 KVM_CAP_IRQ_MPIC 4031 4032 Architectures: ppc 4033 Parameters: args[0] is the MPIC device fd 4034 args[1] is the MPIC CPU number for this vcpu 4035 4036 This capability connects the vcpu to an in-kernel MPIC device. 4037 4038 6.7 KVM_CAP_IRQ_XICS 4039 4040 Architectures: ppc 4041 Target: vcpu 4042 Parameters: args[0] is the XICS device fd 4043 args[1] is the XICS CPU number (server ID) for this vcpu 4044 4045 This capability connects the vcpu to an in-kernel XICS device. 4046 4047 6.8 KVM_CAP_S390_IRQCHIP 4048 4049 Architectures: s390 4050 Target: vm 4051 Parameters: none 4052 4053 This capability enables the in-kernel irqchip for s390. Please refer to 4054 "4.24 KVM_CREATE_IRQCHIP" for details. 4055 4056 6.9 KVM_CAP_MIPS_FPU 4057 4058 Architectures: mips 4059 Target: vcpu 4060 Parameters: args[0] is reserved for future use (should be 0). 4061 4062 This capability allows the use of the host Floating Point Unit by the guest. It 4063 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is 4064 done the KVM_REG_MIPS_FPR_* and KVM_REG_MIPS_FCR_* registers can be accessed 4065 (depending on the current guest FPU register mode), and the Status.FR, 4066 Config5.FRE bits are accessible via the KVM API and also from the guest, 4067 depending on them being supported by the FPU. 4068 4069 6.10 KVM_CAP_MIPS_MSA 4070 4071 Architectures: mips 4072 Target: vcpu 4073 Parameters: args[0] is reserved for future use (should be 0). 4074 4075 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest. 4076 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest. 4077 Once this is done the KVM_REG_MIPS_VEC_* and KVM_REG_MIPS_MSA_* registers can be 4078 accessed, and the Config5.MSAEn bit is accessible via the KVM API and also from 4079 the guest. 4080 4081 7. Capabilities that can be enabled on VMs 4082 ------------------------------------------ 4083 4084 There are certain capabilities that change the behavior of the virtual 4085 machine when enabled. To enable them, please see section 4.37. Below 4086 you can find a list of capabilities and what their effect on the VM 4087 is when enabling them. 4088 4089 The following information is provided along with the description: 4090 4091 Architectures: which instruction set architectures provide this ioctl. 4092 x86 includes both i386 and x86_64. 4093 4094 Parameters: what parameters are accepted by the capability. 4095 4096 Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) 4097 are not detailed, but errors with specific meanings are. 4098 4099 4100 7.1 KVM_CAP_PPC_ENABLE_HCALL 4101 4102 Architectures: ppc 4103 Parameters: args[0] is the sPAPR hcall number 4104 args[1] is 0 to disable, 1 to enable in-kernel handling 4105 4106 This capability controls whether individual sPAPR hypercalls (hcalls) 4107 get handled by the kernel or not. Enabling or disabling in-kernel 4108 handling of an hcall is effective across the VM. On creation, an 4109 initial set of hcalls are enabled for in-kernel handling, which 4110 consists of those hcalls for which in-kernel handlers were implemented 4111 before this capability was implemented. If disabled, the kernel will 4112 not to attempt to handle the hcall, but will always exit to userspace 4113 to handle it. Note that it may not make sense to enable some and 4114 disable others of a group of related hcalls, but KVM does not prevent 4115 userspace from doing that. 4116 4117 If the hcall number specified is not one that has an in-kernel 4118 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL 4119 error. 4120 4121 7.2 KVM_CAP_S390_USER_SIGP 4122 4123 Architectures: s390 4124 Parameters: none 4125 4126 This capability controls which SIGP orders will be handled completely in user 4127 space. With this capability enabled, all fast orders will be handled completely 4128 in the kernel: 4129 - SENSE 4130 - SENSE RUNNING 4131 - EXTERNAL CALL 4132 - EMERGENCY SIGNAL 4133 - CONDITIONAL EMERGENCY SIGNAL 4134 4135 All other orders will be handled completely in user space. 4136 4137 Only privileged operation exceptions will be checked for in the kernel (or even 4138 in the hardware prior to interception). If this capability is not enabled, the 4139 old way of handling SIGP orders is used (partially in kernel and user space). 4140 4141 7.3 KVM_CAP_S390_VECTOR_REGISTERS 4142 4143 Architectures: s390 4144 Parameters: none 4145 Returns: 0 on success, negative value on error 4146 4147 Allows use of the vector registers introduced with z13 processor, and 4148 provides for the synchronization between host and user space. Will 4149 return -EINVAL if the machine does not support vectors. 4150 4151 7.4 KVM_CAP_S390_USER_STSI 4152 4153 Architectures: s390 4154 Parameters: none 4155 4156 This capability allows post-handlers for the STSI instruction. After 4157 initial handling in the kernel, KVM exits to user space with 4158 KVM_EXIT_S390_STSI to allow user space to insert further data. 4159 4160 Before exiting to userspace, kvm handlers should fill in s390_stsi field of 4161 vcpu->run: 4162 struct { 4163 __u64 addr; 4164 __u8 ar; 4165 __u8 reserved; 4166 __u8 fc; 4167 __u8 sel1; 4168 __u16 sel2; 4169 } s390_stsi; 4170 4171 @addr - guest address of STSI SYSIB 4172 @fc - function code 4173 @sel1 - selector 1 4174 @sel2 - selector 2 4175 @ar - access register number 4176 4177 KVM handlers should exit to userspace with rc = -EREMOTE. 4178 4179 7.5 KVM_CAP_SPLIT_IRQCHIP 4180 4181 Architectures: x86 4182 Parameters: args[0] - number of routes reserved for userspace IOAPICs 4183 Returns: 0 on success, -1 on error 4184 4185 Create a local apic for each processor in the kernel. This can be used 4186 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the 4187 IOAPIC and PIC (and also the PIT, even though this has to be enabled 4188 separately). 4189 4190 This capability also enables in kernel routing of interrupt requests; 4191 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are 4192 used in the IRQ routing table. The first args[0] MSI routes are reserved 4193 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes, 4194 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace. 4195 4196 Fails if VCPU has already been created, or if the irqchip is already in the 4197 kernel (i.e. KVM_CREATE_IRQCHIP has already been called). 4198 4199 7.6 KVM_CAP_S390_RI 4200 4201 Architectures: s390 4202 Parameters: none 4203 4204 Allows use of runtime-instrumentation introduced with zEC12 processor. 4205 Will return -EINVAL if the machine does not support runtime-instrumentation. 4206 Will return -EBUSY if a VCPU has already been created. 4207 4208 7.7 KVM_CAP_X2APIC_API 4209 4210 Architectures: x86 4211 Parameters: args[0] - features that should be enabled 4212 Returns: 0 on success, -EINVAL when args[0] contains invalid features 4213 4214 Valid feature flags in args[0] are 4215 4216 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) 4217 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) 4218 4219 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of 4220 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, 4221 allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their 4222 respective sections. 4223 4224 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work 4225 in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff 4226 as a broadcast even in x2APIC mode in order to support physical x2APIC 4227 without interrupt remapping. This is undesirable in logical mode, 4228 where 0xff represents CPUs 0-7 in cluster 0. 4229 4230 7.8 KVM_CAP_S390_USER_INSTR0 4231 4232 Architectures: s390 4233 Parameters: none 4234 4235 With this capability enabled, all illegal instructions 0x0000 (2 bytes) will 4236 be intercepted and forwarded to user space. User space can use this 4237 mechanism e.g. to realize 2-byte software breakpoints. The kernel will 4238 not inject an operating exception for these instructions, user space has 4239 to take care of that. 4240 4241 This capability can be enabled dynamically even if VCPUs were already 4242 created and are running. 4243 4244 7.9 KVM_CAP_S390_GS 4245 4246 Architectures: s390 4247 Parameters: none 4248 Returns: 0 on success; -EINVAL if the machine does not support 4249 guarded storage; -EBUSY if a VCPU has already been created. 4250 4251 Allows use of guarded storage for the KVM guest. 4252 4253 7.10 KVM_CAP_S390_AIS 4254 4255 Architectures: s390 4256 Parameters: none 4257 4258 Allow use of adapter-interruption suppression. 4259 Returns: 0 on success; -EBUSY if a VCPU has already been created. 4260 4261 7.11 KVM_CAP_PPC_SMT 4262 4263 Architectures: ppc 4264 Parameters: vsmt_mode, flags 4265 4266 Enabling this capability on a VM provides userspace with a way to set 4267 the desired virtual SMT mode (i.e. the number of virtual CPUs per 4268 virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2 4269 between 1 and 8. On POWER8, vsmt_mode must also be no greater than 4270 the number of threads per subcore for the host. Currently flags must 4271 be 0. A successful call to enable this capability will result in 4272 vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is 4273 subsequently queried for the VM. This capability is only supported by 4274 HV KVM, and can only be set before any VCPUs have been created. 4275 The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT 4276 modes are available. 4277 4278 7.12 KVM_CAP_PPC_FWNMI 4279 4280 Architectures: ppc 4281 Parameters: none 4282 4283 With this capability a machine check exception in the guest address 4284 space will cause KVM to exit the guest with NMI exit reason. This 4285 enables QEMU to build error log and branch to guest kernel registered 4286 machine check handling routine. Without this capability KVM will 4287 branch to guests' 0x200 interrupt vector. 4288 4289 8. Other capabilities. 4290 ---------------------- 4291 4292 This section lists capabilities that give information about other 4293 features of the KVM implementation. 4294 4295 8.1 KVM_CAP_PPC_HWRNG 4296 4297 Architectures: ppc 4298 4299 This capability, if KVM_CHECK_EXTENSION indicates that it is 4300 available, means that that the kernel has an implementation of the 4301 H_RANDOM hypercall backed by a hardware random-number generator. 4302 If present, the kernel H_RANDOM handler can be enabled for guest use 4303 with the KVM_CAP_PPC_ENABLE_HCALL capability. 4304 4305 8.2 KVM_CAP_HYPERV_SYNIC 4306 4307 Architectures: x86 4308 This capability, if KVM_CHECK_EXTENSION indicates that it is 4309 available, means that that the kernel has an implementation of the 4310 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is 4311 used to support Windows Hyper-V based guest paravirt drivers(VMBus). 4312 4313 In order to use SynIC, it has to be activated by setting this 4314 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this 4315 will disable the use of APIC hardware virtualization even if supported 4316 by the CPU, as it's incompatible with SynIC auto-EOI behavior. 4317 4318 8.3 KVM_CAP_PPC_RADIX_MMU 4319 4320 Architectures: ppc 4321 4322 This capability, if KVM_CHECK_EXTENSION indicates that it is 4323 available, means that that the kernel can support guests using the 4324 radix MMU defined in Power ISA V3.00 (as implemented in the POWER9 4325 processor). 4326 4327 8.4 KVM_CAP_PPC_HASH_MMU_V3 4328 4329 Architectures: ppc 4330 4331 This capability, if KVM_CHECK_EXTENSION indicates that it is 4332 available, means that that the kernel can support guests using the 4333 hashed page table MMU defined in Power ISA V3.00 (as implemented in 4334 the POWER9 processor), including in-memory segment tables. 4335 4336 8.5 KVM_CAP_MIPS_VZ 4337 4338 Architectures: mips 4339 4340 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 4341 it is available, means that full hardware assisted virtualization capabilities 4342 of the hardware are available for use through KVM. An appropriate 4343 KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which 4344 utilises it. 4345 4346 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 4347 available, it means that the VM is using full hardware assisted virtualization 4348 capabilities of the hardware. This is useful to check after creating a VM with 4349 KVM_VM_MIPS_DEFAULT. 4350 4351 The value returned by KVM_CHECK_EXTENSION should be compared against known 4352 values (see below). All other values are reserved. This is to allow for the 4353 possibility of other hardware assisted virtualization implementations which 4354 may be incompatible with the MIPS VZ ASE. 4355 4356 0: The trap & emulate implementation is in use to run guest code in user 4357 mode. Guest virtual memory segments are rearranged to fit the guest in the 4358 user mode address space. 4359 4360 1: The MIPS VZ ASE is in use, providing full hardware assisted 4361 virtualization, including standard guest virtual memory segments. 4362 4363 8.6 KVM_CAP_MIPS_TE 4364 4365 Architectures: mips 4366 4367 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that 4368 it is available, means that the trap & emulate implementation is available to 4369 run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware 4370 assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed 4371 to KVM_CREATE_VM to create a VM which utilises it. 4372 4373 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is 4374 available, it means that the VM is using trap & emulate. 4375 4376 8.7 KVM_CAP_MIPS_64BIT 4377 4378 Architectures: mips 4379 4380 This capability indicates the supported architecture type of the guest, i.e. the 4381 supported register and address width. 4382 4383 The values returned when this capability is checked by KVM_CHECK_EXTENSION on a 4384 kvm VM handle correspond roughly to the CP0_Config.AT register field, and should 4385 be checked specifically against known values (see below). All other values are 4386 reserved. 4387 4388 0: MIPS32 or microMIPS32. 4389 Both registers and addresses are 32-bits wide. 4390 It will only be possible to run 32-bit guest code. 4391 4392 1: MIPS64 or microMIPS64 with access only to 32-bit compatibility segments. 4393 Registers are 64-bits wide, but addresses are 32-bits wide. 4394 64-bit guest code may run but cannot access MIPS64 memory segments. 4395 It will also be possible to run 32-bit guest code. 4396 4397 2: MIPS64 or microMIPS64 with access to all address segments. 4398 Both registers and addresses are 64-bits wide. 4399 It will be possible to run 64-bit or 32-bit guest code. 4400 4401 8.8 KVM_CAP_X86_GUEST_MWAIT 4402 4403 Architectures: x86 4404 4405 This capability indicates that guest using memory monotoring instructions 4406 (MWAIT/MWAITX) to stop the virtual CPU will not cause a VM exit. As such time 4407 spent while virtual CPU is halted in this way will then be accounted for as 4408 guest running time on the host (as opposed to e.g. HLT). 4409 4410 8.9 KVM_CAP_ARM_USER_IRQ 4411 4412 Architectures: arm, arm64 4413 This capability, if KVM_CHECK_EXTENSION indicates that it is available, means 4414 that if userspace creates a VM without an in-kernel interrupt controller, it 4415 will be notified of changes to the output level of in-kernel emulated devices, 4416 which can generate virtual interrupts, presented to the VM. 4417 For such VMs, on every return to userspace, the kernel 4418 updates the vcpu's run->s.regs.device_irq_level field to represent the actual 4419 output level of the device. 4420 4421 Whenever kvm detects a change in the device output level, kvm guarantees at 4422 least one return to userspace before running the VM. This exit could either 4423 be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way, 4424 userspace can always sample the device output level and re-compute the state of 4425 the userspace interrupt controller. Userspace should always check the state 4426 of run->s.regs.device_irq_level on every kvm exit. 4427 The value in run->s.regs.device_irq_level can represent both level and edge 4428 triggered interrupt signals, depending on the device. Edge triggered interrupt 4429 signals will exit to userspace with the bit in run->s.regs.device_irq_level 4430 set exactly once per edge signal. 4431 4432 The field run->s.regs.device_irq_level is available independent of 4433 run->kvm_valid_regs or run->kvm_dirty_regs bits. 4434 4435 If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a 4436 number larger than 0 indicating the version of this capability is implemented 4437 and thereby which bits in in run->s.regs.device_irq_level can signal values. 4438 4439 Currently the following bits are defined for the device_irq_level bitmap: 4440 4441 KVM_CAP_ARM_USER_IRQ >= 1: 4442 4443 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer 4444 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer 4445 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal 4446 4447 Future versions of kvm may implement additional events. These will get 4448 indicated by returning a higher number from KVM_CHECK_EXTENSION and will be 4449 listed above. 4450 4451 8.10 KVM_CAP_PPC_SMT_POSSIBLE 4452 4453 Architectures: ppc 4454 4455 Querying this capability returns a bitmap indicating the possible 4456 virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N 4457 (counting from the right) is set, then a virtual SMT mode of 2^N is 4458 available. 4459 4460 8.11 KVM_CAP_HYPERV_SYNIC2 4461 4462 Architectures: x86 4463 4464 This capability enables a newer version of Hyper-V Synthetic interrupt 4465 controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM 4466 doesn't clear SynIC message and event flags pages when they are enabled by 4467 writing to the respective MSRs. 4468 4469 8.12 KVM_CAP_HYPERV_VP_INDEX 4470 4471 Architectures: x86 4472 4473 This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its 4474 value is used to denote the target vcpu for a SynIC interrupt. For 4475 compatibilty, KVM initializes this msr to KVM's internal vcpu index. When this 4476 capability is absent, userspace can still query this msr's value. 4477 4478 8.13 KVM_CAP_S390_AIS_MIGRATION 4479 4480 Architectures: s390 4481 Parameters: none 4482 4483 This capability indicates if the flic device will be able to get/set the 4484 AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows 4485 to discover this without having to create a flic device.