1 .. SPDX-License-Identifier: GPL-2.0 1 .. SPDX-License-Identifier: GPL-2.0
2 2
3 ============================================== 3 ===================================================================
4 The Definitive KVM (Kernel-based Virtual Machi 4 The Definitive KVM (Kernel-based Virtual Machine) API Documentation
5 ============================================== 5 ===================================================================
6 6
7 1. General description 7 1. General description
8 ====================== 8 ======================
9 9
10 The kvm API is a set of ioctls that are issued 10 The kvm API is a set of ioctls that are issued to control various aspects
11 of a virtual machine. The ioctls belong to th 11 of a virtual machine. The ioctls belong to the following classes:
12 12
13 - System ioctls: These query and set global a 13 - System ioctls: These query and set global attributes which affect the
14 whole kvm subsystem. In addition a system 14 whole kvm subsystem. In addition a system ioctl is used to create
15 virtual machines. 15 virtual machines.
16 16
17 - VM ioctls: These query and set attributes t 17 - VM ioctls: These query and set attributes that affect an entire virtual
18 machine, for example memory layout. In add 18 machine, for example memory layout. In addition a VM ioctl is used to
19 create virtual cpus (vcpus) and devices. 19 create virtual cpus (vcpus) and devices.
20 20
21 VM ioctls must be issued from the same proc 21 VM ioctls must be issued from the same process (address space) that was
22 used to create the VM. 22 used to create the VM.
23 23
24 - vcpu ioctls: These query and set attributes 24 - vcpu ioctls: These query and set attributes that control the operation
25 of a single virtual cpu. 25 of a single virtual cpu.
26 26
27 vcpu ioctls should be issued from the same 27 vcpu ioctls should be issued from the same thread that was used to create
28 the vcpu, except for asynchronous vcpu ioct 28 the vcpu, except for asynchronous vcpu ioctl that are marked as such in
29 the documentation. Otherwise, the first io 29 the documentation. Otherwise, the first ioctl after switching threads
30 could see a performance impact. 30 could see a performance impact.
31 31
32 - device ioctls: These query and set attribut 32 - device ioctls: These query and set attributes that control the operation
33 of a single device. 33 of a single device.
34 34
35 device ioctls must be issued from the same 35 device ioctls must be issued from the same process (address space) that
36 was used to create the VM. 36 was used to create the VM.
37 37
38 2. File descriptors 38 2. File descriptors
39 =================== 39 ===================
40 40
41 The kvm API is centered around file descriptor 41 The kvm API is centered around file descriptors. An initial
42 open("/dev/kvm") obtains a handle to the kvm s 42 open("/dev/kvm") obtains a handle to the kvm subsystem; this handle
43 can be used to issue system ioctls. A KVM_CRE 43 can be used to issue system ioctls. A KVM_CREATE_VM ioctl on this
44 handle will create a VM file descriptor which 44 handle will create a VM file descriptor which can be used to issue VM
45 ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVIC 45 ioctls. A KVM_CREATE_VCPU or KVM_CREATE_DEVICE ioctl on a VM fd will
46 create a virtual cpu or device and return a fi 46 create a virtual cpu or device and return a file descriptor pointing to
47 the new resource. Finally, ioctls on a vcpu o 47 the new resource. Finally, ioctls on a vcpu or device fd can be used
48 to control the vcpu or device. For vcpus, thi 48 to control the vcpu or device. For vcpus, this includes the important
49 task of actually running guest code. 49 task of actually running guest code.
50 50
51 In general file descriptors can be migrated am 51 In general file descriptors can be migrated among processes by means
52 of fork() and the SCM_RIGHTS facility of unix 52 of fork() and the SCM_RIGHTS facility of unix domain socket. These
53 kinds of tricks are explicitly not supported b 53 kinds of tricks are explicitly not supported by kvm. While they will
54 not cause harm to the host, their actual behav 54 not cause harm to the host, their actual behavior is not guaranteed by
55 the API. See "General description" for detail 55 the API. See "General description" for details on the ioctl usage
56 model that is supported by KVM. 56 model that is supported by KVM.
57 57
58 It is important to note that although VM ioctl 58 It is important to note that although VM ioctls may only be issued from
59 the process that created the VM, a VM's lifecy 59 the process that created the VM, a VM's lifecycle is associated with its
60 file descriptor, not its creator (process). I 60 file descriptor, not its creator (process). In other words, the VM and
61 its resources, *including the associated addre 61 its resources, *including the associated address space*, are not freed
62 until the last reference to the VM's file desc 62 until the last reference to the VM's file descriptor has been released.
63 For example, if fork() is issued after ioctl(K 63 For example, if fork() is issued after ioctl(KVM_CREATE_VM), the VM will
64 not be freed until both the parent (original) 64 not be freed until both the parent (original) process and its child have
65 put their references to the VM's file descript 65 put their references to the VM's file descriptor.
66 66
67 Because a VM's resources are not freed until t 67 Because a VM's resources are not freed until the last reference to its
68 file descriptor is released, creating addition 68 file descriptor is released, creating additional references to a VM
69 via fork(), dup(), etc... without careful cons 69 via fork(), dup(), etc... without careful consideration is strongly
70 discouraged and may have unwanted side effects 70 discouraged and may have unwanted side effects, e.g. memory allocated
71 by and on behalf of the VM's process may not b 71 by and on behalf of the VM's process may not be freed/unaccounted when
72 the VM is shut down. 72 the VM is shut down.
73 73
74 74
75 3. Extensions 75 3. Extensions
76 ============= 76 =============
77 77
78 As of Linux 2.6.22, the KVM ABI has been stabi 78 As of Linux 2.6.22, the KVM ABI has been stabilized: no backward
79 incompatible change are allowed. However, the 79 incompatible change are allowed. However, there is an extension
80 facility that allows backward-compatible exten 80 facility that allows backward-compatible extensions to the API to be
81 queried and used. 81 queried and used.
82 82
83 The extension mechanism is not based on the Li 83 The extension mechanism is not based on the Linux version number.
84 Instead, kvm defines extension identifiers and 84 Instead, kvm defines extension identifiers and a facility to query
85 whether a particular extension identifier is a 85 whether a particular extension identifier is available. If it is, a
86 set of ioctls is available for application use 86 set of ioctls is available for application use.
87 87
88 88
89 4. API description 89 4. API description
90 ================== 90 ==================
91 91
92 This section describes ioctls that can be used 92 This section describes ioctls that can be used to control kvm guests.
93 For each ioctl, the following information is p 93 For each ioctl, the following information is provided along with a
94 description: 94 description:
95 95
96 Capability: 96 Capability:
97 which KVM extension provides this ioctl. 97 which KVM extension provides this ioctl. Can be 'basic',
98 which means that is will be provided by 98 which means that is will be provided by any kernel that supports
99 API version 12 (see section 4.1), a KVM_ 99 API version 12 (see section 4.1), a KVM_CAP_xyz constant, which
100 means availability needs to be checked w 100 means availability needs to be checked with KVM_CHECK_EXTENSION
101 (see section 4.4), or 'none' which means 101 (see section 4.4), or 'none' which means that while not all kernels
102 support this ioctl, there's no capabilit 102 support this ioctl, there's no capability bit to check its
103 availability: for kernels that don't sup 103 availability: for kernels that don't support the ioctl,
104 the ioctl returns -ENOTTY. 104 the ioctl returns -ENOTTY.
105 105
106 Architectures: 106 Architectures:
107 which instruction set architectures prov 107 which instruction set architectures provide this ioctl.
108 x86 includes both i386 and x86_64. 108 x86 includes both i386 and x86_64.
109 109
110 Type: 110 Type:
111 system, vm, or vcpu. 111 system, vm, or vcpu.
112 112
113 Parameters: 113 Parameters:
114 what parameters are accepted by the ioct 114 what parameters are accepted by the ioctl.
115 115
116 Returns: 116 Returns:
117 the return value. General error numbers 117 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
118 are not detailed, but errors with specif 118 are not detailed, but errors with specific meanings are.
119 119
120 120
121 4.1 KVM_GET_API_VERSION 121 4.1 KVM_GET_API_VERSION
122 ----------------------- 122 -----------------------
123 123
124 :Capability: basic 124 :Capability: basic
125 :Architectures: all 125 :Architectures: all
126 :Type: system ioctl 126 :Type: system ioctl
127 :Parameters: none 127 :Parameters: none
128 :Returns: the constant KVM_API_VERSION (=12) 128 :Returns: the constant KVM_API_VERSION (=12)
129 129
130 This identifies the API version as the stable 130 This identifies the API version as the stable kvm API. It is not
131 expected that this number will change. Howeve 131 expected that this number will change. However, Linux 2.6.20 and
132 2.6.21 report earlier versions; these are not 132 2.6.21 report earlier versions; these are not documented and not
133 supported. Applications should refuse to run 133 supported. Applications should refuse to run if KVM_GET_API_VERSION
134 returns a value other than 12. If this check 134 returns a value other than 12. If this check passes, all ioctls
135 described as 'basic' will be available. 135 described as 'basic' will be available.
136 136
137 137
138 4.2 KVM_CREATE_VM 138 4.2 KVM_CREATE_VM
139 ----------------- 139 -----------------
140 140
141 :Capability: basic 141 :Capability: basic
142 :Architectures: all 142 :Architectures: all
143 :Type: system ioctl 143 :Type: system ioctl
144 :Parameters: machine type identifier (KVM_VM_* 144 :Parameters: machine type identifier (KVM_VM_*)
145 :Returns: a VM fd that can be used to control 145 :Returns: a VM fd that can be used to control the new virtual machine.
146 146
147 The new VM has no virtual cpus and no memory. 147 The new VM has no virtual cpus and no memory.
148 You probably want to use 0 as machine type. 148 You probably want to use 0 as machine type.
149 149
150 X86: <<
151 ^^^^ <<
152 <<
153 Supported X86 VM types can be queried via KVM_ <<
154 <<
155 S390: <<
156 ^^^^^ <<
157 <<
158 In order to create user controlled virtual mac 150 In order to create user controlled virtual machines on S390, check
159 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_ 151 KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL as
160 privileged user (CAP_SYS_ADMIN). 152 privileged user (CAP_SYS_ADMIN).
161 153
162 MIPS: <<
163 ^^^^^ <<
164 <<
165 To use hardware assisted virtualization on MIP <<
166 the default trap & emulate implementation (whi <<
167 memory layout to fit in user mode), check KVM_ <<
168 flag KVM_VM_MIPS_VZ. <<
169 <<
170 ARM64: <<
171 ^^^^^^ <<
172 <<
173 On arm64, the physical address size for a VM ( 154 On arm64, the physical address size for a VM (IPA Size limit) is limited
174 to 40bits by default. The limit can be configu 155 to 40bits by default. The limit can be configured if the host supports the
175 extension KVM_CAP_ARM_VM_IPA_SIZE. When suppor 156 extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use
176 KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the 157 KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type
177 identifier, where IPA_Bits is the maximum widt 158 identifier, where IPA_Bits is the maximum width of any physical
178 address used by the VM. The IPA_Bits is encode 159 address used by the VM. The IPA_Bits is encoded in bits[7-0] of the
179 machine type identifier. 160 machine type identifier.
180 161
181 e.g, to configure a guest to use 48bit physica 162 e.g, to configure a guest to use 48bit physical address size::
182 163
183 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_V 164 vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48));
184 165
185 The requested size (IPA_Bits) must be: 166 The requested size (IPA_Bits) must be:
186 167
187 == ======================================== 168 == =========================================================
188 0 Implies default size, 40bits (for backwa 169 0 Implies default size, 40bits (for backward compatibility)
189 N Implies N bits, where N is a positive in 170 N Implies N bits, where N is a positive integer such that,
190 32 <= N <= Host_IPA_Limit 171 32 <= N <= Host_IPA_Limit
191 == ======================================== 172 == =========================================================
192 173
193 Host_IPA_Limit is the maximum possible value f 174 Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and
194 is dependent on the CPU capability and the ker 175 is dependent on the CPU capability and the kernel configuration. The limit can
195 be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of 176 be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION
196 ioctl() at run-time. 177 ioctl() at run-time.
197 178
198 Creation of the VM will fail if the requested 179 Creation of the VM will fail if the requested IPA size (whether it is
199 implicit or explicit) is unsupported on the ho 180 implicit or explicit) is unsupported on the host.
200 181
201 Please note that configuring the IPA size does 182 Please note that configuring the IPA size does not affect the capability
202 exposed by the guest CPUs in ID_AA64MMFR0_EL1[ 183 exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects
203 size of the address translated by the stage2 l 184 size of the address translated by the stage2 level (guest physical to
204 host physical address translations). 185 host physical address translations).
205 186
206 187
207 4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATUR 188 4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST
208 ---------------------------------------------- 189 ----------------------------------------------------------
209 190
210 :Capability: basic, KVM_CAP_GET_MSR_FEATURES f 191 :Capability: basic, KVM_CAP_GET_MSR_FEATURES for KVM_GET_MSR_FEATURE_INDEX_LIST
211 :Architectures: x86 192 :Architectures: x86
212 :Type: system ioctl 193 :Type: system ioctl
213 :Parameters: struct kvm_msr_list (in/out) 194 :Parameters: struct kvm_msr_list (in/out)
214 :Returns: 0 on success; -1 on error 195 :Returns: 0 on success; -1 on error
215 196
216 Errors: 197 Errors:
217 198
218 ====== ================================= 199 ====== ============================================================
219 EFAULT the msr index list cannot be read 200 EFAULT the msr index list cannot be read from or written to
220 E2BIG the msr index list is too big to 201 E2BIG the msr index list is too big to fit in the array specified by
221 the user. 202 the user.
222 ====== ================================= 203 ====== ============================================================
223 204
224 :: 205 ::
225 206
226 struct kvm_msr_list { 207 struct kvm_msr_list {
227 __u32 nmsrs; /* number of msrs in entr 208 __u32 nmsrs; /* number of msrs in entries */
228 __u32 indices[0]; 209 __u32 indices[0];
229 }; 210 };
230 211
231 The user fills in the size of the indices arra 212 The user fills in the size of the indices array in nmsrs, and in return
232 kvm adjusts nmsrs to reflect the actual number 213 kvm adjusts nmsrs to reflect the actual number of msrs and fills in the
233 indices array with their numbers. 214 indices array with their numbers.
234 215
235 KVM_GET_MSR_INDEX_LIST returns the guest msrs 216 KVM_GET_MSR_INDEX_LIST returns the guest msrs that are supported. The list
236 varies by kvm version and host processor, but 217 varies by kvm version and host processor, but does not change otherwise.
237 218
238 Note: if kvm indicates supports MCE (KVM_CAP_M 219 Note: if kvm indicates supports MCE (KVM_CAP_MCE), then the MCE bank MSRs are
239 not returned in the MSR list, as different vcp 220 not returned in the MSR list, as different vcpus can have a different number
240 of banks, as set via the KVM_X86_SETUP_MCE ioc 221 of banks, as set via the KVM_X86_SETUP_MCE ioctl.
241 222
242 KVM_GET_MSR_FEATURE_INDEX_LIST returns the lis 223 KVM_GET_MSR_FEATURE_INDEX_LIST returns the list of MSRs that can be passed
243 to the KVM_GET_MSRS system ioctl. This lets u 224 to the KVM_GET_MSRS system ioctl. This lets userspace probe host capabilities
244 and processor features that are exposed via MS 225 and processor features that are exposed via MSRs (e.g., VMX capabilities).
245 This list also varies by kvm version and host 226 This list also varies by kvm version and host processor, but does not change
246 otherwise. 227 otherwise.
247 228
248 229
249 4.4 KVM_CHECK_EXTENSION 230 4.4 KVM_CHECK_EXTENSION
250 ----------------------- 231 -----------------------
251 232
252 :Capability: basic, KVM_CAP_CHECK_EXTENSION_VM 233 :Capability: basic, KVM_CAP_CHECK_EXTENSION_VM for vm ioctl
253 :Architectures: all 234 :Architectures: all
254 :Type: system ioctl, vm ioctl 235 :Type: system ioctl, vm ioctl
255 :Parameters: extension identifier (KVM_CAP_*) 236 :Parameters: extension identifier (KVM_CAP_*)
256 :Returns: 0 if unsupported; 1 (or some other p 237 :Returns: 0 if unsupported; 1 (or some other positive integer) if supported
257 238
258 The API allows the application to query about 239 The API allows the application to query about extensions to the core
259 kvm API. Userspace passes an extension identi 240 kvm API. Userspace passes an extension identifier (an integer) and
260 receives an integer that describes the extensi 241 receives an integer that describes the extension availability.
261 Generally 0 means no and 1 means yes, but some 242 Generally 0 means no and 1 means yes, but some extensions may report
262 additional information in the integer return v 243 additional information in the integer return value.
263 244
264 Based on their initialization different VMs ma 245 Based on their initialization different VMs may have different capabilities.
265 It is thus encouraged to use the vm ioctl to q 246 It is thus encouraged to use the vm ioctl to query for capabilities (available
266 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd) 247 with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
267 248
268 4.5 KVM_GET_VCPU_MMAP_SIZE 249 4.5 KVM_GET_VCPU_MMAP_SIZE
269 -------------------------- 250 --------------------------
270 251
271 :Capability: basic 252 :Capability: basic
272 :Architectures: all 253 :Architectures: all
273 :Type: system ioctl 254 :Type: system ioctl
274 :Parameters: none 255 :Parameters: none
275 :Returns: size of vcpu mmap area, in bytes 256 :Returns: size of vcpu mmap area, in bytes
276 257
277 The KVM_RUN ioctl (cf.) communicates with user 258 The KVM_RUN ioctl (cf.) communicates with userspace via a shared
278 memory region. This ioctl returns the size of 259 memory region. This ioctl returns the size of that region. See the
279 KVM_RUN documentation for details. 260 KVM_RUN documentation for details.
280 261
281 Besides the size of the KVM_RUN communication 262 Besides the size of the KVM_RUN communication region, other areas of
282 the VCPU file descriptor can be mmap-ed, inclu 263 the VCPU file descriptor can be mmap-ed, including:
283 264
284 - if KVM_CAP_COALESCED_MMIO is available, a pa 265 - if KVM_CAP_COALESCED_MMIO is available, a page at
285 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; 266 KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons,
286 this page is included in the result of KVM_G 267 this page is included in the result of KVM_GET_VCPU_MMAP_SIZE.
287 KVM_CAP_COALESCED_MMIO is not documented yet 268 KVM_CAP_COALESCED_MMIO is not documented yet.
288 269
289 - if KVM_CAP_DIRTY_LOG_RING is available, a nu 270 - if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at
290 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For 271 KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE. For more information on
291 KVM_CAP_DIRTY_LOG_RING, see section 8.3. 272 KVM_CAP_DIRTY_LOG_RING, see section 8.3.
292 273
293 274
294 4.7 KVM_CREATE_VCPU 275 4.7 KVM_CREATE_VCPU
295 ------------------- 276 -------------------
296 277
297 :Capability: basic 278 :Capability: basic
298 :Architectures: all 279 :Architectures: all
299 :Type: vm ioctl 280 :Type: vm ioctl
300 :Parameters: vcpu id (apic id on x86) 281 :Parameters: vcpu id (apic id on x86)
301 :Returns: vcpu fd on success, -1 on error 282 :Returns: vcpu fd on success, -1 on error
302 283
303 This API adds a vcpu to a virtual machine. No 284 This API adds a vcpu to a virtual machine. No more than max_vcpus may be added.
304 The vcpu id is an integer in the range [0, max 285 The vcpu id is an integer in the range [0, max_vcpu_id).
305 286
306 The recommended max_vcpus value can be retriev 287 The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of
307 the KVM_CHECK_EXTENSION ioctl() at run-time. 288 the KVM_CHECK_EXTENSION ioctl() at run-time.
308 The maximum possible value for max_vcpus can b 289 The maximum possible value for max_vcpus can be retrieved using the
309 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION i 290 KVM_CAP_MAX_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time.
310 291
311 If the KVM_CAP_NR_VCPUS does not exist, you sh 292 If the KVM_CAP_NR_VCPUS does not exist, you should assume that max_vcpus is 4
312 cpus max. 293 cpus max.
313 If the KVM_CAP_MAX_VCPUS does not exist, you s 294 If the KVM_CAP_MAX_VCPUS does not exist, you should assume that max_vcpus is
314 same as the value returned from KVM_CAP_NR_VCP 295 same as the value returned from KVM_CAP_NR_VCPUS.
315 296
316 The maximum possible value for max_vcpu_id can 297 The maximum possible value for max_vcpu_id can be retrieved using the
317 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION 298 KVM_CAP_MAX_VCPU_ID of the KVM_CHECK_EXTENSION ioctl() at run-time.
318 299
319 If the KVM_CAP_MAX_VCPU_ID does not exist, you 300 If the KVM_CAP_MAX_VCPU_ID does not exist, you should assume that max_vcpu_id
320 is the same as the value returned from KVM_CAP 301 is the same as the value returned from KVM_CAP_MAX_VCPUS.
321 302
322 On powerpc using book3s_hv mode, the vcpus are 303 On powerpc using book3s_hv mode, the vcpus are mapped onto virtual
323 threads in one or more virtual CPU cores. (Th 304 threads in one or more virtual CPU cores. (This is because the
324 hardware requires all the hardware threads in 305 hardware requires all the hardware threads in a CPU core to be in the
325 same partition.) The KVM_CAP_PPC_SMT capabili 306 same partition.) The KVM_CAP_PPC_SMT capability indicates the number
326 of vcpus per virtual core (vcore). The vcore 307 of vcpus per virtual core (vcore). The vcore id is obtained by
327 dividing the vcpu id by the number of vcpus pe 308 dividing the vcpu id by the number of vcpus per vcore. The vcpus in a
328 given vcore will always be in the same physica 309 given vcore will always be in the same physical core as each other
329 (though that might be a different physical cor 310 (though that might be a different physical core from time to time).
330 Userspace can control the threading (SMT) mode 311 Userspace can control the threading (SMT) mode of the guest by its
331 allocation of vcpu ids. For example, if users 312 allocation of vcpu ids. For example, if userspace wants
332 single-threaded guest vcpus, it should make al 313 single-threaded guest vcpus, it should make all vcpu ids be a multiple
333 of the number of vcpus per vcore. 314 of the number of vcpus per vcore.
334 315
335 For virtual cpus that have been created with S 316 For virtual cpus that have been created with S390 user controlled virtual
336 machines, the resulting vcpu fd can be memory 317 machines, the resulting vcpu fd can be memory mapped at page offset
337 KVM_S390_SIE_PAGE_OFFSET in order to obtain a 318 KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of the virtual
338 cpu's hardware control block. 319 cpu's hardware control block.
339 320
340 321
341 4.8 KVM_GET_DIRTY_LOG (vm ioctl) 322 4.8 KVM_GET_DIRTY_LOG (vm ioctl)
342 -------------------------------- 323 --------------------------------
343 324
344 :Capability: basic 325 :Capability: basic
345 :Architectures: all 326 :Architectures: all
346 :Type: vm ioctl 327 :Type: vm ioctl
347 :Parameters: struct kvm_dirty_log (in/out) 328 :Parameters: struct kvm_dirty_log (in/out)
348 :Returns: 0 on success, -1 on error 329 :Returns: 0 on success, -1 on error
349 330
350 :: 331 ::
351 332
352 /* for KVM_GET_DIRTY_LOG */ 333 /* for KVM_GET_DIRTY_LOG */
353 struct kvm_dirty_log { 334 struct kvm_dirty_log {
354 __u32 slot; 335 __u32 slot;
355 __u32 padding; 336 __u32 padding;
356 union { 337 union {
357 void __user *dirty_bitmap; /* 338 void __user *dirty_bitmap; /* one bit per page */
358 __u64 padding; 339 __u64 padding;
359 }; 340 };
360 }; 341 };
361 342
362 Given a memory slot, return a bitmap containin 343 Given a memory slot, return a bitmap containing any pages dirtied
363 since the last call to this ioctl. Bit 0 is t 344 since the last call to this ioctl. Bit 0 is the first page in the
364 memory slot. Ensure the entire structure is c 345 memory slot. Ensure the entire structure is cleared to avoid padding
365 issues. 346 issues.
366 347
367 If KVM_CAP_MULTI_ADDRESS_SPACE is available, b 348 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies
368 the address space for which you want to return 349 the address space for which you want to return the dirty bitmap. See
369 KVM_SET_USER_MEMORY_REGION for details on the 350 KVM_SET_USER_MEMORY_REGION for details on the usage of slot field.
370 351
371 The bits in the dirty bitmap are cleared befor 352 The bits in the dirty bitmap are cleared before the ioctl returns, unless
372 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. 353 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is enabled. For more information,
373 see the description of the capability. 354 see the description of the capability.
374 355
375 Note that the Xen shared_info page, if configu !! 356 Note that the Xen shared info page, if configured, shall always be assumed
376 to be dirty. KVM will not explicitly mark it s 357 to be dirty. KVM will not explicitly mark it such.
377 358
378 359
379 4.10 KVM_RUN 360 4.10 KVM_RUN
380 ------------ 361 ------------
381 362
382 :Capability: basic 363 :Capability: basic
383 :Architectures: all 364 :Architectures: all
384 :Type: vcpu ioctl 365 :Type: vcpu ioctl
385 :Parameters: none 366 :Parameters: none
386 :Returns: 0 on success, -1 on error 367 :Returns: 0 on success, -1 on error
387 368
388 Errors: 369 Errors:
389 370
390 ======= ================================= 371 ======= ==============================================================
391 EINTR an unmasked signal is pending 372 EINTR an unmasked signal is pending
392 ENOEXEC the vcpu hasn't been initialized 373 ENOEXEC the vcpu hasn't been initialized or the guest tried to execute
393 instructions from device memory ( 374 instructions from device memory (arm64)
394 ENOSYS data abort outside memslots with 375 ENOSYS data abort outside memslots with no syndrome info and
395 KVM_CAP_ARM_NISV_TO_USER not enab 376 KVM_CAP_ARM_NISV_TO_USER not enabled (arm64)
396 EPERM SVE feature set but not finalized 377 EPERM SVE feature set but not finalized (arm64)
397 ======= ================================= 378 ======= ==============================================================
398 379
399 This ioctl is used to run a guest virtual cpu. 380 This ioctl is used to run a guest virtual cpu. While there are no
400 explicit parameters, there is an implicit para 381 explicit parameters, there is an implicit parameter block that can be
401 obtained by mmap()ing the vcpu fd at offset 0, 382 obtained by mmap()ing the vcpu fd at offset 0, with the size given by
402 KVM_GET_VCPU_MMAP_SIZE. The parameter block i 383 KVM_GET_VCPU_MMAP_SIZE. The parameter block is formatted as a 'struct
403 kvm_run' (see below). 384 kvm_run' (see below).
404 385
405 386
406 4.11 KVM_GET_REGS 387 4.11 KVM_GET_REGS
407 ----------------- 388 -----------------
408 389
409 :Capability: basic 390 :Capability: basic
410 :Architectures: all except arm64 391 :Architectures: all except arm64
411 :Type: vcpu ioctl 392 :Type: vcpu ioctl
412 :Parameters: struct kvm_regs (out) 393 :Parameters: struct kvm_regs (out)
413 :Returns: 0 on success, -1 on error 394 :Returns: 0 on success, -1 on error
414 395
415 Reads the general purpose registers from the v 396 Reads the general purpose registers from the vcpu.
416 397
417 :: 398 ::
418 399
419 /* x86 */ 400 /* x86 */
420 struct kvm_regs { 401 struct kvm_regs {
421 /* out (KVM_GET_REGS) / in (KVM_SET_RE 402 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
422 __u64 rax, rbx, rcx, rdx; 403 __u64 rax, rbx, rcx, rdx;
423 __u64 rsi, rdi, rsp, rbp; 404 __u64 rsi, rdi, rsp, rbp;
424 __u64 r8, r9, r10, r11; 405 __u64 r8, r9, r10, r11;
425 __u64 r12, r13, r14, r15; 406 __u64 r12, r13, r14, r15;
426 __u64 rip, rflags; 407 __u64 rip, rflags;
427 }; 408 };
428 409
429 /* mips */ 410 /* mips */
430 struct kvm_regs { 411 struct kvm_regs {
431 /* out (KVM_GET_REGS) / in (KVM_SET_RE 412 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
432 __u64 gpr[32]; 413 __u64 gpr[32];
433 __u64 hi; 414 __u64 hi;
434 __u64 lo; 415 __u64 lo;
435 __u64 pc; 416 __u64 pc;
436 }; 417 };
437 418
438 /* LoongArch */ 419 /* LoongArch */
439 struct kvm_regs { 420 struct kvm_regs {
440 /* out (KVM_GET_REGS) / in (KVM_SET_RE 421 /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */
441 unsigned long gpr[32]; 422 unsigned long gpr[32];
442 unsigned long pc; 423 unsigned long pc;
443 }; 424 };
444 425
445 426
446 4.12 KVM_SET_REGS 427 4.12 KVM_SET_REGS
447 ----------------- 428 -----------------
448 429
449 :Capability: basic 430 :Capability: basic
450 :Architectures: all except arm64 431 :Architectures: all except arm64
451 :Type: vcpu ioctl 432 :Type: vcpu ioctl
452 :Parameters: struct kvm_regs (in) 433 :Parameters: struct kvm_regs (in)
453 :Returns: 0 on success, -1 on error 434 :Returns: 0 on success, -1 on error
454 435
455 Writes the general purpose registers into the 436 Writes the general purpose registers into the vcpu.
456 437
457 See KVM_GET_REGS for the data structure. 438 See KVM_GET_REGS for the data structure.
458 439
459 440
460 4.13 KVM_GET_SREGS 441 4.13 KVM_GET_SREGS
461 ------------------ 442 ------------------
462 443
463 :Capability: basic 444 :Capability: basic
464 :Architectures: x86, ppc 445 :Architectures: x86, ppc
465 :Type: vcpu ioctl 446 :Type: vcpu ioctl
466 :Parameters: struct kvm_sregs (out) 447 :Parameters: struct kvm_sregs (out)
467 :Returns: 0 on success, -1 on error 448 :Returns: 0 on success, -1 on error
468 449
469 Reads special registers from the vcpu. 450 Reads special registers from the vcpu.
470 451
471 :: 452 ::
472 453
473 /* x86 */ 454 /* x86 */
474 struct kvm_sregs { 455 struct kvm_sregs {
475 struct kvm_segment cs, ds, es, fs, gs, 456 struct kvm_segment cs, ds, es, fs, gs, ss;
476 struct kvm_segment tr, ldt; 457 struct kvm_segment tr, ldt;
477 struct kvm_dtable gdt, idt; 458 struct kvm_dtable gdt, idt;
478 __u64 cr0, cr2, cr3, cr4, cr8; 459 __u64 cr0, cr2, cr3, cr4, cr8;
479 __u64 efer; 460 __u64 efer;
480 __u64 apic_base; 461 __u64 apic_base;
481 __u64 interrupt_bitmap[(KVM_NR_INTERRU 462 __u64 interrupt_bitmap[(KVM_NR_INTERRUPTS + 63) / 64];
482 }; 463 };
483 464
484 /* ppc -- see arch/powerpc/include/uapi/asm/ 465 /* ppc -- see arch/powerpc/include/uapi/asm/kvm.h */
485 466
486 interrupt_bitmap is a bitmap of pending extern 467 interrupt_bitmap is a bitmap of pending external interrupts. At most
487 one bit may be set. This interrupt has been a 468 one bit may be set. This interrupt has been acknowledged by the APIC
488 but not yet injected into the cpu core. 469 but not yet injected into the cpu core.
489 470
490 471
491 4.14 KVM_SET_SREGS 472 4.14 KVM_SET_SREGS
492 ------------------ 473 ------------------
493 474
494 :Capability: basic 475 :Capability: basic
495 :Architectures: x86, ppc 476 :Architectures: x86, ppc
496 :Type: vcpu ioctl 477 :Type: vcpu ioctl
497 :Parameters: struct kvm_sregs (in) 478 :Parameters: struct kvm_sregs (in)
498 :Returns: 0 on success, -1 on error 479 :Returns: 0 on success, -1 on error
499 480
500 Writes special registers into the vcpu. See K 481 Writes special registers into the vcpu. See KVM_GET_SREGS for the
501 data structures. 482 data structures.
502 483
503 484
504 4.15 KVM_TRANSLATE 485 4.15 KVM_TRANSLATE
505 ------------------ 486 ------------------
506 487
507 :Capability: basic 488 :Capability: basic
508 :Architectures: x86 489 :Architectures: x86
509 :Type: vcpu ioctl 490 :Type: vcpu ioctl
510 :Parameters: struct kvm_translation (in/out) 491 :Parameters: struct kvm_translation (in/out)
511 :Returns: 0 on success, -1 on error 492 :Returns: 0 on success, -1 on error
512 493
513 Translates a virtual address according to the 494 Translates a virtual address according to the vcpu's current address
514 translation mode. 495 translation mode.
515 496
516 :: 497 ::
517 498
518 struct kvm_translation { 499 struct kvm_translation {
519 /* in */ 500 /* in */
520 __u64 linear_address; 501 __u64 linear_address;
521 502
522 /* out */ 503 /* out */
523 __u64 physical_address; 504 __u64 physical_address;
524 __u8 valid; 505 __u8 valid;
525 __u8 writeable; 506 __u8 writeable;
526 __u8 usermode; 507 __u8 usermode;
527 __u8 pad[5]; 508 __u8 pad[5];
528 }; 509 };
529 510
530 511
531 4.16 KVM_INTERRUPT 512 4.16 KVM_INTERRUPT
532 ------------------ 513 ------------------
533 514
534 :Capability: basic 515 :Capability: basic
535 :Architectures: x86, ppc, mips, riscv, loongar 516 :Architectures: x86, ppc, mips, riscv, loongarch
536 :Type: vcpu ioctl 517 :Type: vcpu ioctl
537 :Parameters: struct kvm_interrupt (in) 518 :Parameters: struct kvm_interrupt (in)
538 :Returns: 0 on success, negative on failure. 519 :Returns: 0 on success, negative on failure.
539 520
540 Queues a hardware interrupt vector to be injec 521 Queues a hardware interrupt vector to be injected.
541 522
542 :: 523 ::
543 524
544 /* for KVM_INTERRUPT */ 525 /* for KVM_INTERRUPT */
545 struct kvm_interrupt { 526 struct kvm_interrupt {
546 /* in */ 527 /* in */
547 __u32 irq; 528 __u32 irq;
548 }; 529 };
549 530
550 X86: 531 X86:
551 ^^^^ 532 ^^^^
552 533
553 :Returns: 534 :Returns:
554 535
555 ========= ============================ 536 ========= ===================================
556 0 on success, 537 0 on success,
557 -EEXIST if an interrupt is already e 538 -EEXIST if an interrupt is already enqueued
558 -EINVAL the irq number is invalid 539 -EINVAL the irq number is invalid
559 -ENXIO if the PIC is in the kernel 540 -ENXIO if the PIC is in the kernel
560 -EFAULT if the pointer is invalid 541 -EFAULT if the pointer is invalid
561 ========= ============================ 542 ========= ===================================
562 543
563 Note 'irq' is an interrupt vector, not an inte 544 Note 'irq' is an interrupt vector, not an interrupt pin or line. This
564 ioctl is useful if the in-kernel PIC is not us 545 ioctl is useful if the in-kernel PIC is not used.
565 546
566 PPC: 547 PPC:
567 ^^^^ 548 ^^^^
568 549
569 Queues an external interrupt to be injected. T 550 Queues an external interrupt to be injected. This ioctl is overloaded
570 with 3 different irq values: 551 with 3 different irq values:
571 552
572 a) KVM_INTERRUPT_SET 553 a) KVM_INTERRUPT_SET
573 554
574 This injects an edge type external interrup 555 This injects an edge type external interrupt into the guest once it's ready
575 to receive interrupts. When injected, the i 556 to receive interrupts. When injected, the interrupt is done.
576 557
577 b) KVM_INTERRUPT_UNSET 558 b) KVM_INTERRUPT_UNSET
578 559
579 This unsets any pending interrupt. 560 This unsets any pending interrupt.
580 561
581 Only available with KVM_CAP_PPC_UNSET_IRQ. 562 Only available with KVM_CAP_PPC_UNSET_IRQ.
582 563
583 c) KVM_INTERRUPT_SET_LEVEL 564 c) KVM_INTERRUPT_SET_LEVEL
584 565
585 This injects a level type external interrup 566 This injects a level type external interrupt into the guest context. The
586 interrupt stays pending until a specific io 567 interrupt stays pending until a specific ioctl with KVM_INTERRUPT_UNSET
587 is triggered. 568 is triggered.
588 569
589 Only available with KVM_CAP_PPC_IRQ_LEVEL. 570 Only available with KVM_CAP_PPC_IRQ_LEVEL.
590 571
591 Note that any value for 'irq' other than the o 572 Note that any value for 'irq' other than the ones stated above is invalid
592 and incurs unexpected behavior. 573 and incurs unexpected behavior.
593 574
594 This is an asynchronous vcpu ioctl and can be 575 This is an asynchronous vcpu ioctl and can be invoked from any thread.
595 576
596 MIPS: 577 MIPS:
597 ^^^^^ 578 ^^^^^
598 579
599 Queues an external interrupt to be injected in 580 Queues an external interrupt to be injected into the virtual CPU. A negative
600 interrupt number dequeues the interrupt. 581 interrupt number dequeues the interrupt.
601 582
602 This is an asynchronous vcpu ioctl and can be 583 This is an asynchronous vcpu ioctl and can be invoked from any thread.
603 584
604 RISC-V: 585 RISC-V:
605 ^^^^^^^ 586 ^^^^^^^
606 587
607 Queues an external interrupt to be injected in 588 Queues an external interrupt to be injected into the virtual CPU. This ioctl
608 is overloaded with 2 different irq values: 589 is overloaded with 2 different irq values:
609 590
610 a) KVM_INTERRUPT_SET 591 a) KVM_INTERRUPT_SET
611 592
612 This sets external interrupt for a virtual 593 This sets external interrupt for a virtual CPU and it will receive
613 once it is ready. 594 once it is ready.
614 595
615 b) KVM_INTERRUPT_UNSET 596 b) KVM_INTERRUPT_UNSET
616 597
617 This clears pending external interrupt for 598 This clears pending external interrupt for a virtual CPU.
618 599
619 This is an asynchronous vcpu ioctl and can be 600 This is an asynchronous vcpu ioctl and can be invoked from any thread.
620 601
621 LOONGARCH: 602 LOONGARCH:
622 ^^^^^^^^^^ 603 ^^^^^^^^^^
623 604
624 Queues an external interrupt to be injected in 605 Queues an external interrupt to be injected into the virtual CPU. A negative
625 interrupt number dequeues the interrupt. 606 interrupt number dequeues the interrupt.
626 607
627 This is an asynchronous vcpu ioctl and can be 608 This is an asynchronous vcpu ioctl and can be invoked from any thread.
628 609
629 610
>> 611 4.17 KVM_DEBUG_GUEST
>> 612 --------------------
>> 613
>> 614 :Capability: basic
>> 615 :Architectures: none
>> 616 :Type: vcpu ioctl
>> 617 :Parameters: none)
>> 618 :Returns: -1 on error
>> 619
>> 620 Support for this has been removed. Use KVM_SET_GUEST_DEBUG instead.
>> 621
>> 622
630 4.18 KVM_GET_MSRS 623 4.18 KVM_GET_MSRS
631 ----------------- 624 -----------------
632 625
633 :Capability: basic (vcpu), KVM_CAP_GET_MSR_FEA 626 :Capability: basic (vcpu), KVM_CAP_GET_MSR_FEATURES (system)
634 :Architectures: x86 627 :Architectures: x86
635 :Type: system ioctl, vcpu ioctl 628 :Type: system ioctl, vcpu ioctl
636 :Parameters: struct kvm_msrs (in/out) 629 :Parameters: struct kvm_msrs (in/out)
637 :Returns: number of msrs successfully returned 630 :Returns: number of msrs successfully returned;
638 -1 on error 631 -1 on error
639 632
640 When used as a system ioctl: 633 When used as a system ioctl:
641 Reads the values of MSR-based features that ar 634 Reads the values of MSR-based features that are available for the VM. This
642 is similar to KVM_GET_SUPPORTED_CPUID, but it 635 is similar to KVM_GET_SUPPORTED_CPUID, but it returns MSR indices and values.
643 The list of msr-based features can be obtained 636 The list of msr-based features can be obtained using KVM_GET_MSR_FEATURE_INDEX_LIST
644 in a system ioctl. 637 in a system ioctl.
645 638
646 When used as a vcpu ioctl: 639 When used as a vcpu ioctl:
647 Reads model-specific registers from the vcpu. 640 Reads model-specific registers from the vcpu. Supported msr indices can
648 be obtained using KVM_GET_MSR_INDEX_LIST in a 641 be obtained using KVM_GET_MSR_INDEX_LIST in a system ioctl.
649 642
650 :: 643 ::
651 644
652 struct kvm_msrs { 645 struct kvm_msrs {
653 __u32 nmsrs; /* number of msrs in entr 646 __u32 nmsrs; /* number of msrs in entries */
654 __u32 pad; 647 __u32 pad;
655 648
656 struct kvm_msr_entry entries[0]; 649 struct kvm_msr_entry entries[0];
657 }; 650 };
658 651
659 struct kvm_msr_entry { 652 struct kvm_msr_entry {
660 __u32 index; 653 __u32 index;
661 __u32 reserved; 654 __u32 reserved;
662 __u64 data; 655 __u64 data;
663 }; 656 };
664 657
665 Application code should set the 'nmsrs' member 658 Application code should set the 'nmsrs' member (which indicates the
666 size of the entries array) and the 'index' mem 659 size of the entries array) and the 'index' member of each array entry.
667 kvm will fill in the 'data' member. 660 kvm will fill in the 'data' member.
668 661
669 662
670 4.19 KVM_SET_MSRS 663 4.19 KVM_SET_MSRS
671 ----------------- 664 -----------------
672 665
673 :Capability: basic 666 :Capability: basic
674 :Architectures: x86 667 :Architectures: x86
675 :Type: vcpu ioctl 668 :Type: vcpu ioctl
676 :Parameters: struct kvm_msrs (in) 669 :Parameters: struct kvm_msrs (in)
677 :Returns: number of msrs successfully set (see 670 :Returns: number of msrs successfully set (see below), -1 on error
678 671
679 Writes model-specific registers to the vcpu. 672 Writes model-specific registers to the vcpu. See KVM_GET_MSRS for the
680 data structures. 673 data structures.
681 674
682 Application code should set the 'nmsrs' member 675 Application code should set the 'nmsrs' member (which indicates the
683 size of the entries array), and the 'index' an 676 size of the entries array), and the 'index' and 'data' members of each
684 array entry. 677 array entry.
685 678
686 It tries to set the MSRs in array entries[] on 679 It tries to set the MSRs in array entries[] one by one. If setting an MSR
687 fails, e.g., due to setting reserved bits, the 680 fails, e.g., due to setting reserved bits, the MSR isn't supported/emulated
688 by KVM, etc..., it stops processing the MSR li 681 by KVM, etc..., it stops processing the MSR list and returns the number of
689 MSRs that have been set successfully. 682 MSRs that have been set successfully.
690 683
691 684
692 4.20 KVM_SET_CPUID 685 4.20 KVM_SET_CPUID
693 ------------------ 686 ------------------
694 687
695 :Capability: basic 688 :Capability: basic
696 :Architectures: x86 689 :Architectures: x86
697 :Type: vcpu ioctl 690 :Type: vcpu ioctl
698 :Parameters: struct kvm_cpuid (in) 691 :Parameters: struct kvm_cpuid (in)
699 :Returns: 0 on success, -1 on error 692 :Returns: 0 on success, -1 on error
700 693
701 Defines the vcpu responses to the cpuid instru 694 Defines the vcpu responses to the cpuid instruction. Applications
702 should use the KVM_SET_CPUID2 ioctl if availab 695 should use the KVM_SET_CPUID2 ioctl if available.
703 696
704 Caveat emptor: 697 Caveat emptor:
705 - If this IOCTL fails, KVM gives no guarante 698 - If this IOCTL fails, KVM gives no guarantees that previous valid CPUID
706 configuration (if there is) is not corrupt 699 configuration (if there is) is not corrupted. Userspace can get a copy
707 of the resulting CPUID configuration throu 700 of the resulting CPUID configuration through KVM_GET_CPUID2 in case.
708 - Using KVM_SET_CPUID{,2} after KVM_RUN, i.e 701 - Using KVM_SET_CPUID{,2} after KVM_RUN, i.e. changing the guest vCPU model
709 after running the guest, may cause guest i 702 after running the guest, may cause guest instability.
710 - Using heterogeneous CPUID configurations, 703 - Using heterogeneous CPUID configurations, modulo APIC IDs, topology, etc...
711 may cause guest instability. 704 may cause guest instability.
712 705
713 :: 706 ::
714 707
715 struct kvm_cpuid_entry { 708 struct kvm_cpuid_entry {
716 __u32 function; 709 __u32 function;
717 __u32 eax; 710 __u32 eax;
718 __u32 ebx; 711 __u32 ebx;
719 __u32 ecx; 712 __u32 ecx;
720 __u32 edx; 713 __u32 edx;
721 __u32 padding; 714 __u32 padding;
722 }; 715 };
723 716
724 /* for KVM_SET_CPUID */ 717 /* for KVM_SET_CPUID */
725 struct kvm_cpuid { 718 struct kvm_cpuid {
726 __u32 nent; 719 __u32 nent;
727 __u32 padding; 720 __u32 padding;
728 struct kvm_cpuid_entry entries[0]; 721 struct kvm_cpuid_entry entries[0];
729 }; 722 };
730 723
731 724
732 4.21 KVM_SET_SIGNAL_MASK 725 4.21 KVM_SET_SIGNAL_MASK
733 ------------------------ 726 ------------------------
734 727
735 :Capability: basic 728 :Capability: basic
736 :Architectures: all 729 :Architectures: all
737 :Type: vcpu ioctl 730 :Type: vcpu ioctl
738 :Parameters: struct kvm_signal_mask (in) 731 :Parameters: struct kvm_signal_mask (in)
739 :Returns: 0 on success, -1 on error 732 :Returns: 0 on success, -1 on error
740 733
741 Defines which signals are blocked during execu 734 Defines which signals are blocked during execution of KVM_RUN. This
742 signal mask temporarily overrides the threads 735 signal mask temporarily overrides the threads signal mask. Any
743 unblocked signal received (except SIGKILL and 736 unblocked signal received (except SIGKILL and SIGSTOP, which retain
744 their traditional behaviour) will cause KVM_RU 737 their traditional behaviour) will cause KVM_RUN to return with -EINTR.
745 738
746 Note the signal will only be delivered if not 739 Note the signal will only be delivered if not blocked by the original
747 signal mask. 740 signal mask.
748 741
749 :: 742 ::
750 743
751 /* for KVM_SET_SIGNAL_MASK */ 744 /* for KVM_SET_SIGNAL_MASK */
752 struct kvm_signal_mask { 745 struct kvm_signal_mask {
753 __u32 len; 746 __u32 len;
754 __u8 sigset[0]; 747 __u8 sigset[0];
755 }; 748 };
756 749
757 750
758 4.22 KVM_GET_FPU 751 4.22 KVM_GET_FPU
759 ---------------- 752 ----------------
760 753
761 :Capability: basic 754 :Capability: basic
762 :Architectures: x86, loongarch 755 :Architectures: x86, loongarch
763 :Type: vcpu ioctl 756 :Type: vcpu ioctl
764 :Parameters: struct kvm_fpu (out) 757 :Parameters: struct kvm_fpu (out)
765 :Returns: 0 on success, -1 on error 758 :Returns: 0 on success, -1 on error
766 759
767 Reads the floating point state from the vcpu. 760 Reads the floating point state from the vcpu.
768 761
769 :: 762 ::
770 763
771 /* x86: for KVM_GET_FPU and KVM_SET_FPU */ 764 /* x86: for KVM_GET_FPU and KVM_SET_FPU */
772 struct kvm_fpu { 765 struct kvm_fpu {
773 __u8 fpr[8][16]; 766 __u8 fpr[8][16];
774 __u16 fcw; 767 __u16 fcw;
775 __u16 fsw; 768 __u16 fsw;
776 __u8 ftwx; /* in fxsave format */ 769 __u8 ftwx; /* in fxsave format */
777 __u8 pad1; 770 __u8 pad1;
778 __u16 last_opcode; 771 __u16 last_opcode;
779 __u64 last_ip; 772 __u64 last_ip;
780 __u64 last_dp; 773 __u64 last_dp;
781 __u8 xmm[16][16]; 774 __u8 xmm[16][16];
782 __u32 mxcsr; 775 __u32 mxcsr;
783 __u32 pad2; 776 __u32 pad2;
784 }; 777 };
785 778
786 /* LoongArch: for KVM_GET_FPU and KVM_SET_FP 779 /* LoongArch: for KVM_GET_FPU and KVM_SET_FPU */
787 struct kvm_fpu { 780 struct kvm_fpu {
788 __u32 fcsr; 781 __u32 fcsr;
789 __u64 fcc; 782 __u64 fcc;
790 struct kvm_fpureg { 783 struct kvm_fpureg {
791 __u64 val64[4]; 784 __u64 val64[4];
792 }fpr[32]; 785 }fpr[32];
793 }; 786 };
794 787
795 788
796 4.23 KVM_SET_FPU 789 4.23 KVM_SET_FPU
797 ---------------- 790 ----------------
798 791
799 :Capability: basic 792 :Capability: basic
800 :Architectures: x86, loongarch 793 :Architectures: x86, loongarch
801 :Type: vcpu ioctl 794 :Type: vcpu ioctl
802 :Parameters: struct kvm_fpu (in) 795 :Parameters: struct kvm_fpu (in)
803 :Returns: 0 on success, -1 on error 796 :Returns: 0 on success, -1 on error
804 797
805 Writes the floating point state to the vcpu. 798 Writes the floating point state to the vcpu.
806 799
807 :: 800 ::
808 801
809 /* x86: for KVM_GET_FPU and KVM_SET_FPU */ 802 /* x86: for KVM_GET_FPU and KVM_SET_FPU */
810 struct kvm_fpu { 803 struct kvm_fpu {
811 __u8 fpr[8][16]; 804 __u8 fpr[8][16];
812 __u16 fcw; 805 __u16 fcw;
813 __u16 fsw; 806 __u16 fsw;
814 __u8 ftwx; /* in fxsave format */ 807 __u8 ftwx; /* in fxsave format */
815 __u8 pad1; 808 __u8 pad1;
816 __u16 last_opcode; 809 __u16 last_opcode;
817 __u64 last_ip; 810 __u64 last_ip;
818 __u64 last_dp; 811 __u64 last_dp;
819 __u8 xmm[16][16]; 812 __u8 xmm[16][16];
820 __u32 mxcsr; 813 __u32 mxcsr;
821 __u32 pad2; 814 __u32 pad2;
822 }; 815 };
823 816
824 /* LoongArch: for KVM_GET_FPU and KVM_SET_FP 817 /* LoongArch: for KVM_GET_FPU and KVM_SET_FPU */
825 struct kvm_fpu { 818 struct kvm_fpu {
826 __u32 fcsr; 819 __u32 fcsr;
827 __u64 fcc; 820 __u64 fcc;
828 struct kvm_fpureg { 821 struct kvm_fpureg {
829 __u64 val64[4]; 822 __u64 val64[4];
830 }fpr[32]; 823 }fpr[32];
831 }; 824 };
832 825
833 826
834 4.24 KVM_CREATE_IRQCHIP 827 4.24 KVM_CREATE_IRQCHIP
835 ----------------------- 828 -----------------------
836 829
837 :Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQ 830 :Capability: KVM_CAP_IRQCHIP, KVM_CAP_S390_IRQCHIP (s390)
838 :Architectures: x86, arm64, s390 831 :Architectures: x86, arm64, s390
839 :Type: vm ioctl 832 :Type: vm ioctl
840 :Parameters: none 833 :Parameters: none
841 :Returns: 0 on success, -1 on error 834 :Returns: 0 on success, -1 on error
842 835
843 Creates an interrupt controller model in the k 836 Creates an interrupt controller model in the kernel.
844 On x86, creates a virtual ioapic, a virtual PI 837 On x86, creates a virtual ioapic, a virtual PIC (two PICs, nested), and sets up
845 future vcpus to have a local APIC. IRQ routin 838 future vcpus to have a local APIC. IRQ routing for GSIs 0-15 is set to both
846 PIC and IOAPIC; GSI 16-23 only go to the IOAPI 839 PIC and IOAPIC; GSI 16-23 only go to the IOAPIC.
847 On arm64, a GICv2 is created. Any other GIC ve 840 On arm64, a GICv2 is created. Any other GIC versions require the usage of
848 KVM_CREATE_DEVICE, which also supports creatin 841 KVM_CREATE_DEVICE, which also supports creating a GICv2. Using
849 KVM_CREATE_DEVICE is preferred over KVM_CREATE 842 KVM_CREATE_DEVICE is preferred over KVM_CREATE_IRQCHIP for GICv2.
850 On s390, a dummy irq routing table is created. 843 On s390, a dummy irq routing table is created.
851 844
852 Note that on s390 the KVM_CAP_S390_IRQCHIP vm 845 Note that on s390 the KVM_CAP_S390_IRQCHIP vm capability needs to be enabled
853 before KVM_CREATE_IRQCHIP can be used. 846 before KVM_CREATE_IRQCHIP can be used.
854 847
855 848
856 4.25 KVM_IRQ_LINE 849 4.25 KVM_IRQ_LINE
857 ----------------- 850 -----------------
858 851
859 :Capability: KVM_CAP_IRQCHIP 852 :Capability: KVM_CAP_IRQCHIP
860 :Architectures: x86, arm64 853 :Architectures: x86, arm64
861 :Type: vm ioctl 854 :Type: vm ioctl
862 :Parameters: struct kvm_irq_level 855 :Parameters: struct kvm_irq_level
863 :Returns: 0 on success, -1 on error 856 :Returns: 0 on success, -1 on error
864 857
865 Sets the level of a GSI input to the interrupt 858 Sets the level of a GSI input to the interrupt controller model in the kernel.
866 On some architectures it is required that an i 859 On some architectures it is required that an interrupt controller model has
867 been previously created with KVM_CREATE_IRQCHI 860 been previously created with KVM_CREATE_IRQCHIP. Note that edge-triggered
868 interrupts require the level to be set to 1 an 861 interrupts require the level to be set to 1 and then back to 0.
869 862
870 On real hardware, interrupt pins can be active 863 On real hardware, interrupt pins can be active-low or active-high. This
871 does not matter for the level field of struct 864 does not matter for the level field of struct kvm_irq_level: 1 always
872 means active (asserted), 0 means inactive (dea 865 means active (asserted), 0 means inactive (deasserted).
873 866
874 x86 allows the operating system to program the 867 x86 allows the operating system to program the interrupt polarity
875 (active-low/active-high) for level-triggered i 868 (active-low/active-high) for level-triggered interrupts, and KVM used
876 to consider the polarity. However, due to bit 869 to consider the polarity. However, due to bitrot in the handling of
877 active-low interrupts, the above convention is 870 active-low interrupts, the above convention is now valid on x86 too.
878 This is signaled by KVM_CAP_X86_IOAPIC_POLARIT 871 This is signaled by KVM_CAP_X86_IOAPIC_POLARITY_IGNORED. Userspace
879 should not present interrupts to the guest as 872 should not present interrupts to the guest as active-low unless this
880 capability is present (or unless it is not usi 873 capability is present (or unless it is not using the in-kernel irqchip,
881 of course). 874 of course).
882 875
883 876
884 arm64 can signal an interrupt either at the CP 877 arm64 can signal an interrupt either at the CPU level, or at the
885 in-kernel irqchip (GIC), and for in-kernel irq 878 in-kernel irqchip (GIC), and for in-kernel irqchip can tell the GIC to
886 use PPIs designated for specific cpus. The ir 879 use PPIs designated for specific cpus. The irq field is interpreted
887 like this:: 880 like this::
888 881
889 bits: | 31 ... 28 | 27 ... 24 | 23 ... 1 882 bits: | 31 ... 28 | 27 ... 24 | 23 ... 16 | 15 ... 0 |
890 field: | vcpu2_index | irq_type | vcpu_inde 883 field: | vcpu2_index | irq_type | vcpu_index | irq_id |
891 884
892 The irq_type field has the following values: 885 The irq_type field has the following values:
893 886
894 - KVM_ARM_IRQ_TYPE_CPU: !! 887 - irq_type[0]:
895 out-of-kernel GIC: irq_id 0 is 888 out-of-kernel GIC: irq_id 0 is IRQ, irq_id 1 is FIQ
896 - KVM_ARM_IRQ_TYPE_SPI: !! 889 - irq_type[1]:
897 in-kernel GIC: SPI, irq_id betw 890 in-kernel GIC: SPI, irq_id between 32 and 1019 (incl.)
898 (the vcpu_index field is ignore 891 (the vcpu_index field is ignored)
899 - KVM_ARM_IRQ_TYPE_PPI: !! 892 - irq_type[2]:
900 in-kernel GIC: PPI, irq_id betw 893 in-kernel GIC: PPI, irq_id between 16 and 31 (incl.)
901 894
902 (The irq_id field thus corresponds nicely to t 895 (The irq_id field thus corresponds nicely to the IRQ ID in the ARM GIC specs)
903 896
904 In both cases, level is used to assert/deasser 897 In both cases, level is used to assert/deassert the line.
905 898
906 When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supporte 899 When KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 is supported, the target vcpu is
907 identified as (256 * vcpu2_index + vcpu_index) 900 identified as (256 * vcpu2_index + vcpu_index). Otherwise, vcpu2_index
908 must be zero. 901 must be zero.
909 902
910 Note that on arm64, the KVM_CAP_IRQCHIP capabi 903 Note that on arm64, the KVM_CAP_IRQCHIP capability only conditions
911 injection of interrupts for the in-kernel irqc 904 injection of interrupts for the in-kernel irqchip. KVM_IRQ_LINE can always
912 be used for a userspace interrupt controller. 905 be used for a userspace interrupt controller.
913 906
914 :: 907 ::
915 908
916 struct kvm_irq_level { 909 struct kvm_irq_level {
917 union { 910 union {
918 __u32 irq; /* GSI */ 911 __u32 irq; /* GSI */
919 __s32 status; /* not used for 912 __s32 status; /* not used for KVM_IRQ_LEVEL */
920 }; 913 };
921 __u32 level; /* 0 or 1 */ 914 __u32 level; /* 0 or 1 */
922 }; 915 };
923 916
924 917
925 4.26 KVM_GET_IRQCHIP 918 4.26 KVM_GET_IRQCHIP
926 -------------------- 919 --------------------
927 920
928 :Capability: KVM_CAP_IRQCHIP 921 :Capability: KVM_CAP_IRQCHIP
929 :Architectures: x86 922 :Architectures: x86
930 :Type: vm ioctl 923 :Type: vm ioctl
931 :Parameters: struct kvm_irqchip (in/out) 924 :Parameters: struct kvm_irqchip (in/out)
932 :Returns: 0 on success, -1 on error 925 :Returns: 0 on success, -1 on error
933 926
934 Reads the state of a kernel interrupt controll 927 Reads the state of a kernel interrupt controller created with
935 KVM_CREATE_IRQCHIP into a buffer provided by t 928 KVM_CREATE_IRQCHIP into a buffer provided by the caller.
936 929
937 :: 930 ::
938 931
939 struct kvm_irqchip { 932 struct kvm_irqchip {
940 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 933 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
941 __u32 pad; 934 __u32 pad;
942 union { 935 union {
943 char dummy[512]; /* reserving 936 char dummy[512]; /* reserving space */
944 struct kvm_pic_state pic; 937 struct kvm_pic_state pic;
945 struct kvm_ioapic_state ioapic 938 struct kvm_ioapic_state ioapic;
946 } chip; 939 } chip;
947 }; 940 };
948 941
949 942
950 4.27 KVM_SET_IRQCHIP 943 4.27 KVM_SET_IRQCHIP
951 -------------------- 944 --------------------
952 945
953 :Capability: KVM_CAP_IRQCHIP 946 :Capability: KVM_CAP_IRQCHIP
954 :Architectures: x86 947 :Architectures: x86
955 :Type: vm ioctl 948 :Type: vm ioctl
956 :Parameters: struct kvm_irqchip (in) 949 :Parameters: struct kvm_irqchip (in)
957 :Returns: 0 on success, -1 on error 950 :Returns: 0 on success, -1 on error
958 951
959 Sets the state of a kernel interrupt controlle 952 Sets the state of a kernel interrupt controller created with
960 KVM_CREATE_IRQCHIP from a buffer provided by t 953 KVM_CREATE_IRQCHIP from a buffer provided by the caller.
961 954
962 :: 955 ::
963 956
964 struct kvm_irqchip { 957 struct kvm_irqchip {
965 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 958 __u32 chip_id; /* 0 = PIC1, 1 = PIC2, 2 = IOAPIC */
966 __u32 pad; 959 __u32 pad;
967 union { 960 union {
968 char dummy[512]; /* reserving 961 char dummy[512]; /* reserving space */
969 struct kvm_pic_state pic; 962 struct kvm_pic_state pic;
970 struct kvm_ioapic_state ioapic 963 struct kvm_ioapic_state ioapic;
971 } chip; 964 } chip;
972 }; 965 };
973 966
974 967
975 4.28 KVM_XEN_HVM_CONFIG 968 4.28 KVM_XEN_HVM_CONFIG
976 ----------------------- 969 -----------------------
977 970
978 :Capability: KVM_CAP_XEN_HVM 971 :Capability: KVM_CAP_XEN_HVM
979 :Architectures: x86 972 :Architectures: x86
980 :Type: vm ioctl 973 :Type: vm ioctl
981 :Parameters: struct kvm_xen_hvm_config (in) 974 :Parameters: struct kvm_xen_hvm_config (in)
982 :Returns: 0 on success, -1 on error 975 :Returns: 0 on success, -1 on error
983 976
984 Sets the MSR that the Xen HVM guest uses to in 977 Sets the MSR that the Xen HVM guest uses to initialize its hypercall
985 page, and provides the starting address and si 978 page, and provides the starting address and size of the hypercall
986 blobs in userspace. When the guest writes the 979 blobs in userspace. When the guest writes the MSR, kvm copies one
987 page of a blob (32- or 64-bit, depending on th 980 page of a blob (32- or 64-bit, depending on the vcpu mode) to guest
988 memory. 981 memory.
989 982
990 :: 983 ::
991 984
992 struct kvm_xen_hvm_config { 985 struct kvm_xen_hvm_config {
993 __u32 flags; 986 __u32 flags;
994 __u32 msr; 987 __u32 msr;
995 __u64 blob_addr_32; 988 __u64 blob_addr_32;
996 __u64 blob_addr_64; 989 __u64 blob_addr_64;
997 __u8 blob_size_32; 990 __u8 blob_size_32;
998 __u8 blob_size_64; 991 __u8 blob_size_64;
999 __u8 pad2[30]; 992 __u8 pad2[30];
1000 }; 993 };
1001 994
1002 If certain flags are returned from the KVM_CA 995 If certain flags are returned from the KVM_CAP_XEN_HVM check, they may
1003 be set in the flags field of this ioctl: 996 be set in the flags field of this ioctl:
1004 997
1005 The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag r 998 The KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL flag requests KVM to generate
1006 the contents of the hypercall page automatica 999 the contents of the hypercall page automatically; hypercalls will be
1007 intercepted and passed to userspace through K 1000 intercepted and passed to userspace through KVM_EXIT_XEN. In this
1008 case, all of the blob size and address fields 1001 case, all of the blob size and address fields must be zero.
1009 1002
1010 The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indic 1003 The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates to KVM that userspace
1011 will always use the KVM_XEN_HVM_EVTCHN_SEND i 1004 will always use the KVM_XEN_HVM_EVTCHN_SEND ioctl to deliver event
1012 channel interrupts rather than manipulating t 1005 channel interrupts rather than manipulating the guest's shared_info
1013 structures directly. This, in turn, may allow 1006 structures directly. This, in turn, may allow KVM to enable features
1014 such as intercepting the SCHEDOP_poll hyperca 1007 such as intercepting the SCHEDOP_poll hypercall to accelerate PV
1015 spinlock operation for the guest. Userspace m 1008 spinlock operation for the guest. Userspace may still use the ioctl
1016 to deliver events if it was advertised, even 1009 to deliver events if it was advertised, even if userspace does not
1017 send this indication that it will always do s 1010 send this indication that it will always do so
1018 1011
1019 No other flags are currently valid in the str 1012 No other flags are currently valid in the struct kvm_xen_hvm_config.
1020 1013
1021 4.29 KVM_GET_CLOCK 1014 4.29 KVM_GET_CLOCK
1022 ------------------ 1015 ------------------
1023 1016
1024 :Capability: KVM_CAP_ADJUST_CLOCK 1017 :Capability: KVM_CAP_ADJUST_CLOCK
1025 :Architectures: x86 1018 :Architectures: x86
1026 :Type: vm ioctl 1019 :Type: vm ioctl
1027 :Parameters: struct kvm_clock_data (out) 1020 :Parameters: struct kvm_clock_data (out)
1028 :Returns: 0 on success, -1 on error 1021 :Returns: 0 on success, -1 on error
1029 1022
1030 Gets the current timestamp of kvmclock as see 1023 Gets the current timestamp of kvmclock as seen by the current guest. In
1031 conjunction with KVM_SET_CLOCK, it is used to 1024 conjunction with KVM_SET_CLOCK, it is used to ensure monotonicity on scenarios
1032 such as migration. 1025 such as migration.
1033 1026
1034 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CH 1027 When KVM_CAP_ADJUST_CLOCK is passed to KVM_CHECK_EXTENSION, it returns the
1035 set of bits that KVM can return in struct kvm 1028 set of bits that KVM can return in struct kvm_clock_data's flag member.
1036 1029
1037 The following flags are defined: 1030 The following flags are defined:
1038 1031
1039 KVM_CLOCK_TSC_STABLE 1032 KVM_CLOCK_TSC_STABLE
1040 If set, the returned value is the exact kvm 1033 If set, the returned value is the exact kvmclock
1041 value seen by all VCPUs at the instant when 1034 value seen by all VCPUs at the instant when KVM_GET_CLOCK was called.
1042 If clear, the returned value is simply CLOC 1035 If clear, the returned value is simply CLOCK_MONOTONIC plus a constant
1043 offset; the offset can be modified with KVM 1036 offset; the offset can be modified with KVM_SET_CLOCK. KVM will try
1044 to make all VCPUs follow this clock, but th 1037 to make all VCPUs follow this clock, but the exact value read by each
1045 VCPU could differ, because the host TSC is 1038 VCPU could differ, because the host TSC is not stable.
1046 1039
1047 KVM_CLOCK_REALTIME 1040 KVM_CLOCK_REALTIME
1048 If set, the `realtime` field in the kvm_clo 1041 If set, the `realtime` field in the kvm_clock_data
1049 structure is populated with the value of th 1042 structure is populated with the value of the host's real time
1050 clocksource at the instant when KVM_GET_CLO 1043 clocksource at the instant when KVM_GET_CLOCK was called. If clear,
1051 the `realtime` field does not contain a val 1044 the `realtime` field does not contain a value.
1052 1045
1053 KVM_CLOCK_HOST_TSC 1046 KVM_CLOCK_HOST_TSC
1054 If set, the `host_tsc` field in the kvm_clo 1047 If set, the `host_tsc` field in the kvm_clock_data
1055 structure is populated with the value of th 1048 structure is populated with the value of the host's timestamp counter (TSC)
1056 at the instant when KVM_GET_CLOCK was calle 1049 at the instant when KVM_GET_CLOCK was called. If clear, the `host_tsc` field
1057 does not contain a value. 1050 does not contain a value.
1058 1051
1059 :: 1052 ::
1060 1053
1061 struct kvm_clock_data { 1054 struct kvm_clock_data {
1062 __u64 clock; /* kvmclock current val 1055 __u64 clock; /* kvmclock current value */
1063 __u32 flags; 1056 __u32 flags;
1064 __u32 pad0; 1057 __u32 pad0;
1065 __u64 realtime; 1058 __u64 realtime;
1066 __u64 host_tsc; 1059 __u64 host_tsc;
1067 __u32 pad[4]; 1060 __u32 pad[4];
1068 }; 1061 };
1069 1062
1070 1063
1071 4.30 KVM_SET_CLOCK 1064 4.30 KVM_SET_CLOCK
1072 ------------------ 1065 ------------------
1073 1066
1074 :Capability: KVM_CAP_ADJUST_CLOCK 1067 :Capability: KVM_CAP_ADJUST_CLOCK
1075 :Architectures: x86 1068 :Architectures: x86
1076 :Type: vm ioctl 1069 :Type: vm ioctl
1077 :Parameters: struct kvm_clock_data (in) 1070 :Parameters: struct kvm_clock_data (in)
1078 :Returns: 0 on success, -1 on error 1071 :Returns: 0 on success, -1 on error
1079 1072
1080 Sets the current timestamp of kvmclock to the 1073 Sets the current timestamp of kvmclock to the value specified in its parameter.
1081 In conjunction with KVM_GET_CLOCK, it is used 1074 In conjunction with KVM_GET_CLOCK, it is used to ensure monotonicity on scenarios
1082 such as migration. 1075 such as migration.
1083 1076
1084 The following flags can be passed: 1077 The following flags can be passed:
1085 1078
1086 KVM_CLOCK_REALTIME 1079 KVM_CLOCK_REALTIME
1087 If set, KVM will compare the value of the ` 1080 If set, KVM will compare the value of the `realtime` field
1088 with the value of the host's real time cloc 1081 with the value of the host's real time clocksource at the instant when
1089 KVM_SET_CLOCK was called. The difference in 1082 KVM_SET_CLOCK was called. The difference in elapsed time is added to the final
1090 kvmclock value that will be provided to gue 1083 kvmclock value that will be provided to guests.
1091 1084
1092 Other flags returned by ``KVM_GET_CLOCK`` are 1085 Other flags returned by ``KVM_GET_CLOCK`` are accepted but ignored.
1093 1086
1094 :: 1087 ::
1095 1088
1096 struct kvm_clock_data { 1089 struct kvm_clock_data {
1097 __u64 clock; /* kvmclock current val 1090 __u64 clock; /* kvmclock current value */
1098 __u32 flags; 1091 __u32 flags;
1099 __u32 pad0; 1092 __u32 pad0;
1100 __u64 realtime; 1093 __u64 realtime;
1101 __u64 host_tsc; 1094 __u64 host_tsc;
1102 __u32 pad[4]; 1095 __u32 pad[4];
1103 }; 1096 };
1104 1097
1105 1098
1106 4.31 KVM_GET_VCPU_EVENTS 1099 4.31 KVM_GET_VCPU_EVENTS
1107 ------------------------ 1100 ------------------------
1108 1101
1109 :Capability: KVM_CAP_VCPU_EVENTS 1102 :Capability: KVM_CAP_VCPU_EVENTS
1110 :Extended by: KVM_CAP_INTR_SHADOW 1103 :Extended by: KVM_CAP_INTR_SHADOW
1111 :Architectures: x86, arm64 1104 :Architectures: x86, arm64
1112 :Type: vcpu ioctl 1105 :Type: vcpu ioctl
1113 :Parameters: struct kvm_vcpu_events (out) 1106 :Parameters: struct kvm_vcpu_events (out)
1114 :Returns: 0 on success, -1 on error 1107 :Returns: 0 on success, -1 on error
1115 1108
1116 X86: 1109 X86:
1117 ^^^^ 1110 ^^^^
1118 1111
1119 Gets currently pending exceptions, interrupts 1112 Gets currently pending exceptions, interrupts, and NMIs as well as related
1120 states of the vcpu. 1113 states of the vcpu.
1121 1114
1122 :: 1115 ::
1123 1116
1124 struct kvm_vcpu_events { 1117 struct kvm_vcpu_events {
1125 struct { 1118 struct {
1126 __u8 injected; 1119 __u8 injected;
1127 __u8 nr; 1120 __u8 nr;
1128 __u8 has_error_code; 1121 __u8 has_error_code;
1129 __u8 pending; 1122 __u8 pending;
1130 __u32 error_code; 1123 __u32 error_code;
1131 } exception; 1124 } exception;
1132 struct { 1125 struct {
1133 __u8 injected; 1126 __u8 injected;
1134 __u8 nr; 1127 __u8 nr;
1135 __u8 soft; 1128 __u8 soft;
1136 __u8 shadow; 1129 __u8 shadow;
1137 } interrupt; 1130 } interrupt;
1138 struct { 1131 struct {
1139 __u8 injected; 1132 __u8 injected;
1140 __u8 pending; 1133 __u8 pending;
1141 __u8 masked; 1134 __u8 masked;
1142 __u8 pad; 1135 __u8 pad;
1143 } nmi; 1136 } nmi;
1144 __u32 sipi_vector; 1137 __u32 sipi_vector;
1145 __u32 flags; 1138 __u32 flags;
1146 struct { 1139 struct {
1147 __u8 smm; 1140 __u8 smm;
1148 __u8 pending; 1141 __u8 pending;
1149 __u8 smm_inside_nmi; 1142 __u8 smm_inside_nmi;
1150 __u8 latched_init; 1143 __u8 latched_init;
1151 } smi; 1144 } smi;
1152 __u8 reserved[27]; 1145 __u8 reserved[27];
1153 __u8 exception_has_payload; 1146 __u8 exception_has_payload;
1154 __u64 exception_payload; 1147 __u64 exception_payload;
1155 }; 1148 };
1156 1149
1157 The following bits are defined in the flags f 1150 The following bits are defined in the flags field:
1158 1151
1159 - KVM_VCPUEVENT_VALID_SHADOW may be set to si 1152 - KVM_VCPUEVENT_VALID_SHADOW may be set to signal that
1160 interrupt.shadow contains a valid state. 1153 interrupt.shadow contains a valid state.
1161 1154
1162 - KVM_VCPUEVENT_VALID_SMM may be set to signa 1155 - KVM_VCPUEVENT_VALID_SMM may be set to signal that smi contains a
1163 valid state. 1156 valid state.
1164 1157
1165 - KVM_VCPUEVENT_VALID_PAYLOAD may be set to s 1158 - KVM_VCPUEVENT_VALID_PAYLOAD may be set to signal that the
1166 exception_has_payload, exception_payload, a 1159 exception_has_payload, exception_payload, and exception.pending
1167 fields contain a valid state. This bit will 1160 fields contain a valid state. This bit will be set whenever
1168 KVM_CAP_EXCEPTION_PAYLOAD is enabled. 1161 KVM_CAP_EXCEPTION_PAYLOAD is enabled.
1169 1162
1170 - KVM_VCPUEVENT_VALID_TRIPLE_FAULT may be set 1163 - KVM_VCPUEVENT_VALID_TRIPLE_FAULT may be set to signal that the
1171 triple_fault_pending field contains a valid 1164 triple_fault_pending field contains a valid state. This bit will
1172 be set whenever KVM_CAP_X86_TRIPLE_FAULT_EV 1165 be set whenever KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled.
1173 1166
1174 ARM64: 1167 ARM64:
1175 ^^^^^^ 1168 ^^^^^^
1176 1169
1177 If the guest accesses a device that is being 1170 If the guest accesses a device that is being emulated by the host kernel in
1178 such a way that a real device would generate 1171 such a way that a real device would generate a physical SError, KVM may make
1179 a virtual SError pending for that VCPU. This 1172 a virtual SError pending for that VCPU. This system error interrupt remains
1180 pending until the guest takes the exception b 1173 pending until the guest takes the exception by unmasking PSTATE.A.
1181 1174
1182 Running the VCPU may cause it to take a pendi 1175 Running the VCPU may cause it to take a pending SError, or make an access that
1183 causes an SError to become pending. The event 1176 causes an SError to become pending. The event's description is only valid while
1184 the VPCU is not running. 1177 the VPCU is not running.
1185 1178
1186 This API provides a way to read and write the 1179 This API provides a way to read and write the pending 'event' state that is not
1187 visible to the guest. To save, restore or mig 1180 visible to the guest. To save, restore or migrate a VCPU the struct representing
1188 the state can be read then written using this 1181 the state can be read then written using this GET/SET API, along with the other
1189 guest-visible registers. It is not possible t 1182 guest-visible registers. It is not possible to 'cancel' an SError that has been
1190 made pending. 1183 made pending.
1191 1184
1192 A device being emulated in user-space may als 1185 A device being emulated in user-space may also wish to generate an SError. To do
1193 this the events structure can be populated by 1186 this the events structure can be populated by user-space. The current state
1194 should be read first, to ensure no existing S 1187 should be read first, to ensure no existing SError is pending. If an existing
1195 SError is pending, the architecture's 'Multip 1188 SError is pending, the architecture's 'Multiple SError interrupts' rules should
1196 be followed. (2.5.3 of DDI0587.a "ARM Reliabi 1189 be followed. (2.5.3 of DDI0587.a "ARM Reliability, Availability, and
1197 Serviceability (RAS) Specification"). 1190 Serviceability (RAS) Specification").
1198 1191
1199 SError exceptions always have an ESR value. S 1192 SError exceptions always have an ESR value. Some CPUs have the ability to
1200 specify what the virtual SError's ESR value s 1193 specify what the virtual SError's ESR value should be. These systems will
1201 advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In t 1194 advertise KVM_CAP_ARM_INJECT_SERROR_ESR. In this case exception.has_esr will
1202 always have a non-zero value when read, and t 1195 always have a non-zero value when read, and the agent making an SError pending
1203 should specify the ISS field in the lower 24 1196 should specify the ISS field in the lower 24 bits of exception.serror_esr. If
1204 the system supports KVM_CAP_ARM_INJECT_SERROR 1197 the system supports KVM_CAP_ARM_INJECT_SERROR_ESR, but user-space sets the events
1205 with exception.has_esr as zero, KVM will choo 1198 with exception.has_esr as zero, KVM will choose an ESR.
1206 1199
1207 Specifying exception.has_esr on a system that 1200 Specifying exception.has_esr on a system that does not support it will return
1208 -EINVAL. Setting anything other than the lowe 1201 -EINVAL. Setting anything other than the lower 24bits of exception.serror_esr
1209 will return -EINVAL. 1202 will return -EINVAL.
1210 1203
1211 It is not possible to read back a pending ext 1204 It is not possible to read back a pending external abort (injected via
1212 KVM_SET_VCPU_EVENTS or otherwise) because suc 1205 KVM_SET_VCPU_EVENTS or otherwise) because such an exception is always delivered
1213 directly to the virtual CPU). 1206 directly to the virtual CPU).
1214 1207
1215 :: 1208 ::
1216 1209
1217 struct kvm_vcpu_events { 1210 struct kvm_vcpu_events {
1218 struct { 1211 struct {
1219 __u8 serror_pending; 1212 __u8 serror_pending;
1220 __u8 serror_has_esr; 1213 __u8 serror_has_esr;
1221 __u8 ext_dabt_pending; 1214 __u8 ext_dabt_pending;
1222 /* Align it to 8 bytes */ 1215 /* Align it to 8 bytes */
1223 __u8 pad[5]; 1216 __u8 pad[5];
1224 __u64 serror_esr; 1217 __u64 serror_esr;
1225 } exception; 1218 } exception;
1226 __u32 reserved[12]; 1219 __u32 reserved[12];
1227 }; 1220 };
1228 1221
1229 4.32 KVM_SET_VCPU_EVENTS 1222 4.32 KVM_SET_VCPU_EVENTS
1230 ------------------------ 1223 ------------------------
1231 1224
1232 :Capability: KVM_CAP_VCPU_EVENTS 1225 :Capability: KVM_CAP_VCPU_EVENTS
1233 :Extended by: KVM_CAP_INTR_SHADOW 1226 :Extended by: KVM_CAP_INTR_SHADOW
1234 :Architectures: x86, arm64 1227 :Architectures: x86, arm64
1235 :Type: vcpu ioctl 1228 :Type: vcpu ioctl
1236 :Parameters: struct kvm_vcpu_events (in) 1229 :Parameters: struct kvm_vcpu_events (in)
1237 :Returns: 0 on success, -1 on error 1230 :Returns: 0 on success, -1 on error
1238 1231
1239 X86: 1232 X86:
1240 ^^^^ 1233 ^^^^
1241 1234
1242 Set pending exceptions, interrupts, and NMIs 1235 Set pending exceptions, interrupts, and NMIs as well as related states of the
1243 vcpu. 1236 vcpu.
1244 1237
1245 See KVM_GET_VCPU_EVENTS for the data structur 1238 See KVM_GET_VCPU_EVENTS for the data structure.
1246 1239
1247 Fields that may be modified asynchronously by 1240 Fields that may be modified asynchronously by running VCPUs can be excluded
1248 from the update. These fields are nmi.pending 1241 from the update. These fields are nmi.pending, sipi_vector, smi.smm,
1249 smi.pending. Keep the corresponding bits in t 1242 smi.pending. Keep the corresponding bits in the flags field cleared to
1250 suppress overwriting the current in-kernel st 1243 suppress overwriting the current in-kernel state. The bits are:
1251 1244
1252 =============================== ============ 1245 =============================== ==================================
1253 KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi 1246 KVM_VCPUEVENT_VALID_NMI_PENDING transfer nmi.pending to the kernel
1254 KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sip 1247 KVM_VCPUEVENT_VALID_SIPI_VECTOR transfer sipi_vector
1255 KVM_VCPUEVENT_VALID_SMM transfer the 1248 KVM_VCPUEVENT_VALID_SMM transfer the smi sub-struct.
1256 =============================== ============ 1249 =============================== ==================================
1257 1250
1258 If KVM_CAP_INTR_SHADOW is available, KVM_VCPU 1251 If KVM_CAP_INTR_SHADOW is available, KVM_VCPUEVENT_VALID_SHADOW can be set in
1259 the flags field to signal that interrupt.shad 1252 the flags field to signal that interrupt.shadow contains a valid state and
1260 shall be written into the VCPU. 1253 shall be written into the VCPU.
1261 1254
1262 KVM_VCPUEVENT_VALID_SMM can only be set if KV 1255 KVM_VCPUEVENT_VALID_SMM can only be set if KVM_CAP_X86_SMM is available.
1263 1256
1264 If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_ 1257 If KVM_CAP_EXCEPTION_PAYLOAD is enabled, KVM_VCPUEVENT_VALID_PAYLOAD
1265 can be set in the flags field to signal that 1258 can be set in the flags field to signal that the
1266 exception_has_payload, exception_payload, and 1259 exception_has_payload, exception_payload, and exception.pending fields
1267 contain a valid state and shall be written in 1260 contain a valid state and shall be written into the VCPU.
1268 1261
1269 If KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled, 1262 If KVM_CAP_X86_TRIPLE_FAULT_EVENT is enabled, KVM_VCPUEVENT_VALID_TRIPLE_FAULT
1270 can be set in flags field to signal that the 1263 can be set in flags field to signal that the triple_fault field contains
1271 a valid state and shall be written into the V 1264 a valid state and shall be written into the VCPU.
1272 1265
1273 ARM64: 1266 ARM64:
1274 ^^^^^^ 1267 ^^^^^^
1275 1268
1276 User space may need to inject several types o 1269 User space may need to inject several types of events to the guest.
1277 1270
1278 Set the pending SError exception state for th 1271 Set the pending SError exception state for this VCPU. It is not possible to
1279 'cancel' an Serror that has been made pending 1272 'cancel' an Serror that has been made pending.
1280 1273
1281 If the guest performed an access to I/O memor 1274 If the guest performed an access to I/O memory which could not be handled by
1282 userspace, for example because of missing ins 1275 userspace, for example because of missing instruction syndrome decode
1283 information or because there is no device map 1276 information or because there is no device mapped at the accessed IPA, then
1284 userspace can ask the kernel to inject an ext 1277 userspace can ask the kernel to inject an external abort using the address
1285 from the exiting fault on the VCPU. It is a p 1278 from the exiting fault on the VCPU. It is a programming error to set
1286 ext_dabt_pending after an exit which was not 1279 ext_dabt_pending after an exit which was not either KVM_EXIT_MMIO or
1287 KVM_EXIT_ARM_NISV. This feature is only avail 1280 KVM_EXIT_ARM_NISV. This feature is only available if the system supports
1288 KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper 1281 KVM_CAP_ARM_INJECT_EXT_DABT. This is a helper which provides commonality in
1289 how userspace reports accesses for the above 1282 how userspace reports accesses for the above cases to guests, across different
1290 userspace implementations. Nevertheless, user 1283 userspace implementations. Nevertheless, userspace can still emulate all Arm
1291 exceptions by manipulating individual registe 1284 exceptions by manipulating individual registers using the KVM_SET_ONE_REG API.
1292 1285
1293 See KVM_GET_VCPU_EVENTS for the data structur 1286 See KVM_GET_VCPU_EVENTS for the data structure.
1294 1287
1295 1288
1296 4.33 KVM_GET_DEBUGREGS 1289 4.33 KVM_GET_DEBUGREGS
1297 ---------------------- 1290 ----------------------
1298 1291
1299 :Capability: KVM_CAP_DEBUGREGS 1292 :Capability: KVM_CAP_DEBUGREGS
1300 :Architectures: x86 1293 :Architectures: x86
1301 :Type: vm ioctl 1294 :Type: vm ioctl
1302 :Parameters: struct kvm_debugregs (out) 1295 :Parameters: struct kvm_debugregs (out)
1303 :Returns: 0 on success, -1 on error 1296 :Returns: 0 on success, -1 on error
1304 1297
1305 Reads debug registers from the vcpu. 1298 Reads debug registers from the vcpu.
1306 1299
1307 :: 1300 ::
1308 1301
1309 struct kvm_debugregs { 1302 struct kvm_debugregs {
1310 __u64 db[4]; 1303 __u64 db[4];
1311 __u64 dr6; 1304 __u64 dr6;
1312 __u64 dr7; 1305 __u64 dr7;
1313 __u64 flags; 1306 __u64 flags;
1314 __u64 reserved[9]; 1307 __u64 reserved[9];
1315 }; 1308 };
1316 1309
1317 1310
1318 4.34 KVM_SET_DEBUGREGS 1311 4.34 KVM_SET_DEBUGREGS
1319 ---------------------- 1312 ----------------------
1320 1313
1321 :Capability: KVM_CAP_DEBUGREGS 1314 :Capability: KVM_CAP_DEBUGREGS
1322 :Architectures: x86 1315 :Architectures: x86
1323 :Type: vm ioctl 1316 :Type: vm ioctl
1324 :Parameters: struct kvm_debugregs (in) 1317 :Parameters: struct kvm_debugregs (in)
1325 :Returns: 0 on success, -1 on error 1318 :Returns: 0 on success, -1 on error
1326 1319
1327 Writes debug registers into the vcpu. 1320 Writes debug registers into the vcpu.
1328 1321
1329 See KVM_GET_DEBUGREGS for the data structure. 1322 See KVM_GET_DEBUGREGS for the data structure. The flags field is unused
1330 yet and must be cleared on entry. 1323 yet and must be cleared on entry.
1331 1324
1332 1325
1333 4.35 KVM_SET_USER_MEMORY_REGION 1326 4.35 KVM_SET_USER_MEMORY_REGION
1334 ------------------------------- 1327 -------------------------------
1335 1328
1336 :Capability: KVM_CAP_USER_MEMORY 1329 :Capability: KVM_CAP_USER_MEMORY
1337 :Architectures: all 1330 :Architectures: all
1338 :Type: vm ioctl 1331 :Type: vm ioctl
1339 :Parameters: struct kvm_userspace_memory_regi 1332 :Parameters: struct kvm_userspace_memory_region (in)
1340 :Returns: 0 on success, -1 on error 1333 :Returns: 0 on success, -1 on error
1341 1334
1342 :: 1335 ::
1343 1336
1344 struct kvm_userspace_memory_region { 1337 struct kvm_userspace_memory_region {
1345 __u32 slot; 1338 __u32 slot;
1346 __u32 flags; 1339 __u32 flags;
1347 __u64 guest_phys_addr; 1340 __u64 guest_phys_addr;
1348 __u64 memory_size; /* bytes */ 1341 __u64 memory_size; /* bytes */
1349 __u64 userspace_addr; /* start of the 1342 __u64 userspace_addr; /* start of the userspace allocated memory */
1350 }; 1343 };
1351 1344
1352 /* for kvm_userspace_memory_region::flags * 1345 /* for kvm_userspace_memory_region::flags */
1353 #define KVM_MEM_LOG_DIRTY_PAGES (1UL 1346 #define KVM_MEM_LOG_DIRTY_PAGES (1UL << 0)
1354 #define KVM_MEM_READONLY (1UL << 1) 1347 #define KVM_MEM_READONLY (1UL << 1)
1355 1348
1356 This ioctl allows the user to create, modify 1349 This ioctl allows the user to create, modify or delete a guest physical
1357 memory slot. Bits 0-15 of "slot" specify the 1350 memory slot. Bits 0-15 of "slot" specify the slot id and this value
1358 should be less than the maximum number of use 1351 should be less than the maximum number of user memory slots supported per
1359 VM. The maximum allowed slots can be queried 1352 VM. The maximum allowed slots can be queried using KVM_CAP_NR_MEMSLOTS.
1360 Slots may not overlap in guest physical addre 1353 Slots may not overlap in guest physical address space.
1361 1354
1362 If KVM_CAP_MULTI_ADDRESS_SPACE is available, 1355 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of "slot"
1363 specifies the address space which is being mo 1356 specifies the address space which is being modified. They must be
1364 less than the value that KVM_CHECK_EXTENSION 1357 less than the value that KVM_CHECK_EXTENSION returns for the
1365 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slot 1358 KVM_CAP_MULTI_ADDRESS_SPACE capability. Slots in separate address spaces
1366 are unrelated; the restriction on overlapping 1359 are unrelated; the restriction on overlapping slots only applies within
1367 each address space. 1360 each address space.
1368 1361
1369 Deleting a slot is done by passing zero for m 1362 Deleting a slot is done by passing zero for memory_size. When changing
1370 an existing slot, it may be moved in the gues 1363 an existing slot, it may be moved in the guest physical memory space,
1371 or its flags may be modified, but it may not 1364 or its flags may be modified, but it may not be resized.
1372 1365
1373 Memory for the region is taken starting at th 1366 Memory for the region is taken starting at the address denoted by the
1374 field userspace_addr, which must point at use 1367 field userspace_addr, which must point at user addressable memory for
1375 the entire memory slot size. Any object may 1368 the entire memory slot size. Any object may back this memory, including
1376 anonymous memory, ordinary files, and hugetlb 1369 anonymous memory, ordinary files, and hugetlbfs.
1377 1370
1378 On architectures that support a form of addre 1371 On architectures that support a form of address tagging, userspace_addr must
1379 be an untagged address. 1372 be an untagged address.
1380 1373
1381 It is recommended that the lower 21 bits of g 1374 It is recommended that the lower 21 bits of guest_phys_addr and userspace_addr
1382 be identical. This allows large pages in the 1375 be identical. This allows large pages in the guest to be backed by large
1383 pages in the host. 1376 pages in the host.
1384 1377
1385 The flags field supports two flags: KVM_MEM_L 1378 The flags field supports two flags: KVM_MEM_LOG_DIRTY_PAGES and
1386 KVM_MEM_READONLY. The former can be set to i 1379 KVM_MEM_READONLY. The former can be set to instruct KVM to keep track of
1387 writes to memory within the slot. See KVM_GE 1380 writes to memory within the slot. See KVM_GET_DIRTY_LOG ioctl to know how to
1388 use it. The latter can be set, if KVM_CAP_RE 1381 use it. The latter can be set, if KVM_CAP_READONLY_MEM capability allows it,
1389 to make a new slot read-only. In this case, 1382 to make a new slot read-only. In this case, writes to this memory will be
1390 posted to userspace as KVM_EXIT_MMIO exits. 1383 posted to userspace as KVM_EXIT_MMIO exits.
1391 1384
1392 When the KVM_CAP_SYNC_MMU capability is avail 1385 When the KVM_CAP_SYNC_MMU capability is available, changes in the backing of
1393 the memory region are automatically reflected 1386 the memory region are automatically reflected into the guest. For example, an
1394 mmap() that affects the region will be made v 1387 mmap() that affects the region will be made visible immediately. Another
1395 example is madvise(MADV_DROP). 1388 example is madvise(MADV_DROP).
1396 1389
1397 Note: On arm64, a write generated by the page 1390 Note: On arm64, a write generated by the page-table walker (to update
1398 the Access and Dirty flags, for example) neve 1391 the Access and Dirty flags, for example) never results in a
1399 KVM_EXIT_MMIO exit when the slot has the KVM_ 1392 KVM_EXIT_MMIO exit when the slot has the KVM_MEM_READONLY flag. This
1400 is because KVM cannot provide the data that w 1393 is because KVM cannot provide the data that would be written by the
1401 page-table walker, making it impossible to em 1394 page-table walker, making it impossible to emulate the access.
1402 Instead, an abort (data abort if the cause of 1395 Instead, an abort (data abort if the cause of the page-table update
1403 was a load or a store, instruction abort if i 1396 was a load or a store, instruction abort if it was an instruction
1404 fetch) is injected in the guest. 1397 fetch) is injected in the guest.
1405 1398
1406 S390: <<
1407 ^^^^^ <<
1408 <<
1409 Returns -EINVAL if the VM has the KVM_VM_S390 <<
1410 Returns -EINVAL if called on a protected VM. <<
1411 <<
1412 4.36 KVM_SET_TSS_ADDR 1399 4.36 KVM_SET_TSS_ADDR
1413 --------------------- 1400 ---------------------
1414 1401
1415 :Capability: KVM_CAP_SET_TSS_ADDR 1402 :Capability: KVM_CAP_SET_TSS_ADDR
1416 :Architectures: x86 1403 :Architectures: x86
1417 :Type: vm ioctl 1404 :Type: vm ioctl
1418 :Parameters: unsigned long tss_address (in) 1405 :Parameters: unsigned long tss_address (in)
1419 :Returns: 0 on success, -1 on error 1406 :Returns: 0 on success, -1 on error
1420 1407
1421 This ioctl defines the physical address of a 1408 This ioctl defines the physical address of a three-page region in the guest
1422 physical address space. The region must be w 1409 physical address space. The region must be within the first 4GB of the
1423 guest physical address space and must not con 1410 guest physical address space and must not conflict with any memory slot
1424 or any mmio address. The guest may malfuncti 1411 or any mmio address. The guest may malfunction if it accesses this memory
1425 region. 1412 region.
1426 1413
1427 This ioctl is required on Intel-based hosts. 1414 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1428 because of a quirk in the virtualization impl 1415 because of a quirk in the virtualization implementation (see the internals
1429 documentation when it pops into existence). 1416 documentation when it pops into existence).
1430 1417
1431 1418
1432 4.37 KVM_ENABLE_CAP 1419 4.37 KVM_ENABLE_CAP
1433 ------------------- 1420 -------------------
1434 1421
1435 :Capability: KVM_CAP_ENABLE_CAP 1422 :Capability: KVM_CAP_ENABLE_CAP
1436 :Architectures: mips, ppc, s390, x86, loongar 1423 :Architectures: mips, ppc, s390, x86, loongarch
1437 :Type: vcpu ioctl 1424 :Type: vcpu ioctl
1438 :Parameters: struct kvm_enable_cap (in) 1425 :Parameters: struct kvm_enable_cap (in)
1439 :Returns: 0 on success; -1 on error 1426 :Returns: 0 on success; -1 on error
1440 1427
1441 :Capability: KVM_CAP_ENABLE_CAP_VM 1428 :Capability: KVM_CAP_ENABLE_CAP_VM
1442 :Architectures: all 1429 :Architectures: all
1443 :Type: vm ioctl 1430 :Type: vm ioctl
1444 :Parameters: struct kvm_enable_cap (in) 1431 :Parameters: struct kvm_enable_cap (in)
1445 :Returns: 0 on success; -1 on error 1432 :Returns: 0 on success; -1 on error
1446 1433
1447 .. note:: 1434 .. note::
1448 1435
1449 Not all extensions are enabled by default. 1436 Not all extensions are enabled by default. Using this ioctl the application
1450 can enable an extension, making it availab 1437 can enable an extension, making it available to the guest.
1451 1438
1452 On systems that do not support this ioctl, it 1439 On systems that do not support this ioctl, it always fails. On systems that
1453 do support it, it only works for extensions t 1440 do support it, it only works for extensions that are supported for enablement.
1454 1441
1455 To check if a capability can be enabled, the 1442 To check if a capability can be enabled, the KVM_CHECK_EXTENSION ioctl should
1456 be used. 1443 be used.
1457 1444
1458 :: 1445 ::
1459 1446
1460 struct kvm_enable_cap { 1447 struct kvm_enable_cap {
1461 /* in */ 1448 /* in */
1462 __u32 cap; 1449 __u32 cap;
1463 1450
1464 The capability that is supposed to get enable 1451 The capability that is supposed to get enabled.
1465 1452
1466 :: 1453 ::
1467 1454
1468 __u32 flags; 1455 __u32 flags;
1469 1456
1470 A bitfield indicating future enhancements. Ha 1457 A bitfield indicating future enhancements. Has to be 0 for now.
1471 1458
1472 :: 1459 ::
1473 1460
1474 __u64 args[4]; 1461 __u64 args[4];
1475 1462
1476 Arguments for enabling a feature. If a featur 1463 Arguments for enabling a feature. If a feature needs initial values to
1477 function properly, this is the place to put t 1464 function properly, this is the place to put them.
1478 1465
1479 :: 1466 ::
1480 1467
1481 __u8 pad[64]; 1468 __u8 pad[64];
1482 }; 1469 };
1483 1470
1484 The vcpu ioctl should be used for vcpu-specif 1471 The vcpu ioctl should be used for vcpu-specific capabilities, the vm ioctl
1485 for vm-wide capabilities. 1472 for vm-wide capabilities.
1486 1473
1487 4.38 KVM_GET_MP_STATE 1474 4.38 KVM_GET_MP_STATE
1488 --------------------- 1475 ---------------------
1489 1476
1490 :Capability: KVM_CAP_MP_STATE 1477 :Capability: KVM_CAP_MP_STATE
1491 :Architectures: x86, s390, arm64, riscv, loon 1478 :Architectures: x86, s390, arm64, riscv, loongarch
1492 :Type: vcpu ioctl 1479 :Type: vcpu ioctl
1493 :Parameters: struct kvm_mp_state (out) 1480 :Parameters: struct kvm_mp_state (out)
1494 :Returns: 0 on success; -1 on error 1481 :Returns: 0 on success; -1 on error
1495 1482
1496 :: 1483 ::
1497 1484
1498 struct kvm_mp_state { 1485 struct kvm_mp_state {
1499 __u32 mp_state; 1486 __u32 mp_state;
1500 }; 1487 };
1501 1488
1502 Returns the vcpu's current "multiprocessing s 1489 Returns the vcpu's current "multiprocessing state" (though also valid on
1503 uniprocessor guests). 1490 uniprocessor guests).
1504 1491
1505 Possible values are: 1492 Possible values are:
1506 1493
1507 ========================== ============ 1494 ========================== ===============================================
1508 KVM_MP_STATE_RUNNABLE the vcpu is 1495 KVM_MP_STATE_RUNNABLE the vcpu is currently running
1509 [x86,arm64,r 1496 [x86,arm64,riscv,loongarch]
1510 KVM_MP_STATE_UNINITIALIZED the vcpu is 1497 KVM_MP_STATE_UNINITIALIZED the vcpu is an application processor (AP)
1511 which has no 1498 which has not yet received an INIT signal [x86]
1512 KVM_MP_STATE_INIT_RECEIVED the vcpu has 1499 KVM_MP_STATE_INIT_RECEIVED the vcpu has received an INIT signal, and is
1513 now ready fo 1500 now ready for a SIPI [x86]
1514 KVM_MP_STATE_HALTED the vcpu has 1501 KVM_MP_STATE_HALTED the vcpu has executed a HLT instruction and
1515 is waiting f 1502 is waiting for an interrupt [x86]
1516 KVM_MP_STATE_SIPI_RECEIVED the vcpu has 1503 KVM_MP_STATE_SIPI_RECEIVED the vcpu has just received a SIPI (vector
1517 accessible v 1504 accessible via KVM_GET_VCPU_EVENTS) [x86]
1518 KVM_MP_STATE_STOPPED the vcpu is 1505 KVM_MP_STATE_STOPPED the vcpu is stopped [s390,arm64,riscv]
1519 KVM_MP_STATE_CHECK_STOP the vcpu is 1506 KVM_MP_STATE_CHECK_STOP the vcpu is in a special error state [s390]
1520 KVM_MP_STATE_OPERATING the vcpu is 1507 KVM_MP_STATE_OPERATING the vcpu is operating (running or halted)
1521 [s390] 1508 [s390]
1522 KVM_MP_STATE_LOAD the vcpu is 1509 KVM_MP_STATE_LOAD the vcpu is in a special load/startup state
1523 [s390] 1510 [s390]
1524 KVM_MP_STATE_SUSPENDED the vcpu is 1511 KVM_MP_STATE_SUSPENDED the vcpu is in a suspend state and is waiting
1525 for a wakeup 1512 for a wakeup event [arm64]
1526 ========================== ============ 1513 ========================== ===============================================
1527 1514
1528 On x86, this ioctl is only useful after KVM_C 1515 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1529 in-kernel irqchip, the multiprocessing state 1516 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1530 these architectures. 1517 these architectures.
1531 1518
1532 For arm64: 1519 For arm64:
1533 ^^^^^^^^^^ 1520 ^^^^^^^^^^
1534 1521
1535 If a vCPU is in the KVM_MP_STATE_SUSPENDED st 1522 If a vCPU is in the KVM_MP_STATE_SUSPENDED state, KVM will emulate the
1536 architectural execution of a WFI instruction. 1523 architectural execution of a WFI instruction.
1537 1524
1538 If a wakeup event is recognized, KVM will exi 1525 If a wakeup event is recognized, KVM will exit to userspace with a
1539 KVM_SYSTEM_EVENT exit, where the event type i 1526 KVM_SYSTEM_EVENT exit, where the event type is KVM_SYSTEM_EVENT_WAKEUP. If
1540 userspace wants to honor the wakeup, it must 1527 userspace wants to honor the wakeup, it must set the vCPU's MP state to
1541 KVM_MP_STATE_RUNNABLE. If it does not, KVM wi 1528 KVM_MP_STATE_RUNNABLE. If it does not, KVM will continue to await a wakeup
1542 event in subsequent calls to KVM_RUN. 1529 event in subsequent calls to KVM_RUN.
1543 1530
1544 .. warning:: 1531 .. warning::
1545 1532
1546 If userspace intends to keep the vCPU in 1533 If userspace intends to keep the vCPU in a SUSPENDED state, it is
1547 strongly recommended that userspace take 1534 strongly recommended that userspace take action to suppress the
1548 wakeup event (such as masking an interru 1535 wakeup event (such as masking an interrupt). Otherwise, subsequent
1549 calls to KVM_RUN will immediately exit w 1536 calls to KVM_RUN will immediately exit with a KVM_SYSTEM_EVENT_WAKEUP
1550 event and inadvertently waste CPU cycles 1537 event and inadvertently waste CPU cycles.
1551 1538
1552 Additionally, if userspace takes action 1539 Additionally, if userspace takes action to suppress a wakeup event,
1553 it is strongly recommended that it also 1540 it is strongly recommended that it also restores the vCPU to its
1554 original state when the vCPU is made RUN 1541 original state when the vCPU is made RUNNABLE again. For example,
1555 if userspace masked a pending interrupt 1542 if userspace masked a pending interrupt to suppress the wakeup,
1556 the interrupt should be unmasked before 1543 the interrupt should be unmasked before returning control to the
1557 guest. 1544 guest.
1558 1545
1559 For riscv: 1546 For riscv:
1560 ^^^^^^^^^^ 1547 ^^^^^^^^^^
1561 1548
1562 The only states that are valid are KVM_MP_STA 1549 The only states that are valid are KVM_MP_STATE_STOPPED and
1563 KVM_MP_STATE_RUNNABLE which reflect if the vc 1550 KVM_MP_STATE_RUNNABLE which reflect if the vcpu is paused or not.
1564 1551
1565 On LoongArch, only the KVM_MP_STATE_RUNNABLE 1552 On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect
1566 whether the vcpu is runnable. 1553 whether the vcpu is runnable.
1567 1554
1568 4.39 KVM_SET_MP_STATE 1555 4.39 KVM_SET_MP_STATE
1569 --------------------- 1556 ---------------------
1570 1557
1571 :Capability: KVM_CAP_MP_STATE 1558 :Capability: KVM_CAP_MP_STATE
1572 :Architectures: x86, s390, arm64, riscv, loon 1559 :Architectures: x86, s390, arm64, riscv, loongarch
1573 :Type: vcpu ioctl 1560 :Type: vcpu ioctl
1574 :Parameters: struct kvm_mp_state (in) 1561 :Parameters: struct kvm_mp_state (in)
1575 :Returns: 0 on success; -1 on error 1562 :Returns: 0 on success; -1 on error
1576 1563
1577 Sets the vcpu's current "multiprocessing stat 1564 Sets the vcpu's current "multiprocessing state"; see KVM_GET_MP_STATE for
1578 arguments. 1565 arguments.
1579 1566
1580 On x86, this ioctl is only useful after KVM_C 1567 On x86, this ioctl is only useful after KVM_CREATE_IRQCHIP. Without an
1581 in-kernel irqchip, the multiprocessing state 1568 in-kernel irqchip, the multiprocessing state must be maintained by userspace on
1582 these architectures. 1569 these architectures.
1583 1570
1584 For arm64/riscv: 1571 For arm64/riscv:
1585 ^^^^^^^^^^^^^^^^ 1572 ^^^^^^^^^^^^^^^^
1586 1573
1587 The only states that are valid are KVM_MP_STA 1574 The only states that are valid are KVM_MP_STATE_STOPPED and
1588 KVM_MP_STATE_RUNNABLE which reflect if the vc 1575 KVM_MP_STATE_RUNNABLE which reflect if the vcpu should be paused or not.
1589 1576
1590 On LoongArch, only the KVM_MP_STATE_RUNNABLE 1577 On LoongArch, only the KVM_MP_STATE_RUNNABLE state is used to reflect
1591 whether the vcpu is runnable. 1578 whether the vcpu is runnable.
1592 1579
1593 4.40 KVM_SET_IDENTITY_MAP_ADDR 1580 4.40 KVM_SET_IDENTITY_MAP_ADDR
1594 ------------------------------ 1581 ------------------------------
1595 1582
1596 :Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR 1583 :Capability: KVM_CAP_SET_IDENTITY_MAP_ADDR
1597 :Architectures: x86 1584 :Architectures: x86
1598 :Type: vm ioctl 1585 :Type: vm ioctl
1599 :Parameters: unsigned long identity (in) 1586 :Parameters: unsigned long identity (in)
1600 :Returns: 0 on success, -1 on error 1587 :Returns: 0 on success, -1 on error
1601 1588
1602 This ioctl defines the physical address of a 1589 This ioctl defines the physical address of a one-page region in the guest
1603 physical address space. The region must be w 1590 physical address space. The region must be within the first 4GB of the
1604 guest physical address space and must not con 1591 guest physical address space and must not conflict with any memory slot
1605 or any mmio address. The guest may malfuncti 1592 or any mmio address. The guest may malfunction if it accesses this memory
1606 region. 1593 region.
1607 1594
1608 Setting the address to 0 will result in reset 1595 Setting the address to 0 will result in resetting the address to its default
1609 (0xfffbc000). 1596 (0xfffbc000).
1610 1597
1611 This ioctl is required on Intel-based hosts. 1598 This ioctl is required on Intel-based hosts. This is needed on Intel hardware
1612 because of a quirk in the virtualization impl 1599 because of a quirk in the virtualization implementation (see the internals
1613 documentation when it pops into existence). 1600 documentation when it pops into existence).
1614 1601
1615 Fails if any VCPU has already been created. 1602 Fails if any VCPU has already been created.
1616 1603
1617 4.41 KVM_SET_BOOT_CPU_ID 1604 4.41 KVM_SET_BOOT_CPU_ID
1618 ------------------------ 1605 ------------------------
1619 1606
1620 :Capability: KVM_CAP_SET_BOOT_CPU_ID 1607 :Capability: KVM_CAP_SET_BOOT_CPU_ID
1621 :Architectures: x86 1608 :Architectures: x86
1622 :Type: vm ioctl 1609 :Type: vm ioctl
1623 :Parameters: unsigned long vcpu_id 1610 :Parameters: unsigned long vcpu_id
1624 :Returns: 0 on success, -1 on error 1611 :Returns: 0 on success, -1 on error
1625 1612
1626 Define which vcpu is the Bootstrap Processor 1613 Define which vcpu is the Bootstrap Processor (BSP). Values are the same
1627 as the vcpu id in KVM_CREATE_VCPU. If this i 1614 as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default
1628 is vcpu 0. This ioctl has to be called before 1615 is vcpu 0. This ioctl has to be called before vcpu creation,
1629 otherwise it will return EBUSY error. 1616 otherwise it will return EBUSY error.
1630 1617
1631 1618
1632 4.42 KVM_GET_XSAVE 1619 4.42 KVM_GET_XSAVE
1633 ------------------ 1620 ------------------
1634 1621
1635 :Capability: KVM_CAP_XSAVE 1622 :Capability: KVM_CAP_XSAVE
1636 :Architectures: x86 1623 :Architectures: x86
1637 :Type: vcpu ioctl 1624 :Type: vcpu ioctl
1638 :Parameters: struct kvm_xsave (out) 1625 :Parameters: struct kvm_xsave (out)
1639 :Returns: 0 on success, -1 on error 1626 :Returns: 0 on success, -1 on error
1640 1627
1641 1628
1642 :: 1629 ::
1643 1630
1644 struct kvm_xsave { 1631 struct kvm_xsave {
1645 __u32 region[1024]; 1632 __u32 region[1024];
1646 __u32 extra[0]; 1633 __u32 extra[0];
1647 }; 1634 };
1648 1635
1649 This ioctl would copy current vcpu's xsave st 1636 This ioctl would copy current vcpu's xsave struct to the userspace.
1650 1637
1651 1638
1652 4.43 KVM_SET_XSAVE 1639 4.43 KVM_SET_XSAVE
1653 ------------------ 1640 ------------------
1654 1641
1655 :Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2 1642 :Capability: KVM_CAP_XSAVE and KVM_CAP_XSAVE2
1656 :Architectures: x86 1643 :Architectures: x86
1657 :Type: vcpu ioctl 1644 :Type: vcpu ioctl
1658 :Parameters: struct kvm_xsave (in) 1645 :Parameters: struct kvm_xsave (in)
1659 :Returns: 0 on success, -1 on error 1646 :Returns: 0 on success, -1 on error
1660 1647
1661 :: 1648 ::
1662 1649
1663 1650
1664 struct kvm_xsave { 1651 struct kvm_xsave {
1665 __u32 region[1024]; 1652 __u32 region[1024];
1666 __u32 extra[0]; 1653 __u32 extra[0];
1667 }; 1654 };
1668 1655
1669 This ioctl would copy userspace's xsave struc 1656 This ioctl would copy userspace's xsave struct to the kernel. It copies
1670 as many bytes as are returned by KVM_CHECK_EX 1657 as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2),
1671 when invoked on the vm file descriptor. The s 1658 when invoked on the vm file descriptor. The size value returned by
1672 KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will alwa 1659 KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096.
1673 Currently, it is only greater than 4096 if a 1660 Currently, it is only greater than 4096 if a dynamic feature has been
1674 enabled with ``arch_prctl()``, but this may c 1661 enabled with ``arch_prctl()``, but this may change in the future.
1675 1662
1676 The offsets of the state save areas in struct 1663 The offsets of the state save areas in struct kvm_xsave follow the
1677 contents of CPUID leaf 0xD on the host. 1664 contents of CPUID leaf 0xD on the host.
1678 1665
1679 1666
1680 4.44 KVM_GET_XCRS 1667 4.44 KVM_GET_XCRS
1681 ----------------- 1668 -----------------
1682 1669
1683 :Capability: KVM_CAP_XCRS 1670 :Capability: KVM_CAP_XCRS
1684 :Architectures: x86 1671 :Architectures: x86
1685 :Type: vcpu ioctl 1672 :Type: vcpu ioctl
1686 :Parameters: struct kvm_xcrs (out) 1673 :Parameters: struct kvm_xcrs (out)
1687 :Returns: 0 on success, -1 on error 1674 :Returns: 0 on success, -1 on error
1688 1675
1689 :: 1676 ::
1690 1677
1691 struct kvm_xcr { 1678 struct kvm_xcr {
1692 __u32 xcr; 1679 __u32 xcr;
1693 __u32 reserved; 1680 __u32 reserved;
1694 __u64 value; 1681 __u64 value;
1695 }; 1682 };
1696 1683
1697 struct kvm_xcrs { 1684 struct kvm_xcrs {
1698 __u32 nr_xcrs; 1685 __u32 nr_xcrs;
1699 __u32 flags; 1686 __u32 flags;
1700 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1687 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1701 __u64 padding[16]; 1688 __u64 padding[16];
1702 }; 1689 };
1703 1690
1704 This ioctl would copy current vcpu's xcrs to 1691 This ioctl would copy current vcpu's xcrs to the userspace.
1705 1692
1706 1693
1707 4.45 KVM_SET_XCRS 1694 4.45 KVM_SET_XCRS
1708 ----------------- 1695 -----------------
1709 1696
1710 :Capability: KVM_CAP_XCRS 1697 :Capability: KVM_CAP_XCRS
1711 :Architectures: x86 1698 :Architectures: x86
1712 :Type: vcpu ioctl 1699 :Type: vcpu ioctl
1713 :Parameters: struct kvm_xcrs (in) 1700 :Parameters: struct kvm_xcrs (in)
1714 :Returns: 0 on success, -1 on error 1701 :Returns: 0 on success, -1 on error
1715 1702
1716 :: 1703 ::
1717 1704
1718 struct kvm_xcr { 1705 struct kvm_xcr {
1719 __u32 xcr; 1706 __u32 xcr;
1720 __u32 reserved; 1707 __u32 reserved;
1721 __u64 value; 1708 __u64 value;
1722 }; 1709 };
1723 1710
1724 struct kvm_xcrs { 1711 struct kvm_xcrs {
1725 __u32 nr_xcrs; 1712 __u32 nr_xcrs;
1726 __u32 flags; 1713 __u32 flags;
1727 struct kvm_xcr xcrs[KVM_MAX_XCRS]; 1714 struct kvm_xcr xcrs[KVM_MAX_XCRS];
1728 __u64 padding[16]; 1715 __u64 padding[16];
1729 }; 1716 };
1730 1717
1731 This ioctl would set vcpu's xcr to the value 1718 This ioctl would set vcpu's xcr to the value userspace specified.
1732 1719
1733 1720
1734 4.46 KVM_GET_SUPPORTED_CPUID 1721 4.46 KVM_GET_SUPPORTED_CPUID
1735 ---------------------------- 1722 ----------------------------
1736 1723
1737 :Capability: KVM_CAP_EXT_CPUID 1724 :Capability: KVM_CAP_EXT_CPUID
1738 :Architectures: x86 1725 :Architectures: x86
1739 :Type: system ioctl 1726 :Type: system ioctl
1740 :Parameters: struct kvm_cpuid2 (in/out) 1727 :Parameters: struct kvm_cpuid2 (in/out)
1741 :Returns: 0 on success, -1 on error 1728 :Returns: 0 on success, -1 on error
1742 1729
1743 :: 1730 ::
1744 1731
1745 struct kvm_cpuid2 { 1732 struct kvm_cpuid2 {
1746 __u32 nent; 1733 __u32 nent;
1747 __u32 padding; 1734 __u32 padding;
1748 struct kvm_cpuid_entry2 entries[0]; 1735 struct kvm_cpuid_entry2 entries[0];
1749 }; 1736 };
1750 1737
1751 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 1738 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
1752 #define KVM_CPUID_FLAG_STATEFUL_FUNC 1739 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */
1753 #define KVM_CPUID_FLAG_STATE_READ_NEXT 1740 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */
1754 1741
1755 struct kvm_cpuid_entry2 { 1742 struct kvm_cpuid_entry2 {
1756 __u32 function; 1743 __u32 function;
1757 __u32 index; 1744 __u32 index;
1758 __u32 flags; 1745 __u32 flags;
1759 __u32 eax; 1746 __u32 eax;
1760 __u32 ebx; 1747 __u32 ebx;
1761 __u32 ecx; 1748 __u32 ecx;
1762 __u32 edx; 1749 __u32 edx;
1763 __u32 padding[3]; 1750 __u32 padding[3];
1764 }; 1751 };
1765 1752
1766 This ioctl returns x86 cpuid features which a 1753 This ioctl returns x86 cpuid features which are supported by both the
1767 hardware and kvm in its default configuration 1754 hardware and kvm in its default configuration. Userspace can use the
1768 information returned by this ioctl to constru 1755 information returned by this ioctl to construct cpuid information (for
1769 KVM_SET_CPUID2) that is consistent with hardw 1756 KVM_SET_CPUID2) that is consistent with hardware, kernel, and
1770 userspace capabilities, and with user require 1757 userspace capabilities, and with user requirements (for example, the
1771 user may wish to constrain cpuid to emulate o 1758 user may wish to constrain cpuid to emulate older hardware, or for
1772 feature consistency across a cluster). 1759 feature consistency across a cluster).
1773 1760
1774 Dynamically-enabled feature bits need to be r 1761 Dynamically-enabled feature bits need to be requested with
1775 ``arch_prctl()`` before calling this ioctl. F 1762 ``arch_prctl()`` before calling this ioctl. Feature bits that have not
1776 been requested are excluded from the result. 1763 been requested are excluded from the result.
1777 1764
1778 Note that certain capabilities, such as KVM_C 1765 Note that certain capabilities, such as KVM_CAP_X86_DISABLE_EXITS, may
1779 expose cpuid features (e.g. MONITOR) which ar 1766 expose cpuid features (e.g. MONITOR) which are not supported by kvm in
1780 its default configuration. If userspace enabl 1767 its default configuration. If userspace enables such capabilities, it
1781 is responsible for modifying the results of t 1768 is responsible for modifying the results of this ioctl appropriately.
1782 1769
1783 Userspace invokes KVM_GET_SUPPORTED_CPUID by 1770 Userspace invokes KVM_GET_SUPPORTED_CPUID by passing a kvm_cpuid2 structure
1784 with the 'nent' field indicating the number o 1771 with the 'nent' field indicating the number of entries in the variable-size
1785 array 'entries'. If the number of entries is 1772 array 'entries'. If the number of entries is too low to describe the cpu
1786 capabilities, an error (E2BIG) is returned. 1773 capabilities, an error (E2BIG) is returned. If the number is too high,
1787 the 'nent' field is adjusted and an error (EN 1774 the 'nent' field is adjusted and an error (ENOMEM) is returned. If the
1788 number is just right, the 'nent' field is adj 1775 number is just right, the 'nent' field is adjusted to the number of valid
1789 entries in the 'entries' array, which is then 1776 entries in the 'entries' array, which is then filled.
1790 1777
1791 The entries returned are the host cpuid as re 1778 The entries returned are the host cpuid as returned by the cpuid instruction,
1792 with unknown or unsupported features masked o 1779 with unknown or unsupported features masked out. Some features (for example,
1793 x2apic), may not be present in the host cpu, 1780 x2apic), may not be present in the host cpu, but are exposed by kvm if it can
1794 emulate them efficiently. The fields in each 1781 emulate them efficiently. The fields in each entry are defined as follows:
1795 1782
1796 function: 1783 function:
1797 the eax value used to obtain the ent 1784 the eax value used to obtain the entry
1798 1785
1799 index: 1786 index:
1800 the ecx value used to obtain the ent 1787 the ecx value used to obtain the entry (for entries that are
1801 affected by ecx) 1788 affected by ecx)
1802 1789
1803 flags: 1790 flags:
1804 an OR of zero or more of the following: 1791 an OR of zero or more of the following:
1805 1792
1806 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 1793 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
1807 if the index field is valid 1794 if the index field is valid
1808 1795
1809 eax, ebx, ecx, edx: 1796 eax, ebx, ecx, edx:
1810 the values returned by the cpuid ins 1797 the values returned by the cpuid instruction for
1811 this function/index combination 1798 this function/index combination
1812 1799
1813 The TSC deadline timer feature (CPUID leaf 1, 1800 The TSC deadline timer feature (CPUID leaf 1, ecx[24]) is always returned
1814 as false, since the feature depends on KVM_CR 1801 as false, since the feature depends on KVM_CREATE_IRQCHIP for local APIC
1815 support. Instead it is reported via:: 1802 support. Instead it is reported via::
1816 1803
1817 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEAD 1804 ioctl(KVM_CHECK_EXTENSION, KVM_CAP_TSC_DEADLINE_TIMER)
1818 1805
1819 if that returns true and you use KVM_CREATE_I 1806 if that returns true and you use KVM_CREATE_IRQCHIP, or if you emulate the
1820 feature in userspace, then you can enable the 1807 feature in userspace, then you can enable the feature for KVM_SET_CPUID2.
1821 1808
1822 1809
1823 4.47 KVM_PPC_GET_PVINFO 1810 4.47 KVM_PPC_GET_PVINFO
1824 ----------------------- 1811 -----------------------
1825 1812
1826 :Capability: KVM_CAP_PPC_GET_PVINFO 1813 :Capability: KVM_CAP_PPC_GET_PVINFO
1827 :Architectures: ppc 1814 :Architectures: ppc
1828 :Type: vm ioctl 1815 :Type: vm ioctl
1829 :Parameters: struct kvm_ppc_pvinfo (out) 1816 :Parameters: struct kvm_ppc_pvinfo (out)
1830 :Returns: 0 on success, !0 on error 1817 :Returns: 0 on success, !0 on error
1831 1818
1832 :: 1819 ::
1833 1820
1834 struct kvm_ppc_pvinfo { 1821 struct kvm_ppc_pvinfo {
1835 __u32 flags; 1822 __u32 flags;
1836 __u32 hcall[4]; 1823 __u32 hcall[4];
1837 __u8 pad[108]; 1824 __u8 pad[108];
1838 }; 1825 };
1839 1826
1840 This ioctl fetches PV specific information th 1827 This ioctl fetches PV specific information that need to be passed to the guest
1841 using the device tree or other means from vm 1828 using the device tree or other means from vm context.
1842 1829
1843 The hcall array defines 4 instructions that m 1830 The hcall array defines 4 instructions that make up a hypercall.
1844 1831
1845 If any additional field gets added to this st 1832 If any additional field gets added to this structure later on, a bit for that
1846 additional piece of information will be set i 1833 additional piece of information will be set in the flags bitmap.
1847 1834
1848 The flags bitmap is defined as:: 1835 The flags bitmap is defined as::
1849 1836
1850 /* the host supports the ePAPR idle hcall 1837 /* the host supports the ePAPR idle hcall
1851 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1< 1838 #define KVM_PPC_PVINFO_FLAGS_EV_IDLE (1<<0)
1852 1839
1853 4.52 KVM_SET_GSI_ROUTING 1840 4.52 KVM_SET_GSI_ROUTING
1854 ------------------------ 1841 ------------------------
1855 1842
1856 :Capability: KVM_CAP_IRQ_ROUTING 1843 :Capability: KVM_CAP_IRQ_ROUTING
1857 :Architectures: x86 s390 arm64 1844 :Architectures: x86 s390 arm64
1858 :Type: vm ioctl 1845 :Type: vm ioctl
1859 :Parameters: struct kvm_irq_routing (in) 1846 :Parameters: struct kvm_irq_routing (in)
1860 :Returns: 0 on success, -1 on error 1847 :Returns: 0 on success, -1 on error
1861 1848
1862 Sets the GSI routing table entries, overwriti 1849 Sets the GSI routing table entries, overwriting any previously set entries.
1863 1850
1864 On arm64, GSI routing has the following limit 1851 On arm64, GSI routing has the following limitation:
1865 1852
1866 - GSI routing does not apply to KVM_IRQ_LINE 1853 - GSI routing does not apply to KVM_IRQ_LINE but only to KVM_IRQFD.
1867 1854
1868 :: 1855 ::
1869 1856
1870 struct kvm_irq_routing { 1857 struct kvm_irq_routing {
1871 __u32 nr; 1858 __u32 nr;
1872 __u32 flags; 1859 __u32 flags;
1873 struct kvm_irq_routing_entry entries[ 1860 struct kvm_irq_routing_entry entries[0];
1874 }; 1861 };
1875 1862
1876 No flags are specified so far, the correspond 1863 No flags are specified so far, the corresponding field must be set to zero.
1877 1864
1878 :: 1865 ::
1879 1866
1880 struct kvm_irq_routing_entry { 1867 struct kvm_irq_routing_entry {
1881 __u32 gsi; 1868 __u32 gsi;
1882 __u32 type; 1869 __u32 type;
1883 __u32 flags; 1870 __u32 flags;
1884 __u32 pad; 1871 __u32 pad;
1885 union { 1872 union {
1886 struct kvm_irq_routing_irqchi 1873 struct kvm_irq_routing_irqchip irqchip;
1887 struct kvm_irq_routing_msi ms 1874 struct kvm_irq_routing_msi msi;
1888 struct kvm_irq_routing_s390_a 1875 struct kvm_irq_routing_s390_adapter adapter;
1889 struct kvm_irq_routing_hv_sin 1876 struct kvm_irq_routing_hv_sint hv_sint;
1890 struct kvm_irq_routing_xen_ev 1877 struct kvm_irq_routing_xen_evtchn xen_evtchn;
1891 __u32 pad[8]; 1878 __u32 pad[8];
1892 } u; 1879 } u;
1893 }; 1880 };
1894 1881
1895 /* gsi routing entry types */ 1882 /* gsi routing entry types */
1896 #define KVM_IRQ_ROUTING_IRQCHIP 1 1883 #define KVM_IRQ_ROUTING_IRQCHIP 1
1897 #define KVM_IRQ_ROUTING_MSI 2 1884 #define KVM_IRQ_ROUTING_MSI 2
1898 #define KVM_IRQ_ROUTING_S390_ADAPTER 3 1885 #define KVM_IRQ_ROUTING_S390_ADAPTER 3
1899 #define KVM_IRQ_ROUTING_HV_SINT 4 1886 #define KVM_IRQ_ROUTING_HV_SINT 4
1900 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 1887 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5
1901 1888
1902 flags: 1889 flags:
1903 1890
1904 - KVM_MSI_VALID_DEVID: used along with KVM_IR 1891 - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry
1905 type, specifies that the devid field contai 1892 type, specifies that the devid field contains a valid value. The per-VM
1906 KVM_CAP_MSI_DEVID capability advertises the 1893 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
1907 the device ID. If this capability is not a 1894 the device ID. If this capability is not available, userspace should
1908 never set the KVM_MSI_VALID_DEVID flag as t 1895 never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
1909 - zero otherwise 1896 - zero otherwise
1910 1897
1911 :: 1898 ::
1912 1899
1913 struct kvm_irq_routing_irqchip { 1900 struct kvm_irq_routing_irqchip {
1914 __u32 irqchip; 1901 __u32 irqchip;
1915 __u32 pin; 1902 __u32 pin;
1916 }; 1903 };
1917 1904
1918 struct kvm_irq_routing_msi { 1905 struct kvm_irq_routing_msi {
1919 __u32 address_lo; 1906 __u32 address_lo;
1920 __u32 address_hi; 1907 __u32 address_hi;
1921 __u32 data; 1908 __u32 data;
1922 union { 1909 union {
1923 __u32 pad; 1910 __u32 pad;
1924 __u32 devid; 1911 __u32 devid;
1925 }; 1912 };
1926 }; 1913 };
1927 1914
1928 If KVM_MSI_VALID_DEVID is set, devid contains 1915 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
1929 for the device that wrote the MSI message. F 1916 for the device that wrote the MSI message. For PCI, this is usually a
1930 BDF identifier in the lower 16 bits. !! 1917 BFD identifier in the lower 16 bits.
1931 1918
1932 On x86, address_hi is ignored unless the KVM_ 1919 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
1933 feature of KVM_CAP_X2APIC_API capability is e 1920 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
1934 address_hi bits 31-8 provide bits 31-8 of the 1921 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
1935 address_hi must be zero. 1922 address_hi must be zero.
1936 1923
1937 :: 1924 ::
1938 1925
1939 struct kvm_irq_routing_s390_adapter { 1926 struct kvm_irq_routing_s390_adapter {
1940 __u64 ind_addr; 1927 __u64 ind_addr;
1941 __u64 summary_addr; 1928 __u64 summary_addr;
1942 __u64 ind_offset; 1929 __u64 ind_offset;
1943 __u32 summary_offset; 1930 __u32 summary_offset;
1944 __u32 adapter_id; 1931 __u32 adapter_id;
1945 }; 1932 };
1946 1933
1947 struct kvm_irq_routing_hv_sint { 1934 struct kvm_irq_routing_hv_sint {
1948 __u32 vcpu; 1935 __u32 vcpu;
1949 __u32 sint; 1936 __u32 sint;
1950 }; 1937 };
1951 1938
1952 struct kvm_irq_routing_xen_evtchn { 1939 struct kvm_irq_routing_xen_evtchn {
1953 __u32 port; 1940 __u32 port;
1954 __u32 vcpu; 1941 __u32 vcpu;
1955 __u32 priority; 1942 __u32 priority;
1956 }; 1943 };
1957 1944
1958 1945
1959 When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM 1946 When KVM_CAP_XEN_HVM includes the KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL bit
1960 in its indication of supported features, rout 1947 in its indication of supported features, routing to Xen event channels
1961 is supported. Although the priority field is 1948 is supported. Although the priority field is present, only the value
1962 KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported 1949 KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL is supported, which means delivery by
1963 2 level event channels. FIFO event channel su 1950 2 level event channels. FIFO event channel support may be added in
1964 the future. 1951 the future.
1965 1952
1966 1953
1967 4.55 KVM_SET_TSC_KHZ 1954 4.55 KVM_SET_TSC_KHZ
1968 -------------------- 1955 --------------------
1969 1956
1970 :Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM 1957 :Capability: KVM_CAP_TSC_CONTROL / KVM_CAP_VM_TSC_CONTROL
1971 :Architectures: x86 1958 :Architectures: x86
1972 :Type: vcpu ioctl / vm ioctl 1959 :Type: vcpu ioctl / vm ioctl
1973 :Parameters: virtual tsc_khz 1960 :Parameters: virtual tsc_khz
1974 :Returns: 0 on success, -1 on error 1961 :Returns: 0 on success, -1 on error
1975 1962
1976 Specifies the tsc frequency for the virtual m 1963 Specifies the tsc frequency for the virtual machine. The unit of the
1977 frequency is KHz. 1964 frequency is KHz.
1978 1965
1979 If the KVM_CAP_VM_TSC_CONTROL capability is a 1966 If the KVM_CAP_VM_TSC_CONTROL capability is advertised, this can also
1980 be used as a vm ioctl to set the initial tsc 1967 be used as a vm ioctl to set the initial tsc frequency of subsequently
1981 created vCPUs. 1968 created vCPUs.
1982 1969
1983 4.56 KVM_GET_TSC_KHZ 1970 4.56 KVM_GET_TSC_KHZ
1984 -------------------- 1971 --------------------
1985 1972
1986 :Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM 1973 :Capability: KVM_CAP_GET_TSC_KHZ / KVM_CAP_VM_TSC_CONTROL
1987 :Architectures: x86 1974 :Architectures: x86
1988 :Type: vcpu ioctl / vm ioctl 1975 :Type: vcpu ioctl / vm ioctl
1989 :Parameters: none 1976 :Parameters: none
1990 :Returns: virtual tsc-khz on success, negativ 1977 :Returns: virtual tsc-khz on success, negative value on error
1991 1978
1992 Returns the tsc frequency of the guest. The u 1979 Returns the tsc frequency of the guest. The unit of the return value is
1993 KHz. If the host has unstable tsc this ioctl 1980 KHz. If the host has unstable tsc this ioctl returns -EIO instead as an
1994 error. 1981 error.
1995 1982
1996 1983
1997 4.57 KVM_GET_LAPIC 1984 4.57 KVM_GET_LAPIC
1998 ------------------ 1985 ------------------
1999 1986
2000 :Capability: KVM_CAP_IRQCHIP 1987 :Capability: KVM_CAP_IRQCHIP
2001 :Architectures: x86 1988 :Architectures: x86
2002 :Type: vcpu ioctl 1989 :Type: vcpu ioctl
2003 :Parameters: struct kvm_lapic_state (out) 1990 :Parameters: struct kvm_lapic_state (out)
2004 :Returns: 0 on success, -1 on error 1991 :Returns: 0 on success, -1 on error
2005 1992
2006 :: 1993 ::
2007 1994
2008 #define KVM_APIC_REG_SIZE 0x400 1995 #define KVM_APIC_REG_SIZE 0x400
2009 struct kvm_lapic_state { 1996 struct kvm_lapic_state {
2010 char regs[KVM_APIC_REG_SIZE]; 1997 char regs[KVM_APIC_REG_SIZE];
2011 }; 1998 };
2012 1999
2013 Reads the Local APIC registers and copies the 2000 Reads the Local APIC registers and copies them into the input argument. The
2014 data format and layout are the same as docume 2001 data format and layout are the same as documented in the architecture manual.
2015 2002
2016 If KVM_X2APIC_API_USE_32BIT_IDS feature of KV 2003 If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is
2017 enabled, then the format of APIC_ID register 2004 enabled, then the format of APIC_ID register depends on the APIC mode
2018 (reported by MSR_IA32_APICBASE) of its VCPU. 2005 (reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in
2019 the APIC_ID register (bytes 32-35). xAPIC on 2006 the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID
2020 which is stored in bits 31-24 of the APIC reg 2007 which is stored in bits 31-24 of the APIC register, or equivalently in
2021 byte 35 of struct kvm_lapic_state's regs fiel 2008 byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then
2022 be called after MSR_IA32_APICBASE has been se 2009 be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR.
2023 2010
2024 If KVM_X2APIC_API_USE_32BIT_IDS feature is di 2011 If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state
2025 always uses xAPIC format. 2012 always uses xAPIC format.
2026 2013
2027 2014
2028 4.58 KVM_SET_LAPIC 2015 4.58 KVM_SET_LAPIC
2029 ------------------ 2016 ------------------
2030 2017
2031 :Capability: KVM_CAP_IRQCHIP 2018 :Capability: KVM_CAP_IRQCHIP
2032 :Architectures: x86 2019 :Architectures: x86
2033 :Type: vcpu ioctl 2020 :Type: vcpu ioctl
2034 :Parameters: struct kvm_lapic_state (in) 2021 :Parameters: struct kvm_lapic_state (in)
2035 :Returns: 0 on success, -1 on error 2022 :Returns: 0 on success, -1 on error
2036 2023
2037 :: 2024 ::
2038 2025
2039 #define KVM_APIC_REG_SIZE 0x400 2026 #define KVM_APIC_REG_SIZE 0x400
2040 struct kvm_lapic_state { 2027 struct kvm_lapic_state {
2041 char regs[KVM_APIC_REG_SIZE]; 2028 char regs[KVM_APIC_REG_SIZE];
2042 }; 2029 };
2043 2030
2044 Copies the input argument into the Local APIC 2031 Copies the input argument into the Local APIC registers. The data format
2045 and layout are the same as documented in the 2032 and layout are the same as documented in the architecture manual.
2046 2033
2047 The format of the APIC ID register (bytes 32- 2034 The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's
2048 regs field) depends on the state of the KVM_C 2035 regs field) depends on the state of the KVM_CAP_X2APIC_API capability.
2049 See the note in KVM_GET_LAPIC. 2036 See the note in KVM_GET_LAPIC.
2050 2037
2051 2038
2052 4.59 KVM_IOEVENTFD 2039 4.59 KVM_IOEVENTFD
2053 ------------------ 2040 ------------------
2054 2041
2055 :Capability: KVM_CAP_IOEVENTFD 2042 :Capability: KVM_CAP_IOEVENTFD
2056 :Architectures: all 2043 :Architectures: all
2057 :Type: vm ioctl 2044 :Type: vm ioctl
2058 :Parameters: struct kvm_ioeventfd (in) 2045 :Parameters: struct kvm_ioeventfd (in)
2059 :Returns: 0 on success, !0 on error 2046 :Returns: 0 on success, !0 on error
2060 2047
2061 This ioctl attaches or detaches an ioeventfd 2048 This ioctl attaches or detaches an ioeventfd to a legal pio/mmio address
2062 within the guest. A guest write in the regis 2049 within the guest. A guest write in the registered address will signal the
2063 provided event instead of triggering an exit. 2050 provided event instead of triggering an exit.
2064 2051
2065 :: 2052 ::
2066 2053
2067 struct kvm_ioeventfd { 2054 struct kvm_ioeventfd {
2068 __u64 datamatch; 2055 __u64 datamatch;
2069 __u64 addr; /* legal pio/mmio 2056 __u64 addr; /* legal pio/mmio address */
2070 __u32 len; /* 0, 1, 2, 4, or 2057 __u32 len; /* 0, 1, 2, 4, or 8 bytes */
2071 __s32 fd; 2058 __s32 fd;
2072 __u32 flags; 2059 __u32 flags;
2073 __u8 pad[36]; 2060 __u8 pad[36];
2074 }; 2061 };
2075 2062
2076 For the special case of virtio-ccw devices on 2063 For the special case of virtio-ccw devices on s390, the ioevent is matched
2077 to a subchannel/virtqueue tuple instead. 2064 to a subchannel/virtqueue tuple instead.
2078 2065
2079 The following flags are defined:: 2066 The following flags are defined::
2080 2067
2081 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << 2068 #define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
2082 #define KVM_IOEVENTFD_FLAG_PIO (1 << 2069 #define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
2083 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << 2070 #define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
2084 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIF 2071 #define KVM_IOEVENTFD_FLAG_VIRTIO_CCW_NOTIFY \
2085 (1 << kvm_ioeventfd_flag_nr_virtio_cc 2072 (1 << kvm_ioeventfd_flag_nr_virtio_ccw_notify)
2086 2073
2087 If datamatch flag is set, the event will be s 2074 If datamatch flag is set, the event will be signaled only if the written value
2088 to the registered address is equal to datamat 2075 to the registered address is equal to datamatch in struct kvm_ioeventfd.
2089 2076
2090 For virtio-ccw devices, addr contains the sub 2077 For virtio-ccw devices, addr contains the subchannel id and datamatch the
2091 virtqueue index. 2078 virtqueue index.
2092 2079
2093 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero len 2080 With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and
2094 the kernel will ignore the length of guest wr 2081 the kernel will ignore the length of guest write and may get a faster vmexit.
2095 The speedup may only apply to specific archit 2082 The speedup may only apply to specific architectures, but the ioeventfd will
2096 work anyway. 2083 work anyway.
2097 2084
2098 4.60 KVM_DIRTY_TLB 2085 4.60 KVM_DIRTY_TLB
2099 ------------------ 2086 ------------------
2100 2087
2101 :Capability: KVM_CAP_SW_TLB 2088 :Capability: KVM_CAP_SW_TLB
2102 :Architectures: ppc 2089 :Architectures: ppc
2103 :Type: vcpu ioctl 2090 :Type: vcpu ioctl
2104 :Parameters: struct kvm_dirty_tlb (in) 2091 :Parameters: struct kvm_dirty_tlb (in)
2105 :Returns: 0 on success, -1 on error 2092 :Returns: 0 on success, -1 on error
2106 2093
2107 :: 2094 ::
2108 2095
2109 struct kvm_dirty_tlb { 2096 struct kvm_dirty_tlb {
2110 __u64 bitmap; 2097 __u64 bitmap;
2111 __u32 num_dirty; 2098 __u32 num_dirty;
2112 }; 2099 };
2113 2100
2114 This must be called whenever userspace has ch 2101 This must be called whenever userspace has changed an entry in the shared
2115 TLB, prior to calling KVM_RUN on the associat 2102 TLB, prior to calling KVM_RUN on the associated vcpu.
2116 2103
2117 The "bitmap" field is the userspace address o 2104 The "bitmap" field is the userspace address of an array. This array
2118 consists of a number of bits, equal to the to 2105 consists of a number of bits, equal to the total number of TLB entries as
2119 determined by the last successful call to KVM 2106 determined by the last successful call to KVM_CONFIG_TLB, rounded up to the
2120 nearest multiple of 64. 2107 nearest multiple of 64.
2121 2108
2122 Each bit corresponds to one TLB entry, ordere 2109 Each bit corresponds to one TLB entry, ordered the same as in the shared TLB
2123 array. 2110 array.
2124 2111
2125 The array is little-endian: the bit 0 is the 2112 The array is little-endian: the bit 0 is the least significant bit of the
2126 first byte, bit 8 is the least significant bi 2113 first byte, bit 8 is the least significant bit of the second byte, etc.
2127 This avoids any complications with differing 2114 This avoids any complications with differing word sizes.
2128 2115
2129 The "num_dirty" field is a performance hint f 2116 The "num_dirty" field is a performance hint for KVM to determine whether it
2130 should skip processing the bitmap and just in 2117 should skip processing the bitmap and just invalidate everything. It must
2131 be set to the number of set bits in the bitma 2118 be set to the number of set bits in the bitmap.
2132 2119
2133 2120
2134 4.62 KVM_CREATE_SPAPR_TCE 2121 4.62 KVM_CREATE_SPAPR_TCE
2135 ------------------------- 2122 -------------------------
2136 2123
2137 :Capability: KVM_CAP_SPAPR_TCE 2124 :Capability: KVM_CAP_SPAPR_TCE
2138 :Architectures: powerpc 2125 :Architectures: powerpc
2139 :Type: vm ioctl 2126 :Type: vm ioctl
2140 :Parameters: struct kvm_create_spapr_tce (in) 2127 :Parameters: struct kvm_create_spapr_tce (in)
2141 :Returns: file descriptor for manipulating th 2128 :Returns: file descriptor for manipulating the created TCE table
2142 2129
2143 This creates a virtual TCE (translation contr 2130 This creates a virtual TCE (translation control entry) table, which
2144 is an IOMMU for PAPR-style virtual I/O. It i 2131 is an IOMMU for PAPR-style virtual I/O. It is used to translate
2145 logical addresses used in virtual I/O into gu 2132 logical addresses used in virtual I/O into guest physical addresses,
2146 and provides a scatter/gather capability for 2133 and provides a scatter/gather capability for PAPR virtual I/O.
2147 2134
2148 :: 2135 ::
2149 2136
2150 /* for KVM_CAP_SPAPR_TCE */ 2137 /* for KVM_CAP_SPAPR_TCE */
2151 struct kvm_create_spapr_tce { 2138 struct kvm_create_spapr_tce {
2152 __u64 liobn; 2139 __u64 liobn;
2153 __u32 window_size; 2140 __u32 window_size;
2154 }; 2141 };
2155 2142
2156 The liobn field gives the logical IO bus numb 2143 The liobn field gives the logical IO bus number for which to create a
2157 TCE table. The window_size field specifies t 2144 TCE table. The window_size field specifies the size of the DMA window
2158 which this TCE table will translate - the tab 2145 which this TCE table will translate - the table will contain one 64
2159 bit TCE entry for every 4kiB of the DMA windo 2146 bit TCE entry for every 4kiB of the DMA window.
2160 2147
2161 When the guest issues an H_PUT_TCE hcall on a 2148 When the guest issues an H_PUT_TCE hcall on a liobn for which a TCE
2162 table has been created using this ioctl(), th 2149 table has been created using this ioctl(), the kernel will handle it
2163 in real mode, updating the TCE table. H_PUT_ 2150 in real mode, updating the TCE table. H_PUT_TCE calls for other
2164 liobns will cause a vm exit and must be handl 2151 liobns will cause a vm exit and must be handled by userspace.
2165 2152
2166 The return value is a file descriptor which c 2153 The return value is a file descriptor which can be passed to mmap(2)
2167 to map the created TCE table into userspace. 2154 to map the created TCE table into userspace. This lets userspace read
2168 the entries written by kernel-handled H_PUT_T 2155 the entries written by kernel-handled H_PUT_TCE calls, and also lets
2169 userspace update the TCE table directly which 2156 userspace update the TCE table directly which is useful in some
2170 circumstances. 2157 circumstances.
2171 2158
2172 2159
2173 4.63 KVM_ALLOCATE_RMA 2160 4.63 KVM_ALLOCATE_RMA
2174 --------------------- 2161 ---------------------
2175 2162
2176 :Capability: KVM_CAP_PPC_RMA 2163 :Capability: KVM_CAP_PPC_RMA
2177 :Architectures: powerpc 2164 :Architectures: powerpc
2178 :Type: vm ioctl 2165 :Type: vm ioctl
2179 :Parameters: struct kvm_allocate_rma (out) 2166 :Parameters: struct kvm_allocate_rma (out)
2180 :Returns: file descriptor for mapping the all 2167 :Returns: file descriptor for mapping the allocated RMA
2181 2168
2182 This allocates a Real Mode Area (RMA) from th 2169 This allocates a Real Mode Area (RMA) from the pool allocated at boot
2183 time by the kernel. An RMA is a physically-c 2170 time by the kernel. An RMA is a physically-contiguous, aligned region
2184 of memory used on older POWER processors to p 2171 of memory used on older POWER processors to provide the memory which
2185 will be accessed by real-mode (MMU off) acces 2172 will be accessed by real-mode (MMU off) accesses in a KVM guest.
2186 POWER processors support a set of sizes for t 2173 POWER processors support a set of sizes for the RMA that usually
2187 includes 64MB, 128MB, 256MB and some larger p 2174 includes 64MB, 128MB, 256MB and some larger powers of two.
2188 2175
2189 :: 2176 ::
2190 2177
2191 /* for KVM_ALLOCATE_RMA */ 2178 /* for KVM_ALLOCATE_RMA */
2192 struct kvm_allocate_rma { 2179 struct kvm_allocate_rma {
2193 __u64 rma_size; 2180 __u64 rma_size;
2194 }; 2181 };
2195 2182
2196 The return value is a file descriptor which c 2183 The return value is a file descriptor which can be passed to mmap(2)
2197 to map the allocated RMA into userspace. The 2184 to map the allocated RMA into userspace. The mapped area can then be
2198 passed to the KVM_SET_USER_MEMORY_REGION ioct 2185 passed to the KVM_SET_USER_MEMORY_REGION ioctl to establish it as the
2199 RMA for a virtual machine. The size of the R 2186 RMA for a virtual machine. The size of the RMA in bytes (which is
2200 fixed at host kernel boot time) is returned i 2187 fixed at host kernel boot time) is returned in the rma_size field of
2201 the argument structure. 2188 the argument structure.
2202 2189
2203 The KVM_CAP_PPC_RMA capability is 1 or 2 if t 2190 The KVM_CAP_PPC_RMA capability is 1 or 2 if the KVM_ALLOCATE_RMA ioctl
2204 is supported; 2 if the processor requires all 2191 is supported; 2 if the processor requires all virtual machines to have
2205 an RMA, or 1 if the processor can use an RMA 2192 an RMA, or 1 if the processor can use an RMA but doesn't require it,
2206 because it supports the Virtual RMA (VRMA) fa 2193 because it supports the Virtual RMA (VRMA) facility.
2207 2194
2208 2195
2209 4.64 KVM_NMI 2196 4.64 KVM_NMI
2210 ------------ 2197 ------------
2211 2198
2212 :Capability: KVM_CAP_USER_NMI 2199 :Capability: KVM_CAP_USER_NMI
2213 :Architectures: x86 2200 :Architectures: x86
2214 :Type: vcpu ioctl 2201 :Type: vcpu ioctl
2215 :Parameters: none 2202 :Parameters: none
2216 :Returns: 0 on success, -1 on error 2203 :Returns: 0 on success, -1 on error
2217 2204
2218 Queues an NMI on the thread's vcpu. Note thi 2205 Queues an NMI on the thread's vcpu. Note this is well defined only
2219 when KVM_CREATE_IRQCHIP has not been called, 2206 when KVM_CREATE_IRQCHIP has not been called, since this is an interface
2220 between the virtual cpu core and virtual loca 2207 between the virtual cpu core and virtual local APIC. After KVM_CREATE_IRQCHIP
2221 has been called, this interface is completely 2208 has been called, this interface is completely emulated within the kernel.
2222 2209
2223 To use this to emulate the LINT1 input with K 2210 To use this to emulate the LINT1 input with KVM_CREATE_IRQCHIP, use the
2224 following algorithm: 2211 following algorithm:
2225 2212
2226 - pause the vcpu 2213 - pause the vcpu
2227 - read the local APIC's state (KVM_GET_LAPI 2214 - read the local APIC's state (KVM_GET_LAPIC)
2228 - check whether changing LINT1 will queue a 2215 - check whether changing LINT1 will queue an NMI (see the LVT entry for LINT1)
2229 - if so, issue KVM_NMI 2216 - if so, issue KVM_NMI
2230 - resume the vcpu 2217 - resume the vcpu
2231 2218
2232 Some guests configure the LINT1 NMI input to 2219 Some guests configure the LINT1 NMI input to cause a panic, aiding in
2233 debugging. 2220 debugging.
2234 2221
2235 2222
2236 4.65 KVM_S390_UCAS_MAP 2223 4.65 KVM_S390_UCAS_MAP
2237 ---------------------- 2224 ----------------------
2238 2225
2239 :Capability: KVM_CAP_S390_UCONTROL 2226 :Capability: KVM_CAP_S390_UCONTROL
2240 :Architectures: s390 2227 :Architectures: s390
2241 :Type: vcpu ioctl 2228 :Type: vcpu ioctl
2242 :Parameters: struct kvm_s390_ucas_mapping (in 2229 :Parameters: struct kvm_s390_ucas_mapping (in)
2243 :Returns: 0 in case of success 2230 :Returns: 0 in case of success
2244 2231
2245 The parameter is defined like this:: 2232 The parameter is defined like this::
2246 2233
2247 struct kvm_s390_ucas_mapping { 2234 struct kvm_s390_ucas_mapping {
2248 __u64 user_addr; 2235 __u64 user_addr;
2249 __u64 vcpu_addr; 2236 __u64 vcpu_addr;
2250 __u64 length; 2237 __u64 length;
2251 }; 2238 };
2252 2239
2253 This ioctl maps the memory at "user_addr" wit 2240 This ioctl maps the memory at "user_addr" with the length "length" to
2254 the vcpu's address space starting at "vcpu_ad 2241 the vcpu's address space starting at "vcpu_addr". All parameters need to
2255 be aligned by 1 megabyte. 2242 be aligned by 1 megabyte.
2256 2243
2257 2244
2258 4.66 KVM_S390_UCAS_UNMAP 2245 4.66 KVM_S390_UCAS_UNMAP
2259 ------------------------ 2246 ------------------------
2260 2247
2261 :Capability: KVM_CAP_S390_UCONTROL 2248 :Capability: KVM_CAP_S390_UCONTROL
2262 :Architectures: s390 2249 :Architectures: s390
2263 :Type: vcpu ioctl 2250 :Type: vcpu ioctl
2264 :Parameters: struct kvm_s390_ucas_mapping (in 2251 :Parameters: struct kvm_s390_ucas_mapping (in)
2265 :Returns: 0 in case of success 2252 :Returns: 0 in case of success
2266 2253
2267 The parameter is defined like this:: 2254 The parameter is defined like this::
2268 2255
2269 struct kvm_s390_ucas_mapping { 2256 struct kvm_s390_ucas_mapping {
2270 __u64 user_addr; 2257 __u64 user_addr;
2271 __u64 vcpu_addr; 2258 __u64 vcpu_addr;
2272 __u64 length; 2259 __u64 length;
2273 }; 2260 };
2274 2261
2275 This ioctl unmaps the memory in the vcpu's ad 2262 This ioctl unmaps the memory in the vcpu's address space starting at
2276 "vcpu_addr" with the length "length". The fie 2263 "vcpu_addr" with the length "length". The field "user_addr" is ignored.
2277 All parameters need to be aligned by 1 megaby 2264 All parameters need to be aligned by 1 megabyte.
2278 2265
2279 2266
2280 4.67 KVM_S390_VCPU_FAULT 2267 4.67 KVM_S390_VCPU_FAULT
2281 ------------------------ 2268 ------------------------
2282 2269
2283 :Capability: KVM_CAP_S390_UCONTROL 2270 :Capability: KVM_CAP_S390_UCONTROL
2284 :Architectures: s390 2271 :Architectures: s390
2285 :Type: vcpu ioctl 2272 :Type: vcpu ioctl
2286 :Parameters: vcpu absolute address (in) 2273 :Parameters: vcpu absolute address (in)
2287 :Returns: 0 in case of success 2274 :Returns: 0 in case of success
2288 2275
2289 This call creates a page table entry on the v 2276 This call creates a page table entry on the virtual cpu's address space
2290 (for user controlled virtual machines) or the 2277 (for user controlled virtual machines) or the virtual machine's address
2291 space (for regular virtual machines). This on 2278 space (for regular virtual machines). This only works for minor faults,
2292 thus it's recommended to access subject memor 2279 thus it's recommended to access subject memory page via the user page
2293 table upfront. This is useful to handle valid 2280 table upfront. This is useful to handle validity intercepts for user
2294 controlled virtual machines to fault in the v 2281 controlled virtual machines to fault in the virtual cpu's lowcore pages
2295 prior to calling the KVM_RUN ioctl. 2282 prior to calling the KVM_RUN ioctl.
2296 2283
2297 2284
2298 4.68 KVM_SET_ONE_REG 2285 4.68 KVM_SET_ONE_REG
2299 -------------------- 2286 --------------------
2300 2287
2301 :Capability: KVM_CAP_ONE_REG 2288 :Capability: KVM_CAP_ONE_REG
2302 :Architectures: all 2289 :Architectures: all
2303 :Type: vcpu ioctl 2290 :Type: vcpu ioctl
2304 :Parameters: struct kvm_one_reg (in) 2291 :Parameters: struct kvm_one_reg (in)
2305 :Returns: 0 on success, negative value on fai 2292 :Returns: 0 on success, negative value on failure
2306 2293
2307 Errors: 2294 Errors:
2308 2295
2309 ====== ================================== 2296 ====== ============================================================
2310 ENOENT no such register 2297 ENOENT no such register
2311 EINVAL invalid register ID, or no such re 2298 EINVAL invalid register ID, or no such register or used with VMs in
2312 protected virtualization mode on s 2299 protected virtualization mode on s390
2313 EPERM (arm64) register access not allowe 2300 EPERM (arm64) register access not allowed before vcpu finalization
2314 EBUSY (riscv) changing register value no 2301 EBUSY (riscv) changing register value not allowed after the vcpu
2315 has run at least once 2302 has run at least once
2316 ====== ================================== 2303 ====== ============================================================
2317 2304
2318 (These error codes are indicative only: do no 2305 (These error codes are indicative only: do not rely on a specific error
2319 code being returned in a specific situation.) 2306 code being returned in a specific situation.)
2320 2307
2321 :: 2308 ::
2322 2309
2323 struct kvm_one_reg { 2310 struct kvm_one_reg {
2324 __u64 id; 2311 __u64 id;
2325 __u64 addr; 2312 __u64 addr;
2326 }; 2313 };
2327 2314
2328 Using this ioctl, a single vcpu register can 2315 Using this ioctl, a single vcpu register can be set to a specific value
2329 defined by user space with the passed in stru 2316 defined by user space with the passed in struct kvm_one_reg, where id
2330 refers to the register identifier as describe 2317 refers to the register identifier as described below and addr is a pointer
2331 to a variable with the respective size. There 2318 to a variable with the respective size. There can be architecture agnostic
2332 and architecture specific registers. Each hav 2319 and architecture specific registers. Each have their own range of operation
2333 and their own constants and width. To keep tr 2320 and their own constants and width. To keep track of the implemented
2334 registers, find a list below: 2321 registers, find a list below:
2335 2322
2336 ======= =============================== === 2323 ======= =============================== ============
2337 Arch Register Wid 2324 Arch Register Width (bits)
2338 ======= =============================== === 2325 ======= =============================== ============
2339 PPC KVM_REG_PPC_HIOR 64 2326 PPC KVM_REG_PPC_HIOR 64
2340 PPC KVM_REG_PPC_IAC1 64 2327 PPC KVM_REG_PPC_IAC1 64
2341 PPC KVM_REG_PPC_IAC2 64 2328 PPC KVM_REG_PPC_IAC2 64
2342 PPC KVM_REG_PPC_IAC3 64 2329 PPC KVM_REG_PPC_IAC3 64
2343 PPC KVM_REG_PPC_IAC4 64 2330 PPC KVM_REG_PPC_IAC4 64
2344 PPC KVM_REG_PPC_DAC1 64 2331 PPC KVM_REG_PPC_DAC1 64
2345 PPC KVM_REG_PPC_DAC2 64 2332 PPC KVM_REG_PPC_DAC2 64
2346 PPC KVM_REG_PPC_DABR 64 2333 PPC KVM_REG_PPC_DABR 64
2347 PPC KVM_REG_PPC_DSCR 64 2334 PPC KVM_REG_PPC_DSCR 64
2348 PPC KVM_REG_PPC_PURR 64 2335 PPC KVM_REG_PPC_PURR 64
2349 PPC KVM_REG_PPC_SPURR 64 2336 PPC KVM_REG_PPC_SPURR 64
2350 PPC KVM_REG_PPC_DAR 64 2337 PPC KVM_REG_PPC_DAR 64
2351 PPC KVM_REG_PPC_DSISR 32 2338 PPC KVM_REG_PPC_DSISR 32
2352 PPC KVM_REG_PPC_AMR 64 2339 PPC KVM_REG_PPC_AMR 64
2353 PPC KVM_REG_PPC_UAMOR 64 2340 PPC KVM_REG_PPC_UAMOR 64
2354 PPC KVM_REG_PPC_MMCR0 64 2341 PPC KVM_REG_PPC_MMCR0 64
2355 PPC KVM_REG_PPC_MMCR1 64 2342 PPC KVM_REG_PPC_MMCR1 64
2356 PPC KVM_REG_PPC_MMCRA 64 2343 PPC KVM_REG_PPC_MMCRA 64
2357 PPC KVM_REG_PPC_MMCR2 64 2344 PPC KVM_REG_PPC_MMCR2 64
2358 PPC KVM_REG_PPC_MMCRS 64 2345 PPC KVM_REG_PPC_MMCRS 64
2359 PPC KVM_REG_PPC_MMCR3 64 2346 PPC KVM_REG_PPC_MMCR3 64
2360 PPC KVM_REG_PPC_SIAR 64 2347 PPC KVM_REG_PPC_SIAR 64
2361 PPC KVM_REG_PPC_SDAR 64 2348 PPC KVM_REG_PPC_SDAR 64
2362 PPC KVM_REG_PPC_SIER 64 2349 PPC KVM_REG_PPC_SIER 64
2363 PPC KVM_REG_PPC_SIER2 64 2350 PPC KVM_REG_PPC_SIER2 64
2364 PPC KVM_REG_PPC_SIER3 64 2351 PPC KVM_REG_PPC_SIER3 64
2365 PPC KVM_REG_PPC_PMC1 32 2352 PPC KVM_REG_PPC_PMC1 32
2366 PPC KVM_REG_PPC_PMC2 32 2353 PPC KVM_REG_PPC_PMC2 32
2367 PPC KVM_REG_PPC_PMC3 32 2354 PPC KVM_REG_PPC_PMC3 32
2368 PPC KVM_REG_PPC_PMC4 32 2355 PPC KVM_REG_PPC_PMC4 32
2369 PPC KVM_REG_PPC_PMC5 32 2356 PPC KVM_REG_PPC_PMC5 32
2370 PPC KVM_REG_PPC_PMC6 32 2357 PPC KVM_REG_PPC_PMC6 32
2371 PPC KVM_REG_PPC_PMC7 32 2358 PPC KVM_REG_PPC_PMC7 32
2372 PPC KVM_REG_PPC_PMC8 32 2359 PPC KVM_REG_PPC_PMC8 32
2373 PPC KVM_REG_PPC_FPR0 64 2360 PPC KVM_REG_PPC_FPR0 64
2374 ... 2361 ...
2375 PPC KVM_REG_PPC_FPR31 64 2362 PPC KVM_REG_PPC_FPR31 64
2376 PPC KVM_REG_PPC_VR0 128 2363 PPC KVM_REG_PPC_VR0 128
2377 ... 2364 ...
2378 PPC KVM_REG_PPC_VR31 128 2365 PPC KVM_REG_PPC_VR31 128
2379 PPC KVM_REG_PPC_VSR0 128 2366 PPC KVM_REG_PPC_VSR0 128
2380 ... 2367 ...
2381 PPC KVM_REG_PPC_VSR31 128 2368 PPC KVM_REG_PPC_VSR31 128
2382 PPC KVM_REG_PPC_FPSCR 64 2369 PPC KVM_REG_PPC_FPSCR 64
2383 PPC KVM_REG_PPC_VSCR 32 2370 PPC KVM_REG_PPC_VSCR 32
2384 PPC KVM_REG_PPC_VPA_ADDR 64 2371 PPC KVM_REG_PPC_VPA_ADDR 64
2385 PPC KVM_REG_PPC_VPA_SLB 128 2372 PPC KVM_REG_PPC_VPA_SLB 128
2386 PPC KVM_REG_PPC_VPA_DTL 128 2373 PPC KVM_REG_PPC_VPA_DTL 128
2387 PPC KVM_REG_PPC_EPCR 32 2374 PPC KVM_REG_PPC_EPCR 32
2388 PPC KVM_REG_PPC_EPR 32 2375 PPC KVM_REG_PPC_EPR 32
2389 PPC KVM_REG_PPC_TCR 32 2376 PPC KVM_REG_PPC_TCR 32
2390 PPC KVM_REG_PPC_TSR 32 2377 PPC KVM_REG_PPC_TSR 32
2391 PPC KVM_REG_PPC_OR_TSR 32 2378 PPC KVM_REG_PPC_OR_TSR 32
2392 PPC KVM_REG_PPC_CLEAR_TSR 32 2379 PPC KVM_REG_PPC_CLEAR_TSR 32
2393 PPC KVM_REG_PPC_MAS0 32 2380 PPC KVM_REG_PPC_MAS0 32
2394 PPC KVM_REG_PPC_MAS1 32 2381 PPC KVM_REG_PPC_MAS1 32
2395 PPC KVM_REG_PPC_MAS2 64 2382 PPC KVM_REG_PPC_MAS2 64
2396 PPC KVM_REG_PPC_MAS7_3 64 2383 PPC KVM_REG_PPC_MAS7_3 64
2397 PPC KVM_REG_PPC_MAS4 32 2384 PPC KVM_REG_PPC_MAS4 32
2398 PPC KVM_REG_PPC_MAS6 32 2385 PPC KVM_REG_PPC_MAS6 32
2399 PPC KVM_REG_PPC_MMUCFG 32 2386 PPC KVM_REG_PPC_MMUCFG 32
2400 PPC KVM_REG_PPC_TLB0CFG 32 2387 PPC KVM_REG_PPC_TLB0CFG 32
2401 PPC KVM_REG_PPC_TLB1CFG 32 2388 PPC KVM_REG_PPC_TLB1CFG 32
2402 PPC KVM_REG_PPC_TLB2CFG 32 2389 PPC KVM_REG_PPC_TLB2CFG 32
2403 PPC KVM_REG_PPC_TLB3CFG 32 2390 PPC KVM_REG_PPC_TLB3CFG 32
2404 PPC KVM_REG_PPC_TLB0PS 32 2391 PPC KVM_REG_PPC_TLB0PS 32
2405 PPC KVM_REG_PPC_TLB1PS 32 2392 PPC KVM_REG_PPC_TLB1PS 32
2406 PPC KVM_REG_PPC_TLB2PS 32 2393 PPC KVM_REG_PPC_TLB2PS 32
2407 PPC KVM_REG_PPC_TLB3PS 32 2394 PPC KVM_REG_PPC_TLB3PS 32
2408 PPC KVM_REG_PPC_EPTCFG 32 2395 PPC KVM_REG_PPC_EPTCFG 32
2409 PPC KVM_REG_PPC_ICP_STATE 64 2396 PPC KVM_REG_PPC_ICP_STATE 64
2410 PPC KVM_REG_PPC_VP_STATE 128 2397 PPC KVM_REG_PPC_VP_STATE 128
2411 PPC KVM_REG_PPC_TB_OFFSET 64 2398 PPC KVM_REG_PPC_TB_OFFSET 64
2412 PPC KVM_REG_PPC_SPMC1 32 2399 PPC KVM_REG_PPC_SPMC1 32
2413 PPC KVM_REG_PPC_SPMC2 32 2400 PPC KVM_REG_PPC_SPMC2 32
2414 PPC KVM_REG_PPC_IAMR 64 2401 PPC KVM_REG_PPC_IAMR 64
2415 PPC KVM_REG_PPC_TFHAR 64 2402 PPC KVM_REG_PPC_TFHAR 64
2416 PPC KVM_REG_PPC_TFIAR 64 2403 PPC KVM_REG_PPC_TFIAR 64
2417 PPC KVM_REG_PPC_TEXASR 64 2404 PPC KVM_REG_PPC_TEXASR 64
2418 PPC KVM_REG_PPC_FSCR 64 2405 PPC KVM_REG_PPC_FSCR 64
2419 PPC KVM_REG_PPC_PSPB 32 2406 PPC KVM_REG_PPC_PSPB 32
2420 PPC KVM_REG_PPC_EBBHR 64 2407 PPC KVM_REG_PPC_EBBHR 64
2421 PPC KVM_REG_PPC_EBBRR 64 2408 PPC KVM_REG_PPC_EBBRR 64
2422 PPC KVM_REG_PPC_BESCR 64 2409 PPC KVM_REG_PPC_BESCR 64
2423 PPC KVM_REG_PPC_TAR 64 2410 PPC KVM_REG_PPC_TAR 64
2424 PPC KVM_REG_PPC_DPDES 64 2411 PPC KVM_REG_PPC_DPDES 64
2425 PPC KVM_REG_PPC_DAWR 64 2412 PPC KVM_REG_PPC_DAWR 64
2426 PPC KVM_REG_PPC_DAWRX 64 2413 PPC KVM_REG_PPC_DAWRX 64
2427 PPC KVM_REG_PPC_CIABR 64 2414 PPC KVM_REG_PPC_CIABR 64
2428 PPC KVM_REG_PPC_IC 64 2415 PPC KVM_REG_PPC_IC 64
2429 PPC KVM_REG_PPC_VTB 64 2416 PPC KVM_REG_PPC_VTB 64
2430 PPC KVM_REG_PPC_CSIGR 64 2417 PPC KVM_REG_PPC_CSIGR 64
2431 PPC KVM_REG_PPC_TACR 64 2418 PPC KVM_REG_PPC_TACR 64
2432 PPC KVM_REG_PPC_TCSCR 64 2419 PPC KVM_REG_PPC_TCSCR 64
2433 PPC KVM_REG_PPC_PID 64 2420 PPC KVM_REG_PPC_PID 64
2434 PPC KVM_REG_PPC_ACOP 64 2421 PPC KVM_REG_PPC_ACOP 64
2435 PPC KVM_REG_PPC_VRSAVE 32 2422 PPC KVM_REG_PPC_VRSAVE 32
2436 PPC KVM_REG_PPC_LPCR 32 2423 PPC KVM_REG_PPC_LPCR 32
2437 PPC KVM_REG_PPC_LPCR_64 64 2424 PPC KVM_REG_PPC_LPCR_64 64
2438 PPC KVM_REG_PPC_PPR 64 2425 PPC KVM_REG_PPC_PPR 64
2439 PPC KVM_REG_PPC_ARCH_COMPAT 32 2426 PPC KVM_REG_PPC_ARCH_COMPAT 32
2440 PPC KVM_REG_PPC_DABRX 32 2427 PPC KVM_REG_PPC_DABRX 32
2441 PPC KVM_REG_PPC_WORT 64 2428 PPC KVM_REG_PPC_WORT 64
2442 PPC KVM_REG_PPC_SPRG9 64 2429 PPC KVM_REG_PPC_SPRG9 64
2443 PPC KVM_REG_PPC_DBSR 32 2430 PPC KVM_REG_PPC_DBSR 32
2444 PPC KVM_REG_PPC_TIDR 64 2431 PPC KVM_REG_PPC_TIDR 64
2445 PPC KVM_REG_PPC_PSSCR 64 2432 PPC KVM_REG_PPC_PSSCR 64
2446 PPC KVM_REG_PPC_DEC_EXPIRY 64 2433 PPC KVM_REG_PPC_DEC_EXPIRY 64
2447 PPC KVM_REG_PPC_PTCR 64 2434 PPC KVM_REG_PPC_PTCR 64
2448 PPC KVM_REG_PPC_HASHKEYR 64 <<
2449 PPC KVM_REG_PPC_HASHPKEYR 64 <<
2450 PPC KVM_REG_PPC_DAWR1 64 2435 PPC KVM_REG_PPC_DAWR1 64
2451 PPC KVM_REG_PPC_DAWRX1 64 2436 PPC KVM_REG_PPC_DAWRX1 64
2452 PPC KVM_REG_PPC_DEXCR 64 <<
2453 PPC KVM_REG_PPC_TM_GPR0 64 2437 PPC KVM_REG_PPC_TM_GPR0 64
2454 ... 2438 ...
2455 PPC KVM_REG_PPC_TM_GPR31 64 2439 PPC KVM_REG_PPC_TM_GPR31 64
2456 PPC KVM_REG_PPC_TM_VSR0 128 2440 PPC KVM_REG_PPC_TM_VSR0 128
2457 ... 2441 ...
2458 PPC KVM_REG_PPC_TM_VSR63 128 2442 PPC KVM_REG_PPC_TM_VSR63 128
2459 PPC KVM_REG_PPC_TM_CR 64 2443 PPC KVM_REG_PPC_TM_CR 64
2460 PPC KVM_REG_PPC_TM_LR 64 2444 PPC KVM_REG_PPC_TM_LR 64
2461 PPC KVM_REG_PPC_TM_CTR 64 2445 PPC KVM_REG_PPC_TM_CTR 64
2462 PPC KVM_REG_PPC_TM_FPSCR 64 2446 PPC KVM_REG_PPC_TM_FPSCR 64
2463 PPC KVM_REG_PPC_TM_AMR 64 2447 PPC KVM_REG_PPC_TM_AMR 64
2464 PPC KVM_REG_PPC_TM_PPR 64 2448 PPC KVM_REG_PPC_TM_PPR 64
2465 PPC KVM_REG_PPC_TM_VRSAVE 64 2449 PPC KVM_REG_PPC_TM_VRSAVE 64
2466 PPC KVM_REG_PPC_TM_VSCR 32 2450 PPC KVM_REG_PPC_TM_VSCR 32
2467 PPC KVM_REG_PPC_TM_DSCR 64 2451 PPC KVM_REG_PPC_TM_DSCR 64
2468 PPC KVM_REG_PPC_TM_TAR 64 2452 PPC KVM_REG_PPC_TM_TAR 64
2469 PPC KVM_REG_PPC_TM_XER 64 2453 PPC KVM_REG_PPC_TM_XER 64
2470 2454
2471 MIPS KVM_REG_MIPS_R0 64 2455 MIPS KVM_REG_MIPS_R0 64
2472 ... 2456 ...
2473 MIPS KVM_REG_MIPS_R31 64 2457 MIPS KVM_REG_MIPS_R31 64
2474 MIPS KVM_REG_MIPS_HI 64 2458 MIPS KVM_REG_MIPS_HI 64
2475 MIPS KVM_REG_MIPS_LO 64 2459 MIPS KVM_REG_MIPS_LO 64
2476 MIPS KVM_REG_MIPS_PC 64 2460 MIPS KVM_REG_MIPS_PC 64
2477 MIPS KVM_REG_MIPS_CP0_INDEX 32 2461 MIPS KVM_REG_MIPS_CP0_INDEX 32
2478 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64 2462 MIPS KVM_REG_MIPS_CP0_ENTRYLO0 64
2479 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64 2463 MIPS KVM_REG_MIPS_CP0_ENTRYLO1 64
2480 MIPS KVM_REG_MIPS_CP0_CONTEXT 64 2464 MIPS KVM_REG_MIPS_CP0_CONTEXT 64
2481 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32 2465 MIPS KVM_REG_MIPS_CP0_CONTEXTCONFIG 32
2482 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64 2466 MIPS KVM_REG_MIPS_CP0_USERLOCAL 64
2483 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64 2467 MIPS KVM_REG_MIPS_CP0_XCONTEXTCONFIG 64
2484 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32 2468 MIPS KVM_REG_MIPS_CP0_PAGEMASK 32
2485 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32 2469 MIPS KVM_REG_MIPS_CP0_PAGEGRAIN 32
2486 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64 2470 MIPS KVM_REG_MIPS_CP0_SEGCTL0 64
2487 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64 2471 MIPS KVM_REG_MIPS_CP0_SEGCTL1 64
2488 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64 2472 MIPS KVM_REG_MIPS_CP0_SEGCTL2 64
2489 MIPS KVM_REG_MIPS_CP0_PWBASE 64 2473 MIPS KVM_REG_MIPS_CP0_PWBASE 64
2490 MIPS KVM_REG_MIPS_CP0_PWFIELD 64 2474 MIPS KVM_REG_MIPS_CP0_PWFIELD 64
2491 MIPS KVM_REG_MIPS_CP0_PWSIZE 64 2475 MIPS KVM_REG_MIPS_CP0_PWSIZE 64
2492 MIPS KVM_REG_MIPS_CP0_WIRED 32 2476 MIPS KVM_REG_MIPS_CP0_WIRED 32
2493 MIPS KVM_REG_MIPS_CP0_PWCTL 32 2477 MIPS KVM_REG_MIPS_CP0_PWCTL 32
2494 MIPS KVM_REG_MIPS_CP0_HWRENA 32 2478 MIPS KVM_REG_MIPS_CP0_HWRENA 32
2495 MIPS KVM_REG_MIPS_CP0_BADVADDR 64 2479 MIPS KVM_REG_MIPS_CP0_BADVADDR 64
2496 MIPS KVM_REG_MIPS_CP0_BADINSTR 32 2480 MIPS KVM_REG_MIPS_CP0_BADINSTR 32
2497 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32 2481 MIPS KVM_REG_MIPS_CP0_BADINSTRP 32
2498 MIPS KVM_REG_MIPS_CP0_COUNT 32 2482 MIPS KVM_REG_MIPS_CP0_COUNT 32
2499 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64 2483 MIPS KVM_REG_MIPS_CP0_ENTRYHI 64
2500 MIPS KVM_REG_MIPS_CP0_COMPARE 32 2484 MIPS KVM_REG_MIPS_CP0_COMPARE 32
2501 MIPS KVM_REG_MIPS_CP0_STATUS 32 2485 MIPS KVM_REG_MIPS_CP0_STATUS 32
2502 MIPS KVM_REG_MIPS_CP0_INTCTL 32 2486 MIPS KVM_REG_MIPS_CP0_INTCTL 32
2503 MIPS KVM_REG_MIPS_CP0_CAUSE 32 2487 MIPS KVM_REG_MIPS_CP0_CAUSE 32
2504 MIPS KVM_REG_MIPS_CP0_EPC 64 2488 MIPS KVM_REG_MIPS_CP0_EPC 64
2505 MIPS KVM_REG_MIPS_CP0_PRID 32 2489 MIPS KVM_REG_MIPS_CP0_PRID 32
2506 MIPS KVM_REG_MIPS_CP0_EBASE 64 2490 MIPS KVM_REG_MIPS_CP0_EBASE 64
2507 MIPS KVM_REG_MIPS_CP0_CONFIG 32 2491 MIPS KVM_REG_MIPS_CP0_CONFIG 32
2508 MIPS KVM_REG_MIPS_CP0_CONFIG1 32 2492 MIPS KVM_REG_MIPS_CP0_CONFIG1 32
2509 MIPS KVM_REG_MIPS_CP0_CONFIG2 32 2493 MIPS KVM_REG_MIPS_CP0_CONFIG2 32
2510 MIPS KVM_REG_MIPS_CP0_CONFIG3 32 2494 MIPS KVM_REG_MIPS_CP0_CONFIG3 32
2511 MIPS KVM_REG_MIPS_CP0_CONFIG4 32 2495 MIPS KVM_REG_MIPS_CP0_CONFIG4 32
2512 MIPS KVM_REG_MIPS_CP0_CONFIG5 32 2496 MIPS KVM_REG_MIPS_CP0_CONFIG5 32
2513 MIPS KVM_REG_MIPS_CP0_CONFIG7 32 2497 MIPS KVM_REG_MIPS_CP0_CONFIG7 32
2514 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64 2498 MIPS KVM_REG_MIPS_CP0_XCONTEXT 64
2515 MIPS KVM_REG_MIPS_CP0_ERROREPC 64 2499 MIPS KVM_REG_MIPS_CP0_ERROREPC 64
2516 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64 2500 MIPS KVM_REG_MIPS_CP0_KSCRATCH1 64
2517 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64 2501 MIPS KVM_REG_MIPS_CP0_KSCRATCH2 64
2518 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64 2502 MIPS KVM_REG_MIPS_CP0_KSCRATCH3 64
2519 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64 2503 MIPS KVM_REG_MIPS_CP0_KSCRATCH4 64
2520 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64 2504 MIPS KVM_REG_MIPS_CP0_KSCRATCH5 64
2521 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64 2505 MIPS KVM_REG_MIPS_CP0_KSCRATCH6 64
2522 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64 2506 MIPS KVM_REG_MIPS_CP0_MAAR(0..63) 64
2523 MIPS KVM_REG_MIPS_COUNT_CTL 64 2507 MIPS KVM_REG_MIPS_COUNT_CTL 64
2524 MIPS KVM_REG_MIPS_COUNT_RESUME 64 2508 MIPS KVM_REG_MIPS_COUNT_RESUME 64
2525 MIPS KVM_REG_MIPS_COUNT_HZ 64 2509 MIPS KVM_REG_MIPS_COUNT_HZ 64
2526 MIPS KVM_REG_MIPS_FPR_32(0..31) 32 2510 MIPS KVM_REG_MIPS_FPR_32(0..31) 32
2527 MIPS KVM_REG_MIPS_FPR_64(0..31) 64 2511 MIPS KVM_REG_MIPS_FPR_64(0..31) 64
2528 MIPS KVM_REG_MIPS_VEC_128(0..31) 128 2512 MIPS KVM_REG_MIPS_VEC_128(0..31) 128
2529 MIPS KVM_REG_MIPS_FCR_IR 32 2513 MIPS KVM_REG_MIPS_FCR_IR 32
2530 MIPS KVM_REG_MIPS_FCR_CSR 32 2514 MIPS KVM_REG_MIPS_FCR_CSR 32
2531 MIPS KVM_REG_MIPS_MSA_IR 32 2515 MIPS KVM_REG_MIPS_MSA_IR 32
2532 MIPS KVM_REG_MIPS_MSA_CSR 32 2516 MIPS KVM_REG_MIPS_MSA_CSR 32
2533 ======= =============================== === 2517 ======= =============================== ============
2534 2518
2535 ARM registers are mapped using the lower 32 b 2519 ARM registers are mapped using the lower 32 bits. The upper 16 of that
2536 is the register group type, or coprocessor nu 2520 is the register group type, or coprocessor number:
2537 2521
2538 ARM core registers have the following id bit 2522 ARM core registers have the following id bit patterns::
2539 2523
2540 0x4020 0000 0010 <index into the kvm_regs s 2524 0x4020 0000 0010 <index into the kvm_regs struct:16>
2541 2525
2542 ARM 32-bit CP15 registers have the following 2526 ARM 32-bit CP15 registers have the following id bit patterns::
2543 2527
2544 0x4020 0000 000F <zero:1> <crn:4> <crm:4> < 2528 0x4020 0000 000F <zero:1> <crn:4> <crm:4> <opc1:4> <opc2:3>
2545 2529
2546 ARM 64-bit CP15 registers have the following 2530 ARM 64-bit CP15 registers have the following id bit patterns::
2547 2531
2548 0x4030 0000 000F <zero:1> <zero:4> <crm:4> 2532 0x4030 0000 000F <zero:1> <zero:4> <crm:4> <opc1:4> <zero:3>
2549 2533
2550 ARM CCSIDR registers are demultiplexed by CSS 2534 ARM CCSIDR registers are demultiplexed by CSSELR value::
2551 2535
2552 0x4020 0000 0011 00 <csselr:8> 2536 0x4020 0000 0011 00 <csselr:8>
2553 2537
2554 ARM 32-bit VFP control registers have the fol 2538 ARM 32-bit VFP control registers have the following id bit patterns::
2555 2539
2556 0x4020 0000 0012 1 <regno:12> 2540 0x4020 0000 0012 1 <regno:12>
2557 2541
2558 ARM 64-bit FP registers have the following id 2542 ARM 64-bit FP registers have the following id bit patterns::
2559 2543
2560 0x4030 0000 0012 0 <regno:12> 2544 0x4030 0000 0012 0 <regno:12>
2561 2545
2562 ARM firmware pseudo-registers have the follow 2546 ARM firmware pseudo-registers have the following bit pattern::
2563 2547
2564 0x4030 0000 0014 <regno:16> 2548 0x4030 0000 0014 <regno:16>
2565 2549
2566 2550
2567 arm64 registers are mapped using the lower 32 2551 arm64 registers are mapped using the lower 32 bits. The upper 16 of
2568 that is the register group type, or coprocess 2552 that is the register group type, or coprocessor number:
2569 2553
2570 arm64 core/FP-SIMD registers have the followi 2554 arm64 core/FP-SIMD registers have the following id bit patterns. Note
2571 that the size of the access is variable, as t 2555 that the size of the access is variable, as the kvm_regs structure
2572 contains elements ranging from 32 to 128 bits 2556 contains elements ranging from 32 to 128 bits. The index is a 32bit
2573 value in the kvm_regs structure seen as a 32b 2557 value in the kvm_regs structure seen as a 32bit array::
2574 2558
2575 0x60x0 0000 0010 <index into the kvm_regs s 2559 0x60x0 0000 0010 <index into the kvm_regs struct:16>
2576 2560
2577 Specifically: 2561 Specifically:
2578 2562
2579 ======================= ========= ===== ===== 2563 ======================= ========= ===== =======================================
2580 Encoding Register Bits kvm_r 2564 Encoding Register Bits kvm_regs member
2581 ======================= ========= ===== ===== 2565 ======================= ========= ===== =======================================
2582 0x6030 0000 0010 0000 X0 64 regs. 2566 0x6030 0000 0010 0000 X0 64 regs.regs[0]
2583 0x6030 0000 0010 0002 X1 64 regs. 2567 0x6030 0000 0010 0002 X1 64 regs.regs[1]
2584 ... 2568 ...
2585 0x6030 0000 0010 003c X30 64 regs. 2569 0x6030 0000 0010 003c X30 64 regs.regs[30]
2586 0x6030 0000 0010 003e SP 64 regs. 2570 0x6030 0000 0010 003e SP 64 regs.sp
2587 0x6030 0000 0010 0040 PC 64 regs. 2571 0x6030 0000 0010 0040 PC 64 regs.pc
2588 0x6030 0000 0010 0042 PSTATE 64 regs. 2572 0x6030 0000 0010 0042 PSTATE 64 regs.pstate
2589 0x6030 0000 0010 0044 SP_EL1 64 sp_el 2573 0x6030 0000 0010 0044 SP_EL1 64 sp_el1
2590 0x6030 0000 0010 0046 ELR_EL1 64 elr_e 2574 0x6030 0000 0010 0046 ELR_EL1 64 elr_el1
2591 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[ 2575 0x6030 0000 0010 0048 SPSR_EL1 64 spsr[KVM_SPSR_EL1] (alias SPSR_SVC)
2592 0x6030 0000 0010 004a SPSR_ABT 64 spsr[ 2576 0x6030 0000 0010 004a SPSR_ABT 64 spsr[KVM_SPSR_ABT]
2593 0x6030 0000 0010 004c SPSR_UND 64 spsr[ 2577 0x6030 0000 0010 004c SPSR_UND 64 spsr[KVM_SPSR_UND]
2594 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[ 2578 0x6030 0000 0010 004e SPSR_IRQ 64 spsr[KVM_SPSR_IRQ]
2595 0x6030 0000 0010 0050 SPSR_FIQ 64 spsr[ !! 2579 0x6060 0000 0010 0050 SPSR_FIQ 64 spsr[KVM_SPSR_FIQ]
2596 0x6040 0000 0010 0054 V0 128 fp_re 2580 0x6040 0000 0010 0054 V0 128 fp_regs.vregs[0] [1]_
2597 0x6040 0000 0010 0058 V1 128 fp_re 2581 0x6040 0000 0010 0058 V1 128 fp_regs.vregs[1] [1]_
2598 ... 2582 ...
2599 0x6040 0000 0010 00d0 V31 128 fp_re 2583 0x6040 0000 0010 00d0 V31 128 fp_regs.vregs[31] [1]_
2600 0x6020 0000 0010 00d4 FPSR 32 fp_re 2584 0x6020 0000 0010 00d4 FPSR 32 fp_regs.fpsr
2601 0x6020 0000 0010 00d5 FPCR 32 fp_re 2585 0x6020 0000 0010 00d5 FPCR 32 fp_regs.fpcr
2602 ======================= ========= ===== ===== 2586 ======================= ========= ===== =======================================
2603 2587
2604 .. [1] These encodings are not accepted for S 2588 .. [1] These encodings are not accepted for SVE-enabled vcpus. See
2605 KVM_ARM_VCPU_INIT. 2589 KVM_ARM_VCPU_INIT.
2606 2590
2607 The equivalent register content can be 2591 The equivalent register content can be accessed via bits [127:0] of
2608 the corresponding SVE Zn registers ins 2592 the corresponding SVE Zn registers instead for vcpus that have SVE
2609 enabled (see below). 2593 enabled (see below).
2610 2594
2611 arm64 CCSIDR registers are demultiplexed by C 2595 arm64 CCSIDR registers are demultiplexed by CSSELR value::
2612 2596
2613 0x6020 0000 0011 00 <csselr:8> 2597 0x6020 0000 0011 00 <csselr:8>
2614 2598
2615 arm64 system registers have the following id 2599 arm64 system registers have the following id bit patterns::
2616 2600
2617 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <c 2601 0x6030 0000 0013 <op0:2> <op1:3> <crn:4> <crm:4> <op2:3>
2618 2602
2619 .. warning:: 2603 .. warning::
2620 2604
2621 Two system register IDs do not follow th 2605 Two system register IDs do not follow the specified pattern. These
2622 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_A 2606 are KVM_REG_ARM_TIMER_CVAL and KVM_REG_ARM_TIMER_CNT, which map to
2623 system registers CNTV_CVAL_EL0 and CNTVC 2607 system registers CNTV_CVAL_EL0 and CNTVCT_EL0 respectively. These
2624 two had their values accidentally swappe 2608 two had their values accidentally swapped, which means TIMER_CVAL is
2625 derived from the register encoding for C 2609 derived from the register encoding for CNTVCT_EL0 and TIMER_CNT is
2626 derived from the register encoding for C 2610 derived from the register encoding for CNTV_CVAL_EL0. As this is
2627 API, it must remain this way. 2611 API, it must remain this way.
2628 2612
2629 arm64 firmware pseudo-registers have the foll 2613 arm64 firmware pseudo-registers have the following bit pattern::
2630 2614
2631 0x6030 0000 0014 <regno:16> 2615 0x6030 0000 0014 <regno:16>
2632 2616
2633 arm64 SVE registers have the following bit pa 2617 arm64 SVE registers have the following bit patterns::
2634 2618
2635 0x6080 0000 0015 00 <n:5> <slice:5> Zn bi 2619 0x6080 0000 0015 00 <n:5> <slice:5> Zn bits[2048*slice + 2047 : 2048*slice]
2636 0x6050 0000 0015 04 <n:4> <slice:5> Pn bi 2620 0x6050 0000 0015 04 <n:4> <slice:5> Pn bits[256*slice + 255 : 256*slice]
2637 0x6050 0000 0015 060 <slice:5> FFR b 2621 0x6050 0000 0015 060 <slice:5> FFR bits[256*slice + 255 : 256*slice]
2638 0x6060 0000 0015 ffff KVM_R 2622 0x6060 0000 0015 ffff KVM_REG_ARM64_SVE_VLS pseudo-register
2639 2623
2640 Access to register IDs where 2048 * slice >= 2624 Access to register IDs where 2048 * slice >= 128 * max_vq will fail with
2641 ENOENT. max_vq is the vcpu's maximum support 2625 ENOENT. max_vq is the vcpu's maximum supported vector length in 128-bit
2642 quadwords: see [2]_ below. 2626 quadwords: see [2]_ below.
2643 2627
2644 These registers are only accessible on vcpus 2628 These registers are only accessible on vcpus for which SVE is enabled.
2645 See KVM_ARM_VCPU_INIT for details. 2629 See KVM_ARM_VCPU_INIT for details.
2646 2630
2647 In addition, except for KVM_REG_ARM64_SVE_VLS 2631 In addition, except for KVM_REG_ARM64_SVE_VLS, these registers are not
2648 accessible until the vcpu's SVE configuration 2632 accessible until the vcpu's SVE configuration has been finalized
2649 using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE) 2633 using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE). See KVM_ARM_VCPU_INIT
2650 and KVM_ARM_VCPU_FINALIZE for more informatio 2634 and KVM_ARM_VCPU_FINALIZE for more information about this procedure.
2651 2635
2652 KVM_REG_ARM64_SVE_VLS is a pseudo-register th 2636 KVM_REG_ARM64_SVE_VLS is a pseudo-register that allows the set of vector
2653 lengths supported by the vcpu to be discovere 2637 lengths supported by the vcpu to be discovered and configured by
2654 userspace. When transferred to or from user 2638 userspace. When transferred to or from user memory via KVM_GET_ONE_REG
2655 or KVM_SET_ONE_REG, the value of this registe 2639 or KVM_SET_ONE_REG, the value of this register is of type
2656 __u64[KVM_ARM64_SVE_VLS_WORDS], and encodes t 2640 __u64[KVM_ARM64_SVE_VLS_WORDS], and encodes the set of vector lengths as
2657 follows:: 2641 follows::
2658 2642
2659 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORD 2643 __u64 vector_lengths[KVM_ARM64_SVE_VLS_WORDS];
2660 2644
2661 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX && 2645 if (vq >= SVE_VQ_MIN && vq <= SVE_VQ_MAX &&
2662 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ 2646 ((vector_lengths[(vq - KVM_ARM64_SVE_VQ_MIN) / 64] >>
2663 ((vq - KVM_ARM64_SVE_VQ_MIN) 2647 ((vq - KVM_ARM64_SVE_VQ_MIN) % 64)) & 1))
2664 /* Vector length vq * 16 bytes suppor 2648 /* Vector length vq * 16 bytes supported */
2665 else 2649 else
2666 /* Vector length vq * 16 bytes not su 2650 /* Vector length vq * 16 bytes not supported */
2667 2651
2668 .. [2] The maximum value vq for which the abo 2652 .. [2] The maximum value vq for which the above condition is true is
2669 max_vq. This is the maximum vector le 2653 max_vq. This is the maximum vector length available to the guest on
2670 this vcpu, and determines which regist 2654 this vcpu, and determines which register slices are visible through
2671 this ioctl interface. 2655 this ioctl interface.
2672 2656
2673 (See Documentation/arch/arm64/sve.rst for an 2657 (See Documentation/arch/arm64/sve.rst for an explanation of the "vq"
2674 nomenclature.) 2658 nomenclature.)
2675 2659
2676 KVM_REG_ARM64_SVE_VLS is only accessible afte 2660 KVM_REG_ARM64_SVE_VLS is only accessible after KVM_ARM_VCPU_INIT.
2677 KVM_ARM_VCPU_INIT initialises it to the best 2661 KVM_ARM_VCPU_INIT initialises it to the best set of vector lengths that
2678 the host supports. 2662 the host supports.
2679 2663
2680 Userspace may subsequently modify it if desir 2664 Userspace may subsequently modify it if desired until the vcpu's SVE
2681 configuration is finalized using KVM_ARM_VCPU 2665 configuration is finalized using KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE).
2682 2666
2683 Apart from simply removing all vector lengths 2667 Apart from simply removing all vector lengths from the host set that
2684 exceed some value, support for arbitrarily ch 2668 exceed some value, support for arbitrarily chosen sets of vector lengths
2685 is hardware-dependent and may not be availabl 2669 is hardware-dependent and may not be available. Attempting to configure
2686 an invalid set of vector lengths via KVM_SET_ 2670 an invalid set of vector lengths via KVM_SET_ONE_REG will fail with
2687 EINVAL. 2671 EINVAL.
2688 2672
2689 After the vcpu's SVE configuration is finaliz 2673 After the vcpu's SVE configuration is finalized, further attempts to
2690 write this register will fail with EPERM. 2674 write this register will fail with EPERM.
2691 2675
2692 arm64 bitmap feature firmware pseudo-register 2676 arm64 bitmap feature firmware pseudo-registers have the following bit pattern::
2693 2677
2694 0x6030 0000 0016 <regno:16> 2678 0x6030 0000 0016 <regno:16>
2695 2679
2696 The bitmap feature firmware registers exposes 2680 The bitmap feature firmware registers exposes the hypercall services that
2697 are available for userspace to configure. The 2681 are available for userspace to configure. The set bits corresponds to the
2698 services that are available for the guests to 2682 services that are available for the guests to access. By default, KVM
2699 sets all the supported bits during VM initial 2683 sets all the supported bits during VM initialization. The userspace can
2700 discover the available services via KVM_GET_O 2684 discover the available services via KVM_GET_ONE_REG, and write back the
2701 bitmap corresponding to the features that it 2685 bitmap corresponding to the features that it wishes guests to see via
2702 KVM_SET_ONE_REG. 2686 KVM_SET_ONE_REG.
2703 2687
2704 Note: These registers are immutable once any 2688 Note: These registers are immutable once any of the vCPUs of the VM has
2705 run at least once. A KVM_SET_ONE_REG in such 2689 run at least once. A KVM_SET_ONE_REG in such a scenario will return
2706 a -EBUSY to userspace. 2690 a -EBUSY to userspace.
2707 2691
2708 (See Documentation/virt/kvm/arm/hypercalls.rs 2692 (See Documentation/virt/kvm/arm/hypercalls.rst for more details.)
2709 2693
2710 2694
2711 MIPS registers are mapped using the lower 32 2695 MIPS registers are mapped using the lower 32 bits. The upper 16 of that is
2712 the register group type: 2696 the register group type:
2713 2697
2714 MIPS core registers (see above) have the foll 2698 MIPS core registers (see above) have the following id bit patterns::
2715 2699
2716 0x7030 0000 0000 <reg:16> 2700 0x7030 0000 0000 <reg:16>
2717 2701
2718 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* ab 2702 MIPS CP0 registers (see KVM_REG_MIPS_CP0_* above) have the following id bit
2719 patterns depending on whether they're 32-bit 2703 patterns depending on whether they're 32-bit or 64-bit registers::
2720 2704
2721 0x7020 0000 0001 00 <reg:5> <sel:3> (32-b 2705 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit)
2722 0x7030 0000 0001 00 <reg:5> <sel:3> (64-b 2706 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2723 2707
2724 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_M 2708 Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64
2725 versions of the EntryLo registers regardless 2709 versions of the EntryLo registers regardless of the word size of the host
2726 hardware, host kernel, guest, and whether XPA 2710 hardware, host kernel, guest, and whether XPA is present in the guest, i.e.
2727 with the RI and XI bits (if they exist) in bi 2711 with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and
2728 the PFNX field starting at bit 30. 2712 the PFNX field starting at bit 30.
2729 2713
2730 MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) abov 2714 MIPS MAARs (see KVM_REG_MIPS_CP0_MAAR(*) above) have the following id bit
2731 patterns:: 2715 patterns::
2732 2716
2733 0x7030 0000 0001 01 <reg:8> 2717 0x7030 0000 0001 01 <reg:8>
2734 2718
2735 MIPS KVM control registers (see above) have t 2719 MIPS KVM control registers (see above) have the following id bit patterns::
2736 2720
2737 0x7030 0000 0002 <reg:16> 2721 0x7030 0000 0002 <reg:16>
2738 2722
2739 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32, 2723 MIPS FPU registers (see KVM_REG_MIPS_FPR_{32,64}() above) have the following
2740 id bit patterns depending on the size of the 2724 id bit patterns depending on the size of the register being accessed. They are
2741 always accessed according to the current gues 2725 always accessed according to the current guest FPU mode (Status.FR and
2742 Config5.FRE), i.e. as the guest would see the 2726 Config5.FRE), i.e. as the guest would see them, and they become unpredictable
2743 if the guest FPU mode is changed. MIPS SIMD A 2727 if the guest FPU mode is changed. MIPS SIMD Architecture (MSA) vector
2744 registers (see KVM_REG_MIPS_VEC_128() above) 2728 registers (see KVM_REG_MIPS_VEC_128() above) have similar patterns as they
2745 overlap the FPU registers:: 2729 overlap the FPU registers::
2746 2730
2747 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit F 2731 0x7020 0000 0003 00 <0:3> <reg:5> (32-bit FPU registers)
2748 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit F 2732 0x7030 0000 0003 00 <0:3> <reg:5> (64-bit FPU registers)
2749 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit 2733 0x7040 0000 0003 00 <0:3> <reg:5> (128-bit MSA vector registers)
2750 2734
2751 MIPS FPU control registers (see KVM_REG_MIPS_ 2735 MIPS FPU control registers (see KVM_REG_MIPS_FCR_{IR,CSR} above) have the
2752 following id bit patterns:: 2736 following id bit patterns::
2753 2737
2754 0x7020 0000 0003 01 <0:3> <reg:5> 2738 0x7020 0000 0003 01 <0:3> <reg:5>
2755 2739
2756 MIPS MSA control registers (see KVM_REG_MIPS_ 2740 MIPS MSA control registers (see KVM_REG_MIPS_MSA_{IR,CSR} above) have the
2757 following id bit patterns:: 2741 following id bit patterns::
2758 2742
2759 0x7020 0000 0003 02 <0:3> <reg:5> 2743 0x7020 0000 0003 02 <0:3> <reg:5>
2760 2744
2761 RISC-V registers are mapped using the lower 3 2745 RISC-V registers are mapped using the lower 32 bits. The upper 8 bits of
2762 that is the register group type. 2746 that is the register group type.
2763 2747
2764 RISC-V config registers are meant for configu 2748 RISC-V config registers are meant for configuring a Guest VCPU and it has
2765 the following id bit patterns:: 2749 the following id bit patterns::
2766 2750
2767 0x8020 0000 01 <index into the kvm_riscv_co 2751 0x8020 0000 01 <index into the kvm_riscv_config struct:24> (32bit Host)
2768 0x8030 0000 01 <index into the kvm_riscv_co 2752 0x8030 0000 01 <index into the kvm_riscv_config struct:24> (64bit Host)
2769 2753
2770 Following are the RISC-V config registers: 2754 Following are the RISC-V config registers:
2771 2755
2772 ======================= ========= =========== 2756 ======================= ========= =============================================
2773 Encoding Register Description 2757 Encoding Register Description
2774 ======================= ========= =========== 2758 ======================= ========= =============================================
2775 0x80x0 0000 0100 0000 isa ISA feature 2759 0x80x0 0000 0100 0000 isa ISA feature bitmap of Guest VCPU
2776 ======================= ========= =========== 2760 ======================= ========= =============================================
2777 2761
2778 The isa config register can be read anytime b 2762 The isa config register can be read anytime but can only be written before
2779 a Guest VCPU runs. It will have ISA feature b 2763 a Guest VCPU runs. It will have ISA feature bits matching underlying host
2780 set by default. 2764 set by default.
2781 2765
2782 RISC-V core registers represent the general e 2766 RISC-V core registers represent the general execution state of a Guest VCPU
2783 and it has the following id bit patterns:: 2767 and it has the following id bit patterns::
2784 2768
2785 0x8020 0000 02 <index into the kvm_riscv_co 2769 0x8020 0000 02 <index into the kvm_riscv_core struct:24> (32bit Host)
2786 0x8030 0000 02 <index into the kvm_riscv_co 2770 0x8030 0000 02 <index into the kvm_riscv_core struct:24> (64bit Host)
2787 2771
2788 Following are the RISC-V core registers: 2772 Following are the RISC-V core registers:
2789 2773
2790 ======================= ========= =========== 2774 ======================= ========= =============================================
2791 Encoding Register Description 2775 Encoding Register Description
2792 ======================= ========= =========== 2776 ======================= ========= =============================================
2793 0x80x0 0000 0200 0000 regs.pc Program cou 2777 0x80x0 0000 0200 0000 regs.pc Program counter
2794 0x80x0 0000 0200 0001 regs.ra Return addr 2778 0x80x0 0000 0200 0001 regs.ra Return address
2795 0x80x0 0000 0200 0002 regs.sp Stack point 2779 0x80x0 0000 0200 0002 regs.sp Stack pointer
2796 0x80x0 0000 0200 0003 regs.gp Global poin 2780 0x80x0 0000 0200 0003 regs.gp Global pointer
2797 0x80x0 0000 0200 0004 regs.tp Task pointe 2781 0x80x0 0000 0200 0004 regs.tp Task pointer
2798 0x80x0 0000 0200 0005 regs.t0 Caller save 2782 0x80x0 0000 0200 0005 regs.t0 Caller saved register 0
2799 0x80x0 0000 0200 0006 regs.t1 Caller save 2783 0x80x0 0000 0200 0006 regs.t1 Caller saved register 1
2800 0x80x0 0000 0200 0007 regs.t2 Caller save 2784 0x80x0 0000 0200 0007 regs.t2 Caller saved register 2
2801 0x80x0 0000 0200 0008 regs.s0 Callee save 2785 0x80x0 0000 0200 0008 regs.s0 Callee saved register 0
2802 0x80x0 0000 0200 0009 regs.s1 Callee save 2786 0x80x0 0000 0200 0009 regs.s1 Callee saved register 1
2803 0x80x0 0000 0200 000a regs.a0 Function ar 2787 0x80x0 0000 0200 000a regs.a0 Function argument (or return value) 0
2804 0x80x0 0000 0200 000b regs.a1 Function ar 2788 0x80x0 0000 0200 000b regs.a1 Function argument (or return value) 1
2805 0x80x0 0000 0200 000c regs.a2 Function ar 2789 0x80x0 0000 0200 000c regs.a2 Function argument 2
2806 0x80x0 0000 0200 000d regs.a3 Function ar 2790 0x80x0 0000 0200 000d regs.a3 Function argument 3
2807 0x80x0 0000 0200 000e regs.a4 Function ar 2791 0x80x0 0000 0200 000e regs.a4 Function argument 4
2808 0x80x0 0000 0200 000f regs.a5 Function ar 2792 0x80x0 0000 0200 000f regs.a5 Function argument 5
2809 0x80x0 0000 0200 0010 regs.a6 Function ar 2793 0x80x0 0000 0200 0010 regs.a6 Function argument 6
2810 0x80x0 0000 0200 0011 regs.a7 Function ar 2794 0x80x0 0000 0200 0011 regs.a7 Function argument 7
2811 0x80x0 0000 0200 0012 regs.s2 Callee save 2795 0x80x0 0000 0200 0012 regs.s2 Callee saved register 2
2812 0x80x0 0000 0200 0013 regs.s3 Callee save 2796 0x80x0 0000 0200 0013 regs.s3 Callee saved register 3
2813 0x80x0 0000 0200 0014 regs.s4 Callee save 2797 0x80x0 0000 0200 0014 regs.s4 Callee saved register 4
2814 0x80x0 0000 0200 0015 regs.s5 Callee save 2798 0x80x0 0000 0200 0015 regs.s5 Callee saved register 5
2815 0x80x0 0000 0200 0016 regs.s6 Callee save 2799 0x80x0 0000 0200 0016 regs.s6 Callee saved register 6
2816 0x80x0 0000 0200 0017 regs.s7 Callee save 2800 0x80x0 0000 0200 0017 regs.s7 Callee saved register 7
2817 0x80x0 0000 0200 0018 regs.s8 Callee save 2801 0x80x0 0000 0200 0018 regs.s8 Callee saved register 8
2818 0x80x0 0000 0200 0019 regs.s9 Callee save 2802 0x80x0 0000 0200 0019 regs.s9 Callee saved register 9
2819 0x80x0 0000 0200 001a regs.s10 Callee save 2803 0x80x0 0000 0200 001a regs.s10 Callee saved register 10
2820 0x80x0 0000 0200 001b regs.s11 Callee save 2804 0x80x0 0000 0200 001b regs.s11 Callee saved register 11
2821 0x80x0 0000 0200 001c regs.t3 Caller save 2805 0x80x0 0000 0200 001c regs.t3 Caller saved register 3
2822 0x80x0 0000 0200 001d regs.t4 Caller save 2806 0x80x0 0000 0200 001d regs.t4 Caller saved register 4
2823 0x80x0 0000 0200 001e regs.t5 Caller save 2807 0x80x0 0000 0200 001e regs.t5 Caller saved register 5
2824 0x80x0 0000 0200 001f regs.t6 Caller save 2808 0x80x0 0000 0200 001f regs.t6 Caller saved register 6
2825 0x80x0 0000 0200 0020 mode Privilege m 2809 0x80x0 0000 0200 0020 mode Privilege mode (1 = S-mode or 0 = U-mode)
2826 ======================= ========= =========== 2810 ======================= ========= =============================================
2827 2811
2828 RISC-V csr registers represent the supervisor 2812 RISC-V csr registers represent the supervisor mode control/status registers
2829 of a Guest VCPU and it has the following id b 2813 of a Guest VCPU and it has the following id bit patterns::
2830 2814
2831 0x8020 0000 03 <index into the kvm_riscv_cs 2815 0x8020 0000 03 <index into the kvm_riscv_csr struct:24> (32bit Host)
2832 0x8030 0000 03 <index into the kvm_riscv_cs 2816 0x8030 0000 03 <index into the kvm_riscv_csr struct:24> (64bit Host)
2833 2817
2834 Following are the RISC-V csr registers: 2818 Following are the RISC-V csr registers:
2835 2819
2836 ======================= ========= =========== 2820 ======================= ========= =============================================
2837 Encoding Register Description 2821 Encoding Register Description
2838 ======================= ========= =========== 2822 ======================= ========= =============================================
2839 0x80x0 0000 0300 0000 sstatus Supervisor 2823 0x80x0 0000 0300 0000 sstatus Supervisor status
2840 0x80x0 0000 0300 0001 sie Supervisor 2824 0x80x0 0000 0300 0001 sie Supervisor interrupt enable
2841 0x80x0 0000 0300 0002 stvec Supervisor 2825 0x80x0 0000 0300 0002 stvec Supervisor trap vector base
2842 0x80x0 0000 0300 0003 sscratch Supervisor 2826 0x80x0 0000 0300 0003 sscratch Supervisor scratch register
2843 0x80x0 0000 0300 0004 sepc Supervisor 2827 0x80x0 0000 0300 0004 sepc Supervisor exception program counter
2844 0x80x0 0000 0300 0005 scause Supervisor 2828 0x80x0 0000 0300 0005 scause Supervisor trap cause
2845 0x80x0 0000 0300 0006 stval Supervisor 2829 0x80x0 0000 0300 0006 stval Supervisor bad address or instruction
2846 0x80x0 0000 0300 0007 sip Supervisor 2830 0x80x0 0000 0300 0007 sip Supervisor interrupt pending
2847 0x80x0 0000 0300 0008 satp Supervisor 2831 0x80x0 0000 0300 0008 satp Supervisor address translation and protection
2848 ======================= ========= =========== 2832 ======================= ========= =============================================
2849 2833
2850 RISC-V timer registers represent the timer st 2834 RISC-V timer registers represent the timer state of a Guest VCPU and it has
2851 the following id bit patterns:: 2835 the following id bit patterns::
2852 2836
2853 0x8030 0000 04 <index into the kvm_riscv_ti 2837 0x8030 0000 04 <index into the kvm_riscv_timer struct:24>
2854 2838
2855 Following are the RISC-V timer registers: 2839 Following are the RISC-V timer registers:
2856 2840
2857 ======================= ========= =========== 2841 ======================= ========= =============================================
2858 Encoding Register Description 2842 Encoding Register Description
2859 ======================= ========= =========== 2843 ======================= ========= =============================================
2860 0x8030 0000 0400 0000 frequency Time base f 2844 0x8030 0000 0400 0000 frequency Time base frequency (read-only)
2861 0x8030 0000 0400 0001 time Time value 2845 0x8030 0000 0400 0001 time Time value visible to Guest
2862 0x8030 0000 0400 0002 compare Time compar 2846 0x8030 0000 0400 0002 compare Time compare programmed by Guest
2863 0x8030 0000 0400 0003 state Time compar 2847 0x8030 0000 0400 0003 state Time compare state (1 = ON or 0 = OFF)
2864 ======================= ========= =========== 2848 ======================= ========= =============================================
2865 2849
2866 RISC-V F-extension registers represent the si 2850 RISC-V F-extension registers represent the single precision floating point
2867 state of a Guest VCPU and it has the followin 2851 state of a Guest VCPU and it has the following id bit patterns::
2868 2852
2869 0x8020 0000 05 <index into the __riscv_f_ex 2853 0x8020 0000 05 <index into the __riscv_f_ext_state struct:24>
2870 2854
2871 Following are the RISC-V F-extension register 2855 Following are the RISC-V F-extension registers:
2872 2856
2873 ======================= ========= =========== 2857 ======================= ========= =============================================
2874 Encoding Register Description 2858 Encoding Register Description
2875 ======================= ========= =========== 2859 ======================= ========= =============================================
2876 0x8020 0000 0500 0000 f[0] Floating po 2860 0x8020 0000 0500 0000 f[0] Floating point register 0
2877 ... 2861 ...
2878 0x8020 0000 0500 001f f[31] Floating po 2862 0x8020 0000 0500 001f f[31] Floating point register 31
2879 0x8020 0000 0500 0020 fcsr Floating po 2863 0x8020 0000 0500 0020 fcsr Floating point control and status register
2880 ======================= ========= =========== 2864 ======================= ========= =============================================
2881 2865
2882 RISC-V D-extension registers represent the do 2866 RISC-V D-extension registers represent the double precision floating point
2883 state of a Guest VCPU and it has the followin 2867 state of a Guest VCPU and it has the following id bit patterns::
2884 2868
2885 0x8020 0000 06 <index into the __riscv_d_ex 2869 0x8020 0000 06 <index into the __riscv_d_ext_state struct:24> (fcsr)
2886 0x8030 0000 06 <index into the __riscv_d_ex 2870 0x8030 0000 06 <index into the __riscv_d_ext_state struct:24> (non-fcsr)
2887 2871
2888 Following are the RISC-V D-extension register 2872 Following are the RISC-V D-extension registers:
2889 2873
2890 ======================= ========= =========== 2874 ======================= ========= =============================================
2891 Encoding Register Description 2875 Encoding Register Description
2892 ======================= ========= =========== 2876 ======================= ========= =============================================
2893 0x8030 0000 0600 0000 f[0] Floating po 2877 0x8030 0000 0600 0000 f[0] Floating point register 0
2894 ... 2878 ...
2895 0x8030 0000 0600 001f f[31] Floating po 2879 0x8030 0000 0600 001f f[31] Floating point register 31
2896 0x8020 0000 0600 0020 fcsr Floating po 2880 0x8020 0000 0600 0020 fcsr Floating point control and status register
2897 ======================= ========= =========== 2881 ======================= ========= =============================================
2898 2882
2899 LoongArch registers are mapped using the lowe 2883 LoongArch registers are mapped using the lower 32 bits. The upper 16 bits of
2900 that is the register group type. 2884 that is the register group type.
2901 2885
2902 LoongArch csr registers are used to control g 2886 LoongArch csr registers are used to control guest cpu or get status of guest
2903 cpu, and they have the following id bit patte 2887 cpu, and they have the following id bit patterns::
2904 2888
2905 0x9030 0000 0001 00 <reg:5> <sel:3> (64-b 2889 0x9030 0000 0001 00 <reg:5> <sel:3> (64-bit)
2906 2890
2907 LoongArch KVM control registers are used to i 2891 LoongArch KVM control registers are used to implement some new defined functions
2908 such as set vcpu counter or reset vcpu, and t 2892 such as set vcpu counter or reset vcpu, and they have the following id bit patterns::
2909 2893
2910 0x9030 0000 0002 <reg:16> 2894 0x9030 0000 0002 <reg:16>
2911 2895
2912 2896
2913 4.69 KVM_GET_ONE_REG 2897 4.69 KVM_GET_ONE_REG
2914 -------------------- 2898 --------------------
2915 2899
2916 :Capability: KVM_CAP_ONE_REG 2900 :Capability: KVM_CAP_ONE_REG
2917 :Architectures: all 2901 :Architectures: all
2918 :Type: vcpu ioctl 2902 :Type: vcpu ioctl
2919 :Parameters: struct kvm_one_reg (in and out) 2903 :Parameters: struct kvm_one_reg (in and out)
2920 :Returns: 0 on success, negative value on fai 2904 :Returns: 0 on success, negative value on failure
2921 2905
2922 Errors include: 2906 Errors include:
2923 2907
2924 ======== ================================== 2908 ======== ============================================================
2925 ENOENT no such register 2909 ENOENT no such register
2926 EINVAL invalid register ID, or no such re 2910 EINVAL invalid register ID, or no such register or used with VMs in
2927 protected virtualization mode on s 2911 protected virtualization mode on s390
2928 EPERM (arm64) register access not allowe 2912 EPERM (arm64) register access not allowed before vcpu finalization
2929 ======== ================================== 2913 ======== ============================================================
2930 2914
2931 (These error codes are indicative only: do no 2915 (These error codes are indicative only: do not rely on a specific error
2932 code being returned in a specific situation.) 2916 code being returned in a specific situation.)
2933 2917
2934 This ioctl allows to receive the value of a s 2918 This ioctl allows to receive the value of a single register implemented
2935 in a vcpu. The register to read is indicated 2919 in a vcpu. The register to read is indicated by the "id" field of the
2936 kvm_one_reg struct passed in. On success, the 2920 kvm_one_reg struct passed in. On success, the register value can be found
2937 at the memory location pointed to by "addr". 2921 at the memory location pointed to by "addr".
2938 2922
2939 The list of registers accessible using this i 2923 The list of registers accessible using this interface is identical to the
2940 list in 4.68. 2924 list in 4.68.
2941 2925
2942 2926
2943 4.70 KVM_KVMCLOCK_CTRL 2927 4.70 KVM_KVMCLOCK_CTRL
2944 ---------------------- 2928 ----------------------
2945 2929
2946 :Capability: KVM_CAP_KVMCLOCK_CTRL 2930 :Capability: KVM_CAP_KVMCLOCK_CTRL
2947 :Architectures: Any that implement pvclocks ( 2931 :Architectures: Any that implement pvclocks (currently x86 only)
2948 :Type: vcpu ioctl 2932 :Type: vcpu ioctl
2949 :Parameters: None 2933 :Parameters: None
2950 :Returns: 0 on success, -1 on error 2934 :Returns: 0 on success, -1 on error
2951 2935
2952 This ioctl sets a flag accessible to the gues 2936 This ioctl sets a flag accessible to the guest indicating that the specified
2953 vCPU has been paused by the host userspace. 2937 vCPU has been paused by the host userspace.
2954 2938
2955 The host will set a flag in the pvclock struc 2939 The host will set a flag in the pvclock structure that is checked from the
2956 soft lockup watchdog. The flag is part of th 2940 soft lockup watchdog. The flag is part of the pvclock structure that is
2957 shared between guest and host, specifically t 2941 shared between guest and host, specifically the second bit of the flags
2958 field of the pvclock_vcpu_time_info structure 2942 field of the pvclock_vcpu_time_info structure. It will be set exclusively by
2959 the host and read/cleared exclusively by the 2943 the host and read/cleared exclusively by the guest. The guest operation of
2960 checking and clearing the flag must be an ato 2944 checking and clearing the flag must be an atomic operation so
2961 load-link/store-conditional, or equivalent mu 2945 load-link/store-conditional, or equivalent must be used. There are two cases
2962 where the guest will clear the flag: when the 2946 where the guest will clear the flag: when the soft lockup watchdog timer resets
2963 itself or when a soft lockup is detected. Th 2947 itself or when a soft lockup is detected. This ioctl can be called any time
2964 after pausing the vcpu, but before it is resu 2948 after pausing the vcpu, but before it is resumed.
2965 2949
2966 2950
2967 4.71 KVM_SIGNAL_MSI 2951 4.71 KVM_SIGNAL_MSI
2968 ------------------- 2952 -------------------
2969 2953
2970 :Capability: KVM_CAP_SIGNAL_MSI 2954 :Capability: KVM_CAP_SIGNAL_MSI
2971 :Architectures: x86 arm64 2955 :Architectures: x86 arm64
2972 :Type: vm ioctl 2956 :Type: vm ioctl
2973 :Parameters: struct kvm_msi (in) 2957 :Parameters: struct kvm_msi (in)
2974 :Returns: >0 on delivery, 0 if guest blocked 2958 :Returns: >0 on delivery, 0 if guest blocked the MSI, and -1 on error
2975 2959
2976 Directly inject a MSI message. Only valid wit 2960 Directly inject a MSI message. Only valid with in-kernel irqchip that handles
2977 MSI messages. 2961 MSI messages.
2978 2962
2979 :: 2963 ::
2980 2964
2981 struct kvm_msi { 2965 struct kvm_msi {
2982 __u32 address_lo; 2966 __u32 address_lo;
2983 __u32 address_hi; 2967 __u32 address_hi;
2984 __u32 data; 2968 __u32 data;
2985 __u32 flags; 2969 __u32 flags;
2986 __u32 devid; 2970 __u32 devid;
2987 __u8 pad[12]; 2971 __u8 pad[12];
2988 }; 2972 };
2989 2973
2990 flags: 2974 flags:
2991 KVM_MSI_VALID_DEVID: devid contains a valid 2975 KVM_MSI_VALID_DEVID: devid contains a valid value. The per-VM
2992 KVM_CAP_MSI_DEVID capability advertises the 2976 KVM_CAP_MSI_DEVID capability advertises the requirement to provide
2993 the device ID. If this capability is not a 2977 the device ID. If this capability is not available, userspace
2994 should never set the KVM_MSI_VALID_DEVID fl 2978 should never set the KVM_MSI_VALID_DEVID flag as the ioctl might fail.
2995 2979
2996 If KVM_MSI_VALID_DEVID is set, devid contains 2980 If KVM_MSI_VALID_DEVID is set, devid contains a unique device identifier
2997 for the device that wrote the MSI message. F 2981 for the device that wrote the MSI message. For PCI, this is usually a
2998 BDF identifier in the lower 16 bits. !! 2982 BFD identifier in the lower 16 bits.
2999 2983
3000 On x86, address_hi is ignored unless the KVM_ 2984 On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS
3001 feature of KVM_CAP_X2APIC_API capability is e 2985 feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled,
3002 address_hi bits 31-8 provide bits 31-8 of the 2986 address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of
3003 address_hi must be zero. 2987 address_hi must be zero.
3004 2988
3005 2989
3006 4.71 KVM_CREATE_PIT2 2990 4.71 KVM_CREATE_PIT2
3007 -------------------- 2991 --------------------
3008 2992
3009 :Capability: KVM_CAP_PIT2 2993 :Capability: KVM_CAP_PIT2
3010 :Architectures: x86 2994 :Architectures: x86
3011 :Type: vm ioctl 2995 :Type: vm ioctl
3012 :Parameters: struct kvm_pit_config (in) 2996 :Parameters: struct kvm_pit_config (in)
3013 :Returns: 0 on success, -1 on error 2997 :Returns: 0 on success, -1 on error
3014 2998
3015 Creates an in-kernel device model for the i82 2999 Creates an in-kernel device model for the i8254 PIT. This call is only valid
3016 after enabling in-kernel irqchip support via 3000 after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP. The following
3017 parameters have to be passed:: 3001 parameters have to be passed::
3018 3002
3019 struct kvm_pit_config { 3003 struct kvm_pit_config {
3020 __u32 flags; 3004 __u32 flags;
3021 __u32 pad[15]; 3005 __u32 pad[15];
3022 }; 3006 };
3023 3007
3024 Valid flags are:: 3008 Valid flags are::
3025 3009
3026 #define KVM_PIT_SPEAKER_DUMMY 1 /* emul 3010 #define KVM_PIT_SPEAKER_DUMMY 1 /* emulate speaker port stub */
3027 3011
3028 PIT timer interrupts may use a per-VM kernel 3012 PIT timer interrupts may use a per-VM kernel thread for injection. If it
3029 exists, this thread will have a name of the f 3013 exists, this thread will have a name of the following pattern::
3030 3014
3031 kvm-pit/<owner-process-pid> 3015 kvm-pit/<owner-process-pid>
3032 3016
3033 When running a guest with elevated priorities 3017 When running a guest with elevated priorities, the scheduling parameters of
3034 this thread may have to be adjusted according 3018 this thread may have to be adjusted accordingly.
3035 3019
3036 This IOCTL replaces the obsolete KVM_CREATE_P 3020 This IOCTL replaces the obsolete KVM_CREATE_PIT.
3037 3021
3038 3022
3039 4.72 KVM_GET_PIT2 3023 4.72 KVM_GET_PIT2
3040 ----------------- 3024 -----------------
3041 3025
3042 :Capability: KVM_CAP_PIT_STATE2 3026 :Capability: KVM_CAP_PIT_STATE2
3043 :Architectures: x86 3027 :Architectures: x86
3044 :Type: vm ioctl 3028 :Type: vm ioctl
3045 :Parameters: struct kvm_pit_state2 (out) 3029 :Parameters: struct kvm_pit_state2 (out)
3046 :Returns: 0 on success, -1 on error 3030 :Returns: 0 on success, -1 on error
3047 3031
3048 Retrieves the state of the in-kernel PIT mode 3032 Retrieves the state of the in-kernel PIT model. Only valid after
3049 KVM_CREATE_PIT2. The state is returned in the 3033 KVM_CREATE_PIT2. The state is returned in the following structure::
3050 3034
3051 struct kvm_pit_state2 { 3035 struct kvm_pit_state2 {
3052 struct kvm_pit_channel_state channels 3036 struct kvm_pit_channel_state channels[3];
3053 __u32 flags; 3037 __u32 flags;
3054 __u32 reserved[9]; 3038 __u32 reserved[9];
3055 }; 3039 };
3056 3040
3057 Valid flags are:: 3041 Valid flags are::
3058 3042
3059 /* disable PIT in HPET legacy mode */ 3043 /* disable PIT in HPET legacy mode */
3060 #define KVM_PIT_FLAGS_HPET_LEGACY 0x000 3044 #define KVM_PIT_FLAGS_HPET_LEGACY 0x00000001
3061 /* speaker port data bit enabled */ 3045 /* speaker port data bit enabled */
3062 #define KVM_PIT_FLAGS_SPEAKER_DATA_ON 0x000 3046 #define KVM_PIT_FLAGS_SPEAKER_DATA_ON 0x00000002
3063 3047
3064 This IOCTL replaces the obsolete KVM_GET_PIT. 3048 This IOCTL replaces the obsolete KVM_GET_PIT.
3065 3049
3066 3050
3067 4.73 KVM_SET_PIT2 3051 4.73 KVM_SET_PIT2
3068 ----------------- 3052 -----------------
3069 3053
3070 :Capability: KVM_CAP_PIT_STATE2 3054 :Capability: KVM_CAP_PIT_STATE2
3071 :Architectures: x86 3055 :Architectures: x86
3072 :Type: vm ioctl 3056 :Type: vm ioctl
3073 :Parameters: struct kvm_pit_state2 (in) 3057 :Parameters: struct kvm_pit_state2 (in)
3074 :Returns: 0 on success, -1 on error 3058 :Returns: 0 on success, -1 on error
3075 3059
3076 Sets the state of the in-kernel PIT model. On 3060 Sets the state of the in-kernel PIT model. Only valid after KVM_CREATE_PIT2.
3077 See KVM_GET_PIT2 for details on struct kvm_pi 3061 See KVM_GET_PIT2 for details on struct kvm_pit_state2.
3078 3062
3079 This IOCTL replaces the obsolete KVM_SET_PIT. 3063 This IOCTL replaces the obsolete KVM_SET_PIT.
3080 3064
3081 3065
3082 4.74 KVM_PPC_GET_SMMU_INFO 3066 4.74 KVM_PPC_GET_SMMU_INFO
3083 -------------------------- 3067 --------------------------
3084 3068
3085 :Capability: KVM_CAP_PPC_GET_SMMU_INFO 3069 :Capability: KVM_CAP_PPC_GET_SMMU_INFO
3086 :Architectures: powerpc 3070 :Architectures: powerpc
3087 :Type: vm ioctl 3071 :Type: vm ioctl
3088 :Parameters: None 3072 :Parameters: None
3089 :Returns: 0 on success, -1 on error 3073 :Returns: 0 on success, -1 on error
3090 3074
3091 This populates and returns a structure descri 3075 This populates and returns a structure describing the features of
3092 the "Server" class MMU emulation supported by 3076 the "Server" class MMU emulation supported by KVM.
3093 This can in turn be used by userspace to gene 3077 This can in turn be used by userspace to generate the appropriate
3094 device-tree properties for the guest operatin 3078 device-tree properties for the guest operating system.
3095 3079
3096 The structure contains some global informatio 3080 The structure contains some global information, followed by an
3097 array of supported segment page sizes:: 3081 array of supported segment page sizes::
3098 3082
3099 struct kvm_ppc_smmu_info { 3083 struct kvm_ppc_smmu_info {
3100 __u64 flags; 3084 __u64 flags;
3101 __u32 slb_size; 3085 __u32 slb_size;
3102 __u32 pad; 3086 __u32 pad;
3103 struct kvm_ppc_one_seg_page_size 3087 struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ];
3104 }; 3088 };
3105 3089
3106 The supported flags are: 3090 The supported flags are:
3107 3091
3108 - KVM_PPC_PAGE_SIZES_REAL: 3092 - KVM_PPC_PAGE_SIZES_REAL:
3109 When that flag is set, guest page siz 3093 When that flag is set, guest page sizes must "fit" the backing
3110 store page sizes. When not set, any p 3094 store page sizes. When not set, any page size in the list can
3111 be used regardless of how they are ba 3095 be used regardless of how they are backed by userspace.
3112 3096
3113 - KVM_PPC_1T_SEGMENTS 3097 - KVM_PPC_1T_SEGMENTS
3114 The emulated MMU supports 1T segments 3098 The emulated MMU supports 1T segments in addition to the
3115 standard 256M ones. 3099 standard 256M ones.
3116 3100
3117 - KVM_PPC_NO_HASH 3101 - KVM_PPC_NO_HASH
3118 This flag indicates that HPT guests a 3102 This flag indicates that HPT guests are not supported by KVM,
3119 thus all guests must use radix MMU mo 3103 thus all guests must use radix MMU mode.
3120 3104
3121 The "slb_size" field indicates how many SLB e 3105 The "slb_size" field indicates how many SLB entries are supported
3122 3106
3123 The "sps" array contains 8 entries indicating 3107 The "sps" array contains 8 entries indicating the supported base
3124 page sizes for a segment in increasing order. 3108 page sizes for a segment in increasing order. Each entry is defined
3125 as follow:: 3109 as follow::
3126 3110
3127 struct kvm_ppc_one_seg_page_size { 3111 struct kvm_ppc_one_seg_page_size {
3128 __u32 page_shift; /* Base page 3112 __u32 page_shift; /* Base page shift of segment (or 0) */
3129 __u32 slb_enc; /* SLB encodi 3113 __u32 slb_enc; /* SLB encoding for BookS */
3130 struct kvm_ppc_one_page_size enc[KVM_ 3114 struct kvm_ppc_one_page_size enc[KVM_PPC_PAGE_SIZES_MAX_SZ];
3131 }; 3115 };
3132 3116
3133 An entry with a "page_shift" of 0 is unused. 3117 An entry with a "page_shift" of 0 is unused. Because the array is
3134 organized in increasing order, a lookup can s 3118 organized in increasing order, a lookup can stop when encountering
3135 such an entry. 3119 such an entry.
3136 3120
3137 The "slb_enc" field provides the encoding to 3121 The "slb_enc" field provides the encoding to use in the SLB for the
3138 page size. The bits are in positions such as 3122 page size. The bits are in positions such as the value can directly
3139 be OR'ed into the "vsid" argument of the slbm 3123 be OR'ed into the "vsid" argument of the slbmte instruction.
3140 3124
3141 The "enc" array is a list which for each of t 3125 The "enc" array is a list which for each of those segment base page
3142 size provides the list of supported actual pa 3126 size provides the list of supported actual page sizes (which can be
3143 only larger or equal to the base page size), 3127 only larger or equal to the base page size), along with the
3144 corresponding encoding in the hash PTE. Simil 3128 corresponding encoding in the hash PTE. Similarly, the array is
3145 8 entries sorted by increasing sizes and an e 3129 8 entries sorted by increasing sizes and an entry with a "0" shift
3146 is an empty entry and a terminator:: 3130 is an empty entry and a terminator::
3147 3131
3148 struct kvm_ppc_one_page_size { 3132 struct kvm_ppc_one_page_size {
3149 __u32 page_shift; /* Page shift 3133 __u32 page_shift; /* Page shift (or 0) */
3150 __u32 pte_enc; /* Encoding i 3134 __u32 pte_enc; /* Encoding in the HPTE (>>12) */
3151 }; 3135 };
3152 3136
3153 The "pte_enc" field provides a value that can 3137 The "pte_enc" field provides a value that can OR'ed into the hash
3154 PTE's RPN field (ie, it needs to be shifted l 3138 PTE's RPN field (ie, it needs to be shifted left by 12 to OR it
3155 into the hash PTE second double word). 3139 into the hash PTE second double word).
3156 3140
3157 4.75 KVM_IRQFD 3141 4.75 KVM_IRQFD
3158 -------------- 3142 --------------
3159 3143
3160 :Capability: KVM_CAP_IRQFD 3144 :Capability: KVM_CAP_IRQFD
3161 :Architectures: x86 s390 arm64 3145 :Architectures: x86 s390 arm64
3162 :Type: vm ioctl 3146 :Type: vm ioctl
3163 :Parameters: struct kvm_irqfd (in) 3147 :Parameters: struct kvm_irqfd (in)
3164 :Returns: 0 on success, -1 on error 3148 :Returns: 0 on success, -1 on error
3165 3149
3166 Allows setting an eventfd to directly trigger 3150 Allows setting an eventfd to directly trigger a guest interrupt.
3167 kvm_irqfd.fd specifies the file descriptor to 3151 kvm_irqfd.fd specifies the file descriptor to use as the eventfd and
3168 kvm_irqfd.gsi specifies the irqchip pin toggl 3152 kvm_irqfd.gsi specifies the irqchip pin toggled by this event. When
3169 an event is triggered on the eventfd, an inte 3153 an event is triggered on the eventfd, an interrupt is injected into
3170 the guest using the specified gsi pin. The i 3154 the guest using the specified gsi pin. The irqfd is removed using
3171 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying 3155 the KVM_IRQFD_FLAG_DEASSIGN flag, specifying both kvm_irqfd.fd
3172 and kvm_irqfd.gsi. 3156 and kvm_irqfd.gsi.
3173 3157
3174 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD suppor 3158 With KVM_CAP_IRQFD_RESAMPLE, KVM_IRQFD supports a de-assert and notify
3175 mechanism allowing emulation of level-trigger 3159 mechanism allowing emulation of level-triggered, irqfd-based
3176 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is 3160 interrupts. When KVM_IRQFD_FLAG_RESAMPLE is set the user must pass an
3177 additional eventfd in the kvm_irqfd.resamplef 3161 additional eventfd in the kvm_irqfd.resamplefd field. When operating
3178 in resample mode, posting of an interrupt thr 3162 in resample mode, posting of an interrupt through kvm_irq.fd asserts
3179 the specified gsi in the irqchip. When the i 3163 the specified gsi in the irqchip. When the irqchip is resampled, such
3180 as from an EOI, the gsi is de-asserted and th 3164 as from an EOI, the gsi is de-asserted and the user is notified via
3181 kvm_irqfd.resamplefd. It is the user's respo 3165 kvm_irqfd.resamplefd. It is the user's responsibility to re-queue
3182 the interrupt if the device making use of it 3166 the interrupt if the device making use of it still requires service.
3183 Note that closing the resamplefd is not suffi 3167 Note that closing the resamplefd is not sufficient to disable the
3184 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only n 3168 irqfd. The KVM_IRQFD_FLAG_RESAMPLE is only necessary on assignment
3185 and need not be specified with KVM_IRQFD_FLAG 3169 and need not be specified with KVM_IRQFD_FLAG_DEASSIGN.
3186 3170
3187 On arm64, gsi routing being supported, the fo 3171 On arm64, gsi routing being supported, the following can happen:
3188 3172
3189 - in case no routing entry is associated to t 3173 - in case no routing entry is associated to this gsi, injection fails
3190 - in case the gsi is associated to an irqchip 3174 - in case the gsi is associated to an irqchip routing entry,
3191 irqchip.pin + 32 corresponds to the injecte 3175 irqchip.pin + 32 corresponds to the injected SPI ID.
3192 - in case the gsi is associated to an MSI rou 3176 - in case the gsi is associated to an MSI routing entry, the MSI
3193 message and device ID are translated into a 3177 message and device ID are translated into an LPI (support restricted
3194 to GICv3 ITS in-kernel emulation). 3178 to GICv3 ITS in-kernel emulation).
3195 3179
3196 4.76 KVM_PPC_ALLOCATE_HTAB 3180 4.76 KVM_PPC_ALLOCATE_HTAB
3197 -------------------------- 3181 --------------------------
3198 3182
3199 :Capability: KVM_CAP_PPC_ALLOC_HTAB 3183 :Capability: KVM_CAP_PPC_ALLOC_HTAB
3200 :Architectures: powerpc 3184 :Architectures: powerpc
3201 :Type: vm ioctl 3185 :Type: vm ioctl
3202 :Parameters: Pointer to u32 containing hash t 3186 :Parameters: Pointer to u32 containing hash table order (in/out)
3203 :Returns: 0 on success, -1 on error 3187 :Returns: 0 on success, -1 on error
3204 3188
3205 This requests the host kernel to allocate an 3189 This requests the host kernel to allocate an MMU hash table for a
3206 guest using the PAPR paravirtualization inter 3190 guest using the PAPR paravirtualization interface. This only does
3207 anything if the kernel is configured to use t 3191 anything if the kernel is configured to use the Book 3S HV style of
3208 virtualization. Otherwise the capability doe 3192 virtualization. Otherwise the capability doesn't exist and the ioctl
3209 returns an ENOTTY error. The rest of this de 3193 returns an ENOTTY error. The rest of this description assumes Book 3S
3210 HV. 3194 HV.
3211 3195
3212 There must be no vcpus running when this ioct 3196 There must be no vcpus running when this ioctl is called; if there
3213 are, it will do nothing and return an EBUSY e 3197 are, it will do nothing and return an EBUSY error.
3214 3198
3215 The parameter is a pointer to a 32-bit unsign 3199 The parameter is a pointer to a 32-bit unsigned integer variable
3216 containing the order (log base 2) of the desi 3200 containing the order (log base 2) of the desired size of the hash
3217 table, which must be between 18 and 46. On s 3201 table, which must be between 18 and 46. On successful return from the
3218 ioctl, the value will not be changed by the k 3202 ioctl, the value will not be changed by the kernel.
3219 3203
3220 If no hash table has been allocated when any 3204 If no hash table has been allocated when any vcpu is asked to run
3221 (with the KVM_RUN ioctl), the host kernel wil 3205 (with the KVM_RUN ioctl), the host kernel will allocate a
3222 default-sized hash table (16 MB). 3206 default-sized hash table (16 MB).
3223 3207
3224 If this ioctl is called when a hash table has 3208 If this ioctl is called when a hash table has already been allocated,
3225 with a different order from the existing hash 3209 with a different order from the existing hash table, the existing hash
3226 table will be freed and a new one allocated. 3210 table will be freed and a new one allocated. If this is ioctl is
3227 called when a hash table has already been all 3211 called when a hash table has already been allocated of the same order
3228 as specified, the kernel will clear out the e 3212 as specified, the kernel will clear out the existing hash table (zero
3229 all HPTEs). In either case, if the guest is 3213 all HPTEs). In either case, if the guest is using the virtualized
3230 real-mode area (VRMA) facility, the kernel wi 3214 real-mode area (VRMA) facility, the kernel will re-create the VMRA
3231 HPTEs on the next KVM_RUN of any vcpu. 3215 HPTEs on the next KVM_RUN of any vcpu.
3232 3216
3233 4.77 KVM_S390_INTERRUPT 3217 4.77 KVM_S390_INTERRUPT
3234 ----------------------- 3218 -----------------------
3235 3219
3236 :Capability: basic 3220 :Capability: basic
3237 :Architectures: s390 3221 :Architectures: s390
3238 :Type: vm ioctl, vcpu ioctl 3222 :Type: vm ioctl, vcpu ioctl
3239 :Parameters: struct kvm_s390_interrupt (in) 3223 :Parameters: struct kvm_s390_interrupt (in)
3240 :Returns: 0 on success, -1 on error 3224 :Returns: 0 on success, -1 on error
3241 3225
3242 Allows to inject an interrupt to the guest. I 3226 Allows to inject an interrupt to the guest. Interrupts can be floating
3243 (vm ioctl) or per cpu (vcpu ioctl), depending 3227 (vm ioctl) or per cpu (vcpu ioctl), depending on the interrupt type.
3244 3228
3245 Interrupt parameters are passed via kvm_s390_ 3229 Interrupt parameters are passed via kvm_s390_interrupt::
3246 3230
3247 struct kvm_s390_interrupt { 3231 struct kvm_s390_interrupt {
3248 __u32 type; 3232 __u32 type;
3249 __u32 parm; 3233 __u32 parm;
3250 __u64 parm64; 3234 __u64 parm64;
3251 }; 3235 };
3252 3236
3253 type can be one of the following: 3237 type can be one of the following:
3254 3238
3255 KVM_S390_SIGP_STOP (vcpu) 3239 KVM_S390_SIGP_STOP (vcpu)
3256 - sigp stop; optional flags in parm 3240 - sigp stop; optional flags in parm
3257 KVM_S390_PROGRAM_INT (vcpu) 3241 KVM_S390_PROGRAM_INT (vcpu)
3258 - program check; code in parm 3242 - program check; code in parm
3259 KVM_S390_SIGP_SET_PREFIX (vcpu) 3243 KVM_S390_SIGP_SET_PREFIX (vcpu)
3260 - sigp set prefix; prefix address in parm 3244 - sigp set prefix; prefix address in parm
3261 KVM_S390_RESTART (vcpu) 3245 KVM_S390_RESTART (vcpu)
3262 - restart 3246 - restart
3263 KVM_S390_INT_CLOCK_COMP (vcpu) 3247 KVM_S390_INT_CLOCK_COMP (vcpu)
3264 - clock comparator interrupt 3248 - clock comparator interrupt
3265 KVM_S390_INT_CPU_TIMER (vcpu) 3249 KVM_S390_INT_CPU_TIMER (vcpu)
3266 - CPU timer interrupt 3250 - CPU timer interrupt
3267 KVM_S390_INT_VIRTIO (vm) 3251 KVM_S390_INT_VIRTIO (vm)
3268 - virtio external interrupt; external int 3252 - virtio external interrupt; external interrupt
3269 parameters in parm and parm64 3253 parameters in parm and parm64
3270 KVM_S390_INT_SERVICE (vm) 3254 KVM_S390_INT_SERVICE (vm)
3271 - sclp external interrupt; sclp parameter 3255 - sclp external interrupt; sclp parameter in parm
3272 KVM_S390_INT_EMERGENCY (vcpu) 3256 KVM_S390_INT_EMERGENCY (vcpu)
3273 - sigp emergency; source cpu in parm 3257 - sigp emergency; source cpu in parm
3274 KVM_S390_INT_EXTERNAL_CALL (vcpu) 3258 KVM_S390_INT_EXTERNAL_CALL (vcpu)
3275 - sigp external call; source cpu in parm 3259 - sigp external call; source cpu in parm
3276 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) 3260 KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm)
3277 - compound value to indicate an 3261 - compound value to indicate an
3278 I/O interrupt (ai - adapter interrupt; 3262 I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
3279 I/O interruption parameters in parm (su 3263 I/O interruption parameters in parm (subchannel) and parm64 (intparm,
3280 interruption subclass) 3264 interruption subclass)
3281 KVM_S390_MCHK (vm, vcpu) 3265 KVM_S390_MCHK (vm, vcpu)
3282 - machine check interrupt; cr 14 bits in 3266 - machine check interrupt; cr 14 bits in parm, machine check interrupt
3283 code in parm64 (note that machine check 3267 code in parm64 (note that machine checks needing further payload are not
3284 supported by this ioctl) 3268 supported by this ioctl)
3285 3269
3286 This is an asynchronous vcpu ioctl and can be 3270 This is an asynchronous vcpu ioctl and can be invoked from any thread.
3287 3271
3288 4.78 KVM_PPC_GET_HTAB_FD 3272 4.78 KVM_PPC_GET_HTAB_FD
3289 ------------------------ 3273 ------------------------
3290 3274
3291 :Capability: KVM_CAP_PPC_HTAB_FD 3275 :Capability: KVM_CAP_PPC_HTAB_FD
3292 :Architectures: powerpc 3276 :Architectures: powerpc
3293 :Type: vm ioctl 3277 :Type: vm ioctl
3294 :Parameters: Pointer to struct kvm_get_htab_f 3278 :Parameters: Pointer to struct kvm_get_htab_fd (in)
3295 :Returns: file descriptor number (>= 0) on su 3279 :Returns: file descriptor number (>= 0) on success, -1 on error
3296 3280
3297 This returns a file descriptor that can be us 3281 This returns a file descriptor that can be used either to read out the
3298 entries in the guest's hashed page table (HPT 3282 entries in the guest's hashed page table (HPT), or to write entries to
3299 initialize the HPT. The returned fd can only 3283 initialize the HPT. The returned fd can only be written to if the
3300 KVM_GET_HTAB_WRITE bit is set in the flags fi 3284 KVM_GET_HTAB_WRITE bit is set in the flags field of the argument, and
3301 can only be read if that bit is clear. The a 3285 can only be read if that bit is clear. The argument struct looks like
3302 this:: 3286 this::
3303 3287
3304 /* For KVM_PPC_GET_HTAB_FD */ 3288 /* For KVM_PPC_GET_HTAB_FD */
3305 struct kvm_get_htab_fd { 3289 struct kvm_get_htab_fd {
3306 __u64 flags; 3290 __u64 flags;
3307 __u64 start_index; 3291 __u64 start_index;
3308 __u64 reserved[2]; 3292 __u64 reserved[2];
3309 }; 3293 };
3310 3294
3311 /* Values for kvm_get_htab_fd.flags */ 3295 /* Values for kvm_get_htab_fd.flags */
3312 #define KVM_GET_HTAB_BOLTED_ONLY ((__u 3296 #define KVM_GET_HTAB_BOLTED_ONLY ((__u64)0x1)
3313 #define KVM_GET_HTAB_WRITE ((__u 3297 #define KVM_GET_HTAB_WRITE ((__u64)0x2)
3314 3298
3315 The 'start_index' field gives the index in th 3299 The 'start_index' field gives the index in the HPT of the entry at
3316 which to start reading. It is ignored when w 3300 which to start reading. It is ignored when writing.
3317 3301
3318 Reads on the fd will initially supply informa 3302 Reads on the fd will initially supply information about all
3319 "interesting" HPT entries. Interesting entri 3303 "interesting" HPT entries. Interesting entries are those with the
3320 bolted bit set, if the KVM_GET_HTAB_BOLTED_ON 3304 bolted bit set, if the KVM_GET_HTAB_BOLTED_ONLY bit is set, otherwise
3321 all entries. When the end of the HPT is reac 3305 all entries. When the end of the HPT is reached, the read() will
3322 return. If read() is called again on the fd, 3306 return. If read() is called again on the fd, it will start again from
3323 the beginning of the HPT, but will only retur 3307 the beginning of the HPT, but will only return HPT entries that have
3324 changed since they were last read. 3308 changed since they were last read.
3325 3309
3326 Data read or written is structured as a heade 3310 Data read or written is structured as a header (8 bytes) followed by a
3327 series of valid HPT entries (16 bytes) each. 3311 series of valid HPT entries (16 bytes) each. The header indicates how
3328 many valid HPT entries there are and how many 3312 many valid HPT entries there are and how many invalid entries follow
3329 the valid entries. The invalid entries are n 3313 the valid entries. The invalid entries are not represented explicitly
3330 in the stream. The header format is:: 3314 in the stream. The header format is::
3331 3315
3332 struct kvm_get_htab_header { 3316 struct kvm_get_htab_header {
3333 __u32 index; 3317 __u32 index;
3334 __u16 n_valid; 3318 __u16 n_valid;
3335 __u16 n_invalid; 3319 __u16 n_invalid;
3336 }; 3320 };
3337 3321
3338 Writes to the fd create HPT entries starting 3322 Writes to the fd create HPT entries starting at the index given in the
3339 header; first 'n_valid' valid entries with co 3323 header; first 'n_valid' valid entries with contents from the data
3340 written, then 'n_invalid' invalid entries, in 3324 written, then 'n_invalid' invalid entries, invalidating any previously
3341 valid entries found. 3325 valid entries found.
3342 3326
3343 4.79 KVM_CREATE_DEVICE 3327 4.79 KVM_CREATE_DEVICE
3344 ---------------------- 3328 ----------------------
3345 3329
3346 :Capability: KVM_CAP_DEVICE_CTRL 3330 :Capability: KVM_CAP_DEVICE_CTRL
3347 :Architectures: all 3331 :Architectures: all
3348 :Type: vm ioctl 3332 :Type: vm ioctl
3349 :Parameters: struct kvm_create_device (in/out 3333 :Parameters: struct kvm_create_device (in/out)
3350 :Returns: 0 on success, -1 on error 3334 :Returns: 0 on success, -1 on error
3351 3335
3352 Errors: 3336 Errors:
3353 3337
3354 ====== =================================== 3338 ====== =======================================================
3355 ENODEV The device type is unknown or unsup 3339 ENODEV The device type is unknown or unsupported
3356 EEXIST Device already created, and this ty 3340 EEXIST Device already created, and this type of device may not
3357 be instantiated multiple times 3341 be instantiated multiple times
3358 ====== =================================== 3342 ====== =======================================================
3359 3343
3360 Other error conditions may be defined by in 3344 Other error conditions may be defined by individual device types or
3361 have their standard meanings. 3345 have their standard meanings.
3362 3346
3363 Creates an emulated device in the kernel. Th 3347 Creates an emulated device in the kernel. The file descriptor returned
3364 in fd can be used with KVM_SET/GET/HAS_DEVICE 3348 in fd can be used with KVM_SET/GET/HAS_DEVICE_ATTR.
3365 3349
3366 If the KVM_CREATE_DEVICE_TEST flag is set, on 3350 If the KVM_CREATE_DEVICE_TEST flag is set, only test whether the
3367 device type is supported (not necessarily whe 3351 device type is supported (not necessarily whether it can be created
3368 in the current vm). 3352 in the current vm).
3369 3353
3370 Individual devices should not define flags. 3354 Individual devices should not define flags. Attributes should be used
3371 for specifying any behavior that is not impli 3355 for specifying any behavior that is not implied by the device type
3372 number. 3356 number.
3373 3357
3374 :: 3358 ::
3375 3359
3376 struct kvm_create_device { 3360 struct kvm_create_device {
3377 __u32 type; /* in: KVM_DEV_TYPE_x 3361 __u32 type; /* in: KVM_DEV_TYPE_xxx */
3378 __u32 fd; /* out: device handle 3362 __u32 fd; /* out: device handle */
3379 __u32 flags; /* in: KVM_CREATE_DEV 3363 __u32 flags; /* in: KVM_CREATE_DEVICE_xxx */
3380 }; 3364 };
3381 3365
3382 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR 3366 4.80 KVM_SET_DEVICE_ATTR/KVM_GET_DEVICE_ATTR
3383 -------------------------------------------- 3367 --------------------------------------------
3384 3368
3385 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ 3369 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
3386 KVM_CAP_VCPU_ATTRIBUTES for vcpu 3370 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
3387 KVM_CAP_SYS_ATTRIBUTES for syste 3371 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device (no set)
3388 :Architectures: x86, arm64, s390 3372 :Architectures: x86, arm64, s390
3389 :Type: device ioctl, vm ioctl, vcpu ioctl 3373 :Type: device ioctl, vm ioctl, vcpu ioctl
3390 :Parameters: struct kvm_device_attr 3374 :Parameters: struct kvm_device_attr
3391 :Returns: 0 on success, -1 on error 3375 :Returns: 0 on success, -1 on error
3392 3376
3393 Errors: 3377 Errors:
3394 3378
3395 ===== =================================== 3379 ===== =============================================================
3396 ENXIO The group or attribute is unknown/u 3380 ENXIO The group or attribute is unknown/unsupported for this device
3397 or hardware support is missing. 3381 or hardware support is missing.
3398 EPERM The attribute cannot (currently) be 3382 EPERM The attribute cannot (currently) be accessed this way
3399 (e.g. read-only attribute, or attri 3383 (e.g. read-only attribute, or attribute that only makes
3400 sense when the device is in a diffe 3384 sense when the device is in a different state)
3401 ===== =================================== 3385 ===== =============================================================
3402 3386
3403 Other error conditions may be defined by in 3387 Other error conditions may be defined by individual device types.
3404 3388
3405 Gets/sets a specified piece of device configu 3389 Gets/sets a specified piece of device configuration and/or state. The
3406 semantics are device-specific. See individua 3390 semantics are device-specific. See individual device documentation in
3407 the "devices" directory. As with ONE_REG, th 3391 the "devices" directory. As with ONE_REG, the size of the data
3408 transferred is defined by the particular attr 3392 transferred is defined by the particular attribute.
3409 3393
3410 :: 3394 ::
3411 3395
3412 struct kvm_device_attr { 3396 struct kvm_device_attr {
3413 __u32 flags; /* no flags c 3397 __u32 flags; /* no flags currently defined */
3414 __u32 group; /* device-def 3398 __u32 group; /* device-defined */
3415 __u64 attr; /* group-defi 3399 __u64 attr; /* group-defined */
3416 __u64 addr; /* userspace 3400 __u64 addr; /* userspace address of attr data */
3417 }; 3401 };
3418 3402
3419 4.81 KVM_HAS_DEVICE_ATTR 3403 4.81 KVM_HAS_DEVICE_ATTR
3420 ------------------------ 3404 ------------------------
3421 3405
3422 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ 3406 :Capability: KVM_CAP_DEVICE_CTRL, KVM_CAP_VM_ATTRIBUTES for vm device,
3423 KVM_CAP_VCPU_ATTRIBUTES for vcpu 3407 KVM_CAP_VCPU_ATTRIBUTES for vcpu device
3424 KVM_CAP_SYS_ATTRIBUTES for syste 3408 KVM_CAP_SYS_ATTRIBUTES for system (/dev/kvm) device
3425 :Type: device ioctl, vm ioctl, vcpu ioctl 3409 :Type: device ioctl, vm ioctl, vcpu ioctl
3426 :Parameters: struct kvm_device_attr 3410 :Parameters: struct kvm_device_attr
3427 :Returns: 0 on success, -1 on error 3411 :Returns: 0 on success, -1 on error
3428 3412
3429 Errors: 3413 Errors:
3430 3414
3431 ===== =================================== 3415 ===== =============================================================
3432 ENXIO The group or attribute is unknown/u 3416 ENXIO The group or attribute is unknown/unsupported for this device
3433 or hardware support is missing. 3417 or hardware support is missing.
3434 ===== =================================== 3418 ===== =============================================================
3435 3419
3436 Tests whether a device supports a particular 3420 Tests whether a device supports a particular attribute. A successful
3437 return indicates the attribute is implemented 3421 return indicates the attribute is implemented. It does not necessarily
3438 indicate that the attribute can be read or wr 3422 indicate that the attribute can be read or written in the device's
3439 current state. "addr" is ignored. 3423 current state. "addr" is ignored.
3440 3424
3441 .. _KVM_ARM_VCPU_INIT: 3425 .. _KVM_ARM_VCPU_INIT:
3442 3426
3443 4.82 KVM_ARM_VCPU_INIT 3427 4.82 KVM_ARM_VCPU_INIT
3444 ---------------------- 3428 ----------------------
3445 3429
3446 :Capability: basic 3430 :Capability: basic
3447 :Architectures: arm64 3431 :Architectures: arm64
3448 :Type: vcpu ioctl 3432 :Type: vcpu ioctl
3449 :Parameters: struct kvm_vcpu_init (in) 3433 :Parameters: struct kvm_vcpu_init (in)
3450 :Returns: 0 on success; -1 on error 3434 :Returns: 0 on success; -1 on error
3451 3435
3452 Errors: 3436 Errors:
3453 3437
3454 ====== ================================ 3438 ====== =================================================================
3455 EINVAL the target is unknown, or the co 3439 EINVAL the target is unknown, or the combination of features is invalid.
3456 ENOENT a features bit specified is unkn 3440 ENOENT a features bit specified is unknown.
3457 ====== ================================ 3441 ====== =================================================================
3458 3442
3459 This tells KVM what type of CPU to present to 3443 This tells KVM what type of CPU to present to the guest, and what
3460 optional features it should have. This will 3444 optional features it should have. This will cause a reset of the cpu
3461 registers to their initial values. If this i 3445 registers to their initial values. If this is not called, KVM_RUN will
3462 return ENOEXEC for that vcpu. 3446 return ENOEXEC for that vcpu.
3463 3447
3464 The initial values are defined as: 3448 The initial values are defined as:
3465 - Processor state: 3449 - Processor state:
3466 * AArch64: EL1h, D, A, I and 3450 * AArch64: EL1h, D, A, I and F bits set. All other bits
3467 are cleared. 3451 are cleared.
3468 * AArch32: SVC, A, I and F bi 3452 * AArch32: SVC, A, I and F bits set. All other bits are
3469 cleared. 3453 cleared.
3470 - General Purpose registers, includin 3454 - General Purpose registers, including PC and SP: set to 0
3471 - FPSIMD/NEON registers: set to 0 3455 - FPSIMD/NEON registers: set to 0
3472 - SVE registers: set to 0 3456 - SVE registers: set to 0
3473 - System registers: Reset to their ar 3457 - System registers: Reset to their architecturally defined
3474 values as for a warm reset to EL1 ( 3458 values as for a warm reset to EL1 (resp. SVC)
3475 3459
3476 Note that because some registers reflect mach 3460 Note that because some registers reflect machine topology, all vcpus
3477 should be created before this ioctl is invoke 3461 should be created before this ioctl is invoked.
3478 3462
3479 Userspace can call this function multiple tim 3463 Userspace can call this function multiple times for a given vcpu, including
3480 after the vcpu has been run. This will reset 3464 after the vcpu has been run. This will reset the vcpu to its initial
3481 state. All calls to this function after the i 3465 state. All calls to this function after the initial call must use the same
3482 target and same set of feature flags, otherwi 3466 target and same set of feature flags, otherwise EINVAL will be returned.
3483 3467
3484 Possible features: 3468 Possible features:
3485 3469
3486 - KVM_ARM_VCPU_POWER_OFF: Starts the 3470 - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
3487 Depends on KVM_CAP_ARM_PSCI. If no 3471 Depends on KVM_CAP_ARM_PSCI. If not set, the CPU will be powered on
3488 and execute guest code when KVM_RUN 3472 and execute guest code when KVM_RUN is called.
3489 - KVM_ARM_VCPU_EL1_32BIT: Starts the 3473 - KVM_ARM_VCPU_EL1_32BIT: Starts the CPU in a 32bit mode.
3490 Depends on KVM_CAP_ARM_EL1_32BIT (a 3474 Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
3491 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI 3475 - KVM_ARM_VCPU_PSCI_0_2: Emulate PSCI v0.2 (or a future revision
3492 backward compatible with v0.2) for 3476 backward compatible with v0.2) for the CPU.
3493 Depends on KVM_CAP_ARM_PSCI_0_2. 3477 Depends on KVM_CAP_ARM_PSCI_0_2.
3494 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 3478 - KVM_ARM_VCPU_PMU_V3: Emulate PMUv3 for the CPU.
3495 Depends on KVM_CAP_ARM_PMU_V3. 3479 Depends on KVM_CAP_ARM_PMU_V3.
3496 3480
3497 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enabl 3481 - KVM_ARM_VCPU_PTRAUTH_ADDRESS: Enables Address Pointer authentication
3498 for arm64 only. 3482 for arm64 only.
3499 Depends on KVM_CAP_ARM_PTRAUTH_ADDR 3483 Depends on KVM_CAP_ARM_PTRAUTH_ADDRESS.
3500 If KVM_CAP_ARM_PTRAUTH_ADDRESS and 3484 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are
3501 both present, then both KVM_ARM_VCP 3485 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and
3502 KVM_ARM_VCPU_PTRAUTH_GENERIC must b 3486 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be
3503 requested. 3487 requested.
3504 3488
3505 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enabl 3489 - KVM_ARM_VCPU_PTRAUTH_GENERIC: Enables Generic Pointer authentication
3506 for arm64 only. 3490 for arm64 only.
3507 Depends on KVM_CAP_ARM_PTRAUTH_GENE 3491 Depends on KVM_CAP_ARM_PTRAUTH_GENERIC.
3508 If KVM_CAP_ARM_PTRAUTH_ADDRESS and 3492 If KVM_CAP_ARM_PTRAUTH_ADDRESS and KVM_CAP_ARM_PTRAUTH_GENERIC are
3509 both present, then both KVM_ARM_VCP 3493 both present, then both KVM_ARM_VCPU_PTRAUTH_ADDRESS and
3510 KVM_ARM_VCPU_PTRAUTH_GENERIC must b 3494 KVM_ARM_VCPU_PTRAUTH_GENERIC must be requested or neither must be
3511 requested. 3495 requested.
3512 3496
3513 - KVM_ARM_VCPU_SVE: Enables SVE for t 3497 - KVM_ARM_VCPU_SVE: Enables SVE for the CPU (arm64 only).
3514 Depends on KVM_CAP_ARM_SVE. 3498 Depends on KVM_CAP_ARM_SVE.
3515 Requires KVM_ARM_VCPU_FINALIZE(KVM_ 3499 Requires KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3516 3500
3517 * After KVM_ARM_VCPU_INIT: 3501 * After KVM_ARM_VCPU_INIT:
3518 3502
3519 - KVM_REG_ARM64_SVE_VLS may be 3503 - KVM_REG_ARM64_SVE_VLS may be read using KVM_GET_ONE_REG: the
3520 initial value of this pseudo- 3504 initial value of this pseudo-register indicates the best set of
3521 vector lengths possible for a 3505 vector lengths possible for a vcpu on this host.
3522 3506
3523 * Before KVM_ARM_VCPU_FINALIZE(KVM 3507 * Before KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3524 3508
3525 - KVM_RUN and KVM_GET_REG_LIST 3509 - KVM_RUN and KVM_GET_REG_LIST are not available;
3526 3510
3527 - KVM_GET_ONE_REG and KVM_SET_O 3511 - KVM_GET_ONE_REG and KVM_SET_ONE_REG cannot be used to access
3528 the scalable architectural SV 3512 the scalable architectural SVE registers
3529 KVM_REG_ARM64_SVE_ZREG(), KVM 3513 KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() or
3530 KVM_REG_ARM64_SVE_FFR; 3514 KVM_REG_ARM64_SVE_FFR;
3531 3515
3532 - KVM_REG_ARM64_SVE_VLS may opt 3516 - KVM_REG_ARM64_SVE_VLS may optionally be written using
3533 KVM_SET_ONE_REG, to modify th 3517 KVM_SET_ONE_REG, to modify the set of vector lengths available
3534 for the vcpu. 3518 for the vcpu.
3535 3519
3536 * After KVM_ARM_VCPU_FINALIZE(KVM_ 3520 * After KVM_ARM_VCPU_FINALIZE(KVM_ARM_VCPU_SVE):
3537 3521
3538 - the KVM_REG_ARM64_SVE_VLS pse 3522 - the KVM_REG_ARM64_SVE_VLS pseudo-register is immutable, and can
3539 no longer be written using KV 3523 no longer be written using KVM_SET_ONE_REG.
3540 3524
3541 4.83 KVM_ARM_PREFERRED_TARGET 3525 4.83 KVM_ARM_PREFERRED_TARGET
3542 ----------------------------- 3526 -----------------------------
3543 3527
3544 :Capability: basic 3528 :Capability: basic
3545 :Architectures: arm64 3529 :Architectures: arm64
3546 :Type: vm ioctl 3530 :Type: vm ioctl
3547 :Parameters: struct kvm_vcpu_init (out) 3531 :Parameters: struct kvm_vcpu_init (out)
3548 :Returns: 0 on success; -1 on error 3532 :Returns: 0 on success; -1 on error
3549 3533
3550 Errors: 3534 Errors:
3551 3535
3552 ====== ================================ 3536 ====== ==========================================
3553 ENODEV no preferred target available fo 3537 ENODEV no preferred target available for the host
3554 ====== ================================ 3538 ====== ==========================================
3555 3539
3556 This queries KVM for preferred CPU target typ 3540 This queries KVM for preferred CPU target type which can be emulated
3557 by KVM on underlying host. 3541 by KVM on underlying host.
3558 3542
3559 The ioctl returns struct kvm_vcpu_init instan 3543 The ioctl returns struct kvm_vcpu_init instance containing information
3560 about preferred CPU target type and recommend 3544 about preferred CPU target type and recommended features for it. The
3561 kvm_vcpu_init->features bitmap returned will 3545 kvm_vcpu_init->features bitmap returned will have feature bits set if
3562 the preferred target recommends setting these 3546 the preferred target recommends setting these features, but this is
3563 not mandatory. 3547 not mandatory.
3564 3548
3565 The information returned by this ioctl can be 3549 The information returned by this ioctl can be used to prepare an instance
3566 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT 3550 of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
3567 VCPU matching underlying host. 3551 VCPU matching underlying host.
3568 3552
3569 3553
3570 4.84 KVM_GET_REG_LIST 3554 4.84 KVM_GET_REG_LIST
3571 --------------------- 3555 ---------------------
3572 3556
3573 :Capability: basic 3557 :Capability: basic
3574 :Architectures: arm64, mips, riscv 3558 :Architectures: arm64, mips, riscv
3575 :Type: vcpu ioctl 3559 :Type: vcpu ioctl
3576 :Parameters: struct kvm_reg_list (in/out) 3560 :Parameters: struct kvm_reg_list (in/out)
3577 :Returns: 0 on success; -1 on error 3561 :Returns: 0 on success; -1 on error
3578 3562
3579 Errors: 3563 Errors:
3580 3564
3581 ===== ================================ 3565 ===== ==============================================================
3582 E2BIG the reg index list is too big to 3566 E2BIG the reg index list is too big to fit in the array specified by
3583 the user (the number required wi 3567 the user (the number required will be written into n).
3584 ===== ================================ 3568 ===== ==============================================================
3585 3569
3586 :: 3570 ::
3587 3571
3588 struct kvm_reg_list { 3572 struct kvm_reg_list {
3589 __u64 n; /* number of registers in re 3573 __u64 n; /* number of registers in reg[] */
3590 __u64 reg[0]; 3574 __u64 reg[0];
3591 }; 3575 };
3592 3576
3593 This ioctl returns the guest registers that a 3577 This ioctl returns the guest registers that are supported for the
3594 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls. 3578 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
3595 3579
3596 3580
3597 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated) 3581 4.85 KVM_ARM_SET_DEVICE_ADDR (deprecated)
3598 ----------------------------------------- 3582 -----------------------------------------
3599 3583
3600 :Capability: KVM_CAP_ARM_SET_DEVICE_ADDR 3584 :Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
3601 :Architectures: arm64 3585 :Architectures: arm64
3602 :Type: vm ioctl 3586 :Type: vm ioctl
3603 :Parameters: struct kvm_arm_device_address (i 3587 :Parameters: struct kvm_arm_device_address (in)
3604 :Returns: 0 on success, -1 on error 3588 :Returns: 0 on success, -1 on error
3605 3589
3606 Errors: 3590 Errors:
3607 3591
3608 ====== =================================== 3592 ====== ============================================
3609 ENODEV The device id is unknown 3593 ENODEV The device id is unknown
3610 ENXIO Device not supported on current sys 3594 ENXIO Device not supported on current system
3611 EEXIST Address already set 3595 EEXIST Address already set
3612 E2BIG Address outside guest physical addr 3596 E2BIG Address outside guest physical address space
3613 EBUSY Address overlaps with other device 3597 EBUSY Address overlaps with other device range
3614 ====== =================================== 3598 ====== ============================================
3615 3599
3616 :: 3600 ::
3617 3601
3618 struct kvm_arm_device_addr { 3602 struct kvm_arm_device_addr {
3619 __u64 id; 3603 __u64 id;
3620 __u64 addr; 3604 __u64 addr;
3621 }; 3605 };
3622 3606
3623 Specify a device address in the guest's physi 3607 Specify a device address in the guest's physical address space where guests
3624 can access emulated or directly exposed devic 3608 can access emulated or directly exposed devices, which the host kernel needs
3625 to know about. The id field is an architectur 3609 to know about. The id field is an architecture specific identifier for a
3626 specific device. 3610 specific device.
3627 3611
3628 arm64 divides the id field into two parts, a 3612 arm64 divides the id field into two parts, a device id and an
3629 address type id specific to the individual de 3613 address type id specific to the individual device::
3630 3614
3631 bits: | 63 ... 32 | 31 ... 3615 bits: | 63 ... 32 | 31 ... 16 | 15 ... 0 |
3632 field: | 0x00000000 | devic 3616 field: | 0x00000000 | device id | addr type id |
3633 3617
3634 arm64 currently only require this when using 3618 arm64 currently only require this when using the in-kernel GIC
3635 support for the hardware VGIC features, using 3619 support for the hardware VGIC features, using KVM_ARM_DEVICE_VGIC_V2
3636 as the device id. When setting the base addr 3620 as the device id. When setting the base address for the guest's
3637 mapping of the VGIC virtual CPU and distribut 3621 mapping of the VGIC virtual CPU and distributor interface, the ioctl
3638 must be called after calling KVM_CREATE_IRQCH 3622 must be called after calling KVM_CREATE_IRQCHIP, but before calling
3639 KVM_RUN on any of the VCPUs. Calling this io 3623 KVM_RUN on any of the VCPUs. Calling this ioctl twice for any of the
3640 base addresses will return -EEXIST. 3624 base addresses will return -EEXIST.
3641 3625
3642 Note, this IOCTL is deprecated and the more f 3626 Note, this IOCTL is deprecated and the more flexible SET/GET_DEVICE_ATTR API
3643 should be used instead. 3627 should be used instead.
3644 3628
3645 3629
3646 4.86 KVM_PPC_RTAS_DEFINE_TOKEN 3630 4.86 KVM_PPC_RTAS_DEFINE_TOKEN
3647 ------------------------------ 3631 ------------------------------
3648 3632
3649 :Capability: KVM_CAP_PPC_RTAS 3633 :Capability: KVM_CAP_PPC_RTAS
3650 :Architectures: ppc 3634 :Architectures: ppc
3651 :Type: vm ioctl 3635 :Type: vm ioctl
3652 :Parameters: struct kvm_rtas_token_args 3636 :Parameters: struct kvm_rtas_token_args
3653 :Returns: 0 on success, -1 on error 3637 :Returns: 0 on success, -1 on error
3654 3638
3655 Defines a token value for a RTAS (Run Time Ab 3639 Defines a token value for a RTAS (Run Time Abstraction Services)
3656 service in order to allow it to be handled in 3640 service in order to allow it to be handled in the kernel. The
3657 argument struct gives the name of the service 3641 argument struct gives the name of the service, which must be the name
3658 of a service that has a kernel-side implement 3642 of a service that has a kernel-side implementation. If the token
3659 value is non-zero, it will be associated with 3643 value is non-zero, it will be associated with that service, and
3660 subsequent RTAS calls by the guest specifying 3644 subsequent RTAS calls by the guest specifying that token will be
3661 handled by the kernel. If the token value is 3645 handled by the kernel. If the token value is 0, then any token
3662 associated with the service will be forgotten 3646 associated with the service will be forgotten, and subsequent RTAS
3663 calls by the guest for that service will be p 3647 calls by the guest for that service will be passed to userspace to be
3664 handled. 3648 handled.
3665 3649
3666 4.87 KVM_SET_GUEST_DEBUG 3650 4.87 KVM_SET_GUEST_DEBUG
3667 ------------------------ 3651 ------------------------
3668 3652
3669 :Capability: KVM_CAP_SET_GUEST_DEBUG 3653 :Capability: KVM_CAP_SET_GUEST_DEBUG
3670 :Architectures: x86, s390, ppc, arm64 3654 :Architectures: x86, s390, ppc, arm64
3671 :Type: vcpu ioctl 3655 :Type: vcpu ioctl
3672 :Parameters: struct kvm_guest_debug (in) 3656 :Parameters: struct kvm_guest_debug (in)
3673 :Returns: 0 on success; -1 on error 3657 :Returns: 0 on success; -1 on error
3674 3658
3675 :: 3659 ::
3676 3660
3677 struct kvm_guest_debug { 3661 struct kvm_guest_debug {
3678 __u32 control; 3662 __u32 control;
3679 __u32 pad; 3663 __u32 pad;
3680 struct kvm_guest_debug_arch arch; 3664 struct kvm_guest_debug_arch arch;
3681 }; 3665 };
3682 3666
3683 Set up the processor specific debug registers 3667 Set up the processor specific debug registers and configure vcpu for
3684 handling guest debug events. There are two pa 3668 handling guest debug events. There are two parts to the structure, the
3685 first a control bitfield indicates the type o 3669 first a control bitfield indicates the type of debug events to handle
3686 when running. Common control bits are: 3670 when running. Common control bits are:
3687 3671
3688 - KVM_GUESTDBG_ENABLE: guest debuggi 3672 - KVM_GUESTDBG_ENABLE: guest debugging is enabled
3689 - KVM_GUESTDBG_SINGLESTEP: the next run 3673 - KVM_GUESTDBG_SINGLESTEP: the next run should single-step
3690 3674
3691 The top 16 bits of the control field are arch 3675 The top 16 bits of the control field are architecture specific control
3692 flags which can include the following: 3676 flags which can include the following:
3693 3677
3694 - KVM_GUESTDBG_USE_SW_BP: using softwar 3678 - KVM_GUESTDBG_USE_SW_BP: using software breakpoints [x86, arm64]
3695 - KVM_GUESTDBG_USE_HW_BP: using hardwar 3679 - KVM_GUESTDBG_USE_HW_BP: using hardware breakpoints [x86, s390]
3696 - KVM_GUESTDBG_USE_HW: using hardwar 3680 - KVM_GUESTDBG_USE_HW: using hardware debug events [arm64]
3697 - KVM_GUESTDBG_INJECT_DB: inject DB typ 3681 - KVM_GUESTDBG_INJECT_DB: inject DB type exception [x86]
3698 - KVM_GUESTDBG_INJECT_BP: inject BP typ 3682 - KVM_GUESTDBG_INJECT_BP: inject BP type exception [x86]
3699 - KVM_GUESTDBG_EXIT_PENDING: trigger an im 3683 - KVM_GUESTDBG_EXIT_PENDING: trigger an immediate guest exit [s390]
3700 - KVM_GUESTDBG_BLOCKIRQ: avoid injecti 3684 - KVM_GUESTDBG_BLOCKIRQ: avoid injecting interrupts/NMI/SMI [x86]
3701 3685
3702 For example KVM_GUESTDBG_USE_SW_BP indicates 3686 For example KVM_GUESTDBG_USE_SW_BP indicates that software breakpoints
3703 are enabled in memory so we need to ensure br 3687 are enabled in memory so we need to ensure breakpoint exceptions are
3704 correctly trapped and the KVM run loop exits 3688 correctly trapped and the KVM run loop exits at the breakpoint and not
3705 running off into the normal guest vector. For 3689 running off into the normal guest vector. For KVM_GUESTDBG_USE_HW_BP
3706 we need to ensure the guest vCPUs architectur 3690 we need to ensure the guest vCPUs architecture specific registers are
3707 updated to the correct (supplied) values. 3691 updated to the correct (supplied) values.
3708 3692
3709 The second part of the structure is architect 3693 The second part of the structure is architecture specific and
3710 typically contains a set of debug registers. 3694 typically contains a set of debug registers.
3711 3695
3712 For arm64 the number of debug registers is im 3696 For arm64 the number of debug registers is implementation defined and
3713 can be determined by querying the KVM_CAP_GUE 3697 can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
3714 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which 3698 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
3715 indicating the number of supported registers. 3699 indicating the number of supported registers.
3716 3700
3717 For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP ca 3701 For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether
3718 the single-step debug event (KVM_GUESTDBG_SIN 3702 the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported.
3719 3703
3720 Also when supported, KVM_CAP_SET_GUEST_DEBUG2 3704 Also when supported, KVM_CAP_SET_GUEST_DEBUG2 capability indicates the
3721 supported KVM_GUESTDBG_* bits in the control 3705 supported KVM_GUESTDBG_* bits in the control field.
3722 3706
3723 When debug events exit the main run loop with 3707 When debug events exit the main run loop with the reason
3724 KVM_EXIT_DEBUG with the kvm_debug_exit_arch p 3708 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
3725 structure containing architecture specific de 3709 structure containing architecture specific debug information.
3726 3710
3727 4.88 KVM_GET_EMULATED_CPUID 3711 4.88 KVM_GET_EMULATED_CPUID
3728 --------------------------- 3712 ---------------------------
3729 3713
3730 :Capability: KVM_CAP_EXT_EMUL_CPUID 3714 :Capability: KVM_CAP_EXT_EMUL_CPUID
3731 :Architectures: x86 3715 :Architectures: x86
3732 :Type: system ioctl 3716 :Type: system ioctl
3733 :Parameters: struct kvm_cpuid2 (in/out) 3717 :Parameters: struct kvm_cpuid2 (in/out)
3734 :Returns: 0 on success, -1 on error 3718 :Returns: 0 on success, -1 on error
3735 3719
3736 :: 3720 ::
3737 3721
3738 struct kvm_cpuid2 { 3722 struct kvm_cpuid2 {
3739 __u32 nent; 3723 __u32 nent;
3740 __u32 flags; 3724 __u32 flags;
3741 struct kvm_cpuid_entry2 entries[0]; 3725 struct kvm_cpuid_entry2 entries[0];
3742 }; 3726 };
3743 3727
3744 The member 'flags' is used for passing flags 3728 The member 'flags' is used for passing flags from userspace.
3745 3729
3746 :: 3730 ::
3747 3731
3748 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 3732 #define KVM_CPUID_FLAG_SIGNIFCANT_INDEX BIT(0)
3749 #define KVM_CPUID_FLAG_STATEFUL_FUNC 3733 #define KVM_CPUID_FLAG_STATEFUL_FUNC BIT(1) /* deprecated */
3750 #define KVM_CPUID_FLAG_STATE_READ_NEXT 3734 #define KVM_CPUID_FLAG_STATE_READ_NEXT BIT(2) /* deprecated */
3751 3735
3752 struct kvm_cpuid_entry2 { 3736 struct kvm_cpuid_entry2 {
3753 __u32 function; 3737 __u32 function;
3754 __u32 index; 3738 __u32 index;
3755 __u32 flags; 3739 __u32 flags;
3756 __u32 eax; 3740 __u32 eax;
3757 __u32 ebx; 3741 __u32 ebx;
3758 __u32 ecx; 3742 __u32 ecx;
3759 __u32 edx; 3743 __u32 edx;
3760 __u32 padding[3]; 3744 __u32 padding[3];
3761 }; 3745 };
3762 3746
3763 This ioctl returns x86 cpuid features which a 3747 This ioctl returns x86 cpuid features which are emulated by
3764 kvm.Userspace can use the information returne 3748 kvm.Userspace can use the information returned by this ioctl to query
3765 which features are emulated by kvm instead of 3749 which features are emulated by kvm instead of being present natively.
3766 3750
3767 Userspace invokes KVM_GET_EMULATED_CPUID by p 3751 Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
3768 structure with the 'nent' field indicating th 3752 structure with the 'nent' field indicating the number of entries in
3769 the variable-size array 'entries'. If the num 3753 the variable-size array 'entries'. If the number of entries is too low
3770 to describe the cpu capabilities, an error (E 3754 to describe the cpu capabilities, an error (E2BIG) is returned. If the
3771 number is too high, the 'nent' field is adjus 3755 number is too high, the 'nent' field is adjusted and an error (ENOMEM)
3772 is returned. If the number is just right, the 3756 is returned. If the number is just right, the 'nent' field is adjusted
3773 to the number of valid entries in the 'entrie 3757 to the number of valid entries in the 'entries' array, which is then
3774 filled. 3758 filled.
3775 3759
3776 The entries returned are the set CPUID bits o 3760 The entries returned are the set CPUID bits of the respective features
3777 which kvm emulates, as returned by the CPUID 3761 which kvm emulates, as returned by the CPUID instruction, with unknown
3778 or unsupported feature bits cleared. 3762 or unsupported feature bits cleared.
3779 3763
3780 Features like x2apic, for example, may not be 3764 Features like x2apic, for example, may not be present in the host cpu
3781 but are exposed by kvm in KVM_GET_SUPPORTED_C 3765 but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
3782 emulated efficiently and thus not included he 3766 emulated efficiently and thus not included here.
3783 3767
3784 The fields in each entry are defined as follo 3768 The fields in each entry are defined as follows:
3785 3769
3786 function: 3770 function:
3787 the eax value used to obtain the ent 3771 the eax value used to obtain the entry
3788 index: 3772 index:
3789 the ecx value used to obtain the ent 3773 the ecx value used to obtain the entry (for entries that are
3790 affected by ecx) 3774 affected by ecx)
3791 flags: 3775 flags:
3792 an OR of zero or more of the following: 3776 an OR of zero or more of the following:
3793 3777
3794 KVM_CPUID_FLAG_SIGNIFCANT_INDEX: 3778 KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
3795 if the index field is valid 3779 if the index field is valid
3796 3780
3797 eax, ebx, ecx, edx: 3781 eax, ebx, ecx, edx:
3798 3782
3799 the values returned by the cpuid ins 3783 the values returned by the cpuid instruction for
3800 this function/index combination 3784 this function/index combination
3801 3785
3802 4.89 KVM_S390_MEM_OP 3786 4.89 KVM_S390_MEM_OP
3803 -------------------- 3787 --------------------
3804 3788
3805 :Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S39 3789 :Capability: KVM_CAP_S390_MEM_OP, KVM_CAP_S390_PROTECTED, KVM_CAP_S390_MEM_OP_EXTENSION
3806 :Architectures: s390 3790 :Architectures: s390
3807 :Type: vm ioctl, vcpu ioctl 3791 :Type: vm ioctl, vcpu ioctl
3808 :Parameters: struct kvm_s390_mem_op (in) 3792 :Parameters: struct kvm_s390_mem_op (in)
3809 :Returns: = 0 on success, 3793 :Returns: = 0 on success,
3810 < 0 on generic error (e.g. -EFAULT 3794 < 0 on generic error (e.g. -EFAULT or -ENOMEM),
3811 16 bit program exception code if th 3795 16 bit program exception code if the access causes such an exception
3812 3796
3813 Read or write data from/to the VM's memory. 3797 Read or write data from/to the VM's memory.
3814 The KVM_CAP_S390_MEM_OP_EXTENSION capability 3798 The KVM_CAP_S390_MEM_OP_EXTENSION capability specifies what functionality is
3815 supported. 3799 supported.
3816 3800
3817 Parameters are specified via the following st 3801 Parameters are specified via the following structure::
3818 3802
3819 struct kvm_s390_mem_op { 3803 struct kvm_s390_mem_op {
3820 __u64 gaddr; /* the guest 3804 __u64 gaddr; /* the guest address */
3821 __u64 flags; /* flags */ 3805 __u64 flags; /* flags */
3822 __u32 size; /* amount of 3806 __u32 size; /* amount of bytes */
3823 __u32 op; /* type of op 3807 __u32 op; /* type of operation */
3824 __u64 buf; /* buffer in 3808 __u64 buf; /* buffer in userspace */
3825 union { 3809 union {
3826 struct { 3810 struct {
3827 __u8 ar; /* th 3811 __u8 ar; /* the access register number */
3828 __u8 key; /* ac 3812 __u8 key; /* access key, ignored if flag unset */
3829 __u8 pad1[6]; /* ig 3813 __u8 pad1[6]; /* ignored */
3830 __u64 old_addr; /* ig 3814 __u64 old_addr; /* ignored if flag unset */
3831 }; 3815 };
3832 __u32 sida_offset; /* offset 3816 __u32 sida_offset; /* offset into the sida */
3833 __u8 reserved[32]; /* ignored 3817 __u8 reserved[32]; /* ignored */
3834 }; 3818 };
3835 }; 3819 };
3836 3820
3837 The start address of the memory region has to 3821 The start address of the memory region has to be specified in the "gaddr"
3838 field, and the length of the region in the "s 3822 field, and the length of the region in the "size" field (which must not
3839 be 0). The maximum value for "size" can be ob 3823 be 0). The maximum value for "size" can be obtained by checking the
3840 KVM_CAP_S390_MEM_OP capability. "buf" is the 3824 KVM_CAP_S390_MEM_OP capability. "buf" is the buffer supplied by the
3841 userspace application where the read data sho 3825 userspace application where the read data should be written to for
3842 a read access, or where the data that should 3826 a read access, or where the data that should be written is stored for
3843 a write access. The "reserved" field is mean 3827 a write access. The "reserved" field is meant for future extensions.
3844 Reserved and unused values are ignored. Futur 3828 Reserved and unused values are ignored. Future extension that add members must
3845 introduce new flags. 3829 introduce new flags.
3846 3830
3847 The type of operation is specified in the "op 3831 The type of operation is specified in the "op" field. Flags modifying
3848 their behavior can be set in the "flags" fiel 3832 their behavior can be set in the "flags" field. Undefined flag bits must
3849 be set to 0. 3833 be set to 0.
3850 3834
3851 Possible operations are: 3835 Possible operations are:
3852 * ``KVM_S390_MEMOP_LOGICAL_READ`` 3836 * ``KVM_S390_MEMOP_LOGICAL_READ``
3853 * ``KVM_S390_MEMOP_LOGICAL_WRITE`` 3837 * ``KVM_S390_MEMOP_LOGICAL_WRITE``
3854 * ``KVM_S390_MEMOP_ABSOLUTE_READ`` 3838 * ``KVM_S390_MEMOP_ABSOLUTE_READ``
3855 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE`` 3839 * ``KVM_S390_MEMOP_ABSOLUTE_WRITE``
3856 * ``KVM_S390_MEMOP_SIDA_READ`` 3840 * ``KVM_S390_MEMOP_SIDA_READ``
3857 * ``KVM_S390_MEMOP_SIDA_WRITE`` 3841 * ``KVM_S390_MEMOP_SIDA_WRITE``
3858 * ``KVM_S390_MEMOP_ABSOLUTE_CMPXCHG`` 3842 * ``KVM_S390_MEMOP_ABSOLUTE_CMPXCHG``
3859 3843
3860 Logical read/write: 3844 Logical read/write:
3861 ^^^^^^^^^^^^^^^^^^^ 3845 ^^^^^^^^^^^^^^^^^^^
3862 3846
3863 Access logical memory, i.e. translate the giv 3847 Access logical memory, i.e. translate the given guest address to an absolute
3864 address given the state of the VCPU and use t 3848 address given the state of the VCPU and use the absolute address as target of
3865 the access. "ar" designates the access regist 3849 the access. "ar" designates the access register number to be used; the valid
3866 range is 0..15. 3850 range is 0..15.
3867 Logical accesses are permitted for the VCPU i 3851 Logical accesses are permitted for the VCPU ioctl only.
3868 Logical accesses are permitted for non-protec 3852 Logical accesses are permitted for non-protected guests only.
3869 3853
3870 Supported flags: 3854 Supported flags:
3871 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3855 * ``KVM_S390_MEMOP_F_CHECK_ONLY``
3872 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION`` 3856 * ``KVM_S390_MEMOP_F_INJECT_EXCEPTION``
3873 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3857 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION``
3874 3858
3875 The KVM_S390_MEMOP_F_CHECK_ONLY flag can be s 3859 The KVM_S390_MEMOP_F_CHECK_ONLY flag can be set to check whether the
3876 corresponding memory access would cause an ac 3860 corresponding memory access would cause an access exception; however,
3877 no actual access to the data in memory at the 3861 no actual access to the data in memory at the destination is performed.
3878 In this case, "buf" is unused and can be NULL 3862 In this case, "buf" is unused and can be NULL.
3879 3863
3880 In case an access exception occurred during t 3864 In case an access exception occurred during the access (or would occur
3881 in case of KVM_S390_MEMOP_F_CHECK_ONLY), the 3865 in case of KVM_S390_MEMOP_F_CHECK_ONLY), the ioctl returns a positive
3882 error number indicating the type of exception 3866 error number indicating the type of exception. This exception is also
3883 raised directly at the corresponding VCPU if 3867 raised directly at the corresponding VCPU if the flag
3884 KVM_S390_MEMOP_F_INJECT_EXCEPTION is set. 3868 KVM_S390_MEMOP_F_INJECT_EXCEPTION is set.
3885 On protection exceptions, unless specified ot 3869 On protection exceptions, unless specified otherwise, the injected
3886 translation-exception identifier (TEID) indic 3870 translation-exception identifier (TEID) indicates suppression.
3887 3871
3888 If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag 3872 If the KVM_S390_MEMOP_F_SKEY_PROTECTION flag is set, storage key
3889 protection is also in effect and may cause ex 3873 protection is also in effect and may cause exceptions if accesses are
3890 prohibited given the access key designated by 3874 prohibited given the access key designated by "key"; the valid range is 0..15.
3891 KVM_S390_MEMOP_F_SKEY_PROTECTION is available 3875 KVM_S390_MEMOP_F_SKEY_PROTECTION is available if KVM_CAP_S390_MEM_OP_EXTENSION
3892 is > 0. 3876 is > 0.
3893 Since the accessed memory may span multiple p 3877 Since the accessed memory may span multiple pages and those pages might have
3894 different storage keys, it is possible that a 3878 different storage keys, it is possible that a protection exception occurs
3895 after memory has been modified. In this case, 3879 after memory has been modified. In this case, if the exception is injected,
3896 the TEID does not indicate suppression. 3880 the TEID does not indicate suppression.
3897 3881
3898 Absolute read/write: 3882 Absolute read/write:
3899 ^^^^^^^^^^^^^^^^^^^^ 3883 ^^^^^^^^^^^^^^^^^^^^
3900 3884
3901 Access absolute memory. This operation is int 3885 Access absolute memory. This operation is intended to be used with the
3902 KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to all 3886 KVM_S390_MEMOP_F_SKEY_PROTECTION flag, to allow accessing memory and performing
3903 the checks required for storage key protectio 3887 the checks required for storage key protection as one operation (as opposed to
3904 user space getting the storage keys, performi 3888 user space getting the storage keys, performing the checks, and accessing
3905 memory thereafter, which could lead to a dela 3889 memory thereafter, which could lead to a delay between check and access).
3906 Absolute accesses are permitted for the VM io 3890 Absolute accesses are permitted for the VM ioctl if KVM_CAP_S390_MEM_OP_EXTENSION
3907 has the KVM_S390_MEMOP_EXTENSION_CAP_BASE bit 3891 has the KVM_S390_MEMOP_EXTENSION_CAP_BASE bit set.
3908 Currently absolute accesses are not permitted 3892 Currently absolute accesses are not permitted for VCPU ioctls.
3909 Absolute accesses are permitted for non-prote 3893 Absolute accesses are permitted for non-protected guests only.
3910 3894
3911 Supported flags: 3895 Supported flags:
3912 * ``KVM_S390_MEMOP_F_CHECK_ONLY`` 3896 * ``KVM_S390_MEMOP_F_CHECK_ONLY``
3913 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3897 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION``
3914 3898
3915 The semantics of the flags common with logica 3899 The semantics of the flags common with logical accesses are as for logical
3916 accesses. 3900 accesses.
3917 3901
3918 Absolute cmpxchg: 3902 Absolute cmpxchg:
3919 ^^^^^^^^^^^^^^^^^ 3903 ^^^^^^^^^^^^^^^^^
3920 3904
3921 Perform cmpxchg on absolute guest memory. Int 3905 Perform cmpxchg on absolute guest memory. Intended for use with the
3922 KVM_S390_MEMOP_F_SKEY_PROTECTION flag. 3906 KVM_S390_MEMOP_F_SKEY_PROTECTION flag.
3923 Instead of doing an unconditional write, the 3907 Instead of doing an unconditional write, the access occurs only if the target
3924 location contains the value pointed to by "ol 3908 location contains the value pointed to by "old_addr".
3925 This is performed as an atomic cmpxchg with t 3909 This is performed as an atomic cmpxchg with the length specified by the "size"
3926 parameter. "size" must be a power of two up t 3910 parameter. "size" must be a power of two up to and including 16.
3927 If the exchange did not take place because th 3911 If the exchange did not take place because the target value doesn't match the
3928 old value, the value "old_addr" points to is 3912 old value, the value "old_addr" points to is replaced by the target value.
3929 User space can tell if an exchange took place 3913 User space can tell if an exchange took place by checking if this replacement
3930 occurred. The cmpxchg op is permitted for the 3914 occurred. The cmpxchg op is permitted for the VM ioctl if
3931 KVM_CAP_S390_MEM_OP_EXTENSION has flag KVM_S3 3915 KVM_CAP_S390_MEM_OP_EXTENSION has flag KVM_S390_MEMOP_EXTENSION_CAP_CMPXCHG set.
3932 3916
3933 Supported flags: 3917 Supported flags:
3934 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION`` 3918 * ``KVM_S390_MEMOP_F_SKEY_PROTECTION``
3935 3919
3936 SIDA read/write: 3920 SIDA read/write:
3937 ^^^^^^^^^^^^^^^^ 3921 ^^^^^^^^^^^^^^^^
3938 3922
3939 Access the secure instruction data area which 3923 Access the secure instruction data area which contains memory operands necessary
3940 for instruction emulation for protected guest 3924 for instruction emulation for protected guests.
3941 SIDA accesses are available if the KVM_CAP_S3 3925 SIDA accesses are available if the KVM_CAP_S390_PROTECTED capability is available.
3942 SIDA accesses are permitted for the VCPU ioct 3926 SIDA accesses are permitted for the VCPU ioctl only.
3943 SIDA accesses are permitted for protected gue 3927 SIDA accesses are permitted for protected guests only.
3944 3928
3945 No flags are supported. 3929 No flags are supported.
3946 3930
3947 4.90 KVM_S390_GET_SKEYS 3931 4.90 KVM_S390_GET_SKEYS
3948 ----------------------- 3932 -----------------------
3949 3933
3950 :Capability: KVM_CAP_S390_SKEYS 3934 :Capability: KVM_CAP_S390_SKEYS
3951 :Architectures: s390 3935 :Architectures: s390
3952 :Type: vm ioctl 3936 :Type: vm ioctl
3953 :Parameters: struct kvm_s390_skeys 3937 :Parameters: struct kvm_s390_skeys
3954 :Returns: 0 on success, KVM_S390_GET_SKEYS_NO 3938 :Returns: 0 on success, KVM_S390_GET_SKEYS_NONE if guest is not using storage
3955 keys, negative value on error 3939 keys, negative value on error
3956 3940
3957 This ioctl is used to get guest storage key v 3941 This ioctl is used to get guest storage key values on the s390
3958 architecture. The ioctl takes parameters via 3942 architecture. The ioctl takes parameters via the kvm_s390_skeys struct::
3959 3943
3960 struct kvm_s390_skeys { 3944 struct kvm_s390_skeys {
3961 __u64 start_gfn; 3945 __u64 start_gfn;
3962 __u64 count; 3946 __u64 count;
3963 __u64 skeydata_addr; 3947 __u64 skeydata_addr;
3964 __u32 flags; 3948 __u32 flags;
3965 __u32 reserved[9]; 3949 __u32 reserved[9];
3966 }; 3950 };
3967 3951
3968 The start_gfn field is the number of the firs 3952 The start_gfn field is the number of the first guest frame whose storage keys
3969 you want to get. 3953 you want to get.
3970 3954
3971 The count field is the number of consecutive 3955 The count field is the number of consecutive frames (starting from start_gfn)
3972 whose storage keys to get. The count field mu 3956 whose storage keys to get. The count field must be at least 1 and the maximum
3973 allowed value is defined as KVM_S390_SKEYS_MA 3957 allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range
3974 will cause the ioctl to return -EINVAL. 3958 will cause the ioctl to return -EINVAL.
3975 3959
3976 The skeydata_addr field is the address to a b 3960 The skeydata_addr field is the address to a buffer large enough to hold count
3977 bytes. This buffer will be filled with storag 3961 bytes. This buffer will be filled with storage key data by the ioctl.
3978 3962
3979 4.91 KVM_S390_SET_SKEYS 3963 4.91 KVM_S390_SET_SKEYS
3980 ----------------------- 3964 -----------------------
3981 3965
3982 :Capability: KVM_CAP_S390_SKEYS 3966 :Capability: KVM_CAP_S390_SKEYS
3983 :Architectures: s390 3967 :Architectures: s390
3984 :Type: vm ioctl 3968 :Type: vm ioctl
3985 :Parameters: struct kvm_s390_skeys 3969 :Parameters: struct kvm_s390_skeys
3986 :Returns: 0 on success, negative value on err 3970 :Returns: 0 on success, negative value on error
3987 3971
3988 This ioctl is used to set guest storage key v 3972 This ioctl is used to set guest storage key values on the s390
3989 architecture. The ioctl takes parameters via 3973 architecture. The ioctl takes parameters via the kvm_s390_skeys struct.
3990 See section on KVM_S390_GET_SKEYS for struct 3974 See section on KVM_S390_GET_SKEYS for struct definition.
3991 3975
3992 The start_gfn field is the number of the firs 3976 The start_gfn field is the number of the first guest frame whose storage keys
3993 you want to set. 3977 you want to set.
3994 3978
3995 The count field is the number of consecutive 3979 The count field is the number of consecutive frames (starting from start_gfn)
3996 whose storage keys to get. The count field mu 3980 whose storage keys to get. The count field must be at least 1 and the maximum
3997 allowed value is defined as KVM_S390_SKEYS_MA 3981 allowed value is defined as KVM_S390_SKEYS_MAX. Values outside this range
3998 will cause the ioctl to return -EINVAL. 3982 will cause the ioctl to return -EINVAL.
3999 3983
4000 The skeydata_addr field is the address to a b 3984 The skeydata_addr field is the address to a buffer containing count bytes of
4001 storage keys. Each byte in the buffer will be 3985 storage keys. Each byte in the buffer will be set as the storage key for a
4002 single frame starting at start_gfn for count 3986 single frame starting at start_gfn for count frames.
4003 3987
4004 Note: If any architecturally invalid key valu 3988 Note: If any architecturally invalid key value is found in the given data then
4005 the ioctl will return -EINVAL. 3989 the ioctl will return -EINVAL.
4006 3990
4007 4.92 KVM_S390_IRQ 3991 4.92 KVM_S390_IRQ
4008 ----------------- 3992 -----------------
4009 3993
4010 :Capability: KVM_CAP_S390_INJECT_IRQ 3994 :Capability: KVM_CAP_S390_INJECT_IRQ
4011 :Architectures: s390 3995 :Architectures: s390
4012 :Type: vcpu ioctl 3996 :Type: vcpu ioctl
4013 :Parameters: struct kvm_s390_irq (in) 3997 :Parameters: struct kvm_s390_irq (in)
4014 :Returns: 0 on success, -1 on error 3998 :Returns: 0 on success, -1 on error
4015 3999
4016 Errors: 4000 Errors:
4017 4001
4018 4002
4019 ====== =================================== 4003 ====== =================================================================
4020 EINVAL interrupt type is invalid 4004 EINVAL interrupt type is invalid
4021 type is KVM_S390_SIGP_STOP and flag 4005 type is KVM_S390_SIGP_STOP and flag parameter is invalid value,
4022 type is KVM_S390_INT_EXTERNAL_CALL 4006 type is KVM_S390_INT_EXTERNAL_CALL and code is bigger
4023 than the maximum of VCPUs 4007 than the maximum of VCPUs
4024 EBUSY type is KVM_S390_SIGP_SET_PREFIX an 4008 EBUSY type is KVM_S390_SIGP_SET_PREFIX and vcpu is not stopped,
4025 type is KVM_S390_SIGP_STOP and a st 4009 type is KVM_S390_SIGP_STOP and a stop irq is already pending,
4026 type is KVM_S390_INT_EXTERNAL_CALL 4010 type is KVM_S390_INT_EXTERNAL_CALL and an external call interrupt
4027 is already pending 4011 is already pending
4028 ====== =================================== 4012 ====== =================================================================
4029 4013
4030 Allows to inject an interrupt to the guest. 4014 Allows to inject an interrupt to the guest.
4031 4015
4032 Using struct kvm_s390_irq as a parameter allo 4016 Using struct kvm_s390_irq as a parameter allows
4033 to inject additional payload which is not 4017 to inject additional payload which is not
4034 possible via KVM_S390_INTERRUPT. 4018 possible via KVM_S390_INTERRUPT.
4035 4019
4036 Interrupt parameters are passed via kvm_s390_ 4020 Interrupt parameters are passed via kvm_s390_irq::
4037 4021
4038 struct kvm_s390_irq { 4022 struct kvm_s390_irq {
4039 __u64 type; 4023 __u64 type;
4040 union { 4024 union {
4041 struct kvm_s390_io_info io; 4025 struct kvm_s390_io_info io;
4042 struct kvm_s390_ext_info ext; 4026 struct kvm_s390_ext_info ext;
4043 struct kvm_s390_pgm_info pgm; 4027 struct kvm_s390_pgm_info pgm;
4044 struct kvm_s390_emerg_info em 4028 struct kvm_s390_emerg_info emerg;
4045 struct kvm_s390_extcall_info 4029 struct kvm_s390_extcall_info extcall;
4046 struct kvm_s390_prefix_info p 4030 struct kvm_s390_prefix_info prefix;
4047 struct kvm_s390_stop_info sto 4031 struct kvm_s390_stop_info stop;
4048 struct kvm_s390_mchk_info mch 4032 struct kvm_s390_mchk_info mchk;
4049 char reserved[64]; 4033 char reserved[64];
4050 } u; 4034 } u;
4051 }; 4035 };
4052 4036
4053 type can be one of the following: 4037 type can be one of the following:
4054 4038
4055 - KVM_S390_SIGP_STOP - sigp stop; parameter i 4039 - KVM_S390_SIGP_STOP - sigp stop; parameter in .stop
4056 - KVM_S390_PROGRAM_INT - program check; param 4040 - KVM_S390_PROGRAM_INT - program check; parameters in .pgm
4057 - KVM_S390_SIGP_SET_PREFIX - sigp set prefix; 4041 - KVM_S390_SIGP_SET_PREFIX - sigp set prefix; parameters in .prefix
4058 - KVM_S390_RESTART - restart; no parameters 4042 - KVM_S390_RESTART - restart; no parameters
4059 - KVM_S390_INT_CLOCK_COMP - clock comparator 4043 - KVM_S390_INT_CLOCK_COMP - clock comparator interrupt; no parameters
4060 - KVM_S390_INT_CPU_TIMER - CPU timer interrup 4044 - KVM_S390_INT_CPU_TIMER - CPU timer interrupt; no parameters
4061 - KVM_S390_INT_EMERGENCY - sigp emergency; pa 4045 - KVM_S390_INT_EMERGENCY - sigp emergency; parameters in .emerg
4062 - KVM_S390_INT_EXTERNAL_CALL - sigp external 4046 - KVM_S390_INT_EXTERNAL_CALL - sigp external call; parameters in .extcall
4063 - KVM_S390_MCHK - machine check interrupt; pa 4047 - KVM_S390_MCHK - machine check interrupt; parameters in .mchk
4064 4048
4065 This is an asynchronous vcpu ioctl and can be 4049 This is an asynchronous vcpu ioctl and can be invoked from any thread.
4066 4050
4067 4.94 KVM_S390_GET_IRQ_STATE 4051 4.94 KVM_S390_GET_IRQ_STATE
4068 --------------------------- 4052 ---------------------------
4069 4053
4070 :Capability: KVM_CAP_S390_IRQ_STATE 4054 :Capability: KVM_CAP_S390_IRQ_STATE
4071 :Architectures: s390 4055 :Architectures: s390
4072 :Type: vcpu ioctl 4056 :Type: vcpu ioctl
4073 :Parameters: struct kvm_s390_irq_state (out) 4057 :Parameters: struct kvm_s390_irq_state (out)
4074 :Returns: >= number of bytes copied into buff 4058 :Returns: >= number of bytes copied into buffer,
4075 -EINVAL if buffer size is 0, 4059 -EINVAL if buffer size is 0,
4076 -ENOBUFS if buffer size is too smal 4060 -ENOBUFS if buffer size is too small to fit all pending interrupts,
4077 -EFAULT if the buffer address was i 4061 -EFAULT if the buffer address was invalid
4078 4062
4079 This ioctl allows userspace to retrieve the c 4063 This ioctl allows userspace to retrieve the complete state of all currently
4080 pending interrupts in a single buffer. Use ca 4064 pending interrupts in a single buffer. Use cases include migration
4081 and introspection. The parameter structure co 4065 and introspection. The parameter structure contains the address of a
4082 userspace buffer and its length:: 4066 userspace buffer and its length::
4083 4067
4084 struct kvm_s390_irq_state { 4068 struct kvm_s390_irq_state {
4085 __u64 buf; 4069 __u64 buf;
4086 __u32 flags; /* will stay unus 4070 __u32 flags; /* will stay unused for compatibility reasons */
4087 __u32 len; 4071 __u32 len;
4088 __u32 reserved[4]; /* will stay unus 4072 __u32 reserved[4]; /* will stay unused for compatibility reasons */
4089 }; 4073 };
4090 4074
4091 Userspace passes in the above struct and for 4075 Userspace passes in the above struct and for each pending interrupt a
4092 struct kvm_s390_irq is copied to the provided 4076 struct kvm_s390_irq is copied to the provided buffer.
4093 4077
4094 The structure contains a flags and a reserved 4078 The structure contains a flags and a reserved field for future extensions. As
4095 the kernel never checked for flags == 0 and Q 4079 the kernel never checked for flags == 0 and QEMU never pre-zeroed flags and
4096 reserved, these fields can not be used in the 4080 reserved, these fields can not be used in the future without breaking
4097 compatibility. 4081 compatibility.
4098 4082
4099 If -ENOBUFS is returned the buffer provided w 4083 If -ENOBUFS is returned the buffer provided was too small and userspace
4100 may retry with a bigger buffer. 4084 may retry with a bigger buffer.
4101 4085
4102 4.95 KVM_S390_SET_IRQ_STATE 4086 4.95 KVM_S390_SET_IRQ_STATE
4103 --------------------------- 4087 ---------------------------
4104 4088
4105 :Capability: KVM_CAP_S390_IRQ_STATE 4089 :Capability: KVM_CAP_S390_IRQ_STATE
4106 :Architectures: s390 4090 :Architectures: s390
4107 :Type: vcpu ioctl 4091 :Type: vcpu ioctl
4108 :Parameters: struct kvm_s390_irq_state (in) 4092 :Parameters: struct kvm_s390_irq_state (in)
4109 :Returns: 0 on success, 4093 :Returns: 0 on success,
4110 -EFAULT if the buffer address was i 4094 -EFAULT if the buffer address was invalid,
4111 -EINVAL for an invalid buffer lengt 4095 -EINVAL for an invalid buffer length (see below),
4112 -EBUSY if there were already interr 4096 -EBUSY if there were already interrupts pending,
4113 errors occurring when actually inje 4097 errors occurring when actually injecting the
4114 interrupt. See KVM_S390_IRQ. 4098 interrupt. See KVM_S390_IRQ.
4115 4099
4116 This ioctl allows userspace to set the comple 4100 This ioctl allows userspace to set the complete state of all cpu-local
4117 interrupts currently pending for the vcpu. It 4101 interrupts currently pending for the vcpu. It is intended for restoring
4118 interrupt state after a migration. The input 4102 interrupt state after a migration. The input parameter is a userspace buffer
4119 containing a struct kvm_s390_irq_state:: 4103 containing a struct kvm_s390_irq_state::
4120 4104
4121 struct kvm_s390_irq_state { 4105 struct kvm_s390_irq_state {
4122 __u64 buf; 4106 __u64 buf;
4123 __u32 flags; /* will stay unus 4107 __u32 flags; /* will stay unused for compatibility reasons */
4124 __u32 len; 4108 __u32 len;
4125 __u32 reserved[4]; /* will stay unus 4109 __u32 reserved[4]; /* will stay unused for compatibility reasons */
4126 }; 4110 };
4127 4111
4128 The restrictions for flags and reserved apply 4112 The restrictions for flags and reserved apply as well.
4129 (see KVM_S390_GET_IRQ_STATE) 4113 (see KVM_S390_GET_IRQ_STATE)
4130 4114
4131 The userspace memory referenced by buf contai 4115 The userspace memory referenced by buf contains a struct kvm_s390_irq
4132 for each interrupt to be injected into the gu 4116 for each interrupt to be injected into the guest.
4133 If one of the interrupts could not be injecte 4117 If one of the interrupts could not be injected for some reason the
4134 ioctl aborts. 4118 ioctl aborts.
4135 4119
4136 len must be a multiple of sizeof(struct kvm_s 4120 len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
4137 and it must not exceed (max_vcpus + 32) * siz 4121 and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
4138 which is the maximum number of possibly pendi 4122 which is the maximum number of possibly pending cpu-local interrupts.
4139 4123
4140 4.96 KVM_SMI 4124 4.96 KVM_SMI
4141 ------------ 4125 ------------
4142 4126
4143 :Capability: KVM_CAP_X86_SMM 4127 :Capability: KVM_CAP_X86_SMM
4144 :Architectures: x86 4128 :Architectures: x86
4145 :Type: vcpu ioctl 4129 :Type: vcpu ioctl
4146 :Parameters: none 4130 :Parameters: none
4147 :Returns: 0 on success, -1 on error 4131 :Returns: 0 on success, -1 on error
4148 4132
4149 Queues an SMI on the thread's vcpu. 4133 Queues an SMI on the thread's vcpu.
4150 4134
4151 4.97 KVM_X86_SET_MSR_FILTER 4135 4.97 KVM_X86_SET_MSR_FILTER
4152 ---------------------------- 4136 ----------------------------
4153 4137
4154 :Capability: KVM_CAP_X86_MSR_FILTER 4138 :Capability: KVM_CAP_X86_MSR_FILTER
4155 :Architectures: x86 4139 :Architectures: x86
4156 :Type: vm ioctl 4140 :Type: vm ioctl
4157 :Parameters: struct kvm_msr_filter 4141 :Parameters: struct kvm_msr_filter
4158 :Returns: 0 on success, < 0 on error 4142 :Returns: 0 on success, < 0 on error
4159 4143
4160 :: 4144 ::
4161 4145
4162 struct kvm_msr_filter_range { 4146 struct kvm_msr_filter_range {
4163 #define KVM_MSR_FILTER_READ (1 << 0) 4147 #define KVM_MSR_FILTER_READ (1 << 0)
4164 #define KVM_MSR_FILTER_WRITE (1 << 1) 4148 #define KVM_MSR_FILTER_WRITE (1 << 1)
4165 __u32 flags; 4149 __u32 flags;
4166 __u32 nmsrs; /* number of msrs in bit 4150 __u32 nmsrs; /* number of msrs in bitmap */
4167 __u32 base; /* MSR index the bitmap 4151 __u32 base; /* MSR index the bitmap starts at */
4168 __u8 *bitmap; /* a 1 bit allows the o 4152 __u8 *bitmap; /* a 1 bit allows the operations in flags, 0 denies */
4169 }; 4153 };
4170 4154
4171 #define KVM_MSR_FILTER_MAX_RANGES 16 4155 #define KVM_MSR_FILTER_MAX_RANGES 16
4172 struct kvm_msr_filter { 4156 struct kvm_msr_filter {
4173 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 4157 #define KVM_MSR_FILTER_DEFAULT_ALLOW (0 << 0)
4174 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 4158 #define KVM_MSR_FILTER_DEFAULT_DENY (1 << 0)
4175 __u32 flags; 4159 __u32 flags;
4176 struct kvm_msr_filter_range ranges[KV 4160 struct kvm_msr_filter_range ranges[KVM_MSR_FILTER_MAX_RANGES];
4177 }; 4161 };
4178 4162
4179 flags values for ``struct kvm_msr_filter_rang 4163 flags values for ``struct kvm_msr_filter_range``:
4180 4164
4181 ``KVM_MSR_FILTER_READ`` 4165 ``KVM_MSR_FILTER_READ``
4182 4166
4183 Filter read accesses to MSRs using the give 4167 Filter read accesses to MSRs using the given bitmap. A 0 in the bitmap
4184 indicates that read accesses should be deni 4168 indicates that read accesses should be denied, while a 1 indicates that
4185 a read for a particular MSR should be allow 4169 a read for a particular MSR should be allowed regardless of the default
4186 filter action. 4170 filter action.
4187 4171
4188 ``KVM_MSR_FILTER_WRITE`` 4172 ``KVM_MSR_FILTER_WRITE``
4189 4173
4190 Filter write accesses to MSRs using the giv 4174 Filter write accesses to MSRs using the given bitmap. A 0 in the bitmap
4191 indicates that write accesses should be den 4175 indicates that write accesses should be denied, while a 1 indicates that
4192 a write for a particular MSR should be allo 4176 a write for a particular MSR should be allowed regardless of the default
4193 filter action. 4177 filter action.
4194 4178
4195 flags values for ``struct kvm_msr_filter``: 4179 flags values for ``struct kvm_msr_filter``:
4196 4180
4197 ``KVM_MSR_FILTER_DEFAULT_ALLOW`` 4181 ``KVM_MSR_FILTER_DEFAULT_ALLOW``
4198 4182
4199 If no filter range matches an MSR index tha 4183 If no filter range matches an MSR index that is getting accessed, KVM will
4200 allow accesses to all MSRs by default. 4184 allow accesses to all MSRs by default.
4201 4185
4202 ``KVM_MSR_FILTER_DEFAULT_DENY`` 4186 ``KVM_MSR_FILTER_DEFAULT_DENY``
4203 4187
4204 If no filter range matches an MSR index tha 4188 If no filter range matches an MSR index that is getting accessed, KVM will
4205 deny accesses to all MSRs by default. 4189 deny accesses to all MSRs by default.
4206 4190
4207 This ioctl allows userspace to define up to 1 4191 This ioctl allows userspace to define up to 16 bitmaps of MSR ranges to deny
4208 guest MSR accesses that would normally be all 4192 guest MSR accesses that would normally be allowed by KVM. If an MSR is not
4209 covered by a specific range, the "default" fi 4193 covered by a specific range, the "default" filtering behavior applies. Each
4210 bitmap range covers MSRs from [base .. base+n 4194 bitmap range covers MSRs from [base .. base+nmsrs).
4211 4195
4212 If an MSR access is denied by userspace, the 4196 If an MSR access is denied by userspace, the resulting KVM behavior depends on
4213 whether or not KVM_CAP_X86_USER_SPACE_MSR's K 4197 whether or not KVM_CAP_X86_USER_SPACE_MSR's KVM_MSR_EXIT_REASON_FILTER is
4214 enabled. If KVM_MSR_EXIT_REASON_FILTER is en 4198 enabled. If KVM_MSR_EXIT_REASON_FILTER is enabled, KVM will exit to userspace
4215 on denied accesses, i.e. userspace effectivel 4199 on denied accesses, i.e. userspace effectively intercepts the MSR access. If
4216 KVM_MSR_EXIT_REASON_FILTER is not enabled, KV 4200 KVM_MSR_EXIT_REASON_FILTER is not enabled, KVM will inject a #GP into the guest
4217 on denied accesses. Note, if an MSR access i !! 4201 on denied accesses.
4218 load/stores during VMX transitions, KVM ignor <<
4219 See the below warning for full details. <<
4220 4202
4221 If an MSR access is allowed by userspace, KVM 4203 If an MSR access is allowed by userspace, KVM will emulate and/or virtualize
4222 the access in accordance with the vCPU model. 4204 the access in accordance with the vCPU model. Note, KVM may still ultimately
4223 inject a #GP if an access is allowed by users 4205 inject a #GP if an access is allowed by userspace, e.g. if KVM doesn't support
4224 the MSR, or to follow architectural behavior 4206 the MSR, or to follow architectural behavior for the MSR.
4225 4207
4226 By default, KVM operates in KVM_MSR_FILTER_DE 4208 By default, KVM operates in KVM_MSR_FILTER_DEFAULT_ALLOW mode with no MSR range
4227 filters. 4209 filters.
4228 4210
4229 Calling this ioctl with an empty set of range 4211 Calling this ioctl with an empty set of ranges (all nmsrs == 0) disables MSR
4230 filtering. In that mode, ``KVM_MSR_FILTER_DEF 4212 filtering. In that mode, ``KVM_MSR_FILTER_DEFAULT_DENY`` is invalid and causes
4231 an error. 4213 an error.
4232 4214
4233 .. warning:: 4215 .. warning::
4234 MSR accesses that are side effects of inst !! 4216 MSR accesses as part of nested VM-Enter/VM-Exit are not filtered.
4235 native) are not filtered as hardware does !! 4217 This includes both writes to individual VMCS fields and reads/writes
4236 RDMSR and WRMSR, and KVM mimics that behav !! 4218 through the MSR lists pointed to by the VMCS.
4237 to avoid pointless divergence from hardwar <<
4238 SYSENTER reads the SYSENTER MSRs, etc. <<
4239 <<
4240 MSRs that are loaded/stored via dedicated <<
4241 part of VM-Enter/VM-Exit emulation. <<
4242 <<
4243 MSRs that are loaded/store via VMX's load/ <<
4244 of VM-Enter/VM-Exit emulation. If an MSR <<
4245 synthesizes a consistency check VM-Exit(EX <<
4246 MSR access is denied on VM-Exit, KVM synth <<
4247 extends Intel's architectural list of MSRs <<
4248 the VM-Enter/VM-Exit MSR list. It is plat <<
4249 to communicate any such restrictions to th <<
4250 4219
4251 x2APIC MSR accesses cannot be filtered (KV 4220 x2APIC MSR accesses cannot be filtered (KVM silently ignores filters that
4252 cover any x2APIC MSRs). 4221 cover any x2APIC MSRs).
4253 4222
4254 Note, invoking this ioctl while a vCPU is run 4223 Note, invoking this ioctl while a vCPU is running is inherently racy. However,
4255 KVM does guarantee that vCPUs will see either 4224 KVM does guarantee that vCPUs will see either the previous filter or the new
4256 filter, e.g. MSRs with identical settings in 4225 filter, e.g. MSRs with identical settings in both the old and new filter will
4257 have deterministic behavior. 4226 have deterministic behavior.
4258 4227
4259 Similarly, if userspace wishes to intercept o 4228 Similarly, if userspace wishes to intercept on denied accesses,
4260 KVM_MSR_EXIT_REASON_FILTER must be enabled be 4229 KVM_MSR_EXIT_REASON_FILTER must be enabled before activating any filters, and
4261 left enabled until after all filters are deac 4230 left enabled until after all filters are deactivated. Failure to do so may
4262 result in KVM injecting a #GP instead of exit 4231 result in KVM injecting a #GP instead of exiting to userspace.
4263 4232
4264 4.98 KVM_CREATE_SPAPR_TCE_64 4233 4.98 KVM_CREATE_SPAPR_TCE_64
4265 ---------------------------- 4234 ----------------------------
4266 4235
4267 :Capability: KVM_CAP_SPAPR_TCE_64 4236 :Capability: KVM_CAP_SPAPR_TCE_64
4268 :Architectures: powerpc 4237 :Architectures: powerpc
4269 :Type: vm ioctl 4238 :Type: vm ioctl
4270 :Parameters: struct kvm_create_spapr_tce_64 ( 4239 :Parameters: struct kvm_create_spapr_tce_64 (in)
4271 :Returns: file descriptor for manipulating th 4240 :Returns: file descriptor for manipulating the created TCE table
4272 4241
4273 This is an extension for KVM_CAP_SPAPR_TCE wh 4242 This is an extension for KVM_CAP_SPAPR_TCE which only supports 32bit
4274 windows, described in 4.62 KVM_CREATE_SPAPR_T 4243 windows, described in 4.62 KVM_CREATE_SPAPR_TCE
4275 4244
4276 This capability uses extended struct in ioctl 4245 This capability uses extended struct in ioctl interface::
4277 4246
4278 /* for KVM_CAP_SPAPR_TCE_64 */ 4247 /* for KVM_CAP_SPAPR_TCE_64 */
4279 struct kvm_create_spapr_tce_64 { 4248 struct kvm_create_spapr_tce_64 {
4280 __u64 liobn; 4249 __u64 liobn;
4281 __u32 page_shift; 4250 __u32 page_shift;
4282 __u32 flags; 4251 __u32 flags;
4283 __u64 offset; /* in pages */ 4252 __u64 offset; /* in pages */
4284 __u64 size; /* in pages */ 4253 __u64 size; /* in pages */
4285 }; 4254 };
4286 4255
4287 The aim of extension is to support an additio 4256 The aim of extension is to support an additional bigger DMA window with
4288 a variable page size. 4257 a variable page size.
4289 KVM_CREATE_SPAPR_TCE_64 receives a 64bit wind 4258 KVM_CREATE_SPAPR_TCE_64 receives a 64bit window size, an IOMMU page shift and
4290 a bus offset of the corresponding DMA window, 4259 a bus offset of the corresponding DMA window, @size and @offset are numbers
4291 of IOMMU pages. 4260 of IOMMU pages.
4292 4261
4293 @flags are not used at the moment. 4262 @flags are not used at the moment.
4294 4263
4295 The rest of functionality is identical to KVM 4264 The rest of functionality is identical to KVM_CREATE_SPAPR_TCE.
4296 4265
4297 4.99 KVM_REINJECT_CONTROL 4266 4.99 KVM_REINJECT_CONTROL
4298 ------------------------- 4267 -------------------------
4299 4268
4300 :Capability: KVM_CAP_REINJECT_CONTROL 4269 :Capability: KVM_CAP_REINJECT_CONTROL
4301 :Architectures: x86 4270 :Architectures: x86
4302 :Type: vm ioctl 4271 :Type: vm ioctl
4303 :Parameters: struct kvm_reinject_control (in) 4272 :Parameters: struct kvm_reinject_control (in)
4304 :Returns: 0 on success, 4273 :Returns: 0 on success,
4305 -EFAULT if struct kvm_reinject_contr 4274 -EFAULT if struct kvm_reinject_control cannot be read,
4306 -ENXIO if KVM_CREATE_PIT or KVM_CREA 4275 -ENXIO if KVM_CREATE_PIT or KVM_CREATE_PIT2 didn't succeed earlier.
4307 4276
4308 i8254 (PIT) has two modes, reinject and !rein 4277 i8254 (PIT) has two modes, reinject and !reinject. The default is reinject,
4309 where KVM queues elapsed i8254 ticks and moni 4278 where KVM queues elapsed i8254 ticks and monitors completion of interrupt from
4310 vector(s) that i8254 injects. Reinject mode 4279 vector(s) that i8254 injects. Reinject mode dequeues a tick and injects its
4311 interrupt whenever there isn't a pending inte 4280 interrupt whenever there isn't a pending interrupt from i8254.
4312 !reinject mode injects an interrupt as soon a 4281 !reinject mode injects an interrupt as soon as a tick arrives.
4313 4282
4314 :: 4283 ::
4315 4284
4316 struct kvm_reinject_control { 4285 struct kvm_reinject_control {
4317 __u8 pit_reinject; 4286 __u8 pit_reinject;
4318 __u8 reserved[31]; 4287 __u8 reserved[31];
4319 }; 4288 };
4320 4289
4321 pit_reinject = 0 (!reinject mode) is recommen 4290 pit_reinject = 0 (!reinject mode) is recommended, unless running an old
4322 operating system that uses the PIT for timing 4291 operating system that uses the PIT for timing (e.g. Linux 2.4.x).
4323 4292
4324 4.100 KVM_PPC_CONFIGURE_V3_MMU 4293 4.100 KVM_PPC_CONFIGURE_V3_MMU
4325 ------------------------------ 4294 ------------------------------
4326 4295
4327 :Capability: KVM_CAP_PPC_MMU_RADIX or KVM_CAP !! 4296 :Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3
4328 :Architectures: ppc 4297 :Architectures: ppc
4329 :Type: vm ioctl 4298 :Type: vm ioctl
4330 :Parameters: struct kvm_ppc_mmuv3_cfg (in) 4299 :Parameters: struct kvm_ppc_mmuv3_cfg (in)
4331 :Returns: 0 on success, 4300 :Returns: 0 on success,
4332 -EFAULT if struct kvm_ppc_mmuv3_cfg 4301 -EFAULT if struct kvm_ppc_mmuv3_cfg cannot be read,
4333 -EINVAL if the configuration is inva 4302 -EINVAL if the configuration is invalid
4334 4303
4335 This ioctl controls whether the guest will us 4304 This ioctl controls whether the guest will use radix or HPT (hashed
4336 page table) translation, and sets the pointer 4305 page table) translation, and sets the pointer to the process table for
4337 the guest. 4306 the guest.
4338 4307
4339 :: 4308 ::
4340 4309
4341 struct kvm_ppc_mmuv3_cfg { 4310 struct kvm_ppc_mmuv3_cfg {
4342 __u64 flags; 4311 __u64 flags;
4343 __u64 process_table; 4312 __u64 process_table;
4344 }; 4313 };
4345 4314
4346 There are two bits that can be set in flags; 4315 There are two bits that can be set in flags; KVM_PPC_MMUV3_RADIX and
4347 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if 4316 KVM_PPC_MMUV3_GTSE. KVM_PPC_MMUV3_RADIX, if set, configures the guest
4348 to use radix tree translation, and if clear, 4317 to use radix tree translation, and if clear, to use HPT translation.
4349 KVM_PPC_MMUV3_GTSE, if set and if KVM permits 4318 KVM_PPC_MMUV3_GTSE, if set and if KVM permits it, configures the guest
4350 to be able to use the global TLB and SLB inva 4319 to be able to use the global TLB and SLB invalidation instructions;
4351 if clear, the guest may not use these instruc 4320 if clear, the guest may not use these instructions.
4352 4321
4353 The process_table field specifies the address 4322 The process_table field specifies the address and size of the guest
4354 process table, which is in the guest's space. 4323 process table, which is in the guest's space. This field is formatted
4355 as the second doubleword of the partition tab 4324 as the second doubleword of the partition table entry, as defined in
4356 the Power ISA V3.00, Book III section 5.7.6.1 4325 the Power ISA V3.00, Book III section 5.7.6.1.
4357 4326
4358 4.101 KVM_PPC_GET_RMMU_INFO 4327 4.101 KVM_PPC_GET_RMMU_INFO
4359 --------------------------- 4328 ---------------------------
4360 4329
4361 :Capability: KVM_CAP_PPC_MMU_RADIX !! 4330 :Capability: KVM_CAP_PPC_RADIX_MMU
4362 :Architectures: ppc 4331 :Architectures: ppc
4363 :Type: vm ioctl 4332 :Type: vm ioctl
4364 :Parameters: struct kvm_ppc_rmmu_info (out) 4333 :Parameters: struct kvm_ppc_rmmu_info (out)
4365 :Returns: 0 on success, 4334 :Returns: 0 on success,
4366 -EFAULT if struct kvm_ppc_rmmu_info 4335 -EFAULT if struct kvm_ppc_rmmu_info cannot be written,
4367 -EINVAL if no useful information can 4336 -EINVAL if no useful information can be returned
4368 4337
4369 This ioctl returns a structure containing two 4338 This ioctl returns a structure containing two things: (a) a list
4370 containing supported radix tree geometries, a 4339 containing supported radix tree geometries, and (b) a list that maps
4371 page sizes to put in the "AP" (actual page si 4340 page sizes to put in the "AP" (actual page size) field for the tlbie
4372 (TLB invalidate entry) instruction. 4341 (TLB invalidate entry) instruction.
4373 4342
4374 :: 4343 ::
4375 4344
4376 struct kvm_ppc_rmmu_info { 4345 struct kvm_ppc_rmmu_info {
4377 struct kvm_ppc_radix_geom { 4346 struct kvm_ppc_radix_geom {
4378 __u8 page_shift; 4347 __u8 page_shift;
4379 __u8 level_bits[4]; 4348 __u8 level_bits[4];
4380 __u8 pad[3]; 4349 __u8 pad[3];
4381 } geometries[8]; 4350 } geometries[8];
4382 __u32 ap_encodings[8]; 4351 __u32 ap_encodings[8];
4383 }; 4352 };
4384 4353
4385 The geometries[] field gives up to 8 supporte 4354 The geometries[] field gives up to 8 supported geometries for the
4386 radix page table, in terms of the log base 2 4355 radix page table, in terms of the log base 2 of the smallest page
4387 size, and the number of bits indexed at each 4356 size, and the number of bits indexed at each level of the tree, from
4388 the PTE level up to the PGD level in that ord 4357 the PTE level up to the PGD level in that order. Any unused entries
4389 will have 0 in the page_shift field. 4358 will have 0 in the page_shift field.
4390 4359
4391 The ap_encodings gives the supported page siz 4360 The ap_encodings gives the supported page sizes and their AP field
4392 encodings, encoded with the AP value in the t 4361 encodings, encoded with the AP value in the top 3 bits and the log
4393 base 2 of the page size in the bottom 6 bits. 4362 base 2 of the page size in the bottom 6 bits.
4394 4363
4395 4.102 KVM_PPC_RESIZE_HPT_PREPARE 4364 4.102 KVM_PPC_RESIZE_HPT_PREPARE
4396 -------------------------------- 4365 --------------------------------
4397 4366
4398 :Capability: KVM_CAP_SPAPR_RESIZE_HPT 4367 :Capability: KVM_CAP_SPAPR_RESIZE_HPT
4399 :Architectures: powerpc 4368 :Architectures: powerpc
4400 :Type: vm ioctl 4369 :Type: vm ioctl
4401 :Parameters: struct kvm_ppc_resize_hpt (in) 4370 :Parameters: struct kvm_ppc_resize_hpt (in)
4402 :Returns: 0 on successful completion, 4371 :Returns: 0 on successful completion,
4403 >0 if a new HPT is being prepared, t 4372 >0 if a new HPT is being prepared, the value is an estimated
4404 number of milliseconds until prepara 4373 number of milliseconds until preparation is complete,
4405 -EFAULT if struct kvm_reinject_contr 4374 -EFAULT if struct kvm_reinject_control cannot be read,
4406 -EINVAL if the supplied shift or fla 4375 -EINVAL if the supplied shift or flags are invalid,
4407 -ENOMEM if unable to allocate the ne 4376 -ENOMEM if unable to allocate the new HPT,
4408 4377
4409 Used to implement the PAPR extension for runt 4378 Used to implement the PAPR extension for runtime resizing of a guest's
4410 Hashed Page Table (HPT). Specifically this s 4379 Hashed Page Table (HPT). Specifically this starts, stops or monitors
4411 the preparation of a new potential HPT for th 4380 the preparation of a new potential HPT for the guest, essentially
4412 implementing the H_RESIZE_HPT_PREPARE hyperca 4381 implementing the H_RESIZE_HPT_PREPARE hypercall.
4413 4382
4414 :: 4383 ::
4415 4384
4416 struct kvm_ppc_resize_hpt { 4385 struct kvm_ppc_resize_hpt {
4417 __u64 flags; 4386 __u64 flags;
4418 __u32 shift; 4387 __u32 shift;
4419 __u32 pad; 4388 __u32 pad;
4420 }; 4389 };
4421 4390
4422 If called with shift > 0 when there is no pen 4391 If called with shift > 0 when there is no pending HPT for the guest,
4423 this begins preparation of a new pending HPT 4392 this begins preparation of a new pending HPT of size 2^(shift) bytes.
4424 It then returns a positive integer with the e 4393 It then returns a positive integer with the estimated number of
4425 milliseconds until preparation is complete. 4394 milliseconds until preparation is complete.
4426 4395
4427 If called when there is a pending HPT whose s 4396 If called when there is a pending HPT whose size does not match that
4428 requested in the parameters, discards the exi 4397 requested in the parameters, discards the existing pending HPT and
4429 creates a new one as above. 4398 creates a new one as above.
4430 4399
4431 If called when there is a pending HPT of the 4400 If called when there is a pending HPT of the size requested, will:
4432 4401
4433 * If preparation of the pending HPT is alre 4402 * If preparation of the pending HPT is already complete, return 0
4434 * If preparation of the pending HPT has fai 4403 * If preparation of the pending HPT has failed, return an error
4435 code, then discard the pending HPT. 4404 code, then discard the pending HPT.
4436 * If preparation of the pending HPT is stil 4405 * If preparation of the pending HPT is still in progress, return an
4437 estimated number of milliseconds until pr 4406 estimated number of milliseconds until preparation is complete.
4438 4407
4439 If called with shift == 0, discards any curre 4408 If called with shift == 0, discards any currently pending HPT and
4440 returns 0 (i.e. cancels any in-progress prepa 4409 returns 0 (i.e. cancels any in-progress preparation).
4441 4410
4442 flags is reserved for future expansion, curre 4411 flags is reserved for future expansion, currently setting any bits in
4443 flags will result in an -EINVAL. 4412 flags will result in an -EINVAL.
4444 4413
4445 Normally this will be called repeatedly with 4414 Normally this will be called repeatedly with the same parameters until
4446 it returns <= 0. The first call will initiat 4415 it returns <= 0. The first call will initiate preparation, subsequent
4447 ones will monitor preparation until it comple 4416 ones will monitor preparation until it completes or fails.
4448 4417
4449 4.103 KVM_PPC_RESIZE_HPT_COMMIT 4418 4.103 KVM_PPC_RESIZE_HPT_COMMIT
4450 ------------------------------- 4419 -------------------------------
4451 4420
4452 :Capability: KVM_CAP_SPAPR_RESIZE_HPT 4421 :Capability: KVM_CAP_SPAPR_RESIZE_HPT
4453 :Architectures: powerpc 4422 :Architectures: powerpc
4454 :Type: vm ioctl 4423 :Type: vm ioctl
4455 :Parameters: struct kvm_ppc_resize_hpt (in) 4424 :Parameters: struct kvm_ppc_resize_hpt (in)
4456 :Returns: 0 on successful completion, 4425 :Returns: 0 on successful completion,
4457 -EFAULT if struct kvm_reinject_contr 4426 -EFAULT if struct kvm_reinject_control cannot be read,
4458 -EINVAL if the supplied shift or fla 4427 -EINVAL if the supplied shift or flags are invalid,
4459 -ENXIO is there is no pending HPT, o 4428 -ENXIO is there is no pending HPT, or the pending HPT doesn't
4460 have the requested size, 4429 have the requested size,
4461 -EBUSY if the pending HPT is not ful 4430 -EBUSY if the pending HPT is not fully prepared,
4462 -ENOSPC if there was a hash collisio 4431 -ENOSPC if there was a hash collision when moving existing
4463 HPT entries to the new HPT, 4432 HPT entries to the new HPT,
4464 -EIO on other error conditions 4433 -EIO on other error conditions
4465 4434
4466 Used to implement the PAPR extension for runt 4435 Used to implement the PAPR extension for runtime resizing of a guest's
4467 Hashed Page Table (HPT). Specifically this r 4436 Hashed Page Table (HPT). Specifically this requests that the guest be
4468 transferred to working with the new HPT, esse 4437 transferred to working with the new HPT, essentially implementing the
4469 H_RESIZE_HPT_COMMIT hypercall. 4438 H_RESIZE_HPT_COMMIT hypercall.
4470 4439
4471 :: 4440 ::
4472 4441
4473 struct kvm_ppc_resize_hpt { 4442 struct kvm_ppc_resize_hpt {
4474 __u64 flags; 4443 __u64 flags;
4475 __u32 shift; 4444 __u32 shift;
4476 __u32 pad; 4445 __u32 pad;
4477 }; 4446 };
4478 4447
4479 This should only be called after KVM_PPC_RESI 4448 This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has
4480 returned 0 with the same parameters. In othe 4449 returned 0 with the same parameters. In other cases
4481 KVM_PPC_RESIZE_HPT_COMMIT will return an erro 4450 KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or
4482 -EBUSY, though others may be possible if the 4451 -EBUSY, though others may be possible if the preparation was started,
4483 but failed). 4452 but failed).
4484 4453
4485 This will have undefined effects on the guest 4454 This will have undefined effects on the guest if it has not already
4486 placed itself in a quiescent state where no v 4455 placed itself in a quiescent state where no vcpu will make MMU enabled
4487 memory accesses. 4456 memory accesses.
4488 4457
4489 On successful completion, the pending HPT wil 4458 On successful completion, the pending HPT will become the guest's active
4490 HPT and the previous HPT will be discarded. 4459 HPT and the previous HPT will be discarded.
4491 4460
4492 On failure, the guest will still be operating 4461 On failure, the guest will still be operating on its previous HPT.
4493 4462
4494 4.104 KVM_X86_GET_MCE_CAP_SUPPORTED 4463 4.104 KVM_X86_GET_MCE_CAP_SUPPORTED
4495 ----------------------------------- 4464 -----------------------------------
4496 4465
4497 :Capability: KVM_CAP_MCE 4466 :Capability: KVM_CAP_MCE
4498 :Architectures: x86 4467 :Architectures: x86
4499 :Type: system ioctl 4468 :Type: system ioctl
4500 :Parameters: u64 mce_cap (out) 4469 :Parameters: u64 mce_cap (out)
4501 :Returns: 0 on success, -1 on error 4470 :Returns: 0 on success, -1 on error
4502 4471
4503 Returns supported MCE capabilities. The u64 m 4472 Returns supported MCE capabilities. The u64 mce_cap parameter
4504 has the same format as the MSR_IA32_MCG_CAP r 4473 has the same format as the MSR_IA32_MCG_CAP register. Supported
4505 capabilities will have the corresponding bits 4474 capabilities will have the corresponding bits set.
4506 4475
4507 4.105 KVM_X86_SETUP_MCE 4476 4.105 KVM_X86_SETUP_MCE
4508 ----------------------- 4477 -----------------------
4509 4478
4510 :Capability: KVM_CAP_MCE 4479 :Capability: KVM_CAP_MCE
4511 :Architectures: x86 4480 :Architectures: x86
4512 :Type: vcpu ioctl 4481 :Type: vcpu ioctl
4513 :Parameters: u64 mcg_cap (in) 4482 :Parameters: u64 mcg_cap (in)
4514 :Returns: 0 on success, 4483 :Returns: 0 on success,
4515 -EFAULT if u64 mcg_cap cannot be rea 4484 -EFAULT if u64 mcg_cap cannot be read,
4516 -EINVAL if the requested number of b 4485 -EINVAL if the requested number of banks is invalid,
4517 -EINVAL if requested MCE capability 4486 -EINVAL if requested MCE capability is not supported.
4518 4487
4519 Initializes MCE support for use. The u64 mcg_ 4488 Initializes MCE support for use. The u64 mcg_cap parameter
4520 has the same format as the MSR_IA32_MCG_CAP r 4489 has the same format as the MSR_IA32_MCG_CAP register and
4521 specifies which capabilities should be enable 4490 specifies which capabilities should be enabled. The maximum
4522 supported number of error-reporting banks can 4491 supported number of error-reporting banks can be retrieved when
4523 checking for KVM_CAP_MCE. The supported capab 4492 checking for KVM_CAP_MCE. The supported capabilities can be
4524 retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED. 4493 retrieved with KVM_X86_GET_MCE_CAP_SUPPORTED.
4525 4494
4526 4.106 KVM_X86_SET_MCE 4495 4.106 KVM_X86_SET_MCE
4527 --------------------- 4496 ---------------------
4528 4497
4529 :Capability: KVM_CAP_MCE 4498 :Capability: KVM_CAP_MCE
4530 :Architectures: x86 4499 :Architectures: x86
4531 :Type: vcpu ioctl 4500 :Type: vcpu ioctl
4532 :Parameters: struct kvm_x86_mce (in) 4501 :Parameters: struct kvm_x86_mce (in)
4533 :Returns: 0 on success, 4502 :Returns: 0 on success,
4534 -EFAULT if struct kvm_x86_mce cannot 4503 -EFAULT if struct kvm_x86_mce cannot be read,
4535 -EINVAL if the bank number is invali 4504 -EINVAL if the bank number is invalid,
4536 -EINVAL if VAL bit is not set in sta 4505 -EINVAL if VAL bit is not set in status field.
4537 4506
4538 Inject a machine check error (MCE) into the g 4507 Inject a machine check error (MCE) into the guest. The input
4539 parameter is:: 4508 parameter is::
4540 4509
4541 struct kvm_x86_mce { 4510 struct kvm_x86_mce {
4542 __u64 status; 4511 __u64 status;
4543 __u64 addr; 4512 __u64 addr;
4544 __u64 misc; 4513 __u64 misc;
4545 __u64 mcg_status; 4514 __u64 mcg_status;
4546 __u8 bank; 4515 __u8 bank;
4547 __u8 pad1[7]; 4516 __u8 pad1[7];
4548 __u64 pad2[3]; 4517 __u64 pad2[3];
4549 }; 4518 };
4550 4519
4551 If the MCE being reported is an uncorrected e 4520 If the MCE being reported is an uncorrected error, KVM will
4552 inject it as an MCE exception into the guest. 4521 inject it as an MCE exception into the guest. If the guest
4553 MCG_STATUS register reports that an MCE is in 4522 MCG_STATUS register reports that an MCE is in progress, KVM
4554 causes an KVM_EXIT_SHUTDOWN vmexit. 4523 causes an KVM_EXIT_SHUTDOWN vmexit.
4555 4524
4556 Otherwise, if the MCE is a corrected error, K 4525 Otherwise, if the MCE is a corrected error, KVM will just
4557 store it in the corresponding bank (provided 4526 store it in the corresponding bank (provided this bank is
4558 not holding a previously reported uncorrected 4527 not holding a previously reported uncorrected error).
4559 4528
4560 4.107 KVM_S390_GET_CMMA_BITS 4529 4.107 KVM_S390_GET_CMMA_BITS
4561 ---------------------------- 4530 ----------------------------
4562 4531
4563 :Capability: KVM_CAP_S390_CMMA_MIGRATION 4532 :Capability: KVM_CAP_S390_CMMA_MIGRATION
4564 :Architectures: s390 4533 :Architectures: s390
4565 :Type: vm ioctl 4534 :Type: vm ioctl
4566 :Parameters: struct kvm_s390_cmma_log (in, ou 4535 :Parameters: struct kvm_s390_cmma_log (in, out)
4567 :Returns: 0 on success, a negative value on e 4536 :Returns: 0 on success, a negative value on error
4568 4537
4569 Errors: 4538 Errors:
4570 4539
4571 ====== ================================ 4540 ====== =============================================================
4572 ENOMEM not enough memory can be allocat 4541 ENOMEM not enough memory can be allocated to complete the task
4573 ENXIO if CMMA is not enabled 4542 ENXIO if CMMA is not enabled
4574 EINVAL if KVM_S390_CMMA_PEEK is not set 4543 EINVAL if KVM_S390_CMMA_PEEK is not set but migration mode was not enabled
4575 EINVAL if KVM_S390_CMMA_PEEK is not set 4544 EINVAL if KVM_S390_CMMA_PEEK is not set but dirty tracking has been
4576 disabled (and thus migration mod 4545 disabled (and thus migration mode was automatically disabled)
4577 EFAULT if the userspace address is inva 4546 EFAULT if the userspace address is invalid or if no page table is
4578 present for the addresses (e.g. 4547 present for the addresses (e.g. when using hugepages).
4579 ====== ================================ 4548 ====== =============================================================
4580 4549
4581 This ioctl is used to get the values of the C 4550 This ioctl is used to get the values of the CMMA bits on the s390
4582 architecture. It is meant to be used in two s 4551 architecture. It is meant to be used in two scenarios:
4583 4552
4584 - During live migration to save the CMMA valu 4553 - During live migration to save the CMMA values. Live migration needs
4585 to be enabled via the KVM_REQ_START_MIGRATI 4554 to be enabled via the KVM_REQ_START_MIGRATION VM property.
4586 - To non-destructively peek at the CMMA value 4555 - To non-destructively peek at the CMMA values, with the flag
4587 KVM_S390_CMMA_PEEK set. 4556 KVM_S390_CMMA_PEEK set.
4588 4557
4589 The ioctl takes parameters via the kvm_s390_c 4558 The ioctl takes parameters via the kvm_s390_cmma_log struct. The desired
4590 values are written to a buffer whose location 4559 values are written to a buffer whose location is indicated via the "values"
4591 member in the kvm_s390_cmma_log struct. The 4560 member in the kvm_s390_cmma_log struct. The values in the input struct are
4592 also updated as needed. 4561 also updated as needed.
4593 4562
4594 Each CMMA value takes up one byte. 4563 Each CMMA value takes up one byte.
4595 4564
4596 :: 4565 ::
4597 4566
4598 struct kvm_s390_cmma_log { 4567 struct kvm_s390_cmma_log {
4599 __u64 start_gfn; 4568 __u64 start_gfn;
4600 __u32 count; 4569 __u32 count;
4601 __u32 flags; 4570 __u32 flags;
4602 union { 4571 union {
4603 __u64 remaining; 4572 __u64 remaining;
4604 __u64 mask; 4573 __u64 mask;
4605 }; 4574 };
4606 __u64 values; 4575 __u64 values;
4607 }; 4576 };
4608 4577
4609 start_gfn is the number of the first guest fr 4578 start_gfn is the number of the first guest frame whose CMMA values are
4610 to be retrieved, 4579 to be retrieved,
4611 4580
4612 count is the length of the buffer in bytes, 4581 count is the length of the buffer in bytes,
4613 4582
4614 values points to the buffer where the result 4583 values points to the buffer where the result will be written to.
4615 4584
4616 If count is greater than KVM_S390_SKEYS_MAX, 4585 If count is greater than KVM_S390_SKEYS_MAX, then it is considered to be
4617 KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re- 4586 KVM_S390_SKEYS_MAX. KVM_S390_SKEYS_MAX is re-used for consistency with
4618 other ioctls. 4587 other ioctls.
4619 4588
4620 The result is written in the buffer pointed t 4589 The result is written in the buffer pointed to by the field values, and
4621 the values of the input parameter are updated 4590 the values of the input parameter are updated as follows.
4622 4591
4623 Depending on the flags, different actions are 4592 Depending on the flags, different actions are performed. The only
4624 supported flag so far is KVM_S390_CMMA_PEEK. 4593 supported flag so far is KVM_S390_CMMA_PEEK.
4625 4594
4626 The default behaviour if KVM_S390_CMMA_PEEK i 4595 The default behaviour if KVM_S390_CMMA_PEEK is not set is:
4627 start_gfn will indicate the first page frame 4596 start_gfn will indicate the first page frame whose CMMA bits were dirty.
4628 It is not necessarily the same as the one pas 4597 It is not necessarily the same as the one passed as input, as clean pages
4629 are skipped. 4598 are skipped.
4630 4599
4631 count will indicate the number of bytes actua 4600 count will indicate the number of bytes actually written in the buffer.
4632 It can (and very often will) be smaller than 4601 It can (and very often will) be smaller than the input value, since the
4633 buffer is only filled until 16 bytes of clean 4602 buffer is only filled until 16 bytes of clean values are found (which
4634 are then not copied in the buffer). Since a C 4603 are then not copied in the buffer). Since a CMMA migration block needs
4635 the base address and the length, for a total 4604 the base address and the length, for a total of 16 bytes, we will send
4636 back some clean data if there is some dirty d 4605 back some clean data if there is some dirty data afterwards, as long as
4637 the size of the clean data does not exceed th 4606 the size of the clean data does not exceed the size of the header. This
4638 allows to minimize the amount of data to be s 4607 allows to minimize the amount of data to be saved or transferred over
4639 the network at the expense of more roundtrips 4608 the network at the expense of more roundtrips to userspace. The next
4640 invocation of the ioctl will skip over all th 4609 invocation of the ioctl will skip over all the clean values, saving
4641 potentially more than just the 16 bytes we fo 4610 potentially more than just the 16 bytes we found.
4642 4611
4643 If KVM_S390_CMMA_PEEK is set: 4612 If KVM_S390_CMMA_PEEK is set:
4644 the existing storage attributes are read even 4613 the existing storage attributes are read even when not in migration
4645 mode, and no other action is performed; 4614 mode, and no other action is performed;
4646 4615
4647 the output start_gfn will be equal to the inp 4616 the output start_gfn will be equal to the input start_gfn,
4648 4617
4649 the output count will be equal to the input c 4618 the output count will be equal to the input count, except if the end of
4650 memory has been reached. 4619 memory has been reached.
4651 4620
4652 In both cases: 4621 In both cases:
4653 the field "remaining" will indicate the total 4622 the field "remaining" will indicate the total number of dirty CMMA values
4654 still remaining, or 0 if KVM_S390_CMMA_PEEK i 4623 still remaining, or 0 if KVM_S390_CMMA_PEEK is set and migration mode is
4655 not enabled. 4624 not enabled.
4656 4625
4657 mask is unused. 4626 mask is unused.
4658 4627
4659 values points to the userspace buffer where t 4628 values points to the userspace buffer where the result will be stored.
4660 4629
4661 4.108 KVM_S390_SET_CMMA_BITS 4630 4.108 KVM_S390_SET_CMMA_BITS
4662 ---------------------------- 4631 ----------------------------
4663 4632
4664 :Capability: KVM_CAP_S390_CMMA_MIGRATION 4633 :Capability: KVM_CAP_S390_CMMA_MIGRATION
4665 :Architectures: s390 4634 :Architectures: s390
4666 :Type: vm ioctl 4635 :Type: vm ioctl
4667 :Parameters: struct kvm_s390_cmma_log (in) 4636 :Parameters: struct kvm_s390_cmma_log (in)
4668 :Returns: 0 on success, a negative value on e 4637 :Returns: 0 on success, a negative value on error
4669 4638
4670 This ioctl is used to set the values of the C 4639 This ioctl is used to set the values of the CMMA bits on the s390
4671 architecture. It is meant to be used during l 4640 architecture. It is meant to be used during live migration to restore
4672 the CMMA values, but there are no restriction 4641 the CMMA values, but there are no restrictions on its use.
4673 The ioctl takes parameters via the kvm_s390_c 4642 The ioctl takes parameters via the kvm_s390_cmma_values struct.
4674 Each CMMA value takes up one byte. 4643 Each CMMA value takes up one byte.
4675 4644
4676 :: 4645 ::
4677 4646
4678 struct kvm_s390_cmma_log { 4647 struct kvm_s390_cmma_log {
4679 __u64 start_gfn; 4648 __u64 start_gfn;
4680 __u32 count; 4649 __u32 count;
4681 __u32 flags; 4650 __u32 flags;
4682 union { 4651 union {
4683 __u64 remaining; 4652 __u64 remaining;
4684 __u64 mask; 4653 __u64 mask;
4685 }; 4654 };
4686 __u64 values; 4655 __u64 values;
4687 }; 4656 };
4688 4657
4689 start_gfn indicates the starting guest frame 4658 start_gfn indicates the starting guest frame number,
4690 4659
4691 count indicates how many values are to be con 4660 count indicates how many values are to be considered in the buffer,
4692 4661
4693 flags is not used and must be 0. 4662 flags is not used and must be 0.
4694 4663
4695 mask indicates which PGSTE bits are to be con 4664 mask indicates which PGSTE bits are to be considered.
4696 4665
4697 remaining is not used. 4666 remaining is not used.
4698 4667
4699 values points to the buffer in userspace wher 4668 values points to the buffer in userspace where to store the values.
4700 4669
4701 This ioctl can fail with -ENOMEM if not enoug 4670 This ioctl can fail with -ENOMEM if not enough memory can be allocated to
4702 complete the task, with -ENXIO if CMMA is not 4671 complete the task, with -ENXIO if CMMA is not enabled, with -EINVAL if
4703 the count field is too large (e.g. more than 4672 the count field is too large (e.g. more than KVM_S390_CMMA_SIZE_MAX) or
4704 if the flags field was not 0, with -EFAULT if 4673 if the flags field was not 0, with -EFAULT if the userspace address is
4705 invalid, if invalid pages are written to (e.g 4674 invalid, if invalid pages are written to (e.g. after the end of memory)
4706 or if no page table is present for the addres 4675 or if no page table is present for the addresses (e.g. when using
4707 hugepages). 4676 hugepages).
4708 4677
4709 4.109 KVM_PPC_GET_CPU_CHAR 4678 4.109 KVM_PPC_GET_CPU_CHAR
4710 -------------------------- 4679 --------------------------
4711 4680
4712 :Capability: KVM_CAP_PPC_GET_CPU_CHAR 4681 :Capability: KVM_CAP_PPC_GET_CPU_CHAR
4713 :Architectures: powerpc 4682 :Architectures: powerpc
4714 :Type: vm ioctl 4683 :Type: vm ioctl
4715 :Parameters: struct kvm_ppc_cpu_char (out) 4684 :Parameters: struct kvm_ppc_cpu_char (out)
4716 :Returns: 0 on successful completion, 4685 :Returns: 0 on successful completion,
4717 -EFAULT if struct kvm_ppc_cpu_char c 4686 -EFAULT if struct kvm_ppc_cpu_char cannot be written
4718 4687
4719 This ioctl gives userspace information about 4688 This ioctl gives userspace information about certain characteristics
4720 of the CPU relating to speculative execution 4689 of the CPU relating to speculative execution of instructions and
4721 possible information leakage resulting from s 4690 possible information leakage resulting from speculative execution (see
4722 CVE-2017-5715, CVE-2017-5753 and CVE-2017-575 4691 CVE-2017-5715, CVE-2017-5753 and CVE-2017-5754). The information is
4723 returned in struct kvm_ppc_cpu_char, which lo 4692 returned in struct kvm_ppc_cpu_char, which looks like this::
4724 4693
4725 struct kvm_ppc_cpu_char { 4694 struct kvm_ppc_cpu_char {
4726 __u64 character; /* ch 4695 __u64 character; /* characteristics of the CPU */
4727 __u64 behaviour; /* re 4696 __u64 behaviour; /* recommended software behaviour */
4728 __u64 character_mask; /* va 4697 __u64 character_mask; /* valid bits in character */
4729 __u64 behaviour_mask; /* va 4698 __u64 behaviour_mask; /* valid bits in behaviour */
4730 }; 4699 };
4731 4700
4732 For extensibility, the character_mask and beh 4701 For extensibility, the character_mask and behaviour_mask fields
4733 indicate which bits of character and behaviou 4702 indicate which bits of character and behaviour have been filled in by
4734 the kernel. If the set of defined bits is ex 4703 the kernel. If the set of defined bits is extended in future then
4735 userspace will be able to tell whether it is 4704 userspace will be able to tell whether it is running on a kernel that
4736 knows about the new bits. 4705 knows about the new bits.
4737 4706
4738 The character field describes attributes of t 4707 The character field describes attributes of the CPU which can help
4739 with preventing inadvertent information discl 4708 with preventing inadvertent information disclosure - specifically,
4740 whether there is an instruction to flash-inva 4709 whether there is an instruction to flash-invalidate the L1 data cache
4741 (ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether 4710 (ori 30,30,0 or mtspr SPRN_TRIG2,rN), whether the L1 data cache is set
4742 to a mode where entries can only be used by t 4711 to a mode where entries can only be used by the thread that created
4743 them, whether the bcctr[l] instruction preven 4712 them, whether the bcctr[l] instruction prevents speculation, and
4744 whether a speculation barrier instruction (or 4713 whether a speculation barrier instruction (ori 31,31,0) is provided.
4745 4714
4746 The behaviour field describes actions that so 4715 The behaviour field describes actions that software should take to
4747 prevent inadvertent information disclosure, a 4716 prevent inadvertent information disclosure, and thus describes which
4748 vulnerabilities the hardware is subject to; s 4717 vulnerabilities the hardware is subject to; specifically whether the
4749 L1 data cache should be flushed when returnin 4718 L1 data cache should be flushed when returning to user mode from the
4750 kernel, and whether a speculation barrier sho 4719 kernel, and whether a speculation barrier should be placed between an
4751 array bounds check and the array access. 4720 array bounds check and the array access.
4752 4721
4753 These fields use the same bit definitions as 4722 These fields use the same bit definitions as the new
4754 H_GET_CPU_CHARACTERISTICS hypercall. 4723 H_GET_CPU_CHARACTERISTICS hypercall.
4755 4724
4756 4.110 KVM_MEMORY_ENCRYPT_OP 4725 4.110 KVM_MEMORY_ENCRYPT_OP
4757 --------------------------- 4726 ---------------------------
4758 4727
4759 :Capability: basic 4728 :Capability: basic
4760 :Architectures: x86 4729 :Architectures: x86
4761 :Type: vm 4730 :Type: vm
4762 :Parameters: an opaque platform specific stru 4731 :Parameters: an opaque platform specific structure (in/out)
4763 :Returns: 0 on success; -1 on error 4732 :Returns: 0 on success; -1 on error
4764 4733
4765 If the platform supports creating encrypted V 4734 If the platform supports creating encrypted VMs then this ioctl can be used
4766 for issuing platform-specific memory encrypti 4735 for issuing platform-specific memory encryption commands to manage those
4767 encrypted VMs. 4736 encrypted VMs.
4768 4737
4769 Currently, this ioctl is used for issuing Sec 4738 Currently, this ioctl is used for issuing Secure Encrypted Virtualization
4770 (SEV) commands on AMD Processors. The SEV com 4739 (SEV) commands on AMD Processors. The SEV commands are defined in
4771 Documentation/virt/kvm/x86/amd-memory-encrypt 4740 Documentation/virt/kvm/x86/amd-memory-encryption.rst.
4772 4741
4773 4.111 KVM_MEMORY_ENCRYPT_REG_REGION 4742 4.111 KVM_MEMORY_ENCRYPT_REG_REGION
4774 ----------------------------------- 4743 -----------------------------------
4775 4744
4776 :Capability: basic 4745 :Capability: basic
4777 :Architectures: x86 4746 :Architectures: x86
4778 :Type: system 4747 :Type: system
4779 :Parameters: struct kvm_enc_region (in) 4748 :Parameters: struct kvm_enc_region (in)
4780 :Returns: 0 on success; -1 on error 4749 :Returns: 0 on success; -1 on error
4781 4750
4782 This ioctl can be used to register a guest me 4751 This ioctl can be used to register a guest memory region which may
4783 contain encrypted data (e.g. guest RAM, SMRAM 4752 contain encrypted data (e.g. guest RAM, SMRAM etc).
4784 4753
4785 It is used in the SEV-enabled guest. When enc 4754 It is used in the SEV-enabled guest. When encryption is enabled, a guest
4786 memory region may contain encrypted data. The 4755 memory region may contain encrypted data. The SEV memory encryption
4787 engine uses a tweak such that two identical p 4756 engine uses a tweak such that two identical plaintext pages, each at
4788 different locations will have differing ciphe 4757 different locations will have differing ciphertexts. So swapping or
4789 moving ciphertext of those pages will not res 4758 moving ciphertext of those pages will not result in plaintext being
4790 swapped. So relocating (or migrating) physica 4759 swapped. So relocating (or migrating) physical backing pages for the SEV
4791 guest will require some additional steps. 4760 guest will require some additional steps.
4792 4761
4793 Note: The current SEV key management spec doe 4762 Note: The current SEV key management spec does not provide commands to
4794 swap or migrate (move) ciphertext pages. Henc 4763 swap or migrate (move) ciphertext pages. Hence, for now we pin the guest
4795 memory region registered with the ioctl. 4764 memory region registered with the ioctl.
4796 4765
4797 4.112 KVM_MEMORY_ENCRYPT_UNREG_REGION 4766 4.112 KVM_MEMORY_ENCRYPT_UNREG_REGION
4798 ------------------------------------- 4767 -------------------------------------
4799 4768
4800 :Capability: basic 4769 :Capability: basic
4801 :Architectures: x86 4770 :Architectures: x86
4802 :Type: system 4771 :Type: system
4803 :Parameters: struct kvm_enc_region (in) 4772 :Parameters: struct kvm_enc_region (in)
4804 :Returns: 0 on success; -1 on error 4773 :Returns: 0 on success; -1 on error
4805 4774
4806 This ioctl can be used to unregister the gues 4775 This ioctl can be used to unregister the guest memory region registered
4807 with KVM_MEMORY_ENCRYPT_REG_REGION ioctl abov 4776 with KVM_MEMORY_ENCRYPT_REG_REGION ioctl above.
4808 4777
4809 4.113 KVM_HYPERV_EVENTFD 4778 4.113 KVM_HYPERV_EVENTFD
4810 ------------------------ 4779 ------------------------
4811 4780
4812 :Capability: KVM_CAP_HYPERV_EVENTFD 4781 :Capability: KVM_CAP_HYPERV_EVENTFD
4813 :Architectures: x86 4782 :Architectures: x86
4814 :Type: vm ioctl 4783 :Type: vm ioctl
4815 :Parameters: struct kvm_hyperv_eventfd (in) 4784 :Parameters: struct kvm_hyperv_eventfd (in)
4816 4785
4817 This ioctl (un)registers an eventfd to receiv 4786 This ioctl (un)registers an eventfd to receive notifications from the guest on
4818 the specified Hyper-V connection id through t 4787 the specified Hyper-V connection id through the SIGNAL_EVENT hypercall, without
4819 causing a user exit. SIGNAL_EVENT hypercall 4788 causing a user exit. SIGNAL_EVENT hypercall with non-zero event flag number
4820 (bits 24-31) still triggers a KVM_EXIT_HYPERV 4789 (bits 24-31) still triggers a KVM_EXIT_HYPERV_HCALL user exit.
4821 4790
4822 :: 4791 ::
4823 4792
4824 struct kvm_hyperv_eventfd { 4793 struct kvm_hyperv_eventfd {
4825 __u32 conn_id; 4794 __u32 conn_id;
4826 __s32 fd; 4795 __s32 fd;
4827 __u32 flags; 4796 __u32 flags;
4828 __u32 padding[3]; 4797 __u32 padding[3];
4829 }; 4798 };
4830 4799
4831 The conn_id field should fit within 24 bits:: 4800 The conn_id field should fit within 24 bits::
4832 4801
4833 #define KVM_HYPERV_CONN_ID_MASK 4802 #define KVM_HYPERV_CONN_ID_MASK 0x00ffffff
4834 4803
4835 The acceptable values for the flags field are 4804 The acceptable values for the flags field are::
4836 4805
4837 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 4806 #define KVM_HYPERV_EVENTFD_DEASSIGN (1 << 0)
4838 4807
4839 :Returns: 0 on success, 4808 :Returns: 0 on success,
4840 -EINVAL if conn_id or flags is outs 4809 -EINVAL if conn_id or flags is outside the allowed range,
4841 -ENOENT on deassign if the conn_id 4810 -ENOENT on deassign if the conn_id isn't registered,
4842 -EEXIST on assign if the conn_id is 4811 -EEXIST on assign if the conn_id is already registered
4843 4812
4844 4.114 KVM_GET_NESTED_STATE 4813 4.114 KVM_GET_NESTED_STATE
4845 -------------------------- 4814 --------------------------
4846 4815
4847 :Capability: KVM_CAP_NESTED_STATE 4816 :Capability: KVM_CAP_NESTED_STATE
4848 :Architectures: x86 4817 :Architectures: x86
4849 :Type: vcpu ioctl 4818 :Type: vcpu ioctl
4850 :Parameters: struct kvm_nested_state (in/out) 4819 :Parameters: struct kvm_nested_state (in/out)
4851 :Returns: 0 on success, -1 on error 4820 :Returns: 0 on success, -1 on error
4852 4821
4853 Errors: 4822 Errors:
4854 4823
4855 ===== ================================ 4824 ===== =============================================================
4856 E2BIG the total state size exceeds the 4825 E2BIG the total state size exceeds the value of 'size' specified by
4857 the user; the size required will 4826 the user; the size required will be written into size.
4858 ===== ================================ 4827 ===== =============================================================
4859 4828
4860 :: 4829 ::
4861 4830
4862 struct kvm_nested_state { 4831 struct kvm_nested_state {
4863 __u16 flags; 4832 __u16 flags;
4864 __u16 format; 4833 __u16 format;
4865 __u32 size; 4834 __u32 size;
4866 4835
4867 union { 4836 union {
4868 struct kvm_vmx_nested_state_h 4837 struct kvm_vmx_nested_state_hdr vmx;
4869 struct kvm_svm_nested_state_h 4838 struct kvm_svm_nested_state_hdr svm;
4870 4839
4871 /* Pad the header to 128 byte 4840 /* Pad the header to 128 bytes. */
4872 __u8 pad[120]; 4841 __u8 pad[120];
4873 } hdr; 4842 } hdr;
4874 4843
4875 union { 4844 union {
4876 struct kvm_vmx_nested_state_d 4845 struct kvm_vmx_nested_state_data vmx[0];
4877 struct kvm_svm_nested_state_d 4846 struct kvm_svm_nested_state_data svm[0];
4878 } data; 4847 } data;
4879 }; 4848 };
4880 4849
4881 #define KVM_STATE_NESTED_GUEST_MODE 4850 #define KVM_STATE_NESTED_GUEST_MODE 0x00000001
4882 #define KVM_STATE_NESTED_RUN_PENDING 4851 #define KVM_STATE_NESTED_RUN_PENDING 0x00000002
4883 #define KVM_STATE_NESTED_EVMCS 4852 #define KVM_STATE_NESTED_EVMCS 0x00000004
4884 4853
4885 #define KVM_STATE_NESTED_FORMAT_VMX 4854 #define KVM_STATE_NESTED_FORMAT_VMX 0
4886 #define KVM_STATE_NESTED_FORMAT_SVM 4855 #define KVM_STATE_NESTED_FORMAT_SVM 1
4887 4856
4888 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 4857 #define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000
4889 4858
4890 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 4859 #define KVM_STATE_NESTED_VMX_SMM_GUEST_MODE 0x00000001
4891 #define KVM_STATE_NESTED_VMX_SMM_VMXON 4860 #define KVM_STATE_NESTED_VMX_SMM_VMXON 0x00000002
4892 4861
4893 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEAD 4862 #define KVM_STATE_VMX_PREEMPTION_TIMER_DEADLINE 0x00000001
4894 4863
4895 struct kvm_vmx_nested_state_hdr { 4864 struct kvm_vmx_nested_state_hdr {
4896 __u64 vmxon_pa; 4865 __u64 vmxon_pa;
4897 __u64 vmcs12_pa; 4866 __u64 vmcs12_pa;
4898 4867
4899 struct { 4868 struct {
4900 __u16 flags; 4869 __u16 flags;
4901 } smm; 4870 } smm;
4902 4871
4903 __u32 flags; 4872 __u32 flags;
4904 __u64 preemption_timer_deadline; 4873 __u64 preemption_timer_deadline;
4905 }; 4874 };
4906 4875
4907 struct kvm_vmx_nested_state_data { 4876 struct kvm_vmx_nested_state_data {
4908 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS 4877 __u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE];
4909 __u8 shadow_vmcs12[KVM_STATE_NESTED_V 4878 __u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE];
4910 }; 4879 };
4911 4880
4912 This ioctl copies the vcpu's nested virtualiz 4881 This ioctl copies the vcpu's nested virtualization state from the kernel to
4913 userspace. 4882 userspace.
4914 4883
4915 The maximum size of the state can be retrieve 4884 The maximum size of the state can be retrieved by passing KVM_CAP_NESTED_STATE
4916 to the KVM_CHECK_EXTENSION ioctl(). 4885 to the KVM_CHECK_EXTENSION ioctl().
4917 4886
4918 4.115 KVM_SET_NESTED_STATE 4887 4.115 KVM_SET_NESTED_STATE
4919 -------------------------- 4888 --------------------------
4920 4889
4921 :Capability: KVM_CAP_NESTED_STATE 4890 :Capability: KVM_CAP_NESTED_STATE
4922 :Architectures: x86 4891 :Architectures: x86
4923 :Type: vcpu ioctl 4892 :Type: vcpu ioctl
4924 :Parameters: struct kvm_nested_state (in) 4893 :Parameters: struct kvm_nested_state (in)
4925 :Returns: 0 on success, -1 on error 4894 :Returns: 0 on success, -1 on error
4926 4895
4927 This copies the vcpu's kvm_nested_state struc 4896 This copies the vcpu's kvm_nested_state struct from userspace to the kernel.
4928 For the definition of struct kvm_nested_state 4897 For the definition of struct kvm_nested_state, see KVM_GET_NESTED_STATE.
4929 4898
4930 4.116 KVM_(UN)REGISTER_COALESCED_MMIO 4899 4.116 KVM_(UN)REGISTER_COALESCED_MMIO
4931 ------------------------------------- 4900 -------------------------------------
4932 4901
4933 :Capability: KVM_CAP_COALESCED_MMIO (for coal 4902 :Capability: KVM_CAP_COALESCED_MMIO (for coalesced mmio)
4934 KVM_CAP_COALESCED_PIO (for coale 4903 KVM_CAP_COALESCED_PIO (for coalesced pio)
4935 :Architectures: all 4904 :Architectures: all
4936 :Type: vm ioctl 4905 :Type: vm ioctl
4937 :Parameters: struct kvm_coalesced_mmio_zone 4906 :Parameters: struct kvm_coalesced_mmio_zone
4938 :Returns: 0 on success, < 0 on error 4907 :Returns: 0 on success, < 0 on error
4939 4908
4940 Coalesced I/O is a performance optimization t 4909 Coalesced I/O is a performance optimization that defers hardware
4941 register write emulation so that userspace ex 4910 register write emulation so that userspace exits are avoided. It is
4942 typically used to reduce the overhead of emul 4911 typically used to reduce the overhead of emulating frequently accessed
4943 hardware registers. 4912 hardware registers.
4944 4913
4945 When a hardware register is configured for co 4914 When a hardware register is configured for coalesced I/O, write accesses
4946 do not exit to userspace and their value is r 4915 do not exit to userspace and their value is recorded in a ring buffer
4947 that is shared between kernel and userspace. 4916 that is shared between kernel and userspace.
4948 4917
4949 Coalesced I/O is used if one or more write ac 4918 Coalesced I/O is used if one or more write accesses to a hardware
4950 register can be deferred until a read or a wr 4919 register can be deferred until a read or a write to another hardware
4951 register on the same device. This last acces 4920 register on the same device. This last access will cause a vmexit and
4952 userspace will process accesses from the ring 4921 userspace will process accesses from the ring buffer before emulating
4953 it. That will avoid exiting to userspace on r 4922 it. That will avoid exiting to userspace on repeated writes.
4954 4923
4955 Coalesced pio is based on coalesced mmio. The 4924 Coalesced pio is based on coalesced mmio. There is little difference
4956 between coalesced mmio and pio except that co 4925 between coalesced mmio and pio except that coalesced pio records accesses
4957 to I/O ports. 4926 to I/O ports.
4958 4927
4959 4.117 KVM_CLEAR_DIRTY_LOG (vm ioctl) 4928 4.117 KVM_CLEAR_DIRTY_LOG (vm ioctl)
4960 ------------------------------------ 4929 ------------------------------------
4961 4930
4962 :Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT 4931 :Capability: KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
4963 :Architectures: x86, arm64, mips 4932 :Architectures: x86, arm64, mips
4964 :Type: vm ioctl 4933 :Type: vm ioctl
4965 :Parameters: struct kvm_clear_dirty_log (in) 4934 :Parameters: struct kvm_clear_dirty_log (in)
4966 :Returns: 0 on success, -1 on error 4935 :Returns: 0 on success, -1 on error
4967 4936
4968 :: 4937 ::
4969 4938
4970 /* for KVM_CLEAR_DIRTY_LOG */ 4939 /* for KVM_CLEAR_DIRTY_LOG */
4971 struct kvm_clear_dirty_log { 4940 struct kvm_clear_dirty_log {
4972 __u32 slot; 4941 __u32 slot;
4973 __u32 num_pages; 4942 __u32 num_pages;
4974 __u64 first_page; 4943 __u64 first_page;
4975 union { 4944 union {
4976 void __user *dirty_bitmap; /* 4945 void __user *dirty_bitmap; /* one bit per page */
4977 __u64 padding; 4946 __u64 padding;
4978 }; 4947 };
4979 }; 4948 };
4980 4949
4981 The ioctl clears the dirty status of pages in 4950 The ioctl clears the dirty status of pages in a memory slot, according to
4982 the bitmap that is passed in struct kvm_clear 4951 the bitmap that is passed in struct kvm_clear_dirty_log's dirty_bitmap
4983 field. Bit 0 of the bitmap corresponds to pa 4952 field. Bit 0 of the bitmap corresponds to page "first_page" in the
4984 memory slot, and num_pages is the size in bit 4953 memory slot, and num_pages is the size in bits of the input bitmap.
4985 first_page must be a multiple of 64; num_page 4954 first_page must be a multiple of 64; num_pages must also be a multiple of
4986 64 unless first_page + num_pages is the size 4955 64 unless first_page + num_pages is the size of the memory slot. For each
4987 bit that is set in the input bitmap, the corr 4956 bit that is set in the input bitmap, the corresponding page is marked "clean"
4988 in KVM's dirty bitmap, and dirty tracking is 4957 in KVM's dirty bitmap, and dirty tracking is re-enabled for that page
4989 (for example via write-protection, or by clea 4958 (for example via write-protection, or by clearing the dirty bit in
4990 a page table entry). 4959 a page table entry).
4991 4960
4992 If KVM_CAP_MULTI_ADDRESS_SPACE is available, 4961 If KVM_CAP_MULTI_ADDRESS_SPACE is available, bits 16-31 of slot field specifies
4993 the address space for which you want to clear 4962 the address space for which you want to clear the dirty status. See
4994 KVM_SET_USER_MEMORY_REGION for details on the 4963 KVM_SET_USER_MEMORY_REGION for details on the usage of slot field.
4995 4964
4996 This ioctl is mostly useful when KVM_CAP_MANU 4965 This ioctl is mostly useful when KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
4997 is enabled; for more information, see the des 4966 is enabled; for more information, see the description of the capability.
4998 However, it can always be used as long as KVM 4967 However, it can always be used as long as KVM_CHECK_EXTENSION confirms
4999 that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is pre 4968 that KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is present.
5000 4969
5001 4.118 KVM_GET_SUPPORTED_HV_CPUID 4970 4.118 KVM_GET_SUPPORTED_HV_CPUID
5002 -------------------------------- 4971 --------------------------------
5003 4972
5004 :Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM 4973 :Capability: KVM_CAP_HYPERV_CPUID (vcpu), KVM_CAP_SYS_HYPERV_CPUID (system)
5005 :Architectures: x86 4974 :Architectures: x86
5006 :Type: system ioctl, vcpu ioctl 4975 :Type: system ioctl, vcpu ioctl
5007 :Parameters: struct kvm_cpuid2 (in/out) 4976 :Parameters: struct kvm_cpuid2 (in/out)
5008 :Returns: 0 on success, -1 on error 4977 :Returns: 0 on success, -1 on error
5009 4978
5010 :: 4979 ::
5011 4980
5012 struct kvm_cpuid2 { 4981 struct kvm_cpuid2 {
5013 __u32 nent; 4982 __u32 nent;
5014 __u32 padding; 4983 __u32 padding;
5015 struct kvm_cpuid_entry2 entries[0]; 4984 struct kvm_cpuid_entry2 entries[0];
5016 }; 4985 };
5017 4986
5018 struct kvm_cpuid_entry2 { 4987 struct kvm_cpuid_entry2 {
5019 __u32 function; 4988 __u32 function;
5020 __u32 index; 4989 __u32 index;
5021 __u32 flags; 4990 __u32 flags;
5022 __u32 eax; 4991 __u32 eax;
5023 __u32 ebx; 4992 __u32 ebx;
5024 __u32 ecx; 4993 __u32 ecx;
5025 __u32 edx; 4994 __u32 edx;
5026 __u32 padding[3]; 4995 __u32 padding[3];
5027 }; 4996 };
5028 4997
5029 This ioctl returns x86 cpuid features leaves 4998 This ioctl returns x86 cpuid features leaves related to Hyper-V emulation in
5030 KVM. Userspace can use the information retur 4999 KVM. Userspace can use the information returned by this ioctl to construct
5031 cpuid information presented to guests consumi 5000 cpuid information presented to guests consuming Hyper-V enlightenments (e.g.
5032 Windows or Hyper-V guests). 5001 Windows or Hyper-V guests).
5033 5002
5034 CPUID feature leaves returned by this ioctl a 5003 CPUID feature leaves returned by this ioctl are defined by Hyper-V Top Level
5035 Functional Specification (TLFS). These leaves 5004 Functional Specification (TLFS). These leaves can't be obtained with
5036 KVM_GET_SUPPORTED_CPUID ioctl because some of 5005 KVM_GET_SUPPORTED_CPUID ioctl because some of them intersect with KVM feature
5037 leaves (0x40000000, 0x40000001). 5006 leaves (0x40000000, 0x40000001).
5038 5007
5039 Currently, the following list of CPUID leaves 5008 Currently, the following list of CPUID leaves are returned:
5040 5009
5041 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS 5010 - HYPERV_CPUID_VENDOR_AND_MAX_FUNCTIONS
5042 - HYPERV_CPUID_INTERFACE 5011 - HYPERV_CPUID_INTERFACE
5043 - HYPERV_CPUID_VERSION 5012 - HYPERV_CPUID_VERSION
5044 - HYPERV_CPUID_FEATURES 5013 - HYPERV_CPUID_FEATURES
5045 - HYPERV_CPUID_ENLIGHTMENT_INFO 5014 - HYPERV_CPUID_ENLIGHTMENT_INFO
5046 - HYPERV_CPUID_IMPLEMENT_LIMITS 5015 - HYPERV_CPUID_IMPLEMENT_LIMITS
5047 - HYPERV_CPUID_NESTED_FEATURES 5016 - HYPERV_CPUID_NESTED_FEATURES
5048 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIO 5017 - HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS
5049 - HYPERV_CPUID_SYNDBG_INTERFACE 5018 - HYPERV_CPUID_SYNDBG_INTERFACE
5050 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES 5019 - HYPERV_CPUID_SYNDBG_PLATFORM_CAPABILITIES
5051 5020
5052 Userspace invokes KVM_GET_SUPPORTED_HV_CPUID 5021 Userspace invokes KVM_GET_SUPPORTED_HV_CPUID by passing a kvm_cpuid2 structure
5053 with the 'nent' field indicating the number o 5022 with the 'nent' field indicating the number of entries in the variable-size
5054 array 'entries'. If the number of entries is 5023 array 'entries'. If the number of entries is too low to describe all Hyper-V
5055 feature leaves, an error (E2BIG) is returned. 5024 feature leaves, an error (E2BIG) is returned. If the number is more or equal
5056 to the number of Hyper-V feature leaves, the 5025 to the number of Hyper-V feature leaves, the 'nent' field is adjusted to the
5057 number of valid entries in the 'entries' arra 5026 number of valid entries in the 'entries' array, which is then filled.
5058 5027
5059 'index' and 'flags' fields in 'struct kvm_cpu 5028 'index' and 'flags' fields in 'struct kvm_cpuid_entry2' are currently reserved,
5060 userspace should not expect to get any partic 5029 userspace should not expect to get any particular value there.
5061 5030
5062 Note, vcpu version of KVM_GET_SUPPORTED_HV_CP 5031 Note, vcpu version of KVM_GET_SUPPORTED_HV_CPUID is currently deprecated. Unlike
5063 system ioctl which exposes all supported feat 5032 system ioctl which exposes all supported feature bits unconditionally, vcpu
5064 version has the following quirks: 5033 version has the following quirks:
5065 5034
5066 - HYPERV_CPUID_NESTED_FEATURES leaf and HV_X6 5035 - HYPERV_CPUID_NESTED_FEATURES leaf and HV_X64_ENLIGHTENED_VMCS_RECOMMENDED
5067 feature bit are only exposed when Enlighten 5036 feature bit are only exposed when Enlightened VMCS was previously enabled
5068 on the corresponding vCPU (KVM_CAP_HYPERV_E 5037 on the corresponding vCPU (KVM_CAP_HYPERV_ENLIGHTENED_VMCS).
5069 - HV_STIMER_DIRECT_MODE_AVAILABLE bit is only 5038 - HV_STIMER_DIRECT_MODE_AVAILABLE bit is only exposed with in-kernel LAPIC.
5070 (presumes KVM_CREATE_IRQCHIP has already be 5039 (presumes KVM_CREATE_IRQCHIP has already been called).
5071 5040
5072 4.119 KVM_ARM_VCPU_FINALIZE 5041 4.119 KVM_ARM_VCPU_FINALIZE
5073 --------------------------- 5042 ---------------------------
5074 5043
5075 :Architectures: arm64 5044 :Architectures: arm64
5076 :Type: vcpu ioctl 5045 :Type: vcpu ioctl
5077 :Parameters: int feature (in) 5046 :Parameters: int feature (in)
5078 :Returns: 0 on success, -1 on error 5047 :Returns: 0 on success, -1 on error
5079 5048
5080 Errors: 5049 Errors:
5081 5050
5082 ====== ================================ 5051 ====== ==============================================================
5083 EPERM feature not enabled, needs confi 5052 EPERM feature not enabled, needs configuration, or already finalized
5084 EINVAL feature unknown or not present 5053 EINVAL feature unknown or not present
5085 ====== ================================ 5054 ====== ==============================================================
5086 5055
5087 Recognised values for feature: 5056 Recognised values for feature:
5088 5057
5089 ===== ================================ 5058 ===== ===========================================
5090 arm64 KVM_ARM_VCPU_SVE (requires KVM_C 5059 arm64 KVM_ARM_VCPU_SVE (requires KVM_CAP_ARM_SVE)
5091 ===== ================================ 5060 ===== ===========================================
5092 5061
5093 Finalizes the configuration of the specified 5062 Finalizes the configuration of the specified vcpu feature.
5094 5063
5095 The vcpu must already have been initialised, 5064 The vcpu must already have been initialised, enabling the affected feature, by
5096 means of a successful KVM_ARM_VCPU_INIT call 5065 means of a successful KVM_ARM_VCPU_INIT call with the appropriate flag set in
5097 features[]. 5066 features[].
5098 5067
5099 For affected vcpu features, this is a mandato 5068 For affected vcpu features, this is a mandatory step that must be performed
5100 before the vcpu is fully usable. 5069 before the vcpu is fully usable.
5101 5070
5102 Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FI 5071 Between KVM_ARM_VCPU_INIT and KVM_ARM_VCPU_FINALIZE, the feature may be
5103 configured by use of ioctls such as KVM_SET_O 5072 configured by use of ioctls such as KVM_SET_ONE_REG. The exact configuration
5104 that should be performed and how to do it are 5073 that should be performed and how to do it are feature-dependent.
5105 5074
5106 Other calls that depend on a particular featu 5075 Other calls that depend on a particular feature being finalized, such as
5107 KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG an 5076 KVM_RUN, KVM_GET_REG_LIST, KVM_GET_ONE_REG and KVM_SET_ONE_REG, will fail with
5108 -EPERM unless the feature has already been fi 5077 -EPERM unless the feature has already been finalized by means of a
5109 KVM_ARM_VCPU_FINALIZE call. 5078 KVM_ARM_VCPU_FINALIZE call.
5110 5079
5111 See KVM_ARM_VCPU_INIT for details of vcpu fea 5080 See KVM_ARM_VCPU_INIT for details of vcpu features that require finalization
5112 using this ioctl. 5081 using this ioctl.
5113 5082
5114 4.120 KVM_SET_PMU_EVENT_FILTER 5083 4.120 KVM_SET_PMU_EVENT_FILTER
5115 ------------------------------ 5084 ------------------------------
5116 5085
5117 :Capability: KVM_CAP_PMU_EVENT_FILTER 5086 :Capability: KVM_CAP_PMU_EVENT_FILTER
5118 :Architectures: x86 5087 :Architectures: x86
5119 :Type: vm ioctl 5088 :Type: vm ioctl
5120 :Parameters: struct kvm_pmu_event_filter (in) 5089 :Parameters: struct kvm_pmu_event_filter (in)
5121 :Returns: 0 on success, -1 on error 5090 :Returns: 0 on success, -1 on error
5122 5091
5123 Errors: 5092 Errors:
5124 5093
5125 ====== ================================ 5094 ====== ============================================================
5126 EFAULT args[0] cannot be accessed 5095 EFAULT args[0] cannot be accessed
5127 EINVAL args[0] contains invalid data in 5096 EINVAL args[0] contains invalid data in the filter or filter events
5128 E2BIG nevents is too large 5097 E2BIG nevents is too large
5129 EBUSY not enough memory to allocate th 5098 EBUSY not enough memory to allocate the filter
5130 ====== ================================ 5099 ====== ============================================================
5131 5100
5132 :: 5101 ::
5133 5102
5134 struct kvm_pmu_event_filter { 5103 struct kvm_pmu_event_filter {
5135 __u32 action; 5104 __u32 action;
5136 __u32 nevents; 5105 __u32 nevents;
5137 __u32 fixed_counter_bitmap; 5106 __u32 fixed_counter_bitmap;
5138 __u32 flags; 5107 __u32 flags;
5139 __u32 pad[4]; 5108 __u32 pad[4];
5140 __u64 events[0]; 5109 __u64 events[0];
5141 }; 5110 };
5142 5111
5143 This ioctl restricts the set of PMU events th 5112 This ioctl restricts the set of PMU events the guest can program by limiting
5144 which event select and unit mask combinations 5113 which event select and unit mask combinations are permitted.
5145 5114
5146 The argument holds a list of filter events wh 5115 The argument holds a list of filter events which will be allowed or denied.
5147 5116
5148 Filter events only control general purpose co 5117 Filter events only control general purpose counters; fixed purpose counters
5149 are controlled by the fixed_counter_bitmap. 5118 are controlled by the fixed_counter_bitmap.
5150 5119
5151 Valid values for 'flags':: 5120 Valid values for 'flags'::
5152 5121
5153 ``0`` 5122 ``0``
5154 5123
5155 To use this mode, clear the 'flags' field. 5124 To use this mode, clear the 'flags' field.
5156 5125
5157 In this mode each event will contain an event 5126 In this mode each event will contain an event select + unit mask.
5158 5127
5159 When the guest attempts to program the PMU th 5128 When the guest attempts to program the PMU the guest's event select +
5160 unit mask is compared against the filter even 5129 unit mask is compared against the filter events to determine whether the
5161 guest should have access. 5130 guest should have access.
5162 5131
5163 ``KVM_PMU_EVENT_FLAG_MASKED_EVENTS`` 5132 ``KVM_PMU_EVENT_FLAG_MASKED_EVENTS``
5164 :Capability: KVM_CAP_PMU_EVENT_MASKED_EVENTS 5133 :Capability: KVM_CAP_PMU_EVENT_MASKED_EVENTS
5165 5134
5166 In this mode each filter event will contain a 5135 In this mode each filter event will contain an event select, mask, match, and
5167 exclude value. To encode a masked event use: 5136 exclude value. To encode a masked event use::
5168 5137
5169 KVM_PMU_ENCODE_MASKED_ENTRY() 5138 KVM_PMU_ENCODE_MASKED_ENTRY()
5170 5139
5171 An encoded event will follow this layout:: 5140 An encoded event will follow this layout::
5172 5141
5173 Bits Description 5142 Bits Description
5174 ---- ----------- 5143 ---- -----------
5175 7:0 event select (low bits) 5144 7:0 event select (low bits)
5176 15:8 umask match 5145 15:8 umask match
5177 31:16 unused 5146 31:16 unused
5178 35:32 event select (high bits) 5147 35:32 event select (high bits)
5179 36:54 unused 5148 36:54 unused
5180 55 exclude bit 5149 55 exclude bit
5181 63:56 umask mask 5150 63:56 umask mask
5182 5151
5183 When the guest attempts to program the PMU, t 5152 When the guest attempts to program the PMU, these steps are followed in
5184 determining if the guest should have access: 5153 determining if the guest should have access:
5185 5154
5186 1. Match the event select from the guest aga 5155 1. Match the event select from the guest against the filter events.
5187 2. If a match is found, match the guest's un 5156 2. If a match is found, match the guest's unit mask to the mask and match
5188 values of the included filter events. 5157 values of the included filter events.
5189 I.e. (unit mask & mask) == match && !excl 5158 I.e. (unit mask & mask) == match && !exclude.
5190 3. If a match is found, match the guest's un 5159 3. If a match is found, match the guest's unit mask to the mask and match
5191 values of the excluded filter events. 5160 values of the excluded filter events.
5192 I.e. (unit mask & mask) == match && exclu 5161 I.e. (unit mask & mask) == match && exclude.
5193 4. 5162 4.
5194 a. If an included match is found and an ex 5163 a. If an included match is found and an excluded match is not found, filter
5195 the event. 5164 the event.
5196 b. For everything else, do not filter the 5165 b. For everything else, do not filter the event.
5197 5. 5166 5.
5198 a. If the event is filtered and it's an al 5167 a. If the event is filtered and it's an allow list, allow the guest to
5199 program the event. 5168 program the event.
5200 b. If the event is filtered and it's a den 5169 b. If the event is filtered and it's a deny list, do not allow the guest to
5201 program the event. 5170 program the event.
5202 5171
5203 When setting a new pmu event filter, -EINVAL 5172 When setting a new pmu event filter, -EINVAL will be returned if any of the
5204 unused fields are set or if any of the high b 5173 unused fields are set or if any of the high bits (35:32) in the event
5205 select are set when called on Intel. 5174 select are set when called on Intel.
5206 5175
5207 Valid values for 'action':: 5176 Valid values for 'action'::
5208 5177
5209 #define KVM_PMU_EVENT_ALLOW 0 5178 #define KVM_PMU_EVENT_ALLOW 0
5210 #define KVM_PMU_EVENT_DENY 1 5179 #define KVM_PMU_EVENT_DENY 1
5211 5180
5212 Via this API, KVM userspace can also control 5181 Via this API, KVM userspace can also control the behavior of the VM's fixed
5213 counters (if any) by configuring the "action" 5182 counters (if any) by configuring the "action" and "fixed_counter_bitmap" fields.
5214 5183
5215 Specifically, KVM follows the following pseud 5184 Specifically, KVM follows the following pseudo-code when determining whether to
5216 allow the guest FixCtr[i] to count its pre-de 5185 allow the guest FixCtr[i] to count its pre-defined fixed event::
5217 5186
5218 FixCtr[i]_is_allowed = (action == ALLOW) && 5187 FixCtr[i]_is_allowed = (action == ALLOW) && (bitmap & BIT(i)) ||
5219 (action == DENY) && !(bitmap & BIT(i)); 5188 (action == DENY) && !(bitmap & BIT(i));
5220 FixCtr[i]_is_denied = !FixCtr[i]_is_allowed 5189 FixCtr[i]_is_denied = !FixCtr[i]_is_allowed;
5221 5190
5222 KVM always consumes fixed_counter_bitmap, it' 5191 KVM always consumes fixed_counter_bitmap, it's userspace's responsibility to
5223 ensure fixed_counter_bitmap is set correctly, 5192 ensure fixed_counter_bitmap is set correctly, e.g. if userspace wants to define
5224 a filter that only affects general purpose co 5193 a filter that only affects general purpose counters.
5225 5194
5226 Note, the "events" field also applies to fixe 5195 Note, the "events" field also applies to fixed counters' hardcoded event_select
5227 and unit_mask values. "fixed_counter_bitmap" 5196 and unit_mask values. "fixed_counter_bitmap" has higher priority than "events"
5228 if there is a contradiction between the two. 5197 if there is a contradiction between the two.
5229 5198
5230 4.121 KVM_PPC_SVM_OFF 5199 4.121 KVM_PPC_SVM_OFF
5231 --------------------- 5200 ---------------------
5232 5201
5233 :Capability: basic 5202 :Capability: basic
5234 :Architectures: powerpc 5203 :Architectures: powerpc
5235 :Type: vm ioctl 5204 :Type: vm ioctl
5236 :Parameters: none 5205 :Parameters: none
5237 :Returns: 0 on successful completion, 5206 :Returns: 0 on successful completion,
5238 5207
5239 Errors: 5208 Errors:
5240 5209
5241 ====== ================================ 5210 ====== ================================================================
5242 EINVAL if ultravisor failed to terminat 5211 EINVAL if ultravisor failed to terminate the secure guest
5243 ENOMEM if hypervisor failed to allocate 5212 ENOMEM if hypervisor failed to allocate new radix page tables for guest
5244 ====== ================================ 5213 ====== ================================================================
5245 5214
5246 This ioctl is used to turn off the secure mod 5215 This ioctl is used to turn off the secure mode of the guest or transition
5247 the guest from secure mode to normal mode. Th 5216 the guest from secure mode to normal mode. This is invoked when the guest
5248 is reset. This has no effect if called for a 5217 is reset. This has no effect if called for a normal guest.
5249 5218
5250 This ioctl issues an ultravisor call to termi 5219 This ioctl issues an ultravisor call to terminate the secure guest,
5251 unpins the VPA pages and releases all the dev 5220 unpins the VPA pages and releases all the device pages that are used to
5252 track the secure pages by hypervisor. 5221 track the secure pages by hypervisor.
5253 5222
5254 4.122 KVM_S390_NORMAL_RESET 5223 4.122 KVM_S390_NORMAL_RESET
5255 --------------------------- 5224 ---------------------------
5256 5225
5257 :Capability: KVM_CAP_S390_VCPU_RESETS 5226 :Capability: KVM_CAP_S390_VCPU_RESETS
5258 :Architectures: s390 5227 :Architectures: s390
5259 :Type: vcpu ioctl 5228 :Type: vcpu ioctl
5260 :Parameters: none 5229 :Parameters: none
5261 :Returns: 0 5230 :Returns: 0
5262 5231
5263 This ioctl resets VCPU registers and control 5232 This ioctl resets VCPU registers and control structures according to
5264 the cpu reset definition in the POP (Principl 5233 the cpu reset definition in the POP (Principles Of Operation).
5265 5234
5266 4.123 KVM_S390_INITIAL_RESET 5235 4.123 KVM_S390_INITIAL_RESET
5267 ---------------------------- 5236 ----------------------------
5268 5237
5269 :Capability: none 5238 :Capability: none
5270 :Architectures: s390 5239 :Architectures: s390
5271 :Type: vcpu ioctl 5240 :Type: vcpu ioctl
5272 :Parameters: none 5241 :Parameters: none
5273 :Returns: 0 5242 :Returns: 0
5274 5243
5275 This ioctl resets VCPU registers and control 5244 This ioctl resets VCPU registers and control structures according to
5276 the initial cpu reset definition in the POP. 5245 the initial cpu reset definition in the POP. However, the cpu is not
5277 put into ESA mode. This reset is a superset o 5246 put into ESA mode. This reset is a superset of the normal reset.
5278 5247
5279 4.124 KVM_S390_CLEAR_RESET 5248 4.124 KVM_S390_CLEAR_RESET
5280 -------------------------- 5249 --------------------------
5281 5250
5282 :Capability: KVM_CAP_S390_VCPU_RESETS 5251 :Capability: KVM_CAP_S390_VCPU_RESETS
5283 :Architectures: s390 5252 :Architectures: s390
5284 :Type: vcpu ioctl 5253 :Type: vcpu ioctl
5285 :Parameters: none 5254 :Parameters: none
5286 :Returns: 0 5255 :Returns: 0
5287 5256
5288 This ioctl resets VCPU registers and control 5257 This ioctl resets VCPU registers and control structures according to
5289 the clear cpu reset definition in the POP. Ho 5258 the clear cpu reset definition in the POP. However, the cpu is not put
5290 into ESA mode. This reset is a superset of th 5259 into ESA mode. This reset is a superset of the initial reset.
5291 5260
5292 5261
5293 4.125 KVM_S390_PV_COMMAND 5262 4.125 KVM_S390_PV_COMMAND
5294 ------------------------- 5263 -------------------------
5295 5264
5296 :Capability: KVM_CAP_S390_PROTECTED 5265 :Capability: KVM_CAP_S390_PROTECTED
5297 :Architectures: s390 5266 :Architectures: s390
5298 :Type: vm ioctl 5267 :Type: vm ioctl
5299 :Parameters: struct kvm_pv_cmd 5268 :Parameters: struct kvm_pv_cmd
5300 :Returns: 0 on success, < 0 on error 5269 :Returns: 0 on success, < 0 on error
5301 5270
5302 :: 5271 ::
5303 5272
5304 struct kvm_pv_cmd { 5273 struct kvm_pv_cmd {
5305 __u32 cmd; /* Command to be exec 5274 __u32 cmd; /* Command to be executed */
5306 __u16 rc; /* Ultravisor return 5275 __u16 rc; /* Ultravisor return code */
5307 __u16 rrc; /* Ultravisor return 5276 __u16 rrc; /* Ultravisor return reason code */
5308 __u64 data; /* Data or address */ 5277 __u64 data; /* Data or address */
5309 __u32 flags; /* flags for future e 5278 __u32 flags; /* flags for future extensions. Must be 0 for now */
5310 __u32 reserved[3]; 5279 __u32 reserved[3];
5311 }; 5280 };
5312 5281
5313 **Ultravisor return codes** 5282 **Ultravisor return codes**
5314 The Ultravisor return (reason) codes are prov 5283 The Ultravisor return (reason) codes are provided by the kernel if a
5315 Ultravisor call has been executed to achieve 5284 Ultravisor call has been executed to achieve the results expected by
5316 the command. Therefore they are independent o 5285 the command. Therefore they are independent of the IOCTL return
5317 code. If KVM changes `rc`, its value will alw 5286 code. If KVM changes `rc`, its value will always be greater than 0
5318 hence setting it to 0 before issuing a PV com 5287 hence setting it to 0 before issuing a PV command is advised to be
5319 able to detect a change of `rc`. 5288 able to detect a change of `rc`.
5320 5289
5321 **cmd values:** 5290 **cmd values:**
5322 5291
5323 KVM_PV_ENABLE 5292 KVM_PV_ENABLE
5324 Allocate memory and register the VM with th 5293 Allocate memory and register the VM with the Ultravisor, thereby
5325 donating memory to the Ultravisor that will 5294 donating memory to the Ultravisor that will become inaccessible to
5326 KVM. All existing CPUs are converted to pro 5295 KVM. All existing CPUs are converted to protected ones. After this
5327 command has succeeded, any CPU added via ho 5296 command has succeeded, any CPU added via hotplug will become
5328 protected during its creation as well. 5297 protected during its creation as well.
5329 5298
5330 Errors: 5299 Errors:
5331 5300
5332 ===== ============================= 5301 ===== =============================
5333 EINTR an unmasked signal is pending 5302 EINTR an unmasked signal is pending
5334 ===== ============================= 5303 ===== =============================
5335 5304
5336 KVM_PV_DISABLE 5305 KVM_PV_DISABLE
5337 Deregister the VM from the Ultravisor and r 5306 Deregister the VM from the Ultravisor and reclaim the memory that had
5338 been donated to the Ultravisor, making it u 5307 been donated to the Ultravisor, making it usable by the kernel again.
5339 All registered VCPUs are converted back to 5308 All registered VCPUs are converted back to non-protected ones. If a
5340 previous protected VM had been prepared for 5309 previous protected VM had been prepared for asynchronous teardown with
5341 KVM_PV_ASYNC_CLEANUP_PREPARE and not subseq 5310 KVM_PV_ASYNC_CLEANUP_PREPARE and not subsequently torn down with
5342 KVM_PV_ASYNC_CLEANUP_PERFORM, it will be to 5311 KVM_PV_ASYNC_CLEANUP_PERFORM, it will be torn down in this call
5343 together with the current protected VM. 5312 together with the current protected VM.
5344 5313
5345 KVM_PV_VM_SET_SEC_PARMS 5314 KVM_PV_VM_SET_SEC_PARMS
5346 Pass the image header from VM memory to the 5315 Pass the image header from VM memory to the Ultravisor in
5347 preparation of image unpacking and verifica 5316 preparation of image unpacking and verification.
5348 5317
5349 KVM_PV_VM_UNPACK 5318 KVM_PV_VM_UNPACK
5350 Unpack (protect and decrypt) a page of the 5319 Unpack (protect and decrypt) a page of the encrypted boot image.
5351 5320
5352 KVM_PV_VM_VERIFY 5321 KVM_PV_VM_VERIFY
5353 Verify the integrity of the unpacked image. 5322 Verify the integrity of the unpacked image. Only if this succeeds,
5354 KVM is allowed to start protected VCPUs. 5323 KVM is allowed to start protected VCPUs.
5355 5324
5356 KVM_PV_INFO 5325 KVM_PV_INFO
5357 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5326 :Capability: KVM_CAP_S390_PROTECTED_DUMP
5358 5327
5359 Presents an API that provides Ultravisor re 5328 Presents an API that provides Ultravisor related data to userspace
5360 via subcommands. len_max is the size of the 5329 via subcommands. len_max is the size of the user space buffer,
5361 len_written is KVM's indication of how much 5330 len_written is KVM's indication of how much bytes of that buffer
5362 were actually written to. len_written can b 5331 were actually written to. len_written can be used to determine the
5363 valid fields if more response fields are ad 5332 valid fields if more response fields are added in the future.
5364 5333
5365 :: 5334 ::
5366 5335
5367 enum pv_cmd_info_id { 5336 enum pv_cmd_info_id {
5368 KVM_PV_INFO_VM, 5337 KVM_PV_INFO_VM,
5369 KVM_PV_INFO_DUMP, 5338 KVM_PV_INFO_DUMP,
5370 }; 5339 };
5371 5340
5372 struct kvm_s390_pv_info_header { 5341 struct kvm_s390_pv_info_header {
5373 __u32 id; 5342 __u32 id;
5374 __u32 len_max; 5343 __u32 len_max;
5375 __u32 len_written; 5344 __u32 len_written;
5376 __u32 reserved; 5345 __u32 reserved;
5377 }; 5346 };
5378 5347
5379 struct kvm_s390_pv_info { 5348 struct kvm_s390_pv_info {
5380 struct kvm_s390_pv_info_header header 5349 struct kvm_s390_pv_info_header header;
5381 struct kvm_s390_pv_info_dump dump; 5350 struct kvm_s390_pv_info_dump dump;
5382 struct kvm_s390_pv_info_vm vm; 5351 struct kvm_s390_pv_info_vm vm;
5383 }; 5352 };
5384 5353
5385 **subcommands:** 5354 **subcommands:**
5386 5355
5387 KVM_PV_INFO_VM 5356 KVM_PV_INFO_VM
5388 This subcommand provides basic Ultravisor 5357 This subcommand provides basic Ultravisor information for PV
5389 hosts. These values are likely also expor 5358 hosts. These values are likely also exported as files in the sysfs
5390 firmware UV query interface but they are 5359 firmware UV query interface but they are more easily available to
5391 programs in this API. 5360 programs in this API.
5392 5361
5393 The installed calls and feature_indicatio 5362 The installed calls and feature_indication members provide the
5394 installed UV calls and the UV's other fea 5363 installed UV calls and the UV's other feature indications.
5395 5364
5396 The max_* members provide information abo 5365 The max_* members provide information about the maximum number of PV
5397 vcpus, PV guests and PV guest memory size 5366 vcpus, PV guests and PV guest memory size.
5398 5367
5399 :: 5368 ::
5400 5369
5401 struct kvm_s390_pv_info_vm { 5370 struct kvm_s390_pv_info_vm {
5402 __u64 inst_calls_list[4]; 5371 __u64 inst_calls_list[4];
5403 __u64 max_cpus; 5372 __u64 max_cpus;
5404 __u64 max_guests; 5373 __u64 max_guests;
5405 __u64 max_guest_addr; 5374 __u64 max_guest_addr;
5406 __u64 feature_indication; 5375 __u64 feature_indication;
5407 }; 5376 };
5408 5377
5409 5378
5410 KVM_PV_INFO_DUMP 5379 KVM_PV_INFO_DUMP
5411 This subcommand provides information rela 5380 This subcommand provides information related to dumping PV guests.
5412 5381
5413 :: 5382 ::
5414 5383
5415 struct kvm_s390_pv_info_dump { 5384 struct kvm_s390_pv_info_dump {
5416 __u64 dump_cpu_buffer_len; 5385 __u64 dump_cpu_buffer_len;
5417 __u64 dump_config_mem_buffer_per_1m; 5386 __u64 dump_config_mem_buffer_per_1m;
5418 __u64 dump_config_finalize_len; 5387 __u64 dump_config_finalize_len;
5419 }; 5388 };
5420 5389
5421 KVM_PV_DUMP 5390 KVM_PV_DUMP
5422 :Capability: KVM_CAP_S390_PROTECTED_DUMP 5391 :Capability: KVM_CAP_S390_PROTECTED_DUMP
5423 5392
5424 Presents an API that provides calls which f 5393 Presents an API that provides calls which facilitate dumping a
5425 protected VM. 5394 protected VM.
5426 5395
5427 :: 5396 ::
5428 5397
5429 struct kvm_s390_pv_dmp { 5398 struct kvm_s390_pv_dmp {
5430 __u64 subcmd; 5399 __u64 subcmd;
5431 __u64 buff_addr; 5400 __u64 buff_addr;
5432 __u64 buff_len; 5401 __u64 buff_len;
5433 __u64 gaddr; /* For dump s 5402 __u64 gaddr; /* For dump storage state */
5434 }; 5403 };
5435 5404
5436 **subcommands:** 5405 **subcommands:**
5437 5406
5438 KVM_PV_DUMP_INIT 5407 KVM_PV_DUMP_INIT
5439 Initializes the dump process of a protect 5408 Initializes the dump process of a protected VM. If this call does
5440 not succeed all other subcommands will fa 5409 not succeed all other subcommands will fail with -EINVAL. This
5441 subcommand will return -EINVAL if a dump 5410 subcommand will return -EINVAL if a dump process has not yet been
5442 completed. 5411 completed.
5443 5412
5444 Not all PV vms can be dumped, the owner n 5413 Not all PV vms can be dumped, the owner needs to set `dump
5445 allowed` PCF bit 34 in the SE header to a 5414 allowed` PCF bit 34 in the SE header to allow dumping.
5446 5415
5447 KVM_PV_DUMP_CONFIG_STOR_STATE 5416 KVM_PV_DUMP_CONFIG_STOR_STATE
5448 Stores `buff_len` bytes of tweak compone 5417 Stores `buff_len` bytes of tweak component values starting with
5449 the 1MB block specified by the absolute 5418 the 1MB block specified by the absolute guest address
5450 (`gaddr`). `buff_len` needs to be `conf_ 5419 (`gaddr`). `buff_len` needs to be `conf_dump_storage_state_len`
5451 aligned and at least >= the `conf_dump_s 5420 aligned and at least >= the `conf_dump_storage_state_len` value
5452 provided by the dump uv_info data. buff_ 5421 provided by the dump uv_info data. buff_user might be written to
5453 even if an error rc is returned. For ins 5422 even if an error rc is returned. For instance if we encounter a
5454 fault after writing the first page of da 5423 fault after writing the first page of data.
5455 5424
5456 KVM_PV_DUMP_COMPLETE 5425 KVM_PV_DUMP_COMPLETE
5457 If the subcommand succeeds it completes t 5426 If the subcommand succeeds it completes the dump process and lets
5458 KVM_PV_DUMP_INIT be called again. 5427 KVM_PV_DUMP_INIT be called again.
5459 5428
5460 On success `conf_dump_finalize_len` bytes 5429 On success `conf_dump_finalize_len` bytes of completion data will be
5461 stored to the `buff_addr`. The completion 5430 stored to the `buff_addr`. The completion data contains a key
5462 derivation seed, IV, tweak nonce and encr 5431 derivation seed, IV, tweak nonce and encryption keys as well as an
5463 authentication tag all of which are neede 5432 authentication tag all of which are needed to decrypt the dump at a
5464 later time. 5433 later time.
5465 5434
5466 KVM_PV_ASYNC_CLEANUP_PREPARE 5435 KVM_PV_ASYNC_CLEANUP_PREPARE
5467 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_D 5436 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE
5468 5437
5469 Prepare the current protected VM for asynch 5438 Prepare the current protected VM for asynchronous teardown. Most
5470 resources used by the current protected VM 5439 resources used by the current protected VM will be set aside for a
5471 subsequent asynchronous teardown. The curre 5440 subsequent asynchronous teardown. The current protected VM will then
5472 resume execution immediately as non-protect 5441 resume execution immediately as non-protected. There can be at most
5473 one protected VM prepared for asynchronous 5442 one protected VM prepared for asynchronous teardown at any time. If
5474 a protected VM had already been prepared fo 5443 a protected VM had already been prepared for teardown without
5475 subsequently calling KVM_PV_ASYNC_CLEANUP_P 5444 subsequently calling KVM_PV_ASYNC_CLEANUP_PERFORM, this call will
5476 fail. In that case, the userspace process s 5445 fail. In that case, the userspace process should issue a normal
5477 KVM_PV_DISABLE. The resources set aside wit 5446 KVM_PV_DISABLE. The resources set aside with this call will need to
5478 be cleaned up with a subsequent call to KVM 5447 be cleaned up with a subsequent call to KVM_PV_ASYNC_CLEANUP_PERFORM
5479 or KVM_PV_DISABLE, otherwise they will be c 5448 or KVM_PV_DISABLE, otherwise they will be cleaned up when KVM
5480 terminates. KVM_PV_ASYNC_CLEANUP_PREPARE ca 5449 terminates. KVM_PV_ASYNC_CLEANUP_PREPARE can be called again as soon
5481 as cleanup starts, i.e. before KVM_PV_ASYNC 5450 as cleanup starts, i.e. before KVM_PV_ASYNC_CLEANUP_PERFORM finishes.
5482 5451
5483 KVM_PV_ASYNC_CLEANUP_PERFORM 5452 KVM_PV_ASYNC_CLEANUP_PERFORM
5484 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_D 5453 :Capability: KVM_CAP_S390_PROTECTED_ASYNC_DISABLE
5485 5454
5486 Tear down the protected VM previously prepa 5455 Tear down the protected VM previously prepared for teardown with
5487 KVM_PV_ASYNC_CLEANUP_PREPARE. The resources 5456 KVM_PV_ASYNC_CLEANUP_PREPARE. The resources that had been set aside
5488 will be freed during the execution of this 5457 will be freed during the execution of this command. This PV command
5489 should ideally be issued by userspace from 5458 should ideally be issued by userspace from a separate thread. If a
5490 fatal signal is received (or the process te 5459 fatal signal is received (or the process terminates naturally), the
5491 command will terminate immediately without 5460 command will terminate immediately without completing, and the normal
5492 KVM shutdown procedure will take care of cl 5461 KVM shutdown procedure will take care of cleaning up all remaining
5493 protected VMs, including the ones whose tea 5462 protected VMs, including the ones whose teardown was interrupted by
5494 process termination. 5463 process termination.
5495 5464
5496 4.126 KVM_XEN_HVM_SET_ATTR 5465 4.126 KVM_XEN_HVM_SET_ATTR
5497 -------------------------- 5466 --------------------------
5498 5467
5499 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CO 5468 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO
5500 :Architectures: x86 5469 :Architectures: x86
5501 :Type: vm ioctl 5470 :Type: vm ioctl
5502 :Parameters: struct kvm_xen_hvm_attr 5471 :Parameters: struct kvm_xen_hvm_attr
5503 :Returns: 0 on success, < 0 on error 5472 :Returns: 0 on success, < 0 on error
5504 5473
5505 :: 5474 ::
5506 5475
5507 struct kvm_xen_hvm_attr { 5476 struct kvm_xen_hvm_attr {
5508 __u16 type; 5477 __u16 type;
5509 __u16 pad[3]; 5478 __u16 pad[3];
5510 union { 5479 union {
5511 __u8 long_mode; 5480 __u8 long_mode;
5512 __u8 vector; 5481 __u8 vector;
5513 __u8 runstate_update_flag; 5482 __u8 runstate_update_flag;
5514 union { !! 5483 struct {
5515 __u64 gfn; 5484 __u64 gfn;
5516 __u64 hva; <<
5517 } shared_info; 5485 } shared_info;
5518 struct { 5486 struct {
5519 __u32 send_port; 5487 __u32 send_port;
5520 __u32 type; /* EVTCHN 5488 __u32 type; /* EVTCHNSTAT_ipi / EVTCHNSTAT_interdomain */
5521 __u32 flags; 5489 __u32 flags;
5522 union { 5490 union {
5523 struct { 5491 struct {
5524 __u32 5492 __u32 port;
5525 __u32 5493 __u32 vcpu;
5526 __u32 5494 __u32 priority;
5527 } port; 5495 } port;
5528 struct { 5496 struct {
5529 __u32 5497 __u32 port; /* Zero for eventfd */
5530 __s32 5498 __s32 fd;
5531 } eventfd; 5499 } eventfd;
5532 __u32 padding 5500 __u32 padding[4];
5533 } deliver; 5501 } deliver;
5534 } evtchn; 5502 } evtchn;
5535 __u32 xen_version; 5503 __u32 xen_version;
5536 __u64 pad[8]; 5504 __u64 pad[8];
5537 } u; 5505 } u;
5538 }; 5506 };
5539 5507
5540 type values: 5508 type values:
5541 5509
5542 KVM_XEN_ATTR_TYPE_LONG_MODE 5510 KVM_XEN_ATTR_TYPE_LONG_MODE
5543 Sets the ABI mode of the VM to 32-bit or 64 5511 Sets the ABI mode of the VM to 32-bit or 64-bit (long mode). This
5544 determines the layout of the shared_info pa !! 5512 determines the layout of the shared info pages exposed to the VM.
5545 5513
5546 KVM_XEN_ATTR_TYPE_SHARED_INFO 5514 KVM_XEN_ATTR_TYPE_SHARED_INFO
5547 Sets the guest physical frame number at whi !! 5515 Sets the guest physical frame number at which the Xen "shared info"
5548 page resides. Note that although Xen places 5516 page resides. Note that although Xen places vcpu_info for the first
5549 32 vCPUs in the shared_info page, KVM does 5517 32 vCPUs in the shared_info page, KVM does not automatically do so
5550 and instead requires that KVM_XEN_VCPU_ATTR !! 5518 and instead requires that KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO be used
5551 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA be use !! 5519 explicitly even when the vcpu_info for a given vCPU resides at the
5552 the vcpu_info for a given vCPU resides at t !! 5520 "default" location in the shared_info page. This is because KVM may
5553 in the shared_info page. This is because KV !! 5521 not be aware of the Xen CPU id which is used as the index into the
5554 the Xen CPU id which is used as the index i !! 5522 vcpu_info[] array, so may know the correct default location.
5555 array, so may know the correct default loca <<
5556 5523
5557 Note that the shared_info page may be const !! 5524 Note that the shared info page may be constantly written to by KVM;
5558 it contains the event channel bitmap used t 5525 it contains the event channel bitmap used to deliver interrupts to
5559 a Xen guest, amongst other things. It is ex 5526 a Xen guest, amongst other things. It is exempt from dirty tracking
5560 mechanisms — KVM will not explicitly mark 5527 mechanisms — KVM will not explicitly mark the page as dirty each
5561 time an event channel interrupt is delivere 5528 time an event channel interrupt is delivered to the guest! Thus,
5562 userspace should always assume that the des 5529 userspace should always assume that the designated GFN is dirty if
5563 any vCPU has been running or any event chan 5530 any vCPU has been running or any event channel interrupts can be
5564 routed to the guest. 5531 routed to the guest.
5565 5532
5566 Setting the gfn to KVM_XEN_INVALID_GFN will !! 5533 Setting the gfn to KVM_XEN_INVALID_GFN will disable the shared info
5567 page. 5534 page.
5568 5535
5569 KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA <<
5570 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA f <<
5571 Xen capabilities, then this attribute may b <<
5572 userspace address at which the shared_info <<
5573 will always be fixed in the VMM regardless <<
5574 in guest physical address space. This attri <<
5575 preference to KVM_XEN_ATTR_TYPE_SHARED_INFO <<
5576 unnecessary invalidation of an internal cac <<
5577 re-mapped in guest physcial address space. <<
5578 <<
5579 Setting the hva to zero will disable the sh <<
5580 <<
5581 KVM_XEN_ATTR_TYPE_UPCALL_VECTOR 5536 KVM_XEN_ATTR_TYPE_UPCALL_VECTOR
5582 Sets the exception vector used to deliver X 5537 Sets the exception vector used to deliver Xen event channel upcalls.
5583 This is the HVM-wide vector injected direct 5538 This is the HVM-wide vector injected directly by the hypervisor
5584 (not through the local APIC), typically con 5539 (not through the local APIC), typically configured by a guest via
5585 HVM_PARAM_CALLBACK_IRQ. This can be disable 5540 HVM_PARAM_CALLBACK_IRQ. This can be disabled again (e.g. for guest
5586 SHUTDOWN_soft_reset) by setting it to zero. 5541 SHUTDOWN_soft_reset) by setting it to zero.
5587 5542
5588 KVM_XEN_ATTR_TYPE_EVTCHN 5543 KVM_XEN_ATTR_TYPE_EVTCHN
5589 This attribute is available when the KVM_CA 5544 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5590 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5545 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures
5591 an outbound port number for interception of 5546 an outbound port number for interception of EVTCHNOP_send requests
5592 from the guest. A given sending port number 5547 from the guest. A given sending port number may be directed back to
5593 a specified vCPU (by APIC ID) / port / prio 5548 a specified vCPU (by APIC ID) / port / priority on the guest, or to
5594 trigger events on an eventfd. The vCPU and 5549 trigger events on an eventfd. The vCPU and priority can be changed
5595 by setting KVM_XEN_EVTCHN_UPDATE in a subse 5550 by setting KVM_XEN_EVTCHN_UPDATE in a subsequent call, but other
5596 fields cannot change for a given sending po 5551 fields cannot change for a given sending port. A port mapping is
5597 removed by using KVM_XEN_EVTCHN_DEASSIGN in 5552 removed by using KVM_XEN_EVTCHN_DEASSIGN in the flags field. Passing
5598 KVM_XEN_EVTCHN_RESET in the flags field rem 5553 KVM_XEN_EVTCHN_RESET in the flags field removes all interception of
5599 outbound event channels. The values of the 5554 outbound event channels. The values of the flags field are mutually
5600 exclusive and cannot be combined as a bitma 5555 exclusive and cannot be combined as a bitmask.
5601 5556
5602 KVM_XEN_ATTR_TYPE_XEN_VERSION 5557 KVM_XEN_ATTR_TYPE_XEN_VERSION
5603 This attribute is available when the KVM_CA 5558 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5604 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5559 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It configures
5605 the 32-bit version code returned to the gue 5560 the 32-bit version code returned to the guest when it invokes the
5606 XENVER_version call; typically (XEN_MAJOR < 5561 XENVER_version call; typically (XEN_MAJOR << 16 | XEN_MINOR). PV
5607 Xen guests will often use this to as a dumm 5562 Xen guests will often use this to as a dummy hypercall to trigger
5608 event channel delivery, so responding withi 5563 event channel delivery, so responding within the kernel without
5609 exiting to userspace is beneficial. 5564 exiting to userspace is beneficial.
5610 5565
5611 KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG 5566 KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG
5612 This attribute is available when the KVM_CA 5567 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5613 support for KVM_XEN_HVM_CONFIG_RUNSTATE_UPD 5568 support for KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG. It enables the
5614 XEN_RUNSTATE_UPDATE flag which allows guest 5569 XEN_RUNSTATE_UPDATE flag which allows guest vCPUs to safely read
5615 other vCPUs' vcpu_runstate_info. Xen guests 5570 other vCPUs' vcpu_runstate_info. Xen guests enable this feature via
5616 the VMASST_TYPE_runstate_update_flag of the 5571 the VMASST_TYPE_runstate_update_flag of the HYPERVISOR_vm_assist
5617 hypercall. 5572 hypercall.
5618 5573
5619 4.127 KVM_XEN_HVM_GET_ATTR 5574 4.127 KVM_XEN_HVM_GET_ATTR
5620 -------------------------- 5575 --------------------------
5621 5576
5622 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CO 5577 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO
5623 :Architectures: x86 5578 :Architectures: x86
5624 :Type: vm ioctl 5579 :Type: vm ioctl
5625 :Parameters: struct kvm_xen_hvm_attr 5580 :Parameters: struct kvm_xen_hvm_attr
5626 :Returns: 0 on success, < 0 on error 5581 :Returns: 0 on success, < 0 on error
5627 5582
5628 Allows Xen VM attributes to be read. For the 5583 Allows Xen VM attributes to be read. For the structure and types,
5629 see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_A 5584 see KVM_XEN_HVM_SET_ATTR above. The KVM_XEN_ATTR_TYPE_EVTCHN
5630 attribute cannot be read. 5585 attribute cannot be read.
5631 5586
5632 4.128 KVM_XEN_VCPU_SET_ATTR 5587 4.128 KVM_XEN_VCPU_SET_ATTR
5633 --------------------------- 5588 ---------------------------
5634 5589
5635 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CO 5590 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO
5636 :Architectures: x86 5591 :Architectures: x86
5637 :Type: vcpu ioctl 5592 :Type: vcpu ioctl
5638 :Parameters: struct kvm_xen_vcpu_attr 5593 :Parameters: struct kvm_xen_vcpu_attr
5639 :Returns: 0 on success, < 0 on error 5594 :Returns: 0 on success, < 0 on error
5640 5595
5641 :: 5596 ::
5642 5597
5643 struct kvm_xen_vcpu_attr { 5598 struct kvm_xen_vcpu_attr {
5644 __u16 type; 5599 __u16 type;
5645 __u16 pad[3]; 5600 __u16 pad[3];
5646 union { 5601 union {
5647 __u64 gpa; 5602 __u64 gpa;
5648 __u64 pad[4]; 5603 __u64 pad[4];
5649 struct { 5604 struct {
5650 __u64 state; 5605 __u64 state;
5651 __u64 state_entry_tim 5606 __u64 state_entry_time;
5652 __u64 time_running; 5607 __u64 time_running;
5653 __u64 time_runnable; 5608 __u64 time_runnable;
5654 __u64 time_blocked; 5609 __u64 time_blocked;
5655 __u64 time_offline; 5610 __u64 time_offline;
5656 } runstate; 5611 } runstate;
5657 __u32 vcpu_id; 5612 __u32 vcpu_id;
5658 struct { 5613 struct {
5659 __u32 port; 5614 __u32 port;
5660 __u32 priority; 5615 __u32 priority;
5661 __u64 expires_ns; 5616 __u64 expires_ns;
5662 } timer; 5617 } timer;
5663 __u8 vector; 5618 __u8 vector;
5664 } u; 5619 } u;
5665 }; 5620 };
5666 5621
5667 type values: 5622 type values:
5668 5623
5669 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO 5624 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO
5670 Sets the guest physical address of the vcpu 5625 Sets the guest physical address of the vcpu_info for a given vCPU.
5671 As with the shared_info page for the VM, th 5626 As with the shared_info page for the VM, the corresponding page may be
5672 dirtied at any time if event channel interr 5627 dirtied at any time if event channel interrupt delivery is enabled, so
5673 userspace should always assume that the pag 5628 userspace should always assume that the page is dirty without relying
5674 on dirty logging. Setting the gpa to KVM_XE 5629 on dirty logging. Setting the gpa to KVM_XEN_INVALID_GPA will disable
5675 the vcpu_info. 5630 the vcpu_info.
5676 5631
5677 KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA <<
5678 If the KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA f <<
5679 Xen capabilities, then this attribute may b <<
5680 userspace address of the vcpu_info for a gi <<
5681 only be used when the vcpu_info resides at <<
5682 in the shared_info page. In this case it is <<
5683 userspace address will not change, because <<
5684 an overlay on guest memory and remains at a <<
5685 regardless of where it is mapped in guest p <<
5686 and hence unnecessary invalidation of an in <<
5687 avoided if the guest memory layout is modif <<
5688 If the vcpu_info does not reside at the "de <<
5689 it is not guaranteed to remain at the same <<
5690 hence the aforementioned cache invalidation <<
5691 <<
5692 KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO 5632 KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO
5693 Sets the guest physical address of an addit 5633 Sets the guest physical address of an additional pvclock structure
5694 for a given vCPU. This is typically used fo 5634 for a given vCPU. This is typically used for guest vsyscall support.
5695 Setting the gpa to KVM_XEN_INVALID_GPA will 5635 Setting the gpa to KVM_XEN_INVALID_GPA will disable the structure.
5696 5636
5697 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 5637 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR
5698 Sets the guest physical address of the vcpu 5638 Sets the guest physical address of the vcpu_runstate_info for a given
5699 vCPU. This is how a Xen guest tracks CPU st 5639 vCPU. This is how a Xen guest tracks CPU state such as steal time.
5700 Setting the gpa to KVM_XEN_INVALID_GPA will 5640 Setting the gpa to KVM_XEN_INVALID_GPA will disable the runstate area.
5701 5641
5702 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT 5642 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT
5703 Sets the runstate (RUNSTATE_running/_runnab 5643 Sets the runstate (RUNSTATE_running/_runnable/_blocked/_offline) of
5704 the given vCPU from the .u.runstate.state m 5644 the given vCPU from the .u.runstate.state member of the structure.
5705 KVM automatically accounts running and runn 5645 KVM automatically accounts running and runnable time but blocked
5706 and offline states are only entered explici 5646 and offline states are only entered explicitly.
5707 5647
5708 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA 5648 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA
5709 Sets all fields of the vCPU runstate data f 5649 Sets all fields of the vCPU runstate data from the .u.runstate member
5710 of the structure, including the current run 5650 of the structure, including the current runstate. The state_entry_time
5711 must equal the sum of the other four times. 5651 must equal the sum of the other four times.
5712 5652
5713 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST 5653 KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST
5714 This *adds* the contents of the .u.runstate 5654 This *adds* the contents of the .u.runstate members of the structure
5715 to the corresponding members of the given v 5655 to the corresponding members of the given vCPU's runstate data, thus
5716 permitting atomic adjustments to the runsta 5656 permitting atomic adjustments to the runstate times. The adjustment
5717 to the state_entry_time must equal the sum 5657 to the state_entry_time must equal the sum of the adjustments to the
5718 other four times. The state field must be s 5658 other four times. The state field must be set to -1, or to a valid
5719 runstate value (RUNSTATE_running, RUNSTATE_ 5659 runstate value (RUNSTATE_running, RUNSTATE_runnable, RUNSTATE_blocked
5720 or RUNSTATE_offline) to set the current acc 5660 or RUNSTATE_offline) to set the current accounted state as of the
5721 adjusted state_entry_time. 5661 adjusted state_entry_time.
5722 5662
5723 KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID 5663 KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID
5724 This attribute is available when the KVM_CA 5664 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5725 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5665 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the Xen
5726 vCPU ID of the given vCPU, to allow timer-r 5666 vCPU ID of the given vCPU, to allow timer-related VCPU operations to
5727 be intercepted by KVM. 5667 be intercepted by KVM.
5728 5668
5729 KVM_XEN_VCPU_ATTR_TYPE_TIMER 5669 KVM_XEN_VCPU_ATTR_TYPE_TIMER
5730 This attribute is available when the KVM_CA 5670 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5731 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5671 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the
5732 event channel port/priority for the VIRQ_TI 5672 event channel port/priority for the VIRQ_TIMER of the vCPU, as well
5733 as allowing a pending timer to be saved/res 5673 as allowing a pending timer to be saved/restored. Setting the timer
5734 port to zero disables kernel handling of th 5674 port to zero disables kernel handling of the singleshot timer.
5735 5675
5736 KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR 5676 KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR
5737 This attribute is available when the KVM_CA 5677 This attribute is available when the KVM_CAP_XEN_HVM ioctl indicates
5738 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND 5678 support for KVM_XEN_HVM_CONFIG_EVTCHN_SEND features. It sets the
5739 per-vCPU local APIC upcall vector, configur 5679 per-vCPU local APIC upcall vector, configured by a Xen guest with
5740 the HVMOP_set_evtchn_upcall_vector hypercal 5680 the HVMOP_set_evtchn_upcall_vector hypercall. This is typically
5741 used by Windows guests, and is distinct fro 5681 used by Windows guests, and is distinct from the HVM-wide upcall
5742 vector configured with HVM_PARAM_CALLBACK_I 5682 vector configured with HVM_PARAM_CALLBACK_IRQ. It is disabled by
5743 setting the vector to zero. 5683 setting the vector to zero.
5744 5684
5745 5685
5746 4.129 KVM_XEN_VCPU_GET_ATTR 5686 4.129 KVM_XEN_VCPU_GET_ATTR
5747 --------------------------- 5687 ---------------------------
5748 5688
5749 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CO 5689 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_SHARED_INFO
5750 :Architectures: x86 5690 :Architectures: x86
5751 :Type: vcpu ioctl 5691 :Type: vcpu ioctl
5752 :Parameters: struct kvm_xen_vcpu_attr 5692 :Parameters: struct kvm_xen_vcpu_attr
5753 :Returns: 0 on success, < 0 on error 5693 :Returns: 0 on success, < 0 on error
5754 5694
5755 Allows Xen vCPU attributes to be read. For th 5695 Allows Xen vCPU attributes to be read. For the structure and types,
5756 see KVM_XEN_VCPU_SET_ATTR above. 5696 see KVM_XEN_VCPU_SET_ATTR above.
5757 5697
5758 The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST ty 5698 The KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST type may not be used
5759 with the KVM_XEN_VCPU_GET_ATTR ioctl. 5699 with the KVM_XEN_VCPU_GET_ATTR ioctl.
5760 5700
5761 4.130 KVM_ARM_MTE_COPY_TAGS 5701 4.130 KVM_ARM_MTE_COPY_TAGS
5762 --------------------------- 5702 ---------------------------
5763 5703
5764 :Capability: KVM_CAP_ARM_MTE 5704 :Capability: KVM_CAP_ARM_MTE
5765 :Architectures: arm64 5705 :Architectures: arm64
5766 :Type: vm ioctl 5706 :Type: vm ioctl
5767 :Parameters: struct kvm_arm_copy_mte_tags 5707 :Parameters: struct kvm_arm_copy_mte_tags
5768 :Returns: number of bytes copied, < 0 on erro 5708 :Returns: number of bytes copied, < 0 on error (-EINVAL for incorrect
5769 arguments, -EFAULT if memory cannot 5709 arguments, -EFAULT if memory cannot be accessed).
5770 5710
5771 :: 5711 ::
5772 5712
5773 struct kvm_arm_copy_mte_tags { 5713 struct kvm_arm_copy_mte_tags {
5774 __u64 guest_ipa; 5714 __u64 guest_ipa;
5775 __u64 length; 5715 __u64 length;
5776 void __user *addr; 5716 void __user *addr;
5777 __u64 flags; 5717 __u64 flags;
5778 __u64 reserved[2]; 5718 __u64 reserved[2];
5779 }; 5719 };
5780 5720
5781 Copies Memory Tagging Extension (MTE) tags to 5721 Copies Memory Tagging Extension (MTE) tags to/from guest tag memory. The
5782 ``guest_ipa`` and ``length`` fields must be ` 5722 ``guest_ipa`` and ``length`` fields must be ``PAGE_SIZE`` aligned.
5783 ``length`` must not be bigger than 2^31 - PAG 5723 ``length`` must not be bigger than 2^31 - PAGE_SIZE bytes. The ``addr``
5784 field must point to a buffer which the tags w 5724 field must point to a buffer which the tags will be copied to or from.
5785 5725
5786 ``flags`` specifies the direction of copy, ei 5726 ``flags`` specifies the direction of copy, either ``KVM_ARM_TAGS_TO_GUEST`` or
5787 ``KVM_ARM_TAGS_FROM_GUEST``. 5727 ``KVM_ARM_TAGS_FROM_GUEST``.
5788 5728
5789 The size of the buffer to store the tags is ` 5729 The size of the buffer to store the tags is ``(length / 16)`` bytes
5790 (granules in MTE are 16 bytes long). Each byt 5730 (granules in MTE are 16 bytes long). Each byte contains a single tag
5791 value. This matches the format of ``PTRACE_PE 5731 value. This matches the format of ``PTRACE_PEEKMTETAGS`` and
5792 ``PTRACE_POKEMTETAGS``. 5732 ``PTRACE_POKEMTETAGS``.
5793 5733
5794 If an error occurs before any data is copied 5734 If an error occurs before any data is copied then a negative error code is
5795 returned. If some tags have been copied befor 5735 returned. If some tags have been copied before an error occurs then the number
5796 of bytes successfully copied is returned. If 5736 of bytes successfully copied is returned. If the call completes successfully
5797 then ``length`` is returned. 5737 then ``length`` is returned.
5798 5738
5799 4.131 KVM_GET_SREGS2 5739 4.131 KVM_GET_SREGS2
5800 -------------------- 5740 --------------------
5801 5741
5802 :Capability: KVM_CAP_SREGS2 5742 :Capability: KVM_CAP_SREGS2
5803 :Architectures: x86 5743 :Architectures: x86
5804 :Type: vcpu ioctl 5744 :Type: vcpu ioctl
5805 :Parameters: struct kvm_sregs2 (out) 5745 :Parameters: struct kvm_sregs2 (out)
5806 :Returns: 0 on success, -1 on error 5746 :Returns: 0 on success, -1 on error
5807 5747
5808 Reads special registers from the vcpu. 5748 Reads special registers from the vcpu.
5809 This ioctl (when supported) replaces the KVM_ 5749 This ioctl (when supported) replaces the KVM_GET_SREGS.
5810 5750
5811 :: 5751 ::
5812 5752
5813 struct kvm_sregs2 { 5753 struct kvm_sregs2 {
5814 /* out (KVM_GET_SREGS2) / in 5754 /* out (KVM_GET_SREGS2) / in (KVM_SET_SREGS2) */
5815 struct kvm_segment cs, ds, es 5755 struct kvm_segment cs, ds, es, fs, gs, ss;
5816 struct kvm_segment tr, ldt; 5756 struct kvm_segment tr, ldt;
5817 struct kvm_dtable gdt, idt; 5757 struct kvm_dtable gdt, idt;
5818 __u64 cr0, cr2, cr3, cr4, cr8 5758 __u64 cr0, cr2, cr3, cr4, cr8;
5819 __u64 efer; 5759 __u64 efer;
5820 __u64 apic_base; 5760 __u64 apic_base;
5821 __u64 flags; 5761 __u64 flags;
5822 __u64 pdptrs[4]; 5762 __u64 pdptrs[4];
5823 }; 5763 };
5824 5764
5825 flags values for ``kvm_sregs2``: 5765 flags values for ``kvm_sregs2``:
5826 5766
5827 ``KVM_SREGS2_FLAGS_PDPTRS_VALID`` 5767 ``KVM_SREGS2_FLAGS_PDPTRS_VALID``
5828 5768
5829 Indicates that the struct contains valid PD 5769 Indicates that the struct contains valid PDPTR values.
5830 5770
5831 5771
5832 4.132 KVM_SET_SREGS2 5772 4.132 KVM_SET_SREGS2
5833 -------------------- 5773 --------------------
5834 5774
5835 :Capability: KVM_CAP_SREGS2 5775 :Capability: KVM_CAP_SREGS2
5836 :Architectures: x86 5776 :Architectures: x86
5837 :Type: vcpu ioctl 5777 :Type: vcpu ioctl
5838 :Parameters: struct kvm_sregs2 (in) 5778 :Parameters: struct kvm_sregs2 (in)
5839 :Returns: 0 on success, -1 on error 5779 :Returns: 0 on success, -1 on error
5840 5780
5841 Writes special registers into the vcpu. 5781 Writes special registers into the vcpu.
5842 See KVM_GET_SREGS2 for the data structures. 5782 See KVM_GET_SREGS2 for the data structures.
5843 This ioctl (when supported) replaces the KVM_ 5783 This ioctl (when supported) replaces the KVM_SET_SREGS.
5844 5784
5845 4.133 KVM_GET_STATS_FD 5785 4.133 KVM_GET_STATS_FD
5846 ---------------------- 5786 ----------------------
5847 5787
5848 :Capability: KVM_CAP_STATS_BINARY_FD 5788 :Capability: KVM_CAP_STATS_BINARY_FD
5849 :Architectures: all 5789 :Architectures: all
5850 :Type: vm ioctl, vcpu ioctl 5790 :Type: vm ioctl, vcpu ioctl
5851 :Parameters: none 5791 :Parameters: none
5852 :Returns: statistics file descriptor on succe 5792 :Returns: statistics file descriptor on success, < 0 on error
5853 5793
5854 Errors: 5794 Errors:
5855 5795
5856 ====== ================================ 5796 ====== ======================================================
5857 ENOMEM if the fd could not be created d 5797 ENOMEM if the fd could not be created due to lack of memory
5858 EMFILE if the number of opened files ex 5798 EMFILE if the number of opened files exceeds the limit
5859 ====== ================================ 5799 ====== ======================================================
5860 5800
5861 The returned file descriptor can be used to r 5801 The returned file descriptor can be used to read VM/vCPU statistics data in
5862 binary format. The data in the file descripto 5802 binary format. The data in the file descriptor consists of four blocks
5863 organized as follows: 5803 organized as follows:
5864 5804
5865 +-------------+ 5805 +-------------+
5866 | Header | 5806 | Header |
5867 +-------------+ 5807 +-------------+
5868 | id string | 5808 | id string |
5869 +-------------+ 5809 +-------------+
5870 | Descriptors | 5810 | Descriptors |
5871 +-------------+ 5811 +-------------+
5872 | Stats Data | 5812 | Stats Data |
5873 +-------------+ 5813 +-------------+
5874 5814
5875 Apart from the header starting at offset 0, p 5815 Apart from the header starting at offset 0, please be aware that it is
5876 not guaranteed that the four blocks are adjac 5816 not guaranteed that the four blocks are adjacent or in the above order;
5877 the offsets of the id, descriptors and data b 5817 the offsets of the id, descriptors and data blocks are found in the
5878 header. However, all four blocks are aligned 5818 header. However, all four blocks are aligned to 64 bit offsets in the
5879 file and they do not overlap. 5819 file and they do not overlap.
5880 5820
5881 All blocks except the data block are immutabl 5821 All blocks except the data block are immutable. Userspace can read them
5882 only one time after retrieving the file descr 5822 only one time after retrieving the file descriptor, and then use ``pread`` or
5883 ``lseek`` to read the statistics repeatedly. 5823 ``lseek`` to read the statistics repeatedly.
5884 5824
5885 All data is in system endianness. 5825 All data is in system endianness.
5886 5826
5887 The format of the header is as follows:: 5827 The format of the header is as follows::
5888 5828
5889 struct kvm_stats_header { 5829 struct kvm_stats_header {
5890 __u32 flags; 5830 __u32 flags;
5891 __u32 name_size; 5831 __u32 name_size;
5892 __u32 num_desc; 5832 __u32 num_desc;
5893 __u32 id_offset; 5833 __u32 id_offset;
5894 __u32 desc_offset; 5834 __u32 desc_offset;
5895 __u32 data_offset; 5835 __u32 data_offset;
5896 }; 5836 };
5897 5837
5898 The ``flags`` field is not used at the moment 5838 The ``flags`` field is not used at the moment. It is always read as 0.
5899 5839
5900 The ``name_size`` field is the size (in byte) 5840 The ``name_size`` field is the size (in byte) of the statistics name string
5901 (including trailing '\0') which is contained 5841 (including trailing '\0') which is contained in the "id string" block and
5902 appended at the end of every descriptor. 5842 appended at the end of every descriptor.
5903 5843
5904 The ``num_desc`` field is the number of descr 5844 The ``num_desc`` field is the number of descriptors that are included in the
5905 descriptor block. (The actual number of valu 5845 descriptor block. (The actual number of values in the data block may be
5906 larger, since each descriptor may comprise mo 5846 larger, since each descriptor may comprise more than one value).
5907 5847
5908 The ``id_offset`` field is the offset of the 5848 The ``id_offset`` field is the offset of the id string from the start of the
5909 file indicated by the file descriptor. It is 5849 file indicated by the file descriptor. It is a multiple of 8.
5910 5850
5911 The ``desc_offset`` field is the offset of th 5851 The ``desc_offset`` field is the offset of the Descriptors block from the start
5912 of the file indicated by the file descriptor. 5852 of the file indicated by the file descriptor. It is a multiple of 8.
5913 5853
5914 The ``data_offset`` field is the offset of th 5854 The ``data_offset`` field is the offset of the Stats Data block from the start
5915 of the file indicated by the file descriptor. 5855 of the file indicated by the file descriptor. It is a multiple of 8.
5916 5856
5917 The id string block contains a string which i 5857 The id string block contains a string which identifies the file descriptor on
5918 which KVM_GET_STATS_FD was invoked. The size 5858 which KVM_GET_STATS_FD was invoked. The size of the block, including the
5919 trailing ``'\0'``, is indicated by the ``name 5859 trailing ``'\0'``, is indicated by the ``name_size`` field in the header.
5920 5860
5921 The descriptors block is only needed to be re 5861 The descriptors block is only needed to be read once for the lifetime of the
5922 file descriptor contains a sequence of ``stru 5862 file descriptor contains a sequence of ``struct kvm_stats_desc``, each followed
5923 by a string of size ``name_size``. 5863 by a string of size ``name_size``.
5924 :: 5864 ::
5925 5865
5926 #define KVM_STATS_TYPE_SHIFT 5866 #define KVM_STATS_TYPE_SHIFT 0
5927 #define KVM_STATS_TYPE_MASK 5867 #define KVM_STATS_TYPE_MASK (0xF << KVM_STATS_TYPE_SHIFT)
5928 #define KVM_STATS_TYPE_CUMULATIVE 5868 #define KVM_STATS_TYPE_CUMULATIVE (0x0 << KVM_STATS_TYPE_SHIFT)
5929 #define KVM_STATS_TYPE_INSTANT 5869 #define KVM_STATS_TYPE_INSTANT (0x1 << KVM_STATS_TYPE_SHIFT)
5930 #define KVM_STATS_TYPE_PEAK 5870 #define KVM_STATS_TYPE_PEAK (0x2 << KVM_STATS_TYPE_SHIFT)
5931 #define KVM_STATS_TYPE_LINEAR_HIST 5871 #define KVM_STATS_TYPE_LINEAR_HIST (0x3 << KVM_STATS_TYPE_SHIFT)
5932 #define KVM_STATS_TYPE_LOG_HIST 5872 #define KVM_STATS_TYPE_LOG_HIST (0x4 << KVM_STATS_TYPE_SHIFT)
5933 #define KVM_STATS_TYPE_MAX 5873 #define KVM_STATS_TYPE_MAX KVM_STATS_TYPE_LOG_HIST
5934 5874
5935 #define KVM_STATS_UNIT_SHIFT 5875 #define KVM_STATS_UNIT_SHIFT 4
5936 #define KVM_STATS_UNIT_MASK 5876 #define KVM_STATS_UNIT_MASK (0xF << KVM_STATS_UNIT_SHIFT)
5937 #define KVM_STATS_UNIT_NONE 5877 #define KVM_STATS_UNIT_NONE (0x0 << KVM_STATS_UNIT_SHIFT)
5938 #define KVM_STATS_UNIT_BYTES 5878 #define KVM_STATS_UNIT_BYTES (0x1 << KVM_STATS_UNIT_SHIFT)
5939 #define KVM_STATS_UNIT_SECONDS 5879 #define KVM_STATS_UNIT_SECONDS (0x2 << KVM_STATS_UNIT_SHIFT)
5940 #define KVM_STATS_UNIT_CYCLES 5880 #define KVM_STATS_UNIT_CYCLES (0x3 << KVM_STATS_UNIT_SHIFT)
5941 #define KVM_STATS_UNIT_BOOLEAN 5881 #define KVM_STATS_UNIT_BOOLEAN (0x4 << KVM_STATS_UNIT_SHIFT)
5942 #define KVM_STATS_UNIT_MAX 5882 #define KVM_STATS_UNIT_MAX KVM_STATS_UNIT_BOOLEAN
5943 5883
5944 #define KVM_STATS_BASE_SHIFT 5884 #define KVM_STATS_BASE_SHIFT 8
5945 #define KVM_STATS_BASE_MASK 5885 #define KVM_STATS_BASE_MASK (0xF << KVM_STATS_BASE_SHIFT)
5946 #define KVM_STATS_BASE_POW10 5886 #define KVM_STATS_BASE_POW10 (0x0 << KVM_STATS_BASE_SHIFT)
5947 #define KVM_STATS_BASE_POW2 5887 #define KVM_STATS_BASE_POW2 (0x1 << KVM_STATS_BASE_SHIFT)
5948 #define KVM_STATS_BASE_MAX 5888 #define KVM_STATS_BASE_MAX KVM_STATS_BASE_POW2
5949 5889
5950 struct kvm_stats_desc { 5890 struct kvm_stats_desc {
5951 __u32 flags; 5891 __u32 flags;
5952 __s16 exponent; 5892 __s16 exponent;
5953 __u16 size; 5893 __u16 size;
5954 __u32 offset; 5894 __u32 offset;
5955 __u32 bucket_size; 5895 __u32 bucket_size;
5956 char name[]; 5896 char name[];
5957 }; 5897 };
5958 5898
5959 The ``flags`` field contains the type and uni 5899 The ``flags`` field contains the type and unit of the statistics data described
5960 by this descriptor. Its endianness is CPU nat 5900 by this descriptor. Its endianness is CPU native.
5961 The following flags are supported: 5901 The following flags are supported:
5962 5902
5963 Bits 0-3 of ``flags`` encode the type: 5903 Bits 0-3 of ``flags`` encode the type:
5964 5904
5965 * ``KVM_STATS_TYPE_CUMULATIVE`` 5905 * ``KVM_STATS_TYPE_CUMULATIVE``
5966 The statistics reports a cumulative count 5906 The statistics reports a cumulative count. The value of data can only be increased.
5967 Most of the counters used in KVM are of t 5907 Most of the counters used in KVM are of this type.
5968 The corresponding ``size`` field for this 5908 The corresponding ``size`` field for this type is always 1.
5969 All cumulative statistics data are read/w 5909 All cumulative statistics data are read/write.
5970 * ``KVM_STATS_TYPE_INSTANT`` 5910 * ``KVM_STATS_TYPE_INSTANT``
5971 The statistics reports an instantaneous v 5911 The statistics reports an instantaneous value. Its value can be increased or
5972 decreased. This type is usually used as a 5912 decreased. This type is usually used as a measurement of some resources,
5973 like the number of dirty pages, the numbe 5913 like the number of dirty pages, the number of large pages, etc.
5974 All instant statistics are read only. 5914 All instant statistics are read only.
5975 The corresponding ``size`` field for this 5915 The corresponding ``size`` field for this type is always 1.
5976 * ``KVM_STATS_TYPE_PEAK`` 5916 * ``KVM_STATS_TYPE_PEAK``
5977 The statistics data reports a peak value, 5917 The statistics data reports a peak value, for example the maximum number
5978 of items in a hash table bucket, the long 5918 of items in a hash table bucket, the longest time waited and so on.
5979 The value of data can only be increased. 5919 The value of data can only be increased.
5980 The corresponding ``size`` field for this 5920 The corresponding ``size`` field for this type is always 1.
5981 * ``KVM_STATS_TYPE_LINEAR_HIST`` 5921 * ``KVM_STATS_TYPE_LINEAR_HIST``
5982 The statistic is reported as a linear his 5922 The statistic is reported as a linear histogram. The number of
5983 buckets is specified by the ``size`` fiel 5923 buckets is specified by the ``size`` field. The size of buckets is specified
5984 by the ``hist_param`` field. The range of 5924 by the ``hist_param`` field. The range of the Nth bucket (1 <= N < ``size``)
5985 is [``hist_param``*(N-1), ``hist_param``* 5925 is [``hist_param``*(N-1), ``hist_param``*N), while the range of the last
5986 bucket is [``hist_param``*(``size``-1), + 5926 bucket is [``hist_param``*(``size``-1), +INF). (+INF means positive infinity
5987 value.) 5927 value.)
5988 * ``KVM_STATS_TYPE_LOG_HIST`` 5928 * ``KVM_STATS_TYPE_LOG_HIST``
5989 The statistic is reported as a logarithmi 5929 The statistic is reported as a logarithmic histogram. The number of
5990 buckets is specified by the ``size`` fiel 5930 buckets is specified by the ``size`` field. The range of the first bucket is
5991 [0, 1), while the range of the last bucke 5931 [0, 1), while the range of the last bucket is [pow(2, ``size``-2), +INF).
5992 Otherwise, The Nth bucket (1 < N < ``size 5932 Otherwise, The Nth bucket (1 < N < ``size``) covers
5993 [pow(2, N-2), pow(2, N-1)). 5933 [pow(2, N-2), pow(2, N-1)).
5994 5934
5995 Bits 4-7 of ``flags`` encode the unit: 5935 Bits 4-7 of ``flags`` encode the unit:
5996 5936
5997 * ``KVM_STATS_UNIT_NONE`` 5937 * ``KVM_STATS_UNIT_NONE``
5998 There is no unit for the value of statist 5938 There is no unit for the value of statistics data. This usually means that
5999 the value is a simple counter of an event 5939 the value is a simple counter of an event.
6000 * ``KVM_STATS_UNIT_BYTES`` 5940 * ``KVM_STATS_UNIT_BYTES``
6001 It indicates that the statistics data is 5941 It indicates that the statistics data is used to measure memory size, in the
6002 unit of Byte, KiByte, MiByte, GiByte, etc 5942 unit of Byte, KiByte, MiByte, GiByte, etc. The unit of the data is
6003 determined by the ``exponent`` field in t 5943 determined by the ``exponent`` field in the descriptor.
6004 * ``KVM_STATS_UNIT_SECONDS`` 5944 * ``KVM_STATS_UNIT_SECONDS``
6005 It indicates that the statistics data is 5945 It indicates that the statistics data is used to measure time or latency.
6006 * ``KVM_STATS_UNIT_CYCLES`` 5946 * ``KVM_STATS_UNIT_CYCLES``
6007 It indicates that the statistics data is 5947 It indicates that the statistics data is used to measure CPU clock cycles.
6008 * ``KVM_STATS_UNIT_BOOLEAN`` 5948 * ``KVM_STATS_UNIT_BOOLEAN``
6009 It indicates that the statistic will alwa 5949 It indicates that the statistic will always be either 0 or 1. Boolean
6010 statistics of "peak" type will never go b 5950 statistics of "peak" type will never go back from 1 to 0. Boolean
6011 statistics can be linear histograms (with 5951 statistics can be linear histograms (with two buckets) but not logarithmic
6012 histograms. 5952 histograms.
6013 5953
6014 Note that, in the case of histograms, the uni 5954 Note that, in the case of histograms, the unit applies to the bucket
6015 ranges, while the bucket value indicates how 5955 ranges, while the bucket value indicates how many samples fell in the
6016 bucket's range. 5956 bucket's range.
6017 5957
6018 Bits 8-11 of ``flags``, together with ``expon 5958 Bits 8-11 of ``flags``, together with ``exponent``, encode the scale of the
6019 unit: 5959 unit:
6020 5960
6021 * ``KVM_STATS_BASE_POW10`` 5961 * ``KVM_STATS_BASE_POW10``
6022 The scale is based on power of 10. It is 5962 The scale is based on power of 10. It is used for measurement of time and
6023 CPU clock cycles. For example, an expone 5963 CPU clock cycles. For example, an exponent of -9 can be used with
6024 ``KVM_STATS_UNIT_SECONDS`` to express tha 5964 ``KVM_STATS_UNIT_SECONDS`` to express that the unit is nanoseconds.
6025 * ``KVM_STATS_BASE_POW2`` 5965 * ``KVM_STATS_BASE_POW2``
6026 The scale is based on power of 2. It is u 5966 The scale is based on power of 2. It is used for measurement of memory size.
6027 For example, an exponent of 20 can be use 5967 For example, an exponent of 20 can be used with ``KVM_STATS_UNIT_BYTES`` to
6028 express that the unit is MiB. 5968 express that the unit is MiB.
6029 5969
6030 The ``size`` field is the number of values of 5970 The ``size`` field is the number of values of this statistics data. Its
6031 value is usually 1 for most of simple statist 5971 value is usually 1 for most of simple statistics. 1 means it contains an
6032 unsigned 64bit data. 5972 unsigned 64bit data.
6033 5973
6034 The ``offset`` field is the offset from the s 5974 The ``offset`` field is the offset from the start of Data Block to the start of
6035 the corresponding statistics data. 5975 the corresponding statistics data.
6036 5976
6037 The ``bucket_size`` field is used as a parame 5977 The ``bucket_size`` field is used as a parameter for histogram statistics data.
6038 It is only used by linear histogram statistic 5978 It is only used by linear histogram statistics data, specifying the size of a
6039 bucket in the unit expressed by bits 4-11 of 5979 bucket in the unit expressed by bits 4-11 of ``flags`` together with ``exponent``.
6040 5980
6041 The ``name`` field is the name string of the 5981 The ``name`` field is the name string of the statistics data. The name string
6042 starts at the end of ``struct kvm_stats_desc` 5982 starts at the end of ``struct kvm_stats_desc``. The maximum length including
6043 the trailing ``'\0'``, is indicated by ``name 5983 the trailing ``'\0'``, is indicated by ``name_size`` in the header.
6044 5984
6045 The Stats Data block contains an array of 64- 5985 The Stats Data block contains an array of 64-bit values in the same order
6046 as the descriptors in Descriptors block. 5986 as the descriptors in Descriptors block.
6047 5987
6048 4.134 KVM_GET_XSAVE2 5988 4.134 KVM_GET_XSAVE2
6049 -------------------- 5989 --------------------
6050 5990
6051 :Capability: KVM_CAP_XSAVE2 5991 :Capability: KVM_CAP_XSAVE2
6052 :Architectures: x86 5992 :Architectures: x86
6053 :Type: vcpu ioctl 5993 :Type: vcpu ioctl
6054 :Parameters: struct kvm_xsave (out) 5994 :Parameters: struct kvm_xsave (out)
6055 :Returns: 0 on success, -1 on error 5995 :Returns: 0 on success, -1 on error
6056 5996
6057 5997
6058 :: 5998 ::
6059 5999
6060 struct kvm_xsave { 6000 struct kvm_xsave {
6061 __u32 region[1024]; 6001 __u32 region[1024];
6062 __u32 extra[0]; 6002 __u32 extra[0];
6063 }; 6003 };
6064 6004
6065 This ioctl would copy current vcpu's xsave st 6005 This ioctl would copy current vcpu's xsave struct to the userspace. It
6066 copies as many bytes as are returned by KVM_C 6006 copies as many bytes as are returned by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2)
6067 when invoked on the vm file descriptor. The s 6007 when invoked on the vm file descriptor. The size value returned by
6068 KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will alwa 6008 KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2) will always be at least 4096.
6069 Currently, it is only greater than 4096 if a 6009 Currently, it is only greater than 4096 if a dynamic feature has been
6070 enabled with ``arch_prctl()``, but this may c 6010 enabled with ``arch_prctl()``, but this may change in the future.
6071 6011
6072 The offsets of the state save areas in struct 6012 The offsets of the state save areas in struct kvm_xsave follow the contents
6073 of CPUID leaf 0xD on the host. 6013 of CPUID leaf 0xD on the host.
6074 6014
6075 4.135 KVM_XEN_HVM_EVTCHN_SEND 6015 4.135 KVM_XEN_HVM_EVTCHN_SEND
6076 ----------------------------- 6016 -----------------------------
6077 6017
6078 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CO 6018 :Capability: KVM_CAP_XEN_HVM / KVM_XEN_HVM_CONFIG_EVTCHN_SEND
6079 :Architectures: x86 6019 :Architectures: x86
6080 :Type: vm ioctl 6020 :Type: vm ioctl
6081 :Parameters: struct kvm_irq_routing_xen_evtch 6021 :Parameters: struct kvm_irq_routing_xen_evtchn
6082 :Returns: 0 on success, < 0 on error 6022 :Returns: 0 on success, < 0 on error
6083 6023
6084 6024
6085 :: 6025 ::
6086 6026
6087 struct kvm_irq_routing_xen_evtchn { 6027 struct kvm_irq_routing_xen_evtchn {
6088 __u32 port; 6028 __u32 port;
6089 __u32 vcpu; 6029 __u32 vcpu;
6090 __u32 priority; 6030 __u32 priority;
6091 }; 6031 };
6092 6032
6093 This ioctl injects an event channel interrupt 6033 This ioctl injects an event channel interrupt directly to the guest vCPU.
6094 6034
6095 4.136 KVM_S390_PV_CPU_COMMAND 6035 4.136 KVM_S390_PV_CPU_COMMAND
6096 ----------------------------- 6036 -----------------------------
6097 6037
6098 :Capability: KVM_CAP_S390_PROTECTED_DUMP 6038 :Capability: KVM_CAP_S390_PROTECTED_DUMP
6099 :Architectures: s390 6039 :Architectures: s390
6100 :Type: vcpu ioctl 6040 :Type: vcpu ioctl
6101 :Parameters: none 6041 :Parameters: none
6102 :Returns: 0 on success, < 0 on error 6042 :Returns: 0 on success, < 0 on error
6103 6043
6104 This ioctl closely mirrors `KVM_S390_PV_COMMA 6044 This ioctl closely mirrors `KVM_S390_PV_COMMAND` but handles requests
6105 for vcpus. It re-uses the kvm_s390_pv_dmp str 6045 for vcpus. It re-uses the kvm_s390_pv_dmp struct and hence also shares
6106 the command ids. 6046 the command ids.
6107 6047
6108 **command:** 6048 **command:**
6109 6049
6110 KVM_PV_DUMP 6050 KVM_PV_DUMP
6111 Presents an API that provides calls which f 6051 Presents an API that provides calls which facilitate dumping a vcpu
6112 of a protected VM. 6052 of a protected VM.
6113 6053
6114 **subcommand:** 6054 **subcommand:**
6115 6055
6116 KVM_PV_DUMP_CPU 6056 KVM_PV_DUMP_CPU
6117 Provides encrypted dump data like register 6057 Provides encrypted dump data like register values.
6118 The length of the returned data is provided 6058 The length of the returned data is provided by uv_info.guest_cpu_stor_len.
6119 6059
6120 4.137 KVM_S390_ZPCI_OP 6060 4.137 KVM_S390_ZPCI_OP
6121 ---------------------- 6061 ----------------------
6122 6062
6123 :Capability: KVM_CAP_S390_ZPCI_OP 6063 :Capability: KVM_CAP_S390_ZPCI_OP
6124 :Architectures: s390 6064 :Architectures: s390
6125 :Type: vm ioctl 6065 :Type: vm ioctl
6126 :Parameters: struct kvm_s390_zpci_op (in) 6066 :Parameters: struct kvm_s390_zpci_op (in)
6127 :Returns: 0 on success, <0 on error 6067 :Returns: 0 on success, <0 on error
6128 6068
6129 Used to manage hardware-assisted virtualizati 6069 Used to manage hardware-assisted virtualization features for zPCI devices.
6130 6070
6131 Parameters are specified via the following st 6071 Parameters are specified via the following structure::
6132 6072
6133 struct kvm_s390_zpci_op { 6073 struct kvm_s390_zpci_op {
6134 /* in */ 6074 /* in */
6135 __u32 fh; /* target dev 6075 __u32 fh; /* target device */
6136 __u8 op; /* operation 6076 __u8 op; /* operation to perform */
6137 __u8 pad[3]; 6077 __u8 pad[3];
6138 union { 6078 union {
6139 /* for KVM_S390_ZPCIOP_REG_AE 6079 /* for KVM_S390_ZPCIOP_REG_AEN */
6140 struct { 6080 struct {
6141 __u64 ibv; /* Gu 6081 __u64 ibv; /* Guest addr of interrupt bit vector */
6142 __u64 sb; /* Gu 6082 __u64 sb; /* Guest addr of summary bit */
6143 __u32 flags; 6083 __u32 flags;
6144 __u32 noi; /* Nu 6084 __u32 noi; /* Number of interrupts */
6145 __u8 isc; /* Gu 6085 __u8 isc; /* Guest interrupt subclass */
6146 __u8 sbo; /* Of 6086 __u8 sbo; /* Offset of guest summary bit vector */
6147 __u16 pad; 6087 __u16 pad;
6148 } reg_aen; 6088 } reg_aen;
6149 __u64 reserved[8]; 6089 __u64 reserved[8];
6150 } u; 6090 } u;
6151 }; 6091 };
6152 6092
6153 The type of operation is specified in the "op 6093 The type of operation is specified in the "op" field.
6154 KVM_S390_ZPCIOP_REG_AEN is used to register t 6094 KVM_S390_ZPCIOP_REG_AEN is used to register the VM for adapter event
6155 notification interpretation, which will allow 6095 notification interpretation, which will allow firmware delivery of adapter
6156 events directly to the vm, with KVM providing 6096 events directly to the vm, with KVM providing a backup delivery mechanism;
6157 KVM_S390_ZPCIOP_DEREG_AEN is used to subseque 6097 KVM_S390_ZPCIOP_DEREG_AEN is used to subsequently disable interpretation of
6158 adapter event notifications. 6098 adapter event notifications.
6159 6099
6160 The target zPCI function must also be specifi 6100 The target zPCI function must also be specified via the "fh" field. For the
6161 KVM_S390_ZPCIOP_REG_AEN operation, additional 6101 KVM_S390_ZPCIOP_REG_AEN operation, additional information to establish firmware
6162 delivery must be provided via the "reg_aen" s 6102 delivery must be provided via the "reg_aen" struct.
6163 6103
6164 The "pad" and "reserved" fields may be used f 6104 The "pad" and "reserved" fields may be used for future extensions and should be
6165 set to 0s by userspace. 6105 set to 0s by userspace.
6166 6106
6167 4.138 KVM_ARM_SET_COUNTER_OFFSET 6107 4.138 KVM_ARM_SET_COUNTER_OFFSET
6168 -------------------------------- 6108 --------------------------------
6169 6109
6170 :Capability: KVM_CAP_COUNTER_OFFSET 6110 :Capability: KVM_CAP_COUNTER_OFFSET
6171 :Architectures: arm64 6111 :Architectures: arm64
6172 :Type: vm ioctl 6112 :Type: vm ioctl
6173 :Parameters: struct kvm_arm_counter_offset (i 6113 :Parameters: struct kvm_arm_counter_offset (in)
6174 :Returns: 0 on success, < 0 on error 6114 :Returns: 0 on success, < 0 on error
6175 6115
6176 This capability indicates that userspace is a 6116 This capability indicates that userspace is able to apply a single VM-wide
6177 offset to both the virtual and physical count 6117 offset to both the virtual and physical counters as viewed by the guest
6178 using the KVM_ARM_SET_CNT_OFFSET ioctl and th 6118 using the KVM_ARM_SET_CNT_OFFSET ioctl and the following data structure:
6179 6119
6180 :: 6120 ::
6181 6121
6182 struct kvm_arm_counter_offset { 6122 struct kvm_arm_counter_offset {
6183 __u64 counter_offset; 6123 __u64 counter_offset;
6184 __u64 reserved; 6124 __u64 reserved;
6185 }; 6125 };
6186 6126
6187 The offset describes a number of counter cycl 6127 The offset describes a number of counter cycles that are subtracted from
6188 both virtual and physical counter views (simi 6128 both virtual and physical counter views (similar to the effects of the
6189 CNTVOFF_EL2 and CNTPOFF_EL2 system registers, 6129 CNTVOFF_EL2 and CNTPOFF_EL2 system registers, but only global). The offset
6190 always applies to all vcpus (already created 6130 always applies to all vcpus (already created or created after this ioctl)
6191 for this VM. 6131 for this VM.
6192 6132
6193 It is userspace's responsibility to compute t 6133 It is userspace's responsibility to compute the offset based, for example,
6194 on previous values of the guest counters. 6134 on previous values of the guest counters.
6195 6135
6196 Any value other than 0 for the "reserved" fie 6136 Any value other than 0 for the "reserved" field may result in an error
6197 (-EINVAL) being returned. This ioctl can also 6137 (-EINVAL) being returned. This ioctl can also return -EBUSY if any vcpu
6198 ioctl is issued concurrently. 6138 ioctl is issued concurrently.
6199 6139
6200 Note that using this ioctl results in KVM ign 6140 Note that using this ioctl results in KVM ignoring subsequent userspace
6201 writes to the CNTVCT_EL0 and CNTPCT_EL0 regis 6141 writes to the CNTVCT_EL0 and CNTPCT_EL0 registers using the SET_ONE_REG
6202 interface. No error will be returned, but the 6142 interface. No error will be returned, but the resulting offset will not be
6203 applied. 6143 applied.
6204 6144
6205 .. _KVM_ARM_GET_REG_WRITABLE_MASKS: 6145 .. _KVM_ARM_GET_REG_WRITABLE_MASKS:
6206 6146
6207 4.139 KVM_ARM_GET_REG_WRITABLE_MASKS 6147 4.139 KVM_ARM_GET_REG_WRITABLE_MASKS
6208 ------------------------------------------- 6148 -------------------------------------------
6209 6149
6210 :Capability: KVM_CAP_ARM_SUPPORTED_REG_MASK_R 6150 :Capability: KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES
6211 :Architectures: arm64 6151 :Architectures: arm64
6212 :Type: vm ioctl 6152 :Type: vm ioctl
6213 :Parameters: struct reg_mask_range (in/out) 6153 :Parameters: struct reg_mask_range (in/out)
6214 :Returns: 0 on success, < 0 on error 6154 :Returns: 0 on success, < 0 on error
6215 6155
6216 6156
6217 :: 6157 ::
6218 6158
6219 #define KVM_ARM_FEATURE_ID_RANGE 6159 #define KVM_ARM_FEATURE_ID_RANGE 0
6220 #define KVM_ARM_FEATURE_ID_RANGE_SIZE 6160 #define KVM_ARM_FEATURE_ID_RANGE_SIZE (3 * 8 * 8)
6221 6161
6222 struct reg_mask_range { 6162 struct reg_mask_range {
6223 __u64 addr; /* Po 6163 __u64 addr; /* Pointer to mask array */
6224 __u32 range; /* Re 6164 __u32 range; /* Requested range */
6225 __u32 reserved[13]; 6165 __u32 reserved[13];
6226 }; 6166 };
6227 6167
6228 This ioctl copies the writable masks for a se 6168 This ioctl copies the writable masks for a selected range of registers to
6229 userspace. 6169 userspace.
6230 6170
6231 The ``addr`` field is a pointer to the destin 6171 The ``addr`` field is a pointer to the destination array where KVM copies
6232 the writable masks. 6172 the writable masks.
6233 6173
6234 The ``range`` field indicates the requested r 6174 The ``range`` field indicates the requested range of registers.
6235 ``KVM_CHECK_EXTENSION`` for the ``KVM_CAP_ARM 6175 ``KVM_CHECK_EXTENSION`` for the ``KVM_CAP_ARM_SUPPORTED_REG_MASK_RANGES``
6236 capability returns the supported ranges, expr 6176 capability returns the supported ranges, expressed as a set of flags. Each
6237 flag's bit index represents a possible value 6177 flag's bit index represents a possible value for the ``range`` field.
6238 All other values are reserved for future use 6178 All other values are reserved for future use and KVM may return an error.
6239 6179
6240 The ``reserved[13]`` array is reserved for fu 6180 The ``reserved[13]`` array is reserved for future use and should be 0, or
6241 KVM may return an error. 6181 KVM may return an error.
6242 6182
6243 KVM_ARM_FEATURE_ID_RANGE (0) 6183 KVM_ARM_FEATURE_ID_RANGE (0)
6244 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 6184 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6245 6185
6246 The Feature ID range is defined as the AArch6 6186 The Feature ID range is defined as the AArch64 System register space with
6247 op0==3, op1=={0, 1, 3}, CRn==0, CRm=={0-7}, o 6187 op0==3, op1=={0, 1, 3}, CRn==0, CRm=={0-7}, op2=={0-7}.
6248 6188
6249 The mask returned array pointed to by ``addr` 6189 The mask returned array pointed to by ``addr`` is indexed by the macro
6250 ``ARM64_FEATURE_ID_RANGE_IDX(op0, op1, crn, c 6190 ``ARM64_FEATURE_ID_RANGE_IDX(op0, op1, crn, crm, op2)``, allowing userspace
6251 to know what fields can be changed for the sy 6191 to know what fields can be changed for the system register described by
6252 ``op0, op1, crn, crm, op2``. KVM rejects ID r 6192 ``op0, op1, crn, crm, op2``. KVM rejects ID register values that describe a
6253 superset of the features supported by the sys 6193 superset of the features supported by the system.
6254 6194
6255 4.140 KVM_SET_USER_MEMORY_REGION2 <<
6256 --------------------------------- <<
6257 <<
6258 :Capability: KVM_CAP_USER_MEMORY2 <<
6259 :Architectures: all <<
6260 :Type: vm ioctl <<
6261 :Parameters: struct kvm_userspace_memory_regi <<
6262 :Returns: 0 on success, -1 on error <<
6263 <<
6264 KVM_SET_USER_MEMORY_REGION2 is an extension t <<
6265 allows mapping guest_memfd memory into a gues <<
6266 KVM_SET_USER_MEMORY_REGION identically. User <<
6267 in flags to have KVM bind the memory region t <<
6268 [guest_memfd_offset, guest_memfd_offset + mem <<
6269 must point at a file created via KVM_CREATE_G <<
6270 the target range must not be bound to any oth <<
6271 bounds checks apply (use common sense). <<
6272 <<
6273 :: <<
6274 <<
6275 struct kvm_userspace_memory_region2 { <<
6276 __u32 slot; <<
6277 __u32 flags; <<
6278 __u64 guest_phys_addr; <<
6279 __u64 memory_size; /* bytes */ <<
6280 __u64 userspace_addr; /* start of the <<
6281 __u64 guest_memfd_offset; <<
6282 __u32 guest_memfd; <<
6283 __u32 pad1; <<
6284 __u64 pad2[14]; <<
6285 }; <<
6286 <<
6287 A KVM_MEM_GUEST_MEMFD region _must_ have a va <<
6288 userspace_addr (shared memory). However, "va <<
6289 means that the address itself must be a legal <<
6290 mapping for userspace_addr is not required to <<
6291 KVM_SET_USER_MEMORY_REGION2, e.g. shared memo <<
6292 on-demand. <<
6293 <<
6294 When mapping a gfn into the guest, KVM select <<
6295 userspace_addr vs. guest_memfd, based on the <<
6296 state. At VM creation time, all memory is sh <<
6297 is '0' for all gfns. Userspace can control w <<
6298 toggling KVM_MEMORY_ATTRIBUTE_PRIVATE via KVM <<
6299 <<
6300 S390: <<
6301 ^^^^^ <<
6302 <<
6303 Returns -EINVAL if the VM has the KVM_VM_S390 <<
6304 Returns -EINVAL if called on a protected VM. <<
6305 <<
6306 4.141 KVM_SET_MEMORY_ATTRIBUTES <<
6307 ------------------------------- <<
6308 <<
6309 :Capability: KVM_CAP_MEMORY_ATTRIBUTES <<
6310 :Architectures: x86 <<
6311 :Type: vm ioctl <<
6312 :Parameters: struct kvm_memory_attributes (in <<
6313 :Returns: 0 on success, <0 on error <<
6314 <<
6315 KVM_SET_MEMORY_ATTRIBUTES allows userspace to <<
6316 of guest physical memory. <<
6317 <<
6318 :: <<
6319 <<
6320 struct kvm_memory_attributes { <<
6321 __u64 address; <<
6322 __u64 size; <<
6323 __u64 attributes; <<
6324 __u64 flags; <<
6325 }; <<
6326 <<
6327 #define KVM_MEMORY_ATTRIBUTE_PRIVATE <<
6328 <<
6329 The address and size must be page aligned. T <<
6330 retrieved via ioctl(KVM_CHECK_EXTENSION) on K <<
6331 executed on a VM, KVM_CAP_MEMORY_ATTRIBUTES p <<
6332 supported by that VM. If executed at system <<
6333 returns all attributes supported by KVM. The <<
6334 time is KVM_MEMORY_ATTRIBUTE_PRIVATE, which m <<
6335 guest private memory. <<
6336 <<
6337 Note, there is no "get" API. Userspace is re <<
6338 the state of a gfn/page as needed. <<
6339 <<
6340 The "flags" field is reserved for future exte <<
6341 <<
6342 4.142 KVM_CREATE_GUEST_MEMFD <<
6343 ---------------------------- <<
6344 <<
6345 :Capability: KVM_CAP_GUEST_MEMFD <<
6346 :Architectures: none <<
6347 :Type: vm ioctl <<
6348 :Parameters: struct kvm_create_guest_memfd(in <<
6349 :Returns: A file descriptor on success, <0 on <<
6350 <<
6351 KVM_CREATE_GUEST_MEMFD creates an anonymous f <<
6352 that refers to it. guest_memfd files are rou <<
6353 via memfd_create(), e.g. guest_memfd files li <<
6354 and are automatically released when the last <<
6355 "regular" memfd_create() files, guest_memfd f <<
6356 virtual machine (see below), cannot be mapped <<
6357 and cannot be resized (guest_memfd files do <<
6358 <<
6359 :: <<
6360 <<
6361 struct kvm_create_guest_memfd { <<
6362 __u64 size; <<
6363 __u64 flags; <<
6364 __u64 reserved[6]; <<
6365 }; <<
6366 <<
6367 Conceptually, the inode backing a guest_memfd <<
6368 i.e. is coupled to the virtual machine as a t <<
6369 file itself, which is bound to a "struct kvm" <<
6370 underlying memory, e.g. effectively provides <<
6371 to host memory. This allows for use cases wh <<
6372 used to manage a single virtual machine, e.g. <<
6373 migration of a virtual machine. <<
6374 <<
6375 KVM currently only supports mapping guest_mem <<
6376 and more specifically via the guest_memfd and <<
6377 "struct kvm_userspace_memory_region2", where <<
6378 into the guest_memfd instance. For a given g <<
6379 most one mapping per page, i.e. binding multi <<
6380 guest_memfd range is not allowed (any number <<
6381 a single guest_memfd file, but the bound rang <<
6382 <<
6383 See KVM_SET_USER_MEMORY_REGION2 for additiona <<
6384 <<
6385 4.143 KVM_PRE_FAULT_MEMORY <<
6386 --------------------------- <<
6387 <<
6388 :Capability: KVM_CAP_PRE_FAULT_MEMORY <<
6389 :Architectures: none <<
6390 :Type: vcpu ioctl <<
6391 :Parameters: struct kvm_pre_fault_memory (in/ <<
6392 :Returns: 0 if at least one page is processed <<
6393 <<
6394 Errors: <<
6395 <<
6396 ========== ================================ <<
6397 EINVAL The specified `gpa` and `size` w <<
6398 page aligned, causes an overflow <<
6399 ENOENT The specified `gpa` is outside d <<
6400 EINTR An unmasked signal is pending an <<
6401 EFAULT The parameter address was invali <<
6402 EOPNOTSUPP Mapping memory for a GPA is unsu <<
6403 hypervisor, and/or for the curre <<
6404 EIO unexpected error conditions (als <<
6405 ========== ================================ <<
6406 <<
6407 :: <<
6408 <<
6409 struct kvm_pre_fault_memory { <<
6410 /* in/out */ <<
6411 __u64 gpa; <<
6412 __u64 size; <<
6413 /* in */ <<
6414 __u64 flags; <<
6415 __u64 padding[5]; <<
6416 }; <<
6417 <<
6418 KVM_PRE_FAULT_MEMORY populates KVM's stage-2 <<
6419 for the current vCPU state. KVM maps memory <<
6420 stage-2 read page fault, e.g. faults in memor <<
6421 CoW. However, KVM does not mark any newly cr <<
6422 <<
6423 In the case of confidential VM types where th <<
6424 private guest memory before the guest is 'fin <<
6425 should only be issued after completing all th <<
6426 guest into a 'finalized' state so that the ab <<
6427 ensured. <<
6428 <<
6429 In some cases, multiple vCPUs might share the <<
6430 case, the ioctl can be called in parallel. <<
6431 <<
6432 When the ioctl returns, the input values are <<
6433 remaining range. If `size` > 0 on return, th <<
6434 the ioctl again with the same `struct kvm_map <<
6435 <<
6436 Shadow page tables cannot support this ioctl <<
6437 are indexed by virtual address or nested gues <<
6438 Calling this ioctl when the guest is using sh <<
6439 example because it is running a nested guest <<
6440 will fail with `EOPNOTSUPP` even if `KVM_CHEC <<
6441 the capability to be present. <<
6442 <<
6443 `flags` must currently be zero. <<
6444 <<
6445 <<
6446 5. The kvm_run structure 6195 5. The kvm_run structure
6447 ======================== 6196 ========================
6448 6197
6449 Application code obtains a pointer to the kvm 6198 Application code obtains a pointer to the kvm_run structure by
6450 mmap()ing a vcpu fd. From that point, applic 6199 mmap()ing a vcpu fd. From that point, application code can control
6451 execution by changing fields in kvm_run prior 6200 execution by changing fields in kvm_run prior to calling the KVM_RUN
6452 ioctl, and obtain information about the reaso 6201 ioctl, and obtain information about the reason KVM_RUN returned by
6453 looking up structure members. 6202 looking up structure members.
6454 6203
6455 :: 6204 ::
6456 6205
6457 struct kvm_run { 6206 struct kvm_run {
6458 /* in */ 6207 /* in */
6459 __u8 request_interrupt_window; 6208 __u8 request_interrupt_window;
6460 6209
6461 Request that KVM_RUN return when it becomes p 6210 Request that KVM_RUN return when it becomes possible to inject external
6462 interrupts into the guest. Useful in conjunc 6211 interrupts into the guest. Useful in conjunction with KVM_INTERRUPT.
6463 6212
6464 :: 6213 ::
6465 6214
6466 __u8 immediate_exit; 6215 __u8 immediate_exit;
6467 6216
6468 This field is polled once when KVM_RUN starts 6217 This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN
6469 exits immediately, returning -EINTR. In the 6218 exits immediately, returning -EINTR. In the common scenario where a
6470 signal is used to "kick" a VCPU out of KVM_RU 6219 signal is used to "kick" a VCPU out of KVM_RUN, this field can be used
6471 to avoid usage of KVM_SET_SIGNAL_MASK, which 6220 to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability.
6472 Rather than blocking the signal outside KVM_R 6221 Rather than blocking the signal outside KVM_RUN, userspace can set up
6473 a signal handler that sets run->immediate_exi 6222 a signal handler that sets run->immediate_exit to a non-zero value.
6474 6223
6475 This field is ignored if KVM_CAP_IMMEDIATE_EX 6224 This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available.
6476 6225
6477 :: 6226 ::
6478 6227
6479 __u8 padding1[6]; 6228 __u8 padding1[6];
6480 6229
6481 /* out */ 6230 /* out */
6482 __u32 exit_reason; 6231 __u32 exit_reason;
6483 6232
6484 When KVM_RUN has returned successfully (retur 6233 When KVM_RUN has returned successfully (return value 0), this informs
6485 application code why KVM_RUN has returned. A 6234 application code why KVM_RUN has returned. Allowable values for this
6486 field are detailed below. 6235 field are detailed below.
6487 6236
6488 :: 6237 ::
6489 6238
6490 __u8 ready_for_interrupt_injection; 6239 __u8 ready_for_interrupt_injection;
6491 6240
6492 If request_interrupt_window has been specifie 6241 If request_interrupt_window has been specified, this field indicates
6493 an interrupt can be injected now with KVM_INT 6242 an interrupt can be injected now with KVM_INTERRUPT.
6494 6243
6495 :: 6244 ::
6496 6245
6497 __u8 if_flag; 6246 __u8 if_flag;
6498 6247
6499 The value of the current interrupt flag. Onl 6248 The value of the current interrupt flag. Only valid if in-kernel
6500 local APIC is not used. 6249 local APIC is not used.
6501 6250
6502 :: 6251 ::
6503 6252
6504 __u16 flags; 6253 __u16 flags;
6505 6254
6506 More architecture-specific flags detailing st 6255 More architecture-specific flags detailing state of the VCPU that may
6507 affect the device's behavior. Current defined 6256 affect the device's behavior. Current defined flags::
6508 6257
6509 /* x86, set if the VCPU is in system manage 6258 /* x86, set if the VCPU is in system management mode */
6510 #define KVM_RUN_X86_SMM (1 << 0) !! 6259 #define KVM_RUN_X86_SMM (1 << 0)
6511 /* x86, set if bus lock detected in VM */ 6260 /* x86, set if bus lock detected in VM */
6512 #define KVM_RUN_X86_BUS_LOCK (1 << 1) !! 6261 #define KVM_RUN_BUS_LOCK (1 << 1)
6513 /* x86, set if the VCPU is executing a nest <<
6514 #define KVM_RUN_X86_GUEST_MODE (1 << 2) <<
6515 <<
6516 /* arm64, set for KVM_EXIT_DEBUG */ 6262 /* arm64, set for KVM_EXIT_DEBUG */
6517 #define KVM_DEBUG_ARCH_HSR_HIGH_VALID (1 < 6263 #define KVM_DEBUG_ARCH_HSR_HIGH_VALID (1 << 0)
6518 6264
6519 :: 6265 ::
6520 6266
6521 /* in (pre_kvm_run), out (post_kvm_ru 6267 /* in (pre_kvm_run), out (post_kvm_run) */
6522 __u64 cr8; 6268 __u64 cr8;
6523 6269
6524 The value of the cr8 register. Only valid if 6270 The value of the cr8 register. Only valid if in-kernel local APIC is
6525 not used. Both input and output. 6271 not used. Both input and output.
6526 6272
6527 :: 6273 ::
6528 6274
6529 __u64 apic_base; 6275 __u64 apic_base;
6530 6276
6531 The value of the APIC BASE msr. Only valid i 6277 The value of the APIC BASE msr. Only valid if in-kernel local
6532 APIC is not used. Both input and output. 6278 APIC is not used. Both input and output.
6533 6279
6534 :: 6280 ::
6535 6281
6536 union { 6282 union {
6537 /* KVM_EXIT_UNKNOWN */ 6283 /* KVM_EXIT_UNKNOWN */
6538 struct { 6284 struct {
6539 __u64 hardware_exit_r 6285 __u64 hardware_exit_reason;
6540 } hw; 6286 } hw;
6541 6287
6542 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu 6288 If exit_reason is KVM_EXIT_UNKNOWN, the vcpu has exited due to unknown
6543 reasons. Further architecture-specific infor 6289 reasons. Further architecture-specific information is available in
6544 hardware_exit_reason. 6290 hardware_exit_reason.
6545 6291
6546 :: 6292 ::
6547 6293
6548 /* KVM_EXIT_FAIL_ENTRY */ 6294 /* KVM_EXIT_FAIL_ENTRY */
6549 struct { 6295 struct {
6550 __u64 hardware_entry_ 6296 __u64 hardware_entry_failure_reason;
6551 __u32 cpu; /* if KVM_ 6297 __u32 cpu; /* if KVM_LAST_CPU */
6552 } fail_entry; 6298 } fail_entry;
6553 6299
6554 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vc 6300 If exit_reason is KVM_EXIT_FAIL_ENTRY, the vcpu could not be run due
6555 to unknown reasons. Further architecture-spe 6301 to unknown reasons. Further architecture-specific information is
6556 available in hardware_entry_failure_reason. 6302 available in hardware_entry_failure_reason.
6557 6303
6558 :: 6304 ::
6559 6305
6560 /* KVM_EXIT_EXCEPTION */ 6306 /* KVM_EXIT_EXCEPTION */
6561 struct { 6307 struct {
6562 __u32 exception; 6308 __u32 exception;
6563 __u32 error_code; 6309 __u32 error_code;
6564 } ex; 6310 } ex;
6565 6311
6566 Unused. 6312 Unused.
6567 6313
6568 :: 6314 ::
6569 6315
6570 /* KVM_EXIT_IO */ 6316 /* KVM_EXIT_IO */
6571 struct { 6317 struct {
6572 #define KVM_EXIT_IO_IN 0 6318 #define KVM_EXIT_IO_IN 0
6573 #define KVM_EXIT_IO_OUT 1 6319 #define KVM_EXIT_IO_OUT 1
6574 __u8 direction; 6320 __u8 direction;
6575 __u8 size; /* bytes * 6321 __u8 size; /* bytes */
6576 __u16 port; 6322 __u16 port;
6577 __u32 count; 6323 __u32 count;
6578 __u64 data_offset; /* 6324 __u64 data_offset; /* relative to kvm_run start */
6579 } io; 6325 } io;
6580 6326
6581 If exit_reason is KVM_EXIT_IO, then the vcpu 6327 If exit_reason is KVM_EXIT_IO, then the vcpu has
6582 executed a port I/O instruction which could n 6328 executed a port I/O instruction which could not be satisfied by kvm.
6583 data_offset describes where the data is locat 6329 data_offset describes where the data is located (KVM_EXIT_IO_OUT) or
6584 where kvm expects application code to place t 6330 where kvm expects application code to place the data for the next
6585 KVM_RUN invocation (KVM_EXIT_IO_IN). Data fo 6331 KVM_RUN invocation (KVM_EXIT_IO_IN). Data format is a packed array.
6586 6332
6587 :: 6333 ::
6588 6334
6589 /* KVM_EXIT_DEBUG */ 6335 /* KVM_EXIT_DEBUG */
6590 struct { 6336 struct {
6591 struct kvm_debug_exit 6337 struct kvm_debug_exit_arch arch;
6592 } debug; 6338 } debug;
6593 6339
6594 If the exit_reason is KVM_EXIT_DEBUG, then a 6340 If the exit_reason is KVM_EXIT_DEBUG, then a vcpu is processing a debug event
6595 for which architecture specific information i 6341 for which architecture specific information is returned.
6596 6342
6597 :: 6343 ::
6598 6344
6599 /* KVM_EXIT_MMIO */ 6345 /* KVM_EXIT_MMIO */
6600 struct { 6346 struct {
6601 __u64 phys_addr; 6347 __u64 phys_addr;
6602 __u8 data[8]; 6348 __u8 data[8];
6603 __u32 len; 6349 __u32 len;
6604 __u8 is_write; 6350 __u8 is_write;
6605 } mmio; 6351 } mmio;
6606 6352
6607 If exit_reason is KVM_EXIT_MMIO, then the vcp 6353 If exit_reason is KVM_EXIT_MMIO, then the vcpu has
6608 executed a memory-mapped I/O instruction whic 6354 executed a memory-mapped I/O instruction which could not be satisfied
6609 by kvm. The 'data' member contains the writt 6355 by kvm. The 'data' member contains the written data if 'is_write' is
6610 true, and should be filled by application cod 6356 true, and should be filled by application code otherwise.
6611 6357
6612 The 'data' member contains, in its first 'len 6358 The 'data' member contains, in its first 'len' bytes, the value as it would
6613 appear if the VCPU performed a load or store 6359 appear if the VCPU performed a load or store of the appropriate width directly
6614 to the byte array. 6360 to the byte array.
6615 6361
6616 .. note:: 6362 .. note::
6617 6363
6618 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXI 6364 For KVM_EXIT_IO, KVM_EXIT_MMIO, KVM_EXIT_OSI, KVM_EXIT_PAPR, KVM_EXIT_XEN,
6619 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KV 6365 KVM_EXIT_EPR, KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR the corresponding
6620 operations are complete (and guest stat 6366 operations are complete (and guest state is consistent) only after userspace
6621 has re-entered the kernel with KVM_RUN. 6367 has re-entered the kernel with KVM_RUN. The kernel side will first finish
6622 incomplete operations and then check fo 6368 incomplete operations and then check for pending signals.
6623 6369
6624 The pending state of the operation is n 6370 The pending state of the operation is not preserved in state which is
6625 visible to userspace, thus userspace sh 6371 visible to userspace, thus userspace should ensure that the operation is
6626 completed before performing a live migr 6372 completed before performing a live migration. Userspace can re-enter the
6627 guest with an unmasked signal pending o 6373 guest with an unmasked signal pending or with the immediate_exit field set
6628 to complete pending operations without 6374 to complete pending operations without allowing any further instructions
6629 to be executed. 6375 to be executed.
6630 6376
6631 :: 6377 ::
6632 6378
6633 /* KVM_EXIT_HYPERCALL */ 6379 /* KVM_EXIT_HYPERCALL */
6634 struct { 6380 struct {
6635 __u64 nr; 6381 __u64 nr;
6636 __u64 args[6]; 6382 __u64 args[6];
6637 __u64 ret; 6383 __u64 ret;
6638 __u64 flags; 6384 __u64 flags;
6639 } hypercall; 6385 } hypercall;
6640 6386
6641 6387
6642 It is strongly recommended that userspace use 6388 It is strongly recommended that userspace use ``KVM_EXIT_IO`` (x86) or
6643 ``KVM_EXIT_MMIO`` (all except s390) to implem 6389 ``KVM_EXIT_MMIO`` (all except s390) to implement functionality that
6644 requires a guest to interact with host usersp 6390 requires a guest to interact with host userspace.
6645 6391
6646 .. note:: KVM_EXIT_IO is significantly faster 6392 .. note:: KVM_EXIT_IO is significantly faster than KVM_EXIT_MMIO.
6647 6393
6648 For arm64: 6394 For arm64:
6649 ---------- 6395 ----------
6650 6396
6651 SMCCC exits can be enabled depending on the c 6397 SMCCC exits can be enabled depending on the configuration of the SMCCC
6652 filter. See the Documentation/virt/kvm/device 6398 filter. See the Documentation/virt/kvm/devices/vm.rst
6653 ``KVM_ARM_SMCCC_FILTER`` for more details. 6399 ``KVM_ARM_SMCCC_FILTER`` for more details.
6654 6400
6655 ``nr`` contains the function ID of the guest' 6401 ``nr`` contains the function ID of the guest's SMCCC call. Userspace is
6656 expected to use the ``KVM_GET_ONE_REG`` ioctl 6402 expected to use the ``KVM_GET_ONE_REG`` ioctl to retrieve the call
6657 parameters from the vCPU's GPRs. 6403 parameters from the vCPU's GPRs.
6658 6404
6659 Definition of ``flags``: 6405 Definition of ``flags``:
6660 - ``KVM_HYPERCALL_EXIT_SMC``: Indicates that 6406 - ``KVM_HYPERCALL_EXIT_SMC``: Indicates that the guest used the SMC
6661 conduit to initiate the SMCCC call. If thi 6407 conduit to initiate the SMCCC call. If this bit is 0 then the guest
6662 used the HVC conduit for the SMCCC call. 6408 used the HVC conduit for the SMCCC call.
6663 6409
6664 - ``KVM_HYPERCALL_EXIT_16BIT``: Indicates th 6410 - ``KVM_HYPERCALL_EXIT_16BIT``: Indicates that the guest used a 16bit
6665 instruction to initiate the SMCCC call. If 6411 instruction to initiate the SMCCC call. If this bit is 0 then the
6666 guest used a 32bit instruction. An AArch64 6412 guest used a 32bit instruction. An AArch64 guest always has this
6667 bit set to 0. 6413 bit set to 0.
6668 6414
6669 At the point of exit, PC points to the instru 6415 At the point of exit, PC points to the instruction immediately following
6670 the trapping instruction. 6416 the trapping instruction.
6671 6417
6672 :: 6418 ::
6673 6419
6674 /* KVM_EXIT_TPR_ACCESS */ 6420 /* KVM_EXIT_TPR_ACCESS */
6675 struct { 6421 struct {
6676 __u64 rip; 6422 __u64 rip;
6677 __u32 is_write; 6423 __u32 is_write;
6678 __u32 pad; 6424 __u32 pad;
6679 } tpr_access; 6425 } tpr_access;
6680 6426
6681 To be documented (KVM_TPR_ACCESS_REPORTING). 6427 To be documented (KVM_TPR_ACCESS_REPORTING).
6682 6428
6683 :: 6429 ::
6684 6430
6685 /* KVM_EXIT_S390_SIEIC */ 6431 /* KVM_EXIT_S390_SIEIC */
6686 struct { 6432 struct {
6687 __u8 icptcode; 6433 __u8 icptcode;
6688 __u64 mask; /* psw up 6434 __u64 mask; /* psw upper half */
6689 __u64 addr; /* psw lo 6435 __u64 addr; /* psw lower half */
6690 __u16 ipa; 6436 __u16 ipa;
6691 __u32 ipb; 6437 __u32 ipb;
6692 } s390_sieic; 6438 } s390_sieic;
6693 6439
6694 s390 specific. 6440 s390 specific.
6695 6441
6696 :: 6442 ::
6697 6443
6698 /* KVM_EXIT_S390_RESET */ 6444 /* KVM_EXIT_S390_RESET */
6699 #define KVM_S390_RESET_POR 1 6445 #define KVM_S390_RESET_POR 1
6700 #define KVM_S390_RESET_CLEAR 2 6446 #define KVM_S390_RESET_CLEAR 2
6701 #define KVM_S390_RESET_SUBSYSTEM 4 6447 #define KVM_S390_RESET_SUBSYSTEM 4
6702 #define KVM_S390_RESET_CPU_INIT 8 6448 #define KVM_S390_RESET_CPU_INIT 8
6703 #define KVM_S390_RESET_IPL 16 6449 #define KVM_S390_RESET_IPL 16
6704 __u64 s390_reset_flags; 6450 __u64 s390_reset_flags;
6705 6451
6706 s390 specific. 6452 s390 specific.
6707 6453
6708 :: 6454 ::
6709 6455
6710 /* KVM_EXIT_S390_UCONTROL */ 6456 /* KVM_EXIT_S390_UCONTROL */
6711 struct { 6457 struct {
6712 __u64 trans_exc_code; 6458 __u64 trans_exc_code;
6713 __u32 pgm_code; 6459 __u32 pgm_code;
6714 } s390_ucontrol; 6460 } s390_ucontrol;
6715 6461
6716 s390 specific. A page fault has occurred for 6462 s390 specific. A page fault has occurred for a user controlled virtual
6717 machine (KVM_VM_S390_UNCONTROL) on its host p 6463 machine (KVM_VM_S390_UNCONTROL) on its host page table that cannot be
6718 resolved by the kernel. 6464 resolved by the kernel.
6719 The program code and the translation exceptio 6465 The program code and the translation exception code that were placed
6720 in the cpu's lowcore are presented here as de 6466 in the cpu's lowcore are presented here as defined by the z Architecture
6721 Principles of Operation Book in the Chapter f 6467 Principles of Operation Book in the Chapter for Dynamic Address Translation
6722 (DAT) 6468 (DAT)
6723 6469
6724 :: 6470 ::
6725 6471
6726 /* KVM_EXIT_DCR */ 6472 /* KVM_EXIT_DCR */
6727 struct { 6473 struct {
6728 __u32 dcrn; 6474 __u32 dcrn;
6729 __u32 data; 6475 __u32 data;
6730 __u8 is_write; 6476 __u8 is_write;
6731 } dcr; 6477 } dcr;
6732 6478
6733 Deprecated - was used for 440 KVM. 6479 Deprecated - was used for 440 KVM.
6734 6480
6735 :: 6481 ::
6736 6482
6737 /* KVM_EXIT_OSI */ 6483 /* KVM_EXIT_OSI */
6738 struct { 6484 struct {
6739 __u64 gprs[32]; 6485 __u64 gprs[32];
6740 } osi; 6486 } osi;
6741 6487
6742 MOL uses a special hypercall interface it cal 6488 MOL uses a special hypercall interface it calls 'OSI'. To enable it, we catch
6743 hypercalls and exit with this exit struct tha 6489 hypercalls and exit with this exit struct that contains all the guest gprs.
6744 6490
6745 If exit_reason is KVM_EXIT_OSI, then the vcpu 6491 If exit_reason is KVM_EXIT_OSI, then the vcpu has triggered such a hypercall.
6746 Userspace can now handle the hypercall and wh 6492 Userspace can now handle the hypercall and when it's done modify the gprs as
6747 necessary. Upon guest entry all guest GPRs wi 6493 necessary. Upon guest entry all guest GPRs will then be replaced by the values
6748 in this struct. 6494 in this struct.
6749 6495
6750 :: 6496 ::
6751 6497
6752 /* KVM_EXIT_PAPR_HCALL */ 6498 /* KVM_EXIT_PAPR_HCALL */
6753 struct { 6499 struct {
6754 __u64 nr; 6500 __u64 nr;
6755 __u64 ret; 6501 __u64 ret;
6756 __u64 args[9]; 6502 __u64 args[9];
6757 } papr_hcall; 6503 } papr_hcall;
6758 6504
6759 This is used on 64-bit PowerPC when emulating 6505 This is used on 64-bit PowerPC when emulating a pSeries partition,
6760 e.g. with the 'pseries' machine type in qemu. 6506 e.g. with the 'pseries' machine type in qemu. It occurs when the
6761 guest does a hypercall using the 'sc 1' instr 6507 guest does a hypercall using the 'sc 1' instruction. The 'nr' field
6762 contains the hypercall number (from the guest 6508 contains the hypercall number (from the guest R3), and 'args' contains
6763 the arguments (from the guest R4 - R12). Use 6509 the arguments (from the guest R4 - R12). Userspace should put the
6764 return code in 'ret' and any extra returned v 6510 return code in 'ret' and any extra returned values in args[].
6765 The possible hypercalls are defined in the Po 6511 The possible hypercalls are defined in the Power Architecture Platform
6766 Requirements (PAPR) document available from w 6512 Requirements (PAPR) document available from www.power.org (free
6767 developer registration required to access it) 6513 developer registration required to access it).
6768 6514
6769 :: 6515 ::
6770 6516
6771 /* KVM_EXIT_S390_TSCH */ 6517 /* KVM_EXIT_S390_TSCH */
6772 struct { 6518 struct {
6773 __u16 subchannel_id; 6519 __u16 subchannel_id;
6774 __u16 subchannel_nr; 6520 __u16 subchannel_nr;
6775 __u32 io_int_parm; 6521 __u32 io_int_parm;
6776 __u32 io_int_word; 6522 __u32 io_int_word;
6777 __u32 ipb; 6523 __u32 ipb;
6778 __u8 dequeued; 6524 __u8 dequeued;
6779 } s390_tsch; 6525 } s390_tsch;
6780 6526
6781 s390 specific. This exit occurs when KVM_CAP_ 6527 s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
6782 and TEST SUBCHANNEL was intercepted. If deque 6528 and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
6783 interrupt for the target subchannel has been 6529 interrupt for the target subchannel has been dequeued and subchannel_id,
6784 subchannel_nr, io_int_parm and io_int_word co 6530 subchannel_nr, io_int_parm and io_int_word contain the parameters for that
6785 interrupt. ipb is needed for instruction para 6531 interrupt. ipb is needed for instruction parameter decoding.
6786 6532
6787 :: 6533 ::
6788 6534
6789 /* KVM_EXIT_EPR */ 6535 /* KVM_EXIT_EPR */
6790 struct { 6536 struct {
6791 __u32 epr; 6537 __u32 epr;
6792 } epr; 6538 } epr;
6793 6539
6794 On FSL BookE PowerPC chips, the interrupt con 6540 On FSL BookE PowerPC chips, the interrupt controller has a fast patch
6795 interrupt acknowledge path to the core. When 6541 interrupt acknowledge path to the core. When the core successfully
6796 delivers an interrupt, it automatically popul 6542 delivers an interrupt, it automatically populates the EPR register with
6797 the interrupt vector number and acknowledges 6543 the interrupt vector number and acknowledges the interrupt inside
6798 the interrupt controller. 6544 the interrupt controller.
6799 6545
6800 In case the interrupt controller lives in use 6546 In case the interrupt controller lives in user space, we need to do
6801 the interrupt acknowledge cycle through it to 6547 the interrupt acknowledge cycle through it to fetch the next to be
6802 delivered interrupt vector using this exit. 6548 delivered interrupt vector using this exit.
6803 6549
6804 It gets triggered whenever both KVM_CAP_PPC_E 6550 It gets triggered whenever both KVM_CAP_PPC_EPR are enabled and an
6805 external interrupt has just been delivered in 6551 external interrupt has just been delivered into the guest. User space
6806 should put the acknowledged interrupt vector 6552 should put the acknowledged interrupt vector into the 'epr' field.
6807 6553
6808 :: 6554 ::
6809 6555
6810 /* KVM_EXIT_SYSTEM_EVENT */ 6556 /* KVM_EXIT_SYSTEM_EVENT */
6811 struct { 6557 struct {
6812 #define KVM_SYSTEM_EVENT_SHUTDOWN 1 6558 #define KVM_SYSTEM_EVENT_SHUTDOWN 1
6813 #define KVM_SYSTEM_EVENT_RESET 2 6559 #define KVM_SYSTEM_EVENT_RESET 2
6814 #define KVM_SYSTEM_EVENT_CRASH 3 6560 #define KVM_SYSTEM_EVENT_CRASH 3
6815 #define KVM_SYSTEM_EVENT_WAKEUP 4 6561 #define KVM_SYSTEM_EVENT_WAKEUP 4
6816 #define KVM_SYSTEM_EVENT_SUSPEND 5 6562 #define KVM_SYSTEM_EVENT_SUSPEND 5
6817 #define KVM_SYSTEM_EVENT_SEV_TERM 6 6563 #define KVM_SYSTEM_EVENT_SEV_TERM 6
6818 __u32 type; 6564 __u32 type;
6819 __u32 ndata; 6565 __u32 ndata;
6820 __u64 data[16]; 6566 __u64 data[16];
6821 } system_event; 6567 } system_event;
6822 6568
6823 If exit_reason is KVM_EXIT_SYSTEM_EVENT then 6569 If exit_reason is KVM_EXIT_SYSTEM_EVENT then the vcpu has triggered
6824 a system-level event using some architecture 6570 a system-level event using some architecture specific mechanism (hypercall
6825 or some special instruction). In case of ARM6 6571 or some special instruction). In case of ARM64, this is triggered using
6826 HVC instruction based PSCI call from the vcpu 6572 HVC instruction based PSCI call from the vcpu.
6827 6573
6828 The 'type' field describes the system-level e 6574 The 'type' field describes the system-level event type.
6829 Valid values for 'type' are: 6575 Valid values for 'type' are:
6830 6576
6831 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has 6577 - KVM_SYSTEM_EVENT_SHUTDOWN -- the guest has requested a shutdown of the
6832 VM. Userspace is not obliged to honour thi 6578 VM. Userspace is not obliged to honour this, and if it does honour
6833 this does not need to destroy the VM synch 6579 this does not need to destroy the VM synchronously (ie it may call
6834 KVM_RUN again before shutdown finally occu 6580 KVM_RUN again before shutdown finally occurs).
6835 - KVM_SYSTEM_EVENT_RESET -- the guest has re 6581 - KVM_SYSTEM_EVENT_RESET -- the guest has requested a reset of the VM.
6836 As with SHUTDOWN, userspace can choose to 6582 As with SHUTDOWN, userspace can choose to ignore the request, or
6837 to schedule the reset to occur in the futu 6583 to schedule the reset to occur in the future and may call KVM_RUN again.
6838 - KVM_SYSTEM_EVENT_CRASH -- the guest crash 6584 - KVM_SYSTEM_EVENT_CRASH -- the guest crash occurred and the guest
6839 has requested a crash condition maintenanc 6585 has requested a crash condition maintenance. Userspace can choose
6840 to ignore the request, or to gather VM mem 6586 to ignore the request, or to gather VM memory core dump and/or
6841 reset/shutdown of the VM. 6587 reset/shutdown of the VM.
6842 - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV gu 6588 - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination.
6843 The guest physical address of the guest's 6589 The guest physical address of the guest's GHCB is stored in `data[0]`.
6844 - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCP 6590 - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and
6845 KVM has recognized a wakeup event. Userspa 6591 KVM has recognized a wakeup event. Userspace may honor this event by
6846 marking the exiting vCPU as runnable, or d 6592 marking the exiting vCPU as runnable, or deny it and call KVM_RUN again.
6847 - KVM_SYSTEM_EVENT_SUSPEND -- the guest has 6593 - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested a suspension of
6848 the VM. 6594 the VM.
6849 6595
6850 If KVM_CAP_SYSTEM_EVENT_DATA is present, the 6596 If KVM_CAP_SYSTEM_EVENT_DATA is present, the 'data' field can contain
6851 architecture specific information for the sys 6597 architecture specific information for the system-level event. Only
6852 the first `ndata` items (possibly zero) of th 6598 the first `ndata` items (possibly zero) of the data array are valid.
6853 6599
6854 - for arm64, data[0] is set to KVM_SYSTEM_EV 6600 - for arm64, data[0] is set to KVM_SYSTEM_EVENT_RESET_FLAG_PSCI_RESET2 if
6855 the guest issued a SYSTEM_RESET2 call acco 6601 the guest issued a SYSTEM_RESET2 call according to v1.1 of the PSCI
6856 specification. 6602 specification.
6857 6603
6858 - for RISC-V, data[0] is set to the value of 6604 - for RISC-V, data[0] is set to the value of the second argument of the
6859 ``sbi_system_reset`` call. 6605 ``sbi_system_reset`` call.
6860 6606
6861 Previous versions of Linux defined a `flags` 6607 Previous versions of Linux defined a `flags` member in this struct. The
6862 field is now aliased to `data[0]`. Userspace 6608 field is now aliased to `data[0]`. Userspace can assume that it is only
6863 written if ndata is greater than 0. 6609 written if ndata is greater than 0.
6864 6610
6865 For arm/arm64: 6611 For arm/arm64:
6866 -------------- 6612 --------------
6867 6613
6868 KVM_SYSTEM_EVENT_SUSPEND exits are enabled wi 6614 KVM_SYSTEM_EVENT_SUSPEND exits are enabled with the
6869 KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If 6615 KVM_CAP_ARM_SYSTEM_SUSPEND VM capability. If a guest invokes the PSCI
6870 SYSTEM_SUSPEND function, KVM will exit to use 6616 SYSTEM_SUSPEND function, KVM will exit to userspace with this event
6871 type. 6617 type.
6872 6618
6873 It is the sole responsibility of userspace to 6619 It is the sole responsibility of userspace to implement the PSCI
6874 SYSTEM_SUSPEND call according to ARM DEN0022D 6620 SYSTEM_SUSPEND call according to ARM DEN0022D.b 5.19 "SYSTEM_SUSPEND".
6875 KVM does not change the vCPU's state before e 6621 KVM does not change the vCPU's state before exiting to userspace, so
6876 the call parameters are left in-place in the 6622 the call parameters are left in-place in the vCPU registers.
6877 6623
6878 Userspace is _required_ to take action for su 6624 Userspace is _required_ to take action for such an exit. It must
6879 either: 6625 either:
6880 6626
6881 - Honor the guest request to suspend the VM. 6627 - Honor the guest request to suspend the VM. Userspace can request
6882 in-kernel emulation of suspension by setti 6628 in-kernel emulation of suspension by setting the calling vCPU's
6883 state to KVM_MP_STATE_SUSPENDED. Userspace 6629 state to KVM_MP_STATE_SUSPENDED. Userspace must configure the vCPU's
6884 state according to the parameters passed t 6630 state according to the parameters passed to the PSCI function when
6885 the calling vCPU is resumed. See ARM DEN00 6631 the calling vCPU is resumed. See ARM DEN0022D.b 5.19.1 "Intended use"
6886 for details on the function parameters. 6632 for details on the function parameters.
6887 6633
6888 - Deny the guest request to suspend the VM. 6634 - Deny the guest request to suspend the VM. See ARM DEN0022D.b 5.19.2
6889 "Caller responsibilities" for possible ret 6635 "Caller responsibilities" for possible return values.
6890 6636
6891 :: 6637 ::
6892 6638
6893 /* KVM_EXIT_IOAPIC_EOI */ 6639 /* KVM_EXIT_IOAPIC_EOI */
6894 struct { 6640 struct {
6895 __u8 vector; 6641 __u8 vector;
6896 } eoi; 6642 } eoi;
6897 6643
6898 Indicates that the VCPU's in-kernel local API 6644 Indicates that the VCPU's in-kernel local APIC received an EOI for a
6899 level-triggered IOAPIC interrupt. This exit 6645 level-triggered IOAPIC interrupt. This exit only triggers when the
6900 IOAPIC is implemented in userspace (i.e. KVM_ 6646 IOAPIC is implemented in userspace (i.e. KVM_CAP_SPLIT_IRQCHIP is enabled);
6901 the userspace IOAPIC should process the EOI a 6647 the userspace IOAPIC should process the EOI and retrigger the interrupt if
6902 it is still asserted. Vector is the LAPIC in 6648 it is still asserted. Vector is the LAPIC interrupt vector for which the
6903 EOI was received. 6649 EOI was received.
6904 6650
6905 :: 6651 ::
6906 6652
6907 struct kvm_hyperv_exit { 6653 struct kvm_hyperv_exit {
6908 #define KVM_EXIT_HYPERV_SYNIC 1 6654 #define KVM_EXIT_HYPERV_SYNIC 1
6909 #define KVM_EXIT_HYPERV_HCALL 2 6655 #define KVM_EXIT_HYPERV_HCALL 2
6910 #define KVM_EXIT_HYPERV_SYNDBG 3 6656 #define KVM_EXIT_HYPERV_SYNDBG 3
6911 __u32 type; 6657 __u32 type;
6912 __u32 pad1; 6658 __u32 pad1;
6913 union { 6659 union {
6914 struct { 6660 struct {
6915 __u32 6661 __u32 msr;
6916 __u32 6662 __u32 pad2;
6917 __u64 6663 __u64 control;
6918 __u64 6664 __u64 evt_page;
6919 __u64 6665 __u64 msg_page;
6920 } synic; 6666 } synic;
6921 struct { 6667 struct {
6922 __u64 6668 __u64 input;
6923 __u64 6669 __u64 result;
6924 __u64 6670 __u64 params[2];
6925 } hcall; 6671 } hcall;
6926 struct { 6672 struct {
6927 __u32 6673 __u32 msr;
6928 __u32 6674 __u32 pad2;
6929 __u64 6675 __u64 control;
6930 __u64 6676 __u64 status;
6931 __u64 6677 __u64 send_page;
6932 __u64 6678 __u64 recv_page;
6933 __u64 6679 __u64 pending_page;
6934 } syndbg; 6680 } syndbg;
6935 } u; 6681 } u;
6936 }; 6682 };
6937 /* KVM_EXIT_HYPERV */ 6683 /* KVM_EXIT_HYPERV */
6938 struct kvm_hyperv_exit hyperv 6684 struct kvm_hyperv_exit hyperv;
6939 6685
6940 Indicates that the VCPU exits into userspace 6686 Indicates that the VCPU exits into userspace to process some tasks
6941 related to Hyper-V emulation. 6687 related to Hyper-V emulation.
6942 6688
6943 Valid values for 'type' are: 6689 Valid values for 'type' are:
6944 6690
6945 - KVM_EXIT_HYPERV_SYNIC -- synchronou 6691 - KVM_EXIT_HYPERV_SYNIC -- synchronously notify user-space about
6946 6692
6947 Hyper-V SynIC state change. Notification is u 6693 Hyper-V SynIC state change. Notification is used to remap SynIC
6948 event/message pages and to enable/disable Syn 6694 event/message pages and to enable/disable SynIC messages/events processing
6949 in userspace. 6695 in userspace.
6950 6696
6951 - KVM_EXIT_HYPERV_SYNDBG -- synchrono 6697 - KVM_EXIT_HYPERV_SYNDBG -- synchronously notify user-space about
6952 6698
6953 Hyper-V Synthetic debugger state change. Noti 6699 Hyper-V Synthetic debugger state change. Notification is used to either update
6954 the pending_page location or to send a contro 6700 the pending_page location or to send a control command (send the buffer located
6955 in send_page or recv a buffer to recv_page). 6701 in send_page or recv a buffer to recv_page).
6956 6702
6957 :: 6703 ::
6958 6704
6959 /* KVM_EXIT_ARM_NISV */ 6705 /* KVM_EXIT_ARM_NISV */
6960 struct { 6706 struct {
6961 __u64 esr_iss; 6707 __u64 esr_iss;
6962 __u64 fault_ipa; 6708 __u64 fault_ipa;
6963 } arm_nisv; 6709 } arm_nisv;
6964 6710
6965 Used on arm64 systems. If a guest accesses me 6711 Used on arm64 systems. If a guest accesses memory not in a memslot,
6966 KVM will typically return to userspace and as 6712 KVM will typically return to userspace and ask it to do MMIO emulation on its
6967 behalf. However, for certain classes of instr 6713 behalf. However, for certain classes of instructions, no instruction decode
6968 (direction, length of memory access) is provi 6714 (direction, length of memory access) is provided, and fetching and decoding
6969 the instruction from the VM is overly complic 6715 the instruction from the VM is overly complicated to live in the kernel.
6970 6716
6971 Historically, when this situation occurred, K 6717 Historically, when this situation occurred, KVM would print a warning and kill
6972 the VM. KVM assumed that if the guest accesse 6718 the VM. KVM assumed that if the guest accessed non-memslot memory, it was
6973 trying to do I/O, which just couldn't be emul 6719 trying to do I/O, which just couldn't be emulated, and the warning message was
6974 phrased accordingly. However, what happened m 6720 phrased accordingly. However, what happened more often was that a guest bug
6975 caused access outside the guest memory areas 6721 caused access outside the guest memory areas which should lead to a more
6976 meaningful warning message and an external ab 6722 meaningful warning message and an external abort in the guest, if the access
6977 did not fall within an I/O window. 6723 did not fall within an I/O window.
6978 6724
6979 Userspace implementations can query for KVM_C 6725 Userspace implementations can query for KVM_CAP_ARM_NISV_TO_USER, and enable
6980 this capability at VM creation. Once this is 6726 this capability at VM creation. Once this is done, these types of errors will
6981 instead return to userspace with KVM_EXIT_ARM 6727 instead return to userspace with KVM_EXIT_ARM_NISV, with the valid bits from
6982 the ESR_EL2 in the esr_iss field, and the fau 6728 the ESR_EL2 in the esr_iss field, and the faulting IPA in the fault_ipa field.
6983 Userspace can either fix up the access if it' 6729 Userspace can either fix up the access if it's actually an I/O access by
6984 decoding the instruction from guest memory (i 6730 decoding the instruction from guest memory (if it's very brave) and continue
6985 executing the guest, or it can decide to susp 6731 executing the guest, or it can decide to suspend, dump, or restart the guest.
6986 6732
6987 Note that KVM does not skip the faulting inst 6733 Note that KVM does not skip the faulting instruction as it does for
6988 KVM_EXIT_MMIO, but userspace has to emulate a 6734 KVM_EXIT_MMIO, but userspace has to emulate any change to the processing state
6989 if it decides to decode and emulate the instr 6735 if it decides to decode and emulate the instruction.
6990 6736
6991 This feature isn't available to protected VMs <<
6992 have access to the state that is required to <<
6993 Instead, a data abort exception is directly i <<
6994 Note that although KVM_CAP_ARM_NISV_TO_USER w <<
6995 queried outside of a protected VM context, th <<
6996 exposed if queried on a protected VM file des <<
6997 <<
6998 :: 6737 ::
6999 6738
7000 /* KVM_EXIT_X86_RDMSR / KVM_E 6739 /* KVM_EXIT_X86_RDMSR / KVM_EXIT_X86_WRMSR */
7001 struct { 6740 struct {
7002 __u8 error; /* user - 6741 __u8 error; /* user -> kernel */
7003 __u8 pad[7]; 6742 __u8 pad[7];
7004 __u32 reason; /* kern 6743 __u32 reason; /* kernel -> user */
7005 __u32 index; /* kerne 6744 __u32 index; /* kernel -> user */
7006 __u64 data; /* kernel 6745 __u64 data; /* kernel <-> user */
7007 } msr; 6746 } msr;
7008 6747
7009 Used on x86 systems. When the VM capability K 6748 Used on x86 systems. When the VM capability KVM_CAP_X86_USER_SPACE_MSR is
7010 enabled, MSR accesses to registers that would 6749 enabled, MSR accesses to registers that would invoke a #GP by KVM kernel code
7011 may instead trigger a KVM_EXIT_X86_RDMSR exit 6750 may instead trigger a KVM_EXIT_X86_RDMSR exit for reads and KVM_EXIT_X86_WRMSR
7012 exit for writes. 6751 exit for writes.
7013 6752
7014 The "reason" field specifies why the MSR inte 6753 The "reason" field specifies why the MSR interception occurred. Userspace will
7015 only receive MSR exits when a particular reas 6754 only receive MSR exits when a particular reason was requested during through
7016 ENABLE_CAP. Currently valid exit reasons are: 6755 ENABLE_CAP. Currently valid exit reasons are:
7017 6756
7018 ============================ ================ 6757 ============================ ========================================
7019 KVM_MSR_EXIT_REASON_UNKNOWN access to MSR th 6758 KVM_MSR_EXIT_REASON_UNKNOWN access to MSR that is unknown to KVM
7020 KVM_MSR_EXIT_REASON_INVAL access to invali 6759 KVM_MSR_EXIT_REASON_INVAL access to invalid MSRs or reserved bits
7021 KVM_MSR_EXIT_REASON_FILTER access blocked b 6760 KVM_MSR_EXIT_REASON_FILTER access blocked by KVM_X86_SET_MSR_FILTER
7022 ============================ ================ 6761 ============================ ========================================
7023 6762
7024 For KVM_EXIT_X86_RDMSR, the "index" field tel 6763 For KVM_EXIT_X86_RDMSR, the "index" field tells userspace which MSR the guest
7025 wants to read. To respond to this request wit 6764 wants to read. To respond to this request with a successful read, userspace
7026 writes the respective data into the "data" fi 6765 writes the respective data into the "data" field and must continue guest
7027 execution to ensure the read data is transfer 6766 execution to ensure the read data is transferred into guest register state.
7028 6767
7029 If the RDMSR request was unsuccessful, usersp 6768 If the RDMSR request was unsuccessful, userspace indicates that with a "1" in
7030 the "error" field. This will inject a #GP int 6769 the "error" field. This will inject a #GP into the guest when the VCPU is
7031 executed again. 6770 executed again.
7032 6771
7033 For KVM_EXIT_X86_WRMSR, the "index" field tel 6772 For KVM_EXIT_X86_WRMSR, the "index" field tells userspace which MSR the guest
7034 wants to write. Once finished processing the 6773 wants to write. Once finished processing the event, userspace must continue
7035 vCPU execution. If the MSR write was unsucces 6774 vCPU execution. If the MSR write was unsuccessful, userspace also sets the
7036 "error" field to "1". 6775 "error" field to "1".
7037 6776
7038 See KVM_X86_SET_MSR_FILTER for details on the 6777 See KVM_X86_SET_MSR_FILTER for details on the interaction with MSR filtering.
7039 6778
7040 :: 6779 ::
7041 6780
7042 6781
7043 struct kvm_xen_exit { 6782 struct kvm_xen_exit {
7044 #define KVM_EXIT_XEN_HCALL 1 6783 #define KVM_EXIT_XEN_HCALL 1
7045 __u32 type; 6784 __u32 type;
7046 union { 6785 union {
7047 struct { 6786 struct {
7048 __u32 6787 __u32 longmode;
7049 __u32 6788 __u32 cpl;
7050 __u64 6789 __u64 input;
7051 __u64 6790 __u64 result;
7052 __u64 6791 __u64 params[6];
7053 } hcall; 6792 } hcall;
7054 } u; 6793 } u;
7055 }; 6794 };
7056 /* KVM_EXIT_XEN */ 6795 /* KVM_EXIT_XEN */
7057 struct kvm_hyperv_exit xen; 6796 struct kvm_hyperv_exit xen;
7058 6797
7059 Indicates that the VCPU exits into userspace 6798 Indicates that the VCPU exits into userspace to process some tasks
7060 related to Xen emulation. 6799 related to Xen emulation.
7061 6800
7062 Valid values for 'type' are: 6801 Valid values for 'type' are:
7063 6802
7064 - KVM_EXIT_XEN_HCALL -- synchronously notif 6803 - KVM_EXIT_XEN_HCALL -- synchronously notify user-space about Xen hypercall.
7065 Userspace is expected to place the hyperc 6804 Userspace is expected to place the hypercall result into the appropriate
7066 field before invoking KVM_RUN again. 6805 field before invoking KVM_RUN again.
7067 6806
7068 :: 6807 ::
7069 6808
7070 /* KVM_EXIT_RISCV_SBI */ 6809 /* KVM_EXIT_RISCV_SBI */
7071 struct { 6810 struct {
7072 unsigned long extensi 6811 unsigned long extension_id;
7073 unsigned long functio 6812 unsigned long function_id;
7074 unsigned long args[6] 6813 unsigned long args[6];
7075 unsigned long ret[2]; 6814 unsigned long ret[2];
7076 } riscv_sbi; 6815 } riscv_sbi;
7077 6816
7078 If exit reason is KVM_EXIT_RISCV_SBI then it 6817 If exit reason is KVM_EXIT_RISCV_SBI then it indicates that the VCPU has
7079 done a SBI call which is not handled by KVM R 6818 done a SBI call which is not handled by KVM RISC-V kernel module. The details
7080 of the SBI call are available in 'riscv_sbi' 6819 of the SBI call are available in 'riscv_sbi' member of kvm_run structure. The
7081 'extension_id' field of 'riscv_sbi' represent 6820 'extension_id' field of 'riscv_sbi' represents SBI extension ID whereas the
7082 'function_id' field represents function ID of 6821 'function_id' field represents function ID of given SBI extension. The 'args'
7083 array field of 'riscv_sbi' represents paramet 6822 array field of 'riscv_sbi' represents parameters for the SBI call and 'ret'
7084 array field represents return values. The use 6823 array field represents return values. The userspace should update the return
7085 values of SBI call before resuming the VCPU. 6824 values of SBI call before resuming the VCPU. For more details on RISC-V SBI
7086 spec refer, https://github.com/riscv/riscv-sb 6825 spec refer, https://github.com/riscv/riscv-sbi-doc.
7087 6826
7088 :: 6827 ::
7089 6828
7090 /* KVM_EXIT_MEMORY_FAULT */ <<
7091 struct { <<
7092 #define KVM_MEMORY_EXIT_FLAG_PRIVATE (1ULL <<
7093 __u64 flags; <<
7094 __u64 gpa; <<
7095 __u64 size; <<
7096 } memory_fault; <<
7097 <<
7098 KVM_EXIT_MEMORY_FAULT indicates the vCPU has <<
7099 could not be resolved by KVM. The 'gpa' and <<
7100 guest physical address range [gpa, gpa + size <<
7101 describes properties of the faulting access t <<
7102 <<
7103 - KVM_MEMORY_EXIT_FLAG_PRIVATE - When set, i <<
7104 on a private memory access. When clear, i <<
7105 shared access. <<
7106 <<
7107 Note! KVM_EXIT_MEMORY_FAULT is unique among <<
7108 accompanies a return code of '-1', not '0'! <<
7109 or EHWPOISON when KVM exits with KVM_EXIT_MEM <<
7110 kvm_run.exit_reason is stale/undefined for al <<
7111 <<
7112 :: <<
7113 <<
7114 /* KVM_EXIT_NOTIFY */ 6829 /* KVM_EXIT_NOTIFY */
7115 struct { 6830 struct {
7116 #define KVM_NOTIFY_CONTEXT_INVALID (1 << 6831 #define KVM_NOTIFY_CONTEXT_INVALID (1 << 0)
7117 __u32 flags; 6832 __u32 flags;
7118 } notify; 6833 } notify;
7119 6834
7120 Used on x86 systems. When the VM capability K 6835 Used on x86 systems. When the VM capability KVM_CAP_X86_NOTIFY_VMEXIT is
7121 enabled, a VM exit generated if no event wind 6836 enabled, a VM exit generated if no event window occurs in VM non-root mode
7122 for a specified amount of time. Once KVM_X86_ 6837 for a specified amount of time. Once KVM_X86_NOTIFY_VMEXIT_USER is set when
7123 enabling the cap, it would exit to userspace 6838 enabling the cap, it would exit to userspace with the exit reason
7124 KVM_EXIT_NOTIFY for further handling. The "fl 6839 KVM_EXIT_NOTIFY for further handling. The "flags" field contains more
7125 detailed info. 6840 detailed info.
7126 6841
7127 The valid value for 'flags' is: 6842 The valid value for 'flags' is:
7128 6843
7129 - KVM_NOTIFY_CONTEXT_INVALID -- the VM cont 6844 - KVM_NOTIFY_CONTEXT_INVALID -- the VM context is corrupted and not valid
7130 in VMCS. It would run into unknown result 6845 in VMCS. It would run into unknown result if resume the target VM.
7131 6846
7132 :: 6847 ::
7133 6848
7134 /* Fix the size of the union. 6849 /* Fix the size of the union. */
7135 char padding[256]; 6850 char padding[256];
7136 }; 6851 };
7137 6852
7138 /* 6853 /*
7139 * shared registers between kvm and u 6854 * shared registers between kvm and userspace.
7140 * kvm_valid_regs specifies the regis 6855 * kvm_valid_regs specifies the register classes set by the host
7141 * kvm_dirty_regs specified the regis 6856 * kvm_dirty_regs specified the register classes dirtied by userspace
7142 * struct kvm_sync_regs is architectu 6857 * struct kvm_sync_regs is architecture specific, as well as the
7143 * bits for kvm_valid_regs and kvm_di 6858 * bits for kvm_valid_regs and kvm_dirty_regs
7144 */ 6859 */
7145 __u64 kvm_valid_regs; 6860 __u64 kvm_valid_regs;
7146 __u64 kvm_dirty_regs; 6861 __u64 kvm_dirty_regs;
7147 union { 6862 union {
7148 struct kvm_sync_regs regs; 6863 struct kvm_sync_regs regs;
7149 char padding[SYNC_REGS_SIZE_B 6864 char padding[SYNC_REGS_SIZE_BYTES];
7150 } s; 6865 } s;
7151 6866
7152 If KVM_CAP_SYNC_REGS is defined, these fields 6867 If KVM_CAP_SYNC_REGS is defined, these fields allow userspace to access
7153 certain guest registers without having to cal 6868 certain guest registers without having to call SET/GET_*REGS. Thus we can
7154 avoid some system call overhead if userspace 6869 avoid some system call overhead if userspace has to handle the exit.
7155 Userspace can query the validity of the struc 6870 Userspace can query the validity of the structure by checking
7156 kvm_valid_regs for specific bits. These bits 6871 kvm_valid_regs for specific bits. These bits are architecture specific
7157 and usually define the validity of a groups o 6872 and usually define the validity of a groups of registers. (e.g. one bit
7158 for general purpose registers) 6873 for general purpose registers)
7159 6874
7160 Please note that the kernel is allowed to use 6875 Please note that the kernel is allowed to use the kvm_run structure as the
7161 primary storage for certain register types. T 6876 primary storage for certain register types. Therefore, the kernel may use the
7162 values in kvm_run even if the corresponding b 6877 values in kvm_run even if the corresponding bit in kvm_dirty_regs is not set.
7163 6878
7164 6879
7165 6. Capabilities that can be enabled on vCPUs 6880 6. Capabilities that can be enabled on vCPUs
7166 ============================================ 6881 ============================================
7167 6882
7168 There are certain capabilities that change th 6883 There are certain capabilities that change the behavior of the virtual CPU or
7169 the virtual machine when enabled. To enable t 6884 the virtual machine when enabled. To enable them, please see section 4.37.
7170 Below you can find a list of capabilities and 6885 Below you can find a list of capabilities and what their effect on the vCPU or
7171 the virtual machine is when enabling them. 6886 the virtual machine is when enabling them.
7172 6887
7173 The following information is provided along w 6888 The following information is provided along with the description:
7174 6889
7175 Architectures: 6890 Architectures:
7176 which instruction set architectures pro 6891 which instruction set architectures provide this ioctl.
7177 x86 includes both i386 and x86_64. 6892 x86 includes both i386 and x86_64.
7178 6893
7179 Target: 6894 Target:
7180 whether this is a per-vcpu or per-vm ca 6895 whether this is a per-vcpu or per-vm capability.
7181 6896
7182 Parameters: 6897 Parameters:
7183 what parameters are accepted by the cap 6898 what parameters are accepted by the capability.
7184 6899
7185 Returns: 6900 Returns:
7186 the return value. General error number 6901 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
7187 are not detailed, but errors with speci 6902 are not detailed, but errors with specific meanings are.
7188 6903
7189 6904
7190 6.1 KVM_CAP_PPC_OSI 6905 6.1 KVM_CAP_PPC_OSI
7191 ------------------- 6906 -------------------
7192 6907
7193 :Architectures: ppc 6908 :Architectures: ppc
7194 :Target: vcpu 6909 :Target: vcpu
7195 :Parameters: none 6910 :Parameters: none
7196 :Returns: 0 on success; -1 on error 6911 :Returns: 0 on success; -1 on error
7197 6912
7198 This capability enables interception of OSI h 6913 This capability enables interception of OSI hypercalls that otherwise would
7199 be treated as normal system calls to be injec 6914 be treated as normal system calls to be injected into the guest. OSI hypercalls
7200 were invented by Mac-on-Linux to have a stand 6915 were invented by Mac-on-Linux to have a standardized communication mechanism
7201 between the guest and the host. 6916 between the guest and the host.
7202 6917
7203 When this capability is enabled, KVM_EXIT_OSI 6918 When this capability is enabled, KVM_EXIT_OSI can occur.
7204 6919
7205 6920
7206 6.2 KVM_CAP_PPC_PAPR 6921 6.2 KVM_CAP_PPC_PAPR
7207 -------------------- 6922 --------------------
7208 6923
7209 :Architectures: ppc 6924 :Architectures: ppc
7210 :Target: vcpu 6925 :Target: vcpu
7211 :Parameters: none 6926 :Parameters: none
7212 :Returns: 0 on success; -1 on error 6927 :Returns: 0 on success; -1 on error
7213 6928
7214 This capability enables interception of PAPR 6929 This capability enables interception of PAPR hypercalls. PAPR hypercalls are
7215 done using the hypercall instruction "sc 1". 6930 done using the hypercall instruction "sc 1".
7216 6931
7217 It also sets the guest privilege level to "su 6932 It also sets the guest privilege level to "supervisor" mode. Usually the guest
7218 runs in "hypervisor" privilege mode with a fe 6933 runs in "hypervisor" privilege mode with a few missing features.
7219 6934
7220 In addition to the above, it changes the sema 6935 In addition to the above, it changes the semantics of SDR1. In this mode, the
7221 HTAB address part of SDR1 contains an HVA ins 6936 HTAB address part of SDR1 contains an HVA instead of a GPA, as PAPR keeps the
7222 HTAB invisible to the guest. 6937 HTAB invisible to the guest.
7223 6938
7224 When this capability is enabled, KVM_EXIT_PAP 6939 When this capability is enabled, KVM_EXIT_PAPR_HCALL can occur.
7225 6940
7226 6941
7227 6.3 KVM_CAP_SW_TLB 6942 6.3 KVM_CAP_SW_TLB
7228 ------------------ 6943 ------------------
7229 6944
7230 :Architectures: ppc 6945 :Architectures: ppc
7231 :Target: vcpu 6946 :Target: vcpu
7232 :Parameters: args[0] is the address of a stru 6947 :Parameters: args[0] is the address of a struct kvm_config_tlb
7233 :Returns: 0 on success; -1 on error 6948 :Returns: 0 on success; -1 on error
7234 6949
7235 :: 6950 ::
7236 6951
7237 struct kvm_config_tlb { 6952 struct kvm_config_tlb {
7238 __u64 params; 6953 __u64 params;
7239 __u64 array; 6954 __u64 array;
7240 __u32 mmu_type; 6955 __u32 mmu_type;
7241 __u32 array_len; 6956 __u32 array_len;
7242 }; 6957 };
7243 6958
7244 Configures the virtual CPU's TLB array, estab 6959 Configures the virtual CPU's TLB array, establishing a shared memory area
7245 between userspace and KVM. The "params" and 6960 between userspace and KVM. The "params" and "array" fields are userspace
7246 addresses of mmu-type-specific data structure 6961 addresses of mmu-type-specific data structures. The "array_len" field is an
7247 safety mechanism, and should be set to the si 6962 safety mechanism, and should be set to the size in bytes of the memory that
7248 userspace has reserved for the array. It mus 6963 userspace has reserved for the array. It must be at least the size dictated
7249 by "mmu_type" and "params". 6964 by "mmu_type" and "params".
7250 6965
7251 While KVM_RUN is active, the shared region is 6966 While KVM_RUN is active, the shared region is under control of KVM. Its
7252 contents are undefined, and any modification 6967 contents are undefined, and any modification by userspace results in
7253 boundedly undefined behavior. 6968 boundedly undefined behavior.
7254 6969
7255 On return from KVM_RUN, the shared region wil 6970 On return from KVM_RUN, the shared region will reflect the current state of
7256 the guest's TLB. If userspace makes any chan 6971 the guest's TLB. If userspace makes any changes, it must call KVM_DIRTY_TLB
7257 to tell KVM which entries have been changed, 6972 to tell KVM which entries have been changed, prior to calling KVM_RUN again
7258 on this vcpu. 6973 on this vcpu.
7259 6974
7260 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_ 6975 For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
7261 6976
7262 - The "params" field is of type "struct kvm_ 6977 - The "params" field is of type "struct kvm_book3e_206_tlb_params".
7263 - The "array" field points to an array of ty 6978 - The "array" field points to an array of type "struct
7264 kvm_book3e_206_tlb_entry". 6979 kvm_book3e_206_tlb_entry".
7265 - The array consists of all entries in the f 6980 - The array consists of all entries in the first TLB, followed by all
7266 entries in the second TLB. 6981 entries in the second TLB.
7267 - Within a TLB, entries are ordered first by 6982 - Within a TLB, entries are ordered first by increasing set number. Within a
7268 set, entries are ordered by way (increasin 6983 set, entries are ordered by way (increasing ESEL).
7269 - The hash for determining set number in TLB 6984 - The hash for determining set number in TLB0 is: (MAS2 >> 12) & (num_sets - 1)
7270 where "num_sets" is the tlb_sizes[] value 6985 where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
7271 - The tsize field of mas1 shall be set to 4K 6986 - The tsize field of mas1 shall be set to 4K on TLB0, even though the
7272 hardware ignores this value for TLB0. 6987 hardware ignores this value for TLB0.
7273 6988
7274 6.4 KVM_CAP_S390_CSS_SUPPORT 6989 6.4 KVM_CAP_S390_CSS_SUPPORT
7275 ---------------------------- 6990 ----------------------------
7276 6991
7277 :Architectures: s390 6992 :Architectures: s390
7278 :Target: vcpu 6993 :Target: vcpu
7279 :Parameters: none 6994 :Parameters: none
7280 :Returns: 0 on success; -1 on error 6995 :Returns: 0 on success; -1 on error
7281 6996
7282 This capability enables support for handling 6997 This capability enables support for handling of channel I/O instructions.
7283 6998
7284 TEST PENDING INTERRUPTION and the interrupt p 6999 TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
7285 handled in-kernel, while the other I/O instru 7000 handled in-kernel, while the other I/O instructions are passed to userspace.
7286 7001
7287 When this capability is enabled, KVM_EXIT_S39 7002 When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
7288 SUBCHANNEL intercepts. 7003 SUBCHANNEL intercepts.
7289 7004
7290 Note that even though this capability is enab 7005 Note that even though this capability is enabled per-vcpu, the complete
7291 virtual machine is affected. 7006 virtual machine is affected.
7292 7007
7293 6.5 KVM_CAP_PPC_EPR 7008 6.5 KVM_CAP_PPC_EPR
7294 ------------------- 7009 -------------------
7295 7010
7296 :Architectures: ppc 7011 :Architectures: ppc
7297 :Target: vcpu 7012 :Target: vcpu
7298 :Parameters: args[0] defines whether the prox 7013 :Parameters: args[0] defines whether the proxy facility is active
7299 :Returns: 0 on success; -1 on error 7014 :Returns: 0 on success; -1 on error
7300 7015
7301 This capability enables or disables the deliv 7016 This capability enables or disables the delivery of interrupts through the
7302 external proxy facility. 7017 external proxy facility.
7303 7018
7304 When enabled (args[0] != 0), every time the g 7019 When enabled (args[0] != 0), every time the guest gets an external interrupt
7305 delivered, it automatically exits into user s 7020 delivered, it automatically exits into user space with a KVM_EXIT_EPR exit
7306 to receive the topmost interrupt vector. 7021 to receive the topmost interrupt vector.
7307 7022
7308 When disabled (args[0] == 0), behavior is as 7023 When disabled (args[0] == 0), behavior is as if this facility is unsupported.
7309 7024
7310 When this capability is enabled, KVM_EXIT_EPR 7025 When this capability is enabled, KVM_EXIT_EPR can occur.
7311 7026
7312 6.6 KVM_CAP_IRQ_MPIC 7027 6.6 KVM_CAP_IRQ_MPIC
7313 -------------------- 7028 --------------------
7314 7029
7315 :Architectures: ppc 7030 :Architectures: ppc
7316 :Parameters: args[0] is the MPIC device fd; 7031 :Parameters: args[0] is the MPIC device fd;
7317 args[1] is the MPIC CPU number f 7032 args[1] is the MPIC CPU number for this vcpu
7318 7033
7319 This capability connects the vcpu to an in-ke 7034 This capability connects the vcpu to an in-kernel MPIC device.
7320 7035
7321 6.7 KVM_CAP_IRQ_XICS 7036 6.7 KVM_CAP_IRQ_XICS
7322 -------------------- 7037 --------------------
7323 7038
7324 :Architectures: ppc 7039 :Architectures: ppc
7325 :Target: vcpu 7040 :Target: vcpu
7326 :Parameters: args[0] is the XICS device fd; 7041 :Parameters: args[0] is the XICS device fd;
7327 args[1] is the XICS CPU number ( 7042 args[1] is the XICS CPU number (server ID) for this vcpu
7328 7043
7329 This capability connects the vcpu to an in-ke 7044 This capability connects the vcpu to an in-kernel XICS device.
7330 7045
7331 6.8 KVM_CAP_S390_IRQCHIP 7046 6.8 KVM_CAP_S390_IRQCHIP
7332 ------------------------ 7047 ------------------------
7333 7048
7334 :Architectures: s390 7049 :Architectures: s390
7335 :Target: vm 7050 :Target: vm
7336 :Parameters: none 7051 :Parameters: none
7337 7052
7338 This capability enables the in-kernel irqchip 7053 This capability enables the in-kernel irqchip for s390. Please refer to
7339 "4.24 KVM_CREATE_IRQCHIP" for details. 7054 "4.24 KVM_CREATE_IRQCHIP" for details.
7340 7055
7341 6.9 KVM_CAP_MIPS_FPU 7056 6.9 KVM_CAP_MIPS_FPU
7342 -------------------- 7057 --------------------
7343 7058
7344 :Architectures: mips 7059 :Architectures: mips
7345 :Target: vcpu 7060 :Target: vcpu
7346 :Parameters: args[0] is reserved for future u 7061 :Parameters: args[0] is reserved for future use (should be 0).
7347 7062
7348 This capability allows the use of the host Fl 7063 This capability allows the use of the host Floating Point Unit by the guest. It
7349 allows the Config1.FP bit to be set to enable 7064 allows the Config1.FP bit to be set to enable the FPU in the guest. Once this is
7350 done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG 7065 done the ``KVM_REG_MIPS_FPR_*`` and ``KVM_REG_MIPS_FCR_*`` registers can be
7351 accessed (depending on the current guest FPU 7066 accessed (depending on the current guest FPU register mode), and the Status.FR,
7352 Config5.FRE bits are accessible via the KVM A 7067 Config5.FRE bits are accessible via the KVM API and also from the guest,
7353 depending on them being supported by the FPU. 7068 depending on them being supported by the FPU.
7354 7069
7355 6.10 KVM_CAP_MIPS_MSA 7070 6.10 KVM_CAP_MIPS_MSA
7356 --------------------- 7071 ---------------------
7357 7072
7358 :Architectures: mips 7073 :Architectures: mips
7359 :Target: vcpu 7074 :Target: vcpu
7360 :Parameters: args[0] is reserved for future u 7075 :Parameters: args[0] is reserved for future use (should be 0).
7361 7076
7362 This capability allows the use of the MIPS SI 7077 This capability allows the use of the MIPS SIMD Architecture (MSA) by the guest.
7363 It allows the Config3.MSAP bit to be set to e 7078 It allows the Config3.MSAP bit to be set to enable the use of MSA by the guest.
7364 Once this is done the ``KVM_REG_MIPS_VEC_*`` 7079 Once this is done the ``KVM_REG_MIPS_VEC_*`` and ``KVM_REG_MIPS_MSA_*``
7365 registers can be accessed, and the Config5.MS 7080 registers can be accessed, and the Config5.MSAEn bit is accessible via the
7366 KVM API and also from the guest. 7081 KVM API and also from the guest.
7367 7082
7368 6.74 KVM_CAP_SYNC_REGS 7083 6.74 KVM_CAP_SYNC_REGS
7369 ---------------------- 7084 ----------------------
7370 7085
7371 :Architectures: s390, x86 7086 :Architectures: s390, x86
7372 :Target: s390: always enabled, x86: vcpu 7087 :Target: s390: always enabled, x86: vcpu
7373 :Parameters: none 7088 :Parameters: none
7374 :Returns: x86: KVM_CHECK_EXTENSION returns a 7089 :Returns: x86: KVM_CHECK_EXTENSION returns a bit-array indicating which register
7375 sets are supported 7090 sets are supported
7376 (bitfields defined in arch/x86/incl 7091 (bitfields defined in arch/x86/include/uapi/asm/kvm.h).
7377 7092
7378 As described above in the kvm_sync_regs struc 7093 As described above in the kvm_sync_regs struct info in section 5 (kvm_run):
7379 KVM_CAP_SYNC_REGS "allow[s] userspace to acce 7094 KVM_CAP_SYNC_REGS "allow[s] userspace to access certain guest registers
7380 without having to call SET/GET_*REGS". This r 7095 without having to call SET/GET_*REGS". This reduces overhead by eliminating
7381 repeated ioctl calls for setting and/or getti 7096 repeated ioctl calls for setting and/or getting register values. This is
7382 particularly important when userspace is maki 7097 particularly important when userspace is making synchronous guest state
7383 modifications, e.g. when emulating and/or int 7098 modifications, e.g. when emulating and/or intercepting instructions in
7384 userspace. 7099 userspace.
7385 7100
7386 For s390 specifics, please refer to the sourc 7101 For s390 specifics, please refer to the source code.
7387 7102
7388 For x86: 7103 For x86:
7389 7104
7390 - the register sets to be copied out to kvm_r 7105 - the register sets to be copied out to kvm_run are selectable
7391 by userspace (rather that all sets being co 7106 by userspace (rather that all sets being copied out for every exit).
7392 - vcpu_events are available in addition to re 7107 - vcpu_events are available in addition to regs and sregs.
7393 7108
7394 For x86, the 'kvm_valid_regs' field of struct 7109 For x86, the 'kvm_valid_regs' field of struct kvm_run is overloaded to
7395 function as an input bit-array field set by u 7110 function as an input bit-array field set by userspace to indicate the
7396 specific register sets to be copied out on th 7111 specific register sets to be copied out on the next exit.
7397 7112
7398 To indicate when userspace has modified value 7113 To indicate when userspace has modified values that should be copied into
7399 the vCPU, the all architecture bitarray field 7114 the vCPU, the all architecture bitarray field, 'kvm_dirty_regs' must be set.
7400 This is done using the same bitflags as for t 7115 This is done using the same bitflags as for the 'kvm_valid_regs' field.
7401 If the dirty bit is not set, then the registe 7116 If the dirty bit is not set, then the register set values will not be copied
7402 into the vCPU even if they've been modified. 7117 into the vCPU even if they've been modified.
7403 7118
7404 Unused bitfields in the bitarrays must be set 7119 Unused bitfields in the bitarrays must be set to zero.
7405 7120
7406 :: 7121 ::
7407 7122
7408 struct kvm_sync_regs { 7123 struct kvm_sync_regs {
7409 struct kvm_regs regs; 7124 struct kvm_regs regs;
7410 struct kvm_sregs sregs; 7125 struct kvm_sregs sregs;
7411 struct kvm_vcpu_events events; 7126 struct kvm_vcpu_events events;
7412 }; 7127 };
7413 7128
7414 6.75 KVM_CAP_PPC_IRQ_XIVE 7129 6.75 KVM_CAP_PPC_IRQ_XIVE
7415 ------------------------- 7130 -------------------------
7416 7131
7417 :Architectures: ppc 7132 :Architectures: ppc
7418 :Target: vcpu 7133 :Target: vcpu
7419 :Parameters: args[0] is the XIVE device fd; 7134 :Parameters: args[0] is the XIVE device fd;
7420 args[1] is the XIVE CPU number ( 7135 args[1] is the XIVE CPU number (server ID) for this vcpu
7421 7136
7422 This capability connects the vcpu to an in-ke 7137 This capability connects the vcpu to an in-kernel XIVE device.
7423 7138
7424 7. Capabilities that can be enabled on VMs 7139 7. Capabilities that can be enabled on VMs
7425 ========================================== 7140 ==========================================
7426 7141
7427 There are certain capabilities that change th 7142 There are certain capabilities that change the behavior of the virtual
7428 machine when enabled. To enable them, please 7143 machine when enabled. To enable them, please see section 4.37. Below
7429 you can find a list of capabilities and what 7144 you can find a list of capabilities and what their effect on the VM
7430 is when enabling them. 7145 is when enabling them.
7431 7146
7432 The following information is provided along w 7147 The following information is provided along with the description:
7433 7148
7434 Architectures: 7149 Architectures:
7435 which instruction set architectures pro 7150 which instruction set architectures provide this ioctl.
7436 x86 includes both i386 and x86_64. 7151 x86 includes both i386 and x86_64.
7437 7152
7438 Parameters: 7153 Parameters:
7439 what parameters are accepted by the cap 7154 what parameters are accepted by the capability.
7440 7155
7441 Returns: 7156 Returns:
7442 the return value. General error number 7157 the return value. General error numbers (EBADF, ENOMEM, EINVAL)
7443 are not detailed, but errors with speci 7158 are not detailed, but errors with specific meanings are.
7444 7159
7445 7160
7446 7.1 KVM_CAP_PPC_ENABLE_HCALL 7161 7.1 KVM_CAP_PPC_ENABLE_HCALL
7447 ---------------------------- 7162 ----------------------------
7448 7163
7449 :Architectures: ppc 7164 :Architectures: ppc
7450 :Parameters: args[0] is the sPAPR hcall numbe 7165 :Parameters: args[0] is the sPAPR hcall number;
7451 args[1] is 0 to disable, 1 to en 7166 args[1] is 0 to disable, 1 to enable in-kernel handling
7452 7167
7453 This capability controls whether individual s 7168 This capability controls whether individual sPAPR hypercalls (hcalls)
7454 get handled by the kernel or not. Enabling o 7169 get handled by the kernel or not. Enabling or disabling in-kernel
7455 handling of an hcall is effective across the 7170 handling of an hcall is effective across the VM. On creation, an
7456 initial set of hcalls are enabled for in-kern 7171 initial set of hcalls are enabled for in-kernel handling, which
7457 consists of those hcalls for which in-kernel 7172 consists of those hcalls for which in-kernel handlers were implemented
7458 before this capability was implemented. If d 7173 before this capability was implemented. If disabled, the kernel will
7459 not to attempt to handle the hcall, but will 7174 not to attempt to handle the hcall, but will always exit to userspace
7460 to handle it. Note that it may not make sens 7175 to handle it. Note that it may not make sense to enable some and
7461 disable others of a group of related hcalls, 7176 disable others of a group of related hcalls, but KVM does not prevent
7462 userspace from doing that. 7177 userspace from doing that.
7463 7178
7464 If the hcall number specified is not one that 7179 If the hcall number specified is not one that has an in-kernel
7465 implementation, the KVM_ENABLE_CAP ioctl will 7180 implementation, the KVM_ENABLE_CAP ioctl will fail with an EINVAL
7466 error. 7181 error.
7467 7182
7468 7.2 KVM_CAP_S390_USER_SIGP 7183 7.2 KVM_CAP_S390_USER_SIGP
7469 -------------------------- 7184 --------------------------
7470 7185
7471 :Architectures: s390 7186 :Architectures: s390
7472 :Parameters: none 7187 :Parameters: none
7473 7188
7474 This capability controls which SIGP orders wi 7189 This capability controls which SIGP orders will be handled completely in user
7475 space. With this capability enabled, all fast 7190 space. With this capability enabled, all fast orders will be handled completely
7476 in the kernel: 7191 in the kernel:
7477 7192
7478 - SENSE 7193 - SENSE
7479 - SENSE RUNNING 7194 - SENSE RUNNING
7480 - EXTERNAL CALL 7195 - EXTERNAL CALL
7481 - EMERGENCY SIGNAL 7196 - EMERGENCY SIGNAL
7482 - CONDITIONAL EMERGENCY SIGNAL 7197 - CONDITIONAL EMERGENCY SIGNAL
7483 7198
7484 All other orders will be handled completely i 7199 All other orders will be handled completely in user space.
7485 7200
7486 Only privileged operation exceptions will be 7201 Only privileged operation exceptions will be checked for in the kernel (or even
7487 in the hardware prior to interception). If th 7202 in the hardware prior to interception). If this capability is not enabled, the
7488 old way of handling SIGP orders is used (part 7203 old way of handling SIGP orders is used (partially in kernel and user space).
7489 7204
7490 7.3 KVM_CAP_S390_VECTOR_REGISTERS 7205 7.3 KVM_CAP_S390_VECTOR_REGISTERS
7491 --------------------------------- 7206 ---------------------------------
7492 7207
7493 :Architectures: s390 7208 :Architectures: s390
7494 :Parameters: none 7209 :Parameters: none
7495 :Returns: 0 on success, negative value on err 7210 :Returns: 0 on success, negative value on error
7496 7211
7497 Allows use of the vector registers introduced 7212 Allows use of the vector registers introduced with z13 processor, and
7498 provides for the synchronization between host 7213 provides for the synchronization between host and user space. Will
7499 return -EINVAL if the machine does not suppor 7214 return -EINVAL if the machine does not support vectors.
7500 7215
7501 7.4 KVM_CAP_S390_USER_STSI 7216 7.4 KVM_CAP_S390_USER_STSI
7502 -------------------------- 7217 --------------------------
7503 7218
7504 :Architectures: s390 7219 :Architectures: s390
7505 :Parameters: none 7220 :Parameters: none
7506 7221
7507 This capability allows post-handlers for the 7222 This capability allows post-handlers for the STSI instruction. After
7508 initial handling in the kernel, KVM exits to 7223 initial handling in the kernel, KVM exits to user space with
7509 KVM_EXIT_S390_STSI to allow user space to ins 7224 KVM_EXIT_S390_STSI to allow user space to insert further data.
7510 7225
7511 Before exiting to userspace, kvm handlers sho 7226 Before exiting to userspace, kvm handlers should fill in s390_stsi field of
7512 vcpu->run:: 7227 vcpu->run::
7513 7228
7514 struct { 7229 struct {
7515 __u64 addr; 7230 __u64 addr;
7516 __u8 ar; 7231 __u8 ar;
7517 __u8 reserved; 7232 __u8 reserved;
7518 __u8 fc; 7233 __u8 fc;
7519 __u8 sel1; 7234 __u8 sel1;
7520 __u16 sel2; 7235 __u16 sel2;
7521 } s390_stsi; 7236 } s390_stsi;
7522 7237
7523 @addr - guest address of STSI SYSIB 7238 @addr - guest address of STSI SYSIB
7524 @fc - function code 7239 @fc - function code
7525 @sel1 - selector 1 7240 @sel1 - selector 1
7526 @sel2 - selector 2 7241 @sel2 - selector 2
7527 @ar - access register number 7242 @ar - access register number
7528 7243
7529 KVM handlers should exit to userspace with rc 7244 KVM handlers should exit to userspace with rc = -EREMOTE.
7530 7245
7531 7.5 KVM_CAP_SPLIT_IRQCHIP 7246 7.5 KVM_CAP_SPLIT_IRQCHIP
7532 ------------------------- 7247 -------------------------
7533 7248
7534 :Architectures: x86 7249 :Architectures: x86
7535 :Parameters: args[0] - number of routes reser 7250 :Parameters: args[0] - number of routes reserved for userspace IOAPICs
7536 :Returns: 0 on success, -1 on error 7251 :Returns: 0 on success, -1 on error
7537 7252
7538 Create a local apic for each processor in the 7253 Create a local apic for each processor in the kernel. This can be used
7539 instead of KVM_CREATE_IRQCHIP if the userspac 7254 instead of KVM_CREATE_IRQCHIP if the userspace VMM wishes to emulate the
7540 IOAPIC and PIC (and also the PIT, even though 7255 IOAPIC and PIC (and also the PIT, even though this has to be enabled
7541 separately). 7256 separately).
7542 7257
7543 This capability also enables in kernel routin 7258 This capability also enables in kernel routing of interrupt requests;
7544 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM 7259 when KVM_CAP_SPLIT_IRQCHIP only routes of KVM_IRQ_ROUTING_MSI type are
7545 used in the IRQ routing table. The first arg 7260 used in the IRQ routing table. The first args[0] MSI routes are reserved
7546 for the IOAPIC pins. Whenever the LAPIC rece 7261 for the IOAPIC pins. Whenever the LAPIC receives an EOI for these routes,
7547 a KVM_EXIT_IOAPIC_EOI vmexit will be reported 7262 a KVM_EXIT_IOAPIC_EOI vmexit will be reported to userspace.
7548 7263
7549 Fails if VCPU has already been created, or if 7264 Fails if VCPU has already been created, or if the irqchip is already in the
7550 kernel (i.e. KVM_CREATE_IRQCHIP has already b 7265 kernel (i.e. KVM_CREATE_IRQCHIP has already been called).
7551 7266
7552 7.6 KVM_CAP_S390_RI 7267 7.6 KVM_CAP_S390_RI
7553 ------------------- 7268 -------------------
7554 7269
7555 :Architectures: s390 7270 :Architectures: s390
7556 :Parameters: none 7271 :Parameters: none
7557 7272
7558 Allows use of runtime-instrumentation introdu 7273 Allows use of runtime-instrumentation introduced with zEC12 processor.
7559 Will return -EINVAL if the machine does not s 7274 Will return -EINVAL if the machine does not support runtime-instrumentation.
7560 Will return -EBUSY if a VCPU has already been 7275 Will return -EBUSY if a VCPU has already been created.
7561 7276
7562 7.7 KVM_CAP_X2APIC_API 7277 7.7 KVM_CAP_X2APIC_API
7563 ---------------------- 7278 ----------------------
7564 7279
7565 :Architectures: x86 7280 :Architectures: x86
7566 :Parameters: args[0] - features that should b 7281 :Parameters: args[0] - features that should be enabled
7567 :Returns: 0 on success, -EINVAL when args[0] 7282 :Returns: 0 on success, -EINVAL when args[0] contains invalid features
7568 7283
7569 Valid feature flags in args[0] are:: 7284 Valid feature flags in args[0] are::
7570 7285
7571 #define KVM_X2APIC_API_USE_32BIT_IDS 7286 #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0)
7572 #define KVM_X2APIC_API_DISABLE_BROADCAST_QU 7287 #define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1)
7573 7288
7574 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes 7289 Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of
7575 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_ 7290 KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC,
7576 allowing the use of 32-bit APIC IDs. See KVM 7291 allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their
7577 respective sections. 7292 respective sections.
7578 7293
7579 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must b 7294 KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work
7580 in logical mode or with more than 255 VCPUs. 7295 in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff
7581 as a broadcast even in x2APIC mode in order t 7296 as a broadcast even in x2APIC mode in order to support physical x2APIC
7582 without interrupt remapping. This is undesir 7297 without interrupt remapping. This is undesirable in logical mode,
7583 where 0xff represents CPUs 0-7 in cluster 0. 7298 where 0xff represents CPUs 0-7 in cluster 0.
7584 7299
7585 7.8 KVM_CAP_S390_USER_INSTR0 7300 7.8 KVM_CAP_S390_USER_INSTR0
7586 ---------------------------- 7301 ----------------------------
7587 7302
7588 :Architectures: s390 7303 :Architectures: s390
7589 :Parameters: none 7304 :Parameters: none
7590 7305
7591 With this capability enabled, all illegal ins 7306 With this capability enabled, all illegal instructions 0x0000 (2 bytes) will
7592 be intercepted and forwarded to user space. U 7307 be intercepted and forwarded to user space. User space can use this
7593 mechanism e.g. to realize 2-byte software bre 7308 mechanism e.g. to realize 2-byte software breakpoints. The kernel will
7594 not inject an operating exception for these i 7309 not inject an operating exception for these instructions, user space has
7595 to take care of that. 7310 to take care of that.
7596 7311
7597 This capability can be enabled dynamically ev 7312 This capability can be enabled dynamically even if VCPUs were already
7598 created and are running. 7313 created and are running.
7599 7314
7600 7.9 KVM_CAP_S390_GS 7315 7.9 KVM_CAP_S390_GS
7601 ------------------- 7316 -------------------
7602 7317
7603 :Architectures: s390 7318 :Architectures: s390
7604 :Parameters: none 7319 :Parameters: none
7605 :Returns: 0 on success; -EINVAL if the machin 7320 :Returns: 0 on success; -EINVAL if the machine does not support
7606 guarded storage; -EBUSY if a VCPU h 7321 guarded storage; -EBUSY if a VCPU has already been created.
7607 7322
7608 Allows use of guarded storage for the KVM gue 7323 Allows use of guarded storage for the KVM guest.
7609 7324
7610 7.10 KVM_CAP_S390_AIS 7325 7.10 KVM_CAP_S390_AIS
7611 --------------------- 7326 ---------------------
7612 7327
7613 :Architectures: s390 7328 :Architectures: s390
7614 :Parameters: none 7329 :Parameters: none
7615 7330
7616 Allow use of adapter-interruption suppression 7331 Allow use of adapter-interruption suppression.
7617 :Returns: 0 on success; -EBUSY if a VCPU has 7332 :Returns: 0 on success; -EBUSY if a VCPU has already been created.
7618 7333
7619 7.11 KVM_CAP_PPC_SMT 7334 7.11 KVM_CAP_PPC_SMT
7620 -------------------- 7335 --------------------
7621 7336
7622 :Architectures: ppc 7337 :Architectures: ppc
7623 :Parameters: vsmt_mode, flags 7338 :Parameters: vsmt_mode, flags
7624 7339
7625 Enabling this capability on a VM provides use 7340 Enabling this capability on a VM provides userspace with a way to set
7626 the desired virtual SMT mode (i.e. the number 7341 the desired virtual SMT mode (i.e. the number of virtual CPUs per
7627 virtual core). The virtual SMT mode, vsmt_mo 7342 virtual core). The virtual SMT mode, vsmt_mode, must be a power of 2
7628 between 1 and 8. On POWER8, vsmt_mode must a 7343 between 1 and 8. On POWER8, vsmt_mode must also be no greater than
7629 the number of threads per subcore for the hos 7344 the number of threads per subcore for the host. Currently flags must
7630 be 0. A successful call to enable this capab 7345 be 0. A successful call to enable this capability will result in
7631 vsmt_mode being returned when the KVM_CAP_PPC 7346 vsmt_mode being returned when the KVM_CAP_PPC_SMT capability is
7632 subsequently queried for the VM. This capabi 7347 subsequently queried for the VM. This capability is only supported by
7633 HV KVM, and can only be set before any VCPUs 7348 HV KVM, and can only be set before any VCPUs have been created.
7634 The KVM_CAP_PPC_SMT_POSSIBLE capability indic 7349 The KVM_CAP_PPC_SMT_POSSIBLE capability indicates which virtual SMT
7635 modes are available. 7350 modes are available.
7636 7351
7637 7.12 KVM_CAP_PPC_FWNMI 7352 7.12 KVM_CAP_PPC_FWNMI
7638 ---------------------- 7353 ----------------------
7639 7354
7640 :Architectures: ppc 7355 :Architectures: ppc
7641 :Parameters: none 7356 :Parameters: none
7642 7357
7643 With this capability a machine check exceptio 7358 With this capability a machine check exception in the guest address
7644 space will cause KVM to exit the guest with N 7359 space will cause KVM to exit the guest with NMI exit reason. This
7645 enables QEMU to build error log and branch to 7360 enables QEMU to build error log and branch to guest kernel registered
7646 machine check handling routine. Without this 7361 machine check handling routine. Without this capability KVM will
7647 branch to guests' 0x200 interrupt vector. 7362 branch to guests' 0x200 interrupt vector.
7648 7363
7649 7.13 KVM_CAP_X86_DISABLE_EXITS 7364 7.13 KVM_CAP_X86_DISABLE_EXITS
7650 ------------------------------ 7365 ------------------------------
7651 7366
7652 :Architectures: x86 7367 :Architectures: x86
7653 :Parameters: args[0] defines which exits are 7368 :Parameters: args[0] defines which exits are disabled
7654 :Returns: 0 on success, -EINVAL when args[0] 7369 :Returns: 0 on success, -EINVAL when args[0] contains invalid exits
7655 7370
7656 Valid bits in args[0] are:: 7371 Valid bits in args[0] are::
7657 7372
7658 #define KVM_X86_DISABLE_EXITS_MWAIT 7373 #define KVM_X86_DISABLE_EXITS_MWAIT (1 << 0)
7659 #define KVM_X86_DISABLE_EXITS_HLT 7374 #define KVM_X86_DISABLE_EXITS_HLT (1 << 1)
7660 #define KVM_X86_DISABLE_EXITS_PAUSE 7375 #define KVM_X86_DISABLE_EXITS_PAUSE (1 << 2)
7661 #define KVM_X86_DISABLE_EXITS_CSTATE 7376 #define KVM_X86_DISABLE_EXITS_CSTATE (1 << 3)
7662 7377
7663 Enabling this capability on a VM provides use 7378 Enabling this capability on a VM provides userspace with a way to no
7664 longer intercept some instructions for improv 7379 longer intercept some instructions for improved latency in some
7665 workloads, and is suggested when vCPUs are as 7380 workloads, and is suggested when vCPUs are associated to dedicated
7666 physical CPUs. More bits can be added in the 7381 physical CPUs. More bits can be added in the future; userspace can
7667 just pass the KVM_CHECK_EXTENSION result to K 7382 just pass the KVM_CHECK_EXTENSION result to KVM_ENABLE_CAP to disable
7668 all such vmexits. 7383 all such vmexits.
7669 7384
7670 Do not enable KVM_FEATURE_PV_UNHALT if you di 7385 Do not enable KVM_FEATURE_PV_UNHALT if you disable HLT exits.
7671 7386
7672 7.14 KVM_CAP_S390_HPAGE_1M 7387 7.14 KVM_CAP_S390_HPAGE_1M
7673 -------------------------- 7388 --------------------------
7674 7389
7675 :Architectures: s390 7390 :Architectures: s390
7676 :Parameters: none 7391 :Parameters: none
7677 :Returns: 0 on success, -EINVAL if hpage modu 7392 :Returns: 0 on success, -EINVAL if hpage module parameter was not set
7678 or cmma is enabled, or the VM has t 7393 or cmma is enabled, or the VM has the KVM_VM_S390_UCONTROL
7679 flag set 7394 flag set
7680 7395
7681 With this capability the KVM support for memo 7396 With this capability the KVM support for memory backing with 1m pages
7682 through hugetlbfs can be enabled for a VM. Af 7397 through hugetlbfs can be enabled for a VM. After the capability is
7683 enabled, cmma can't be enabled anymore and pf 7398 enabled, cmma can't be enabled anymore and pfmfi and the storage key
7684 interpretation are disabled. If cmma has alre 7399 interpretation are disabled. If cmma has already been enabled or the
7685 hpage module parameter is not set to 1, -EINV 7400 hpage module parameter is not set to 1, -EINVAL is returned.
7686 7401
7687 While it is generally possible to create a hu 7402 While it is generally possible to create a huge page backed VM without
7688 this capability, the VM will not be able to r 7403 this capability, the VM will not be able to run.
7689 7404
7690 7.15 KVM_CAP_MSR_PLATFORM_INFO 7405 7.15 KVM_CAP_MSR_PLATFORM_INFO
7691 ------------------------------ 7406 ------------------------------
7692 7407
7693 :Architectures: x86 7408 :Architectures: x86
7694 :Parameters: args[0] whether feature should b 7409 :Parameters: args[0] whether feature should be enabled or not
7695 7410
7696 With this capability, a guest may read the MS 7411 With this capability, a guest may read the MSR_PLATFORM_INFO MSR. Otherwise,
7697 a #GP would be raised when the guest tries to 7412 a #GP would be raised when the guest tries to access. Currently, this
7698 capability does not enable write permissions 7413 capability does not enable write permissions of this MSR for the guest.
7699 7414
7700 7.16 KVM_CAP_PPC_NESTED_HV 7415 7.16 KVM_CAP_PPC_NESTED_HV
7701 -------------------------- 7416 --------------------------
7702 7417
7703 :Architectures: ppc 7418 :Architectures: ppc
7704 :Parameters: none 7419 :Parameters: none
7705 :Returns: 0 on success, -EINVAL when the impl 7420 :Returns: 0 on success, -EINVAL when the implementation doesn't support
7706 nested-HV virtualization. 7421 nested-HV virtualization.
7707 7422
7708 HV-KVM on POWER9 and later systems allows for 7423 HV-KVM on POWER9 and later systems allows for "nested-HV"
7709 virtualization, which provides a way for a gu 7424 virtualization, which provides a way for a guest VM to run guests that
7710 can run using the CPU's supervisor mode (priv 7425 can run using the CPU's supervisor mode (privileged non-hypervisor
7711 state). Enabling this capability on a VM dep 7426 state). Enabling this capability on a VM depends on the CPU having
7712 the necessary functionality and on the facili 7427 the necessary functionality and on the facility being enabled with a
7713 kvm-hv module parameter. 7428 kvm-hv module parameter.
7714 7429
7715 7.17 KVM_CAP_EXCEPTION_PAYLOAD 7430 7.17 KVM_CAP_EXCEPTION_PAYLOAD
7716 ------------------------------ 7431 ------------------------------
7717 7432
7718 :Architectures: x86 7433 :Architectures: x86
7719 :Parameters: args[0] whether feature should b 7434 :Parameters: args[0] whether feature should be enabled or not
7720 7435
7721 With this capability enabled, CR2 will not be 7436 With this capability enabled, CR2 will not be modified prior to the
7722 emulated VM-exit when L1 intercepts a #PF exc 7437 emulated VM-exit when L1 intercepts a #PF exception that occurs in
7723 L2. Similarly, for kvm-intel only, DR6 will n 7438 L2. Similarly, for kvm-intel only, DR6 will not be modified prior to
7724 the emulated VM-exit when L1 intercepts a #DB 7439 the emulated VM-exit when L1 intercepts a #DB exception that occurs in
7725 L2. As a result, when KVM_GET_VCPU_EVENTS rep 7440 L2. As a result, when KVM_GET_VCPU_EVENTS reports a pending #PF (or
7726 #DB) exception for L2, exception.has_payload 7441 #DB) exception for L2, exception.has_payload will be set and the
7727 faulting address (or the new DR6 bits*) will 7442 faulting address (or the new DR6 bits*) will be reported in the
7728 exception_payload field. Similarly, when user 7443 exception_payload field. Similarly, when userspace injects a #PF (or
7729 #DB) into L2 using KVM_SET_VCPU_EVENTS, it is 7444 #DB) into L2 using KVM_SET_VCPU_EVENTS, it is expected to set
7730 exception.has_payload and to put the faulting 7445 exception.has_payload and to put the faulting address - or the new DR6
7731 bits\ [#]_ - in the exception_payload field. 7446 bits\ [#]_ - in the exception_payload field.
7732 7447
7733 This capability also enables exception.pendin 7448 This capability also enables exception.pending in struct
7734 kvm_vcpu_events, which allows userspace to di 7449 kvm_vcpu_events, which allows userspace to distinguish between pending
7735 and injected exceptions. 7450 and injected exceptions.
7736 7451
7737 7452
7738 .. [#] For the new DR6 bits, note that bit 16 7453 .. [#] For the new DR6 bits, note that bit 16 is set iff the #DB exception
7739 will clear DR6.RTM. 7454 will clear DR6.RTM.
7740 7455
7741 7.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 7456 7.18 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2
7742 -------------------------------------- 7457 --------------------------------------
7743 7458
7744 :Architectures: x86, arm64, mips 7459 :Architectures: x86, arm64, mips
7745 :Parameters: args[0] whether feature should b 7460 :Parameters: args[0] whether feature should be enabled or not
7746 7461
7747 Valid flags are:: 7462 Valid flags are::
7748 7463
7749 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE 7464 #define KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (1 << 0)
7750 #define KVM_DIRTY_LOG_INITIALLY_SET 7465 #define KVM_DIRTY_LOG_INITIALLY_SET (1 << 1)
7751 7466
7752 With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is s 7467 With KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE is set, KVM_GET_DIRTY_LOG will not
7753 automatically clear and write-protect all pag 7468 automatically clear and write-protect all pages that are returned as dirty.
7754 Rather, userspace will have to do this operat 7469 Rather, userspace will have to do this operation separately using
7755 KVM_CLEAR_DIRTY_LOG. 7470 KVM_CLEAR_DIRTY_LOG.
7756 7471
7757 At the cost of a slightly more complicated op 7472 At the cost of a slightly more complicated operation, this provides better
7758 scalability and responsiveness for two reason 7473 scalability and responsiveness for two reasons. First,
7759 KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64 7474 KVM_CLEAR_DIRTY_LOG ioctl can operate on a 64-page granularity rather
7760 than requiring to sync a full memslot; this e 7475 than requiring to sync a full memslot; this ensures that KVM does not
7761 take spinlocks for an extended period of time 7476 take spinlocks for an extended period of time. Second, in some cases a
7762 large amount of time can pass between a call 7477 large amount of time can pass between a call to KVM_GET_DIRTY_LOG and
7763 userspace actually using the data in the page 7478 userspace actually using the data in the page. Pages can be modified
7764 during this time, which is inefficient for bo 7479 during this time, which is inefficient for both the guest and userspace:
7765 the guest will incur a higher penalty due to 7480 the guest will incur a higher penalty due to write protection faults,
7766 while userspace can see false reports of dirt 7481 while userspace can see false reports of dirty pages. Manual reprotection
7767 helps reducing this time, improving guest per 7482 helps reducing this time, improving guest performance and reducing the
7768 number of dirty log false positives. 7483 number of dirty log false positives.
7769 7484
7770 With KVM_DIRTY_LOG_INITIALLY_SET set, all the 7485 With KVM_DIRTY_LOG_INITIALLY_SET set, all the bits of the dirty bitmap
7771 will be initialized to 1 when created. This 7486 will be initialized to 1 when created. This also improves performance because
7772 dirty logging can be enabled gradually in sma 7487 dirty logging can be enabled gradually in small chunks on the first call
7773 to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIA 7488 to KVM_CLEAR_DIRTY_LOG. KVM_DIRTY_LOG_INITIALLY_SET depends on
7774 KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is al 7489 KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE (it is also only available on
7775 x86 and arm64 for now). 7490 x86 and arm64 for now).
7776 7491
7777 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previou 7492 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 was previously available under the name
7778 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the imp 7493 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT, but the implementation had bugs that make
7779 it hard or impossible to use it correctly. T 7494 it hard or impossible to use it correctly. The availability of
7780 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals tha 7495 KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 signals that those bugs are fixed.
7781 Userspace should not try to use KVM_CAP_MANUA 7496 Userspace should not try to use KVM_CAP_MANUAL_DIRTY_LOG_PROTECT.
7782 7497
7783 7.19 KVM_CAP_PPC_SECURE_GUEST 7498 7.19 KVM_CAP_PPC_SECURE_GUEST
7784 ------------------------------ 7499 ------------------------------
7785 7500
7786 :Architectures: ppc 7501 :Architectures: ppc
7787 7502
7788 This capability indicates that KVM is running 7503 This capability indicates that KVM is running on a host that has
7789 ultravisor firmware and thus can support a se 7504 ultravisor firmware and thus can support a secure guest. On such a
7790 system, a guest can ask the ultravisor to mak 7505 system, a guest can ask the ultravisor to make it a secure guest,
7791 one whose memory is inaccessible to the host 7506 one whose memory is inaccessible to the host except for pages which
7792 are explicitly requested to be shared with th 7507 are explicitly requested to be shared with the host. The ultravisor
7793 notifies KVM when a guest requests to become 7508 notifies KVM when a guest requests to become a secure guest, and KVM
7794 has the opportunity to veto the transition. 7509 has the opportunity to veto the transition.
7795 7510
7796 If present, this capability can be enabled fo 7511 If present, this capability can be enabled for a VM, meaning that KVM
7797 will allow the transition to secure guest mod 7512 will allow the transition to secure guest mode. Otherwise KVM will
7798 veto the transition. 7513 veto the transition.
7799 7514
7800 7.20 KVM_CAP_HALT_POLL 7515 7.20 KVM_CAP_HALT_POLL
7801 ---------------------- 7516 ----------------------
7802 7517
7803 :Architectures: all 7518 :Architectures: all
7804 :Target: VM 7519 :Target: VM
7805 :Parameters: args[0] is the maximum poll time 7520 :Parameters: args[0] is the maximum poll time in nanoseconds
7806 :Returns: 0 on success; -1 on error 7521 :Returns: 0 on success; -1 on error
7807 7522
7808 KVM_CAP_HALT_POLL overrides the kvm.halt_poll 7523 KVM_CAP_HALT_POLL overrides the kvm.halt_poll_ns module parameter to set the
7809 maximum halt-polling time for all vCPUs in th 7524 maximum halt-polling time for all vCPUs in the target VM. This capability can
7810 be invoked at any time and any number of time 7525 be invoked at any time and any number of times to dynamically change the
7811 maximum halt-polling time. 7526 maximum halt-polling time.
7812 7527
7813 See Documentation/virt/kvm/halt-polling.rst f 7528 See Documentation/virt/kvm/halt-polling.rst for more information on halt
7814 polling. 7529 polling.
7815 7530
7816 7.21 KVM_CAP_X86_USER_SPACE_MSR 7531 7.21 KVM_CAP_X86_USER_SPACE_MSR
7817 ------------------------------- 7532 -------------------------------
7818 7533
7819 :Architectures: x86 7534 :Architectures: x86
7820 :Target: VM 7535 :Target: VM
7821 :Parameters: args[0] contains the mask of KVM 7536 :Parameters: args[0] contains the mask of KVM_MSR_EXIT_REASON_* events to report
7822 :Returns: 0 on success; -1 on error 7537 :Returns: 0 on success; -1 on error
7823 7538
7824 This capability allows userspace to intercept 7539 This capability allows userspace to intercept RDMSR and WRMSR instructions if
7825 access to an MSR is denied. By default, KVM 7540 access to an MSR is denied. By default, KVM injects #GP on denied accesses.
7826 7541
7827 When a guest requests to read or write an MSR 7542 When a guest requests to read or write an MSR, KVM may not implement all MSRs
7828 that are relevant to a respective system. It 7543 that are relevant to a respective system. It also does not differentiate by
7829 CPU type. 7544 CPU type.
7830 7545
7831 To allow more fine grained control over MSR h 7546 To allow more fine grained control over MSR handling, userspace may enable
7832 this capability. With it enabled, MSR accesse 7547 this capability. With it enabled, MSR accesses that match the mask specified in
7833 args[0] and would trigger a #GP inside the gu 7548 args[0] and would trigger a #GP inside the guest will instead trigger
7834 KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exi 7549 KVM_EXIT_X86_RDMSR and KVM_EXIT_X86_WRMSR exit notifications. Userspace
7835 can then implement model specific MSR handlin 7550 can then implement model specific MSR handling and/or user notifications
7836 to inform a user that an MSR was not emulated 7551 to inform a user that an MSR was not emulated/virtualized by KVM.
7837 7552
7838 The valid mask flags are: 7553 The valid mask flags are:
7839 7554
7840 ============================ ================ 7555 ============================ ===============================================
7841 KVM_MSR_EXIT_REASON_UNKNOWN intercept access 7556 KVM_MSR_EXIT_REASON_UNKNOWN intercept accesses to unknown (to KVM) MSRs
7842 KVM_MSR_EXIT_REASON_INVAL intercept access 7557 KVM_MSR_EXIT_REASON_INVAL intercept accesses that are architecturally
7843 invalid accordin 7558 invalid according to the vCPU model and/or mode
7844 KVM_MSR_EXIT_REASON_FILTER intercept access 7559 KVM_MSR_EXIT_REASON_FILTER intercept accesses that are denied by userspace
7845 via KVM_X86_SET_ 7560 via KVM_X86_SET_MSR_FILTER
7846 ============================ ================ 7561 ============================ ===============================================
7847 7562
7848 7.22 KVM_CAP_X86_BUS_LOCK_EXIT 7563 7.22 KVM_CAP_X86_BUS_LOCK_EXIT
7849 ------------------------------- 7564 -------------------------------
7850 7565
7851 :Architectures: x86 7566 :Architectures: x86
7852 :Target: VM 7567 :Target: VM
7853 :Parameters: args[0] defines the policy used 7568 :Parameters: args[0] defines the policy used when bus locks detected in guest
7854 :Returns: 0 on success, -EINVAL when args[0] 7569 :Returns: 0 on success, -EINVAL when args[0] contains invalid bits
7855 7570
7856 Valid bits in args[0] are:: 7571 Valid bits in args[0] are::
7857 7572
7858 #define KVM_BUS_LOCK_DETECTION_OFF (1 7573 #define KVM_BUS_LOCK_DETECTION_OFF (1 << 0)
7859 #define KVM_BUS_LOCK_DETECTION_EXIT (1 7574 #define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1)
7860 7575
7861 Enabling this capability on a VM provides use !! 7576 Enabling this capability on a VM provides userspace with a way to select
7862 policy to handle the bus locks detected in gu !! 7577 a policy to handle the bus locks detected in guest. Userspace can obtain
7863 supported modes from the result of KVM_CHECK_ !! 7578 the supported modes from the result of KVM_CHECK_EXTENSION and define it
7864 the KVM_ENABLE_CAP. The supported modes are m !! 7579 through the KVM_ENABLE_CAP.
7865 !! 7580
7866 This capability allows userspace to force VM !! 7581 KVM_BUS_LOCK_DETECTION_OFF and KVM_BUS_LOCK_DETECTION_EXIT are supported
7867 guest, irrespective whether or not the host h !! 7582 currently and mutually exclusive with each other. More bits can be added in
7868 (which triggers an #AC exception that KVM int !! 7583 the future.
7869 intended to mitigate attacks where a maliciou !! 7584
7870 locks to degrade the performance of the whole !! 7585 With KVM_BUS_LOCK_DETECTION_OFF set, bus locks in guest will not cause vm exits
7871 !! 7586 so that no additional actions are needed. This is the default mode.
7872 If KVM_BUS_LOCK_DETECTION_OFF is set, KVM doe !! 7587
7873 exit, although the host kernel's split-lock # !! 7588 With KVM_BUS_LOCK_DETECTION_EXIT set, vm exits happen when bus lock detected
7874 enabled. !! 7589 in VM. KVM just exits to userspace when handling them. Userspace can enforce
7875 !! 7590 its own throttling or other policy based mitigations.
7876 If KVM_BUS_LOCK_DETECTION_EXIT is set, KVM en !! 7591
7877 bus locks in the guest trigger a VM exit, and !! 7592 This capability is aimed to address the thread that VM can exploit bus locks to
7878 such VM exits, e.g. to allow userspace to thr !! 7593 degree the performance of the whole system. Once the userspace enable this
7879 apply some other policy-based mitigation. Whe !! 7594 capability and select the KVM_BUS_LOCK_DETECTION_EXIT mode, KVM will set the
7880 KVM_RUN_X86_BUS_LOCK in vcpu-run->flags, and !! 7595 KVM_RUN_BUS_LOCK flag in vcpu-run->flags field and exit to userspace. Concerning
7881 to KVM_EXIT_X86_BUS_LOCK. !! 7596 the bus lock vm exit can be preempted by a higher priority VM exit, the exit
7882 !! 7597 notifications to userspace can be KVM_EXIT_BUS_LOCK or other reasons.
7883 Note! Detected bus locks may be coincident wi !! 7598 KVM_RUN_BUS_LOCK flag is used to distinguish between them.
7884 KVM_RUN_X86_BUS_LOCK should be checked regard <<
7885 userspace wants to take action on all detecte <<
7886 7599
7887 7.23 KVM_CAP_PPC_DAWR1 7600 7.23 KVM_CAP_PPC_DAWR1
7888 ---------------------- 7601 ----------------------
7889 7602
7890 :Architectures: ppc 7603 :Architectures: ppc
7891 :Parameters: none 7604 :Parameters: none
7892 :Returns: 0 on success, -EINVAL when CPU does 7605 :Returns: 0 on success, -EINVAL when CPU doesn't support 2nd DAWR
7893 7606
7894 This capability can be used to check / enable 7607 This capability can be used to check / enable 2nd DAWR feature provided
7895 by POWER10 processor. 7608 by POWER10 processor.
7896 7609
7897 7610
7898 7.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM 7611 7.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM
7899 ------------------------------------- 7612 -------------------------------------
7900 7613
7901 Architectures: x86 SEV enabled 7614 Architectures: x86 SEV enabled
7902 Type: vm 7615 Type: vm
7903 Parameters: args[0] is the fd of the source v 7616 Parameters: args[0] is the fd of the source vm
7904 Returns: 0 on success; ENOTTY on error 7617 Returns: 0 on success; ENOTTY on error
7905 7618
7906 This capability enables userspace to copy enc 7619 This capability enables userspace to copy encryption context from the vm
7907 indicated by the fd to the vm this is called 7620 indicated by the fd to the vm this is called on.
7908 7621
7909 This is intended to support in-guest workload 7622 This is intended to support in-guest workloads scheduled by the host. This
7910 allows the in-guest workload to maintain its 7623 allows the in-guest workload to maintain its own NPTs and keeps the two vms
7911 from accidentally clobbering each other with 7624 from accidentally clobbering each other with interrupts and the like (separate
7912 APIC/MSRs/etc). 7625 APIC/MSRs/etc).
7913 7626
7914 7.25 KVM_CAP_SGX_ATTRIBUTE 7627 7.25 KVM_CAP_SGX_ATTRIBUTE
7915 -------------------------- 7628 --------------------------
7916 7629
7917 :Architectures: x86 7630 :Architectures: x86
7918 :Target: VM 7631 :Target: VM
7919 :Parameters: args[0] is a file handle of a SG 7632 :Parameters: args[0] is a file handle of a SGX attribute file in securityfs
7920 :Returns: 0 on success, -EINVAL if the file h 7633 :Returns: 0 on success, -EINVAL if the file handle is invalid or if a requested
7921 attribute is not supported by KVM. 7634 attribute is not supported by KVM.
7922 7635
7923 KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM 7636 KVM_CAP_SGX_ATTRIBUTE enables a userspace VMM to grant a VM access to one or
7924 more privileged enclave attributes. args[0] 7637 more privileged enclave attributes. args[0] must hold a file handle to a valid
7925 SGX attribute file corresponding to an attrib 7638 SGX attribute file corresponding to an attribute that is supported/restricted
7926 by KVM (currently only PROVISIONKEY). 7639 by KVM (currently only PROVISIONKEY).
7927 7640
7928 The SGX subsystem restricts access to a subse 7641 The SGX subsystem restricts access to a subset of enclave attributes to provide
7929 additional security for an uncompromised kern 7642 additional security for an uncompromised kernel, e.g. use of the PROVISIONKEY
7930 is restricted to deter malware from using the 7643 is restricted to deter malware from using the PROVISIONKEY to obtain a stable
7931 system fingerprint. To prevent userspace fro 7644 system fingerprint. To prevent userspace from circumventing such restrictions
7932 by running an enclave in a VM, KVM prevents a 7645 by running an enclave in a VM, KVM prevents access to privileged attributes by
7933 default. 7646 default.
7934 7647
7935 See Documentation/arch/x86/sgx.rst for more d 7648 See Documentation/arch/x86/sgx.rst for more details.
7936 7649
7937 7.26 KVM_CAP_PPC_RPT_INVALIDATE 7650 7.26 KVM_CAP_PPC_RPT_INVALIDATE
7938 ------------------------------- 7651 -------------------------------
7939 7652
7940 :Capability: KVM_CAP_PPC_RPT_INVALIDATE 7653 :Capability: KVM_CAP_PPC_RPT_INVALIDATE
7941 :Architectures: ppc 7654 :Architectures: ppc
7942 :Type: vm 7655 :Type: vm
7943 7656
7944 This capability indicates that the kernel is 7657 This capability indicates that the kernel is capable of handling
7945 H_RPT_INVALIDATE hcall. 7658 H_RPT_INVALIDATE hcall.
7946 7659
7947 In order to enable the use of H_RPT_INVALIDAT 7660 In order to enable the use of H_RPT_INVALIDATE in the guest,
7948 user space might have to advertise it for the 7661 user space might have to advertise it for the guest. For example,
7949 IBM pSeries (sPAPR) guest starts using it if 7662 IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is
7950 present in the "ibm,hypertas-functions" devic 7663 present in the "ibm,hypertas-functions" device-tree property.
7951 7664
7952 This capability is enabled for hypervisors on 7665 This capability is enabled for hypervisors on platforms like POWER9
7953 that support radix MMU. 7666 that support radix MMU.
7954 7667
7955 7.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE 7668 7.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE
7956 -------------------------------------- 7669 --------------------------------------
7957 7670
7958 :Architectures: x86 7671 :Architectures: x86
7959 :Parameters: args[0] whether the feature shou 7672 :Parameters: args[0] whether the feature should be enabled or not
7960 7673
7961 When this capability is enabled, an emulation 7674 When this capability is enabled, an emulation failure will result in an exit
7962 to userspace with KVM_INTERNAL_ERROR (except 7675 to userspace with KVM_INTERNAL_ERROR (except when the emulator was invoked
7963 to handle a VMware backdoor instruction). Fur 7676 to handle a VMware backdoor instruction). Furthermore, KVM will now provide up
7964 to 15 instruction bytes for any exit to users 7677 to 15 instruction bytes for any exit to userspace resulting from an emulation
7965 failure. When these exits to userspace occur 7678 failure. When these exits to userspace occur use the emulation_failure struct
7966 instead of the internal struct. They both ha 7679 instead of the internal struct. They both have the same layout, but the
7967 emulation_failure struct matches the content 7680 emulation_failure struct matches the content better. It also explicitly
7968 defines the 'flags' field which is used to de 7681 defines the 'flags' field which is used to describe the fields in the struct
7969 that are valid (ie: if KVM_INTERNAL_ERROR_EMU 7682 that are valid (ie: if KVM_INTERNAL_ERROR_EMULATION_FLAG_INSTRUCTION_BYTES is
7970 set in the 'flags' field then both 'insn_size 7683 set in the 'flags' field then both 'insn_size' and 'insn_bytes' have valid data
7971 in them.) 7684 in them.)
7972 7685
7973 7.28 KVM_CAP_ARM_MTE 7686 7.28 KVM_CAP_ARM_MTE
7974 -------------------- 7687 --------------------
7975 7688
7976 :Architectures: arm64 7689 :Architectures: arm64
7977 :Parameters: none 7690 :Parameters: none
7978 7691
7979 This capability indicates that KVM (and the h 7692 This capability indicates that KVM (and the hardware) supports exposing the
7980 Memory Tagging Extensions (MTE) to the guest. 7693 Memory Tagging Extensions (MTE) to the guest. It must also be enabled by the
7981 VMM before creating any VCPUs to allow the gu 7694 VMM before creating any VCPUs to allow the guest access. Note that MTE is only
7982 available to a guest running in AArch64 mode 7695 available to a guest running in AArch64 mode and enabling this capability will
7983 cause attempts to create AArch32 VCPUs to fai 7696 cause attempts to create AArch32 VCPUs to fail.
7984 7697
7985 When enabled the guest is able to access tags 7698 When enabled the guest is able to access tags associated with any memory given
7986 to the guest. KVM will ensure that the tags a 7699 to the guest. KVM will ensure that the tags are maintained during swap or
7987 hibernation of the host; however the VMM need 7700 hibernation of the host; however the VMM needs to manually save/restore the
7988 tags as appropriate if the VM is migrated. 7701 tags as appropriate if the VM is migrated.
7989 7702
7990 When this capability is enabled all memory in 7703 When this capability is enabled all memory in memslots must be mapped as
7991 ``MAP_ANONYMOUS`` or with a RAM-based file ma 7704 ``MAP_ANONYMOUS`` or with a RAM-based file mapping (``tmpfs``, ``memfd``),
7992 attempts to create a memslot with an invalid 7705 attempts to create a memslot with an invalid mmap will result in an
7993 -EINVAL return. 7706 -EINVAL return.
7994 7707
7995 When enabled the VMM may make use of the ``KV 7708 When enabled the VMM may make use of the ``KVM_ARM_MTE_COPY_TAGS`` ioctl to
7996 perform a bulk copy of tags to/from the guest 7709 perform a bulk copy of tags to/from the guest.
7997 7710
7998 7.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 7711 7.29 KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM
7999 ------------------------------------- 7712 -------------------------------------
8000 7713
8001 :Architectures: x86 SEV enabled !! 7714 Architectures: x86 SEV enabled
8002 :Type: vm !! 7715 Type: vm
8003 :Parameters: args[0] is the fd of the source !! 7716 Parameters: args[0] is the fd of the source vm
8004 :Returns: 0 on success !! 7717 Returns: 0 on success
8005 7718
8006 This capability enables userspace to migrate 7719 This capability enables userspace to migrate the encryption context from the VM
8007 indicated by the fd to the VM this is called 7720 indicated by the fd to the VM this is called on.
8008 7721
8009 This is intended to support intra-host migrat 7722 This is intended to support intra-host migration of VMs between userspace VMMs,
8010 upgrading the VMM process without interruptin 7723 upgrading the VMM process without interrupting the guest.
8011 7724
8012 7.30 KVM_CAP_PPC_AIL_MODE_3 7725 7.30 KVM_CAP_PPC_AIL_MODE_3
8013 ------------------------------- 7726 -------------------------------
8014 7727
8015 :Capability: KVM_CAP_PPC_AIL_MODE_3 7728 :Capability: KVM_CAP_PPC_AIL_MODE_3
8016 :Architectures: ppc 7729 :Architectures: ppc
8017 :Type: vm 7730 :Type: vm
8018 7731
8019 This capability indicates that the kernel sup 7732 This capability indicates that the kernel supports the mode 3 setting for the
8020 "Address Translation Mode on Interrupt" aka " 7733 "Address Translation Mode on Interrupt" aka "Alternate Interrupt Location"
8021 resource that is controlled with the H_SET_MO 7734 resource that is controlled with the H_SET_MODE hypercall.
8022 7735
8023 This capability allows a guest kernel to use 7736 This capability allows a guest kernel to use a better-performance mode for
8024 handling interrupts and system calls. 7737 handling interrupts and system calls.
8025 7738
8026 7.31 KVM_CAP_DISABLE_QUIRKS2 7739 7.31 KVM_CAP_DISABLE_QUIRKS2
8027 ---------------------------- 7740 ----------------------------
8028 7741
8029 :Capability: KVM_CAP_DISABLE_QUIRKS2 7742 :Capability: KVM_CAP_DISABLE_QUIRKS2
8030 :Parameters: args[0] - set of KVM quirks to d 7743 :Parameters: args[0] - set of KVM quirks to disable
8031 :Architectures: x86 7744 :Architectures: x86
8032 :Type: vm 7745 :Type: vm
8033 7746
8034 This capability, if enabled, will cause KVM t 7747 This capability, if enabled, will cause KVM to disable some behavior
8035 quirks. 7748 quirks.
8036 7749
8037 Calling KVM_CHECK_EXTENSION for this capabili 7750 Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of
8038 quirks that can be disabled in KVM. 7751 quirks that can be disabled in KVM.
8039 7752
8040 The argument to KVM_ENABLE_CAP for this capab 7753 The argument to KVM_ENABLE_CAP for this capability is a bitmask of
8041 quirks to disable, and must be a subset of th 7754 quirks to disable, and must be a subset of the bitmask returned by
8042 KVM_CHECK_EXTENSION. 7755 KVM_CHECK_EXTENSION.
8043 7756
8044 The valid bits in cap.args[0] are: 7757 The valid bits in cap.args[0] are:
8045 7758
8046 =================================== ========= 7759 =================================== ============================================
8047 KVM_X86_QUIRK_LINT0_REENABLED By defaul 7760 KVM_X86_QUIRK_LINT0_REENABLED By default, the reset value for the LVT
8048 LINT0 reg 7761 LINT0 register is 0x700 (APIC_MODE_EXTINT).
8049 When this 7762 When this quirk is disabled, the reset value
8050 is 0x1000 7763 is 0x10000 (APIC_LVT_MASKED).
8051 7764
8052 KVM_X86_QUIRK_CD_NW_CLEARED By defaul !! 7765 KVM_X86_QUIRK_CD_NW_CLEARED By default, KVM clears CR0.CD and CR0.NW.
8053 AMD CPUs <<
8054 that runs <<
8055 with cach <<
8056 <<
8057 When this 7766 When this quirk is disabled, KVM does not
8058 change th 7767 change the value of CR0.CD and CR0.NW.
8059 7768
8060 KVM_X86_QUIRK_LAPIC_MMIO_HOLE By defaul 7769 KVM_X86_QUIRK_LAPIC_MMIO_HOLE By default, the MMIO LAPIC interface is
8061 available 7770 available even when configured for x2APIC
8062 mode. Whe 7771 mode. When this quirk is disabled, KVM
8063 disables 7772 disables the MMIO LAPIC interface if the
8064 LAPIC is 7773 LAPIC is in x2APIC mode.
8065 7774
8066 KVM_X86_QUIRK_OUT_7E_INC_RIP By defaul 7775 KVM_X86_QUIRK_OUT_7E_INC_RIP By default, KVM pre-increments %rip before
8067 exiting t 7776 exiting to userspace for an OUT instruction
8068 to port 0 7777 to port 0x7e. When this quirk is disabled,
8069 KVM does 7778 KVM does not pre-increment %rip before
8070 exiting t 7779 exiting to userspace.
8071 7780
8072 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this 7781 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT When this quirk is disabled, KVM sets
8073 CPUID.01H 7782 CPUID.01H:ECX[bit 3] (MONITOR/MWAIT) if
8074 IA32_MISC 7783 IA32_MISC_ENABLE[bit 18] (MWAIT) is set.
8075 Additiona 7784 Additionally, when this quirk is disabled,
8076 KVM clear 7785 KVM clears CPUID.01H:ECX[bit 3] if
8077 IA32_MISC 7786 IA32_MISC_ENABLE[bit 18] is cleared.
8078 7787
8079 KVM_X86_QUIRK_FIX_HYPERCALL_INSN By defaul 7788 KVM_X86_QUIRK_FIX_HYPERCALL_INSN By default, KVM rewrites guest
8080 VMMCALL/V 7789 VMMCALL/VMCALL instructions to match the
8081 vendor's 7790 vendor's hypercall instruction for the
8082 system. W 7791 system. When this quirk is disabled, KVM
8083 will no l 7792 will no longer rewrite invalid guest
8084 hypercall 7793 hypercall instructions. Executing the
8085 incorrect 7794 incorrect hypercall instruction will
8086 generate 7795 generate a #UD within the guest.
8087 7796
8088 KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By defaul 7797 KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By default, KVM emulates MONITOR/MWAIT (if
8089 they are 7798 they are intercepted) as NOPs regardless of
8090 whether o 7799 whether or not MONITOR/MWAIT are supported
8091 according 7800 according to guest CPUID. When this quirk
8092 is disabl 7801 is disabled and KVM_X86_DISABLE_EXITS_MWAIT
8093 is not se 7802 is not set (MONITOR/MWAIT are intercepted),
8094 KVM will 7803 KVM will inject a #UD on MONITOR/MWAIT if
8095 they're u 7804 they're unsupported per guest CPUID. Note,
8096 KVM will 7805 KVM will modify MONITOR/MWAIT support in
8097 guest CPU 7806 guest CPUID on writes to MISC_ENABLE if
8098 KVM_X86_Q 7807 KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT is
8099 disabled. 7808 disabled.
8100 <<
8101 KVM_X86_QUIRK_SLOT_ZAP_ALL By defaul <<
8102 invalidat <<
8103 address s <<
8104 moved. W <<
8105 VM type i <<
8106 ensures t <<
8107 or moved <<
8108 _may_ inv <<
8109 memslot. <<
8110 =================================== ========= 7809 =================================== ============================================
8111 7810
8112 7.32 KVM_CAP_MAX_VCPU_ID 7811 7.32 KVM_CAP_MAX_VCPU_ID
8113 ------------------------ 7812 ------------------------
8114 7813
8115 :Architectures: x86 7814 :Architectures: x86
8116 :Target: VM 7815 :Target: VM
8117 :Parameters: args[0] - maximum APIC ID value 7816 :Parameters: args[0] - maximum APIC ID value set for current VM
8118 :Returns: 0 on success, -EINVAL if args[0] is 7817 :Returns: 0 on success, -EINVAL if args[0] is beyond KVM_MAX_VCPU_IDS
8119 supported in KVM or if it has been 7818 supported in KVM or if it has been set.
8120 7819
8121 This capability allows userspace to specify m 7820 This capability allows userspace to specify maximum possible APIC ID
8122 assigned for current VM session prior to the 7821 assigned for current VM session prior to the creation of vCPUs, saving
8123 memory for data structures indexed by the API 7822 memory for data structures indexed by the APIC ID. Userspace is able
8124 to calculate the limit to APIC ID values from 7823 to calculate the limit to APIC ID values from designated
8125 CPU topology. 7824 CPU topology.
8126 7825
8127 The value can be changed only until KVM_ENABL 7826 The value can be changed only until KVM_ENABLE_CAP is set to a nonzero
8128 value or until a vCPU is created. Upon creat 7827 value or until a vCPU is created. Upon creation of the first vCPU,
8129 if the value was set to zero or KVM_ENABLE_CA 7828 if the value was set to zero or KVM_ENABLE_CAP was not invoked, KVM
8130 uses the return value of KVM_CHECK_EXTENSION( 7829 uses the return value of KVM_CHECK_EXTENSION(KVM_CAP_MAX_VCPU_ID) as
8131 the maximum APIC ID. 7830 the maximum APIC ID.
8132 7831
8133 7.33 KVM_CAP_X86_NOTIFY_VMEXIT 7832 7.33 KVM_CAP_X86_NOTIFY_VMEXIT
8134 ------------------------------ 7833 ------------------------------
8135 7834
8136 :Architectures: x86 7835 :Architectures: x86
8137 :Target: VM 7836 :Target: VM
8138 :Parameters: args[0] is the value of notify w 7837 :Parameters: args[0] is the value of notify window as well as some flags
8139 :Returns: 0 on success, -EINVAL if args[0] co 7838 :Returns: 0 on success, -EINVAL if args[0] contains invalid flags or notify
8140 VM exit is unsupported. 7839 VM exit is unsupported.
8141 7840
8142 Bits 63:32 of args[0] are used for notify win 7841 Bits 63:32 of args[0] are used for notify window.
8143 Bits 31:0 of args[0] are for some flags. Vali 7842 Bits 31:0 of args[0] are for some flags. Valid bits are::
8144 7843
8145 #define KVM_X86_NOTIFY_VMEXIT_ENABLED (1 7844 #define KVM_X86_NOTIFY_VMEXIT_ENABLED (1 << 0)
8146 #define KVM_X86_NOTIFY_VMEXIT_USER (1 7845 #define KVM_X86_NOTIFY_VMEXIT_USER (1 << 1)
8147 7846
8148 This capability allows userspace to configure 7847 This capability allows userspace to configure the notify VM exit on/off
8149 in per-VM scope during VM creation. Notify VM 7848 in per-VM scope during VM creation. Notify VM exit is disabled by default.
8150 When userspace sets KVM_X86_NOTIFY_VMEXIT_ENA 7849 When userspace sets KVM_X86_NOTIFY_VMEXIT_ENABLED bit in args[0], VMM will
8151 enable this feature with the notify window pr 7850 enable this feature with the notify window provided, which will generate
8152 a VM exit if no event window occurs in VM non 7851 a VM exit if no event window occurs in VM non-root mode for a specified of
8153 time (notify window). 7852 time (notify window).
8154 7853
8155 If KVM_X86_NOTIFY_VMEXIT_USER is set in args[ 7854 If KVM_X86_NOTIFY_VMEXIT_USER is set in args[0], upon notify VM exits happen,
8156 KVM would exit to userspace for handling. 7855 KVM would exit to userspace for handling.
8157 7856
8158 This capability is aimed to mitigate the thre 7857 This capability is aimed to mitigate the threat that malicious VMs can
8159 cause CPU stuck (due to event windows don't o 7858 cause CPU stuck (due to event windows don't open up) and make the CPU
8160 unavailable to host or other VMs. 7859 unavailable to host or other VMs.
8161 7860
8162 7.34 KVM_CAP_MEMORY_FAULT_INFO <<
8163 ------------------------------ <<
8164 <<
8165 :Architectures: x86 <<
8166 :Returns: Informational only, -EINVAL on dire <<
8167 <<
8168 The presence of this capability indicates tha <<
8169 kvm_run.memory_fault if KVM cannot resolve a <<
8170 there is a valid memslot but no backing VMA f <<
8171 address. <<
8172 <<
8173 The information in kvm_run.memory_fault is va <<
8174 an error with errno=EFAULT or errno=EHWPOISON <<
8175 to KVM_EXIT_MEMORY_FAULT. <<
8176 <<
8177 Note: Userspaces which attempt to resolve mem <<
8178 KVM_RUN are encouraged to guard against repea <<
8179 error/annotated fault. <<
8180 <<
8181 See KVM_EXIT_MEMORY_FAULT for more informatio <<
8182 <<
8183 7.35 KVM_CAP_X86_APIC_BUS_CYCLES_NS <<
8184 ----------------------------------- <<
8185 <<
8186 :Architectures: x86 <<
8187 :Target: VM <<
8188 :Parameters: args[0] is the desired APIC bus <<
8189 :Returns: 0 on success, -EINVAL if args[0] co <<
8190 frequency or if any vCPUs have been <<
8191 local APIC has not been created usi <<
8192 <<
8193 This capability sets the VM's APIC bus clock <<
8194 virtual APIC when emulating APIC timers. KVM <<
8195 by KVM_CHECK_EXTENSION. <<
8196 <<
8197 Note: Userspace is responsible for correctly <<
8198 core crystal clock frequency, if a non-zero C <<
8199 <<
8200 7.36 KVM_CAP_X86_GUEST_MODE <<
8201 ------------------------------ <<
8202 <<
8203 :Architectures: x86 <<
8204 :Returns: Informational only, -EINVAL on dire <<
8205 <<
8206 The presence of this capability indicates tha <<
8207 KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags t <<
8208 vCPU was executing nested guest code when it <<
8209 <<
8210 KVM exits with the register state of either t <<
8211 depending on which executed at the time of an <<
8212 take care to differentiate between these case <<
8213 <<
8214 8. Other capabilities. 7861 8. Other capabilities.
8215 ====================== 7862 ======================
8216 7863
8217 This section lists capabilities that give inf 7864 This section lists capabilities that give information about other
8218 features of the KVM implementation. 7865 features of the KVM implementation.
8219 7866
8220 8.1 KVM_CAP_PPC_HWRNG 7867 8.1 KVM_CAP_PPC_HWRNG
8221 --------------------- 7868 ---------------------
8222 7869
8223 :Architectures: ppc 7870 :Architectures: ppc
8224 7871
8225 This capability, if KVM_CHECK_EXTENSION indic 7872 This capability, if KVM_CHECK_EXTENSION indicates that it is
8226 available, means that the kernel has an imple 7873 available, means that the kernel has an implementation of the
8227 H_RANDOM hypercall backed by a hardware rando 7874 H_RANDOM hypercall backed by a hardware random-number generator.
8228 If present, the kernel H_RANDOM handler can b 7875 If present, the kernel H_RANDOM handler can be enabled for guest use
8229 with the KVM_CAP_PPC_ENABLE_HCALL capability. 7876 with the KVM_CAP_PPC_ENABLE_HCALL capability.
8230 7877
8231 8.2 KVM_CAP_HYPERV_SYNIC 7878 8.2 KVM_CAP_HYPERV_SYNIC
8232 ------------------------ 7879 ------------------------
8233 7880
8234 :Architectures: x86 7881 :Architectures: x86
8235 7882
8236 This capability, if KVM_CHECK_EXTENSION indic 7883 This capability, if KVM_CHECK_EXTENSION indicates that it is
8237 available, means that the kernel has an imple 7884 available, means that the kernel has an implementation of the
8238 Hyper-V Synthetic interrupt controller(SynIC) 7885 Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
8239 used to support Windows Hyper-V based guest p 7886 used to support Windows Hyper-V based guest paravirt drivers(VMBus).
8240 7887
8241 In order to use SynIC, it has to be activated 7888 In order to use SynIC, it has to be activated by setting this
8242 capability via KVM_ENABLE_CAP ioctl on the vc 7889 capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
8243 will disable the use of APIC hardware virtual 7890 will disable the use of APIC hardware virtualization even if supported
8244 by the CPU, as it's incompatible with SynIC a 7891 by the CPU, as it's incompatible with SynIC auto-EOI behavior.
8245 7892
8246 8.3 KVM_CAP_PPC_MMU_RADIX !! 7893 8.3 KVM_CAP_PPC_RADIX_MMU
8247 ------------------------- 7894 -------------------------
8248 7895
8249 :Architectures: ppc 7896 :Architectures: ppc
8250 7897
8251 This capability, if KVM_CHECK_EXTENSION indic 7898 This capability, if KVM_CHECK_EXTENSION indicates that it is
8252 available, means that the kernel can support 7899 available, means that the kernel can support guests using the
8253 radix MMU defined in Power ISA V3.00 (as impl 7900 radix MMU defined in Power ISA V3.00 (as implemented in the POWER9
8254 processor). 7901 processor).
8255 7902
8256 8.4 KVM_CAP_PPC_MMU_HASH_V3 !! 7903 8.4 KVM_CAP_PPC_HASH_MMU_V3
8257 --------------------------- 7904 ---------------------------
8258 7905
8259 :Architectures: ppc 7906 :Architectures: ppc
8260 7907
8261 This capability, if KVM_CHECK_EXTENSION indic 7908 This capability, if KVM_CHECK_EXTENSION indicates that it is
8262 available, means that the kernel can support 7909 available, means that the kernel can support guests using the
8263 hashed page table MMU defined in Power ISA V3 7910 hashed page table MMU defined in Power ISA V3.00 (as implemented in
8264 the POWER9 processor), including in-memory se 7911 the POWER9 processor), including in-memory segment tables.
8265 7912
8266 8.5 KVM_CAP_MIPS_VZ 7913 8.5 KVM_CAP_MIPS_VZ
8267 ------------------- 7914 -------------------
8268 7915
8269 :Architectures: mips 7916 :Architectures: mips
8270 7917
8271 This capability, if KVM_CHECK_EXTENSION on th 7918 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
8272 it is available, means that full hardware ass 7919 it is available, means that full hardware assisted virtualization capabilities
8273 of the hardware are available for use through 7920 of the hardware are available for use through KVM. An appropriate
8274 KVM_VM_MIPS_* type must be passed to KVM_CREA 7921 KVM_VM_MIPS_* type must be passed to KVM_CREATE_VM to create a VM which
8275 utilises it. 7922 utilises it.
8276 7923
8277 If KVM_CHECK_EXTENSION on a kvm VM handle ind 7924 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
8278 available, it means that the VM is using full 7925 available, it means that the VM is using full hardware assisted virtualization
8279 capabilities of the hardware. This is useful 7926 capabilities of the hardware. This is useful to check after creating a VM with
8280 KVM_VM_MIPS_DEFAULT. 7927 KVM_VM_MIPS_DEFAULT.
8281 7928
8282 The value returned by KVM_CHECK_EXTENSION sho 7929 The value returned by KVM_CHECK_EXTENSION should be compared against known
8283 values (see below). All other values are rese 7930 values (see below). All other values are reserved. This is to allow for the
8284 possibility of other hardware assisted virtua 7931 possibility of other hardware assisted virtualization implementations which
8285 may be incompatible with the MIPS VZ ASE. 7932 may be incompatible with the MIPS VZ ASE.
8286 7933
8287 == ========================================= 7934 == ==========================================================================
8288 0 The trap & emulate implementation is in u 7935 0 The trap & emulate implementation is in use to run guest code in user
8289 mode. Guest virtual memory segments are r 7936 mode. Guest virtual memory segments are rearranged to fit the guest in the
8290 user mode address space. 7937 user mode address space.
8291 7938
8292 1 The MIPS VZ ASE is in use, providing full 7939 1 The MIPS VZ ASE is in use, providing full hardware assisted
8293 virtualization, including standard guest 7940 virtualization, including standard guest virtual memory segments.
8294 == ========================================= 7941 == ==========================================================================
8295 7942
8296 8.6 KVM_CAP_MIPS_TE 7943 8.6 KVM_CAP_MIPS_TE
8297 ------------------- 7944 -------------------
8298 7945
8299 :Architectures: mips 7946 :Architectures: mips
8300 7947
8301 This capability, if KVM_CHECK_EXTENSION on th 7948 This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
8302 it is available, means that the trap & emulat 7949 it is available, means that the trap & emulate implementation is available to
8303 run guest code in user mode, even if KVM_CAP_ 7950 run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware
8304 assisted virtualisation is also available. KV 7951 assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed
8305 to KVM_CREATE_VM to create a VM which utilise 7952 to KVM_CREATE_VM to create a VM which utilises it.
8306 7953
8307 If KVM_CHECK_EXTENSION on a kvm VM handle ind 7954 If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
8308 available, it means that the VM is using trap 7955 available, it means that the VM is using trap & emulate.
8309 7956
8310 8.7 KVM_CAP_MIPS_64BIT 7957 8.7 KVM_CAP_MIPS_64BIT
8311 ---------------------- 7958 ----------------------
8312 7959
8313 :Architectures: mips 7960 :Architectures: mips
8314 7961
8315 This capability indicates the supported archi 7962 This capability indicates the supported architecture type of the guest, i.e. the
8316 supported register and address width. 7963 supported register and address width.
8317 7964
8318 The values returned when this capability is c 7965 The values returned when this capability is checked by KVM_CHECK_EXTENSION on a
8319 kvm VM handle correspond roughly to the CP0_C 7966 kvm VM handle correspond roughly to the CP0_Config.AT register field, and should
8320 be checked specifically against known values 7967 be checked specifically against known values (see below). All other values are
8321 reserved. 7968 reserved.
8322 7969
8323 == ========================================= 7970 == ========================================================================
8324 0 MIPS32 or microMIPS32. 7971 0 MIPS32 or microMIPS32.
8325 Both registers and addresses are 32-bits 7972 Both registers and addresses are 32-bits wide.
8326 It will only be possible to run 32-bit gu 7973 It will only be possible to run 32-bit guest code.
8327 7974
8328 1 MIPS64 or microMIPS64 with access only to 7975 1 MIPS64 or microMIPS64 with access only to 32-bit compatibility segments.
8329 Registers are 64-bits wide, but addresses 7976 Registers are 64-bits wide, but addresses are 32-bits wide.
8330 64-bit guest code may run but cannot acce 7977 64-bit guest code may run but cannot access MIPS64 memory segments.
8331 It will also be possible to run 32-bit gu 7978 It will also be possible to run 32-bit guest code.
8332 7979
8333 2 MIPS64 or microMIPS64 with access to all 7980 2 MIPS64 or microMIPS64 with access to all address segments.
8334 Both registers and addresses are 64-bits 7981 Both registers and addresses are 64-bits wide.
8335 It will be possible to run 64-bit or 32-b 7982 It will be possible to run 64-bit or 32-bit guest code.
8336 == ========================================= 7983 == ========================================================================
8337 7984
8338 8.9 KVM_CAP_ARM_USER_IRQ 7985 8.9 KVM_CAP_ARM_USER_IRQ
8339 ------------------------ 7986 ------------------------
8340 7987
8341 :Architectures: arm64 7988 :Architectures: arm64
8342 7989
8343 This capability, if KVM_CHECK_EXTENSION indic 7990 This capability, if KVM_CHECK_EXTENSION indicates that it is available, means
8344 that if userspace creates a VM without an in- 7991 that if userspace creates a VM without an in-kernel interrupt controller, it
8345 will be notified of changes to the output lev 7992 will be notified of changes to the output level of in-kernel emulated devices,
8346 which can generate virtual interrupts, presen 7993 which can generate virtual interrupts, presented to the VM.
8347 For such VMs, on every return to userspace, t 7994 For such VMs, on every return to userspace, the kernel
8348 updates the vcpu's run->s.regs.device_irq_lev 7995 updates the vcpu's run->s.regs.device_irq_level field to represent the actual
8349 output level of the device. 7996 output level of the device.
8350 7997
8351 Whenever kvm detects a change in the device o 7998 Whenever kvm detects a change in the device output level, kvm guarantees at
8352 least one return to userspace before running 7999 least one return to userspace before running the VM. This exit could either
8353 be a KVM_EXIT_INTR or any other exit event, l 8000 be a KVM_EXIT_INTR or any other exit event, like KVM_EXIT_MMIO. This way,
8354 userspace can always sample the device output 8001 userspace can always sample the device output level and re-compute the state of
8355 the userspace interrupt controller. Userspac 8002 the userspace interrupt controller. Userspace should always check the state
8356 of run->s.regs.device_irq_level on every kvm 8003 of run->s.regs.device_irq_level on every kvm exit.
8357 The value in run->s.regs.device_irq_level can 8004 The value in run->s.regs.device_irq_level can represent both level and edge
8358 triggered interrupt signals, depending on the 8005 triggered interrupt signals, depending on the device. Edge triggered interrupt
8359 signals will exit to userspace with the bit i 8006 signals will exit to userspace with the bit in run->s.regs.device_irq_level
8360 set exactly once per edge signal. 8007 set exactly once per edge signal.
8361 8008
8362 The field run->s.regs.device_irq_level is ava 8009 The field run->s.regs.device_irq_level is available independent of
8363 run->kvm_valid_regs or run->kvm_dirty_regs bi 8010 run->kvm_valid_regs or run->kvm_dirty_regs bits.
8364 8011
8365 If KVM_CAP_ARM_USER_IRQ is supported, the KVM 8012 If KVM_CAP_ARM_USER_IRQ is supported, the KVM_CHECK_EXTENSION ioctl returns a
8366 number larger than 0 indicating the version o 8013 number larger than 0 indicating the version of this capability is implemented
8367 and thereby which bits in run->s.regs.device_ 8014 and thereby which bits in run->s.regs.device_irq_level can signal values.
8368 8015
8369 Currently the following bits are defined for 8016 Currently the following bits are defined for the device_irq_level bitmap::
8370 8017
8371 KVM_CAP_ARM_USER_IRQ >= 1: 8018 KVM_CAP_ARM_USER_IRQ >= 1:
8372 8019
8373 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual tim 8020 KVM_ARM_DEV_EL1_VTIMER - EL1 virtual timer
8374 KVM_ARM_DEV_EL1_PTIMER - EL1 physical ti 8021 KVM_ARM_DEV_EL1_PTIMER - EL1 physical timer
8375 KVM_ARM_DEV_PMU - ARM PMU overflo 8022 KVM_ARM_DEV_PMU - ARM PMU overflow interrupt signal
8376 8023
8377 Future versions of kvm may implement addition 8024 Future versions of kvm may implement additional events. These will get
8378 indicated by returning a higher number from K 8025 indicated by returning a higher number from KVM_CHECK_EXTENSION and will be
8379 listed above. 8026 listed above.
8380 8027
8381 8.10 KVM_CAP_PPC_SMT_POSSIBLE 8028 8.10 KVM_CAP_PPC_SMT_POSSIBLE
8382 ----------------------------- 8029 -----------------------------
8383 8030
8384 :Architectures: ppc 8031 :Architectures: ppc
8385 8032
8386 Querying this capability returns a bitmap ind 8033 Querying this capability returns a bitmap indicating the possible
8387 virtual SMT modes that can be set using KVM_C 8034 virtual SMT modes that can be set using KVM_CAP_PPC_SMT. If bit N
8388 (counting from the right) is set, then a virt 8035 (counting from the right) is set, then a virtual SMT mode of 2^N is
8389 available. 8036 available.
8390 8037
8391 8.11 KVM_CAP_HYPERV_SYNIC2 8038 8.11 KVM_CAP_HYPERV_SYNIC2
8392 -------------------------- 8039 --------------------------
8393 8040
8394 :Architectures: x86 8041 :Architectures: x86
8395 8042
8396 This capability enables a newer version of Hy 8043 This capability enables a newer version of Hyper-V Synthetic interrupt
8397 controller (SynIC). The only difference with 8044 controller (SynIC). The only difference with KVM_CAP_HYPERV_SYNIC is that KVM
8398 doesn't clear SynIC message and event flags p 8045 doesn't clear SynIC message and event flags pages when they are enabled by
8399 writing to the respective MSRs. 8046 writing to the respective MSRs.
8400 8047
8401 8.12 KVM_CAP_HYPERV_VP_INDEX 8048 8.12 KVM_CAP_HYPERV_VP_INDEX
8402 ---------------------------- 8049 ----------------------------
8403 8050
8404 :Architectures: x86 8051 :Architectures: x86
8405 8052
8406 This capability indicates that userspace can 8053 This capability indicates that userspace can load HV_X64_MSR_VP_INDEX msr. Its
8407 value is used to denote the target vcpu for a 8054 value is used to denote the target vcpu for a SynIC interrupt. For
8408 compatibility, KVM initializes this msr to KV 8055 compatibility, KVM initializes this msr to KVM's internal vcpu index. When this
8409 capability is absent, userspace can still que 8056 capability is absent, userspace can still query this msr's value.
8410 8057
8411 8.13 KVM_CAP_S390_AIS_MIGRATION 8058 8.13 KVM_CAP_S390_AIS_MIGRATION
8412 ------------------------------- 8059 -------------------------------
8413 8060
8414 :Architectures: s390 8061 :Architectures: s390
8415 :Parameters: none 8062 :Parameters: none
8416 8063
8417 This capability indicates if the flic device 8064 This capability indicates if the flic device will be able to get/set the
8418 AIS states for migration via the KVM_DEV_FLIC 8065 AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows
8419 to discover this without having to create a f 8066 to discover this without having to create a flic device.
8420 8067
8421 8.14 KVM_CAP_S390_PSW 8068 8.14 KVM_CAP_S390_PSW
8422 --------------------- 8069 ---------------------
8423 8070
8424 :Architectures: s390 8071 :Architectures: s390
8425 8072
8426 This capability indicates that the PSW is exp 8073 This capability indicates that the PSW is exposed via the kvm_run structure.
8427 8074
8428 8.15 KVM_CAP_S390_GMAP 8075 8.15 KVM_CAP_S390_GMAP
8429 ---------------------- 8076 ----------------------
8430 8077
8431 :Architectures: s390 8078 :Architectures: s390
8432 8079
8433 This capability indicates that the user space 8080 This capability indicates that the user space memory used as guest mapping can
8434 be anywhere in the user memory address space, 8081 be anywhere in the user memory address space, as long as the memory slots are
8435 aligned and sized to a segment (1MB) boundary 8082 aligned and sized to a segment (1MB) boundary.
8436 8083
8437 8.16 KVM_CAP_S390_COW 8084 8.16 KVM_CAP_S390_COW
8438 --------------------- 8085 ---------------------
8439 8086
8440 :Architectures: s390 8087 :Architectures: s390
8441 8088
8442 This capability indicates that the user space 8089 This capability indicates that the user space memory used as guest mapping can
8443 use copy-on-write semantics as well as dirty 8090 use copy-on-write semantics as well as dirty pages tracking via read-only page
8444 tables. 8091 tables.
8445 8092
8446 8.17 KVM_CAP_S390_BPB 8093 8.17 KVM_CAP_S390_BPB
8447 --------------------- 8094 ---------------------
8448 8095
8449 :Architectures: s390 8096 :Architectures: s390
8450 8097
8451 This capability indicates that kvm will imple 8098 This capability indicates that kvm will implement the interfaces to handle
8452 reset, migration and nested KVM for branch pr 8099 reset, migration and nested KVM for branch prediction blocking. The stfle
8453 facility 82 should not be provided to the gue 8100 facility 82 should not be provided to the guest without this capability.
8454 8101
8455 8.18 KVM_CAP_HYPERV_TLBFLUSH 8102 8.18 KVM_CAP_HYPERV_TLBFLUSH
8456 ---------------------------- 8103 ----------------------------
8457 8104
8458 :Architectures: x86 8105 :Architectures: x86
8459 8106
8460 This capability indicates that KVM supports p 8107 This capability indicates that KVM supports paravirtualized Hyper-V TLB Flush
8461 hypercalls: 8108 hypercalls:
8462 HvFlushVirtualAddressSpace, HvFlushVirtualAdd 8109 HvFlushVirtualAddressSpace, HvFlushVirtualAddressSpaceEx,
8463 HvFlushVirtualAddressList, HvFlushVirtualAddr 8110 HvFlushVirtualAddressList, HvFlushVirtualAddressListEx.
8464 8111
8465 8.19 KVM_CAP_ARM_INJECT_SERROR_ESR 8112 8.19 KVM_CAP_ARM_INJECT_SERROR_ESR
8466 ---------------------------------- 8113 ----------------------------------
8467 8114
8468 :Architectures: arm64 8115 :Architectures: arm64
8469 8116
8470 This capability indicates that userspace can 8117 This capability indicates that userspace can specify (via the
8471 KVM_SET_VCPU_EVENTS ioctl) the syndrome value 8118 KVM_SET_VCPU_EVENTS ioctl) the syndrome value reported to the guest when it
8472 takes a virtual SError interrupt exception. 8119 takes a virtual SError interrupt exception.
8473 If KVM advertises this capability, userspace 8120 If KVM advertises this capability, userspace can only specify the ISS field for
8474 the ESR syndrome. Other parts of the ESR, suc 8121 the ESR syndrome. Other parts of the ESR, such as the EC are generated by the
8475 CPU when the exception is taken. If this virt 8122 CPU when the exception is taken. If this virtual SError is taken to EL1 using
8476 AArch64, this value will be reported in the I 8123 AArch64, this value will be reported in the ISS field of ESR_ELx.
8477 8124
8478 See KVM_CAP_VCPU_EVENTS for more details. 8125 See KVM_CAP_VCPU_EVENTS for more details.
8479 8126
8480 8.20 KVM_CAP_HYPERV_SEND_IPI 8127 8.20 KVM_CAP_HYPERV_SEND_IPI
8481 ---------------------------- 8128 ----------------------------
8482 8129
8483 :Architectures: x86 8130 :Architectures: x86
8484 8131
8485 This capability indicates that KVM supports p 8132 This capability indicates that KVM supports paravirtualized Hyper-V IPI send
8486 hypercalls: 8133 hypercalls:
8487 HvCallSendSyntheticClusterIpi, HvCallSendSynt 8134 HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx.
8488 8135
8489 8.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH 8136 8.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH
8490 ----------------------------------- 8137 -----------------------------------
8491 8138
8492 :Architectures: x86 8139 :Architectures: x86
8493 8140
8494 This capability indicates that KVM running on 8141 This capability indicates that KVM running on top of Hyper-V hypervisor
8495 enables Direct TLB flush for its guests meani 8142 enables Direct TLB flush for its guests meaning that TLB flush
8496 hypercalls are handled by Level 0 hypervisor 8143 hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM.
8497 Due to the different ABI for hypercall parame 8144 Due to the different ABI for hypercall parameters between Hyper-V and
8498 KVM, enabling this capability effectively dis 8145 KVM, enabling this capability effectively disables all hypercall
8499 handling by KVM (as some KVM hypercall may be 8146 handling by KVM (as some KVM hypercall may be mistakenly treated as TLB
8500 flush hypercalls by Hyper-V) so userspace sho 8147 flush hypercalls by Hyper-V) so userspace should disable KVM identification
8501 in CPUID and only exposes Hyper-V identificat 8148 in CPUID and only exposes Hyper-V identification. In this case, guest
8502 thinks it's running on Hyper-V and only use H 8149 thinks it's running on Hyper-V and only use Hyper-V hypercalls.
8503 8150
8504 8.22 KVM_CAP_S390_VCPU_RESETS 8151 8.22 KVM_CAP_S390_VCPU_RESETS
8505 ----------------------------- 8152 -----------------------------
8506 8153
8507 :Architectures: s390 8154 :Architectures: s390
8508 8155
8509 This capability indicates that the KVM_S390_N 8156 This capability indicates that the KVM_S390_NORMAL_RESET and
8510 KVM_S390_CLEAR_RESET ioctls are available. 8157 KVM_S390_CLEAR_RESET ioctls are available.
8511 8158
8512 8.23 KVM_CAP_S390_PROTECTED 8159 8.23 KVM_CAP_S390_PROTECTED
8513 --------------------------- 8160 ---------------------------
8514 8161
8515 :Architectures: s390 8162 :Architectures: s390
8516 8163
8517 This capability indicates that the Ultravisor 8164 This capability indicates that the Ultravisor has been initialized and
8518 KVM can therefore start protected VMs. 8165 KVM can therefore start protected VMs.
8519 This capability governs the KVM_S390_PV_COMMA 8166 This capability governs the KVM_S390_PV_COMMAND ioctl and the
8520 KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE 8167 KVM_MP_STATE_LOAD MP_STATE. KVM_SET_MP_STATE can fail for protected
8521 guests when the state change is invalid. 8168 guests when the state change is invalid.
8522 8169
8523 8.24 KVM_CAP_STEAL_TIME 8170 8.24 KVM_CAP_STEAL_TIME
8524 ----------------------- 8171 -----------------------
8525 8172
8526 :Architectures: arm64, x86 8173 :Architectures: arm64, x86
8527 8174
8528 This capability indicates that KVM supports s 8175 This capability indicates that KVM supports steal time accounting.
8529 When steal time accounting is supported it ma 8176 When steal time accounting is supported it may be enabled with
8530 architecture-specific interfaces. This capab 8177 architecture-specific interfaces. This capability and the architecture-
8531 specific interfaces must be consistent, i.e. 8178 specific interfaces must be consistent, i.e. if one says the feature
8532 is supported, than the other should as well a 8179 is supported, than the other should as well and vice versa. For arm64
8533 see Documentation/virt/kvm/devices/vcpu.rst " 8180 see Documentation/virt/kvm/devices/vcpu.rst "KVM_ARM_VCPU_PVTIME_CTRL".
8534 For x86 see Documentation/virt/kvm/x86/msr.rs 8181 For x86 see Documentation/virt/kvm/x86/msr.rst "MSR_KVM_STEAL_TIME".
8535 8182
8536 8.25 KVM_CAP_S390_DIAG318 8183 8.25 KVM_CAP_S390_DIAG318
8537 ------------------------- 8184 -------------------------
8538 8185
8539 :Architectures: s390 8186 :Architectures: s390
8540 8187
8541 This capability enables a guest to set inform 8188 This capability enables a guest to set information about its control program
8542 (i.e. guest kernel type and version). The inf 8189 (i.e. guest kernel type and version). The information is helpful during
8543 system/firmware service events, providing add 8190 system/firmware service events, providing additional data about the guest
8544 environments running on the machine. 8191 environments running on the machine.
8545 8192
8546 The information is associated with the DIAGNO 8193 The information is associated with the DIAGNOSE 0x318 instruction, which sets
8547 an 8-byte value consisting of a one-byte Cont 8194 an 8-byte value consisting of a one-byte Control Program Name Code (CPNC) and
8548 a 7-byte Control Program Version Code (CPVC). 8195 a 7-byte Control Program Version Code (CPVC). The CPNC determines what
8549 environment the control program is running in 8196 environment the control program is running in (e.g. Linux, z/VM...), and the
8550 CPVC is used for information specific to OS ( 8197 CPVC is used for information specific to OS (e.g. Linux version, Linux
8551 distribution...) 8198 distribution...)
8552 8199
8553 If this capability is available, then the CPN 8200 If this capability is available, then the CPNC and CPVC can be synchronized
8554 between KVM and userspace via the sync regs m 8201 between KVM and userspace via the sync regs mechanism (KVM_SYNC_DIAG318).
8555 8202
8556 8.26 KVM_CAP_X86_USER_SPACE_MSR 8203 8.26 KVM_CAP_X86_USER_SPACE_MSR
8557 ------------------------------- 8204 -------------------------------
8558 8205
8559 :Architectures: x86 8206 :Architectures: x86
8560 8207
8561 This capability indicates that KVM supports d 8208 This capability indicates that KVM supports deflection of MSR reads and
8562 writes to user space. It can be enabled on a 8209 writes to user space. It can be enabled on a VM level. If enabled, MSR
8563 accesses that would usually trigger a #GP by 8210 accesses that would usually trigger a #GP by KVM into the guest will
8564 instead get bounced to user space through the 8211 instead get bounced to user space through the KVM_EXIT_X86_RDMSR and
8565 KVM_EXIT_X86_WRMSR exit notifications. 8212 KVM_EXIT_X86_WRMSR exit notifications.
8566 8213
8567 8.27 KVM_CAP_X86_MSR_FILTER 8214 8.27 KVM_CAP_X86_MSR_FILTER
8568 --------------------------- 8215 ---------------------------
8569 8216
8570 :Architectures: x86 8217 :Architectures: x86
8571 8218
8572 This capability indicates that KVM supports t 8219 This capability indicates that KVM supports that accesses to user defined MSRs
8573 may be rejected. With this capability exposed 8220 may be rejected. With this capability exposed, KVM exports new VM ioctl
8574 KVM_X86_SET_MSR_FILTER which user space can c 8221 KVM_X86_SET_MSR_FILTER which user space can call to specify bitmaps of MSR
8575 ranges that KVM should deny access to. 8222 ranges that KVM should deny access to.
8576 8223
8577 In combination with KVM_CAP_X86_USER_SPACE_MS 8224 In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to
8578 trap and emulate MSRs that are outside of the 8225 trap and emulate MSRs that are outside of the scope of KVM as well as
8579 limit the attack surface on KVM's MSR emulati 8226 limit the attack surface on KVM's MSR emulation code.
8580 8227
8581 8.28 KVM_CAP_ENFORCE_PV_FEATURE_CPUID 8228 8.28 KVM_CAP_ENFORCE_PV_FEATURE_CPUID
8582 ------------------------------------- 8229 -------------------------------------
8583 8230
8584 Architectures: x86 8231 Architectures: x86
8585 8232
8586 When enabled, KVM will disable paravirtual fe 8233 When enabled, KVM will disable paravirtual features provided to the
8587 guest according to the bits in the KVM_CPUID_ 8234 guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf
8588 (0x40000001). Otherwise, a guest may use the 8235 (0x40000001). Otherwise, a guest may use the paravirtual features
8589 regardless of what has actually been exposed 8236 regardless of what has actually been exposed through the CPUID leaf.
8590 8237
8591 8.29 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG 8238 8.29 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG_RING_ACQ_REL
8592 --------------------------------------------- 8239 ----------------------------------------------------------
8593 8240
8594 :Architectures: x86, arm64 8241 :Architectures: x86, arm64
8595 :Parameters: args[0] - size of the dirty log 8242 :Parameters: args[0] - size of the dirty log ring
8596 8243
8597 KVM is capable of tracking dirty memory using 8244 KVM is capable of tracking dirty memory using ring buffers that are
8598 mmapped into userspace; there is one dirty ri 8245 mmapped into userspace; there is one dirty ring per vcpu.
8599 8246
8600 The dirty ring is available to userspace as a 8247 The dirty ring is available to userspace as an array of
8601 ``struct kvm_dirty_gfn``. Each dirty entry i 8248 ``struct kvm_dirty_gfn``. Each dirty entry is defined as::
8602 8249
8603 struct kvm_dirty_gfn { 8250 struct kvm_dirty_gfn {
8604 __u32 flags; 8251 __u32 flags;
8605 __u32 slot; /* as_id | slot_id */ 8252 __u32 slot; /* as_id | slot_id */
8606 __u64 offset; 8253 __u64 offset;
8607 }; 8254 };
8608 8255
8609 The following values are defined for the flag 8256 The following values are defined for the flags field to define the
8610 current state of the entry:: 8257 current state of the entry::
8611 8258
8612 #define KVM_DIRTY_GFN_F_DIRTY BIT 8259 #define KVM_DIRTY_GFN_F_DIRTY BIT(0)
8613 #define KVM_DIRTY_GFN_F_RESET BIT 8260 #define KVM_DIRTY_GFN_F_RESET BIT(1)
8614 #define KVM_DIRTY_GFN_F_MASK 0x3 8261 #define KVM_DIRTY_GFN_F_MASK 0x3
8615 8262
8616 Userspace should call KVM_ENABLE_CAP ioctl ri 8263 Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM
8617 ioctl to enable this capability for the new g 8264 ioctl to enable this capability for the new guest and set the size of
8618 the rings. Enabling the capability is only a 8265 the rings. Enabling the capability is only allowed before creating any
8619 vCPU, and the size of the ring must be a powe 8266 vCPU, and the size of the ring must be a power of two. The larger the
8620 ring buffer, the less likely the ring is full 8267 ring buffer, the less likely the ring is full and the VM is forced to
8621 exit to userspace. The optimal size depends o 8268 exit to userspace. The optimal size depends on the workload, but it is
8622 recommended that it be at least 64 KiB (4096 8269 recommended that it be at least 64 KiB (4096 entries).
8623 8270
8624 Just like for dirty page bitmaps, the buffer 8271 Just like for dirty page bitmaps, the buffer tracks writes to
8625 all user memory regions for which the KVM_MEM 8272 all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was
8626 set in KVM_SET_USER_MEMORY_REGION. Once a me 8273 set in KVM_SET_USER_MEMORY_REGION. Once a memory region is registered
8627 with the flag set, userspace can start harves 8274 with the flag set, userspace can start harvesting dirty pages from the
8628 ring buffer. 8275 ring buffer.
8629 8276
8630 An entry in the ring buffer can be unused (fl 8277 An entry in the ring buffer can be unused (flag bits ``00``),
8631 dirty (flag bits ``01``) or harvested (flag b 8278 dirty (flag bits ``01``) or harvested (flag bits ``1X``). The
8632 state machine for the entry is as follows:: 8279 state machine for the entry is as follows::
8633 8280
8634 dirtied harvested re 8281 dirtied harvested reset
8635 00 -----------> 01 -------------> 1X --- 8282 00 -----------> 01 -------------> 1X -------+
8636 ^ 8283 ^ |
8637 | 8284 | |
8638 +-------------------------------------- 8285 +------------------------------------------+
8639 8286
8640 To harvest the dirty pages, userspace accesse 8287 To harvest the dirty pages, userspace accesses the mmapped ring buffer
8641 to read the dirty GFNs. If the flags has the 8288 to read the dirty GFNs. If the flags has the DIRTY bit set (at this stage
8642 the RESET bit must be cleared), then it means 8289 the RESET bit must be cleared), then it means this GFN is a dirty GFN.
8643 The userspace should harvest this GFN and mar 8290 The userspace should harvest this GFN and mark the flags from state
8644 ``01b`` to ``1Xb`` (bit 0 will be ignored by 8291 ``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set
8645 to show that this GFN is harvested and waitin 8292 to show that this GFN is harvested and waiting for a reset), and move
8646 on to the next GFN. The userspace should con 8293 on to the next GFN. The userspace should continue to do this until the
8647 flags of a GFN have the DIRTY bit cleared, me 8294 flags of a GFN have the DIRTY bit cleared, meaning that it has harvested
8648 all the dirty GFNs that were available. 8295 all the dirty GFNs that were available.
8649 8296
8650 Note that on weakly ordered architectures, us 8297 Note that on weakly ordered architectures, userspace accesses to the
8651 ring buffer (and more specifically the 'flags 8298 ring buffer (and more specifically the 'flags' field) must be ordered,
8652 using load-acquire/store-release accessors wh 8299 using load-acquire/store-release accessors when available, or any
8653 other memory barrier that will ensure this or 8300 other memory barrier that will ensure this ordering.
8654 8301
8655 It's not necessary for userspace to harvest t 8302 It's not necessary for userspace to harvest the all dirty GFNs at once.
8656 However it must collect the dirty GFNs in seq 8303 However it must collect the dirty GFNs in sequence, i.e., the userspace
8657 program cannot skip one dirty GFN to collect 8304 program cannot skip one dirty GFN to collect the one next to it.
8658 8305
8659 After processing one or more entries in the r 8306 After processing one or more entries in the ring buffer, userspace
8660 calls the VM ioctl KVM_RESET_DIRTY_RINGS to n 8307 calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about
8661 it, so that the kernel will reprotect those c 8308 it, so that the kernel will reprotect those collected GFNs.
8662 Therefore, the ioctl must be called *before* 8309 Therefore, the ioctl must be called *before* reading the content of
8663 the dirty pages. 8310 the dirty pages.
8664 8311
8665 The dirty ring can get full. When it happens 8312 The dirty ring can get full. When it happens, the KVM_RUN of the
8666 vcpu will return with exit reason KVM_EXIT_DI 8313 vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL.
8667 8314
8668 The dirty ring interface has a major differen 8315 The dirty ring interface has a major difference comparing to the
8669 KVM_GET_DIRTY_LOG interface in that, when rea 8316 KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from
8670 userspace, it's still possible that the kerne 8317 userspace, it's still possible that the kernel has not yet flushed the
8671 processor's dirty page buffers into the kerne 8318 processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the
8672 flushing is done by the KVM_GET_DIRTY_LOG ioc 8319 flushing is done by the KVM_GET_DIRTY_LOG ioctl). To achieve that, one
8673 needs to kick the vcpu out of KVM_RUN using a 8320 needs to kick the vcpu out of KVM_RUN using a signal. The resulting
8674 vmexit ensures that all dirty GFNs are flushe 8321 vmexit ensures that all dirty GFNs are flushed to the dirty rings.
8675 8322
8676 NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the o 8323 NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the only capability that
8677 should be exposed by weakly ordered architect 8324 should be exposed by weakly ordered architecture, in order to indicate
8678 the additional memory ordering requirements i 8325 the additional memory ordering requirements imposed on userspace when
8679 reading the state of an entry and mutating it 8326 reading the state of an entry and mutating it from DIRTY to HARVESTED.
8680 Architecture with TSO-like ordering (such as 8327 Architecture with TSO-like ordering (such as x86) are allowed to
8681 expose both KVM_CAP_DIRTY_LOG_RING and KVM_CA 8328 expose both KVM_CAP_DIRTY_LOG_RING and KVM_CAP_DIRTY_LOG_RING_ACQ_REL
8682 to userspace. 8329 to userspace.
8683 8330
8684 After enabling the dirty rings, the userspace 8331 After enabling the dirty rings, the userspace needs to detect the
8685 capability of KVM_CAP_DIRTY_LOG_RING_WITH_BIT 8332 capability of KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP to see whether the
8686 ring structures can be backed by per-slot bit 8333 ring structures can be backed by per-slot bitmaps. With this capability
8687 advertised, it means the architecture can dir 8334 advertised, it means the architecture can dirty guest pages without
8688 vcpu/ring context, so that some of the dirty 8335 vcpu/ring context, so that some of the dirty information will still be
8689 maintained in the bitmap structure. KVM_CAP_D 8336 maintained in the bitmap structure. KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP
8690 can't be enabled if the capability of KVM_CAP 8337 can't be enabled if the capability of KVM_CAP_DIRTY_LOG_RING_ACQ_REL
8691 hasn't been enabled, or any memslot has been 8338 hasn't been enabled, or any memslot has been existing.
8692 8339
8693 Note that the bitmap here is only a backup of 8340 Note that the bitmap here is only a backup of the ring structure. The
8694 use of the ring and bitmap combination is onl 8341 use of the ring and bitmap combination is only beneficial if there is
8695 only a very small amount of memory that is di 8342 only a very small amount of memory that is dirtied out of vcpu/ring
8696 context. Otherwise, the stand-alone per-slot 8343 context. Otherwise, the stand-alone per-slot bitmap mechanism needs to
8697 be considered. 8344 be considered.
8698 8345
8699 To collect dirty bits in the backup bitmap, u 8346 To collect dirty bits in the backup bitmap, userspace can use the same
8700 KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG 8347 KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG isn't needed as long as all
8701 the generation of the dirty bits is done in a 8348 the generation of the dirty bits is done in a single pass. Collecting
8702 the dirty bitmap should be the very last thin 8349 the dirty bitmap should be the very last thing that the VMM does before
8703 considering the state as complete. VMM needs 8350 considering the state as complete. VMM needs to ensure that the dirty
8704 state is final and avoid missing dirty pages 8351 state is final and avoid missing dirty pages from another ioctl ordered
8705 after the bitmap collection. 8352 after the bitmap collection.
8706 8353
8707 NOTE: Multiple examples of using the backup b 8354 NOTE: Multiple examples of using the backup bitmap: (1) save vgic/its
8708 tables through command KVM_DEV_ARM_{VGIC_GRP_ 8355 tables through command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} on
8709 KVM device "kvm-arm-vgic-its". (2) restore vg 8356 KVM device "kvm-arm-vgic-its". (2) restore vgic/its tables through
8710 command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTO 8357 command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} on KVM device
8711 "kvm-arm-vgic-its". VGICv3 LPI pending status 8358 "kvm-arm-vgic-its". VGICv3 LPI pending status is restored. (3) save
8712 vgic3 pending table through KVM_DEV_ARM_VGIC_ 8359 vgic3 pending table through KVM_DEV_ARM_VGIC_{GRP_CTRL, SAVE_PENDING_TABLES}
8713 command on KVM device "kvm-arm-vgic-v3". 8360 command on KVM device "kvm-arm-vgic-v3".
8714 8361
8715 8.30 KVM_CAP_XEN_HVM 8362 8.30 KVM_CAP_XEN_HVM
8716 -------------------- 8363 --------------------
8717 8364
8718 :Architectures: x86 8365 :Architectures: x86
8719 8366
8720 This capability indicates the features that X 8367 This capability indicates the features that Xen supports for hosting Xen
8721 PVHVM guests. Valid flags are:: 8368 PVHVM guests. Valid flags are::
8722 8369
8723 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR 8370 #define KVM_XEN_HVM_CONFIG_HYPERCALL_MSR (1 << 0)
8724 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL 8371 #define KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL (1 << 1)
8725 #define KVM_XEN_HVM_CONFIG_SHARED_INFO 8372 #define KVM_XEN_HVM_CONFIG_SHARED_INFO (1 << 2)
8726 #define KVM_XEN_HVM_CONFIG_RUNSTATE 8373 #define KVM_XEN_HVM_CONFIG_RUNSTATE (1 << 3)
8727 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL 8374 #define KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL (1 << 4)
8728 #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND 8375 #define KVM_XEN_HVM_CONFIG_EVTCHN_SEND (1 << 5)
8729 #define KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_ 8376 #define KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG (1 << 6)
8730 #define KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNST <<
8731 8377
8732 The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag ind 8378 The KVM_XEN_HVM_CONFIG_HYPERCALL_MSR flag indicates that the KVM_XEN_HVM_CONFIG
8733 ioctl is available, for the guest to set its 8379 ioctl is available, for the guest to set its hypercall page.
8734 8380
8735 If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also 8381 If KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL is also set, the same flag may also be
8736 provided in the flags to KVM_XEN_HVM_CONFIG, 8382 provided in the flags to KVM_XEN_HVM_CONFIG, without providing hypercall page
8737 contents, to request that KVM generate hyperc 8383 contents, to request that KVM generate hypercall page content automatically
8738 and also enable interception of guest hyperca 8384 and also enable interception of guest hypercalls with KVM_EXIT_XEN.
8739 8385
8740 The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indic 8386 The KVM_XEN_HVM_CONFIG_SHARED_INFO flag indicates the availability of the
8741 KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, K 8387 KVM_XEN_HVM_SET_ATTR, KVM_XEN_HVM_GET_ATTR, KVM_XEN_VCPU_SET_ATTR and
8742 KVM_XEN_VCPU_GET_ATTR ioctls, as well as the 8388 KVM_XEN_VCPU_GET_ATTR ioctls, as well as the delivery of exception vectors
8743 for event channel upcalls when the evtchn_upc 8389 for event channel upcalls when the evtchn_upcall_pending field of a vcpu's
8744 vcpu_info is set. 8390 vcpu_info is set.
8745 8391
8746 The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicate 8392 The KVM_XEN_HVM_CONFIG_RUNSTATE flag indicates that the runstate-related
8747 features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR 8393 features KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR/_CURRENT/_DATA/_ADJUST are
8748 supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XE 8394 supported by the KVM_XEN_VCPU_SET_ATTR/KVM_XEN_VCPU_GET_ATTR ioctls.
8749 8395
8750 The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag ind 8396 The KVM_XEN_HVM_CONFIG_EVTCHN_2LEVEL flag indicates that IRQ routing entries
8751 of the type KVM_IRQ_ROUTING_XEN_EVTCHN are su 8397 of the type KVM_IRQ_ROUTING_XEN_EVTCHN are supported, with the priority
8752 field set to indicate 2 level event channel d 8398 field set to indicate 2 level event channel delivery.
8753 8399
8754 The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indic 8400 The KVM_XEN_HVM_CONFIG_EVTCHN_SEND flag indicates that KVM supports
8755 injecting event channel events directly into 8401 injecting event channel events directly into the guest with the
8756 KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indica 8402 KVM_XEN_HVM_EVTCHN_SEND ioctl. It also indicates support for the
8757 KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attr 8403 KVM_XEN_ATTR_TYPE_EVTCHN/XEN_VERSION HVM attributes and the
8758 KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_V 8404 KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID/TIMER/UPCALL_VECTOR vCPU attributes.
8759 related to event channel delivery, timers, an 8405 related to event channel delivery, timers, and the XENVER_version
8760 interception. 8406 interception.
8761 8407
8762 The KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG f 8408 The KVM_XEN_HVM_CONFIG_RUNSTATE_UPDATE_FLAG flag indicates that KVM supports
8763 the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG at 8409 the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute in the KVM_XEN_SET_ATTR
8764 and KVM_XEN_GET_ATTR ioctls. This controls wh 8410 and KVM_XEN_GET_ATTR ioctls. This controls whether KVM will set the
8765 XEN_RUNSTATE_UPDATE flag in guest memory mapp 8411 XEN_RUNSTATE_UPDATE flag in guest memory mapped vcpu_runstate_info during
8766 updates of the runstate information. Note tha 8412 updates of the runstate information. Note that versions of KVM which support
8767 the RUNSTATE feature above, but not the RUNST 8413 the RUNSTATE feature above, but not the RUNSTATE_UPDATE_FLAG feature, will
8768 always set the XEN_RUNSTATE_UPDATE flag when 8414 always set the XEN_RUNSTATE_UPDATE flag when updating the guest structure,
8769 which is perhaps counterintuitive. When this 8415 which is perhaps counterintuitive. When this flag is advertised, KVM will
8770 behave more correctly, not using the XEN_RUNS 8416 behave more correctly, not using the XEN_RUNSTATE_UPDATE flag until/unless
8771 specifically enabled (by the guest making the 8417 specifically enabled (by the guest making the hypercall, causing the VMM
8772 to enable the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDA 8418 to enable the KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG attribute).
8773 8419
8774 The KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE f <<
8775 clearing the PVCLOCK_TSC_STABLE_BIT flag in X <<
8776 done when the KVM_CAP_XEN_HVM ioctl sets the <<
8777 KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag. <<
8778 <<
8779 8.31 KVM_CAP_PPC_MULTITCE 8420 8.31 KVM_CAP_PPC_MULTITCE
8780 ------------------------- 8421 -------------------------
8781 8422
8782 :Capability: KVM_CAP_PPC_MULTITCE 8423 :Capability: KVM_CAP_PPC_MULTITCE
8783 :Architectures: ppc 8424 :Architectures: ppc
8784 :Type: vm 8425 :Type: vm
8785 8426
8786 This capability means the kernel is capable o 8427 This capability means the kernel is capable of handling hypercalls
8787 H_PUT_TCE_INDIRECT and H_STUFF_TCE without pa 8428 H_PUT_TCE_INDIRECT and H_STUFF_TCE without passing those into the user
8788 space. This significantly accelerates DMA ope 8429 space. This significantly accelerates DMA operations for PPC KVM guests.
8789 User space should expect that its handlers fo 8430 User space should expect that its handlers for these hypercalls
8790 are not going to be called if user space prev 8431 are not going to be called if user space previously registered LIOBN
8791 in KVM (via KVM_CREATE_SPAPR_TCE or similar c 8432 in KVM (via KVM_CREATE_SPAPR_TCE or similar calls).
8792 8433
8793 In order to enable H_PUT_TCE_INDIRECT and H_S 8434 In order to enable H_PUT_TCE_INDIRECT and H_STUFF_TCE use in the guest,
8794 user space might have to advertise it for the 8435 user space might have to advertise it for the guest. For example,
8795 IBM pSeries (sPAPR) guest starts using them i 8436 IBM pSeries (sPAPR) guest starts using them if "hcall-multi-tce" is
8796 present in the "ibm,hypertas-functions" devic 8437 present in the "ibm,hypertas-functions" device-tree property.
8797 8438
8798 The hypercalls mentioned above may or may not 8439 The hypercalls mentioned above may or may not be processed successfully
8799 in the kernel based fast path. If they can no 8440 in the kernel based fast path. If they can not be handled by the kernel,
8800 they will get passed on to user space. So use 8441 they will get passed on to user space. So user space still has to have
8801 an implementation for these despite the in ke 8442 an implementation for these despite the in kernel acceleration.
8802 8443
8803 This capability is always enabled. 8444 This capability is always enabled.
8804 8445
8805 8.32 KVM_CAP_PTP_KVM 8446 8.32 KVM_CAP_PTP_KVM
8806 -------------------- 8447 --------------------
8807 8448
8808 :Architectures: arm64 8449 :Architectures: arm64
8809 8450
8810 This capability indicates that the KVM virtua 8451 This capability indicates that the KVM virtual PTP service is
8811 supported in the host. A VMM can check whethe 8452 supported in the host. A VMM can check whether the service is
8812 available to the guest on migration. 8453 available to the guest on migration.
8813 8454
8814 8.33 KVM_CAP_HYPERV_ENFORCE_CPUID 8455 8.33 KVM_CAP_HYPERV_ENFORCE_CPUID
8815 --------------------------------- 8456 ---------------------------------
8816 8457
8817 Architectures: x86 8458 Architectures: x86
8818 8459
8819 When enabled, KVM will disable emulated Hyper 8460 When enabled, KVM will disable emulated Hyper-V features provided to the
8820 guest according to the bits Hyper-V CPUID fea 8461 guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all
8821 currently implemented Hyper-V features are pr 8462 currently implemented Hyper-V features are provided unconditionally when
8822 Hyper-V identification is set in the HYPERV_C 8463 Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001)
8823 leaf. 8464 leaf.
8824 8465
8825 8.34 KVM_CAP_EXIT_HYPERCALL 8466 8.34 KVM_CAP_EXIT_HYPERCALL
8826 --------------------------- 8467 ---------------------------
8827 8468
8828 :Capability: KVM_CAP_EXIT_HYPERCALL 8469 :Capability: KVM_CAP_EXIT_HYPERCALL
8829 :Architectures: x86 8470 :Architectures: x86
8830 :Type: vm 8471 :Type: vm
8831 8472
8832 This capability, if enabled, will cause KVM t 8473 This capability, if enabled, will cause KVM to exit to userspace
8833 with KVM_EXIT_HYPERCALL exit reason to proces 8474 with KVM_EXIT_HYPERCALL exit reason to process some hypercalls.
8834 8475
8835 Calling KVM_CHECK_EXTENSION for this capabili 8476 Calling KVM_CHECK_EXTENSION for this capability will return a bitmask
8836 of hypercalls that can be configured to exit 8477 of hypercalls that can be configured to exit to userspace.
8837 Right now, the only such hypercall is KVM_HC_ 8478 Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE.
8838 8479
8839 The argument to KVM_ENABLE_CAP is also a bitm 8480 The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset
8840 of the result of KVM_CHECK_EXTENSION. KVM wi 8481 of the result of KVM_CHECK_EXTENSION. KVM will forward to userspace
8841 the hypercalls whose corresponding bit is in 8482 the hypercalls whose corresponding bit is in the argument, and return
8842 ENOSYS for the others. 8483 ENOSYS for the others.
8843 8484
8844 8.35 KVM_CAP_PMU_CAPABILITY 8485 8.35 KVM_CAP_PMU_CAPABILITY
8845 --------------------------- 8486 ---------------------------
8846 8487
8847 :Capability: KVM_CAP_PMU_CAPABILITY 8488 :Capability: KVM_CAP_PMU_CAPABILITY
8848 :Architectures: x86 8489 :Architectures: x86
8849 :Type: vm 8490 :Type: vm
8850 :Parameters: arg[0] is bitmask of PMU virtual 8491 :Parameters: arg[0] is bitmask of PMU virtualization capabilities.
8851 :Returns: 0 on success, -EINVAL when arg[0] c 8492 :Returns: 0 on success, -EINVAL when arg[0] contains invalid bits
8852 8493
8853 This capability alters PMU virtualization in 8494 This capability alters PMU virtualization in KVM.
8854 8495
8855 Calling KVM_CHECK_EXTENSION for this capabili 8496 Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of
8856 PMU virtualization capabilities that can be a 8497 PMU virtualization capabilities that can be adjusted on a VM.
8857 8498
8858 The argument to KVM_ENABLE_CAP is also a bitm 8499 The argument to KVM_ENABLE_CAP is also a bitmask and selects specific
8859 PMU virtualization capabilities to be applied 8500 PMU virtualization capabilities to be applied to the VM. This can
8860 only be invoked on a VM prior to the creation 8501 only be invoked on a VM prior to the creation of VCPUs.
8861 8502
8862 At this time, KVM_PMU_CAP_DISABLE is the only 8503 At this time, KVM_PMU_CAP_DISABLE is the only capability. Setting
8863 this capability will disable PMU virtualizati 8504 this capability will disable PMU virtualization for that VM. Usermode
8864 should adjust CPUID leaf 0xA to reflect that 8505 should adjust CPUID leaf 0xA to reflect that the PMU is disabled.
8865 8506
8866 8.36 KVM_CAP_ARM_SYSTEM_SUSPEND 8507 8.36 KVM_CAP_ARM_SYSTEM_SUSPEND
8867 ------------------------------- 8508 -------------------------------
8868 8509
8869 :Capability: KVM_CAP_ARM_SYSTEM_SUSPEND 8510 :Capability: KVM_CAP_ARM_SYSTEM_SUSPEND
8870 :Architectures: arm64 8511 :Architectures: arm64
8871 :Type: vm 8512 :Type: vm
8872 8513
8873 When enabled, KVM will exit to userspace with 8514 When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of
8874 type KVM_SYSTEM_EVENT_SUSPEND to process the 8515 type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request.
8875 8516
8876 8.37 KVM_CAP_S390_PROTECTED_DUMP 8517 8.37 KVM_CAP_S390_PROTECTED_DUMP
8877 -------------------------------- 8518 --------------------------------
8878 8519
8879 :Capability: KVM_CAP_S390_PROTECTED_DUMP 8520 :Capability: KVM_CAP_S390_PROTECTED_DUMP
8880 :Architectures: s390 8521 :Architectures: s390
8881 :Type: vm 8522 :Type: vm
8882 8523
8883 This capability indicates that KVM and the Ul 8524 This capability indicates that KVM and the Ultravisor support dumping
8884 PV guests. The `KVM_PV_DUMP` command is avail 8525 PV guests. The `KVM_PV_DUMP` command is available for the
8885 `KVM_S390_PV_COMMAND` ioctl and the `KVM_PV_I 8526 `KVM_S390_PV_COMMAND` ioctl and the `KVM_PV_INFO` command provides
8886 dump related UV data. Also the vcpu ioctl `KV 8527 dump related UV data. Also the vcpu ioctl `KVM_S390_PV_CPU_COMMAND` is
8887 available and supports the `KVM_PV_DUMP_CPU` 8528 available and supports the `KVM_PV_DUMP_CPU` subcommand.
8888 8529
8889 8.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES 8530 8.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES
8890 ------------------------------------- 8531 -------------------------------------
8891 8532
8892 :Capability: KVM_CAP_VM_DISABLE_NX_HUGE_PAGES 8533 :Capability: KVM_CAP_VM_DISABLE_NX_HUGE_PAGES
8893 :Architectures: x86 8534 :Architectures: x86
8894 :Type: vm 8535 :Type: vm
8895 :Parameters: arg[0] must be 0. 8536 :Parameters: arg[0] must be 0.
8896 :Returns: 0 on success, -EPERM if the userspa 8537 :Returns: 0 on success, -EPERM if the userspace process does not
8897 have CAP_SYS_BOOT, -EINVAL if args[ 8538 have CAP_SYS_BOOT, -EINVAL if args[0] is not 0 or any vCPUs have been
8898 created. 8539 created.
8899 8540
8900 This capability disables the NX huge pages mi 8541 This capability disables the NX huge pages mitigation for iTLB MULTIHIT.
8901 8542
8902 The capability has no effect if the nx_huge_p 8543 The capability has no effect if the nx_huge_pages module parameter is not set.
8903 8544
8904 This capability may only be set before any vC 8545 This capability may only be set before any vCPUs are created.
8905 8546
8906 8.39 KVM_CAP_S390_CPU_TOPOLOGY 8547 8.39 KVM_CAP_S390_CPU_TOPOLOGY
8907 ------------------------------ 8548 ------------------------------
8908 8549
8909 :Capability: KVM_CAP_S390_CPU_TOPOLOGY 8550 :Capability: KVM_CAP_S390_CPU_TOPOLOGY
8910 :Architectures: s390 8551 :Architectures: s390
8911 :Type: vm 8552 :Type: vm
8912 8553
8913 This capability indicates that KVM will provi 8554 This capability indicates that KVM will provide the S390 CPU Topology
8914 facility which consist of the interpretation 8555 facility which consist of the interpretation of the PTF instruction for
8915 the function code 2 along with interception a 8556 the function code 2 along with interception and forwarding of both the
8916 PTF instruction with function codes 0 or 1 an 8557 PTF instruction with function codes 0 or 1 and the STSI(15,1,x)
8917 instruction to the userland hypervisor. 8558 instruction to the userland hypervisor.
8918 8559
8919 The stfle facility 11, CPU Topology facility, 8560 The stfle facility 11, CPU Topology facility, should not be indicated
8920 to the guest without this capability. 8561 to the guest without this capability.
8921 8562
8922 When this capability is present, KVM provides 8563 When this capability is present, KVM provides a new attribute group
8923 on vm fd, KVM_S390_VM_CPU_TOPOLOGY. 8564 on vm fd, KVM_S390_VM_CPU_TOPOLOGY.
8924 This new attribute allows to get, set or clea 8565 This new attribute allows to get, set or clear the Modified Change
8925 Topology Report (MTCR) bit of the SCA through 8566 Topology Report (MTCR) bit of the SCA through the kvm_device_attr
8926 structure. 8567 structure.
8927 8568
8928 When getting the Modified Change Topology Rep 8569 When getting the Modified Change Topology Report value, the attr->addr
8929 must point to a byte where the value will be 8570 must point to a byte where the value will be stored or retrieved from.
8930 8571
8931 8.40 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE 8572 8.40 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE
8932 --------------------------------------- 8573 ---------------------------------------
8933 8574
8934 :Capability: KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SI 8575 :Capability: KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE
8935 :Architectures: arm64 8576 :Architectures: arm64
8936 :Type: vm 8577 :Type: vm
8937 :Parameters: arg[0] is the new split chunk si 8578 :Parameters: arg[0] is the new split chunk size.
8938 :Returns: 0 on success, -EINVAL if any memslo 8579 :Returns: 0 on success, -EINVAL if any memslot was already created.
8939 8580
8940 This capability sets the chunk size used in E 8581 This capability sets the chunk size used in Eager Page Splitting.
8941 8582
8942 Eager Page Splitting improves the performance 8583 Eager Page Splitting improves the performance of dirty-logging (used
8943 in live migrations) when guest memory is back 8584 in live migrations) when guest memory is backed by huge-pages. It
8944 avoids splitting huge-pages (into PAGE_SIZE p 8585 avoids splitting huge-pages (into PAGE_SIZE pages) on fault, by doing
8945 it eagerly when enabling dirty logging (with 8586 it eagerly when enabling dirty logging (with the
8946 KVM_MEM_LOG_DIRTY_PAGES flag for a memory reg 8587 KVM_MEM_LOG_DIRTY_PAGES flag for a memory region), or when using
8947 KVM_CLEAR_DIRTY_LOG. 8588 KVM_CLEAR_DIRTY_LOG.
8948 8589
8949 The chunk size specifies how many pages to br 8590 The chunk size specifies how many pages to break at a time, using a
8950 single allocation for each chunk. Bigger the 8591 single allocation for each chunk. Bigger the chunk size, more pages
8951 need to be allocated ahead of time. 8592 need to be allocated ahead of time.
8952 8593
8953 The chunk size needs to be a valid block size 8594 The chunk size needs to be a valid block size. The list of acceptable
8954 block sizes is exposed in KVM_CAP_ARM_SUPPORT 8595 block sizes is exposed in KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES as a
8955 64-bit bitmap (each bit describing a block si 8596 64-bit bitmap (each bit describing a block size). The default value is
8956 0, to disable the eager page splitting. 8597 0, to disable the eager page splitting.
8957 <<
8958 8.41 KVM_CAP_VM_TYPES <<
8959 --------------------- <<
8960 <<
8961 :Capability: KVM_CAP_MEMORY_ATTRIBUTES <<
8962 :Architectures: x86 <<
8963 :Type: system ioctl <<
8964 <<
8965 This capability returns a bitmap of support V <<
8966 means the VM type with value @n is supported. <<
8967 <<
8968 #define KVM_X86_DEFAULT_VM 0 <<
8969 #define KVM_X86_SW_PROTECTED_VM 1 <<
8970 #define KVM_X86_SEV_VM 2 <<
8971 #define KVM_X86_SEV_ES_VM 3 <<
8972 <<
8973 Note, KVM_X86_SW_PROTECTED_VM is currently on <<
8974 Do not use KVM_X86_SW_PROTECTED_VM for "real" <<
8975 production. The behavior and effective ABI f <<
8976 unstable. <<
8977 8598
8978 9. Known KVM API problems 8599 9. Known KVM API problems
8979 ========================= 8600 =========================
8980 8601
8981 In some cases, KVM's API has some inconsisten 8602 In some cases, KVM's API has some inconsistencies or common pitfalls
8982 that userspace need to be aware of. This sec 8603 that userspace need to be aware of. This section details some of
8983 these issues. 8604 these issues.
8984 8605
8985 Most of them are architecture specific, so th 8606 Most of them are architecture specific, so the section is split by
8986 architecture. 8607 architecture.
8987 8608
8988 9.1. x86 8609 9.1. x86
8989 -------- 8610 --------
8990 8611
8991 ``KVM_GET_SUPPORTED_CPUID`` issues 8612 ``KVM_GET_SUPPORTED_CPUID`` issues
8992 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 8613 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8993 8614
8994 In general, ``KVM_GET_SUPPORTED_CPUID`` is de 8615 In general, ``KVM_GET_SUPPORTED_CPUID`` is designed so that it is possible
8995 to take its result and pass it directly to `` 8616 to take its result and pass it directly to ``KVM_SET_CPUID2``. This section
8996 documents some cases in which that requires s 8617 documents some cases in which that requires some care.
8997 8618
8998 Local APIC features 8619 Local APIC features
8999 ~~~~~~~~~~~~~~~~~~~ 8620 ~~~~~~~~~~~~~~~~~~~
9000 8621
9001 CPU[EAX=1]:ECX[21] (X2APIC) is reported by `` 8622 CPU[EAX=1]:ECX[21] (X2APIC) is reported by ``KVM_GET_SUPPORTED_CPUID``,
9002 but it can only be enabled if ``KVM_CREATE_IR 8623 but it can only be enabled if ``KVM_CREATE_IRQCHIP`` or
9003 ``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are 8624 ``KVM_ENABLE_CAP(KVM_CAP_IRQCHIP_SPLIT)`` are used to enable in-kernel emulation of
9004 the local APIC. 8625 the local APIC.
9005 8626
9006 The same is true for the ``KVM_FEATURE_PV_UNH 8627 The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature.
9007 8628
9008 CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not repo 8629 CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by ``KVM_GET_SUPPORTED_CPUID``.
9009 It can be enabled if ``KVM_CAP_TSC_DEADLINE_T 8630 It can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` is present and the kernel
9010 has enabled in-kernel emulation of the local 8631 has enabled in-kernel emulation of the local APIC.
9011 8632
9012 CPU topology 8633 CPU topology
9013 ~~~~~~~~~~~~ 8634 ~~~~~~~~~~~~
9014 8635
9015 Several CPUID values include topology informa 8636 Several CPUID values include topology information for the host CPU:
9016 0x0b and 0x1f for Intel systems, 0x8000001e f 8637 0x0b and 0x1f for Intel systems, 0x8000001e for AMD systems. Different
9017 versions of KVM return different values for t 8638 versions of KVM return different values for this information and userspace
9018 should not rely on it. Currently they return 8639 should not rely on it. Currently they return all zeroes.
9019 8640
9020 If userspace wishes to set up a guest topolog 8641 If userspace wishes to set up a guest topology, it should be careful that
9021 the values of these three leaves differ for e 8642 the values of these three leaves differ for each CPU. In particular,
9022 the APIC ID is found in EDX for all subleaves 8643 the APIC ID is found in EDX for all subleaves of 0x0b and 0x1f, and in EAX
9023 for 0x8000001e; the latter also encodes the c 8644 for 0x8000001e; the latter also encodes the core id and node id in bits
9024 7:0 of EBX and ECX respectively. 8645 7:0 of EBX and ECX respectively.
9025 8646
9026 Obsolete ioctls and capabilities 8647 Obsolete ioctls and capabilities
9027 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 8648 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
9028 8649
9029 KVM_CAP_DISABLE_QUIRKS does not let userspace 8650 KVM_CAP_DISABLE_QUIRKS does not let userspace know which quirks are actually
9030 available. Use ``KVM_CHECK_EXTENSION(KVM_CAP 8651 available. Use ``KVM_CHECK_EXTENSION(KVM_CAP_DISABLE_QUIRKS2)`` instead if
9031 available. 8652 available.
9032 8653
9033 Ordering of KVM_GET_*/KVM_SET_* ioctls 8654 Ordering of KVM_GET_*/KVM_SET_* ioctls
9034 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 8655 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
9035 8656
9036 TBD 8657 TBD
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.