]>
Commit | Line | Data |
---|---|---|
9c1b96e3 AK |
1 | The Definitive KVM (Kernel-based Virtual Machine) API Documentation |
2 | =================================================================== | |
3 | ||
4 | 1. General description | |
414fa985 | 5 | ---------------------- |
9c1b96e3 AK |
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 | ||
414fa985 | 27 | |
2044892d | 28 | 2. File descriptors |
414fa985 | 29 | ------------------- |
9c1b96e3 AK |
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 | |
2044892d | 34 | handle will create a VM file descriptor which can be used to issue VM |
9c1b96e3 AK |
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 | ||
414fa985 | 47 | |
9c1b96e3 | 48 | 3. Extensions |
414fa985 | 49 | ------------- |
9c1b96e3 AK |
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 | ||
c9f3f2d8 | 56 | The extension mechanism is not based on the Linux version number. |
9c1b96e3 AK |
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 | ||
414fa985 | 61 | |
9c1b96e3 | 62 | 4. API description |
414fa985 | 63 | ------------------ |
9c1b96e3 AK |
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 | |
7f05db6a | 71 | API version 12 (see section 4.1), a KVM_CAP_xyz constant, which |
9c1b96e3 | 72 | means availability needs to be checked with KVM_CHECK_EXTENSION |
7f05db6a MT |
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. | |
9c1b96e3 AK |
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 | ||
414fa985 | 88 | |
9c1b96e3 AK |
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 | ||
414fa985 | 104 | |
9c1b96e3 AK |
105 | 4.2 KVM_CREATE_VM |
106 | ||
107 | Capability: basic | |
108 | Architectures: all | |
109 | Type: system ioctl | |
e08b9637 | 110 | Parameters: machine type identifier (KVM_VM_*) |
9c1b96e3 AK |
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. An mmap() of a VM fd | |
114 | will access the virtual machine's physical address space; offset zero | |
115 | corresponds to guest physical address zero. Use of mmap() on a VM fd | |
116 | is discouraged if userspace memory allocation (KVM_CAP_USER_MEMORY) is | |
117 | available. | |
e08b9637 CO |
118 | You most certainly want to use 0 as machine type. |
119 | ||
120 | In order to create user controlled virtual machines on S390, check | |
121 | KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as | |
122 | privileged user (CAP_SYS_ADMIN). | |
9c1b96e3 | 123 | |
414fa985 | 124 | |
9c1b96e3 AK |
125 | 4.3 KVM_GET_MSR_INDEX_LIST |
126 | ||
127 | Capability: basic | |
128 | Architectures: x86 | |
129 | Type: system | |
130 | Parameters: struct kvm_msr_list (in/out) | |
131 | Returns: 0 on success; -1 on error | |
132 | Errors: | |
133 | E2BIG: the msr index list is to be to fit in the array specified by | |
134 | the user. | |
135 | ||
136 | struct kvm_msr_list { | |
137 | __u32 nmsrs; /* number of msrs in entries */ | |
138 | __u32 indices[0]; | |
139 | }; | |
140 | ||
141 | This ioctl returns the guest msrs that are supported. The list varies | |
142 | by kvm version and host processor, but does not change otherwise. The | |
143 | 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 | |
145 | the indices array with their numbers. | |
146 | ||
2e2602ca AK |
147 | Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are |
148 | not returned in the MSR list, as different vcpus can have a different number | |
149 | of banks, as set via the KVM_X86_SETUP_MCE ioctl. | |
150 | ||
414fa985 | 151 | |
9c1b96e3 AK |
152 | 4.4 KVM_CHECK_EXTENSION |
153 | ||
92b591a4 | 154 | Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl |
9c1b96e3 | 155 | Architectures: all |
92b591a4 | 156 | Type: system ioctl, vm ioctl |
9c1b96e3 AK |
157 | Parameters: extension identifier (KVM_CAP_*) |
158 | Returns: 0 if unsupported; 1 (or some other positive integer) if supported | |
159 | ||
160 | The API allows the application to query about extensions to the core | |
161 | kvm API. Userspace passes an extension identifier (an integer) and | |
162 | receives an integer that describes the extension availability. | |
163 | Generally 0 means no and 1 means yes, but some extensions may report | |
164 | additional information in the integer return value. | |
165 | ||
92b591a4 AG |
166 | Based on their initialization different VMs may have different capabilities. |
167 | It is thus encouraged to use the vm ioctl to query for capabilities (available | |
168 | with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) | |
414fa985 | 169 | |
9c1b96e3 AK |
170 | 4.5 KVM_GET_VCPU_MMAP_SIZE |
171 | ||
172 | Capability: basic | |
173 | Architectures: all | |
174 | Type: system ioctl | |
175 | Parameters: none | |
176 | Returns: size of vcpu mmap area, in bytes | |
177 | ||
178 | The KVM_RUN ioctl (cf.) communicates with userspace via a shared | |
179 | memory region. This ioctl returns the size of that region. See the | |
180 | KVM_RUN documentation for details. | |
181 | ||
414fa985 | 182 | |
9c1b96e3 AK |
183 | 4.6 KVM_SET_MEMORY_REGION |
184 | ||
185 | Capability: basic | |
186 | Architectures: all | |
187 | Type: vm ioctl | |
188 | Parameters: struct kvm_memory_region (in) | |
189 | Returns: 0 on success, -1 on error | |
190 | ||
b74a07be | 191 | This ioctl is obsolete and has been removed. |
9c1b96e3 | 192 | |
414fa985 | 193 | |
68ba6974 | 194 | 4.7 KVM_CREATE_VCPU |
9c1b96e3 AK |
195 | |
196 | Capability: basic | |
197 | Architectures: all | |
198 | Type: vm ioctl | |
199 | Parameters: vcpu id (apic id on x86) | |
200 | Returns: vcpu fd on success, -1 on error | |
201 | ||
202 | This API adds a vcpu to a virtual machine. The vcpu id is a small integer | |
8c3ba334 SL |
203 | in the range [0, max_vcpus). |
204 | ||
205 | The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of | |
206 | the KVM_CHECK_EXTENSION ioctl() at run-time. | |
207 | The maximum possible value for max_vcpus can be retrieved using the | |
208 | KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. | |
209 | ||
76d25402 PE |
210 | If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4 |
211 | cpus max. | |
8c3ba334 SL |
212 | If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is |
213 | same as the value returned from KVM_CAP_NR_VCPUS. | |
9c1b96e3 | 214 | |
371fefd6 PM |
215 | On powerpc using book3s_hv mode, the vcpus are mapped onto virtual |
216 | threads in one or more virtual CPU cores. (This is because the | |
217 | hardware requires all the hardware threads in a CPU core to be in the | |
218 | same partition.) The KVM_CAP_PPC_SMT capability indicates the number | |
36442687 AK |
219 | of vcpus per virtual core (vcore). The vcore id is obtained by |
220 | dividing the vcpu id by the number of vcpus per vcore. The vcpus in a | |
221 | given vcore will always be in the same physical core as each other | |
222 | (though that might be a different physical core from time to time). | |
223 | Userspace can control the threading (SMT) mode of the guest by its | |
224 | allocation of vcpu ids. For example, if userspace wants | |
225 | single-threaded guest vcpus, it should make all vcpu ids be a multiple | |
226 | of the number of vcpus per vcore. | |
227 | ||
5b1c1493 CO |
228 | For virtual cpus that have been created with S390 user controlled virtual |
229 | machines, the resulting vcpu fd can be memory mapped at page offset | |
230 | KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual | |
231 | cpu's hardware control block. | |
232 | ||
414fa985 | 233 | |
68ba6974 | 234 | 4.8 KVM_GET_DIRTY_LOG (vm ioctl) |
9c1b96e3 AK |
235 | |
236 | Capability: basic | |
237 | Architectures: x86 | |
238 | Type: vm ioctl | |
239 | Parameters: struct kvm_dirty_log (in/out) | |
240 | Returns: 0 on success, -1 on error | |
241 | ||
242 | /* for KVM_GET_DIRTY_LOG */ | |
243 | struct kvm_dirty_log { | |
244 | __u32 slot; | |
245 | __u32 padding; | |
246 | union { | |
247 | void __user *dirty_bitmap; /* one bit per page */ | |
248 | __u64 padding; | |
249 | }; | |
250 | }; | |
251 | ||
252 | Given a memory slot, return a bitmap containing any pages dirtied | |
253 | since the last call to this ioctl. Bit 0 is the first page in the | |
254 | memory slot. Ensure the entire structure is cleared to avoid padding | |
255 | issues. | |
256 | ||
414fa985 | 257 | |
68ba6974 | 258 | 4.9 KVM_SET_MEMORY_ALIAS |
9c1b96e3 AK |
259 | |
260 | Capability: basic | |
261 | Architectures: x86 | |
262 | Type: vm ioctl | |
263 | Parameters: struct kvm_memory_alias (in) | |
264 | Returns: 0 (success), -1 (error) | |
265 | ||
a1f4d395 | 266 | This ioctl is obsolete and has been removed. |
9c1b96e3 | 267 | |
414fa985 | 268 | |
68ba6974 | 269 | 4.10 KVM_RUN |
9c1b96e3 AK |
270 | |
271 | Capability: basic | |
272 | Architectures: all | |
273 | Type: vcpu ioctl | |
274 | Parameters: none | |
275 | Returns: 0 on success, -1 on error | |
276 | Errors: | |
277 | EINTR: an unmasked signal is pending | |
278 | ||
279 | This ioctl is used to run a guest virtual cpu. While there are no | |
280 | explicit parameters, there is an implicit parameter block that can be | |
281 | obtained by mmap()ing the vcpu fd at offset 0, with the size given by | |
282 | KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct | |
283 | kvm_run' (see below). | |
284 | ||
414fa985 | 285 | |
68ba6974 | 286 | 4.11 KVM_GET_REGS |
9c1b96e3 AK |
287 | |
288 | Capability: basic | |
379e04c7 | 289 | Architectures: all except ARM, arm64 |
9c1b96e3 AK |
290 | Type: vcpu ioctl |
291 | Parameters: struct kvm_regs (out) | |
292 | Returns: 0 on success, -1 on error | |
293 | ||
294 | Reads the general purpose registers from the vcpu. | |
295 | ||
296 | /* x86 */ | |
297 | struct kvm_regs { | |
298 | /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ | |
299 | __u64 rax, rbx, rcx, rdx; | |
300 | __u64 rsi, rdi, rsp, rbp; | |
301 | __u64 r8, r9, r10, r11; | |
302 | __u64 r12, r13, r14, r15; | |
303 | __u64 rip, rflags; | |
304 | }; | |
305 | ||
c2d2c21b JH |
306 | /* mips */ |
307 | struct kvm_regs { | |
308 | /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ | |
309 | __u64 gpr[32]; | |
310 | __u64 hi; | |
311 | __u64 lo; | |
312 | __u64 pc; | |
313 | }; | |
314 | ||
414fa985 | 315 | |
68ba6974 | 316 | 4.12 KVM_SET_REGS |
9c1b96e3 AK |
317 | |
318 | Capability: basic | |
379e04c7 | 319 | Architectures: all except ARM, arm64 |
9c1b96e3 AK |
320 | Type: vcpu ioctl |
321 | Parameters: struct kvm_regs (in) | |
322 | Returns: 0 on success, -1 on error | |
323 | ||
324 | Writes the general purpose registers into the vcpu. | |
325 | ||
326 | See KVM_GET_REGS for the data structure. | |
327 | ||
414fa985 | 328 | |
68ba6974 | 329 | 4.13 KVM_GET_SREGS |
9c1b96e3 AK |
330 | |
331 | Capability: basic | |
5ce941ee | 332 | Architectures: x86, ppc |
9c1b96e3 AK |
333 | Type: vcpu ioctl |
334 | Parameters: struct kvm_sregs (out) | |
335 | Returns: 0 on success, -1 on error | |
336 | ||
337 | Reads special registers from the vcpu. | |
338 | ||
339 | /* x86 */ | |
340 | struct kvm_sregs { | |
341 | struct kvm_segment cs, ds, es, fs, gs, ss; | |
342 | struct kvm_segment tr, ldt; | |
343 | struct kvm_dtable gdt, idt; | |
344 | __u64 cr0, cr2, cr3, cr4, cr8; | |
345 | __u64 efer; | |
346 | __u64 apic_base; | |
347 | __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64]; | |
348 | }; | |
349 | ||
68e2ffed | 350 | /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */ |
5ce941ee | 351 | |
9c1b96e3 AK |
352 | interrupt_bitmap is a bitmap of pending external interrupts. At most |
353 | one bit may be set. This interrupt has been acknowledged by the APIC | |
354 | but not yet injected into the cpu core. | |
355 | ||
414fa985 | 356 | |
68ba6974 | 357 | 4.14 KVM_SET_SREGS |
9c1b96e3 AK |
358 | |
359 | Capability: basic | |
5ce941ee | 360 | Architectures: x86, ppc |
9c1b96e3 AK |
361 | Type: vcpu ioctl |
362 | Parameters: struct kvm_sregs (in) | |
363 | Returns: 0 on success, -1 on error | |
364 | ||
365 | Writes special registers into the vcpu. See KVM_GET_SREGS for the | |
366 | data structures. | |
367 | ||
414fa985 | 368 | |
68ba6974 | 369 | 4.15 KVM_TRANSLATE |
9c1b96e3 AK |
370 | |
371 | Capability: basic | |
372 | Architectures: x86 | |
373 | Type: vcpu ioctl | |
374 | Parameters: struct kvm_translation (in/out) | |
375 | Returns: 0 on success, -1 on error | |
376 | ||
377 | Translates a virtual address according to the vcpu's current address | |
378 | translation mode. | |
379 | ||
380 | struct kvm_translation { | |
381 | /* in */ | |
382 | __u64 linear_address; | |
383 | ||
384 | /* out */ | |
385 | __u64 physical_address; | |
386 | __u8 valid; | |
387 | __u8 writeable; | |
388 | __u8 usermode; | |
389 | __u8 pad[5]; | |
390 | }; | |
391 | ||
414fa985 | 392 | |
68ba6974 | 393 | 4.16 KVM_INTERRUPT |
9c1b96e3 AK |
394 | |
395 | Capability: basic | |
c2d2c21b | 396 | Architectures: x86, ppc, mips |
9c1b96e3 AK |
397 | Type: vcpu ioctl |
398 | Parameters: struct kvm_interrupt (in) | |
399 | Returns: 0 on success, -1 on error | |
400 | ||
401 | Queues a hardware interrupt vector to be injected. This is only | |
6f7a2bd4 | 402 | useful if in-kernel local APIC or equivalent is not used. |
9c1b96e3 AK |
403 | |
404 | /* for KVM_INTERRUPT */ | |
405 | struct kvm_interrupt { | |
406 | /* in */ | |
407 | __u32 irq; | |
408 | }; | |
409 | ||
6f7a2bd4 AG |
410 | X86: |
411 | ||
9c1b96e3 AK |
412 | Note 'irq' is an interrupt vector, not an interrupt pin or line. |
413 | ||
6f7a2bd4 AG |
414 | PPC: |
415 | ||
416 | Queues an external interrupt to be injected. This ioctl is overleaded | |
417 | with 3 different irq values: | |
418 | ||
419 | a) KVM_INTERRUPT_SET | |
420 | ||
421 | This injects an edge type external interrupt into the guest once it's ready | |
422 | to receive interrupts. When injected, the interrupt is done. | |
423 | ||
424 | b) KVM_INTERRUPT_UNSET | |
425 | ||
426 | This unsets any pending interrupt. | |
427 | ||
428 | Only available with KVM_CAP_PPC_UNSET_IRQ. | |
429 | ||
430 | c) KVM_INTERRUPT_SET_LEVEL | |
431 | ||
432 | This injects a level type external interrupt into the guest context. The | |
433 | interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET | |
434 | is triggered. | |
435 | ||
436 | Only available with KVM_CAP_PPC_IRQ_LEVEL. | |
437 | ||
438 | Note that any value for 'irq' other than the ones stated above is invalid | |
439 | and incurs unexpected behavior. | |
440 | ||
c2d2c21b JH |
441 | MIPS: |
442 | ||
443 | Queues an external interrupt to be injected into the virtual CPU. A negative | |
444 | interrupt number dequeues the interrupt. | |
445 | ||
414fa985 | 446 | |
68ba6974 | 447 | 4.17 KVM_DEBUG_GUEST |
9c1b96e3 AK |
448 | |
449 | Capability: basic | |
450 | Architectures: none | |
451 | Type: vcpu ioctl | |
452 | Parameters: none) | |
453 | Returns: -1 on error | |
454 | ||
455 | Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead. | |
456 | ||
414fa985 | 457 | |
68ba6974 | 458 | 4.18 KVM_GET_MSRS |
9c1b96e3 AK |
459 | |
460 | Capability: basic | |
461 | Architectures: x86 | |
462 | Type: vcpu ioctl | |
463 | Parameters: struct kvm_msrs (in/out) | |
464 | Returns: 0 on success, -1 on error | |
465 | ||
466 | Reads model-specific registers from the vcpu. Supported msr indices can | |
467 | be obtained using KVM_GET_MSR_INDEX_LIST. | |
468 | ||
469 | struct kvm_msrs { | |
470 | __u32 nmsrs; /* number of msrs in entries */ | |
471 | __u32 pad; | |
472 | ||
473 | struct kvm_msr_entry entries[0]; | |
474 | }; | |
475 | ||
476 | struct kvm_msr_entry { | |
477 | __u32 index; | |
478 | __u32 reserved; | |
479 | __u64 data; | |
480 | }; | |
481 | ||
482 | Application code should set the 'nmsrs' member (which indicates the | |
483 | size of the entries array) and the 'index' member of each array entry. | |
484 | kvm will fill in the 'data' member. | |
485 | ||
414fa985 | 486 | |
68ba6974 | 487 | 4.19 KVM_SET_MSRS |
9c1b96e3 AK |
488 | |
489 | Capability: basic | |
490 | Architectures: x86 | |
491 | Type: vcpu ioctl | |
492 | Parameters: struct kvm_msrs (in) | |
493 | Returns: 0 on success, -1 on error | |
494 | ||
495 | Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the | |
496 | data structures. | |
497 | ||
498 | Application code should set the 'nmsrs' member (which indicates the | |
499 | size of the entries array), and the 'index' and 'data' members of each | |
500 | array entry. | |
501 | ||
414fa985 | 502 | |
68ba6974 | 503 | 4.20 KVM_SET_CPUID |
9c1b96e3 AK |
504 | |
505 | Capability: basic | |
506 | Architectures: x86 | |
507 | Type: vcpu ioctl | |
508 | Parameters: struct kvm_cpuid (in) | |
509 | Returns: 0 on success, -1 on error | |
510 | ||
511 | Defines the vcpu responses to the cpuid instruction. Applications | |
512 | should use the KVM_SET_CPUID2 ioctl if available. | |
513 | ||
514 | ||
515 | struct kvm_cpuid_entry { | |
516 | __u32 function; | |
517 | __u32 eax; | |
518 | __u32 ebx; | |
519 | __u32 ecx; | |
520 | __u32 edx; | |
521 | __u32 padding; | |
522 | }; | |
523 | ||
524 | /* for KVM_SET_CPUID */ | |
525 | struct kvm_cpuid { | |
526 | __u32 nent; | |
527 | __u32 padding; | |
528 | struct kvm_cpuid_entry entries[0]; | |
529 | }; | |
530 | ||
414fa985 | 531 | |
68ba6974 | 532 | 4.21 KVM_SET_SIGNAL_MASK |
9c1b96e3 AK |
533 | |
534 | Capability: basic | |
572e0929 | 535 | Architectures: all |
9c1b96e3 AK |
536 | Type: vcpu ioctl |
537 | Parameters: struct kvm_signal_mask (in) | |
538 | Returns: 0 on success, -1 on error | |
539 | ||
540 | Defines which signals are blocked during execution of KVM_RUN. This | |
541 | signal mask temporarily overrides the threads signal mask. Any | |
542 | unblocked signal received (except SIGKILL and SIGSTOP, which retain | |
543 | their traditional behaviour) will cause KVM_RUN to return with -EINTR. | |
544 | ||
545 | Note the signal will only be delivered if not blocked by the original | |
546 | signal mask. | |
547 | ||
548 | /* for KVM_SET_SIGNAL_MASK */ | |
549 | struct kvm_signal_mask { | |
550 | __u32 len; | |
551 | __u8 sigset[0]; | |
552 | }; | |
553 | ||
414fa985 | 554 | |
68ba6974 | 555 | 4.22 KVM_GET_FPU |
9c1b96e3 AK |
556 | |
557 | Capability: basic | |
558 | Architectures: x86 | |
559 | Type: vcpu ioctl | |
560 | Parameters: struct kvm_fpu (out) | |
561 | Returns: 0 on success, -1 on error | |
562 | ||
563 | Reads the floating point state from the vcpu. | |
564 | ||
565 | /* for KVM_GET_FPU and KVM_SET_FPU */ | |
566 | struct kvm_fpu { | |
567 | __u8 fpr[8][16]; | |
568 | __u16 fcw; | |
569 | __u16 fsw; | |
570 | __u8 ftwx; /* in fxsave format */ | |
571 | __u8 pad1; | |
572 | __u16 last_opcode; | |
573 | __u64 last_ip; | |
574 | __u64 last_dp; | |
575 | __u8 xmm[16][16]; | |
576 | __u32 mxcsr; | |
577 | __u32 pad2; | |
578 | }; | |
579 | ||
414fa985 | 580 | |
68ba6974 | 581 | 4.23 KVM_SET_FPU |
9c1b96e3 AK |
582 | |
583 | Capability: basic | |
584 | Architectures: x86 | |
585 | Type: vcpu ioctl | |
586 | Parameters: struct kvm_fpu (in) | |
587 | Returns: 0 on success, -1 on error | |
588 | ||
589 | Writes the floating point state to the vcpu. | |
590 | ||
591 | /* for KVM_GET_FPU and KVM_SET_FPU */ | |
592 | struct kvm_fpu { | |
593 | __u8 fpr[8][16]; | |
594 | __u16 fcw; | |
595 | __u16 fsw; | |
596 | __u8 ftwx; /* in fxsave format */ | |
597 | __u8 pad1; | |
598 | __u16 last_opcode; | |
599 | __u64 last_ip; | |
600 | __u64 last_dp; | |
601 | __u8 xmm[16][16]; | |
602 | __u32 mxcsr; | |
603 | __u32 pad2; | |
604 | }; | |
605 | ||
414fa985 | 606 | |
68ba6974 | 607 | 4.24 KVM_CREATE_IRQCHIP |
5dadbfd6 | 608 | |
84223598 | 609 | Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390) |
c32a4272 | 610 | Architectures: x86, ARM, arm64, s390 |
5dadbfd6 AK |
611 | Type: vm ioctl |
612 | Parameters: none | |
613 | Returns: 0 on success, -1 on error | |
614 | ||
ac3d3735 AP |
615 | Creates an interrupt controller model in the kernel. |
616 | On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up | |
617 | future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both | |
618 | PIC and IOAPIC; GSI 16-23 only go to the IOAPIC. | |
619 | On ARM/arm64, a GICv2 is created. Any other GIC versions require the usage of | |
620 | KVM_CREATE_DEVICE, which also supports creating a GICv2. Using | |
621 | KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2. | |
622 | On s390, a dummy irq routing table is created. | |
84223598 CH |
623 | |
624 | Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled | |
625 | before KVM_CREATE_IRQCHIP can be used. | |
5dadbfd6 | 626 | |
414fa985 | 627 | |
68ba6974 | 628 | 4.25 KVM_IRQ_LINE |
5dadbfd6 AK |
629 | |
630 | Capability: KVM_CAP_IRQCHIP | |
c32a4272 | 631 | Architectures: x86, arm, arm64 |
5dadbfd6 AK |
632 | Type: vm ioctl |
633 | Parameters: struct kvm_irq_level | |
634 | Returns: 0 on success, -1 on error | |
635 | ||
636 | Sets the level of a GSI input to the interrupt controller model in the kernel. | |
86ce8535 CD |
637 | On some architectures it is required that an interrupt controller model has |
638 | been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered | |
639 | interrupts require the level to be set to 1 and then back to 0. | |
640 | ||
100943c5 GS |
641 | On real hardware, interrupt pins can be active-low or active-high. This |
642 | does not matter for the level field of struct kvm_irq_level: 1 always | |
643 | means active (asserted), 0 means inactive (deasserted). | |
644 | ||
645 | x86 allows the operating system to program the interrupt polarity | |
646 | (active-low/active-high) for level-triggered interrupts, and KVM used | |
647 | to consider the polarity. However, due to bitrot in the handling of | |
648 | active-low interrupts, the above convention is now valid on x86 too. | |
649 | This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace | |
650 | should not present interrupts to the guest as active-low unless this | |
651 | capability is present (or unless it is not using the in-kernel irqchip, | |
652 | of course). | |
653 | ||
654 | ||
379e04c7 MZ |
655 | ARM/arm64 can signal an interrupt either at the CPU level, or at the |
656 | in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to | |
657 | use PPIs designated for specific cpus. The irq field is interpreted | |
658 | like this: | |
86ce8535 CD |
659 | |
660 | Â bits: | 31 ... 24 | 23 ... 16 | 15 ... 0 | | |
661 | field: | irq_type | vcpu_index | irq_id | | |
662 | ||
663 | The irq_type field has the following values: | |
664 | - irq_type[0]: out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ | |
665 | - irq_type[1]: in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.) | |
666 | (the vcpu_index field is ignored) | |
667 | - irq_type[2]: in-kernel GIC: PPI, irq_id between 16 and 31 (incl.) | |
668 | ||
669 | (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs) | |
670 | ||
100943c5 | 671 | In both cases, level is used to assert/deassert the line. |
5dadbfd6 AK |
672 | |
673 | struct kvm_irq_level { | |
674 | union { | |
675 | __u32 irq; /* GSI */ | |
676 | __s32 status; /* not used for KVM_IRQ_LEVEL */ | |
677 | }; | |
678 | __u32 level; /* 0 or 1 */ | |
679 | }; | |
680 | ||
414fa985 | 681 | |
68ba6974 | 682 | 4.26 KVM_GET_IRQCHIP |
5dadbfd6 AK |
683 | |
684 | Capability: KVM_CAP_IRQCHIP | |
c32a4272 | 685 | Architectures: x86 |
5dadbfd6 AK |
686 | Type: vm ioctl |
687 | Parameters: struct kvm_irqchip (in/out) | |
688 | Returns: 0 on success, -1 on error | |
689 | ||
690 | Reads the state of a kernel interrupt controller created with | |
691 | KVM_CREATE_IRQCHIP into a buffer provided by the caller. | |
692 | ||
693 | struct kvm_irqchip { | |
694 | __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ | |
695 | __u32 pad; | |
696 | union { | |
697 | char dummy[512]; /* reserving space */ | |
698 | struct kvm_pic_state pic; | |
699 | struct kvm_ioapic_state ioapic; | |
700 | } chip; | |
701 | }; | |
702 | ||
414fa985 | 703 | |
68ba6974 | 704 | 4.27 KVM_SET_IRQCHIP |
5dadbfd6 AK |
705 | |
706 | Capability: KVM_CAP_IRQCHIP | |
c32a4272 | 707 | Architectures: x86 |
5dadbfd6 AK |
708 | Type: vm ioctl |
709 | Parameters: struct kvm_irqchip (in) | |
710 | Returns: 0 on success, -1 on error | |
711 | ||
712 | Sets the state of a kernel interrupt controller created with | |
713 | KVM_CREATE_IRQCHIP from a buffer provided by the caller. | |
714 | ||
715 | struct kvm_irqchip { | |
716 | __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */ | |
717 | __u32 pad; | |
718 | union { | |
719 | char dummy[512]; /* reserving space */ | |
720 | struct kvm_pic_state pic; | |
721 | struct kvm_ioapic_state ioapic; | |
722 | } chip; | |
723 | }; | |
724 | ||
414fa985 | 725 | |
68ba6974 | 726 | 4.28 KVM_XEN_HVM_CONFIG |
ffde22ac ES |
727 | |
728 | Capability: KVM_CAP_XEN_HVM | |
729 | Architectures: x86 | |
730 | Type: vm ioctl | |
731 | Parameters: struct kvm_xen_hvm_config (in) | |
732 | Returns: 0 on success, -1 on error | |
733 | ||
734 | Sets the MSR that the Xen HVM guest uses to initialize its hypercall | |
735 | page, and provides the starting address and size of the hypercall | |
736 | blobs in userspace. When the guest writes the MSR, kvm copies one | |
737 | page of a blob (32- or 64-bit, depending on the vcpu mode) to guest | |
738 | memory. | |
739 | ||
740 | struct kvm_xen_hvm_config { | |
741 | __u32 flags; | |
742 | __u32 msr; | |
743 | __u64 blob_addr_32; | |
744 | __u64 blob_addr_64; | |
745 | __u8 blob_size_32; | |
746 | __u8 blob_size_64; | |
747 | __u8 pad2[30]; | |
748 | }; | |
749 | ||
414fa985 | 750 | |
68ba6974 | 751 | 4.29 KVM_GET_CLOCK |
afbcf7ab GC |
752 | |
753 | Capability: KVM_CAP_ADJUST_CLOCK | |
754 | Architectures: x86 | |
755 | Type: vm ioctl | |
756 | Parameters: struct kvm_clock_data (out) | |
757 | Returns: 0 on success, -1 on error | |
758 | ||
759 | Gets the current timestamp of kvmclock as seen by the current guest. In | |
760 | conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios | |
761 | such as migration. | |
762 | ||
763 | struct kvm_clock_data { | |
764 | __u64 clock; /* kvmclock current value */ | |
765 | __u32 flags; | |
766 | __u32 pad[9]; | |
767 | }; | |
768 | ||
414fa985 | 769 | |
68ba6974 | 770 | 4.30 KVM_SET_CLOCK |
afbcf7ab GC |
771 | |
772 | Capability: KVM_CAP_ADJUST_CLOCK | |
773 | Architectures: x86 | |
774 | Type: vm ioctl | |
775 | Parameters: struct kvm_clock_data (in) | |
776 | Returns: 0 on success, -1 on error | |
777 | ||
2044892d | 778 | Sets the current timestamp of kvmclock to the value specified in its parameter. |
afbcf7ab GC |
779 | In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios |
780 | such as migration. | |
781 | ||
782 | struct kvm_clock_data { | |
783 | __u64 clock; /* kvmclock current value */ | |
784 | __u32 flags; | |
785 | __u32 pad[9]; | |
786 | }; | |
787 | ||
414fa985 | 788 | |
68ba6974 | 789 | 4.31 KVM_GET_VCPU_EVENTS |
3cfc3092 JK |
790 | |
791 | Capability: KVM_CAP_VCPU_EVENTS | |
48005f64 | 792 | Extended by: KVM_CAP_INTR_SHADOW |
3cfc3092 JK |
793 | Architectures: x86 |
794 | Type: vm ioctl | |
795 | Parameters: struct kvm_vcpu_event (out) | |
796 | Returns: 0 on success, -1 on error | |
797 | ||
798 | Gets currently pending exceptions, interrupts, and NMIs as well as related | |
799 | states of the vcpu. | |
800 | ||
801 | struct kvm_vcpu_events { | |
802 | struct { | |
803 | __u8 injected; | |
804 | __u8 nr; | |
805 | __u8 has_error_code; | |
806 | __u8 pad; | |
807 | __u32 error_code; | |
808 | } exception; | |
809 | struct { | |
810 | __u8 injected; | |
811 | __u8 nr; | |
812 | __u8 soft; | |
48005f64 | 813 | __u8 shadow; |
3cfc3092 JK |
814 | } interrupt; |
815 | struct { | |
816 | __u8 injected; | |
817 | __u8 pending; | |
818 | __u8 masked; | |
819 | __u8 pad; | |
820 | } nmi; | |
821 | __u32 sipi_vector; | |
dab4b911 | 822 | __u32 flags; |
3cfc3092 JK |
823 | }; |
824 | ||
48005f64 JK |
825 | KVM_VCPUEVENT_VALID_SHADOW may be set in the flags field to signal that |
826 | interrupt.shadow contains a valid state. Otherwise, this field is undefined. | |
827 | ||
414fa985 | 828 | |
68ba6974 | 829 | 4.32 KVM_SET_VCPU_EVENTS |
3cfc3092 JK |
830 | |
831 | Capability: KVM_CAP_VCPU_EVENTS | |
48005f64 | 832 | Extended by: KVM_CAP_INTR_SHADOW |
3cfc3092 JK |
833 | Architectures: x86 |
834 | Type: vm ioctl | |
835 | Parameters: struct kvm_vcpu_event (in) | |
836 | Returns: 0 on success, -1 on error | |
837 | ||
838 | Set pending exceptions, interrupts, and NMIs as well as related states of the | |
839 | vcpu. | |
840 | ||
841 | See KVM_GET_VCPU_EVENTS for the data structure. | |
842 | ||
dab4b911 JK |
843 | Fields that may be modified asynchronously by running VCPUs can be excluded |
844 | from the update. These fields are nmi.pending and sipi_vector. Keep the | |
845 | corresponding bits in the flags field cleared to suppress overwriting the | |
846 | current in-kernel state. The bits are: | |
847 | ||
848 | KVM_VCPUEVENT_VALID_NMI_PENDING - transfer nmi.pending to the kernel | |
849 | KVM_VCPUEVENT_VALID_SIPI_VECTOR - transfer sipi_vector | |
850 | ||
48005f64 JK |
851 | If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in |
852 | the flags field to signal that interrupt.shadow contains a valid state and | |
853 | shall be written into the VCPU. | |
854 | ||
414fa985 | 855 | |
68ba6974 | 856 | 4.33 KVM_GET_DEBUGREGS |
a1efbe77 JK |
857 | |
858 | Capability: KVM_CAP_DEBUGREGS | |
859 | Architectures: x86 | |
860 | Type: vm ioctl | |
861 | Parameters: struct kvm_debugregs (out) | |
862 | Returns: 0 on success, -1 on error | |
863 | ||
864 | Reads debug registers from the vcpu. | |
865 | ||
866 | struct kvm_debugregs { | |
867 | __u64 db[4]; | |
868 | __u64 dr6; | |
869 | __u64 dr7; | |
870 | __u64 flags; | |
871 | __u64 reserved[9]; | |
872 | }; | |
873 | ||
414fa985 | 874 | |
68ba6974 | 875 | 4.34 KVM_SET_DEBUGREGS |
a1efbe77 JK |
876 | |
877 | Capability: KVM_CAP_DEBUGREGS | |
878 | Architectures: x86 | |
879 | Type: vm ioctl | |
880 | Parameters: struct kvm_debugregs (in) | |
881 | Returns: 0 on success, -1 on error | |
882 | ||
883 | Writes debug registers into the vcpu. | |
884 | ||
885 | See KVM_GET_DEBUGREGS for the data structure. The flags field is unused | |
886 | yet and must be cleared on entry. | |
887 | ||
414fa985 | 888 | |
68ba6974 | 889 | 4.35 KVM_SET_USER_MEMORY_REGION |
0f2d8f4d AK |
890 | |
891 | Capability: KVM_CAP_USER_MEM | |
892 | Architectures: all | |
893 | Type: vm ioctl | |
894 | Parameters: struct kvm_userspace_memory_region (in) | |
895 | Returns: 0 on success, -1 on error | |
896 | ||
897 | struct kvm_userspace_memory_region { | |
898 | __u32 slot; | |
899 | __u32 flags; | |
900 | __u64 guest_phys_addr; | |
901 | __u64 memory_size; /* bytes */ | |
902 | __u64 userspace_addr; /* start of the userspace allocated memory */ | |
903 | }; | |
904 | ||
905 | /* for kvm_memory_region::flags */ | |
4d8b81ab XG |
906 | #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0) |
907 | #define KVM_MEM_READONLY (1UL << 1) | |
0f2d8f4d AK |
908 | |
909 | This ioctl allows the user to create or modify a guest physical memory | |
910 | slot. When changing an existing slot, it may be moved in the guest | |
911 | physical memory space, or its flags may be modified. It may not be | |
912 | resized. Slots may not overlap in guest physical address space. | |
913 | ||
914 | Memory for the region is taken starting at the address denoted by the | |
915 | field userspace_addr, which must point at user addressable memory for | |
916 | the entire memory slot size. Any object may back this memory, including | |
917 | anonymous memory, ordinary files, and hugetlbfs. | |
918 | ||
919 | It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr | |
920 | be identical. This allows large pages in the guest to be backed by large | |
921 | pages in the host. | |
922 | ||
75d61fbc TY |
923 | The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and |
924 | KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of | |
925 | writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to | |
926 | use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it, | |
927 | to make a new slot read-only. In this case, writes to this memory will be | |
928 | posted to userspace as KVM_EXIT_MMIO exits. | |
7efd8fa1 JK |
929 | |
930 | When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of | |
931 | the memory region are automatically reflected into the guest. For example, an | |
932 | mmap() that affects the region will be made visible immediately. Another | |
933 | example is madvise(MADV_DROP). | |
0f2d8f4d AK |
934 | |
935 | It is recommended to use this API instead of the KVM_SET_MEMORY_REGION ioctl. | |
936 | The KVM_SET_MEMORY_REGION does not allow fine grained control over memory | |
937 | allocation and is deprecated. | |
3cfc3092 | 938 | |
414fa985 | 939 | |
68ba6974 | 940 | 4.36 KVM_SET_TSS_ADDR |
8a5416db AK |
941 | |
942 | Capability: KVM_CAP_SET_TSS_ADDR | |
943 | Architectures: x86 | |
944 | Type: vm ioctl | |
945 | Parameters: unsigned long tss_address (in) | |
946 | Returns: 0 on success, -1 on error | |
947 | ||
948 | This ioctl defines the physical address of a three-page region in the guest | |
949 | physical address space. The region must be within the first 4GB of the | |
950 | guest physical address space and must not conflict with any memory slot | |
951 | or any mmio address. The guest may malfunction if it accesses this memory | |
952 | region. | |
953 | ||
954 | This ioctl is required on Intel-based hosts. This is needed on Intel hardware | |
955 | because of a quirk in the virtualization implementation (see the internals | |
956 | documentation when it pops into existence). | |
957 | ||
414fa985 | 958 | |
68ba6974 | 959 | 4.37 KVM_ENABLE_CAP |
71fbfd5f | 960 | |
d938dc55 | 961 | Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM |
d6712df9 | 962 | Architectures: ppc, s390 |
d938dc55 | 963 | Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM) |
71fbfd5f AG |
964 | Parameters: struct kvm_enable_cap (in) |
965 | Returns: 0 on success; -1 on error | |
966 | ||
967 | +Not all extensions are enabled by default. Using this ioctl the application | |
968 | can enable an extension, making it available to the guest. | |
969 | ||
970 | On systems that do not support this ioctl, it always fails. On systems that | |
971 | do support it, it only works for extensions that are supported for enablement. | |
972 | ||
973 | To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should | |
974 | be used. | |
975 | ||
976 | struct kvm_enable_cap { | |
977 | /* in */ | |
978 | __u32 cap; | |
979 | ||
980 | The capability that is supposed to get enabled. | |
981 | ||
982 | __u32 flags; | |
983 | ||
984 | A bitfield indicating future enhancements. Has to be 0 for now. | |
985 | ||
986 | __u64 args[4]; | |
987 | ||
988 | Arguments for enabling a feature. If a feature needs initial values to | |
989 | function properly, this is the place to put them. | |
990 | ||
991 | __u8 pad[64]; | |
992 | }; | |
993 | ||
d938dc55 CH |
994 | The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl |
995 | for vm-wide capabilities. | |
414fa985 | 996 | |
68ba6974 | 997 | 4.38 KVM_GET_MP_STATE |
b843f065 AK |
998 | |
999 | Capability: KVM_CAP_MP_STATE | |
c32a4272 | 1000 | Architectures: x86, s390 |
b843f065 AK |
1001 | Type: vcpu ioctl |
1002 | Parameters: struct kvm_mp_state (out) | |
1003 | Returns: 0 on success; -1 on error | |
1004 | ||
1005 | struct kvm_mp_state { | |
1006 | __u32 mp_state; | |
1007 | }; | |
1008 | ||
1009 | Returns the vcpu's current "multiprocessing state" (though also valid on | |
1010 | uniprocessor guests). | |
1011 | ||
1012 | Possible values are: | |
1013 | ||
c32a4272 | 1014 | - KVM_MP_STATE_RUNNABLE: the vcpu is currently running [x86] |
b843f065 | 1015 | - KVM_MP_STATE_UNINITIALIZED: the vcpu is an application processor (AP) |
c32a4272 | 1016 | which has not yet received an INIT signal [x86] |
b843f065 | 1017 | - KVM_MP_STATE_INIT_RECEIVED: the vcpu has received an INIT signal, and is |
c32a4272 | 1018 | now ready for a SIPI [x86] |
b843f065 | 1019 | - KVM_MP_STATE_HALTED: the vcpu has executed a HLT instruction and |
c32a4272 | 1020 | is waiting for an interrupt [x86] |
b843f065 | 1021 | - KVM_MP_STATE_SIPI_RECEIVED: the vcpu has just received a SIPI (vector |
c32a4272 | 1022 | accessible via KVM_GET_VCPU_EVENTS) [x86] |
6352e4d2 DH |
1023 | - KVM_MP_STATE_STOPPED: the vcpu is stopped [s390] |
1024 | - KVM_MP_STATE_CHECK_STOP: the vcpu is in a special error state [s390] | |
1025 | - KVM_MP_STATE_OPERATING: the vcpu is operating (running or halted) | |
1026 | [s390] | |
1027 | - KVM_MP_STATE_LOAD: the vcpu is in a special load/startup state | |
1028 | [s390] | |
b843f065 | 1029 | |
c32a4272 | 1030 | On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an |
0b4820d6 DH |
1031 | in-kernel irqchip, the multiprocessing state must be maintained by userspace on |
1032 | these architectures. | |
b843f065 | 1033 | |
414fa985 | 1034 | |
68ba6974 | 1035 | 4.39 KVM_SET_MP_STATE |
b843f065 AK |
1036 | |
1037 | Capability: KVM_CAP_MP_STATE | |
c32a4272 | 1038 | Architectures: x86, s390 |
b843f065 AK |
1039 | Type: vcpu ioctl |
1040 | Parameters: struct kvm_mp_state (in) | |
1041 | Returns: 0 on success; -1 on error | |
1042 | ||
1043 | Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for | |
1044 | arguments. | |
1045 | ||
c32a4272 | 1046 | On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an |
0b4820d6 DH |
1047 | in-kernel irqchip, the multiprocessing state must be maintained by userspace on |
1048 | these architectures. | |
b843f065 | 1049 | |
414fa985 | 1050 | |
68ba6974 | 1051 | 4.40 KVM_SET_IDENTITY_MAP_ADDR |
47dbb84f AK |
1052 | |
1053 | Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR | |
1054 | Architectures: x86 | |
1055 | Type: vm ioctl | |
1056 | Parameters: unsigned long identity (in) | |
1057 | Returns: 0 on success, -1 on error | |
1058 | ||
1059 | This ioctl defines the physical address of a one-page region in the guest | |
1060 | physical address space. The region must be within the first 4GB of the | |
1061 | guest physical address space and must not conflict with any memory slot | |
1062 | or any mmio address. The guest may malfunction if it accesses this memory | |
1063 | region. | |
1064 | ||
1065 | This ioctl is required on Intel-based hosts. This is needed on Intel hardware | |
1066 | because of a quirk in the virtualization implementation (see the internals | |
1067 | documentation when it pops into existence). | |
1068 | ||
414fa985 | 1069 | |
68ba6974 | 1070 | 4.41 KVM_SET_BOOT_CPU_ID |
57bc24cf AK |
1071 | |
1072 | Capability: KVM_CAP_SET_BOOT_CPU_ID | |
c32a4272 | 1073 | Architectures: x86 |
57bc24cf AK |
1074 | Type: vm ioctl |
1075 | Parameters: unsigned long vcpu_id | |
1076 | Returns: 0 on success, -1 on error | |
1077 | ||
1078 | Define which vcpu is the Bootstrap Processor (BSP). Values are the same | |
1079 | as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default | |
1080 | is vcpu 0. | |
1081 | ||
414fa985 | 1082 | |
68ba6974 | 1083 | 4.42 KVM_GET_XSAVE |
2d5b5a66 SY |
1084 | |
1085 | Capability: KVM_CAP_XSAVE | |
1086 | Architectures: x86 | |
1087 | Type: vcpu ioctl | |
1088 | Parameters: struct kvm_xsave (out) | |
1089 | Returns: 0 on success, -1 on error | |
1090 | ||
1091 | struct kvm_xsave { | |
1092 | __u32 region[1024]; | |
1093 | }; | |
1094 | ||
1095 | This ioctl would copy current vcpu's xsave struct to the userspace. | |
1096 | ||
414fa985 | 1097 | |
68ba6974 | 1098 | 4.43 KVM_SET_XSAVE |
2d5b5a66 SY |
1099 | |
1100 | Capability: KVM_CAP_XSAVE | |
1101 | Architectures: x86 | |
1102 | Type: vcpu ioctl | |
1103 | Parameters: struct kvm_xsave (in) | |
1104 | Returns: 0 on success, -1 on error | |
1105 | ||
1106 | struct kvm_xsave { | |
1107 | __u32 region[1024]; | |
1108 | }; | |
1109 | ||
1110 | This ioctl would copy userspace's xsave struct to the kernel. | |
1111 | ||
414fa985 | 1112 | |
68ba6974 | 1113 | 4.44 KVM_GET_XCRS |
2d5b5a66 SY |
1114 | |
1115 | Capability: KVM_CAP_XCRS | |
1116 | Architectures: x86 | |
1117 | Type: vcpu ioctl | |
1118 | Parameters: struct kvm_xcrs (out) | |
1119 | Returns: 0 on success, -1 on error | |
1120 | ||
1121 | struct kvm_xcr { | |
1122 | __u32 xcr; | |
1123 | __u32 reserved; | |
1124 | __u64 value; | |
1125 | }; | |
1126 | ||
1127 | struct kvm_xcrs { | |
1128 | __u32 nr_xcrs; | |
1129 | __u32 flags; | |
1130 | struct kvm_xcr xcrs[KVM_MAX_XCRS]; | |
1131 | __u64 padding[16]; | |
1132 | }; | |
1133 | ||
1134 | This ioctl would copy current vcpu's xcrs to the userspace. | |
1135 | ||
414fa985 | 1136 | |
68ba6974 | 1137 | 4.45 KVM_SET_XCRS |
2d5b5a66 SY |
1138 | |
1139 | Capability: KVM_CAP_XCRS | |
1140 | Architectures: x86 | |
1141 | Type: vcpu ioctl | |
1142 | Parameters: struct kvm_xcrs (in) | |
1143 | Returns: 0 on success, -1 on error | |
1144 | ||
1145 | struct kvm_xcr { | |
1146 | __u32 xcr; | |
1147 | __u32 reserved; | |
1148 | __u64 value; | |
1149 | }; | |
1150 | ||
1151 | struct kvm_xcrs { | |
1152 | __u32 nr_xcrs; | |
1153 | __u32 flags; | |
1154 | struct kvm_xcr xcrs[KVM_MAX_XCRS]; | |
1155 | __u64 padding[16]; | |
1156 | }; | |
1157 | ||
1158 | This ioctl would set vcpu's xcr to the value userspace specified. | |
1159 | ||
414fa985 | 1160 | |
68ba6974 | 1161 | 4.46 KVM_GET_SUPPORTED_CPUID |
d153513d AK |
1162 | |
1163 | Capability: KVM_CAP_EXT_CPUID | |
1164 | Architectures: x86 | |
1165 | Type: system ioctl | |
1166 | Parameters: struct kvm_cpuid2 (in/out) | |
1167 | Returns: 0 on success, -1 on error | |
1168 | ||
1169 | struct kvm_cpuid2 { | |
1170 | __u32 nent; | |
1171 | __u32 padding; | |
1172 | struct kvm_cpuid_entry2 entries[0]; | |
1173 | }; | |
1174 | ||
9c15bb1d BP |
1175 | #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) |
1176 | #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) | |
1177 | #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) | |
d153513d AK |
1178 | |
1179 | struct kvm_cpuid_entry2 { | |
1180 | __u32 function; | |
1181 | __u32 index; | |
1182 | __u32 flags; | |
1183 | __u32 eax; | |
1184 | __u32 ebx; | |
1185 | __u32 ecx; | |
1186 | __u32 edx; | |
1187 | __u32 padding[3]; | |
1188 | }; | |
1189 | ||
1190 | This ioctl returns x86 cpuid features which are supported by both the hardware | |
1191 | and kvm. Userspace can use the information returned by this ioctl to | |
1192 | construct cpuid information (for KVM_SET_CPUID2) that is consistent with | |
1193 | hardware, kernel, and userspace capabilities, and with user requirements (for | |
1194 | example, the user may wish to constrain cpuid to emulate older hardware, | |
1195 | or for feature consistency across a cluster). | |
1196 | ||
1197 | Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure | |
1198 | with the 'nent' field indicating the number of entries in the variable-size | |
1199 | array 'entries'. If the number of entries is too low to describe the cpu | |
1200 | capabilities, an error (E2BIG) is returned. If the number is too high, | |
1201 | the 'nent' field is adjusted and an error (ENOMEM) is returned. If the | |
1202 | number is just right, the 'nent' field is adjusted to the number of valid | |
1203 | entries in the 'entries' array, which is then filled. | |
1204 | ||
1205 | The entries returned are the host cpuid as returned by the cpuid instruction, | |
c39cbd2a AK |
1206 | with unknown or unsupported features masked out. Some features (for example, |
1207 | x2apic), may not be present in the host cpu, but are exposed by kvm if it can | |
1208 | emulate them efficiently. The fields in each entry are defined as follows: | |
d153513d AK |
1209 | |
1210 | function: the eax value used to obtain the entry | |
1211 | index: the ecx value used to obtain the entry (for entries that are | |
1212 | affected by ecx) | |
1213 | flags: an OR of zero or more of the following: | |
1214 | KVM_CPUID_FLAG_SIGNIFCANT_INDEX: | |
1215 | if the index field is valid | |
1216 | KVM_CPUID_FLAG_STATEFUL_FUNC: | |
1217 | if cpuid for this function returns different values for successive | |
1218 | invocations; there will be several entries with the same function, | |
1219 | all with this flag set | |
1220 | KVM_CPUID_FLAG_STATE_READ_NEXT: | |
1221 | for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is | |
1222 | the first entry to be read by a cpu | |
1223 | eax, ebx, ecx, edx: the values returned by the cpuid instruction for | |
1224 | this function/index combination | |
1225 | ||
4d25a066 JK |
1226 | The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned |
1227 | as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC | |
1228 | support. Instead it is reported via | |
1229 | ||
1230 | ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER) | |
1231 | ||
1232 | if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the | |
1233 | feature in userspace, then you can enable the feature for KVM_SET_CPUID2. | |
1234 | ||
414fa985 | 1235 | |
68ba6974 | 1236 | 4.47 KVM_PPC_GET_PVINFO |
15711e9c AG |
1237 | |
1238 | Capability: KVM_CAP_PPC_GET_PVINFO | |
1239 | Architectures: ppc | |
1240 | Type: vm ioctl | |
1241 | Parameters: struct kvm_ppc_pvinfo (out) | |
1242 | Returns: 0 on success, !0 on error | |
1243 | ||
1244 | struct kvm_ppc_pvinfo { | |
1245 | __u32 flags; | |
1246 | __u32 hcall[4]; | |
1247 | __u8 pad[108]; | |
1248 | }; | |
1249 | ||
1250 | This ioctl fetches PV specific information that need to be passed to the guest | |
1251 | using the device tree or other means from vm context. | |
1252 | ||
9202e076 | 1253 | The hcall array defines 4 instructions that make up a hypercall. |
15711e9c AG |
1254 | |
1255 | If any additional field gets added to this structure later on, a bit for that | |
1256 | additional piece of information will be set in the flags bitmap. | |
1257 | ||
9202e076 LYB |
1258 | The flags bitmap is defined as: |
1259 | ||
1260 | /* the host supports the ePAPR idle hcall | |
1261 | #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0) | |
414fa985 | 1262 | |
68ba6974 | 1263 | 4.48 KVM_ASSIGN_PCI_DEVICE |
49f48172 | 1264 | |
7f05db6a | 1265 | Capability: none |
c32a4272 | 1266 | Architectures: x86 |
49f48172 JK |
1267 | Type: vm ioctl |
1268 | Parameters: struct kvm_assigned_pci_dev (in) | |
1269 | Returns: 0 on success, -1 on error | |
1270 | ||
1271 | Assigns a host PCI device to the VM. | |
1272 | ||
1273 | struct kvm_assigned_pci_dev { | |
1274 | __u32 assigned_dev_id; | |
1275 | __u32 busnr; | |
1276 | __u32 devfn; | |
1277 | __u32 flags; | |
1278 | __u32 segnr; | |
1279 | union { | |
1280 | __u32 reserved[11]; | |
1281 | }; | |
1282 | }; | |
1283 | ||
1284 | The PCI device is specified by the triple segnr, busnr, and devfn. | |
1285 | Identification in succeeding service requests is done via assigned_dev_id. The | |
1286 | following flags are specified: | |
1287 | ||
1288 | /* Depends on KVM_CAP_IOMMU */ | |
1289 | #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0) | |
07700a94 JK |
1290 | /* The following two depend on KVM_CAP_PCI_2_3 */ |
1291 | #define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1) | |
1292 | #define KVM_DEV_ASSIGN_MASK_INTX (1 << 2) | |
1293 | ||
1294 | If KVM_DEV_ASSIGN_PCI_2_3 is set, the kernel will manage legacy INTx interrupts | |
1295 | via the PCI-2.3-compliant device-level mask, thus enable IRQ sharing with other | |
1296 | assigned devices or host devices. KVM_DEV_ASSIGN_MASK_INTX specifies the | |
1297 | guest's view on the INTx mask, see KVM_ASSIGN_SET_INTX_MASK for details. | |
49f48172 | 1298 | |
42387373 AW |
1299 | The KVM_DEV_ASSIGN_ENABLE_IOMMU flag is a mandatory option to ensure |
1300 | isolation of the device. Usages not specifying this flag are deprecated. | |
1301 | ||
3d27e23b AW |
1302 | Only PCI header type 0 devices with PCI BAR resources are supported by |
1303 | device assignment. The user requesting this ioctl must have read/write | |
1304 | access to the PCI sysfs resource files associated with the device. | |
1305 | ||
7f05db6a MT |
1306 | Errors: |
1307 | ENOTTY: kernel does not support this ioctl | |
1308 | ||
1309 | Other error conditions may be defined by individual device types or | |
1310 | have their standard meanings. | |
1311 | ||
414fa985 | 1312 | |
68ba6974 | 1313 | 4.49 KVM_DEASSIGN_PCI_DEVICE |
49f48172 | 1314 | |
7f05db6a | 1315 | Capability: none |
c32a4272 | 1316 | Architectures: x86 |
49f48172 JK |
1317 | Type: vm ioctl |
1318 | Parameters: struct kvm_assigned_pci_dev (in) | |
1319 | Returns: 0 on success, -1 on error | |
1320 | ||
1321 | Ends PCI device assignment, releasing all associated resources. | |
1322 | ||
7f05db6a | 1323 | See KVM_ASSIGN_PCI_DEVICE for the data structure. Only assigned_dev_id is |
49f48172 JK |
1324 | used in kvm_assigned_pci_dev to identify the device. |
1325 | ||
7f05db6a MT |
1326 | Errors: |
1327 | ENOTTY: kernel does not support this ioctl | |
1328 | ||
1329 | Other error conditions may be defined by individual device types or | |
1330 | have their standard meanings. | |
414fa985 | 1331 | |
68ba6974 | 1332 | 4.50 KVM_ASSIGN_DEV_IRQ |
49f48172 JK |
1333 | |
1334 | Capability: KVM_CAP_ASSIGN_DEV_IRQ | |
c32a4272 | 1335 | Architectures: x86 |
49f48172 JK |
1336 | Type: vm ioctl |
1337 | Parameters: struct kvm_assigned_irq (in) | |
1338 | Returns: 0 on success, -1 on error | |
1339 | ||
1340 | Assigns an IRQ to a passed-through device. | |
1341 | ||
1342 | struct kvm_assigned_irq { | |
1343 | __u32 assigned_dev_id; | |
91e3d71d | 1344 | __u32 host_irq; /* ignored (legacy field) */ |
49f48172 JK |
1345 | __u32 guest_irq; |
1346 | __u32 flags; | |
1347 | union { | |
49f48172 JK |
1348 | __u32 reserved[12]; |
1349 | }; | |
1350 | }; | |
1351 | ||
1352 | The following flags are defined: | |
1353 | ||
1354 | #define KVM_DEV_IRQ_HOST_INTX (1 << 0) | |
1355 | #define KVM_DEV_IRQ_HOST_MSI (1 << 1) | |
1356 | #define KVM_DEV_IRQ_HOST_MSIX (1 << 2) | |
1357 | ||
1358 | #define KVM_DEV_IRQ_GUEST_INTX (1 << 8) | |
1359 | #define KVM_DEV_IRQ_GUEST_MSI (1 << 9) | |
1360 | #define KVM_DEV_IRQ_GUEST_MSIX (1 << 10) | |
1361 | ||
1362 | It is not valid to specify multiple types per host or guest IRQ. However, the | |
1363 | IRQ type of host and guest can differ or can even be null. | |
1364 | ||
7f05db6a MT |
1365 | Errors: |
1366 | ENOTTY: kernel does not support this ioctl | |
1367 | ||
1368 | Other error conditions may be defined by individual device types or | |
1369 | have their standard meanings. | |
1370 | ||
414fa985 | 1371 | |
68ba6974 | 1372 | 4.51 KVM_DEASSIGN_DEV_IRQ |
49f48172 JK |
1373 | |
1374 | Capability: KVM_CAP_ASSIGN_DEV_IRQ | |
c32a4272 | 1375 | Architectures: x86 |
49f48172 JK |
1376 | Type: vm ioctl |
1377 | Parameters: struct kvm_assigned_irq (in) | |
1378 | Returns: 0 on success, -1 on error | |
1379 | ||
1380 | Ends an IRQ assignment to a passed-through device. | |
1381 | ||
1382 | See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified | |
1383 | by assigned_dev_id, flags must correspond to the IRQ type specified on | |
1384 | KVM_ASSIGN_DEV_IRQ. Partial deassignment of host or guest IRQ is allowed. | |
1385 | ||
414fa985 | 1386 | |
68ba6974 | 1387 | 4.52 KVM_SET_GSI_ROUTING |
49f48172 JK |
1388 | |
1389 | Capability: KVM_CAP_IRQ_ROUTING | |
c32a4272 | 1390 | Architectures: x86 s390 |
49f48172 JK |
1391 | Type: vm ioctl |
1392 | Parameters: struct kvm_irq_routing (in) | |
1393 | Returns: 0 on success, -1 on error | |
1394 | ||
1395 | Sets the GSI routing table entries, overwriting any previously set entries. | |
1396 | ||
1397 | struct kvm_irq_routing { | |
1398 | __u32 nr; | |
1399 | __u32 flags; | |
1400 | struct kvm_irq_routing_entry entries[0]; | |
1401 | }; | |
1402 | ||
1403 | No flags are specified so far, the corresponding field must be set to zero. | |
1404 | ||
1405 | struct kvm_irq_routing_entry { | |
1406 | __u32 gsi; | |
1407 | __u32 type; | |
1408 | __u32 flags; | |
1409 | __u32 pad; | |
1410 | union { | |
1411 | struct kvm_irq_routing_irqchip irqchip; | |
1412 | struct kvm_irq_routing_msi msi; | |
84223598 | 1413 | struct kvm_irq_routing_s390_adapter adapter; |
49f48172 JK |
1414 | __u32 pad[8]; |
1415 | } u; | |
1416 | }; | |
1417 | ||
1418 | /* gsi routing entry types */ | |
1419 | #define KVM_IRQ_ROUTING_IRQCHIP 1 | |
1420 | #define KVM_IRQ_ROUTING_MSI 2 | |
84223598 | 1421 | #define KVM_IRQ_ROUTING_S390_ADAPTER 3 |
49f48172 JK |
1422 | |
1423 | No flags are specified so far, the corresponding field must be set to zero. | |
1424 | ||
1425 | struct kvm_irq_routing_irqchip { | |
1426 | __u32 irqchip; | |
1427 | __u32 pin; | |
1428 | }; | |
1429 | ||
1430 | struct kvm_irq_routing_msi { | |
1431 | __u32 address_lo; | |
1432 | __u32 address_hi; | |
1433 | __u32 data; | |
1434 | __u32 pad; | |
1435 | }; | |
1436 | ||
84223598 CH |
1437 | struct kvm_irq_routing_s390_adapter { |
1438 | __u64 ind_addr; | |
1439 | __u64 summary_addr; | |
1440 | __u64 ind_offset; | |
1441 | __u32 summary_offset; | |
1442 | __u32 adapter_id; | |
1443 | }; | |
1444 | ||
414fa985 | 1445 | |
68ba6974 | 1446 | 4.53 KVM_ASSIGN_SET_MSIX_NR |
49f48172 | 1447 | |
7f05db6a | 1448 | Capability: none |
c32a4272 | 1449 | Architectures: x86 |
49f48172 JK |
1450 | Type: vm ioctl |
1451 | Parameters: struct kvm_assigned_msix_nr (in) | |
1452 | Returns: 0 on success, -1 on error | |
1453 | ||
58f0964e JK |
1454 | Set the number of MSI-X interrupts for an assigned device. The number is |
1455 | reset again by terminating the MSI-X assignment of the device via | |
1456 | KVM_DEASSIGN_DEV_IRQ. Calling this service more than once at any earlier | |
1457 | point will fail. | |
49f48172 JK |
1458 | |
1459 | struct kvm_assigned_msix_nr { | |
1460 | __u32 assigned_dev_id; | |
1461 | __u16 entry_nr; | |
1462 | __u16 padding; | |
1463 | }; | |
1464 | ||
1465 | #define KVM_MAX_MSIX_PER_DEV 256 | |
1466 | ||
414fa985 | 1467 | |
68ba6974 | 1468 | 4.54 KVM_ASSIGN_SET_MSIX_ENTRY |
49f48172 | 1469 | |
7f05db6a | 1470 | Capability: none |
c32a4272 | 1471 | Architectures: x86 |
49f48172 JK |
1472 | Type: vm ioctl |
1473 | Parameters: struct kvm_assigned_msix_entry (in) | |
1474 | Returns: 0 on success, -1 on error | |
1475 | ||
1476 | Specifies the routing of an MSI-X assigned device interrupt to a GSI. Setting | |
1477 | the GSI vector to zero means disabling the interrupt. | |
1478 | ||
1479 | struct kvm_assigned_msix_entry { | |
1480 | __u32 assigned_dev_id; | |
1481 | __u32 gsi; | |
1482 | __u16 entry; /* The index of entry in the MSI-X table */ | |
1483 | __u16 padding[3]; | |
1484 | }; | |
1485 | ||
7f05db6a MT |
1486 | Errors: |
1487 | ENOTTY: kernel does not support this ioctl | |
1488 | ||
1489 | Other error conditions may be defined by individual device types or | |
1490 | have their standard meanings. | |
1491 | ||
414fa985 JK |
1492 | |
1493 | 4.55 KVM_SET_TSC_KHZ | |
92a1f12d JR |
1494 | |
1495 | Capability: KVM_CAP_TSC_CONTROL | |
1496 | Architectures: x86 | |
1497 | Type: vcpu ioctl | |
1498 | Parameters: virtual tsc_khz | |
1499 | Returns: 0 on success, -1 on error | |
1500 | ||
1501 | Specifies the tsc frequency for the virtual machine. The unit of the | |
1502 | frequency is KHz. | |
1503 | ||
414fa985 JK |
1504 | |
1505 | 4.56 KVM_GET_TSC_KHZ | |
92a1f12d JR |
1506 | |
1507 | Capability: KVM_CAP_GET_TSC_KHZ | |
1508 | Architectures: x86 | |
1509 | Type: vcpu ioctl | |
1510 | Parameters: none | |
1511 | Returns: virtual tsc-khz on success, negative value on error | |
1512 | ||
1513 | Returns the tsc frequency of the guest. The unit of the return value is | |
1514 | KHz. If the host has unstable tsc this ioctl returns -EIO instead as an | |
1515 | error. | |
1516 | ||
414fa985 JK |
1517 | |
1518 | 4.57 KVM_GET_LAPIC | |
e7677933 AK |
1519 | |
1520 | Capability: KVM_CAP_IRQCHIP | |
1521 | Architectures: x86 | |
1522 | Type: vcpu ioctl | |
1523 | Parameters: struct kvm_lapic_state (out) | |
1524 | Returns: 0 on success, -1 on error | |
1525 | ||
1526 | #define KVM_APIC_REG_SIZE 0x400 | |
1527 | struct kvm_lapic_state { | |
1528 | char regs[KVM_APIC_REG_SIZE]; | |
1529 | }; | |
1530 | ||
1531 | Reads the Local APIC registers and copies them into the input argument. The | |
1532 | data format and layout are the same as documented in the architecture manual. | |
1533 | ||
414fa985 JK |
1534 | |
1535 | 4.58 KVM_SET_LAPIC | |
e7677933 AK |
1536 | |
1537 | Capability: KVM_CAP_IRQCHIP | |
1538 | Architectures: x86 | |
1539 | Type: vcpu ioctl | |
1540 | Parameters: struct kvm_lapic_state (in) | |
1541 | Returns: 0 on success, -1 on error | |
1542 | ||
1543 | #define KVM_APIC_REG_SIZE 0x400 | |
1544 | struct kvm_lapic_state { | |
1545 | char regs[KVM_APIC_REG_SIZE]; | |
1546 | }; | |
1547 | ||
df5cbb27 | 1548 | Copies the input argument into the Local APIC registers. The data format |
e7677933 AK |
1549 | and layout are the same as documented in the architecture manual. |
1550 | ||
414fa985 JK |
1551 | |
1552 | 4.59 KVM_IOEVENTFD | |
55399a02 SL |
1553 | |
1554 | Capability: KVM_CAP_IOEVENTFD | |
1555 | Architectures: all | |
1556 | Type: vm ioctl | |
1557 | Parameters: struct kvm_ioeventfd (in) | |
1558 | Returns: 0 on success, !0 on error | |
1559 | ||
1560 | This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address | |
1561 | within the guest. A guest write in the registered address will signal the | |
1562 | provided event instead of triggering an exit. | |
1563 | ||
1564 | struct kvm_ioeventfd { | |
1565 | __u64 datamatch; | |
1566 | __u64 addr; /* legal pio/mmio address */ | |
1567 | __u32 len; /* 1, 2, 4, or 8 bytes */ | |
1568 | __s32 fd; | |
1569 | __u32 flags; | |
1570 | __u8 pad[36]; | |
1571 | }; | |
1572 | ||
2b83451b CH |
1573 | For the special case of virtio-ccw devices on s390, the ioevent is matched |
1574 | to a subchannel/virtqueue tuple instead. | |
1575 | ||
55399a02 SL |
1576 | The following flags are defined: |
1577 | ||
1578 | #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch) | |
1579 | #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio) | |
1580 | #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign) | |
2b83451b CH |
1581 | #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \ |
1582 | (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify) | |
55399a02 SL |
1583 | |
1584 | If datamatch flag is set, the event will be signaled only if the written value | |
1585 | to the registered address is equal to datamatch in struct kvm_ioeventfd. | |
1586 | ||
2b83451b CH |
1587 | For virtio-ccw devices, addr contains the subchannel id and datamatch the |
1588 | virtqueue index. | |
1589 | ||
414fa985 JK |
1590 | |
1591 | 4.60 KVM_DIRTY_TLB | |
dc83b8bc SW |
1592 | |
1593 | Capability: KVM_CAP_SW_TLB | |
1594 | Architectures: ppc | |
1595 | Type: vcpu ioctl | |
1596 | Parameters: struct kvm_dirty_tlb (in) | |
1597 | Returns: 0 on success, -1 on error | |
1598 | ||
1599 | struct kvm_dirty_tlb { | |
1600 | __u64 bitmap; | |
1601 | __u32 num_dirty; | |
1602 | }; | |
1603 | ||
1604 | This must be called whenever userspace has changed an entry in the shared | |
1605 | TLB, prior to calling KVM_RUN on the associated vcpu. | |
1606 | ||
1607 | The "bitmap" field is the userspace address of an array. This array | |
1608 | consists of a number of bits, equal to the total number of TLB entries as | |
1609 | determined by the last successful call to KVM_CONFIG_TLB, rounded up to the | |
1610 | nearest multiple of 64. | |
1611 | ||
1612 | Each bit corresponds to one TLB entry, ordered the same as in the shared TLB | |
1613 | array. | |
1614 | ||
1615 | The array is little-endian: the bit 0 is the least significant bit of the | |
1616 | first byte, bit 8 is the least significant bit of the second byte, etc. | |
1617 | This avoids any complications with differing word sizes. | |
1618 | ||
1619 | The "num_dirty" field is a performance hint for KVM to determine whether it | |
1620 | should skip processing the bitmap and just invalidate everything. It must | |
1621 | be set to the number of set bits in the bitmap. | |
1622 | ||
414fa985 JK |
1623 | |
1624 | 4.61 KVM_ASSIGN_SET_INTX_MASK | |
07700a94 JK |
1625 | |
1626 | Capability: KVM_CAP_PCI_2_3 | |
1627 | Architectures: x86 | |
1628 | Type: vm ioctl | |
1629 | Parameters: struct kvm_assigned_pci_dev (in) | |
1630 | Returns: 0 on success, -1 on error | |
1631 | ||
1632 | Allows userspace to mask PCI INTx interrupts from the assigned device. The | |
1633 | kernel will not deliver INTx interrupts to the guest between setting and | |
1634 | clearing of KVM_ASSIGN_SET_INTX_MASK via this interface. This enables use of | |
1635 | and emulation of PCI 2.3 INTx disable command register behavior. | |
1636 | ||
1637 | This may be used for both PCI 2.3 devices supporting INTx disable natively and | |
1638 | older devices lacking this support. Userspace is responsible for emulating the | |
1639 | read value of the INTx disable bit in the guest visible PCI command register. | |
1640 | When modifying the INTx disable state, userspace should precede updating the | |
1641 | physical device command register by calling this ioctl to inform the kernel of | |
1642 | the new intended INTx mask state. | |
1643 | ||
1644 | Note that the kernel uses the device INTx disable bit to internally manage the | |
1645 | device interrupt state for PCI 2.3 devices. Reads of this register may | |
1646 | therefore not match the expected value. Writes should always use the guest | |
1647 | intended INTx disable value rather than attempting to read-copy-update the | |
1648 | current physical device state. Races between user and kernel updates to the | |
1649 | INTx disable bit are handled lazily in the kernel. It's possible the device | |
1650 | may generate unintended interrupts, but they will not be injected into the | |
1651 | guest. | |
1652 | ||
1653 | See KVM_ASSIGN_DEV_IRQ for the data structure. The target device is specified | |
1654 | by assigned_dev_id. In the flags field, only KVM_DEV_ASSIGN_MASK_INTX is | |
1655 | evaluated. | |
1656 | ||
414fa985 | 1657 | |
54738c09 DG |
1658 | 4.62 KVM_CREATE_SPAPR_TCE |
1659 | ||
1660 | Capability: KVM_CAP_SPAPR_TCE | |
1661 | Architectures: powerpc | |
1662 | Type: vm ioctl | |
1663 | Parameters: struct kvm_create_spapr_tce (in) | |
1664 | Returns: file descriptor for manipulating the created TCE table | |
1665 | ||
1666 | This creates a virtual TCE (translation control entry) table, which | |
1667 | is an IOMMU for PAPR-style virtual I/O. It is used to translate | |
1668 | logical addresses used in virtual I/O into guest physical addresses, | |
1669 | and provides a scatter/gather capability for PAPR virtual I/O. | |
1670 | ||
1671 | /* for KVM_CAP_SPAPR_TCE */ | |
1672 | struct kvm_create_spapr_tce { | |
1673 | __u64 liobn; | |
1674 | __u32 window_size; | |
1675 | }; | |
1676 | ||
1677 | The liobn field gives the logical IO bus number for which to create a | |
1678 | TCE table. The window_size field specifies the size of the DMA window | |
1679 | which this TCE table will translate - the table will contain one 64 | |
1680 | bit TCE entry for every 4kiB of the DMA window. | |
1681 | ||
1682 | When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE | |
1683 | table has been created using this ioctl(), the kernel will handle it | |
1684 | in real mode, updating the TCE table. H_PUT_TCE calls for other | |
1685 | liobns will cause a vm exit and must be handled by userspace. | |
1686 | ||
1687 | The return value is a file descriptor which can be passed to mmap(2) | |
1688 | to map the created TCE table into userspace. This lets userspace read | |
1689 | the entries written by kernel-handled H_PUT_TCE calls, and also lets | |
1690 | userspace update the TCE table directly which is useful in some | |
1691 | circumstances. | |
1692 | ||
414fa985 | 1693 | |
aa04b4cc PM |
1694 | 4.63 KVM_ALLOCATE_RMA |
1695 | ||
1696 | Capability: KVM_CAP_PPC_RMA | |
1697 | Architectures: powerpc | |
1698 | Type: vm ioctl | |
1699 | Parameters: struct kvm_allocate_rma (out) | |
1700 | Returns: file descriptor for mapping the allocated RMA | |
1701 | ||
1702 | This allocates a Real Mode Area (RMA) from the pool allocated at boot | |
1703 | time by the kernel. An RMA is a physically-contiguous, aligned region | |
1704 | of memory used on older POWER processors to provide the memory which | |
1705 | will be accessed by real-mode (MMU off) accesses in a KVM guest. | |
1706 | POWER processors support a set of sizes for the RMA that usually | |
1707 | includes 64MB, 128MB, 256MB and some larger powers of two. | |
1708 | ||
1709 | /* for KVM_ALLOCATE_RMA */ | |
1710 | struct kvm_allocate_rma { | |
1711 | __u64 rma_size; | |
1712 | }; | |
1713 | ||
1714 | The return value is a file descriptor which can be passed to mmap(2) | |
1715 | to map the allocated RMA into userspace. The mapped area can then be | |
1716 | passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the | |
1717 | RMA for a virtual machine. The size of the RMA in bytes (which is | |
1718 | fixed at host kernel boot time) is returned in the rma_size field of | |
1719 | the argument structure. | |
1720 | ||
1721 | The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl | |
1722 | is supported; 2 if the processor requires all virtual machines to have | |
1723 | an RMA, or 1 if the processor can use an RMA but doesn't require it, | |
1724 | because it supports the Virtual RMA (VRMA) facility. | |
1725 | ||
414fa985 | 1726 | |
3f745f1e AK |
1727 | 4.64 KVM_NMI |
1728 | ||
1729 | Capability: KVM_CAP_USER_NMI | |
1730 | Architectures: x86 | |
1731 | Type: vcpu ioctl | |
1732 | Parameters: none | |
1733 | Returns: 0 on success, -1 on error | |
1734 | ||
1735 | Queues an NMI on the thread's vcpu. Note this is well defined only | |
1736 | when KVM_CREATE_IRQCHIP has not been called, since this is an interface | |
1737 | between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP | |
1738 | has been called, this interface is completely emulated within the kernel. | |
1739 | ||
1740 | To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the | |
1741 | following algorithm: | |
1742 | ||
1743 | - pause the vpcu | |
1744 | - read the local APIC's state (KVM_GET_LAPIC) | |
1745 | - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1) | |
1746 | - if so, issue KVM_NMI | |
1747 | - resume the vcpu | |
1748 | ||
1749 | Some guests configure the LINT1 NMI input to cause a panic, aiding in | |
1750 | debugging. | |
1751 | ||
414fa985 | 1752 | |
e24ed81f | 1753 | 4.65 KVM_S390_UCAS_MAP |
27e0393f CO |
1754 | |
1755 | Capability: KVM_CAP_S390_UCONTROL | |
1756 | Architectures: s390 | |
1757 | Type: vcpu ioctl | |
1758 | Parameters: struct kvm_s390_ucas_mapping (in) | |
1759 | Returns: 0 in case of success | |
1760 | ||
1761 | The parameter is defined like this: | |
1762 | struct kvm_s390_ucas_mapping { | |
1763 | __u64 user_addr; | |
1764 | __u64 vcpu_addr; | |
1765 | __u64 length; | |
1766 | }; | |
1767 | ||
1768 | This ioctl maps the memory at "user_addr" with the length "length" to | |
1769 | the vcpu's address space starting at "vcpu_addr". All parameters need to | |
f884ab15 | 1770 | be aligned by 1 megabyte. |
27e0393f | 1771 | |
414fa985 | 1772 | |
e24ed81f | 1773 | 4.66 KVM_S390_UCAS_UNMAP |
27e0393f CO |
1774 | |
1775 | Capability: KVM_CAP_S390_UCONTROL | |
1776 | Architectures: s390 | |
1777 | Type: vcpu ioctl | |
1778 | Parameters: struct kvm_s390_ucas_mapping (in) | |
1779 | Returns: 0 in case of success | |
1780 | ||
1781 | The parameter is defined like this: | |
1782 | struct kvm_s390_ucas_mapping { | |
1783 | __u64 user_addr; | |
1784 | __u64 vcpu_addr; | |
1785 | __u64 length; | |
1786 | }; | |
1787 | ||
1788 | This ioctl unmaps the memory in the vcpu's address space starting at | |
1789 | "vcpu_addr" with the length "length". The field "user_addr" is ignored. | |
f884ab15 | 1790 | All parameters need to be aligned by 1 megabyte. |
27e0393f | 1791 | |
414fa985 | 1792 | |
e24ed81f | 1793 | 4.67 KVM_S390_VCPU_FAULT |
ccc7910f CO |
1794 | |
1795 | Capability: KVM_CAP_S390_UCONTROL | |
1796 | Architectures: s390 | |
1797 | Type: vcpu ioctl | |
1798 | Parameters: vcpu absolute address (in) | |
1799 | Returns: 0 in case of success | |
1800 | ||
1801 | This call creates a page table entry on the virtual cpu's address space | |
1802 | (for user controlled virtual machines) or the virtual machine's address | |
1803 | space (for regular virtual machines). This only works for minor faults, | |
1804 | thus it's recommended to access subject memory page via the user page | |
1805 | table upfront. This is useful to handle validity intercepts for user | |
1806 | controlled virtual machines to fault in the virtual cpu's lowcore pages | |
1807 | prior to calling the KVM_RUN ioctl. | |
1808 | ||
414fa985 | 1809 | |
e24ed81f AG |
1810 | 4.68 KVM_SET_ONE_REG |
1811 | ||
1812 | Capability: KVM_CAP_ONE_REG | |
1813 | Architectures: all | |
1814 | Type: vcpu ioctl | |
1815 | Parameters: struct kvm_one_reg (in) | |
1816 | Returns: 0 on success, negative value on failure | |
1817 | ||
1818 | struct kvm_one_reg { | |
1819 | __u64 id; | |
1820 | __u64 addr; | |
1821 | }; | |
1822 | ||
1823 | Using this ioctl, a single vcpu register can be set to a specific value | |
1824 | defined by user space with the passed in struct kvm_one_reg, where id | |
1825 | refers to the register identifier as described below and addr is a pointer | |
1826 | to a variable with the respective size. There can be architecture agnostic | |
1827 | and architecture specific registers. Each have their own range of operation | |
1828 | and their own constants and width. To keep track of the implemented | |
1829 | registers, find a list below: | |
1830 | ||
bf5590f3 JH |
1831 | Arch | Register | Width (bits) |
1832 | | | | |
1833 | PPC | KVM_REG_PPC_HIOR | 64 | |
1834 | PPC | KVM_REG_PPC_IAC1 | 64 | |
1835 | PPC | KVM_REG_PPC_IAC2 | 64 | |
1836 | PPC | KVM_REG_PPC_IAC3 | 64 | |
1837 | PPC | KVM_REG_PPC_IAC4 | 64 | |
1838 | PPC | KVM_REG_PPC_DAC1 | 64 | |
1839 | PPC | KVM_REG_PPC_DAC2 | 64 | |
1840 | PPC | KVM_REG_PPC_DABR | 64 | |
1841 | PPC | KVM_REG_PPC_DSCR | 64 | |
1842 | PPC | KVM_REG_PPC_PURR | 64 | |
1843 | PPC | KVM_REG_PPC_SPURR | 64 | |
1844 | PPC | KVM_REG_PPC_DAR | 64 | |
1845 | PPC | KVM_REG_PPC_DSISR | 32 | |
1846 | PPC | KVM_REG_PPC_AMR | 64 | |
1847 | PPC | KVM_REG_PPC_UAMOR | 64 | |
1848 | PPC | KVM_REG_PPC_MMCR0 | 64 | |
1849 | PPC | KVM_REG_PPC_MMCR1 | 64 | |
1850 | PPC | KVM_REG_PPC_MMCRA | 64 | |
1851 | PPC | KVM_REG_PPC_MMCR2 | 64 | |
1852 | PPC | KVM_REG_PPC_MMCRS | 64 | |
1853 | PPC | KVM_REG_PPC_SIAR | 64 | |
1854 | PPC | KVM_REG_PPC_SDAR | 64 | |
1855 | PPC | KVM_REG_PPC_SIER | 64 | |
1856 | PPC | KVM_REG_PPC_PMC1 | 32 | |
1857 | PPC | KVM_REG_PPC_PMC2 | 32 | |
1858 | PPC | KVM_REG_PPC_PMC3 | 32 | |
1859 | PPC | KVM_REG_PPC_PMC4 | 32 | |
1860 | PPC | KVM_REG_PPC_PMC5 | 32 | |
1861 | PPC | KVM_REG_PPC_PMC6 | 32 | |
1862 | PPC | KVM_REG_PPC_PMC7 | 32 | |
1863 | PPC | KVM_REG_PPC_PMC8 | 32 | |
1864 | PPC | KVM_REG_PPC_FPR0 | 64 | |
a8bd19ef | 1865 | ... |
bf5590f3 JH |
1866 | PPC | KVM_REG_PPC_FPR31 | 64 |
1867 | PPC | KVM_REG_PPC_VR0 | 128 | |
a8bd19ef | 1868 | ... |
bf5590f3 JH |
1869 | PPC | KVM_REG_PPC_VR31 | 128 |
1870 | PPC | KVM_REG_PPC_VSR0 | 128 | |
a8bd19ef | 1871 | ... |
bf5590f3 JH |
1872 | PPC | KVM_REG_PPC_VSR31 | 128 |
1873 | PPC | KVM_REG_PPC_FPSCR | 64 | |
1874 | PPC | KVM_REG_PPC_VSCR | 32 | |
1875 | PPC | KVM_REG_PPC_VPA_ADDR | 64 | |
1876 | PPC | KVM_REG_PPC_VPA_SLB | 128 | |
1877 | PPC | KVM_REG_PPC_VPA_DTL | 128 | |
1878 | PPC | KVM_REG_PPC_EPCR | 32 | |
1879 | PPC | KVM_REG_PPC_EPR | 32 | |
1880 | PPC | KVM_REG_PPC_TCR | 32 | |
1881 | PPC | KVM_REG_PPC_TSR | 32 | |
1882 | PPC | KVM_REG_PPC_OR_TSR | 32 | |
1883 | PPC | KVM_REG_PPC_CLEAR_TSR | 32 | |
1884 | PPC | KVM_REG_PPC_MAS0 | 32 | |
1885 | PPC | KVM_REG_PPC_MAS1 | 32 | |
1886 | PPC | KVM_REG_PPC_MAS2 | 64 | |
1887 | PPC | KVM_REG_PPC_MAS7_3 | 64 | |
1888 | PPC | KVM_REG_PPC_MAS4 | 32 | |
1889 | PPC | KVM_REG_PPC_MAS6 | 32 | |
1890 | PPC | KVM_REG_PPC_MMUCFG | 32 | |
1891 | PPC | KVM_REG_PPC_TLB0CFG | 32 | |
1892 | PPC | KVM_REG_PPC_TLB1CFG | 32 | |
1893 | PPC | KVM_REG_PPC_TLB2CFG | 32 | |
1894 | PPC | KVM_REG_PPC_TLB3CFG | 32 | |
1895 | PPC | KVM_REG_PPC_TLB0PS | 32 | |
1896 | PPC | KVM_REG_PPC_TLB1PS | 32 | |
1897 | PPC | KVM_REG_PPC_TLB2PS | 32 | |
1898 | PPC | KVM_REG_PPC_TLB3PS | 32 | |
1899 | PPC | KVM_REG_PPC_EPTCFG | 32 | |
1900 | PPC | KVM_REG_PPC_ICP_STATE | 64 | |
1901 | PPC | KVM_REG_PPC_TB_OFFSET | 64 | |
1902 | PPC | KVM_REG_PPC_SPMC1 | 32 | |
1903 | PPC | KVM_REG_PPC_SPMC2 | 32 | |
1904 | PPC | KVM_REG_PPC_IAMR | 64 | |
1905 | PPC | KVM_REG_PPC_TFHAR | 64 | |
1906 | PPC | KVM_REG_PPC_TFIAR | 64 | |
1907 | PPC | KVM_REG_PPC_TEXASR | 64 | |
1908 | PPC | KVM_REG_PPC_FSCR | 64 | |
1909 | PPC | KVM_REG_PPC_PSPB | 32 | |
1910 | PPC | KVM_REG_PPC_EBBHR | 64 | |
1911 | PPC | KVM_REG_PPC_EBBRR | 64 | |
1912 | PPC | KVM_REG_PPC_BESCR | 64 | |
1913 | PPC | KVM_REG_PPC_TAR | 64 | |
1914 | PPC | KVM_REG_PPC_DPDES | 64 | |
1915 | PPC | KVM_REG_PPC_DAWR | 64 | |
1916 | PPC | KVM_REG_PPC_DAWRX | 64 | |
1917 | PPC | KVM_REG_PPC_CIABR | 64 | |
1918 | PPC | KVM_REG_PPC_IC | 64 | |
1919 | PPC | KVM_REG_PPC_VTB | 64 | |
1920 | PPC | KVM_REG_PPC_CSIGR | 64 | |
1921 | PPC | KVM_REG_PPC_TACR | 64 | |
1922 | PPC | KVM_REG_PPC_TCSCR | 64 | |
1923 | PPC | KVM_REG_PPC_PID | 64 | |
1924 | PPC | KVM_REG_PPC_ACOP | 64 | |
1925 | PPC | KVM_REG_PPC_VRSAVE | 32 | |
cc568ead PB |
1926 | PPC | KVM_REG_PPC_LPCR | 32 |
1927 | PPC | KVM_REG_PPC_LPCR_64 | 64 | |
bf5590f3 JH |
1928 | PPC | KVM_REG_PPC_PPR | 64 |
1929 | PPC | KVM_REG_PPC_ARCH_COMPAT | 32 | |
1930 | PPC | KVM_REG_PPC_DABRX | 32 | |
1931 | PPC | KVM_REG_PPC_WORT | 64 | |
bc8a4e5c BB |
1932 | PPC | KVM_REG_PPC_SPRG9 | 64 |
1933 | PPC | KVM_REG_PPC_DBSR | 32 | |
bf5590f3 | 1934 | PPC | KVM_REG_PPC_TM_GPR0 | 64 |
3b783474 | 1935 | ... |
bf5590f3 JH |
1936 | PPC | KVM_REG_PPC_TM_GPR31 | 64 |
1937 | PPC | KVM_REG_PPC_TM_VSR0 | 128 | |
3b783474 | 1938 | ... |
bf5590f3 JH |
1939 | PPC | KVM_REG_PPC_TM_VSR63 | 128 |
1940 | PPC | KVM_REG_PPC_TM_CR | 64 | |
1941 | PPC | KVM_REG_PPC_TM_LR | 64 | |
1942 | PPC | KVM_REG_PPC_TM_CTR | 64 | |
1943 | PPC | KVM_REG_PPC_TM_FPSCR | 64 | |
1944 | PPC | KVM_REG_PPC_TM_AMR | 64 | |
1945 | PPC | KVM_REG_PPC_TM_PPR | 64 | |
1946 | PPC | KVM_REG_PPC_TM_VRSAVE | 64 | |
1947 | PPC | KVM_REG_PPC_TM_VSCR | 32 | |
1948 | PPC | KVM_REG_PPC_TM_DSCR | 64 | |
1949 | PPC | KVM_REG_PPC_TM_TAR | 64 | |
c2d2c21b JH |
1950 | | | |
1951 | MIPS | KVM_REG_MIPS_R0 | 64 | |
1952 | ... | |
1953 | MIPS | KVM_REG_MIPS_R31 | 64 | |
1954 | MIPS | KVM_REG_MIPS_HI | 64 | |
1955 | MIPS | KVM_REG_MIPS_LO | 64 | |
1956 | MIPS | KVM_REG_MIPS_PC | 64 | |
1957 | MIPS | KVM_REG_MIPS_CP0_INDEX | 32 | |
1958 | MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64 | |
1959 | MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64 | |
1960 | MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32 | |
1961 | MIPS | KVM_REG_MIPS_CP0_WIRED | 32 | |
1962 | MIPS | KVM_REG_MIPS_CP0_HWRENA | 32 | |
1963 | MIPS | KVM_REG_MIPS_CP0_BADVADDR | 64 | |
1964 | MIPS | KVM_REG_MIPS_CP0_COUNT | 32 | |
1965 | MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64 | |
1966 | MIPS | KVM_REG_MIPS_CP0_COMPARE | 32 | |
1967 | MIPS | KVM_REG_MIPS_CP0_STATUS | 32 | |
1968 | MIPS | KVM_REG_MIPS_CP0_CAUSE | 32 | |
1969 | MIPS | KVM_REG_MIPS_CP0_EPC | 64 | |
1970 | MIPS | KVM_REG_MIPS_CP0_CONFIG | 32 | |
1971 | MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32 | |
1972 | MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32 | |
1973 | MIPS | KVM_REG_MIPS_CP0_CONFIG3 | 32 | |
1974 | MIPS | KVM_REG_MIPS_CP0_CONFIG7 | 32 | |
1975 | MIPS | KVM_REG_MIPS_CP0_ERROREPC | 64 | |
1976 | MIPS | KVM_REG_MIPS_COUNT_CTL | 64 | |
1977 | MIPS | KVM_REG_MIPS_COUNT_RESUME | 64 | |
1978 | MIPS | KVM_REG_MIPS_COUNT_HZ | 64 | |
414fa985 | 1979 | |
749cf76c CD |
1980 | ARM registers are mapped using the lower 32 bits. The upper 16 of that |
1981 | is the register group type, or coprocessor number: | |
1982 | ||
1983 | ARM core registers have the following id bit patterns: | |
aa404ddf | 1984 | 0x4020 0000 0010 <index into the kvm_regs struct:16> |
749cf76c | 1985 | |
1138245c | 1986 | ARM 32-bit CP15 registers have the following id bit patterns: |
aa404ddf | 1987 | 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3> |
1138245c CD |
1988 | |
1989 | ARM 64-bit CP15 registers have the following id bit patterns: | |
aa404ddf | 1990 | 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3> |
749cf76c | 1991 | |
c27581ed | 1992 | ARM CCSIDR registers are demultiplexed by CSSELR value: |
aa404ddf | 1993 | 0x4020 0000 0011 00 <csselr:8> |
749cf76c | 1994 | |
4fe21e4c | 1995 | ARM 32-bit VFP control registers have the following id bit patterns: |
aa404ddf | 1996 | 0x4020 0000 0012 1 <regno:12> |
4fe21e4c RR |
1997 | |
1998 | ARM 64-bit FP registers have the following id bit patterns: | |
aa404ddf | 1999 | 0x4030 0000 0012 0 <regno:12> |
4fe21e4c | 2000 | |
379e04c7 MZ |
2001 | |
2002 | arm64 registers are mapped using the lower 32 bits. The upper 16 of | |
2003 | that is the register group type, or coprocessor number: | |
2004 | ||
2005 | arm64 core/FP-SIMD registers have the following id bit patterns. Note | |
2006 | that the size of the access is variable, as the kvm_regs structure | |
2007 | contains elements ranging from 32 to 128 bits. The index is a 32bit | |
2008 | value in the kvm_regs structure seen as a 32bit array. | |
2009 | 0x60x0 0000 0010 <index into the kvm_regs struct:16> | |
2010 | ||
2011 | arm64 CCSIDR registers are demultiplexed by CSSELR value: | |
2012 | 0x6020 0000 0011 00 <csselr:8> | |
2013 | ||
2014 | arm64 system registers have the following id bit patterns: | |
2015 | 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3> | |
2016 | ||
c2d2c21b JH |
2017 | |
2018 | MIPS registers are mapped using the lower 32 bits. The upper 16 of that is | |
2019 | the register group type: | |
2020 | ||
2021 | MIPS core registers (see above) have the following id bit patterns: | |
2022 | 0x7030 0000 0000 <reg:16> | |
2023 | ||
2024 | MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit | |
2025 | patterns depending on whether they're 32-bit or 64-bit registers: | |
2026 | 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) | |
2027 | 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) | |
2028 | ||
2029 | MIPS KVM control registers (see above) have the following id bit patterns: | |
2030 | 0x7030 0000 0002 <reg:16> | |
2031 | ||
2032 | ||
e24ed81f AG |
2033 | 4.69 KVM_GET_ONE_REG |
2034 | ||
2035 | Capability: KVM_CAP_ONE_REG | |
2036 | Architectures: all | |
2037 | Type: vcpu ioctl | |
2038 | Parameters: struct kvm_one_reg (in and out) | |
2039 | Returns: 0 on success, negative value on failure | |
2040 | ||
2041 | This ioctl allows to receive the value of a single register implemented | |
2042 | in a vcpu. The register to read is indicated by the "id" field of the | |
2043 | kvm_one_reg struct passed in. On success, the register value can be found | |
2044 | at the memory location pointed to by "addr". | |
2045 | ||
2046 | The list of registers accessible using this interface is identical to the | |
2e232702 | 2047 | list in 4.68. |
e24ed81f | 2048 | |
414fa985 | 2049 | |
1c0b28c2 EM |
2050 | 4.70 KVM_KVMCLOCK_CTRL |
2051 | ||
2052 | Capability: KVM_CAP_KVMCLOCK_CTRL | |
2053 | Architectures: Any that implement pvclocks (currently x86 only) | |
2054 | Type: vcpu ioctl | |
2055 | Parameters: None | |
2056 | Returns: 0 on success, -1 on error | |
2057 | ||
2058 | This signals to the host kernel that the specified guest is being paused by | |
2059 | userspace. The host will set a flag in the pvclock structure that is checked | |
2060 | from the soft lockup watchdog. The flag is part of the pvclock structure that | |
2061 | is shared between guest and host, specifically the second bit of the flags | |
2062 | field of the pvclock_vcpu_time_info structure. It will be set exclusively by | |
2063 | the host and read/cleared exclusively by the guest. The guest operation of | |
2064 | checking and clearing the flag must an atomic operation so | |
2065 | load-link/store-conditional, or equivalent must be used. There are two cases | |
2066 | where the guest will clear the flag: when the soft lockup watchdog timer resets | |
2067 | itself or when a soft lockup is detected. This ioctl can be called any time | |
2068 | after pausing the vcpu, but before it is resumed. | |
2069 | ||
414fa985 | 2070 | |
07975ad3 JK |
2071 | 4.71 KVM_SIGNAL_MSI |
2072 | ||
2073 | Capability: KVM_CAP_SIGNAL_MSI | |
2074 | Architectures: x86 | |
2075 | Type: vm ioctl | |
2076 | Parameters: struct kvm_msi (in) | |
2077 | Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error | |
2078 | ||
2079 | Directly inject a MSI message. Only valid with in-kernel irqchip that handles | |
2080 | MSI messages. | |
2081 | ||
2082 | struct kvm_msi { | |
2083 | __u32 address_lo; | |
2084 | __u32 address_hi; | |
2085 | __u32 data; | |
2086 | __u32 flags; | |
2087 | __u8 pad[16]; | |
2088 | }; | |
2089 | ||
2090 | No flags are defined so far. The corresponding field must be 0. | |
2091 | ||
414fa985 | 2092 | |
0589ff6c JK |
2093 | 4.71 KVM_CREATE_PIT2 |
2094 | ||
2095 | Capability: KVM_CAP_PIT2 | |
2096 | Architectures: x86 | |
2097 | Type: vm ioctl | |
2098 | Parameters: struct kvm_pit_config (in) | |
2099 | Returns: 0 on success, -1 on error | |
2100 | ||
2101 | Creates an in-kernel device model for the i8254 PIT. This call is only valid | |
2102 | after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following | |
2103 | parameters have to be passed: | |
2104 | ||
2105 | struct kvm_pit_config { | |
2106 | __u32 flags; | |
2107 | __u32 pad[15]; | |
2108 | }; | |
2109 | ||
2110 | Valid flags are: | |
2111 | ||
2112 | #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */ | |
2113 | ||
b6ddf05f JK |
2114 | PIT timer interrupts may use a per-VM kernel thread for injection. If it |
2115 | exists, this thread will have a name of the following pattern: | |
2116 | ||
2117 | kvm-pit/<owner-process-pid> | |
2118 | ||
2119 | When running a guest with elevated priorities, the scheduling parameters of | |
2120 | this thread may have to be adjusted accordingly. | |
2121 | ||
0589ff6c JK |
2122 | This IOCTL replaces the obsolete KVM_CREATE_PIT. |
2123 | ||
2124 | ||
2125 | 4.72 KVM_GET_PIT2 | |
2126 | ||
2127 | Capability: KVM_CAP_PIT_STATE2 | |
2128 | Architectures: x86 | |
2129 | Type: vm ioctl | |
2130 | Parameters: struct kvm_pit_state2 (out) | |
2131 | Returns: 0 on success, -1 on error | |
2132 | ||
2133 | Retrieves the state of the in-kernel PIT model. Only valid after | |
2134 | KVM_CREATE_PIT2. The state is returned in the following structure: | |
2135 | ||
2136 | struct kvm_pit_state2 { | |
2137 | struct kvm_pit_channel_state channels[3]; | |
2138 | __u32 flags; | |
2139 | __u32 reserved[9]; | |
2140 | }; | |
2141 | ||
2142 | Valid flags are: | |
2143 | ||
2144 | /* disable PIT in HPET legacy mode */ | |
2145 | #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001 | |
2146 | ||
2147 | This IOCTL replaces the obsolete KVM_GET_PIT. | |
2148 | ||
2149 | ||
2150 | 4.73 KVM_SET_PIT2 | |
2151 | ||
2152 | Capability: KVM_CAP_PIT_STATE2 | |
2153 | Architectures: x86 | |
2154 | Type: vm ioctl | |
2155 | Parameters: struct kvm_pit_state2 (in) | |
2156 | Returns: 0 on success, -1 on error | |
2157 | ||
2158 | Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2. | |
2159 | See KVM_GET_PIT2 for details on struct kvm_pit_state2. | |
2160 | ||
2161 | This IOCTL replaces the obsolete KVM_SET_PIT. | |
2162 | ||
2163 | ||
5b74716e BH |
2164 | 4.74 KVM_PPC_GET_SMMU_INFO |
2165 | ||
2166 | Capability: KVM_CAP_PPC_GET_SMMU_INFO | |
2167 | Architectures: powerpc | |
2168 | Type: vm ioctl | |
2169 | Parameters: None | |
2170 | Returns: 0 on success, -1 on error | |
2171 | ||
2172 | This populates and returns a structure describing the features of | |
2173 | the "Server" class MMU emulation supported by KVM. | |
cc22c354 | 2174 | This can in turn be used by userspace to generate the appropriate |
5b74716e BH |
2175 | device-tree properties for the guest operating system. |
2176 | ||
c98be0c9 | 2177 | The structure contains some global information, followed by an |
5b74716e BH |
2178 | array of supported segment page sizes: |
2179 | ||
2180 | struct kvm_ppc_smmu_info { | |
2181 | __u64 flags; | |
2182 | __u32 slb_size; | |
2183 | __u32 pad; | |
2184 | struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; | |
2185 | }; | |
2186 | ||
2187 | The supported flags are: | |
2188 | ||
2189 | - KVM_PPC_PAGE_SIZES_REAL: | |
2190 | When that flag is set, guest page sizes must "fit" the backing | |
2191 | store page sizes. When not set, any page size in the list can | |
2192 | be used regardless of how they are backed by userspace. | |
2193 | ||
2194 | - KVM_PPC_1T_SEGMENTS | |
2195 | The emulated MMU supports 1T segments in addition to the | |
2196 | standard 256M ones. | |
2197 | ||
2198 | The "slb_size" field indicates how many SLB entries are supported | |
2199 | ||
2200 | The "sps" array contains 8 entries indicating the supported base | |
2201 | page sizes for a segment in increasing order. Each entry is defined | |
2202 | as follow: | |
2203 | ||
2204 | struct kvm_ppc_one_seg_page_size { | |
2205 | __u32 page_shift; /* Base page shift of segment (or 0) */ | |
2206 | __u32 slb_enc; /* SLB encoding for BookS */ | |
2207 | struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ]; | |
2208 | }; | |
2209 | ||
2210 | An entry with a "page_shift" of 0 is unused. Because the array is | |
2211 | organized in increasing order, a lookup can stop when encoutering | |
2212 | such an entry. | |
2213 | ||
2214 | The "slb_enc" field provides the encoding to use in the SLB for the | |
2215 | page size. The bits are in positions such as the value can directly | |
2216 | be OR'ed into the "vsid" argument of the slbmte instruction. | |
2217 | ||
2218 | The "enc" array is a list which for each of those segment base page | |
2219 | size provides the list of supported actual page sizes (which can be | |
2220 | only larger or equal to the base page size), along with the | |
f884ab15 | 2221 | corresponding encoding in the hash PTE. Similarly, the array is |
5b74716e BH |
2222 | 8 entries sorted by increasing sizes and an entry with a "0" shift |
2223 | is an empty entry and a terminator: | |
2224 | ||
2225 | struct kvm_ppc_one_page_size { | |
2226 | __u32 page_shift; /* Page shift (or 0) */ | |
2227 | __u32 pte_enc; /* Encoding in the HPTE (>>12) */ | |
2228 | }; | |
2229 | ||
2230 | The "pte_enc" field provides a value that can OR'ed into the hash | |
2231 | PTE's RPN field (ie, it needs to be shifted left by 12 to OR it | |
2232 | into the hash PTE second double word). | |
2233 | ||
f36992e3 AW |
2234 | 4.75 KVM_IRQFD |
2235 | ||
2236 | Capability: KVM_CAP_IRQFD | |
174178fe | 2237 | Architectures: x86 s390 arm arm64 |
f36992e3 AW |
2238 | Type: vm ioctl |
2239 | Parameters: struct kvm_irqfd (in) | |
2240 | Returns: 0 on success, -1 on error | |
2241 | ||
2242 | Allows setting an eventfd to directly trigger a guest interrupt. | |
2243 | kvm_irqfd.fd specifies the file descriptor to use as the eventfd and | |
2244 | kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When | |
17180032 | 2245 | an event is triggered on the eventfd, an interrupt is injected into |
f36992e3 AW |
2246 | the guest using the specified gsi pin. The irqfd is removed using |
2247 | the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd | |
2248 | and kvm_irqfd.gsi. | |
2249 | ||
7a84428a AW |
2250 | With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify |
2251 | mechanism allowing emulation of level-triggered, irqfd-based | |
2252 | interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an | |
2253 | additional eventfd in the kvm_irqfd.resamplefd field. When operating | |
2254 | in resample mode, posting of an interrupt through kvm_irq.fd asserts | |
2255 | the specified gsi in the irqchip. When the irqchip is resampled, such | |
17180032 | 2256 | as from an EOI, the gsi is de-asserted and the user is notified via |
7a84428a AW |
2257 | kvm_irqfd.resamplefd. It is the user's responsibility to re-queue |
2258 | the interrupt if the device making use of it still requires service. | |
2259 | Note that closing the resamplefd is not sufficient to disable the | |
2260 | irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment | |
2261 | and need not be specified with KVM_IRQFD_FLAG_DEASSIGN. | |
2262 | ||
174178fe EA |
2263 | On ARM/ARM64, the gsi field in the kvm_irqfd struct specifies the Shared |
2264 | Peripheral Interrupt (SPI) index, such that the GIC interrupt ID is | |
2265 | given by gsi + 32. | |
2266 | ||
5fecc9d8 | 2267 | 4.76 KVM_PPC_ALLOCATE_HTAB |
32fad281 PM |
2268 | |
2269 | Capability: KVM_CAP_PPC_ALLOC_HTAB | |
2270 | Architectures: powerpc | |
2271 | Type: vm ioctl | |
2272 | Parameters: Pointer to u32 containing hash table order (in/out) | |
2273 | Returns: 0 on success, -1 on error | |
2274 | ||
2275 | This requests the host kernel to allocate an MMU hash table for a | |
2276 | guest using the PAPR paravirtualization interface. This only does | |
2277 | anything if the kernel is configured to use the Book 3S HV style of | |
2278 | virtualization. Otherwise the capability doesn't exist and the ioctl | |
2279 | returns an ENOTTY error. The rest of this description assumes Book 3S | |
2280 | HV. | |
2281 | ||
2282 | There must be no vcpus running when this ioctl is called; if there | |
2283 | are, it will do nothing and return an EBUSY error. | |
2284 | ||
2285 | The parameter is a pointer to a 32-bit unsigned integer variable | |
2286 | containing the order (log base 2) of the desired size of the hash | |
2287 | table, which must be between 18 and 46. On successful return from the | |
2288 | ioctl, it will have been updated with the order of the hash table that | |
2289 | was allocated. | |
2290 | ||
2291 | If no hash table has been allocated when any vcpu is asked to run | |
2292 | (with the KVM_RUN ioctl), the host kernel will allocate a | |
2293 | default-sized hash table (16 MB). | |
2294 | ||
2295 | If this ioctl is called when a hash table has already been allocated, | |
2296 | the kernel will clear out the existing hash table (zero all HPTEs) and | |
2297 | return the hash table order in the parameter. (If the guest is using | |
2298 | the virtualized real-mode area (VRMA) facility, the kernel will | |
2299 | re-create the VMRA HPTEs on the next KVM_RUN of any vcpu.) | |
2300 | ||
416ad65f CH |
2301 | 4.77 KVM_S390_INTERRUPT |
2302 | ||
2303 | Capability: basic | |
2304 | Architectures: s390 | |
2305 | Type: vm ioctl, vcpu ioctl | |
2306 | Parameters: struct kvm_s390_interrupt (in) | |
2307 | Returns: 0 on success, -1 on error | |
2308 | ||
2309 | Allows to inject an interrupt to the guest. Interrupts can be floating | |
2310 | (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type. | |
2311 | ||
2312 | Interrupt parameters are passed via kvm_s390_interrupt: | |
2313 | ||
2314 | struct kvm_s390_interrupt { | |
2315 | __u32 type; | |
2316 | __u32 parm; | |
2317 | __u64 parm64; | |
2318 | }; | |
2319 | ||
2320 | type can be one of the following: | |
2321 | ||
2822545f | 2322 | KVM_S390_SIGP_STOP (vcpu) - sigp stop; optional flags in parm |
416ad65f CH |
2323 | KVM_S390_PROGRAM_INT (vcpu) - program check; code in parm |
2324 | KVM_S390_SIGP_SET_PREFIX (vcpu) - sigp set prefix; prefix address in parm | |
2325 | KVM_S390_RESTART (vcpu) - restart | |
e029ae5b TH |
2326 | KVM_S390_INT_CLOCK_COMP (vcpu) - clock comparator interrupt |
2327 | KVM_S390_INT_CPU_TIMER (vcpu) - CPU timer interrupt | |
416ad65f CH |
2328 | KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt |
2329 | parameters in parm and parm64 | |
2330 | KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm | |
2331 | KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm | |
2332 | KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm | |
d8346b7d CH |
2333 | KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an |
2334 | I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel); | |
2335 | I/O interruption parameters in parm (subchannel) and parm64 (intparm, | |
2336 | interruption subclass) | |
48a3e950 CH |
2337 | KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm, |
2338 | machine check interrupt code in parm64 (note that | |
2339 | machine checks needing further payload are not | |
2340 | supported by this ioctl) | |
416ad65f CH |
2341 | |
2342 | Note that the vcpu ioctl is asynchronous to vcpu execution. | |
2343 | ||
a2932923 PM |
2344 | 4.78 KVM_PPC_GET_HTAB_FD |
2345 | ||
2346 | Capability: KVM_CAP_PPC_HTAB_FD | |
2347 | Architectures: powerpc | |
2348 | Type: vm ioctl | |
2349 | Parameters: Pointer to struct kvm_get_htab_fd (in) | |
2350 | Returns: file descriptor number (>= 0) on success, -1 on error | |
2351 | ||
2352 | This returns a file descriptor that can be used either to read out the | |
2353 | entries in the guest's hashed page table (HPT), or to write entries to | |
2354 | initialize the HPT. The returned fd can only be written to if the | |
2355 | KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and | |
2356 | can only be read if that bit is clear. The argument struct looks like | |
2357 | this: | |
2358 | ||
2359 | /* For KVM_PPC_GET_HTAB_FD */ | |
2360 | struct kvm_get_htab_fd { | |
2361 | __u64 flags; | |
2362 | __u64 start_index; | |
2363 | __u64 reserved[2]; | |
2364 | }; | |
2365 | ||
2366 | /* Values for kvm_get_htab_fd.flags */ | |
2367 | #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1) | |
2368 | #define KVM_GET_HTAB_WRITE ((__u64)0x2) | |
2369 | ||
2370 | The `start_index' field gives the index in the HPT of the entry at | |
2371 | which to start reading. It is ignored when writing. | |
2372 | ||
2373 | Reads on the fd will initially supply information about all | |
2374 | "interesting" HPT entries. Interesting entries are those with the | |
2375 | bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise | |
2376 | all entries. When the end of the HPT is reached, the read() will | |
2377 | return. If read() is called again on the fd, it will start again from | |
2378 | the beginning of the HPT, but will only return HPT entries that have | |
2379 | changed since they were last read. | |
2380 | ||
2381 | Data read or written is structured as a header (8 bytes) followed by a | |
2382 | series of valid HPT entries (16 bytes) each. The header indicates how | |
2383 | many valid HPT entries there are and how many invalid entries follow | |
2384 | the valid entries. The invalid entries are not represented explicitly | |
2385 | in the stream. The header format is: | |
2386 | ||
2387 | struct kvm_get_htab_header { | |
2388 | __u32 index; | |
2389 | __u16 n_valid; | |
2390 | __u16 n_invalid; | |
2391 | }; | |
2392 | ||
2393 | Writes to the fd create HPT entries starting at the index given in the | |
2394 | header; first `n_valid' valid entries with contents from the data | |
2395 | written, then `n_invalid' invalid entries, invalidating any previously | |
2396 | valid entries found. | |
2397 | ||
852b6d57 SW |
2398 | 4.79 KVM_CREATE_DEVICE |
2399 | ||
2400 | Capability: KVM_CAP_DEVICE_CTRL | |
2401 | Type: vm ioctl | |
2402 | Parameters: struct kvm_create_device (in/out) | |
2403 | Returns: 0 on success, -1 on error | |
2404 | Errors: | |
2405 | ENODEV: The device type is unknown or unsupported | |
2406 | EEXIST: Device already created, and this type of device may not | |
2407 | be instantiated multiple times | |
2408 | ||
2409 | Other error conditions may be defined by individual device types or | |
2410 | have their standard meanings. | |
2411 | ||
2412 | Creates an emulated device in the kernel. The file descriptor returned | |
2413 | in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR. | |
2414 | ||
2415 | If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the | |
2416 | device type is supported (not necessarily whether it can be created | |
2417 | in the current vm). | |
2418 | ||
2419 | Individual devices should not define flags. Attributes should be used | |
2420 | for specifying any behavior that is not implied by the device type | |
2421 | number. | |
2422 | ||
2423 | struct kvm_create_device { | |
2424 | __u32 type; /* in: KVM_DEV_TYPE_xxx */ | |
2425 | __u32 fd; /* out: device handle */ | |
2426 | __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */ | |
2427 | }; | |
2428 | ||
2429 | 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR | |
2430 | ||
f2061656 DD |
2431 | Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device |
2432 | Type: device ioctl, vm ioctl | |
852b6d57 SW |
2433 | Parameters: struct kvm_device_attr |
2434 | Returns: 0 on success, -1 on error | |
2435 | Errors: | |
2436 | ENXIO: The group or attribute is unknown/unsupported for this device | |
2437 | EPERM: The attribute cannot (currently) be accessed this way | |
2438 | (e.g. read-only attribute, or attribute that only makes | |
2439 | sense when the device is in a different state) | |
2440 | ||
2441 | Other error conditions may be defined by individual device types. | |
2442 | ||
2443 | Gets/sets a specified piece of device configuration and/or state. The | |
2444 | semantics are device-specific. See individual device documentation in | |
2445 | the "devices" directory. As with ONE_REG, the size of the data | |
2446 | transferred is defined by the particular attribute. | |
2447 | ||
2448 | struct kvm_device_attr { | |
2449 | __u32 flags; /* no flags currently defined */ | |
2450 | __u32 group; /* device-defined */ | |
2451 | __u64 attr; /* group-defined */ | |
2452 | __u64 addr; /* userspace address of attr data */ | |
2453 | }; | |
2454 | ||
2455 | 4.81 KVM_HAS_DEVICE_ATTR | |
2456 | ||
f2061656 DD |
2457 | Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device |
2458 | Type: device ioctl, vm ioctl | |
852b6d57 SW |
2459 | Parameters: struct kvm_device_attr |
2460 | Returns: 0 on success, -1 on error | |
2461 | Errors: | |
2462 | ENXIO: The group or attribute is unknown/unsupported for this device | |
2463 | ||
2464 | Tests whether a device supports a particular attribute. A successful | |
2465 | return indicates the attribute is implemented. It does not necessarily | |
2466 | indicate that the attribute can be read or written in the device's | |
2467 | current state. "addr" is ignored. | |
f36992e3 | 2468 | |
d8968f1f | 2469 | 4.82 KVM_ARM_VCPU_INIT |
749cf76c CD |
2470 | |
2471 | Capability: basic | |
379e04c7 | 2472 | Architectures: arm, arm64 |
749cf76c | 2473 | Type: vcpu ioctl |
beb11fc7 | 2474 | Parameters: struct kvm_vcpu_init (in) |
749cf76c CD |
2475 | Returns: 0 on success; -1 on error |
2476 | Errors: | |
2477 | Â EINVAL: Â Â Â the target is unknown, or the combination of features is invalid. | |
2478 | Â ENOENT: Â Â Â a features bit specified is unknown. | |
2479 | ||
2480 | This tells KVM what type of CPU to present to the guest, and what | |
2481 | optional features it should have. Â This will cause a reset of the cpu | |
2482 | registers to their initial values. Â If this is not called, KVM_RUN will | |
2483 | return ENOEXEC for that vcpu. | |
2484 | ||
2485 | Note that because some registers reflect machine topology, all vcpus | |
2486 | should be created before this ioctl is invoked. | |
2487 | ||
f7fa034d CD |
2488 | Userspace can call this function multiple times for a given vcpu, including |
2489 | after the vcpu has been run. This will reset the vcpu to its initial | |
2490 | state. All calls to this function after the initial call must use the same | |
2491 | target and same set of feature flags, otherwise EINVAL will be returned. | |
2492 | ||
aa024c2f MZ |
2493 | Possible features: |
2494 | - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state. | |
3ad8b3de CD |
2495 | Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on |
2496 | and execute guest code when KVM_RUN is called. | |
379e04c7 MZ |
2497 | - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode. |
2498 | Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only). | |
50bb0c94 AP |
2499 | - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 for the CPU. |
2500 | Depends on KVM_CAP_ARM_PSCI_0_2. | |
aa024c2f | 2501 | |
749cf76c | 2502 | |
740edfc0 AP |
2503 | 4.83 KVM_ARM_PREFERRED_TARGET |
2504 | ||
2505 | Capability: basic | |
2506 | Architectures: arm, arm64 | |
2507 | Type: vm ioctl | |
2508 | Parameters: struct struct kvm_vcpu_init (out) | |
2509 | Returns: 0 on success; -1 on error | |
2510 | Errors: | |
a7265fb1 | 2511 | ENODEV: no preferred target available for the host |
740edfc0 AP |
2512 | |
2513 | This queries KVM for preferred CPU target type which can be emulated | |
2514 | by KVM on underlying host. | |
2515 | ||
2516 | The ioctl returns struct kvm_vcpu_init instance containing information | |
2517 | about preferred CPU target type and recommended features for it. The | |
2518 | kvm_vcpu_init->features bitmap returned will have feature bits set if | |
2519 | the preferred target recommends setting these features, but this is | |
2520 | not mandatory. | |
2521 | ||
2522 | The information returned by this ioctl can be used to prepare an instance | |
2523 | of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in | |
2524 | in VCPU matching underlying host. | |
2525 | ||
2526 | ||
2527 | 4.84 KVM_GET_REG_LIST | |
749cf76c CD |
2528 | |
2529 | Capability: basic | |
c2d2c21b | 2530 | Architectures: arm, arm64, mips |
749cf76c CD |
2531 | Type: vcpu ioctl |
2532 | Parameters: struct kvm_reg_list (in/out) | |
2533 | Returns: 0 on success; -1 on error | |
2534 | Errors: | |
2535 | Â E2BIG: Â Â Â Â the reg index list is too big to fit in the array specified by | |
2536 | Â Â Â Â Â Â Â Â Â Â Â Â the user (the number required will be written into n). | |
2537 | ||
2538 | struct kvm_reg_list { | |
2539 | __u64 n; /* number of registers in reg[] */ | |
2540 | __u64 reg[0]; | |
2541 | }; | |
2542 | ||
2543 | This ioctl returns the guest registers that are supported for the | |
2544 | KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. | |
2545 | ||
ce01e4e8 CD |
2546 | |
2547 | 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) | |
3401d546 CD |
2548 | |
2549 | Capability: KVM_CAP_ARM_SET_DEVICE_ADDR | |
379e04c7 | 2550 | Architectures: arm, arm64 |
3401d546 CD |
2551 | Type: vm ioctl |
2552 | Parameters: struct kvm_arm_device_address (in) | |
2553 | Returns: 0 on success, -1 on error | |
2554 | Errors: | |
2555 | ENODEV: The device id is unknown | |
2556 | ENXIO: Device not supported on current system | |
2557 | EEXIST: Address already set | |
2558 | E2BIG: Address outside guest physical address space | |
330690cd | 2559 | EBUSY: Address overlaps with other device range |
3401d546 CD |
2560 | |
2561 | struct kvm_arm_device_addr { | |
2562 | __u64 id; | |
2563 | __u64 addr; | |
2564 | }; | |
2565 | ||
2566 | Specify a device address in the guest's physical address space where guests | |
2567 | can access emulated or directly exposed devices, which the host kernel needs | |
2568 | to know about. The id field is an architecture specific identifier for a | |
2569 | specific device. | |
2570 | ||
379e04c7 MZ |
2571 | ARM/arm64 divides the id field into two parts, a device id and an |
2572 | address type id specific to the individual device. | |
3401d546 CD |
2573 | |
2574 | Â bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 | | |
2575 | field: | 0x00000000 | device id | addr type id | | |
2576 | ||
379e04c7 MZ |
2577 | ARM/arm64 currently only require this when using the in-kernel GIC |
2578 | support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2 | |
2579 | as the device id. When setting the base address for the guest's | |
2580 | mapping of the VGIC virtual CPU and distributor interface, the ioctl | |
2581 | must be called after calling KVM_CREATE_IRQCHIP, but before calling | |
2582 | KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the | |
2583 | base addresses will return -EEXIST. | |
3401d546 | 2584 | |
ce01e4e8 CD |
2585 | Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API |
2586 | should be used instead. | |
2587 | ||
2588 | ||
740edfc0 | 2589 | 4.86 KVM_PPC_RTAS_DEFINE_TOKEN |
8e591cb7 ME |
2590 | |
2591 | Capability: KVM_CAP_PPC_RTAS | |
2592 | Architectures: ppc | |
2593 | Type: vm ioctl | |
2594 | Parameters: struct kvm_rtas_token_args | |
2595 | Returns: 0 on success, -1 on error | |
2596 | ||
2597 | Defines a token value for a RTAS (Run Time Abstraction Services) | |
2598 | service in order to allow it to be handled in the kernel. The | |
2599 | argument struct gives the name of the service, which must be the name | |
2600 | of a service that has a kernel-side implementation. If the token | |
2601 | value is non-zero, it will be associated with that service, and | |
2602 | subsequent RTAS calls by the guest specifying that token will be | |
2603 | handled by the kernel. If the token value is 0, then any token | |
2604 | associated with the service will be forgotten, and subsequent RTAS | |
2605 | calls by the guest for that service will be passed to userspace to be | |
2606 | handled. | |
2607 | ||
4bd9d344 AB |
2608 | 4.87 KVM_SET_GUEST_DEBUG |
2609 | ||
2610 | Capability: KVM_CAP_SET_GUEST_DEBUG | |
2611 | Architectures: x86, s390, ppc | |
2612 | Type: vcpu ioctl | |
2613 | Parameters: struct kvm_guest_debug (in) | |
2614 | Returns: 0 on success; -1 on error | |
2615 | ||
2616 | struct kvm_guest_debug { | |
2617 | __u32 control; | |
2618 | __u32 pad; | |
2619 | struct kvm_guest_debug_arch arch; | |
2620 | }; | |
2621 | ||
2622 | Set up the processor specific debug registers and configure vcpu for | |
2623 | handling guest debug events. There are two parts to the structure, the | |
2624 | first a control bitfield indicates the type of debug events to handle | |
2625 | when running. Common control bits are: | |
2626 | ||
2627 | - KVM_GUESTDBG_ENABLE: guest debugging is enabled | |
2628 | - KVM_GUESTDBG_SINGLESTEP: the next run should single-step | |
2629 | ||
2630 | The top 16 bits of the control field are architecture specific control | |
2631 | flags which can include the following: | |
2632 | ||
2633 | - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86] | |
2634 | - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390] | |
2635 | - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86] | |
2636 | - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86] | |
2637 | - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390] | |
2638 | ||
2639 | For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints | |
2640 | are enabled in memory so we need to ensure breakpoint exceptions are | |
2641 | correctly trapped and the KVM run loop exits at the breakpoint and not | |
2642 | running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP | |
2643 | we need to ensure the guest vCPUs architecture specific registers are | |
2644 | updated to the correct (supplied) values. | |
2645 | ||
2646 | The second part of the structure is architecture specific and | |
2647 | typically contains a set of debug registers. | |
2648 | ||
2649 | When debug events exit the main run loop with the reason | |
2650 | KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run | |
2651 | structure containing architecture specific debug information. | |
3401d546 | 2652 | |
209cf19f AB |
2653 | 4.88 KVM_GET_EMULATED_CPUID |
2654 | ||
2655 | Capability: KVM_CAP_EXT_EMUL_CPUID | |
2656 | Architectures: x86 | |
2657 | Type: system ioctl | |
2658 | Parameters: struct kvm_cpuid2 (in/out) | |
2659 | Returns: 0 on success, -1 on error | |
2660 | ||
2661 | struct kvm_cpuid2 { | |
2662 | __u32 nent; | |
2663 | __u32 flags; | |
2664 | struct kvm_cpuid_entry2 entries[0]; | |
2665 | }; | |
2666 | ||
2667 | The member 'flags' is used for passing flags from userspace. | |
2668 | ||
2669 | #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0) | |
2670 | #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) | |
2671 | #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) | |
2672 | ||
2673 | struct kvm_cpuid_entry2 { | |
2674 | __u32 function; | |
2675 | __u32 index; | |
2676 | __u32 flags; | |
2677 | __u32 eax; | |
2678 | __u32 ebx; | |
2679 | __u32 ecx; | |
2680 | __u32 edx; | |
2681 | __u32 padding[3]; | |
2682 | }; | |
2683 | ||
2684 | This ioctl returns x86 cpuid features which are emulated by | |
2685 | kvm.Userspace can use the information returned by this ioctl to query | |
2686 | which features are emulated by kvm instead of being present natively. | |
2687 | ||
2688 | Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2 | |
2689 | structure with the 'nent' field indicating the number of entries in | |
2690 | the variable-size array 'entries'. If the number of entries is too low | |
2691 | to describe the cpu capabilities, an error (E2BIG) is returned. If the | |
2692 | number is too high, the 'nent' field is adjusted and an error (ENOMEM) | |
2693 | is returned. If the number is just right, the 'nent' field is adjusted | |
2694 | to the number of valid entries in the 'entries' array, which is then | |
2695 | filled. | |
2696 | ||
2697 | The entries returned are the set CPUID bits of the respective features | |
2698 | which kvm emulates, as returned by the CPUID instruction, with unknown | |
2699 | or unsupported feature bits cleared. | |
2700 | ||
2701 | Features like x2apic, for example, may not be present in the host cpu | |
2702 | but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be | |
2703 | emulated efficiently and thus not included here. | |
2704 | ||
2705 | The fields in each entry are defined as follows: | |
2706 | ||
2707 | function: the eax value used to obtain the entry | |
2708 | index: the ecx value used to obtain the entry (for entries that are | |
2709 | affected by ecx) | |
2710 | flags: an OR of zero or more of the following: | |
2711 | KVM_CPUID_FLAG_SIGNIFCANT_INDEX: | |
2712 | if the index field is valid | |
2713 | KVM_CPUID_FLAG_STATEFUL_FUNC: | |
2714 | if cpuid for this function returns different values for successive | |
2715 | invocations; there will be several entries with the same function, | |
2716 | all with this flag set | |
2717 | KVM_CPUID_FLAG_STATE_READ_NEXT: | |
2718 | for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is | |
2719 | the first entry to be read by a cpu | |
2720 | eax, ebx, ecx, edx: the values returned by the cpuid instruction for | |
2721 | this function/index combination | |
2722 | ||
9c1b96e3 | 2723 | 5. The kvm_run structure |
414fa985 | 2724 | ------------------------ |
9c1b96e3 AK |
2725 | |
2726 | Application code obtains a pointer to the kvm_run structure by | |
2727 | mmap()ing a vcpu fd. From that point, application code can control | |
2728 | execution by changing fields in kvm_run prior to calling the KVM_RUN | |
2729 | ioctl, and obtain information about the reason KVM_RUN returned by | |
2730 | looking up structure members. | |
2731 | ||
2732 | struct kvm_run { | |
2733 | /* in */ | |
2734 | __u8 request_interrupt_window; | |
2735 | ||
2736 | Request that KVM_RUN return when it becomes possible to inject external | |
2737 | interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. | |
2738 | ||
2739 | __u8 padding1[7]; | |
2740 | ||
2741 | /* out */ | |
2742 | __u32 exit_reason; | |
2743 | ||
2744 | When KVM_RUN has returned successfully (return value 0), this informs | |
2745 | application code why KVM_RUN has returned. Allowable values for this | |
2746 | field are detailed below. | |
2747 | ||
2748 | __u8 ready_for_interrupt_injection; | |
2749 | ||
2750 | If request_interrupt_window has been specified, this field indicates | |
2751 | an interrupt can be injected now with KVM_INTERRUPT. | |
2752 | ||
2753 | __u8 if_flag; | |
2754 | ||
2755 | The value of the current interrupt flag. Only valid if in-kernel | |
2756 | local APIC is not used. | |
2757 | ||
2758 | __u8 padding2[2]; | |
2759 | ||
2760 | /* in (pre_kvm_run), out (post_kvm_run) */ | |
2761 | __u64 cr8; | |
2762 | ||
2763 | The value of the cr8 register. Only valid if in-kernel local APIC is | |
2764 | not used. Both input and output. | |
2765 | ||
2766 | __u64 apic_base; | |
2767 | ||
2768 | The value of the APIC BASE msr. Only valid if in-kernel local | |
2769 | APIC is not used. Both input and output. | |
2770 | ||
2771 | union { | |
2772 | /* KVM_EXIT_UNKNOWN */ | |
2773 | struct { | |
2774 | __u64 hardware_exit_reason; | |
2775 | } hw; | |
2776 | ||
2777 | If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown | |
2778 | reasons. Further architecture-specific information is available in | |
2779 | hardware_exit_reason. | |
2780 | ||
2781 | /* KVM_EXIT_FAIL_ENTRY */ | |
2782 | struct { | |
2783 | __u64 hardware_entry_failure_reason; | |
2784 | } fail_entry; | |
2785 | ||
2786 | If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due | |
2787 | to unknown reasons. Further architecture-specific information is | |
2788 | available in hardware_entry_failure_reason. | |
2789 | ||
2790 | /* KVM_EXIT_EXCEPTION */ | |
2791 | struct { | |
2792 | __u32 exception; | |
2793 | __u32 error_code; | |
2794 | } ex; | |
2795 | ||
2796 | Unused. | |
2797 | ||
2798 | /* KVM_EXIT_IO */ | |
2799 | struct { | |
2800 | #define KVM_EXIT_IO_IN 0 | |
2801 | #define KVM_EXIT_IO_OUT 1 | |
2802 | __u8 direction; | |
2803 | __u8 size; /* bytes */ | |
2804 | __u16 port; | |
2805 | __u32 count; | |
2806 | __u64 data_offset; /* relative to kvm_run start */ | |
2807 | } io; | |
2808 | ||
2044892d | 2809 | If exit_reason is KVM_EXIT_IO, then the vcpu has |
9c1b96e3 AK |
2810 | executed a port I/O instruction which could not be satisfied by kvm. |
2811 | data_offset describes where the data is located (KVM_EXIT_IO_OUT) or | |
2812 | where kvm expects application code to place the data for the next | |
2044892d | 2813 | KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array. |
9c1b96e3 AK |
2814 | |
2815 | struct { | |
2816 | struct kvm_debug_exit_arch arch; | |
2817 | } debug; | |
2818 | ||
2819 | Unused. | |
2820 | ||
2821 | /* KVM_EXIT_MMIO */ | |
2822 | struct { | |
2823 | __u64 phys_addr; | |
2824 | __u8 data[8]; | |
2825 | __u32 len; | |
2826 | __u8 is_write; | |
2827 | } mmio; | |
2828 | ||
2044892d | 2829 | If exit_reason is KVM_EXIT_MMIO, then the vcpu has |
9c1b96e3 AK |
2830 | executed a memory-mapped I/O instruction which could not be satisfied |
2831 | by kvm. The 'data' member contains the written data if 'is_write' is | |
2832 | true, and should be filled by application code otherwise. | |
2833 | ||
6acdb160 CD |
2834 | The 'data' member contains, in its first 'len' bytes, the value as it would |
2835 | appear if the VCPU performed a load or store of the appropriate width directly | |
2836 | to the byte array. | |
2837 | ||
cc568ead | 2838 | NOTE: For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR and |
ce91ddc4 | 2839 | KVM_EXIT_EPR the corresponding |
ad0a048b AG |
2840 | operations are complete (and guest state is consistent) only after userspace |
2841 | has re-entered the kernel with KVM_RUN. The kernel side will first finish | |
67961344 MT |
2842 | incomplete operations and then check for pending signals. Userspace |
2843 | can re-enter the guest with an unmasked signal pending to complete | |
2844 | pending operations. | |
2845 | ||
9c1b96e3 AK |
2846 | /* KVM_EXIT_HYPERCALL */ |
2847 | struct { | |
2848 | __u64 nr; | |
2849 | __u64 args[6]; | |
2850 | __u64 ret; | |
2851 | __u32 longmode; | |
2852 | __u32 pad; | |
2853 | } hypercall; | |
2854 | ||
647dc49e AK |
2855 | Unused. This was once used for 'hypercall to userspace'. To implement |
2856 | such functionality, use KVM_EXIT_IO (x86) or KVM_EXIT_MMIO (all except s390). | |
2857 | Note KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO. | |
9c1b96e3 AK |
2858 | |
2859 | /* KVM_EXIT_TPR_ACCESS */ | |
2860 | struct { | |
2861 | __u64 rip; | |
2862 | __u32 is_write; | |
2863 | __u32 pad; | |
2864 | } tpr_access; | |
2865 | ||
2866 | To be documented (KVM_TPR_ACCESS_REPORTING). | |
2867 | ||
2868 | /* KVM_EXIT_S390_SIEIC */ | |
2869 | struct { | |
2870 | __u8 icptcode; | |
2871 | __u64 mask; /* psw upper half */ | |
2872 | __u64 addr; /* psw lower half */ | |
2873 | __u16 ipa; | |
2874 | __u32 ipb; | |
2875 | } s390_sieic; | |
2876 | ||
2877 | s390 specific. | |
2878 | ||
2879 | /* KVM_EXIT_S390_RESET */ | |
2880 | #define KVM_S390_RESET_POR 1 | |
2881 | #define KVM_S390_RESET_CLEAR 2 | |
2882 | #define KVM_S390_RESET_SUBSYSTEM 4 | |
2883 | #define KVM_S390_RESET_CPU_INIT 8 | |
2884 | #define KVM_S390_RESET_IPL 16 | |
2885 | __u64 s390_reset_flags; | |
2886 | ||
2887 | s390 specific. | |
2888 | ||
e168bf8d CO |
2889 | /* KVM_EXIT_S390_UCONTROL */ |
2890 | struct { | |
2891 | __u64 trans_exc_code; | |
2892 | __u32 pgm_code; | |
2893 | } s390_ucontrol; | |
2894 | ||
2895 | s390 specific. A page fault has occurred for a user controlled virtual | |
2896 | machine (KVM_VM_S390_UNCONTROL) on it's host page table that cannot be | |
2897 | resolved by the kernel. | |
2898 | The program code and the translation exception code that were placed | |
2899 | in the cpu's lowcore are presented here as defined by the z Architecture | |
2900 | Principles of Operation Book in the Chapter for Dynamic Address Translation | |
2901 | (DAT) | |
2902 | ||
9c1b96e3 AK |
2903 | /* KVM_EXIT_DCR */ |
2904 | struct { | |
2905 | __u32 dcrn; | |
2906 | __u32 data; | |
2907 | __u8 is_write; | |
2908 | } dcr; | |
2909 | ||
ce91ddc4 | 2910 | Deprecated - was used for 440 KVM. |
9c1b96e3 | 2911 | |
ad0a048b AG |
2912 | /* KVM_EXIT_OSI */ |
2913 | struct { | |
2914 | __u64 gprs[32]; | |
2915 | } osi; | |
2916 | ||
2917 | MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch | |
2918 | hypercalls and exit with this exit struct that contains all the guest gprs. | |
2919 | ||
2920 | If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall. | |
2921 | Userspace can now handle the hypercall and when it's done modify the gprs as | |
2922 | necessary. Upon guest entry all guest GPRs will then be replaced by the values | |
2923 | in this struct. | |
2924 | ||
de56a948 PM |
2925 | /* KVM_EXIT_PAPR_HCALL */ |
2926 | struct { | |
2927 | __u64 nr; | |
2928 | __u64 ret; | |
2929 | __u64 args[9]; | |
2930 | } papr_hcall; | |
2931 | ||
2932 | This is used on 64-bit PowerPC when emulating a pSeries partition, | |
2933 | e.g. with the 'pseries' machine type in qemu. It occurs when the | |
2934 | guest does a hypercall using the 'sc 1' instruction. The 'nr' field | |
2935 | contains the hypercall number (from the guest R3), and 'args' contains | |
2936 | the arguments (from the guest R4 - R12). Userspace should put the | |
2937 | return code in 'ret' and any extra returned values in args[]. | |
2938 | The possible hypercalls are defined in the Power Architecture Platform | |
2939 | Requirements (PAPR) document available from www.power.org (free | |
2940 | developer registration required to access it). | |
2941 | ||
fa6b7fe9 CH |
2942 | /* KVM_EXIT_S390_TSCH */ |
2943 | struct { | |
2944 | __u16 subchannel_id; | |
2945 | __u16 subchannel_nr; | |
2946 | __u32 io_int_parm; | |
2947 | __u32 io_int_word; | |
2948 | __u32 ipb; | |
2949 | __u8 dequeued; | |
2950 | } s390_tsch; | |
2951 | ||
2952 | s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled | |
2953 | and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O | |
2954 | interrupt for the target subchannel has been dequeued and subchannel_id, | |
2955 | subchannel_nr, io_int_parm and io_int_word contain the parameters for that | |
2956 | interrupt. ipb is needed for instruction parameter decoding. | |
2957 | ||
1c810636 AG |
2958 | /* KVM_EXIT_EPR */ |
2959 | struct { | |
2960 | __u32 epr; | |
2961 | } epr; | |
2962 | ||
2963 | On FSL BookE PowerPC chips, the interrupt controller has a fast patch | |
2964 | interrupt acknowledge path to the core. When the core successfully | |
2965 | delivers an interrupt, it automatically populates the EPR register with | |
2966 | the interrupt vector number and acknowledges the interrupt inside | |
2967 | the interrupt controller. | |
2968 | ||
2969 | In case the interrupt controller lives in user space, we need to do | |
2970 | the interrupt acknowledge cycle through it to fetch the next to be | |
2971 | delivered interrupt vector using this exit. | |
2972 | ||
2973 | It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an | |
2974 | external interrupt has just been delivered into the guest. User space | |
2975 | should put the acknowledged interrupt vector into the 'epr' field. | |
2976 | ||
8ad6b634 AP |
2977 | /* KVM_EXIT_SYSTEM_EVENT */ |
2978 | struct { | |
2979 | #define KVM_SYSTEM_EVENT_SHUTDOWN 1 | |
2980 | #define KVM_SYSTEM_EVENT_RESET 2 | |
2981 | __u32 type; | |
2982 | __u64 flags; | |
2983 | } system_event; | |
2984 | ||
2985 | If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered | |
2986 | a system-level event using some architecture specific mechanism (hypercall | |
2987 | or some special instruction). In case of ARM/ARM64, this is triggered using | |
2988 | HVC instruction based PSCI call from the vcpu. The 'type' field describes | |
2989 | the system-level event type. The 'flags' field describes architecture | |
2990 | specific flags for the system-level event. | |
2991 | ||
cf5d3188 CD |
2992 | Valid values for 'type' are: |
2993 | KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the | |
2994 | VM. Userspace is not obliged to honour this, and if it does honour | |
2995 | this does not need to destroy the VM synchronously (ie it may call | |
2996 | KVM_RUN again before shutdown finally occurs). | |
2997 | KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM. | |
2998 | As with SHUTDOWN, userspace can choose to ignore the request, or | |
2999 | to schedule the reset to occur in the future and may call KVM_RUN again. | |
3000 | ||
9c1b96e3 AK |
3001 | /* Fix the size of the union. */ |
3002 | char padding[256]; | |
3003 | }; | |
b9e5dc8d CB |
3004 | |
3005 | /* | |
3006 | * shared registers between kvm and userspace. | |
3007 | * kvm_valid_regs specifies the register classes set by the host | |
3008 | * kvm_dirty_regs specified the register classes dirtied by userspace | |
3009 | * struct kvm_sync_regs is architecture specific, as well as the | |
3010 | * bits for kvm_valid_regs and kvm_dirty_regs | |
3011 | */ | |
3012 | __u64 kvm_valid_regs; | |
3013 | __u64 kvm_dirty_regs; | |
3014 | union { | |
3015 | struct kvm_sync_regs regs; | |
3016 | char padding[1024]; | |
3017 | } s; | |
3018 | ||
3019 | If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access | |
3020 | certain guest registers without having to call SET/GET_*REGS. Thus we can | |
3021 | avoid some system call overhead if userspace has to handle the exit. | |
3022 | Userspace can query the validity of the structure by checking | |
3023 | kvm_valid_regs for specific bits. These bits are architecture specific | |
3024 | and usually define the validity of a groups of registers. (e.g. one bit | |
3025 | for general purpose registers) | |
3026 | ||
d8482c0d DH |
3027 | Please note that the kernel is allowed to use the kvm_run structure as the |
3028 | primary storage for certain register types. Therefore, the kernel may use the | |
3029 | values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set. | |
3030 | ||
9c1b96e3 | 3031 | }; |
821246a5 | 3032 | |
414fa985 | 3033 | |
9c15bb1d | 3034 | |
699a0ea0 PM |
3035 | 6. Capabilities that can be enabled on vCPUs |
3036 | -------------------------------------------- | |
821246a5 | 3037 | |
0907c855 CH |
3038 | There are certain capabilities that change the behavior of the virtual CPU or |
3039 | the virtual machine when enabled. To enable them, please see section 4.37. | |
3040 | Below you can find a list of capabilities and what their effect on the vCPU or | |
3041 | the virtual machine is when enabling them. | |
821246a5 AG |
3042 | |
3043 | The following information is provided along with the description: | |
3044 | ||
3045 | Architectures: which instruction set architectures provide this ioctl. | |
3046 | x86 includes both i386 and x86_64. | |
3047 | ||
0907c855 CH |
3048 | Target: whether this is a per-vcpu or per-vm capability. |
3049 | ||
821246a5 AG |
3050 | Parameters: what parameters are accepted by the capability. |
3051 | ||
3052 | Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) | |
3053 | are not detailed, but errors with specific meanings are. | |
3054 | ||
414fa985 | 3055 | |
821246a5 AG |
3056 | 6.1 KVM_CAP_PPC_OSI |
3057 | ||
3058 | Architectures: ppc | |
0907c855 | 3059 | Target: vcpu |
821246a5 AG |
3060 | Parameters: none |
3061 | Returns: 0 on success; -1 on error | |
3062 | ||
3063 | This capability enables interception of OSI hypercalls that otherwise would | |
3064 | be treated as normal system calls to be injected into the guest. OSI hypercalls | |
3065 | were invented by Mac-on-Linux to have a standardized communication mechanism | |
3066 | between the guest and the host. | |
3067 | ||
3068 | When this capability is enabled, KVM_EXIT_OSI can occur. | |
3069 | ||
414fa985 | 3070 | |
821246a5 AG |
3071 | 6.2 KVM_CAP_PPC_PAPR |
3072 | ||
3073 | Architectures: ppc | |
0907c855 | 3074 | Target: vcpu |
821246a5 AG |
3075 | Parameters: none |
3076 | Returns: 0 on success; -1 on error | |
3077 | ||
3078 | This capability enables interception of PAPR hypercalls. PAPR hypercalls are | |
3079 | done using the hypercall instruction "sc 1". | |
3080 | ||
3081 | It also sets the guest privilege level to "supervisor" mode. Usually the guest | |
3082 | runs in "hypervisor" privilege mode with a few missing features. | |
3083 | ||
3084 | In addition to the above, it changes the semantics of SDR1. In this mode, the | |
3085 | HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the | |
3086 | HTAB invisible to the guest. | |
3087 | ||
3088 | When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur. | |
dc83b8bc | 3089 | |
414fa985 | 3090 | |
dc83b8bc SW |
3091 | 6.3 KVM_CAP_SW_TLB |
3092 | ||
3093 | Architectures: ppc | |
0907c855 | 3094 | Target: vcpu |
dc83b8bc SW |
3095 | Parameters: args[0] is the address of a struct kvm_config_tlb |
3096 | Returns: 0 on success; -1 on error | |
3097 | ||
3098 | struct kvm_config_tlb { | |
3099 | __u64 params; | |
3100 | __u64 array; | |
3101 | __u32 mmu_type; | |
3102 | __u32 array_len; | |
3103 | }; | |
3104 | ||
3105 | Configures the virtual CPU's TLB array, establishing a shared memory area | |
3106 | between userspace and KVM. The "params" and "array" fields are userspace | |
3107 | addresses of mmu-type-specific data structures. The "array_len" field is an | |
3108 | safety mechanism, and should be set to the size in bytes of the memory that | |
3109 | userspace has reserved for the array. It must be at least the size dictated | |
3110 | by "mmu_type" and "params". | |
3111 | ||
3112 | While KVM_RUN is active, the shared region is under control of KVM. Its | |
3113 | contents are undefined, and any modification by userspace results in | |
3114 | boundedly undefined behavior. | |
3115 | ||
3116 | On return from KVM_RUN, the shared region will reflect the current state of | |
3117 | the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB | |
3118 | to tell KVM which entries have been changed, prior to calling KVM_RUN again | |
3119 | on this vcpu. | |
3120 | ||
3121 | For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV: | |
3122 | - The "params" field is of type "struct kvm_book3e_206_tlb_params". | |
3123 | - The "array" field points to an array of type "struct | |
3124 | kvm_book3e_206_tlb_entry". | |
3125 | - The array consists of all entries in the first TLB, followed by all | |
3126 | entries in the second TLB. | |
3127 | - Within a TLB, entries are ordered first by increasing set number. Within a | |
3128 | set, entries are ordered by way (increasing ESEL). | |
3129 | - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1) | |
3130 | where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value. | |
3131 | - The tsize field of mas1 shall be set to 4K on TLB0, even though the | |
3132 | hardware ignores this value for TLB0. | |
fa6b7fe9 CH |
3133 | |
3134 | 6.4 KVM_CAP_S390_CSS_SUPPORT | |
3135 | ||
3136 | Architectures: s390 | |
0907c855 | 3137 | Target: vcpu |
fa6b7fe9 CH |
3138 | Parameters: none |
3139 | Returns: 0 on success; -1 on error | |
3140 | ||
3141 | This capability enables support for handling of channel I/O instructions. | |
3142 | ||
3143 | TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are | |
3144 | handled in-kernel, while the other I/O instructions are passed to userspace. | |
3145 | ||
3146 | When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST | |
3147 | SUBCHANNEL intercepts. | |
1c810636 | 3148 | |
0907c855 CH |
3149 | Note that even though this capability is enabled per-vcpu, the complete |
3150 | virtual machine is affected. | |
3151 | ||
1c810636 AG |
3152 | 6.5 KVM_CAP_PPC_EPR |
3153 | ||
3154 | Architectures: ppc | |
0907c855 | 3155 | Target: vcpu |
1c810636 AG |
3156 | Parameters: args[0] defines whether the proxy facility is active |
3157 | Returns: 0 on success; -1 on error | |
3158 | ||
3159 | This capability enables or disables the delivery of interrupts through the | |
3160 | external proxy facility. | |
3161 | ||
3162 | When enabled (args[0] != 0), every time the guest gets an external interrupt | |
3163 | delivered, it automatically exits into user space with a KVM_EXIT_EPR exit | |
3164 | to receive the topmost interrupt vector. | |
3165 | ||
3166 | When disabled (args[0] == 0), behavior is as if this facility is unsupported. | |
3167 | ||
3168 | When this capability is enabled, KVM_EXIT_EPR can occur. | |
eb1e4f43 SW |
3169 | |
3170 | 6.6 KVM_CAP_IRQ_MPIC | |
3171 | ||
3172 | Architectures: ppc | |
3173 | Parameters: args[0] is the MPIC device fd | |
3174 | args[1] is the MPIC CPU number for this vcpu | |
3175 | ||
3176 | This capability connects the vcpu to an in-kernel MPIC device. | |
5975a2e0 PM |
3177 | |
3178 | 6.7 KVM_CAP_IRQ_XICS | |
3179 | ||
3180 | Architectures: ppc | |
0907c855 | 3181 | Target: vcpu |
5975a2e0 PM |
3182 | Parameters: args[0] is the XICS device fd |
3183 | args[1] is the XICS CPU number (server ID) for this vcpu | |
3184 | ||
3185 | This capability connects the vcpu to an in-kernel XICS device. | |
8a366a4b CH |
3186 | |
3187 | 6.8 KVM_CAP_S390_IRQCHIP | |
3188 | ||
3189 | Architectures: s390 | |
3190 | Target: vm | |
3191 | Parameters: none | |
3192 | ||
3193 | This capability enables the in-kernel irqchip for s390. Please refer to | |
3194 | "4.24 KVM_CREATE_IRQCHIP" for details. | |
699a0ea0 PM |
3195 | |
3196 | 7. Capabilities that can be enabled on VMs | |
3197 | ------------------------------------------ | |
3198 | ||
3199 | There are certain capabilities that change the behavior of the virtual | |
3200 | machine when enabled. To enable them, please see section 4.37. Below | |
3201 | you can find a list of capabilities and what their effect on the VM | |
3202 | is when enabling them. | |
3203 | ||
3204 | The following information is provided along with the description: | |
3205 | ||
3206 | Architectures: which instruction set architectures provide this ioctl. | |
3207 | x86 includes both i386 and x86_64. | |
3208 | ||
3209 | Parameters: what parameters are accepted by the capability. | |
3210 | ||
3211 | Returns: the return value. General error numbers (EBADF, ENOMEM, EINVAL) | |
3212 | are not detailed, but errors with specific meanings are. | |
3213 | ||
3214 | ||
3215 | 7.1 KVM_CAP_PPC_ENABLE_HCALL | |
3216 | ||
3217 | Architectures: ppc | |
3218 | Parameters: args[0] is the sPAPR hcall number | |
3219 | args[1] is 0 to disable, 1 to enable in-kernel handling | |
3220 | ||
3221 | This capability controls whether individual sPAPR hypercalls (hcalls) | |
3222 | get handled by the kernel or not. Enabling or disabling in-kernel | |
3223 | handling of an hcall is effective across the VM. On creation, an | |
3224 | initial set of hcalls are enabled for in-kernel handling, which | |
3225 | consists of those hcalls for which in-kernel handlers were implemented | |
3226 | before this capability was implemented. If disabled, the kernel will | |
3227 | not to attempt to handle the hcall, but will always exit to userspace | |
3228 | to handle it. Note that it may not make sense to enable some and | |
3229 | disable others of a group of related hcalls, but KVM does not prevent | |
3230 | userspace from doing that. | |
ae2113a4 PM |
3231 | |
3232 | If the hcall number specified is not one that has an in-kernel | |
3233 | implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL | |
3234 | error. | |
2444b352 DH |
3235 | |
3236 | 7.2 KVM_CAP_S390_USER_SIGP | |
3237 | ||
3238 | Architectures: s390 | |
3239 | Parameters: none | |
3240 | ||
3241 | This capability controls which SIGP orders will be handled completely in user | |
3242 | space. With this capability enabled, all fast orders will be handled completely | |
3243 | in the kernel: | |
3244 | - SENSE | |
3245 | - SENSE RUNNING | |
3246 | - EXTERNAL CALL | |
3247 | - EMERGENCY SIGNAL | |
3248 | - CONDITIONAL EMERGENCY SIGNAL | |
3249 | ||
3250 | All other orders will be handled completely in user space. | |
3251 | ||
3252 | Only privileged operation exceptions will be checked for in the kernel (or even | |
3253 | in the hardware prior to interception). If this capability is not enabled, the | |
3254 | old way of handling SIGP orders is used (partially in kernel and user space). |