Blame view
arch/x86/kvm/x86.c
156 KB
043405e10
|
1 2 3 4 5 6 |
/* * Kernel-based Virtual Machine driver for Linux * * derived from drivers/kvm/kvm_main.c * * Copyright (C) 2006 Qumranet, Inc. |
4d5c5d0fe
|
7 8 |
* Copyright (C) 2008 Qumranet, Inc. * Copyright IBM Corporation, 2008 |
9611c1877
|
9 |
* Copyright 2010 Red Hat, Inc. and/or its affiliates. |
043405e10
|
10 11 12 13 |
* * Authors: * Avi Kivity <avi@qumranet.com> * Yaniv Kamay <yaniv@qumranet.com> |
4d5c5d0fe
|
14 15 |
* Amit Shah <amit.shah@qumranet.com> * Ben-Ami Yassour <benami@il.ibm.com> |
043405e10
|
16 17 18 19 20 |
* * This work is licensed under the terms of the GNU GPL, version 2. See * the COPYING file in the top-level directory. * */ |
edf884172
|
21 |
#include <linux/kvm_host.h> |
313a3dc75
|
22 |
#include "irq.h" |
1d737c8a6
|
23 |
#include "mmu.h" |
7837699fa
|
24 |
#include "i8254.h" |
37817f298
|
25 |
#include "tss.h" |
5fdbf9765
|
26 |
#include "kvm_cache_regs.h" |
26eef70c3
|
27 |
#include "x86.h" |
00b27a3ef
|
28 |
#include "cpuid.h" |
313a3dc75
|
29 |
|
18068523d
|
30 |
#include <linux/clocksource.h> |
4d5c5d0fe
|
31 |
#include <linux/interrupt.h> |
313a3dc75
|
32 33 34 |
#include <linux/kvm.h> #include <linux/fs.h> #include <linux/vmalloc.h> |
5fb76f9be
|
35 |
#include <linux/module.h> |
0de10343b
|
36 |
#include <linux/mman.h> |
2bacc55c7
|
37 |
#include <linux/highmem.h> |
19de40a84
|
38 |
#include <linux/iommu.h> |
62c476c7c
|
39 |
#include <linux/intel-iommu.h> |
c80766040
|
40 |
#include <linux/cpufreq.h> |
18863bdd6
|
41 |
#include <linux/user-return-notifier.h> |
a983fb238
|
42 |
#include <linux/srcu.h> |
5a0e3ad6a
|
43 |
#include <linux/slab.h> |
ff9d07a0e
|
44 |
#include <linux/perf_event.h> |
7bee342a9
|
45 |
#include <linux/uaccess.h> |
af585b921
|
46 |
#include <linux/hash.h> |
a1b60c1cd
|
47 |
#include <linux/pci.h> |
aec51dc4f
|
48 |
#include <trace/events/kvm.h> |
2ed152afc
|
49 |
|
229456fc3
|
50 51 |
#define CREATE_TRACE_POINTS #include "trace.h" |
043405e10
|
52 |
|
24f1e32c6
|
53 |
#include <asm/debugreg.h> |
d825ed0a9
|
54 |
#include <asm/msr.h> |
a5f61300c
|
55 |
#include <asm/desc.h> |
0bed3b568
|
56 |
#include <asm/mtrr.h> |
890ca9aef
|
57 |
#include <asm/mce.h> |
7cf30855e
|
58 |
#include <asm/i387.h> |
98918833a
|
59 |
#include <asm/xcr.h> |
1d5f066e0
|
60 |
#include <asm/pvclock.h> |
217fc9cfc
|
61 |
#include <asm/div64.h> |
043405e10
|
62 |
|
313a3dc75
|
63 |
#define MAX_IO_MSRS 256 |
890ca9aef
|
64 |
#define KVM_MAX_MCE_BANKS 32 |
5854dbca9
|
65 |
#define KVM_MCE_CAP_SUPPORTED (MCG_CTL_P | MCG_SER_P) |
890ca9aef
|
66 |
|
0f65dd70a
|
67 68 |
#define emul_to_vcpu(ctxt) \ container_of(ctxt, struct kvm_vcpu, arch.emulate_ctxt) |
50a37eb4e
|
69 70 71 72 73 |
/* EFER defaults: * - enable syscall per default because its emulated by KVM * - enable LME and LMA per default on 64 bit KVM */ #ifdef CONFIG_X86_64 |
1260edbe7
|
74 75 |
static u64 __read_mostly efer_reserved_bits = ~((u64)(EFER_SCE | EFER_LME | EFER_LMA)); |
50a37eb4e
|
76 |
#else |
1260edbe7
|
77 |
static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE); |
50a37eb4e
|
78 |
#endif |
313a3dc75
|
79 |
|
ba1389b7a
|
80 81 |
#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU |
417bc3041
|
82 |
|
cb142eb74
|
83 |
static void update_cr8_intercept(struct kvm_vcpu *vcpu); |
7460fb4a3
|
84 |
static void process_nmi(struct kvm_vcpu *vcpu); |
674eea0fc
|
85 |
|
97896d04a
|
86 |
struct kvm_x86_ops *kvm_x86_ops; |
5fdbf9765
|
87 |
EXPORT_SYMBOL_GPL(kvm_x86_ops); |
97896d04a
|
88 |
|
476bc0015
|
89 90 |
static bool ignore_msrs = 0; module_param(ignore_msrs, bool, S_IRUGO | S_IWUSR); |
ed85c0685
|
91 |
|
92a1f12d2
|
92 93 94 95 |
bool kvm_has_tsc_control; EXPORT_SYMBOL_GPL(kvm_has_tsc_control); u32 kvm_max_guest_tsc_khz; EXPORT_SYMBOL_GPL(kvm_max_guest_tsc_khz); |
18863bdd6
|
96 97 98 99 |
#define KVM_NR_SHARED_MSRS 16 struct kvm_shared_msrs_global { int nr; |
2bf78fa7b
|
100 |
u32 msrs[KVM_NR_SHARED_MSRS]; |
18863bdd6
|
101 102 103 104 105 |
}; struct kvm_shared_msrs { struct user_return_notifier urn; bool registered; |
2bf78fa7b
|
106 107 108 109 |
struct kvm_shared_msr_values { u64 host; u64 curr; } values[KVM_NR_SHARED_MSRS]; |
18863bdd6
|
110 111 112 113 |
}; static struct kvm_shared_msrs_global __read_mostly shared_msrs_global; static DEFINE_PER_CPU(struct kvm_shared_msrs, shared_msrs); |
417bc3041
|
114 |
struct kvm_stats_debugfs_item debugfs_entries[] = { |
ba1389b7a
|
115 116 117 118 119 120 121 122 123 |
{ "pf_fixed", VCPU_STAT(pf_fixed) }, { "pf_guest", VCPU_STAT(pf_guest) }, { "tlb_flush", VCPU_STAT(tlb_flush) }, { "invlpg", VCPU_STAT(invlpg) }, { "exits", VCPU_STAT(exits) }, { "io_exits", VCPU_STAT(io_exits) }, { "mmio_exits", VCPU_STAT(mmio_exits) }, { "signal_exits", VCPU_STAT(signal_exits) }, { "irq_window", VCPU_STAT(irq_window_exits) }, |
f08864b42
|
124 |
{ "nmi_window", VCPU_STAT(nmi_window_exits) }, |
ba1389b7a
|
125 126 |
{ "halt_exits", VCPU_STAT(halt_exits) }, { "halt_wakeup", VCPU_STAT(halt_wakeup) }, |
f11c3a8d8
|
127 |
{ "hypercalls", VCPU_STAT(hypercalls) }, |
ba1389b7a
|
128 129 130 131 132 133 134 |
{ "request_irq", VCPU_STAT(request_irq_exits) }, { "irq_exits", VCPU_STAT(irq_exits) }, { "host_state_reload", VCPU_STAT(host_state_reload) }, { "efer_reload", VCPU_STAT(efer_reload) }, { "fpu_reload", VCPU_STAT(fpu_reload) }, { "insn_emulation", VCPU_STAT(insn_emulation) }, { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, |
fa89a8176
|
135 |
{ "irq_injections", VCPU_STAT(irq_injections) }, |
c4abb7c9c
|
136 |
{ "nmi_injections", VCPU_STAT(nmi_injections) }, |
4cee57649
|
137 138 139 140 141 142 |
{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, { "mmu_pte_write", VM_STAT(mmu_pte_write) }, { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, { "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) }, { "mmu_flooded", VM_STAT(mmu_flooded) }, { "mmu_recycled", VM_STAT(mmu_recycled) }, |
dfc5aa00c
|
143 |
{ "mmu_cache_miss", VM_STAT(mmu_cache_miss) }, |
4731d4c7a
|
144 |
{ "mmu_unsync", VM_STAT(mmu_unsync) }, |
0f74a24c5
|
145 |
{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, |
05da45583
|
146 |
{ "largepages", VM_STAT(lpages) }, |
417bc3041
|
147 148 |
{ NULL } }; |
2acf923e3
|
149 |
u64 __read_mostly host_xcr0; |
d6aa10003
|
150 |
int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt); |
af585b921
|
151 152 153 154 155 156 |
static inline void kvm_async_pf_hash_reset(struct kvm_vcpu *vcpu) { int i; for (i = 0; i < roundup_pow_of_two(ASYNC_PF_PER_VCPU); i++) vcpu->arch.apf.gfns[i] = ~0; } |
18863bdd6
|
157 158 159 |
static void kvm_on_user_return(struct user_return_notifier *urn) { unsigned slot; |
18863bdd6
|
160 161 |
struct kvm_shared_msrs *locals = container_of(urn, struct kvm_shared_msrs, urn); |
2bf78fa7b
|
162 |
struct kvm_shared_msr_values *values; |
18863bdd6
|
163 164 |
for (slot = 0; slot < shared_msrs_global.nr; ++slot) { |
2bf78fa7b
|
165 166 167 168 |
values = &locals->values[slot]; if (values->host != values->curr) { wrmsrl(shared_msrs_global.msrs[slot], values->host); values->curr = values->host; |
18863bdd6
|
169 170 171 172 173 |
} } locals->registered = false; user_return_notifier_unregister(urn); } |
2bf78fa7b
|
174 |
static void shared_msr_update(unsigned slot, u32 msr) |
18863bdd6
|
175 |
{ |
2bf78fa7b
|
176 |
struct kvm_shared_msrs *smsr; |
18863bdd6
|
177 |
u64 value; |
2bf78fa7b
|
178 179 180 181 182 183 184 185 186 187 188 189 190 191 |
smsr = &__get_cpu_var(shared_msrs); /* only read, and nobody should modify it at this time, * so don't need lock */ if (slot >= shared_msrs_global.nr) { printk(KERN_ERR "kvm: invalid MSR slot!"); return; } rdmsrl_safe(msr, &value); smsr->values[slot].host = value; smsr->values[slot].curr = value; } void kvm_define_shared_msr(unsigned slot, u32 msr) { |
18863bdd6
|
192 193 |
if (slot >= shared_msrs_global.nr) shared_msrs_global.nr = slot + 1; |
2bf78fa7b
|
194 195 196 |
shared_msrs_global.msrs[slot] = msr; /* we need ensured the shared_msr_global have been updated */ smp_wmb(); |
18863bdd6
|
197 198 199 200 201 202 |
} EXPORT_SYMBOL_GPL(kvm_define_shared_msr); static void kvm_shared_msr_cpu_online(void) { unsigned i; |
18863bdd6
|
203 204 |
for (i = 0; i < shared_msrs_global.nr; ++i) |
2bf78fa7b
|
205 |
shared_msr_update(i, shared_msrs_global.msrs[i]); |
18863bdd6
|
206 |
} |
d5696725b
|
207 |
void kvm_set_shared_msr(unsigned slot, u64 value, u64 mask) |
18863bdd6
|
208 209 |
{ struct kvm_shared_msrs *smsr = &__get_cpu_var(shared_msrs); |
2bf78fa7b
|
210 |
if (((value ^ smsr->values[slot].curr) & mask) == 0) |
18863bdd6
|
211 |
return; |
2bf78fa7b
|
212 213 |
smsr->values[slot].curr = value; wrmsrl(shared_msrs_global.msrs[slot], value); |
18863bdd6
|
214 215 216 217 218 219 220 |
if (!smsr->registered) { smsr->urn.on_user_return = kvm_on_user_return; user_return_notifier_register(&smsr->urn); smsr->registered = true; } } EXPORT_SYMBOL_GPL(kvm_set_shared_msr); |
3548bab50
|
221 222 223 224 225 226 227 |
static void drop_user_return_notifiers(void *ignore) { struct kvm_shared_msrs *smsr = &__get_cpu_var(shared_msrs); if (smsr->registered) kvm_on_user_return(&smsr->urn); } |
6866b83ed
|
228 229 230 |
u64 kvm_get_apic_base(struct kvm_vcpu *vcpu) { if (irqchip_in_kernel(vcpu->kvm)) |
ad312c7c7
|
231 |
return vcpu->arch.apic_base; |
6866b83ed
|
232 |
else |
ad312c7c7
|
233 |
return vcpu->arch.apic_base; |
6866b83ed
|
234 235 236 237 238 239 240 241 242 |
} EXPORT_SYMBOL_GPL(kvm_get_apic_base); void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data) { /* TODO: reserve bits check */ if (irqchip_in_kernel(vcpu->kvm)) kvm_lapic_set_base(vcpu, data); else |
ad312c7c7
|
243 |
vcpu->arch.apic_base = data; |
6866b83ed
|
244 245 |
} EXPORT_SYMBOL_GPL(kvm_set_apic_base); |
3fd28fce7
|
246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
#define EXCPT_BENIGN 0 #define EXCPT_CONTRIBUTORY 1 #define EXCPT_PF 2 static int exception_class(int vector) { switch (vector) { case PF_VECTOR: return EXCPT_PF; case DE_VECTOR: case TS_VECTOR: case NP_VECTOR: case SS_VECTOR: case GP_VECTOR: return EXCPT_CONTRIBUTORY; default: break; } return EXCPT_BENIGN; } static void kvm_multiple_exception(struct kvm_vcpu *vcpu, |
ce7ddec4b
|
268 269 |
unsigned nr, bool has_error, u32 error_code, bool reinject) |
3fd28fce7
|
270 271 272 |
{ u32 prev_nr; int class1, class2; |
3842d135f
|
273 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
3fd28fce7
|
274 275 276 277 278 279 |
if (!vcpu->arch.exception.pending) { queue: vcpu->arch.exception.pending = true; vcpu->arch.exception.has_error_code = has_error; vcpu->arch.exception.nr = nr; vcpu->arch.exception.error_code = error_code; |
3f0fd2927
|
280 |
vcpu->arch.exception.reinject = reinject; |
3fd28fce7
|
281 282 283 284 285 286 287 |
return; } /* to check exception */ prev_nr = vcpu->arch.exception.nr; if (prev_nr == DF_VECTOR) { /* triple fault -> shutdown */ |
a8eeb04a4
|
288 |
kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); |
3fd28fce7
|
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 |
return; } class1 = exception_class(prev_nr); class2 = exception_class(nr); if ((class1 == EXCPT_CONTRIBUTORY && class2 == EXCPT_CONTRIBUTORY) || (class1 == EXCPT_PF && class2 != EXCPT_BENIGN)) { /* generate double fault per SDM Table 5-5 */ vcpu->arch.exception.pending = true; vcpu->arch.exception.has_error_code = true; vcpu->arch.exception.nr = DF_VECTOR; vcpu->arch.exception.error_code = 0; } else /* replace previous exception with a new one in a hope that instruction re-execution will regenerate lost exception */ goto queue; } |
298101da2
|
306 307 |
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr) { |
ce7ddec4b
|
308 |
kvm_multiple_exception(vcpu, nr, false, 0, false); |
298101da2
|
309 310 |
} EXPORT_SYMBOL_GPL(kvm_queue_exception); |
ce7ddec4b
|
311 312 313 314 315 |
void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr) { kvm_multiple_exception(vcpu, nr, false, 0, true); } EXPORT_SYMBOL_GPL(kvm_requeue_exception); |
db8fcefaa
|
316 |
void kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err) |
c3c91fee5
|
317 |
{ |
db8fcefaa
|
318 319 320 321 322 323 |
if (err) kvm_inject_gp(vcpu, 0); else kvm_x86_ops->skip_emulated_instruction(vcpu); } EXPORT_SYMBOL_GPL(kvm_complete_insn_gp); |
8df25a328
|
324 |
|
6389ee946
|
325 |
void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault) |
c3c91fee5
|
326 327 |
{ ++vcpu->stat.pf_guest; |
6389ee946
|
328 329 |
vcpu->arch.cr2 = fault->address; kvm_queue_exception_e(vcpu, PF_VECTOR, fault->error_code); |
c3c91fee5
|
330 |
} |
27d6c8652
|
331 |
EXPORT_SYMBOL_GPL(kvm_inject_page_fault); |
c3c91fee5
|
332 |
|
6389ee946
|
333 |
void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault) |
d4f8cf664
|
334 |
{ |
6389ee946
|
335 336 |
if (mmu_is_nested(vcpu) && !fault->nested_page_fault) vcpu->arch.nested_mmu.inject_page_fault(vcpu, fault); |
d4f8cf664
|
337 |
else |
6389ee946
|
338 |
vcpu->arch.mmu.inject_page_fault(vcpu, fault); |
d4f8cf664
|
339 |
} |
3419ffc8e
|
340 341 |
void kvm_inject_nmi(struct kvm_vcpu *vcpu) { |
7460fb4a3
|
342 343 |
atomic_inc(&vcpu->arch.nmi_queued); kvm_make_request(KVM_REQ_NMI, vcpu); |
3419ffc8e
|
344 345 |
} EXPORT_SYMBOL_GPL(kvm_inject_nmi); |
298101da2
|
346 347 |
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code) { |
ce7ddec4b
|
348 |
kvm_multiple_exception(vcpu, nr, true, error_code, false); |
298101da2
|
349 350 |
} EXPORT_SYMBOL_GPL(kvm_queue_exception_e); |
ce7ddec4b
|
351 352 353 354 355 |
void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code) { kvm_multiple_exception(vcpu, nr, true, error_code, true); } EXPORT_SYMBOL_GPL(kvm_requeue_exception_e); |
a03490ed2
|
356 |
/* |
0a79b0095
|
357 358 359 360 |
* Checks if cpl <= required_cpl; if true, return true. Otherwise queue * a #GP and return false. */ bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl) |
298101da2
|
361 |
{ |
0a79b0095
|
362 363 364 365 |
if (kvm_x86_ops->get_cpl(vcpu) <= required_cpl) return true; kvm_queue_exception_e(vcpu, GP_VECTOR, 0); return false; |
298101da2
|
366 |
} |
0a79b0095
|
367 |
EXPORT_SYMBOL_GPL(kvm_require_cpl); |
298101da2
|
368 |
|
a03490ed2
|
369 |
/* |
ec92fe44e
|
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 |
* This function will be used to read from the physical memory of the currently * running guest. The difference to kvm_read_guest_page is that this function * can read from guest physical or from the guest's guest physical memory. */ int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, gfn_t ngfn, void *data, int offset, int len, u32 access) { gfn_t real_gfn; gpa_t ngpa; ngpa = gfn_to_gpa(ngfn); real_gfn = mmu->translate_gpa(vcpu, ngpa, access); if (real_gfn == UNMAPPED_GVA) return -EFAULT; real_gfn = gpa_to_gfn(real_gfn); return kvm_read_guest_page(vcpu->kvm, real_gfn, data, offset, len); } EXPORT_SYMBOL_GPL(kvm_read_guest_page_mmu); |
3d06b8bfd
|
391 392 393 394 395 396 |
int kvm_read_nested_guest_page(struct kvm_vcpu *vcpu, gfn_t gfn, void *data, int offset, int len, u32 access) { return kvm_read_guest_page_mmu(vcpu, vcpu->arch.walk_mmu, gfn, data, offset, len, access); } |
ec92fe44e
|
397 |
/* |
a03490ed2
|
398 399 |
* Load the pae pdptrs. Return true is they are all valid. */ |
ff03a073e
|
400 |
int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3) |
a03490ed2
|
401 402 403 404 405 |
{ gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT; unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2; int i; int ret; |
ff03a073e
|
406 |
u64 pdpte[ARRAY_SIZE(mmu->pdptrs)]; |
a03490ed2
|
407 |
|
ff03a073e
|
408 409 410 |
ret = kvm_read_guest_page_mmu(vcpu, mmu, pdpt_gfn, pdpte, offset * sizeof(u64), sizeof(pdpte), PFERR_USER_MASK|PFERR_WRITE_MASK); |
a03490ed2
|
411 412 413 414 415 |
if (ret < 0) { ret = 0; goto out; } for (i = 0; i < ARRAY_SIZE(pdpte); ++i) { |
43a3795a3
|
416 |
if (is_present_gpte(pdpte[i]) && |
20c466b56
|
417 |
(pdpte[i] & vcpu->arch.mmu.rsvd_bits_mask[0][2])) { |
a03490ed2
|
418 419 420 421 422 |
ret = 0; goto out; } } ret = 1; |
ff03a073e
|
423 |
memcpy(mmu->pdptrs, pdpte, sizeof(mmu->pdptrs)); |
6de4f3ada
|
424 425 426 427 |
__set_bit(VCPU_EXREG_PDPTR, (unsigned long *)&vcpu->arch.regs_avail); __set_bit(VCPU_EXREG_PDPTR, (unsigned long *)&vcpu->arch.regs_dirty); |
a03490ed2
|
428 |
out: |
a03490ed2
|
429 430 431 |
return ret; } |
cc4b6871e
|
432 |
EXPORT_SYMBOL_GPL(load_pdptrs); |
a03490ed2
|
433 |
|
d835dfecd
|
434 435 |
static bool pdptrs_changed(struct kvm_vcpu *vcpu) { |
ff03a073e
|
436 |
u64 pdpte[ARRAY_SIZE(vcpu->arch.walk_mmu->pdptrs)]; |
d835dfecd
|
437 |
bool changed = true; |
3d06b8bfd
|
438 439 |
int offset; gfn_t gfn; |
d835dfecd
|
440 441 442 443 |
int r; if (is_long_mode(vcpu) || !is_pae(vcpu)) return false; |
6de4f3ada
|
444 445 446 |
if (!test_bit(VCPU_EXREG_PDPTR, (unsigned long *)&vcpu->arch.regs_avail)) return true; |
9f8fe5043
|
447 448 |
gfn = (kvm_read_cr3(vcpu) & ~31u) >> PAGE_SHIFT; offset = (kvm_read_cr3(vcpu) & ~31u) & (PAGE_SIZE - 1); |
3d06b8bfd
|
449 450 |
r = kvm_read_nested_guest_page(vcpu, gfn, pdpte, offset, sizeof(pdpte), PFERR_USER_MASK | PFERR_WRITE_MASK); |
d835dfecd
|
451 452 |
if (r < 0) goto out; |
ff03a073e
|
453 |
changed = memcmp(pdpte, vcpu->arch.walk_mmu->pdptrs, sizeof(pdpte)) != 0; |
d835dfecd
|
454 |
out: |
d835dfecd
|
455 456 457 |
return changed; } |
49a9b07ed
|
458 |
int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) |
a03490ed2
|
459 |
{ |
aad827034
|
460 461 462 |
unsigned long old_cr0 = kvm_read_cr0(vcpu); unsigned long update_bits = X86_CR0_PG | X86_CR0_WP | X86_CR0_CD | X86_CR0_NW; |
f9a48e6a1
|
463 |
cr0 |= X86_CR0_ET; |
ab344828e
|
464 |
#ifdef CONFIG_X86_64 |
0f12244fe
|
465 466 |
if (cr0 & 0xffffffff00000000UL) return 1; |
ab344828e
|
467 468 469 |
#endif cr0 &= ~CR0_RESERVED_BITS; |
a03490ed2
|
470 |
|
0f12244fe
|
471 472 |
if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) return 1; |
a03490ed2
|
473 |
|
0f12244fe
|
474 475 |
if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) return 1; |
a03490ed2
|
476 477 478 |
if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) { #ifdef CONFIG_X86_64 |
f6801dff2
|
479 |
if ((vcpu->arch.efer & EFER_LME)) { |
a03490ed2
|
480 |
int cs_db, cs_l; |
0f12244fe
|
481 482 |
if (!is_pae(vcpu)) return 1; |
a03490ed2
|
483 |
kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); |
0f12244fe
|
484 485 |
if (cs_l) return 1; |
a03490ed2
|
486 487 |
} else #endif |
ff03a073e
|
488 |
if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.walk_mmu, |
9f8fe5043
|
489 |
kvm_read_cr3(vcpu))) |
0f12244fe
|
490 |
return 1; |
a03490ed2
|
491 492 493 |
} kvm_x86_ops->set_cr0(vcpu, cr0); |
a03490ed2
|
494 |
|
d170c4190
|
495 |
if ((cr0 ^ old_cr0) & X86_CR0_PG) { |
e5f3f0279
|
496 |
kvm_clear_async_pf_completion_queue(vcpu); |
d170c4190
|
497 498 |
kvm_async_pf_hash_reset(vcpu); } |
e5f3f0279
|
499 |
|
aad827034
|
500 501 |
if ((cr0 ^ old_cr0) & update_bits) kvm_mmu_reset_context(vcpu); |
0f12244fe
|
502 503 |
return 0; } |
2d3ad1f40
|
504 |
EXPORT_SYMBOL_GPL(kvm_set_cr0); |
a03490ed2
|
505 |
|
2d3ad1f40
|
506 |
void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw) |
a03490ed2
|
507 |
{ |
49a9b07ed
|
508 |
(void)kvm_set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~0x0eul) | (msw & 0x0f)); |
a03490ed2
|
509 |
} |
2d3ad1f40
|
510 |
EXPORT_SYMBOL_GPL(kvm_lmsw); |
a03490ed2
|
511 |
|
2acf923e3
|
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 |
int __kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr) { u64 xcr0; /* Only support XCR_XFEATURE_ENABLED_MASK(xcr0) now */ if (index != XCR_XFEATURE_ENABLED_MASK) return 1; xcr0 = xcr; if (kvm_x86_ops->get_cpl(vcpu) != 0) return 1; if (!(xcr0 & XSTATE_FP)) return 1; if ((xcr0 & XSTATE_YMM) && !(xcr0 & XSTATE_SSE)) return 1; if (xcr0 & ~host_xcr0) return 1; vcpu->arch.xcr0 = xcr0; vcpu->guest_xcr0_loaded = 0; return 0; } int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr) { if (__kvm_set_xcr(vcpu, index, xcr)) { kvm_inject_gp(vcpu, 0); return 1; } return 0; } EXPORT_SYMBOL_GPL(kvm_set_xcr); |
a83b29c6a
|
542 |
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) |
a03490ed2
|
543 |
{ |
fc78f5193
|
544 |
unsigned long old_cr4 = kvm_read_cr4(vcpu); |
c68b734fb
|
545 546 |
unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_SMEP; |
0f12244fe
|
547 548 |
if (cr4 & CR4_RESERVED_BITS) return 1; |
a03490ed2
|
549 |
|
2acf923e3
|
550 551 |
if (!guest_cpuid_has_xsave(vcpu) && (cr4 & X86_CR4_OSXSAVE)) return 1; |
c68b734fb
|
552 553 |
if (!guest_cpuid_has_smep(vcpu) && (cr4 & X86_CR4_SMEP)) return 1; |
74dc2b4ff
|
554 555 |
if (!guest_cpuid_has_fsgsbase(vcpu) && (cr4 & X86_CR4_RDWRGSFS)) return 1; |
a03490ed2
|
556 |
if (is_long_mode(vcpu)) { |
0f12244fe
|
557 558 |
if (!(cr4 & X86_CR4_PAE)) return 1; |
a2edf57f5
|
559 560 |
} else if (is_paging(vcpu) && (cr4 & X86_CR4_PAE) && ((cr4 ^ old_cr4) & pdptr_bits) |
9f8fe5043
|
561 562 |
&& !load_pdptrs(vcpu, vcpu->arch.walk_mmu, kvm_read_cr3(vcpu))) |
0f12244fe
|
563 |
return 1; |
5e1746d62
|
564 |
if (kvm_x86_ops->set_cr4(vcpu, cr4)) |
0f12244fe
|
565 |
return 1; |
a03490ed2
|
566 |
|
aad827034
|
567 568 |
if ((cr4 ^ old_cr4) & pdptr_bits) kvm_mmu_reset_context(vcpu); |
0f12244fe
|
569 |
|
2acf923e3
|
570 |
if ((cr4 ^ old_cr4) & X86_CR4_OSXSAVE) |
00b27a3ef
|
571 |
kvm_update_cpuid(vcpu); |
2acf923e3
|
572 |
|
0f12244fe
|
573 574 |
return 0; } |
2d3ad1f40
|
575 |
EXPORT_SYMBOL_GPL(kvm_set_cr4); |
a03490ed2
|
576 |
|
2390218b6
|
577 |
int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) |
a03490ed2
|
578 |
{ |
9f8fe5043
|
579 |
if (cr3 == kvm_read_cr3(vcpu) && !pdptrs_changed(vcpu)) { |
0ba73cdad
|
580 |
kvm_mmu_sync_roots(vcpu); |
d835dfecd
|
581 |
kvm_mmu_flush_tlb(vcpu); |
0f12244fe
|
582 |
return 0; |
d835dfecd
|
583 |
} |
a03490ed2
|
584 |
if (is_long_mode(vcpu)) { |
0f12244fe
|
585 586 |
if (cr3 & CR3_L_MODE_RESERVED_BITS) return 1; |
a03490ed2
|
587 588 |
} else { if (is_pae(vcpu)) { |
0f12244fe
|
589 590 |
if (cr3 & CR3_PAE_RESERVED_BITS) return 1; |
ff03a073e
|
591 592 |
if (is_paging(vcpu) && !load_pdptrs(vcpu, vcpu->arch.walk_mmu, cr3)) |
0f12244fe
|
593 |
return 1; |
a03490ed2
|
594 595 596 597 598 599 |
} /* * We don't check reserved bits in nonpae mode, because * this isn't enforced, and VMware depends on this. */ } |
a03490ed2
|
600 601 602 603 604 605 606 607 608 609 |
/* * Does the new cr3 value map to physical memory? (Note, we * catch an invalid cr3 even in real-mode, because it would * cause trouble later on when we turn on paging anyway.) * * A real CPU would silently accept an invalid cr3 and would * attempt to use it - with largely undefined (and often hard * to debug) behavior on the guest side. */ if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT))) |
0f12244fe
|
610 611 |
return 1; vcpu->arch.cr3 = cr3; |
aff48baa3
|
612 |
__set_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail); |
0f12244fe
|
613 614 615 |
vcpu->arch.mmu.new_cr3(vcpu); return 0; } |
2d3ad1f40
|
616 |
EXPORT_SYMBOL_GPL(kvm_set_cr3); |
a03490ed2
|
617 |
|
eea1cff9a
|
618 |
int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) |
a03490ed2
|
619 |
{ |
0f12244fe
|
620 621 |
if (cr8 & CR8_RESERVED_BITS) return 1; |
a03490ed2
|
622 623 624 |
if (irqchip_in_kernel(vcpu->kvm)) kvm_lapic_set_tpr(vcpu, cr8); else |
ad312c7c7
|
625 |
vcpu->arch.cr8 = cr8; |
0f12244fe
|
626 627 |
return 0; } |
2d3ad1f40
|
628 |
EXPORT_SYMBOL_GPL(kvm_set_cr8); |
a03490ed2
|
629 |
|
2d3ad1f40
|
630 |
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu) |
a03490ed2
|
631 632 633 634 |
{ if (irqchip_in_kernel(vcpu->kvm)) return kvm_lapic_get_cr8(vcpu); else |
ad312c7c7
|
635 |
return vcpu->arch.cr8; |
a03490ed2
|
636 |
} |
2d3ad1f40
|
637 |
EXPORT_SYMBOL_GPL(kvm_get_cr8); |
a03490ed2
|
638 |
|
338dbc978
|
639 |
static int __kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val) |
020df0794
|
640 641 642 643 644 645 646 647 |
{ switch (dr) { case 0 ... 3: vcpu->arch.db[dr] = val; if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) vcpu->arch.eff_db[dr] = val; break; case 4: |
338dbc978
|
648 649 |
if (kvm_read_cr4_bits(vcpu, X86_CR4_DE)) return 1; /* #UD */ |
020df0794
|
650 651 |
/* fall through */ case 6: |
338dbc978
|
652 653 |
if (val & 0xffffffff00000000ULL) return -1; /* #GP */ |
020df0794
|
654 655 656 |
vcpu->arch.dr6 = (val & DR6_VOLATILE) | DR6_FIXED_1; break; case 5: |
338dbc978
|
657 658 |
if (kvm_read_cr4_bits(vcpu, X86_CR4_DE)) return 1; /* #UD */ |
020df0794
|
659 660 |
/* fall through */ default: /* 7 */ |
338dbc978
|
661 662 |
if (val & 0xffffffff00000000ULL) return -1; /* #GP */ |
020df0794
|
663 664 665 666 667 668 669 670 671 672 |
vcpu->arch.dr7 = (val & DR7_VOLATILE) | DR7_FIXED_1; if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) { kvm_x86_ops->set_dr7(vcpu, vcpu->arch.dr7); vcpu->arch.switch_db_regs = (val & DR7_BP_EN_MASK); } break; } return 0; } |
338dbc978
|
673 674 675 676 677 678 679 680 681 682 683 684 685 |
int kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val) { int res; res = __kvm_set_dr(vcpu, dr, val); if (res > 0) kvm_queue_exception(vcpu, UD_VECTOR); else if (res < 0) kvm_inject_gp(vcpu, 0); return res; } |
020df0794
|
686 |
EXPORT_SYMBOL_GPL(kvm_set_dr); |
338dbc978
|
687 |
static int _kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val) |
020df0794
|
688 689 690 691 692 693 |
{ switch (dr) { case 0 ... 3: *val = vcpu->arch.db[dr]; break; case 4: |
338dbc978
|
694 |
if (kvm_read_cr4_bits(vcpu, X86_CR4_DE)) |
020df0794
|
695 |
return 1; |
020df0794
|
696 697 698 699 700 |
/* fall through */ case 6: *val = vcpu->arch.dr6; break; case 5: |
338dbc978
|
701 |
if (kvm_read_cr4_bits(vcpu, X86_CR4_DE)) |
020df0794
|
702 |
return 1; |
020df0794
|
703 704 705 706 707 708 709 710 |
/* fall through */ default: /* 7 */ *val = vcpu->arch.dr7; break; } return 0; } |
338dbc978
|
711 712 713 714 715 716 717 718 719 |
int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val) { if (_kvm_get_dr(vcpu, dr, val)) { kvm_queue_exception(vcpu, UD_VECTOR); return 1; } return 0; } |
020df0794
|
720 |
EXPORT_SYMBOL_GPL(kvm_get_dr); |
022cd0e84
|
721 722 723 724 725 726 727 728 729 730 731 732 733 734 |
bool kvm_rdpmc(struct kvm_vcpu *vcpu) { u32 ecx = kvm_register_read(vcpu, VCPU_REGS_RCX); u64 data; int err; err = kvm_pmu_read_pmc(vcpu, ecx, &data); if (err) return err; kvm_register_write(vcpu, VCPU_REGS_RAX, (u32)data); kvm_register_write(vcpu, VCPU_REGS_RDX, data >> 32); return err; } EXPORT_SYMBOL_GPL(kvm_rdpmc); |
043405e10
|
735 736 737 738 739 |
/* * List of msr numbers which we expose to userspace through KVM_GET_MSRS * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST. * * This list is modified at module load time to reflect the |
e3267cbbb
|
740 741 |
* capabilities of the host cpu. This capabilities test skips MSRs that are * kvm-specific. Those are put in the beginning of the list. |
043405e10
|
742 |
*/ |
e3267cbbb
|
743 |
|
c9aaa8957
|
744 |
#define KVM_SAVE_MSRS_BEGIN 9 |
043405e10
|
745 |
static u32 msrs_to_save[] = { |
e3267cbbb
|
746 |
MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK, |
11c6bffa4
|
747 |
MSR_KVM_SYSTEM_TIME_NEW, MSR_KVM_WALL_CLOCK_NEW, |
55cd8e5a4
|
748 |
HV_X64_MSR_GUEST_OS_ID, HV_X64_MSR_HYPERCALL, |
c9aaa8957
|
749 |
HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, MSR_KVM_STEAL_TIME, |
043405e10
|
750 |
MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP, |
8c06585d6
|
751 |
MSR_STAR, |
043405e10
|
752 753 754 |
#ifdef CONFIG_X86_64 MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR, #endif |
e90aa41e6
|
755 |
MSR_IA32_TSC, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA |
043405e10
|
756 757 758 759 760 |
}; static unsigned num_msrs_to_save; static u32 emulated_msrs[] = { |
a3e06bbe8
|
761 |
MSR_IA32_TSCDEADLINE, |
043405e10
|
762 |
MSR_IA32_MISC_ENABLE, |
908e75f3e
|
763 764 |
MSR_IA32_MCG_STATUS, MSR_IA32_MCG_CTL, |
043405e10
|
765 |
}; |
b69e8caef
|
766 |
static int set_efer(struct kvm_vcpu *vcpu, u64 efer) |
15c4a6406
|
767 |
{ |
aad827034
|
768 |
u64 old_efer = vcpu->arch.efer; |
b69e8caef
|
769 770 |
if (efer & efer_reserved_bits) return 1; |
15c4a6406
|
771 772 |
if (is_paging(vcpu) |
b69e8caef
|
773 774 |
&& (vcpu->arch.efer & EFER_LME) != (efer & EFER_LME)) return 1; |
15c4a6406
|
775 |
|
1b2fd70c4
|
776 777 778 779 |
if (efer & EFER_FFXSR) { struct kvm_cpuid_entry2 *feat; feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0); |
b69e8caef
|
780 781 |
if (!feat || !(feat->edx & bit(X86_FEATURE_FXSR_OPT))) return 1; |
1b2fd70c4
|
782 |
} |
d80174745
|
783 784 785 786 |
if (efer & EFER_SVME) { struct kvm_cpuid_entry2 *feat; feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0); |
b69e8caef
|
787 788 |
if (!feat || !(feat->ecx & bit(X86_FEATURE_SVM))) return 1; |
d80174745
|
789 |
} |
15c4a6406
|
790 |
efer &= ~EFER_LMA; |
f6801dff2
|
791 |
efer |= vcpu->arch.efer & EFER_LMA; |
15c4a6406
|
792 |
|
a3d204e28
|
793 |
kvm_x86_ops->set_efer(vcpu, efer); |
9645bb56b
|
794 |
vcpu->arch.mmu.base_role.nxe = (efer & EFER_NX) && !tdp_enabled; |
b69e8caef
|
795 |
|
aad827034
|
796 797 798 |
/* Update reserved bits */ if ((efer ^ old_efer) & EFER_NX) kvm_mmu_reset_context(vcpu); |
b69e8caef
|
799 |
return 0; |
15c4a6406
|
800 |
} |
f2b4b7ddf
|
801 802 803 804 805 |
void kvm_enable_efer_bits(u64 mask) { efer_reserved_bits &= ~mask; } EXPORT_SYMBOL_GPL(kvm_enable_efer_bits); |
15c4a6406
|
806 807 808 809 810 811 812 813 814 |
/* * Writes msr value into into the appropriate "register". * Returns 0 on success, non-0 otherwise. * Assumes vcpu_load() was already called. */ int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) { return kvm_x86_ops->set_msr(vcpu, msr_index, data); } |
313a3dc75
|
815 816 817 818 819 820 821 |
/* * Adapt set_msr() to msr_io()'s calling convention */ static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data) { return kvm_set_msr(vcpu, index, *data); } |
18068523d
|
822 823 |
static void kvm_write_wall_clock(struct kvm *kvm, gpa_t wall_clock) { |
9ed3c444a
|
824 825 |
int version; int r; |
50d0a0f98
|
826 |
struct pvclock_wall_clock wc; |
923de3cf5
|
827 |
struct timespec boot; |
18068523d
|
828 829 830 |
if (!wall_clock) return; |
9ed3c444a
|
831 832 833 834 835 836 837 838 |
r = kvm_read_guest(kvm, wall_clock, &version, sizeof(version)); if (r) return; if (version & 1) ++version; /* first time write, random junk */ ++version; |
18068523d
|
839 |
|
18068523d
|
840 |
kvm_write_guest(kvm, wall_clock, &version, sizeof(version)); |
50d0a0f98
|
841 842 |
/* * The guest calculates current wall clock time by adding |
34c238a1d
|
843 |
* system time (updated by kvm_guest_time_update below) to the |
50d0a0f98
|
844 845 846 |
* wall clock specified here. guest system time equals host * system time for us, thus we must fill in host boot time here. */ |
923de3cf5
|
847 |
getboottime(&boot); |
50d0a0f98
|
848 849 850 851 |
wc.sec = boot.tv_sec; wc.nsec = boot.tv_nsec; wc.version = version; |
18068523d
|
852 853 854 855 856 |
kvm_write_guest(kvm, wall_clock, &wc, sizeof(wc)); version++; kvm_write_guest(kvm, wall_clock, &version, sizeof(version)); |
18068523d
|
857 |
} |
50d0a0f98
|
858 859 860 861 862 863 864 865 866 867 868 |
static uint32_t div_frac(uint32_t dividend, uint32_t divisor) { uint32_t quotient, remainder; /* Don't try to replace with do_div(), this one calculates * "(dividend << 32) / divisor" */ __asm__ ( "divl %4" : "=a" (quotient), "=d" (remainder) : "0" (0), "1" (dividend), "r" (divisor) ); return quotient; } |
5f4e3f882
|
869 870 |
static void kvm_get_time_scale(uint32_t scaled_khz, uint32_t base_khz, s8 *pshift, u32 *pmultiplier) |
50d0a0f98
|
871 |
{ |
5f4e3f882
|
872 |
uint64_t scaled64; |
50d0a0f98
|
873 874 875 |
int32_t shift = 0; uint64_t tps64; uint32_t tps32; |
5f4e3f882
|
876 877 |
tps64 = base_khz * 1000LL; scaled64 = scaled_khz * 1000LL; |
50933623e
|
878 |
while (tps64 > scaled64*2 || tps64 & 0xffffffff00000000ULL) { |
50d0a0f98
|
879 880 881 882 883 |
tps64 >>= 1; shift--; } tps32 = (uint32_t)tps64; |
50933623e
|
884 885 |
while (tps32 <= scaled64 || scaled64 & 0xffffffff00000000ULL) { if (scaled64 & 0xffffffff00000000ULL || tps32 & 0x80000000) |
5f4e3f882
|
886 887 888 |
scaled64 >>= 1; else tps32 <<= 1; |
50d0a0f98
|
889 890 |
shift++; } |
5f4e3f882
|
891 892 |
*pshift = shift; *pmultiplier = div_frac(scaled64, tps32); |
50d0a0f98
|
893 |
|
5f4e3f882
|
894 895 896 |
pr_debug("%s: base_khz %u => %u, shift %d, mul %u ", __func__, base_khz, scaled_khz, shift, *pmultiplier); |
50d0a0f98
|
897 |
} |
759379dd6
|
898 899 900 901 902 903 904 905 |
static inline u64 get_kernel_ns(void) { struct timespec ts; WARN_ON(preemptible()); ktime_get_ts(&ts); monotonic_to_bootbased(&ts); return timespec_to_ns(&ts); |
50d0a0f98
|
906 |
} |
c80766040
|
907 |
static DEFINE_PER_CPU(unsigned long, cpu_tsc_khz); |
c285545f8
|
908 |
unsigned long max_tsc_khz; |
c80766040
|
909 |
|
8cfdc0008
|
910 911 912 913 914 915 916 917 |
static inline int kvm_tsc_changes_freq(void) { int cpu = get_cpu(); int ret = !boot_cpu_has(X86_FEATURE_CONSTANT_TSC) && cpufreq_quick_get(cpu) != 0; put_cpu(); return ret; } |
a3e06bbe8
|
918 |
u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu) |
1e993611d
|
919 920 921 922 923 924 |
{ if (vcpu->arch.virtual_tsc_khz) return vcpu->arch.virtual_tsc_khz; else return __this_cpu_read(cpu_tsc_khz); } |
857e40999
|
925 |
static inline u64 nsec_to_cycles(struct kvm_vcpu *vcpu, u64 nsec) |
759379dd6
|
926 |
{ |
217fc9cfc
|
927 |
u64 ret; |
759379dd6
|
928 929 930 931 932 |
WARN_ON(preemptible()); if (kvm_tsc_changes_freq()) printk_once(KERN_WARNING "kvm: unreliable cycle conversion on adjustable rate TSC "); |
857e40999
|
933 |
ret = nsec * vcpu_tsc_khz(vcpu); |
217fc9cfc
|
934 935 |
do_div(ret, USEC_PER_SEC); return ret; |
759379dd6
|
936 |
} |
1e993611d
|
937 |
static void kvm_init_tsc_catchup(struct kvm_vcpu *vcpu, u32 this_tsc_khz) |
c285545f8
|
938 939 940 |
{ /* Compute a scale to convert nanoseconds in TSC cycles */ kvm_get_time_scale(this_tsc_khz, NSEC_PER_SEC / 1000, |
1e993611d
|
941 942 |
&vcpu->arch.tsc_catchup_shift, &vcpu->arch.tsc_catchup_mult); |
c285545f8
|
943 944 945 946 947 |
} static u64 compute_guest_tsc(struct kvm_vcpu *vcpu, s64 kernel_ns) { u64 tsc = pvclock_scale_delta(kernel_ns-vcpu->arch.last_tsc_nsec, |
1e993611d
|
948 949 |
vcpu->arch.tsc_catchup_mult, vcpu->arch.tsc_catchup_shift); |
c285545f8
|
950 951 952 |
tsc += vcpu->arch.last_tsc_write; return tsc; } |
99e3e30ae
|
953 954 955 |
void kvm_write_tsc(struct kvm_vcpu *vcpu, u64 data) { struct kvm *kvm = vcpu->kvm; |
f38e098ff
|
956 |
u64 offset, ns, elapsed; |
99e3e30ae
|
957 |
unsigned long flags; |
46543ba45
|
958 |
s64 sdiff; |
99e3e30ae
|
959 |
|
038f8c110
|
960 |
raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags); |
857e40999
|
961 |
offset = kvm_x86_ops->compute_tsc_offset(vcpu, data); |
759379dd6
|
962 |
ns = get_kernel_ns(); |
f38e098ff
|
963 |
elapsed = ns - kvm->arch.last_tsc_nsec; |
46543ba45
|
964 965 966 |
sdiff = data - kvm->arch.last_tsc_write; if (sdiff < 0) sdiff = -sdiff; |
f38e098ff
|
967 968 |
/* |
46543ba45
|
969 |
* Special case: close write to TSC within 5 seconds of |
f38e098ff
|
970 |
* another CPU is interpreted as an attempt to synchronize |
0d2eb44f6
|
971 |
* The 5 seconds is to accommodate host load / swapping as |
46543ba45
|
972 |
* well as any reset of TSC during the boot process. |
f38e098ff
|
973 974 |
* * In that case, for a reliable TSC, we can match TSC offsets, |
46543ba45
|
975 |
* or make a best guest using elapsed value. |
f38e098ff
|
976 |
*/ |
857e40999
|
977 |
if (sdiff < nsec_to_cycles(vcpu, 5ULL * NSEC_PER_SEC) && |
46543ba45
|
978 |
elapsed < 5ULL * NSEC_PER_SEC) { |
f38e098ff
|
979 980 981 982 983 |
if (!check_tsc_unstable()) { offset = kvm->arch.last_tsc_offset; pr_debug("kvm: matched tsc offset for %llu ", data); } else { |
857e40999
|
984 |
u64 delta = nsec_to_cycles(vcpu, elapsed); |
759379dd6
|
985 986 987 |
offset += delta; pr_debug("kvm: adjusted tsc offset by %llu ", delta); |
f38e098ff
|
988 989 990 991 992 993 |
} ns = kvm->arch.last_tsc_nsec; } kvm->arch.last_tsc_nsec = ns; kvm->arch.last_tsc_write = data; kvm->arch.last_tsc_offset = offset; |
99e3e30ae
|
994 |
kvm_x86_ops->write_tsc_offset(vcpu, offset); |
038f8c110
|
995 |
raw_spin_unlock_irqrestore(&kvm->arch.tsc_write_lock, flags); |
99e3e30ae
|
996 997 998 |
/* Reset of TSC must disable overshoot protection below */ vcpu->arch.hv_clock.tsc_timestamp = 0; |
c285545f8
|
999 1000 |
vcpu->arch.last_tsc_write = data; vcpu->arch.last_tsc_nsec = ns; |
99e3e30ae
|
1001 1002 |
} EXPORT_SYMBOL_GPL(kvm_write_tsc); |
34c238a1d
|
1003 |
static int kvm_guest_time_update(struct kvm_vcpu *v) |
18068523d
|
1004 |
{ |
18068523d
|
1005 1006 1007 |
unsigned long flags; struct kvm_vcpu_arch *vcpu = &v->arch; void *shared_kaddr; |
463656c00
|
1008 |
unsigned long this_tsc_khz; |
1d5f066e0
|
1009 1010 |
s64 kernel_ns, max_kernel_ns; u64 tsc_timestamp; |
18068523d
|
1011 |
|
18068523d
|
1012 1013 |
/* Keep irq disabled to prevent changes to the clock */ local_irq_save(flags); |
d5c1785d2
|
1014 |
tsc_timestamp = kvm_x86_ops->read_l1_tsc(v); |
759379dd6
|
1015 |
kernel_ns = get_kernel_ns(); |
1e993611d
|
1016 |
this_tsc_khz = vcpu_tsc_khz(v); |
8cfdc0008
|
1017 |
if (unlikely(this_tsc_khz == 0)) { |
c285545f8
|
1018 |
local_irq_restore(flags); |
34c238a1d
|
1019 |
kvm_make_request(KVM_REQ_CLOCK_UPDATE, v); |
8cfdc0008
|
1020 1021 |
return 1; } |
18068523d
|
1022 |
|
1d5f066e0
|
1023 |
/* |
c285545f8
|
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 |
* We may have to catch up the TSC to match elapsed wall clock * time for two reasons, even if kvmclock is used. * 1) CPU could have been running below the maximum TSC rate * 2) Broken TSC compensation resets the base at each VCPU * entry to avoid unknown leaps of TSC even when running * again on the same CPU. This may cause apparent elapsed * time to disappear, and the guest to stand still or run * very slowly. */ if (vcpu->tsc_catchup) { u64 tsc = compute_guest_tsc(v, kernel_ns); if (tsc > tsc_timestamp) { kvm_x86_ops->adjust_tsc_offset(v, tsc - tsc_timestamp); tsc_timestamp = tsc; } |
50d0a0f98
|
1039 |
} |
18068523d
|
1040 |
local_irq_restore(flags); |
c285545f8
|
1041 1042 |
if (!vcpu->time_page) return 0; |
18068523d
|
1043 |
|
c285545f8
|
1044 |
/* |
1d5f066e0
|
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 |
* Time as measured by the TSC may go backwards when resetting the base * tsc_timestamp. The reason for this is that the TSC resolution is * higher than the resolution of the other clock scales. Thus, many * possible measurments of the TSC correspond to one measurement of any * other clock, and so a spread of values is possible. This is not a * problem for the computation of the nanosecond clock; with TSC rates * around 1GHZ, there can only be a few cycles which correspond to one * nanosecond value, and any path through this code will inevitably * take longer than that. However, with the kernel_ns value itself, * the precision may be much lower, down to HZ granularity. If the * first sampling of TSC against kernel_ns ends in the low part of the * range, and the second in the high end of the range, we can get: * * (TSC - offset_low) * S + kns_old > (TSC - offset_high) * S + kns_new * * As the sampling errors potentially range in the thousands of cycles, * it is possible such a time value has already been observed by the * guest. To protect against this, we must compute the system time as * observed by the guest and ensure the new system time is greater. */ max_kernel_ns = 0; if (vcpu->hv_clock.tsc_timestamp && vcpu->last_guest_tsc) { max_kernel_ns = vcpu->last_guest_tsc - vcpu->hv_clock.tsc_timestamp; max_kernel_ns = pvclock_scale_delta(max_kernel_ns, vcpu->hv_clock.tsc_to_system_mul, vcpu->hv_clock.tsc_shift); max_kernel_ns += vcpu->last_kernel_ns; } |
afbcf7ab8
|
1074 |
|
e48672fa2
|
1075 |
if (unlikely(vcpu->hw_tsc_khz != this_tsc_khz)) { |
5f4e3f882
|
1076 1077 1078 |
kvm_get_time_scale(NSEC_PER_SEC / 1000, this_tsc_khz, &vcpu->hv_clock.tsc_shift, &vcpu->hv_clock.tsc_to_system_mul); |
e48672fa2
|
1079 |
vcpu->hw_tsc_khz = this_tsc_khz; |
8cfdc0008
|
1080 |
} |
1d5f066e0
|
1081 1082 |
if (max_kernel_ns > kernel_ns) kernel_ns = max_kernel_ns; |
8cfdc0008
|
1083 |
/* With all the info we got, fill in the values */ |
1d5f066e0
|
1084 |
vcpu->hv_clock.tsc_timestamp = tsc_timestamp; |
759379dd6
|
1085 |
vcpu->hv_clock.system_time = kernel_ns + v->kvm->arch.kvmclock_offset; |
1d5f066e0
|
1086 |
vcpu->last_kernel_ns = kernel_ns; |
28e4639ad
|
1087 |
vcpu->last_guest_tsc = tsc_timestamp; |
371bcf646
|
1088 |
vcpu->hv_clock.flags = 0; |
18068523d
|
1089 1090 1091 |
/* * The interface expects us to write an even number signaling that the * update is finished. Since the guest won't see the intermediate |
50d0a0f98
|
1092 |
* state, we just increase by 2 at the end. |
18068523d
|
1093 |
*/ |
50d0a0f98
|
1094 |
vcpu->hv_clock.version += 2; |
18068523d
|
1095 1096 1097 1098 |
shared_kaddr = kmap_atomic(vcpu->time_page, KM_USER0); memcpy(shared_kaddr + vcpu->time_offset, &vcpu->hv_clock, |
50d0a0f98
|
1099 |
sizeof(vcpu->hv_clock)); |
18068523d
|
1100 1101 1102 1103 |
kunmap_atomic(shared_kaddr, KM_USER0); mark_page_dirty(v->kvm, vcpu->time >> PAGE_SHIFT); |
8cfdc0008
|
1104 |
return 0; |
c80766040
|
1105 |
} |
9ba075a66
|
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 |
static bool msr_mtrr_valid(unsigned msr) { switch (msr) { case 0x200 ... 0x200 + 2 * KVM_NR_VAR_MTRR - 1: case MSR_MTRRfix64K_00000: case MSR_MTRRfix16K_80000: case MSR_MTRRfix16K_A0000: case MSR_MTRRfix4K_C0000: case MSR_MTRRfix4K_C8000: case MSR_MTRRfix4K_D0000: case MSR_MTRRfix4K_D8000: case MSR_MTRRfix4K_E0000: case MSR_MTRRfix4K_E8000: case MSR_MTRRfix4K_F0000: case MSR_MTRRfix4K_F8000: case MSR_MTRRdefType: case MSR_IA32_CR_PAT: return true; case 0x2f8: return true; } return false; } |
d6289b936
|
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 |
static bool valid_pat_type(unsigned t) { return t < 8 && (1 << t) & 0xf3; /* 0, 1, 4, 5, 6, 7 */ } static bool valid_mtrr_type(unsigned t) { return t < 8 && (1 << t) & 0x73; /* 0, 1, 4, 5, 6 */ } static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data) { int i; if (!msr_mtrr_valid(msr)) return false; if (msr == MSR_IA32_CR_PAT) { for (i = 0; i < 8; i++) if (!valid_pat_type((data >> (i * 8)) & 0xff)) return false; return true; } else if (msr == MSR_MTRRdefType) { if (data & ~0xcff) return false; return valid_mtrr_type(data & 0xff); } else if (msr >= MSR_MTRRfix64K_00000 && msr <= MSR_MTRRfix4K_F8000) { for (i = 0; i < 8 ; i++) if (!valid_mtrr_type((data >> (i * 8)) & 0xff)) return false; return true; } /* variable MTRRs */ return valid_mtrr_type(data & 0xff); } |
9ba075a66
|
1165 1166 |
static int set_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 data) { |
0bed3b568
|
1167 |
u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges; |
d6289b936
|
1168 |
if (!mtrr_valid(vcpu, msr, data)) |
9ba075a66
|
1169 |
return 1; |
0bed3b568
|
1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 |
if (msr == MSR_MTRRdefType) { vcpu->arch.mtrr_state.def_type = data; vcpu->arch.mtrr_state.enabled = (data & 0xc00) >> 10; } else if (msr == MSR_MTRRfix64K_00000) p[0] = data; else if (msr == MSR_MTRRfix16K_80000 || msr == MSR_MTRRfix16K_A0000) p[1 + msr - MSR_MTRRfix16K_80000] = data; else if (msr >= MSR_MTRRfix4K_C0000 && msr <= MSR_MTRRfix4K_F8000) p[3 + msr - MSR_MTRRfix4K_C0000] = data; else if (msr == MSR_IA32_CR_PAT) vcpu->arch.pat = data; else { /* Variable MTRRs */ int idx, is_mtrr_mask; u64 *pt; idx = (msr - 0x200) / 2; is_mtrr_mask = msr - 0x200 - 2 * idx; if (!is_mtrr_mask) pt = (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].base_lo; else pt = (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].mask_lo; *pt = data; } kvm_mmu_reset_context(vcpu); |
9ba075a66
|
1197 1198 |
return 0; } |
15c4a6406
|
1199 |
|
890ca9aef
|
1200 |
static int set_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 data) |
15c4a6406
|
1201 |
{ |
890ca9aef
|
1202 1203 |
u64 mcg_cap = vcpu->arch.mcg_cap; unsigned bank_num = mcg_cap & 0xff; |
15c4a6406
|
1204 |
switch (msr) { |
15c4a6406
|
1205 |
case MSR_IA32_MCG_STATUS: |
890ca9aef
|
1206 |
vcpu->arch.mcg_status = data; |
15c4a6406
|
1207 |
break; |
c7ac679c1
|
1208 |
case MSR_IA32_MCG_CTL: |
890ca9aef
|
1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 |
if (!(mcg_cap & MCG_CTL_P)) return 1; if (data != 0 && data != ~(u64)0) return -1; vcpu->arch.mcg_ctl = data; break; default: if (msr >= MSR_IA32_MC0_CTL && msr < MSR_IA32_MC0_CTL + 4 * bank_num) { u32 offset = msr - MSR_IA32_MC0_CTL; |
114be429c
|
1219 1220 1221 1222 1223 |
/* only 0 or all 1s can be written to IA32_MCi_CTL * some Linux kernels though clear bit 10 in bank 4 to * workaround a BIOS/GART TBL issue on AMD K8s, ignore * this to avoid an uncatched #GP in the guest */ |
890ca9aef
|
1224 |
if ((offset & 0x3) == 0 && |
114be429c
|
1225 |
data != 0 && (data | (1 << 10)) != ~(u64)0) |
890ca9aef
|
1226 1227 1228 1229 1230 1231 1232 1233 |
return -1; vcpu->arch.mce_banks[offset] = data; break; } return 1; } return 0; } |
ffde22ac5
|
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 |
static int xen_hvm_config(struct kvm_vcpu *vcpu, u64 data) { struct kvm *kvm = vcpu->kvm; int lm = is_long_mode(vcpu); u8 *blob_addr = lm ? (u8 *)(long)kvm->arch.xen_hvm_config.blob_addr_64 : (u8 *)(long)kvm->arch.xen_hvm_config.blob_addr_32; u8 blob_size = lm ? kvm->arch.xen_hvm_config.blob_size_64 : kvm->arch.xen_hvm_config.blob_size_32; u32 page_num = data & ~PAGE_MASK; u64 page_addr = data & PAGE_MASK; u8 *page; int r; r = -E2BIG; if (page_num >= blob_size) goto out; r = -ENOMEM; |
ff5c2c031
|
1251 1252 1253 |
page = memdup_user(blob_addr + (page_num * PAGE_SIZE), PAGE_SIZE); if (IS_ERR(page)) { r = PTR_ERR(page); |
ffde22ac5
|
1254 |
goto out; |
ff5c2c031
|
1255 |
} |
ffde22ac5
|
1256 1257 1258 1259 1260 1261 1262 1263 |
if (kvm_write_guest(kvm, page_addr, page, PAGE_SIZE)) goto out_free; r = 0; out_free: kfree(page); out: return r; } |
55cd8e5a4
|
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 |
static bool kvm_hv_hypercall_enabled(struct kvm *kvm) { return kvm->arch.hv_hypercall & HV_X64_MSR_HYPERCALL_ENABLE; } static bool kvm_hv_msr_partition_wide(u32 msr) { bool r = false; switch (msr) { case HV_X64_MSR_GUEST_OS_ID: case HV_X64_MSR_HYPERCALL: r = true; break; } return r; } static int set_msr_hyperv_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data) { struct kvm *kvm = vcpu->kvm; switch (msr) { case HV_X64_MSR_GUEST_OS_ID: kvm->arch.hv_guest_os_id = data; /* setting guest os id to zero disables hypercall page */ if (!kvm->arch.hv_guest_os_id) kvm->arch.hv_hypercall &= ~HV_X64_MSR_HYPERCALL_ENABLE; break; case HV_X64_MSR_HYPERCALL: { u64 gfn; unsigned long addr; u8 instructions[4]; /* if guest os id is not set hypercall should remain disabled */ if (!kvm->arch.hv_guest_os_id) break; if (!(data & HV_X64_MSR_HYPERCALL_ENABLE)) { kvm->arch.hv_hypercall = data; break; } gfn = data >> HV_X64_MSR_HYPERCALL_PAGE_ADDRESS_SHIFT; addr = gfn_to_hva(kvm, gfn); if (kvm_is_error_hva(addr)) return 1; kvm_x86_ops->patch_hypercall(vcpu, instructions); ((unsigned char *)instructions)[3] = 0xc3; /* ret */ |
8b0cedff0
|
1311 |
if (__copy_to_user((void __user *)addr, instructions, 4)) |
55cd8e5a4
|
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 |
return 1; kvm->arch.hv_hypercall = data; break; } default: pr_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x " "data 0x%llx ", msr, data); return 1; } return 0; } static int set_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 data) { |
10388a071
|
1327 1328 1329 |
switch (msr) { case HV_X64_MSR_APIC_ASSIST_PAGE: { unsigned long addr; |
55cd8e5a4
|
1330 |
|
10388a071
|
1331 1332 1333 1334 1335 1336 1337 1338 |
if (!(data & HV_X64_MSR_APIC_ASSIST_PAGE_ENABLE)) { vcpu->arch.hv_vapic = data; break; } addr = gfn_to_hva(vcpu->kvm, data >> HV_X64_MSR_APIC_ASSIST_PAGE_ADDRESS_SHIFT); if (kvm_is_error_hva(addr)) return 1; |
8b0cedff0
|
1339 |
if (__clear_user((void __user *)addr, PAGE_SIZE)) |
10388a071
|
1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 |
return 1; vcpu->arch.hv_vapic = data; break; } case HV_X64_MSR_EOI: return kvm_hv_vapic_msr_write(vcpu, APIC_EOI, data); case HV_X64_MSR_ICR: return kvm_hv_vapic_msr_write(vcpu, APIC_ICR, data); case HV_X64_MSR_TPR: return kvm_hv_vapic_msr_write(vcpu, APIC_TASKPRI, data); default: pr_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x " "data 0x%llx ", msr, data); return 1; } return 0; |
55cd8e5a4
|
1358 |
} |
344d9588a
|
1359 1360 1361 |
static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data) { gpa_t gpa = data & ~0x3f; |
6adba5274
|
1362 1363 |
/* Bits 2:5 are resrved, Should be zero */ if (data & 0x3c) |
344d9588a
|
1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 |
return 1; vcpu->arch.apf.msr_val = data; if (!(data & KVM_ASYNC_PF_ENABLED)) { kvm_clear_async_pf_completion_queue(vcpu); kvm_async_pf_hash_reset(vcpu); return 0; } if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.apf.data, gpa)) return 1; |
6adba5274
|
1376 |
vcpu->arch.apf.send_user_only = !(data & KVM_ASYNC_PF_SEND_ALWAYS); |
344d9588a
|
1377 1378 1379 |
kvm_async_pf_wakeup_all(vcpu); return 0; } |
12f9a48f7
|
1380 1381 1382 1383 1384 1385 1386 |
static void kvmclock_reset(struct kvm_vcpu *vcpu) { if (vcpu->arch.time_page) { kvm_release_page_dirty(vcpu->arch.time_page); vcpu->arch.time_page = NULL; } } |
c9aaa8957
|
1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 |
static void accumulate_steal_time(struct kvm_vcpu *vcpu) { u64 delta; if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) return; delta = current->sched_info.run_delay - vcpu->arch.st.last_steal; vcpu->arch.st.last_steal = current->sched_info.run_delay; vcpu->arch.st.accum_steal = delta; } static void record_steal_time(struct kvm_vcpu *vcpu) { if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) return; if (unlikely(kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)))) return; vcpu->arch.st.steal.steal += vcpu->arch.st.accum_steal; vcpu->arch.st.steal.version += 2; vcpu->arch.st.accum_steal = 0; kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)); } |
15c4a6406
|
1415 1416 1417 |
int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) { switch (msr) { |
15c4a6406
|
1418 |
case MSR_EFER: |
b69e8caef
|
1419 |
return set_efer(vcpu, data); |
8f1589d95
|
1420 1421 |
case MSR_K7_HWCR: data &= ~(u64)0x40; /* ignore flush filter disable */ |
82494028d
|
1422 |
data &= ~(u64)0x100; /* ignore ignne emulation enable */ |
8f1589d95
|
1423 1424 1425 1426 1427 1428 |
if (data != 0) { pr_unimpl(vcpu, "unimplemented HWCR wrmsr: 0x%llx ", data); return 1; } |
15c4a6406
|
1429 |
break; |
f7c6d1400
|
1430 1431 1432 1433 1434 1435 1436 |
case MSR_FAM10H_MMIO_CONF_BASE: if (data != 0) { pr_unimpl(vcpu, "unimplemented MMIO_CONF_BASE wrmsr: " "0x%llx ", data); return 1; } |
15c4a6406
|
1437 |
break; |
c323c0e5f
|
1438 |
case MSR_AMD64_NB_CFG: |
c7ac679c1
|
1439 |
break; |
b5e2fec0e
|
1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 |
case MSR_IA32_DEBUGCTLMSR: if (!data) { /* We support the non-activated case already */ break; } else if (data & ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_BTF)) { /* Values other than LBR and BTF are vendor-specific, thus reserved and should throw a #GP */ return 1; } pr_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTLMSR 0x%llx, nop ", __func__, data); break; |
15c4a6406
|
1453 1454 |
case MSR_IA32_UCODE_REV: case MSR_IA32_UCODE_WRITE: |
61a6bd672
|
1455 |
case MSR_VM_HSAVE_PA: |
6098ca939
|
1456 |
case MSR_AMD64_PATCH_LOADER: |
15c4a6406
|
1457 |
break; |
9ba075a66
|
1458 1459 |
case 0x200 ... 0x2ff: return set_msr_mtrr(vcpu, msr, data); |
15c4a6406
|
1460 1461 1462 |
case MSR_IA32_APICBASE: kvm_set_apic_base(vcpu, data); break; |
0105d1a52
|
1463 1464 |
case APIC_BASE_MSR ... APIC_BASE_MSR + 0x3ff: return kvm_x2apic_msr_write(vcpu, msr, data); |
a3e06bbe8
|
1465 1466 1467 |
case MSR_IA32_TSCDEADLINE: kvm_set_lapic_tscdeadline_msr(vcpu, data); break; |
15c4a6406
|
1468 |
case MSR_IA32_MISC_ENABLE: |
ad312c7c7
|
1469 |
vcpu->arch.ia32_misc_enable_msr = data; |
15c4a6406
|
1470 |
break; |
11c6bffa4
|
1471 |
case MSR_KVM_WALL_CLOCK_NEW: |
18068523d
|
1472 1473 1474 1475 |
case MSR_KVM_WALL_CLOCK: vcpu->kvm->arch.wall_clock = data; kvm_write_wall_clock(vcpu->kvm, data); break; |
11c6bffa4
|
1476 |
case MSR_KVM_SYSTEM_TIME_NEW: |
18068523d
|
1477 |
case MSR_KVM_SYSTEM_TIME: { |
12f9a48f7
|
1478 |
kvmclock_reset(vcpu); |
18068523d
|
1479 1480 |
vcpu->arch.time = data; |
c285545f8
|
1481 |
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); |
18068523d
|
1482 1483 1484 1485 1486 1487 1488 |
/* we verify if the enable bit is set... */ if (!(data & 1)) break; /* ...but clean it before doing the actual write */ vcpu->arch.time_offset = data & ~(PAGE_MASK | 1); |
18068523d
|
1489 1490 |
vcpu->arch.time_page = gfn_to_page(vcpu->kvm, data >> PAGE_SHIFT); |
18068523d
|
1491 1492 1493 1494 1495 |
if (is_error_page(vcpu->arch.time_page)) { kvm_release_page_clean(vcpu->arch.time_page); vcpu->arch.time_page = NULL; } |
18068523d
|
1496 1497 |
break; } |
344d9588a
|
1498 1499 1500 1501 |
case MSR_KVM_ASYNC_PF_EN: if (kvm_pv_enable_async_pf(vcpu, data)) return 1; break; |
c9aaa8957
|
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 |
case MSR_KVM_STEAL_TIME: if (unlikely(!sched_info_on())) return 1; if (data & KVM_STEAL_RESERVED_MASK) return 1; if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.st.stime, data & KVM_STEAL_VALID_BITS)) return 1; vcpu->arch.st.msr_val = data; if (!(data & KVM_MSR_ENABLED)) break; vcpu->arch.st.last_steal = current->sched_info.run_delay; preempt_disable(); accumulate_steal_time(vcpu); preempt_enable(); kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); break; |
890ca9aef
|
1528 1529 1530 1531 |
case MSR_IA32_MCG_CTL: case MSR_IA32_MCG_STATUS: case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1: return set_msr_mce(vcpu, msr, data); |
71db60232
|
1532 1533 1534 1535 1536 1537 1538 1539 |
/* Performance counters are not protected by a CPUID bit, * so we should check all of them in the generic path for the sake of * cross vendor migration. * Writing a zero into the event select MSRs disables them, * which we perfectly emulate ;-). Any other value should be at least * reported, some guests depend on them. */ |
71db60232
|
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 |
case MSR_K7_EVNTSEL0: case MSR_K7_EVNTSEL1: case MSR_K7_EVNTSEL2: case MSR_K7_EVNTSEL3: if (data != 0) pr_unimpl(vcpu, "unimplemented perfctr wrmsr: " "0x%x data 0x%llx ", msr, data); break; /* at least RHEL 4 unconditionally writes to the perfctr registers, * so we ignore writes to make it happy. */ |
71db60232
|
1552 1553 1554 1555 1556 1557 1558 1559 |
case MSR_K7_PERFCTR0: case MSR_K7_PERFCTR1: case MSR_K7_PERFCTR2: case MSR_K7_PERFCTR3: pr_unimpl(vcpu, "unimplemented perfctr wrmsr: " "0x%x data 0x%llx ", msr, data); break; |
84e0cefa8
|
1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 |
case MSR_K7_CLK_CTL: /* * Ignore all writes to this no longer documented MSR. * Writes are only relevant for old K7 processors, * all pre-dating SVM, but a recommended workaround from * AMD for these chips. It is possible to speicify the * affected processor models on the command line, hence * the need to ignore the workaround. */ break; |
55cd8e5a4
|
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 |
case HV_X64_MSR_GUEST_OS_ID ... HV_X64_MSR_SINT15: if (kvm_hv_msr_partition_wide(msr)) { int r; mutex_lock(&vcpu->kvm->lock); r = set_msr_hyperv_pw(vcpu, msr, data); mutex_unlock(&vcpu->kvm->lock); return r; } else return set_msr_hyperv(vcpu, msr, data); break; |
91c9c3eda
|
1580 1581 1582 1583 1584 1585 1586 |
case MSR_IA32_BBL_CR_CTL3: /* Drop writes to this legacy MSR -- see rdmsr * counterpart for further detail. */ pr_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx ", msr, data); break; |
15c4a6406
|
1587 |
default: |
ffde22ac5
|
1588 1589 |
if (msr && (msr == vcpu->kvm->arch.xen_hvm_config.msr)) return xen_hvm_config(vcpu, data); |
f5132b013
|
1590 1591 |
if (kvm_pmu_msr(vcpu, msr)) return kvm_pmu_set_msr(vcpu, msr, data); |
ed85c0685
|
1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 |
if (!ignore_msrs) { pr_unimpl(vcpu, "unhandled wrmsr: 0x%x data %llx ", msr, data); return 1; } else { pr_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx ", msr, data); break; } |
15c4a6406
|
1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 |
} return 0; } EXPORT_SYMBOL_GPL(kvm_set_msr_common); /* * Reads an msr value (of 'msr_index') into 'pdata'. * Returns 0 on success, non-0 otherwise. * Assumes vcpu_load() was already called. */ int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) { return kvm_x86_ops->get_msr(vcpu, msr_index, pdata); } |
9ba075a66
|
1618 1619 |
static int get_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) { |
0bed3b568
|
1620 |
u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges; |
9ba075a66
|
1621 1622 |
if (!msr_mtrr_valid(msr)) return 1; |
0bed3b568
|
1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 |
if (msr == MSR_MTRRdefType) *pdata = vcpu->arch.mtrr_state.def_type + (vcpu->arch.mtrr_state.enabled << 10); else if (msr == MSR_MTRRfix64K_00000) *pdata = p[0]; else if (msr == MSR_MTRRfix16K_80000 || msr == MSR_MTRRfix16K_A0000) *pdata = p[1 + msr - MSR_MTRRfix16K_80000]; else if (msr >= MSR_MTRRfix4K_C0000 && msr <= MSR_MTRRfix4K_F8000) *pdata = p[3 + msr - MSR_MTRRfix4K_C0000]; else if (msr == MSR_IA32_CR_PAT) *pdata = vcpu->arch.pat; else { /* Variable MTRRs */ int idx, is_mtrr_mask; u64 *pt; idx = (msr - 0x200) / 2; is_mtrr_mask = msr - 0x200 - 2 * idx; if (!is_mtrr_mask) pt = (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].base_lo; else pt = (u64 *)&vcpu->arch.mtrr_state.var_ranges[idx].mask_lo; *pdata = *pt; } |
9ba075a66
|
1648 1649 |
return 0; } |
890ca9aef
|
1650 |
static int get_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) |
15c4a6406
|
1651 1652 |
{ u64 data; |
890ca9aef
|
1653 1654 |
u64 mcg_cap = vcpu->arch.mcg_cap; unsigned bank_num = mcg_cap & 0xff; |
15c4a6406
|
1655 1656 |
switch (msr) { |
15c4a6406
|
1657 1658 |
case MSR_IA32_P5_MC_ADDR: case MSR_IA32_P5_MC_TYPE: |
890ca9aef
|
1659 1660 |
data = 0; break; |
15c4a6406
|
1661 |
case MSR_IA32_MCG_CAP: |
890ca9aef
|
1662 1663 |
data = vcpu->arch.mcg_cap; break; |
c7ac679c1
|
1664 |
case MSR_IA32_MCG_CTL: |
890ca9aef
|
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 |
if (!(mcg_cap & MCG_CTL_P)) return 1; data = vcpu->arch.mcg_ctl; break; case MSR_IA32_MCG_STATUS: data = vcpu->arch.mcg_status; break; default: if (msr >= MSR_IA32_MC0_CTL && msr < MSR_IA32_MC0_CTL + 4 * bank_num) { u32 offset = msr - MSR_IA32_MC0_CTL; data = vcpu->arch.mce_banks[offset]; break; } return 1; } *pdata = data; return 0; } |
55cd8e5a4
|
1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 |
static int get_msr_hyperv_pw(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) { u64 data = 0; struct kvm *kvm = vcpu->kvm; switch (msr) { case HV_X64_MSR_GUEST_OS_ID: data = kvm->arch.hv_guest_os_id; break; case HV_X64_MSR_HYPERCALL: data = kvm->arch.hv_hypercall; break; default: pr_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x ", msr); return 1; } *pdata = data; return 0; } static int get_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) { u64 data = 0; switch (msr) { case HV_X64_MSR_VP_INDEX: { int r; struct kvm_vcpu *v; kvm_for_each_vcpu(r, v, vcpu->kvm) if (v == vcpu) data = r; break; } |
10388a071
|
1719 1720 1721 1722 1723 1724 |
case HV_X64_MSR_EOI: return kvm_hv_vapic_msr_read(vcpu, APIC_EOI, pdata); case HV_X64_MSR_ICR: return kvm_hv_vapic_msr_read(vcpu, APIC_ICR, pdata); case HV_X64_MSR_TPR: return kvm_hv_vapic_msr_read(vcpu, APIC_TASKPRI, pdata); |
14fa67ee9
|
1725 |
case HV_X64_MSR_APIC_ASSIST_PAGE: |
d1613ad5d
|
1726 1727 |
data = vcpu->arch.hv_vapic; break; |
55cd8e5a4
|
1728 1729 1730 1731 1732 1733 1734 1735 |
default: pr_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x ", msr); return 1; } *pdata = data; return 0; } |
890ca9aef
|
1736 1737 1738 1739 1740 |
int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) { u64 data; switch (msr) { |
890ca9aef
|
1741 |
case MSR_IA32_PLATFORM_ID: |
15c4a6406
|
1742 |
case MSR_IA32_EBL_CR_POWERON: |
b5e2fec0e
|
1743 1744 1745 1746 1747 |
case MSR_IA32_DEBUGCTLMSR: case MSR_IA32_LASTBRANCHFROMIP: case MSR_IA32_LASTBRANCHTOIP: case MSR_IA32_LASTINTFROMIP: case MSR_IA32_LASTINTTOIP: |
60af2ecdc
|
1748 1749 |
case MSR_K8_SYSCFG: case MSR_K7_HWCR: |
61a6bd672
|
1750 |
case MSR_VM_HSAVE_PA: |
9e6996240
|
1751 |
case MSR_K7_EVNTSEL0: |
1f3ee616d
|
1752 |
case MSR_K7_PERFCTR0: |
1fdbd48c2
|
1753 |
case MSR_K8_INT_PENDING_MSG: |
c323c0e5f
|
1754 |
case MSR_AMD64_NB_CFG: |
f7c6d1400
|
1755 |
case MSR_FAM10H_MMIO_CONF_BASE: |
15c4a6406
|
1756 1757 |
data = 0; break; |
742bc6704
|
1758 1759 1760 |
case MSR_IA32_UCODE_REV: data = 0x100000000ULL; break; |
9ba075a66
|
1761 1762 1763 1764 1765 |
case MSR_MTRRcap: data = 0x500 | KVM_NR_VAR_MTRR; break; case 0x200 ... 0x2ff: return get_msr_mtrr(vcpu, msr, pdata); |
15c4a6406
|
1766 1767 1768 |
case 0xcd: /* fsb frequency */ data = 3; break; |
7b9140982
|
1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 |
/* * MSR_EBC_FREQUENCY_ID * Conservative value valid for even the basic CPU models. * Models 0,1: 000 in bits 23:21 indicating a bus speed of * 100MHz, model 2 000 in bits 18:16 indicating 100MHz, * and 266MHz for model 3, or 4. Set Core Clock * Frequency to System Bus Frequency Ratio to 1 (bits * 31:24) even though these are only valid for CPU * models > 2, however guests may end up dividing or * multiplying by zero otherwise. */ case MSR_EBC_FREQUENCY_ID: data = 1 << 24; break; |
15c4a6406
|
1783 1784 1785 |
case MSR_IA32_APICBASE: data = kvm_get_apic_base(vcpu); break; |
0105d1a52
|
1786 1787 1788 |
case APIC_BASE_MSR ... APIC_BASE_MSR + 0x3ff: return kvm_x2apic_msr_read(vcpu, msr, pdata); break; |
a3e06bbe8
|
1789 1790 1791 |
case MSR_IA32_TSCDEADLINE: data = kvm_get_lapic_tscdeadline_msr(vcpu); break; |
15c4a6406
|
1792 |
case MSR_IA32_MISC_ENABLE: |
ad312c7c7
|
1793 |
data = vcpu->arch.ia32_misc_enable_msr; |
15c4a6406
|
1794 |
break; |
847f0ad8c
|
1795 1796 1797 1798 1799 1800 |
case MSR_IA32_PERF_STATUS: /* TSC increment by tick */ data = 1000ULL; /* CPU multiplier */ data |= (((uint64_t)4ULL) << 40); break; |
15c4a6406
|
1801 |
case MSR_EFER: |
f6801dff2
|
1802 |
data = vcpu->arch.efer; |
15c4a6406
|
1803 |
break; |
18068523d
|
1804 |
case MSR_KVM_WALL_CLOCK: |
11c6bffa4
|
1805 |
case MSR_KVM_WALL_CLOCK_NEW: |
18068523d
|
1806 1807 1808 |
data = vcpu->kvm->arch.wall_clock; break; case MSR_KVM_SYSTEM_TIME: |
11c6bffa4
|
1809 |
case MSR_KVM_SYSTEM_TIME_NEW: |
18068523d
|
1810 1811 |
data = vcpu->arch.time; break; |
344d9588a
|
1812 1813 1814 |
case MSR_KVM_ASYNC_PF_EN: data = vcpu->arch.apf.msr_val; break; |
c9aaa8957
|
1815 1816 1817 |
case MSR_KVM_STEAL_TIME: data = vcpu->arch.st.msr_val; break; |
890ca9aef
|
1818 1819 1820 1821 1822 1823 1824 |
case MSR_IA32_P5_MC_ADDR: case MSR_IA32_P5_MC_TYPE: case MSR_IA32_MCG_CAP: case MSR_IA32_MCG_CTL: case MSR_IA32_MCG_STATUS: case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1: return get_msr_mce(vcpu, msr, pdata); |
84e0cefa8
|
1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 |
case MSR_K7_CLK_CTL: /* * Provide expected ramp-up count for K7. All other * are set to zero, indicating minimum divisors for * every field. * * This prevents guest kernels on AMD host with CPU * type 6, model 8 and higher from exploding due to * the rdmsr failing. */ data = 0x20000000; break; |
55cd8e5a4
|
1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 |
case HV_X64_MSR_GUEST_OS_ID ... HV_X64_MSR_SINT15: if (kvm_hv_msr_partition_wide(msr)) { int r; mutex_lock(&vcpu->kvm->lock); r = get_msr_hyperv_pw(vcpu, msr, pdata); mutex_unlock(&vcpu->kvm->lock); return r; } else return get_msr_hyperv(vcpu, msr, pdata); break; |
91c9c3eda
|
1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 |
case MSR_IA32_BBL_CR_CTL3: /* This legacy MSR exists but isn't fully documented in current * silicon. It is however accessed by winxp in very narrow * scenarios where it sets bit #19, itself documented as * a "reserved" bit. Best effort attempt to source coherent * read data here should the balance of the register be * interpreted by the guest: * * L2 cache control register 3: 64GB range, 256KB size, * enabled, latency 0x1, configured */ data = 0xbe702111; break; |
15c4a6406
|
1860 |
default: |
f5132b013
|
1861 1862 |
if (kvm_pmu_msr(vcpu, msr)) return kvm_pmu_get_msr(vcpu, msr, pdata); |
ed85c0685
|
1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 |
if (!ignore_msrs) { pr_unimpl(vcpu, "unhandled rdmsr: 0x%x ", msr); return 1; } else { pr_unimpl(vcpu, "ignored rdmsr: 0x%x ", msr); data = 0; } break; |
15c4a6406
|
1873 1874 1875 1876 1877 |
} *pdata = data; return 0; } EXPORT_SYMBOL_GPL(kvm_get_msr_common); |
313a3dc75
|
1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 |
/* * Read or write a bunch of msrs. All parameters are kernel addresses. * * @return number of msrs set successfully. */ static int __msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs *msrs, struct kvm_msr_entry *entries, int (*do_msr)(struct kvm_vcpu *vcpu, unsigned index, u64 *data)) { |
f656ce018
|
1888 |
int i, idx; |
313a3dc75
|
1889 |
|
f656ce018
|
1890 |
idx = srcu_read_lock(&vcpu->kvm->srcu); |
313a3dc75
|
1891 1892 1893 |
for (i = 0; i < msrs->nmsrs; ++i) if (do_msr(vcpu, entries[i].index, &entries[i].data)) break; |
f656ce018
|
1894 |
srcu_read_unlock(&vcpu->kvm->srcu, idx); |
313a3dc75
|
1895 |
|
313a3dc75
|
1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 |
return i; } /* * Read or write a bunch of msrs. Parameters are user addresses. * * @return number of msrs set successfully. */ static int msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs __user *user_msrs, int (*do_msr)(struct kvm_vcpu *vcpu, unsigned index, u64 *data), int writeback) { struct kvm_msrs msrs; struct kvm_msr_entry *entries; int r, n; unsigned size; r = -EFAULT; if (copy_from_user(&msrs, user_msrs, sizeof msrs)) goto out; r = -E2BIG; if (msrs.nmsrs >= MAX_IO_MSRS) goto out; |
313a3dc75
|
1921 |
size = sizeof(struct kvm_msr_entry) * msrs.nmsrs; |
ff5c2c031
|
1922 1923 1924 |
entries = memdup_user(user_msrs->entries, size); if (IS_ERR(entries)) { r = PTR_ERR(entries); |
313a3dc75
|
1925 |
goto out; |
ff5c2c031
|
1926 |
} |
313a3dc75
|
1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 |
r = n = __msr_io(vcpu, &msrs, entries, do_msr); if (r < 0) goto out_free; r = -EFAULT; if (writeback && copy_to_user(user_msrs->entries, entries, size)) goto out_free; r = n; out_free: |
7a73c0283
|
1939 |
kfree(entries); |
313a3dc75
|
1940 1941 1942 |
out: return r; } |
018d00d2f
|
1943 1944 1945 1946 1947 1948 1949 1950 |
int kvm_dev_ioctl_check_extension(long ext) { int r; switch (ext) { case KVM_CAP_IRQCHIP: case KVM_CAP_HLT: case KVM_CAP_MMU_SHADOW_CACHE_CONTROL: |
018d00d2f
|
1951 |
case KVM_CAP_SET_TSS_ADDR: |
077167174
|
1952 |
case KVM_CAP_EXT_CPUID: |
c80766040
|
1953 |
case KVM_CAP_CLOCKSOURCE: |
7837699fa
|
1954 |
case KVM_CAP_PIT: |
a28e4f5a6
|
1955 |
case KVM_CAP_NOP_IO_DELAY: |
62d9f0dbc
|
1956 |
case KVM_CAP_MP_STATE: |
ed8486243
|
1957 |
case KVM_CAP_SYNC_MMU: |
a355c85c5
|
1958 |
case KVM_CAP_USER_NMI: |
52d939a0b
|
1959 |
case KVM_CAP_REINJECT_CONTROL: |
4925663a0
|
1960 |
case KVM_CAP_IRQ_INJECT_STATUS: |
e56d532f2
|
1961 |
case KVM_CAP_ASSIGN_DEV_IRQ: |
721eecbf4
|
1962 |
case KVM_CAP_IRQFD: |
d34e6b175
|
1963 |
case KVM_CAP_IOEVENTFD: |
c5ff41ce6
|
1964 |
case KVM_CAP_PIT2: |
e9f427573
|
1965 |
case KVM_CAP_PIT_STATE2: |
b927a3cec
|
1966 |
case KVM_CAP_SET_IDENTITY_MAP_ADDR: |
ffde22ac5
|
1967 |
case KVM_CAP_XEN_HVM: |
afbcf7ab8
|
1968 |
case KVM_CAP_ADJUST_CLOCK: |
3cfc3092f
|
1969 |
case KVM_CAP_VCPU_EVENTS: |
55cd8e5a4
|
1970 |
case KVM_CAP_HYPERV: |
10388a071
|
1971 |
case KVM_CAP_HYPERV_VAPIC: |
c25bc1638
|
1972 |
case KVM_CAP_HYPERV_SPIN: |
ab9f4ecbb
|
1973 |
case KVM_CAP_PCI_SEGMENT: |
a1efbe77c
|
1974 |
case KVM_CAP_DEBUGREGS: |
d2be1651b
|
1975 |
case KVM_CAP_X86_ROBUST_SINGLESTEP: |
2d5b5a665
|
1976 |
case KVM_CAP_XSAVE: |
344d9588a
|
1977 |
case KVM_CAP_ASYNC_PF: |
92a1f12d2
|
1978 |
case KVM_CAP_GET_TSC_KHZ: |
018d00d2f
|
1979 1980 |
r = 1; break; |
542472b53
|
1981 1982 1983 |
case KVM_CAP_COALESCED_MMIO: r = KVM_COALESCED_MMIO_PAGE_OFFSET; break; |
774ead3ad
|
1984 1985 1986 |
case KVM_CAP_VAPIC: r = !kvm_x86_ops->cpu_has_accelerated_tpr(); break; |
f725230af
|
1987 |
case KVM_CAP_NR_VCPUS: |
8c3ba334f
|
1988 1989 1990 |
r = KVM_SOFT_MAX_VCPUS; break; case KVM_CAP_MAX_VCPUS: |
f725230af
|
1991 1992 |
r = KVM_MAX_VCPUS; break; |
a988b910e
|
1993 1994 1995 |
case KVM_CAP_NR_MEMSLOTS: r = KVM_MEMORY_SLOTS; break; |
a68a6a728
|
1996 1997 |
case KVM_CAP_PV_MMU: /* obsolete */ r = 0; |
2f333bcb4
|
1998 |
break; |
62c476c7c
|
1999 |
case KVM_CAP_IOMMU: |
a1b60c1cd
|
2000 |
r = iommu_present(&pci_bus_type); |
62c476c7c
|
2001 |
break; |
890ca9aef
|
2002 2003 2004 |
case KVM_CAP_MCE: r = KVM_MAX_MCE_BANKS; break; |
2d5b5a665
|
2005 2006 2007 |
case KVM_CAP_XCRS: r = cpu_has_xsave; break; |
92a1f12d2
|
2008 2009 2010 |
case KVM_CAP_TSC_CONTROL: r = kvm_has_tsc_control; break; |
4d25a066b
|
2011 2012 2013 |
case KVM_CAP_TSC_DEADLINE_TIMER: r = boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER); break; |
018d00d2f
|
2014 2015 2016 2017 2018 2019 2020 |
default: r = 0; break; } return r; } |
043405e10
|
2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 |
long kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { void __user *argp = (void __user *)arg; long r; switch (ioctl) { case KVM_GET_MSR_INDEX_LIST: { struct kvm_msr_list __user *user_msr_list = argp; struct kvm_msr_list msr_list; unsigned n; r = -EFAULT; if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list)) goto out; n = msr_list.nmsrs; msr_list.nmsrs = num_msrs_to_save + ARRAY_SIZE(emulated_msrs); if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list)) goto out; r = -E2BIG; |
e125e7b69
|
2041 |
if (n < msr_list.nmsrs) |
043405e10
|
2042 2043 2044 2045 2046 |
goto out; r = -EFAULT; if (copy_to_user(user_msr_list->indices, &msrs_to_save, num_msrs_to_save * sizeof(u32))) goto out; |
e125e7b69
|
2047 |
if (copy_to_user(user_msr_list->indices + num_msrs_to_save, |
043405e10
|
2048 2049 2050 2051 2052 2053 |
&emulated_msrs, ARRAY_SIZE(emulated_msrs) * sizeof(u32))) goto out; r = 0; break; } |
674eea0fc
|
2054 2055 2056 2057 2058 2059 2060 2061 |
case KVM_GET_SUPPORTED_CPUID: { struct kvm_cpuid2 __user *cpuid_arg = argp; struct kvm_cpuid2 cpuid; r = -EFAULT; if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_dev_ioctl_get_supported_cpuid(&cpuid, |
193554750
|
2062 |
cpuid_arg->entries); |
674eea0fc
|
2063 2064 2065 2066 2067 2068 2069 2070 2071 |
if (r) goto out; r = -EFAULT; if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid)) goto out; r = 0; break; } |
890ca9aef
|
2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 |
case KVM_X86_GET_MCE_CAP_SUPPORTED: { u64 mce_cap; mce_cap = KVM_MCE_CAP_SUPPORTED; r = -EFAULT; if (copy_to_user(argp, &mce_cap, sizeof mce_cap)) goto out; r = 0; break; } |
043405e10
|
2082 2083 2084 2085 2086 2087 |
default: r = -EINVAL; } out: return r; } |
f5f48ee15
|
2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 |
static void wbinvd_ipi(void *garbage) { wbinvd(); } static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu) { return vcpu->kvm->arch.iommu_domain && !(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY); } |
313a3dc75
|
2098 2099 |
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { |
f5f48ee15
|
2100 2101 2102 2103 2104 2105 2106 2107 |
/* Address WBINVD may be executed by guest */ if (need_emulate_wbinvd(vcpu)) { if (kvm_x86_ops->has_wbinvd_exit()) cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask); else if (vcpu->cpu != -1 && vcpu->cpu != cpu) smp_call_function_single(vcpu->cpu, wbinvd_ipi, NULL, 1); } |
313a3dc75
|
2108 |
kvm_x86_ops->vcpu_load(vcpu, cpu); |
48434c20e
|
2109 |
if (unlikely(vcpu->cpu != cpu) || check_tsc_unstable()) { |
e48672fa2
|
2110 |
/* Make sure TSC doesn't go backwards */ |
8f6055cba
|
2111 2112 |
s64 tsc_delta; u64 tsc; |
d5c1785d2
|
2113 |
tsc = kvm_x86_ops->read_l1_tsc(vcpu); |
8f6055cba
|
2114 2115 |
tsc_delta = !vcpu->arch.last_guest_tsc ? 0 : tsc - vcpu->arch.last_guest_tsc; |
e48672fa2
|
2116 2117 |
if (tsc_delta < 0) mark_tsc_unstable("KVM discovered backwards TSC"); |
c285545f8
|
2118 |
if (check_tsc_unstable()) { |
e48672fa2
|
2119 |
kvm_x86_ops->adjust_tsc_offset(vcpu, -tsc_delta); |
c285545f8
|
2120 |
vcpu->arch.tsc_catchup = 1; |
c285545f8
|
2121 |
} |
1aa8ceef0
|
2122 |
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); |
c285545f8
|
2123 2124 |
if (vcpu->cpu != cpu) kvm_migrate_timers(vcpu); |
e48672fa2
|
2125 |
vcpu->cpu = cpu; |
6b7d7e762
|
2126 |
} |
c9aaa8957
|
2127 2128 2129 |
accumulate_steal_time(vcpu); kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); |
313a3dc75
|
2130 2131 2132 2133 |
} void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) { |
02daab21d
|
2134 |
kvm_x86_ops->vcpu_put(vcpu); |
1c11e7135
|
2135 |
kvm_put_guest_fpu(vcpu); |
d5c1785d2
|
2136 |
vcpu->arch.last_guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu); |
313a3dc75
|
2137 |
} |
313a3dc75
|
2138 2139 2140 |
static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) { |
ad312c7c7
|
2141 |
memcpy(s->regs, vcpu->arch.apic->regs, sizeof *s); |
313a3dc75
|
2142 2143 2144 2145 2146 2147 2148 |
return 0; } static int kvm_vcpu_ioctl_set_lapic(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) { |
ad312c7c7
|
2149 |
memcpy(vcpu->arch.apic->regs, s->regs, sizeof *s); |
313a3dc75
|
2150 |
kvm_apic_post_state_restore(vcpu); |
cb142eb74
|
2151 |
update_cr8_intercept(vcpu); |
313a3dc75
|
2152 2153 2154 |
return 0; } |
f77bc6a42
|
2155 2156 2157 2158 2159 2160 2161 |
static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq) { if (irq->irq < 0 || irq->irq >= 256) return -EINVAL; if (irqchip_in_kernel(vcpu->kvm)) return -ENXIO; |
f77bc6a42
|
2162 |
|
66fd3f7f9
|
2163 |
kvm_queue_interrupt(vcpu, irq->irq, false); |
3842d135f
|
2164 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
f77bc6a42
|
2165 |
|
f77bc6a42
|
2166 2167 |
return 0; } |
c4abb7c9c
|
2168 2169 |
static int kvm_vcpu_ioctl_nmi(struct kvm_vcpu *vcpu) { |
c4abb7c9c
|
2170 |
kvm_inject_nmi(vcpu); |
c4abb7c9c
|
2171 2172 2173 |
return 0; } |
b209749f5
|
2174 2175 2176 2177 2178 2179 2180 2181 |
static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu, struct kvm_tpr_access_ctl *tac) { if (tac->flags) return -EINVAL; vcpu->arch.tpr_access_reporting = !!tac->enabled; return 0; } |
890ca9aef
|
2182 2183 2184 2185 2186 2187 2188 |
static int kvm_vcpu_ioctl_x86_setup_mce(struct kvm_vcpu *vcpu, u64 mcg_cap) { int r; unsigned bank_num = mcg_cap & 0xff, bank; r = -EINVAL; |
a9e38c3e0
|
2189 |
if (!bank_num || bank_num >= KVM_MAX_MCE_BANKS) |
890ca9aef
|
2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 |
goto out; if (mcg_cap & ~(KVM_MCE_CAP_SUPPORTED | 0xff | 0xff0000)) goto out; r = 0; vcpu->arch.mcg_cap = mcg_cap; /* Init IA32_MCG_CTL to all 1s */ if (mcg_cap & MCG_CTL_P) vcpu->arch.mcg_ctl = ~(u64)0; /* Init IA32_MCi_CTL to all 1s */ for (bank = 0; bank < bank_num; bank++) vcpu->arch.mce_banks[bank*4] = ~(u64)0; out: return r; } static int kvm_vcpu_ioctl_x86_set_mce(struct kvm_vcpu *vcpu, struct kvm_x86_mce *mce) { u64 mcg_cap = vcpu->arch.mcg_cap; unsigned bank_num = mcg_cap & 0xff; u64 *banks = vcpu->arch.mce_banks; if (mce->bank >= bank_num || !(mce->status & MCI_STATUS_VAL)) return -EINVAL; /* * if IA32_MCG_CTL is not all 1s, the uncorrected error * reporting is disabled */ if ((mce->status & MCI_STATUS_UC) && (mcg_cap & MCG_CTL_P) && vcpu->arch.mcg_ctl != ~(u64)0) return 0; banks += 4 * mce->bank; /* * if IA32_MCi_CTL is not all 1s, the uncorrected error * reporting is disabled for the bank */ if ((mce->status & MCI_STATUS_UC) && banks[0] != ~(u64)0) return 0; if (mce->status & MCI_STATUS_UC) { if ((vcpu->arch.mcg_status & MCG_STATUS_MCIP) || |
fc78f5193
|
2230 |
!kvm_read_cr4_bits(vcpu, X86_CR4_MCE)) { |
a8eeb04a4
|
2231 |
kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); |
890ca9aef
|
2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 |
return 0; } if (banks[1] & MCI_STATUS_VAL) mce->status |= MCI_STATUS_OVER; banks[2] = mce->addr; banks[3] = mce->misc; vcpu->arch.mcg_status = mce->mcg_status; banks[1] = mce->status; kvm_queue_exception(vcpu, MC_VECTOR); } else if (!(banks[1] & MCI_STATUS_VAL) || !(banks[1] & MCI_STATUS_UC)) { if (banks[1] & MCI_STATUS_VAL) mce->status |= MCI_STATUS_OVER; banks[2] = mce->addr; banks[3] = mce->misc; banks[1] = mce->status; } else banks[1] |= MCI_STATUS_OVER; return 0; } |
3cfc3092f
|
2252 2253 2254 |
static void kvm_vcpu_ioctl_x86_get_vcpu_events(struct kvm_vcpu *vcpu, struct kvm_vcpu_events *events) { |
7460fb4a3
|
2255 |
process_nmi(vcpu); |
03b82a30e
|
2256 2257 2258 |
events->exception.injected = vcpu->arch.exception.pending && !kvm_exception_is_soft(vcpu->arch.exception.nr); |
3cfc3092f
|
2259 2260 |
events->exception.nr = vcpu->arch.exception.nr; events->exception.has_error_code = vcpu->arch.exception.has_error_code; |
97e69aa62
|
2261 |
events->exception.pad = 0; |
3cfc3092f
|
2262 |
events->exception.error_code = vcpu->arch.exception.error_code; |
03b82a30e
|
2263 2264 |
events->interrupt.injected = vcpu->arch.interrupt.pending && !vcpu->arch.interrupt.soft; |
3cfc3092f
|
2265 |
events->interrupt.nr = vcpu->arch.interrupt.nr; |
03b82a30e
|
2266 |
events->interrupt.soft = 0; |
48005f64d
|
2267 2268 2269 |
events->interrupt.shadow = kvm_x86_ops->get_interrupt_shadow(vcpu, KVM_X86_SHADOW_INT_MOV_SS | KVM_X86_SHADOW_INT_STI); |
3cfc3092f
|
2270 2271 |
events->nmi.injected = vcpu->arch.nmi_injected; |
7460fb4a3
|
2272 |
events->nmi.pending = vcpu->arch.nmi_pending != 0; |
3cfc3092f
|
2273 |
events->nmi.masked = kvm_x86_ops->get_nmi_mask(vcpu); |
97e69aa62
|
2274 |
events->nmi.pad = 0; |
3cfc3092f
|
2275 2276 |
events->sipi_vector = vcpu->arch.sipi_vector; |
dab4b911a
|
2277 |
events->flags = (KVM_VCPUEVENT_VALID_NMI_PENDING |
48005f64d
|
2278 2279 |
| KVM_VCPUEVENT_VALID_SIPI_VECTOR | KVM_VCPUEVENT_VALID_SHADOW); |
97e69aa62
|
2280 |
memset(&events->reserved, 0, sizeof(events->reserved)); |
3cfc3092f
|
2281 2282 2283 2284 2285 |
} static int kvm_vcpu_ioctl_x86_set_vcpu_events(struct kvm_vcpu *vcpu, struct kvm_vcpu_events *events) { |
dab4b911a
|
2286 |
if (events->flags & ~(KVM_VCPUEVENT_VALID_NMI_PENDING |
48005f64d
|
2287 2288 |
| KVM_VCPUEVENT_VALID_SIPI_VECTOR | KVM_VCPUEVENT_VALID_SHADOW)) |
3cfc3092f
|
2289 |
return -EINVAL; |
7460fb4a3
|
2290 |
process_nmi(vcpu); |
3cfc3092f
|
2291 2292 2293 2294 2295 2296 2297 2298 |
vcpu->arch.exception.pending = events->exception.injected; vcpu->arch.exception.nr = events->exception.nr; vcpu->arch.exception.has_error_code = events->exception.has_error_code; vcpu->arch.exception.error_code = events->exception.error_code; vcpu->arch.interrupt.pending = events->interrupt.injected; vcpu->arch.interrupt.nr = events->interrupt.nr; vcpu->arch.interrupt.soft = events->interrupt.soft; |
48005f64d
|
2299 2300 2301 |
if (events->flags & KVM_VCPUEVENT_VALID_SHADOW) kvm_x86_ops->set_interrupt_shadow(vcpu, events->interrupt.shadow); |
3cfc3092f
|
2302 2303 |
vcpu->arch.nmi_injected = events->nmi.injected; |
dab4b911a
|
2304 2305 |
if (events->flags & KVM_VCPUEVENT_VALID_NMI_PENDING) vcpu->arch.nmi_pending = events->nmi.pending; |
3cfc3092f
|
2306 |
kvm_x86_ops->set_nmi_mask(vcpu, events->nmi.masked); |
dab4b911a
|
2307 2308 |
if (events->flags & KVM_VCPUEVENT_VALID_SIPI_VECTOR) vcpu->arch.sipi_vector = events->sipi_vector; |
3cfc3092f
|
2309 |
|
3842d135f
|
2310 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
3cfc3092f
|
2311 2312 |
return 0; } |
a1efbe77c
|
2313 2314 2315 |
static void kvm_vcpu_ioctl_x86_get_debugregs(struct kvm_vcpu *vcpu, struct kvm_debugregs *dbgregs) { |
a1efbe77c
|
2316 2317 2318 2319 |
memcpy(dbgregs->db, vcpu->arch.db, sizeof(vcpu->arch.db)); dbgregs->dr6 = vcpu->arch.dr6; dbgregs->dr7 = vcpu->arch.dr7; dbgregs->flags = 0; |
97e69aa62
|
2320 |
memset(&dbgregs->reserved, 0, sizeof(dbgregs->reserved)); |
a1efbe77c
|
2321 2322 2323 2324 2325 2326 2327 |
} static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu, struct kvm_debugregs *dbgregs) { if (dbgregs->flags) return -EINVAL; |
a1efbe77c
|
2328 2329 2330 |
memcpy(vcpu->arch.db, dbgregs->db, sizeof(vcpu->arch.db)); vcpu->arch.dr6 = dbgregs->dr6; vcpu->arch.dr7 = dbgregs->dr7; |
a1efbe77c
|
2331 2332 |
return 0; } |
2d5b5a665
|
2333 2334 2335 2336 2337 2338 |
static void kvm_vcpu_ioctl_x86_get_xsave(struct kvm_vcpu *vcpu, struct kvm_xsave *guest_xsave) { if (cpu_has_xsave) memcpy(guest_xsave->region, &vcpu->arch.guest_fpu.state->xsave, |
f45755b83
|
2339 |
xstate_size); |
2d5b5a665
|
2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 |
else { memcpy(guest_xsave->region, &vcpu->arch.guest_fpu.state->fxsave, sizeof(struct i387_fxsave_struct)); *(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)] = XSTATE_FPSSE; } } static int kvm_vcpu_ioctl_x86_set_xsave(struct kvm_vcpu *vcpu, struct kvm_xsave *guest_xsave) { u64 xstate_bv = *(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)]; if (cpu_has_xsave) memcpy(&vcpu->arch.guest_fpu.state->xsave, |
f45755b83
|
2357 |
guest_xsave->region, xstate_size); |
2d5b5a665
|
2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 |
else { if (xstate_bv & ~XSTATE_FPSSE) return -EINVAL; memcpy(&vcpu->arch.guest_fpu.state->fxsave, guest_xsave->region, sizeof(struct i387_fxsave_struct)); } return 0; } static void kvm_vcpu_ioctl_x86_get_xcrs(struct kvm_vcpu *vcpu, struct kvm_xcrs *guest_xcrs) { if (!cpu_has_xsave) { guest_xcrs->nr_xcrs = 0; return; } guest_xcrs->nr_xcrs = 1; guest_xcrs->flags = 0; guest_xcrs->xcrs[0].xcr = XCR_XFEATURE_ENABLED_MASK; guest_xcrs->xcrs[0].value = vcpu->arch.xcr0; } static int kvm_vcpu_ioctl_x86_set_xcrs(struct kvm_vcpu *vcpu, struct kvm_xcrs *guest_xcrs) { int i, r = 0; if (!cpu_has_xsave) return -EINVAL; if (guest_xcrs->nr_xcrs > KVM_MAX_XCRS || guest_xcrs->flags) return -EINVAL; for (i = 0; i < guest_xcrs->nr_xcrs; i++) /* Only support XCR0 currently */ if (guest_xcrs->xcrs[0].xcr == XCR_XFEATURE_ENABLED_MASK) { r = __kvm_set_xcr(vcpu, XCR_XFEATURE_ENABLED_MASK, guest_xcrs->xcrs[0].value); break; } if (r) r = -EINVAL; return r; } |
313a3dc75
|
2403 2404 2405 2406 2407 2408 |
long kvm_arch_vcpu_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { struct kvm_vcpu *vcpu = filp->private_data; void __user *argp = (void __user *)arg; int r; |
d1ac91d8a
|
2409 2410 2411 2412 2413 2414 2415 2416 |
union { struct kvm_lapic_state *lapic; struct kvm_xsave *xsave; struct kvm_xcrs *xcrs; void *buffer; } u; u.buffer = NULL; |
313a3dc75
|
2417 2418 |
switch (ioctl) { case KVM_GET_LAPIC: { |
2204ae3c9
|
2419 2420 2421 |
r = -EINVAL; if (!vcpu->arch.apic) goto out; |
d1ac91d8a
|
2422 |
u.lapic = kzalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL); |
313a3dc75
|
2423 |
|
b772ff362
|
2424 |
r = -ENOMEM; |
d1ac91d8a
|
2425 |
if (!u.lapic) |
b772ff362
|
2426 |
goto out; |
d1ac91d8a
|
2427 |
r = kvm_vcpu_ioctl_get_lapic(vcpu, u.lapic); |
313a3dc75
|
2428 2429 2430 |
if (r) goto out; r = -EFAULT; |
d1ac91d8a
|
2431 |
if (copy_to_user(argp, u.lapic, sizeof(struct kvm_lapic_state))) |
313a3dc75
|
2432 2433 2434 2435 2436 |
goto out; r = 0; break; } case KVM_SET_LAPIC: { |
2204ae3c9
|
2437 2438 2439 |
r = -EINVAL; if (!vcpu->arch.apic) goto out; |
ff5c2c031
|
2440 2441 2442 |
u.lapic = memdup_user(argp, sizeof(*u.lapic)); if (IS_ERR(u.lapic)) { r = PTR_ERR(u.lapic); |
313a3dc75
|
2443 |
goto out; |
ff5c2c031
|
2444 |
} |
d1ac91d8a
|
2445 |
r = kvm_vcpu_ioctl_set_lapic(vcpu, u.lapic); |
313a3dc75
|
2446 2447 2448 2449 2450 |
if (r) goto out; r = 0; break; } |
f77bc6a42
|
2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 |
case KVM_INTERRUPT: { struct kvm_interrupt irq; r = -EFAULT; if (copy_from_user(&irq, argp, sizeof irq)) goto out; r = kvm_vcpu_ioctl_interrupt(vcpu, &irq); if (r) goto out; r = 0; break; } |
c4abb7c9c
|
2463 2464 2465 2466 2467 2468 2469 |
case KVM_NMI: { r = kvm_vcpu_ioctl_nmi(vcpu); if (r) goto out; r = 0; break; } |
313a3dc75
|
2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 |
case KVM_SET_CPUID: { struct kvm_cpuid __user *cpuid_arg = argp; struct kvm_cpuid cpuid; r = -EFAULT; if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_vcpu_ioctl_set_cpuid(vcpu, &cpuid, cpuid_arg->entries); if (r) goto out; break; } |
077167174
|
2482 2483 2484 2485 2486 2487 2488 2489 |
case KVM_SET_CPUID2: { struct kvm_cpuid2 __user *cpuid_arg = argp; struct kvm_cpuid2 cpuid; r = -EFAULT; if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid, |
193554750
|
2490 |
cpuid_arg->entries); |
077167174
|
2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 |
if (r) goto out; break; } case KVM_GET_CPUID2: { struct kvm_cpuid2 __user *cpuid_arg = argp; struct kvm_cpuid2 cpuid; r = -EFAULT; if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid, |
193554750
|
2503 |
cpuid_arg->entries); |
077167174
|
2504 2505 2506 2507 2508 2509 2510 2511 |
if (r) goto out; r = -EFAULT; if (copy_to_user(cpuid_arg, &cpuid, sizeof cpuid)) goto out; r = 0; break; } |
313a3dc75
|
2512 2513 2514 2515 2516 2517 |
case KVM_GET_MSRS: r = msr_io(vcpu, argp, kvm_get_msr, 1); break; case KVM_SET_MSRS: r = msr_io(vcpu, argp, do_set_msr, 0); break; |
b209749f5
|
2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 |
case KVM_TPR_ACCESS_REPORTING: { struct kvm_tpr_access_ctl tac; r = -EFAULT; if (copy_from_user(&tac, argp, sizeof tac)) goto out; r = vcpu_ioctl_tpr_access_reporting(vcpu, &tac); if (r) goto out; r = -EFAULT; if (copy_to_user(argp, &tac, sizeof tac)) goto out; r = 0; break; }; |
b93463aa5
|
2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 |
case KVM_SET_VAPIC_ADDR: { struct kvm_vapic_addr va; r = -EINVAL; if (!irqchip_in_kernel(vcpu->kvm)) goto out; r = -EFAULT; if (copy_from_user(&va, argp, sizeof va)) goto out; r = 0; kvm_lapic_set_vapic_addr(vcpu, va.vapic_addr); break; } |
890ca9aef
|
2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 |
case KVM_X86_SETUP_MCE: { u64 mcg_cap; r = -EFAULT; if (copy_from_user(&mcg_cap, argp, sizeof mcg_cap)) goto out; r = kvm_vcpu_ioctl_x86_setup_mce(vcpu, mcg_cap); break; } case KVM_X86_SET_MCE: { struct kvm_x86_mce mce; r = -EFAULT; if (copy_from_user(&mce, argp, sizeof mce)) goto out; r = kvm_vcpu_ioctl_x86_set_mce(vcpu, &mce); break; } |
3cfc3092f
|
2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 |
case KVM_GET_VCPU_EVENTS: { struct kvm_vcpu_events events; kvm_vcpu_ioctl_x86_get_vcpu_events(vcpu, &events); r = -EFAULT; if (copy_to_user(argp, &events, sizeof(struct kvm_vcpu_events))) break; r = 0; break; } case KVM_SET_VCPU_EVENTS: { struct kvm_vcpu_events events; r = -EFAULT; if (copy_from_user(&events, argp, sizeof(struct kvm_vcpu_events))) break; r = kvm_vcpu_ioctl_x86_set_vcpu_events(vcpu, &events); break; } |
a1efbe77c
|
2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 |
case KVM_GET_DEBUGREGS: { struct kvm_debugregs dbgregs; kvm_vcpu_ioctl_x86_get_debugregs(vcpu, &dbgregs); r = -EFAULT; if (copy_to_user(argp, &dbgregs, sizeof(struct kvm_debugregs))) break; r = 0; break; } case KVM_SET_DEBUGREGS: { struct kvm_debugregs dbgregs; r = -EFAULT; if (copy_from_user(&dbgregs, argp, sizeof(struct kvm_debugregs))) break; r = kvm_vcpu_ioctl_x86_set_debugregs(vcpu, &dbgregs); break; } |
2d5b5a665
|
2608 |
case KVM_GET_XSAVE: { |
d1ac91d8a
|
2609 |
u.xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL); |
2d5b5a665
|
2610 |
r = -ENOMEM; |
d1ac91d8a
|
2611 |
if (!u.xsave) |
2d5b5a665
|
2612 |
break; |
d1ac91d8a
|
2613 |
kvm_vcpu_ioctl_x86_get_xsave(vcpu, u.xsave); |
2d5b5a665
|
2614 2615 |
r = -EFAULT; |
d1ac91d8a
|
2616 |
if (copy_to_user(argp, u.xsave, sizeof(struct kvm_xsave))) |
2d5b5a665
|
2617 2618 2619 2620 2621 |
break; r = 0; break; } case KVM_SET_XSAVE: { |
ff5c2c031
|
2622 2623 2624 2625 2626 |
u.xsave = memdup_user(argp, sizeof(*u.xsave)); if (IS_ERR(u.xsave)) { r = PTR_ERR(u.xsave); goto out; } |
2d5b5a665
|
2627 |
|
d1ac91d8a
|
2628 |
r = kvm_vcpu_ioctl_x86_set_xsave(vcpu, u.xsave); |
2d5b5a665
|
2629 2630 2631 |
break; } case KVM_GET_XCRS: { |
d1ac91d8a
|
2632 |
u.xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL); |
2d5b5a665
|
2633 |
r = -ENOMEM; |
d1ac91d8a
|
2634 |
if (!u.xcrs) |
2d5b5a665
|
2635 |
break; |
d1ac91d8a
|
2636 |
kvm_vcpu_ioctl_x86_get_xcrs(vcpu, u.xcrs); |
2d5b5a665
|
2637 2638 |
r = -EFAULT; |
d1ac91d8a
|
2639 |
if (copy_to_user(argp, u.xcrs, |
2d5b5a665
|
2640 2641 2642 2643 2644 2645 |
sizeof(struct kvm_xcrs))) break; r = 0; break; } case KVM_SET_XCRS: { |
ff5c2c031
|
2646 2647 2648 2649 2650 |
u.xcrs = memdup_user(argp, sizeof(*u.xcrs)); if (IS_ERR(u.xcrs)) { r = PTR_ERR(u.xcrs); goto out; } |
2d5b5a665
|
2651 |
|
d1ac91d8a
|
2652 |
r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, u.xcrs); |
2d5b5a665
|
2653 2654 |
break; } |
92a1f12d2
|
2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 |
case KVM_SET_TSC_KHZ: { u32 user_tsc_khz; r = -EINVAL; if (!kvm_has_tsc_control) break; user_tsc_khz = (u32)arg; if (user_tsc_khz >= kvm_max_guest_tsc_khz) goto out; kvm_x86_ops->set_tsc_khz(vcpu, user_tsc_khz); r = 0; goto out; } case KVM_GET_TSC_KHZ: { r = -EIO; if (check_tsc_unstable()) goto out; r = vcpu_tsc_khz(vcpu); goto out; } |
313a3dc75
|
2681 2682 2683 2684 |
default: r = -EINVAL; } out: |
d1ac91d8a
|
2685 |
kfree(u.buffer); |
313a3dc75
|
2686 2687 |
return r; } |
1fe779f8e
|
2688 2689 2690 2691 2692 2693 2694 2695 2696 |
static int kvm_vm_ioctl_set_tss_addr(struct kvm *kvm, unsigned long addr) { int ret; if (addr > (unsigned int)(-3 * PAGE_SIZE)) return -1; ret = kvm_x86_ops->set_tss_addr(kvm, addr); return ret; } |
b927a3cec
|
2697 2698 2699 2700 2701 2702 |
static int kvm_vm_ioctl_set_identity_map_addr(struct kvm *kvm, u64 ident_addr) { kvm->arch.ept_identity_map_addr = ident_addr; return 0; } |
1fe779f8e
|
2703 2704 2705 2706 2707 |
static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm, u32 kvm_nr_mmu_pages) { if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES) return -EINVAL; |
79fac95ec
|
2708 |
mutex_lock(&kvm->slots_lock); |
7c8a83b75
|
2709 |
spin_lock(&kvm->mmu_lock); |
1fe779f8e
|
2710 2711 |
kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages); |
f05e70ac0
|
2712 |
kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages; |
1fe779f8e
|
2713 |
|
7c8a83b75
|
2714 |
spin_unlock(&kvm->mmu_lock); |
79fac95ec
|
2715 |
mutex_unlock(&kvm->slots_lock); |
1fe779f8e
|
2716 2717 2718 2719 2720 |
return 0; } static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm) { |
39de71ec5
|
2721 |
return kvm->arch.n_max_mmu_pages; |
1fe779f8e
|
2722 |
} |
1fe779f8e
|
2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 |
static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm, struct kvm_irqchip *chip) { int r; r = 0; switch (chip->chip_id) { case KVM_IRQCHIP_PIC_MASTER: memcpy(&chip->chip.pic, &pic_irqchip(kvm)->pics[0], sizeof(struct kvm_pic_state)); break; case KVM_IRQCHIP_PIC_SLAVE: memcpy(&chip->chip.pic, &pic_irqchip(kvm)->pics[1], sizeof(struct kvm_pic_state)); break; case KVM_IRQCHIP_IOAPIC: |
eba0226bd
|
2740 |
r = kvm_get_ioapic(kvm, &chip->chip.ioapic); |
1fe779f8e
|
2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 |
break; default: r = -EINVAL; break; } return r; } static int kvm_vm_ioctl_set_irqchip(struct kvm *kvm, struct kvm_irqchip *chip) { int r; r = 0; switch (chip->chip_id) { case KVM_IRQCHIP_PIC_MASTER: |
f4f510508
|
2756 |
spin_lock(&pic_irqchip(kvm)->lock); |
1fe779f8e
|
2757 2758 2759 |
memcpy(&pic_irqchip(kvm)->pics[0], &chip->chip.pic, sizeof(struct kvm_pic_state)); |
f4f510508
|
2760 |
spin_unlock(&pic_irqchip(kvm)->lock); |
1fe779f8e
|
2761 2762 |
break; case KVM_IRQCHIP_PIC_SLAVE: |
f4f510508
|
2763 |
spin_lock(&pic_irqchip(kvm)->lock); |
1fe779f8e
|
2764 2765 2766 |
memcpy(&pic_irqchip(kvm)->pics[1], &chip->chip.pic, sizeof(struct kvm_pic_state)); |
f4f510508
|
2767 |
spin_unlock(&pic_irqchip(kvm)->lock); |
1fe779f8e
|
2768 2769 |
break; case KVM_IRQCHIP_IOAPIC: |
eba0226bd
|
2770 |
r = kvm_set_ioapic(kvm, &chip->chip.ioapic); |
1fe779f8e
|
2771 2772 2773 2774 2775 2776 2777 2778 |
break; default: r = -EINVAL; break; } kvm_pic_update_irq(pic_irqchip(kvm)); return r; } |
e0f63cb92
|
2779 2780 2781 |
static int kvm_vm_ioctl_get_pit(struct kvm *kvm, struct kvm_pit_state *ps) { int r = 0; |
894a9c554
|
2782 |
mutex_lock(&kvm->arch.vpit->pit_state.lock); |
e0f63cb92
|
2783 |
memcpy(ps, &kvm->arch.vpit->pit_state, sizeof(struct kvm_pit_state)); |
894a9c554
|
2784 |
mutex_unlock(&kvm->arch.vpit->pit_state.lock); |
e0f63cb92
|
2785 2786 2787 2788 2789 2790 |
return r; } static int kvm_vm_ioctl_set_pit(struct kvm *kvm, struct kvm_pit_state *ps) { int r = 0; |
894a9c554
|
2791 |
mutex_lock(&kvm->arch.vpit->pit_state.lock); |
e0f63cb92
|
2792 |
memcpy(&kvm->arch.vpit->pit_state, ps, sizeof(struct kvm_pit_state)); |
e9f427573
|
2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 |
kvm_pit_load_count(kvm, 0, ps->channels[0].count, 0); mutex_unlock(&kvm->arch.vpit->pit_state.lock); return r; } static int kvm_vm_ioctl_get_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) { int r = 0; mutex_lock(&kvm->arch.vpit->pit_state.lock); memcpy(ps->channels, &kvm->arch.vpit->pit_state.channels, sizeof(ps->channels)); ps->flags = kvm->arch.vpit->pit_state.flags; mutex_unlock(&kvm->arch.vpit->pit_state.lock); |
97e69aa62
|
2807 |
memset(&ps->reserved, 0, sizeof(ps->reserved)); |
e9f427573
|
2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 |
return r; } static int kvm_vm_ioctl_set_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) { int r = 0, start = 0; u32 prev_legacy, cur_legacy; mutex_lock(&kvm->arch.vpit->pit_state.lock); prev_legacy = kvm->arch.vpit->pit_state.flags & KVM_PIT_FLAGS_HPET_LEGACY; cur_legacy = ps->flags & KVM_PIT_FLAGS_HPET_LEGACY; if (!prev_legacy && cur_legacy) start = 1; memcpy(&kvm->arch.vpit->pit_state.channels, &ps->channels, sizeof(kvm->arch.vpit->pit_state.channels)); kvm->arch.vpit->pit_state.flags = ps->flags; kvm_pit_load_count(kvm, 0, kvm->arch.vpit->pit_state.channels[0].count, start); |
894a9c554
|
2824 |
mutex_unlock(&kvm->arch.vpit->pit_state.lock); |
e0f63cb92
|
2825 2826 |
return r; } |
52d939a0b
|
2827 2828 2829 2830 2831 |
static int kvm_vm_ioctl_reinject(struct kvm *kvm, struct kvm_reinject_control *control) { if (!kvm->arch.vpit) return -ENXIO; |
894a9c554
|
2832 |
mutex_lock(&kvm->arch.vpit->pit_state.lock); |
52d939a0b
|
2833 |
kvm->arch.vpit->pit_state.pit_timer.reinject = control->pit_reinject; |
894a9c554
|
2834 |
mutex_unlock(&kvm->arch.vpit->pit_state.lock); |
52d939a0b
|
2835 2836 |
return 0; } |
95d4c16ce
|
2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 |
/** * write_protect_slot - write protect a slot for dirty logging * @kvm: the kvm instance * @memslot: the slot we protect * @dirty_bitmap: the bitmap indicating which pages are dirty * @nr_dirty_pages: the number of dirty pages * * We have two ways to find all sptes to protect: * 1. Use kvm_mmu_slot_remove_write_access() which walks all shadow pages and * checks ones that have a spte mapping a page in the slot. * 2. Use kvm_mmu_rmap_write_protect() for each gfn found in the bitmap. * * Generally speaking, if there are not so many dirty pages compared to the * number of shadow pages, we should use the latter. * * Note that letting others write into a page marked dirty in the old bitmap * by using the remaining tlb entry is not a problem. That page will become * write protected again when we flush the tlb and then be reported dirty to * the user space by copying the old bitmap. */ static void write_protect_slot(struct kvm *kvm, struct kvm_memory_slot *memslot, unsigned long *dirty_bitmap, unsigned long nr_dirty_pages) { /* Not many dirty pages compared to # of shadow pages. */ if (nr_dirty_pages < kvm->arch.n_used_mmu_pages) { unsigned long gfn_offset; for_each_set_bit(gfn_offset, dirty_bitmap, memslot->npages) { unsigned long gfn = memslot->base_gfn + gfn_offset; spin_lock(&kvm->mmu_lock); kvm_mmu_rmap_write_protect(kvm, gfn, memslot); spin_unlock(&kvm->mmu_lock); } kvm_flush_remote_tlbs(kvm); } else { spin_lock(&kvm->mmu_lock); kvm_mmu_slot_remove_write_access(kvm, memslot->id); spin_unlock(&kvm->mmu_lock); } } |
5bb064dcd
|
2880 2881 2882 2883 2884 2885 |
/* * Get (and clear) the dirty memory log for a memory slot. */ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { |
7850ac542
|
2886 |
int r; |
5bb064dcd
|
2887 |
struct kvm_memory_slot *memslot; |
95d4c16ce
|
2888 |
unsigned long n, nr_dirty_pages; |
5bb064dcd
|
2889 |
|
79fac95ec
|
2890 |
mutex_lock(&kvm->slots_lock); |
5bb064dcd
|
2891 |
|
b050b015a
|
2892 2893 2894 |
r = -EINVAL; if (log->slot >= KVM_MEMORY_SLOTS) goto out; |
28a37544f
|
2895 |
memslot = id_to_memslot(kvm->memslots, log->slot); |
b050b015a
|
2896 2897 2898 |
r = -ENOENT; if (!memslot->dirty_bitmap) goto out; |
87bf6e7de
|
2899 |
n = kvm_dirty_bitmap_bytes(memslot); |
95d4c16ce
|
2900 |
nr_dirty_pages = memslot->nr_dirty_pages; |
b050b015a
|
2901 |
|
5bb064dcd
|
2902 |
/* If nothing is dirty, don't bother messing with page tables. */ |
95d4c16ce
|
2903 |
if (nr_dirty_pages) { |
b050b015a
|
2904 |
struct kvm_memslots *slots, *old_slots; |
28a37544f
|
2905 |
unsigned long *dirty_bitmap, *dirty_bitmap_head; |
b050b015a
|
2906 |
|
28a37544f
|
2907 2908 2909 2910 2911 |
dirty_bitmap = memslot->dirty_bitmap; dirty_bitmap_head = memslot->dirty_bitmap_head; if (dirty_bitmap == dirty_bitmap_head) dirty_bitmap_head += n / sizeof(long); memset(dirty_bitmap_head, 0, n); |
b050b015a
|
2912 |
|
914ebccd2
|
2913 |
r = -ENOMEM; |
cdfca7b34
|
2914 |
slots = kmemdup(kvm->memslots, sizeof(*kvm->memslots), GFP_KERNEL); |
515a01279
|
2915 |
if (!slots) |
914ebccd2
|
2916 |
goto out; |
cdfca7b34
|
2917 |
|
28a37544f
|
2918 |
memslot = id_to_memslot(slots, log->slot); |
95d4c16ce
|
2919 |
memslot->nr_dirty_pages = 0; |
28a37544f
|
2920 |
memslot->dirty_bitmap = dirty_bitmap_head; |
be593d628
|
2921 |
update_memslots(slots, NULL); |
b050b015a
|
2922 2923 2924 2925 |
old_slots = kvm->memslots; rcu_assign_pointer(kvm->memslots, slots); synchronize_srcu_expedited(&kvm->srcu); |
b050b015a
|
2926 |
kfree(old_slots); |
914ebccd2
|
2927 |
|
95d4c16ce
|
2928 |
write_protect_slot(kvm, memslot, dirty_bitmap, nr_dirty_pages); |
edde99ce0
|
2929 |
|
914ebccd2
|
2930 |
r = -EFAULT; |
515a01279
|
2931 |
if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) |
914ebccd2
|
2932 |
goto out; |
914ebccd2
|
2933 2934 2935 2936 |
} else { r = -EFAULT; if (clear_user(log->dirty_bitmap, n)) goto out; |
5bb064dcd
|
2937 |
} |
b050b015a
|
2938 |
|
5bb064dcd
|
2939 2940 |
r = 0; out: |
79fac95ec
|
2941 |
mutex_unlock(&kvm->slots_lock); |
5bb064dcd
|
2942 2943 |
return r; } |
1fe779f8e
|
2944 2945 2946 2947 2948 |
long kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { struct kvm *kvm = filp->private_data; void __user *argp = (void __user *)arg; |
367e1319b
|
2949 |
int r = -ENOTTY; |
f0d662759
|
2950 2951 2952 2953 2954 2955 2956 |
/* * This union makes it completely explicit to gcc-3.x * that these two variables' stack usage should be * combined, not added together. */ union { struct kvm_pit_state ps; |
e9f427573
|
2957 |
struct kvm_pit_state2 ps2; |
c5ff41ce6
|
2958 |
struct kvm_pit_config pit_config; |
f0d662759
|
2959 |
} u; |
1fe779f8e
|
2960 2961 2962 2963 2964 2965 2966 |
switch (ioctl) { case KVM_SET_TSS_ADDR: r = kvm_vm_ioctl_set_tss_addr(kvm, arg); if (r < 0) goto out; break; |
b927a3cec
|
2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 |
case KVM_SET_IDENTITY_MAP_ADDR: { u64 ident_addr; r = -EFAULT; if (copy_from_user(&ident_addr, argp, sizeof ident_addr)) goto out; r = kvm_vm_ioctl_set_identity_map_addr(kvm, ident_addr); if (r < 0) goto out; break; } |
1fe779f8e
|
2978 2979 2980 2981 2982 2983 2984 2985 |
case KVM_SET_NR_MMU_PAGES: r = kvm_vm_ioctl_set_nr_mmu_pages(kvm, arg); if (r) goto out; break; case KVM_GET_NR_MMU_PAGES: r = kvm_vm_ioctl_get_nr_mmu_pages(kvm); break; |
3ddea128a
|
2986 2987 2988 2989 2990 2991 2992 |
case KVM_CREATE_IRQCHIP: { struct kvm_pic *vpic; mutex_lock(&kvm->lock); r = -EEXIST; if (kvm->arch.vpic) goto create_irqchip_unlock; |
1fe779f8e
|
2993 |
r = -ENOMEM; |
3ddea128a
|
2994 2995 |
vpic = kvm_create_pic(kvm); if (vpic) { |
1fe779f8e
|
2996 2997 |
r = kvm_ioapic_init(kvm); if (r) { |
175504cdb
|
2998 |
mutex_lock(&kvm->slots_lock); |
72bb2fcd2
|
2999 |
kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, |
743eeb0b0
|
3000 3001 3002 3003 3004 |
&vpic->dev_master); kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, &vpic->dev_slave); kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, &vpic->dev_eclr); |
175504cdb
|
3005 |
mutex_unlock(&kvm->slots_lock); |
3ddea128a
|
3006 3007 |
kfree(vpic); goto create_irqchip_unlock; |
1fe779f8e
|
3008 3009 |
} } else |
3ddea128a
|
3010 3011 3012 3013 |
goto create_irqchip_unlock; smp_wmb(); kvm->arch.vpic = vpic; smp_wmb(); |
399ec807d
|
3014 3015 |
r = kvm_setup_default_irq_routing(kvm); if (r) { |
175504cdb
|
3016 |
mutex_lock(&kvm->slots_lock); |
3ddea128a
|
3017 |
mutex_lock(&kvm->irq_lock); |
72bb2fcd2
|
3018 3019 |
kvm_ioapic_destroy(kvm); kvm_destroy_pic(kvm); |
3ddea128a
|
3020 |
mutex_unlock(&kvm->irq_lock); |
175504cdb
|
3021 |
mutex_unlock(&kvm->slots_lock); |
399ec807d
|
3022 |
} |
3ddea128a
|
3023 3024 |
create_irqchip_unlock: mutex_unlock(&kvm->lock); |
1fe779f8e
|
3025 |
break; |
3ddea128a
|
3026 |
} |
7837699fa
|
3027 |
case KVM_CREATE_PIT: |
c5ff41ce6
|
3028 3029 3030 3031 3032 3033 3034 3035 |
u.pit_config.flags = KVM_PIT_SPEAKER_DUMMY; goto create_pit; case KVM_CREATE_PIT2: r = -EFAULT; if (copy_from_user(&u.pit_config, argp, sizeof(struct kvm_pit_config))) goto out; create_pit: |
79fac95ec
|
3036 |
mutex_lock(&kvm->slots_lock); |
269e05e48
|
3037 3038 3039 |
r = -EEXIST; if (kvm->arch.vpit) goto create_pit_unlock; |
7837699fa
|
3040 |
r = -ENOMEM; |
c5ff41ce6
|
3041 |
kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags); |
7837699fa
|
3042 3043 |
if (kvm->arch.vpit) r = 0; |
269e05e48
|
3044 |
create_pit_unlock: |
79fac95ec
|
3045 |
mutex_unlock(&kvm->slots_lock); |
7837699fa
|
3046 |
break; |
4925663a0
|
3047 |
case KVM_IRQ_LINE_STATUS: |
1fe779f8e
|
3048 3049 3050 3051 3052 3053 |
case KVM_IRQ_LINE: { struct kvm_irq_level irq_event; r = -EFAULT; if (copy_from_user(&irq_event, argp, sizeof irq_event)) goto out; |
160d2f6c0
|
3054 |
r = -ENXIO; |
1fe779f8e
|
3055 |
if (irqchip_in_kernel(kvm)) { |
4925663a0
|
3056 |
__s32 status; |
4925663a0
|
3057 3058 |
status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, irq_event.irq, irq_event.level); |
4925663a0
|
3059 |
if (ioctl == KVM_IRQ_LINE_STATUS) { |
160d2f6c0
|
3060 |
r = -EFAULT; |
4925663a0
|
3061 3062 3063 3064 3065 |
irq_event.status = status; if (copy_to_user(argp, &irq_event, sizeof irq_event)) goto out; } |
1fe779f8e
|
3066 3067 3068 3069 3070 3071 |
r = 0; } break; } case KVM_GET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ |
ff5c2c031
|
3072 |
struct kvm_irqchip *chip; |
1fe779f8e
|
3073 |
|
ff5c2c031
|
3074 3075 3076 |
chip = memdup_user(argp, sizeof(*chip)); if (IS_ERR(chip)) { r = PTR_ERR(chip); |
1fe779f8e
|
3077 |
goto out; |
ff5c2c031
|
3078 |
} |
1fe779f8e
|
3079 3080 |
r = -ENXIO; if (!irqchip_in_kernel(kvm)) |
f0d662759
|
3081 3082 |
goto get_irqchip_out; r = kvm_vm_ioctl_get_irqchip(kvm, chip); |
1fe779f8e
|
3083 |
if (r) |
f0d662759
|
3084 |
goto get_irqchip_out; |
1fe779f8e
|
3085 |
r = -EFAULT; |
f0d662759
|
3086 3087 |
if (copy_to_user(argp, chip, sizeof *chip)) goto get_irqchip_out; |
1fe779f8e
|
3088 |
r = 0; |
f0d662759
|
3089 3090 3091 3092 |
get_irqchip_out: kfree(chip); if (r) goto out; |
1fe779f8e
|
3093 3094 3095 3096 |
break; } case KVM_SET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ |
ff5c2c031
|
3097 |
struct kvm_irqchip *chip; |
1fe779f8e
|
3098 |
|
ff5c2c031
|
3099 3100 3101 |
chip = memdup_user(argp, sizeof(*chip)); if (IS_ERR(chip)) { r = PTR_ERR(chip); |
1fe779f8e
|
3102 |
goto out; |
ff5c2c031
|
3103 |
} |
1fe779f8e
|
3104 3105 |
r = -ENXIO; if (!irqchip_in_kernel(kvm)) |
f0d662759
|
3106 3107 |
goto set_irqchip_out; r = kvm_vm_ioctl_set_irqchip(kvm, chip); |
1fe779f8e
|
3108 |
if (r) |
f0d662759
|
3109 |
goto set_irqchip_out; |
1fe779f8e
|
3110 |
r = 0; |
f0d662759
|
3111 3112 3113 3114 |
set_irqchip_out: kfree(chip); if (r) goto out; |
1fe779f8e
|
3115 3116 |
break; } |
e0f63cb92
|
3117 |
case KVM_GET_PIT: { |
e0f63cb92
|
3118 |
r = -EFAULT; |
f0d662759
|
3119 |
if (copy_from_user(&u.ps, argp, sizeof(struct kvm_pit_state))) |
e0f63cb92
|
3120 3121 3122 3123 |
goto out; r = -ENXIO; if (!kvm->arch.vpit) goto out; |
f0d662759
|
3124 |
r = kvm_vm_ioctl_get_pit(kvm, &u.ps); |
e0f63cb92
|
3125 3126 3127 |
if (r) goto out; r = -EFAULT; |
f0d662759
|
3128 |
if (copy_to_user(argp, &u.ps, sizeof(struct kvm_pit_state))) |
e0f63cb92
|
3129 3130 3131 3132 3133 |
goto out; r = 0; break; } case KVM_SET_PIT: { |
e0f63cb92
|
3134 |
r = -EFAULT; |
f0d662759
|
3135 |
if (copy_from_user(&u.ps, argp, sizeof u.ps)) |
e0f63cb92
|
3136 3137 3138 3139 |
goto out; r = -ENXIO; if (!kvm->arch.vpit) goto out; |
f0d662759
|
3140 |
r = kvm_vm_ioctl_set_pit(kvm, &u.ps); |
e0f63cb92
|
3141 3142 3143 3144 3145 |
if (r) goto out; r = 0; break; } |
e9f427573
|
3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 |
case KVM_GET_PIT2: { r = -ENXIO; if (!kvm->arch.vpit) goto out; r = kvm_vm_ioctl_get_pit2(kvm, &u.ps2); if (r) goto out; r = -EFAULT; if (copy_to_user(argp, &u.ps2, sizeof(u.ps2))) goto out; r = 0; break; } case KVM_SET_PIT2: { r = -EFAULT; if (copy_from_user(&u.ps2, argp, sizeof(u.ps2))) goto out; r = -ENXIO; if (!kvm->arch.vpit) goto out; r = kvm_vm_ioctl_set_pit2(kvm, &u.ps2); if (r) goto out; r = 0; break; } |
52d939a0b
|
3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 |
case KVM_REINJECT_CONTROL: { struct kvm_reinject_control control; r = -EFAULT; if (copy_from_user(&control, argp, sizeof(control))) goto out; r = kvm_vm_ioctl_reinject(kvm, &control); if (r) goto out; r = 0; break; } |
ffde22ac5
|
3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 |
case KVM_XEN_HVM_CONFIG: { r = -EFAULT; if (copy_from_user(&kvm->arch.xen_hvm_config, argp, sizeof(struct kvm_xen_hvm_config))) goto out; r = -EINVAL; if (kvm->arch.xen_hvm_config.flags) goto out; r = 0; break; } |
afbcf7ab8
|
3194 |
case KVM_SET_CLOCK: { |
afbcf7ab8
|
3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 |
struct kvm_clock_data user_ns; u64 now_ns; s64 delta; r = -EFAULT; if (copy_from_user(&user_ns, argp, sizeof(user_ns))) goto out; r = -EINVAL; if (user_ns.flags) goto out; r = 0; |
395c6b0a9
|
3208 |
local_irq_disable(); |
759379dd6
|
3209 |
now_ns = get_kernel_ns(); |
afbcf7ab8
|
3210 |
delta = user_ns.clock - now_ns; |
395c6b0a9
|
3211 |
local_irq_enable(); |
afbcf7ab8
|
3212 3213 3214 3215 |
kvm->arch.kvmclock_offset = delta; break; } case KVM_GET_CLOCK: { |
afbcf7ab8
|
3216 3217 |
struct kvm_clock_data user_ns; u64 now_ns; |
395c6b0a9
|
3218 |
local_irq_disable(); |
759379dd6
|
3219 |
now_ns = get_kernel_ns(); |
afbcf7ab8
|
3220 |
user_ns.clock = kvm->arch.kvmclock_offset + now_ns; |
395c6b0a9
|
3221 |
local_irq_enable(); |
afbcf7ab8
|
3222 |
user_ns.flags = 0; |
97e69aa62
|
3223 |
memset(&user_ns.pad, 0, sizeof(user_ns.pad)); |
afbcf7ab8
|
3224 3225 3226 3227 3228 3229 3230 |
r = -EFAULT; if (copy_to_user(argp, &user_ns, sizeof(user_ns))) goto out; r = 0; break; } |
1fe779f8e
|
3231 3232 3233 3234 3235 3236 |
default: ; } out: return r; } |
a16b043cc
|
3237 |
static void kvm_init_msr_list(void) |
043405e10
|
3238 3239 3240 |
{ u32 dummy[2]; unsigned i, j; |
e3267cbbb
|
3241 3242 |
/* skip the first msrs in the list. KVM-specific */ for (i = j = KVM_SAVE_MSRS_BEGIN; i < ARRAY_SIZE(msrs_to_save); i++) { |
043405e10
|
3243 3244 3245 3246 3247 3248 3249 3250 |
if (rdmsr_safe(msrs_to_save[i], &dummy[0], &dummy[1]) < 0) continue; if (j < i) msrs_to_save[j] = msrs_to_save[i]; j++; } num_msrs_to_save = j; } |
bda9020e2
|
3251 3252 |
static int vcpu_mmio_write(struct kvm_vcpu *vcpu, gpa_t addr, int len, const void *v) |
bbd9b64e3
|
3253 |
{ |
70252a105
|
3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 |
int handled = 0; int n; do { n = min(len, 8); if (!(vcpu->arch.apic && !kvm_iodevice_write(&vcpu->arch.apic->dev, addr, n, v)) && kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, addr, n, v)) break; handled += n; addr += n; len -= n; v += n; } while (len); |
bbd9b64e3
|
3268 |
|
70252a105
|
3269 |
return handled; |
bbd9b64e3
|
3270 |
} |
bda9020e2
|
3271 |
static int vcpu_mmio_read(struct kvm_vcpu *vcpu, gpa_t addr, int len, void *v) |
bbd9b64e3
|
3272 |
{ |
70252a105
|
3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 |
int handled = 0; int n; do { n = min(len, 8); if (!(vcpu->arch.apic && !kvm_iodevice_read(&vcpu->arch.apic->dev, addr, n, v)) && kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, addr, n, v)) break; trace_kvm_mmio(KVM_TRACE_MMIO_READ, n, addr, *(u64 *)v); handled += n; addr += n; len -= n; v += n; } while (len); |
bbd9b64e3
|
3288 |
|
70252a105
|
3289 |
return handled; |
bbd9b64e3
|
3290 |
} |
2dafc6c23
|
3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 |
static void kvm_set_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg) { kvm_x86_ops->set_segment(vcpu, var, seg); } void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg) { kvm_x86_ops->get_segment(vcpu, var, seg); } |
e459e3228
|
3302 |
gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access) |
02f59dc9f
|
3303 3304 |
{ gpa_t t_gpa; |
ab9ae3138
|
3305 |
struct x86_exception exception; |
02f59dc9f
|
3306 3307 3308 3309 3310 |
BUG_ON(!mmu_is_nested(vcpu)); /* NPT walks are always user-walks */ access |= PFERR_USER_MASK; |
ab9ae3138
|
3311 |
t_gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gpa, access, &exception); |
02f59dc9f
|
3312 3313 3314 |
return t_gpa; } |
ab9ae3138
|
3315 3316 |
gpa_t kvm_mmu_gva_to_gpa_read(struct kvm_vcpu *vcpu, gva_t gva, struct x86_exception *exception) |
1871c6020
|
3317 3318 |
{ u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; |
ab9ae3138
|
3319 |
return vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, access, exception); |
1871c6020
|
3320 |
} |
ab9ae3138
|
3321 3322 |
gpa_t kvm_mmu_gva_to_gpa_fetch(struct kvm_vcpu *vcpu, gva_t gva, struct x86_exception *exception) |
1871c6020
|
3323 3324 3325 |
{ u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; access |= PFERR_FETCH_MASK; |
ab9ae3138
|
3326 |
return vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, access, exception); |
1871c6020
|
3327 |
} |
ab9ae3138
|
3328 3329 |
gpa_t kvm_mmu_gva_to_gpa_write(struct kvm_vcpu *vcpu, gva_t gva, struct x86_exception *exception) |
1871c6020
|
3330 3331 3332 |
{ u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; access |= PFERR_WRITE_MASK; |
ab9ae3138
|
3333 |
return vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, access, exception); |
1871c6020
|
3334 3335 3336 |
} /* uses this to access any guest's mapped memory without checking CPL */ |
ab9ae3138
|
3337 3338 |
gpa_t kvm_mmu_gva_to_gpa_system(struct kvm_vcpu *vcpu, gva_t gva, struct x86_exception *exception) |
1871c6020
|
3339 |
{ |
ab9ae3138
|
3340 |
return vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, 0, exception); |
1871c6020
|
3341 3342 3343 3344 |
} static int kvm_read_guest_virt_helper(gva_t addr, void *val, unsigned int bytes, struct kvm_vcpu *vcpu, u32 access, |
bcc55cba9
|
3345 |
struct x86_exception *exception) |
bbd9b64e3
|
3346 3347 |
{ void *data = val; |
10589a469
|
3348 |
int r = X86EMUL_CONTINUE; |
bbd9b64e3
|
3349 3350 |
while (bytes) { |
14dfe855f
|
3351 |
gpa_t gpa = vcpu->arch.walk_mmu->gva_to_gpa(vcpu, addr, access, |
ab9ae3138
|
3352 |
exception); |
bbd9b64e3
|
3353 |
unsigned offset = addr & (PAGE_SIZE-1); |
77c2002e7
|
3354 |
unsigned toread = min(bytes, (unsigned)PAGE_SIZE - offset); |
bbd9b64e3
|
3355 |
int ret; |
bcc55cba9
|
3356 |
if (gpa == UNMAPPED_GVA) |
ab9ae3138
|
3357 |
return X86EMUL_PROPAGATE_FAULT; |
77c2002e7
|
3358 |
ret = kvm_read_guest(vcpu->kvm, gpa, data, toread); |
10589a469
|
3359 |
if (ret < 0) { |
c3cd7ffaf
|
3360 |
r = X86EMUL_IO_NEEDED; |
10589a469
|
3361 3362 |
goto out; } |
bbd9b64e3
|
3363 |
|
77c2002e7
|
3364 3365 3366 |
bytes -= toread; data += toread; addr += toread; |
bbd9b64e3
|
3367 |
} |
10589a469
|
3368 |
out: |
10589a469
|
3369 |
return r; |
bbd9b64e3
|
3370 |
} |
77c2002e7
|
3371 |
|
1871c6020
|
3372 |
/* used for instruction fetching */ |
0f65dd70a
|
3373 3374 |
static int kvm_fetch_guest_virt(struct x86_emulate_ctxt *ctxt, gva_t addr, void *val, unsigned int bytes, |
bcc55cba9
|
3375 |
struct x86_exception *exception) |
1871c6020
|
3376 |
{ |
0f65dd70a
|
3377 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
1871c6020
|
3378 |
u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; |
0f65dd70a
|
3379 |
|
1871c6020
|
3380 |
return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, |
bcc55cba9
|
3381 3382 |
access | PFERR_FETCH_MASK, exception); |
1871c6020
|
3383 |
} |
064aea774
|
3384 |
int kvm_read_guest_virt(struct x86_emulate_ctxt *ctxt, |
0f65dd70a
|
3385 |
gva_t addr, void *val, unsigned int bytes, |
bcc55cba9
|
3386 |
struct x86_exception *exception) |
1871c6020
|
3387 |
{ |
0f65dd70a
|
3388 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
1871c6020
|
3389 |
u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; |
0f65dd70a
|
3390 |
|
1871c6020
|
3391 |
return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, access, |
bcc55cba9
|
3392 |
exception); |
1871c6020
|
3393 |
} |
064aea774
|
3394 |
EXPORT_SYMBOL_GPL(kvm_read_guest_virt); |
1871c6020
|
3395 |
|
0f65dd70a
|
3396 3397 |
static int kvm_read_guest_virt_system(struct x86_emulate_ctxt *ctxt, gva_t addr, void *val, unsigned int bytes, |
bcc55cba9
|
3398 |
struct x86_exception *exception) |
1871c6020
|
3399 |
{ |
0f65dd70a
|
3400 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
bcc55cba9
|
3401 |
return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, 0, exception); |
1871c6020
|
3402 |
} |
6a4d75506
|
3403 |
int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt, |
0f65dd70a
|
3404 |
gva_t addr, void *val, |
2dafc6c23
|
3405 |
unsigned int bytes, |
bcc55cba9
|
3406 |
struct x86_exception *exception) |
77c2002e7
|
3407 |
{ |
0f65dd70a
|
3408 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
77c2002e7
|
3409 3410 3411 3412 |
void *data = val; int r = X86EMUL_CONTINUE; while (bytes) { |
14dfe855f
|
3413 3414 |
gpa_t gpa = vcpu->arch.walk_mmu->gva_to_gpa(vcpu, addr, PFERR_WRITE_MASK, |
ab9ae3138
|
3415 |
exception); |
77c2002e7
|
3416 3417 3418 |
unsigned offset = addr & (PAGE_SIZE-1); unsigned towrite = min(bytes, (unsigned)PAGE_SIZE - offset); int ret; |
bcc55cba9
|
3419 |
if (gpa == UNMAPPED_GVA) |
ab9ae3138
|
3420 |
return X86EMUL_PROPAGATE_FAULT; |
77c2002e7
|
3421 3422 |
ret = kvm_write_guest(vcpu->kvm, gpa, data, towrite); if (ret < 0) { |
c3cd7ffaf
|
3423 |
r = X86EMUL_IO_NEEDED; |
77c2002e7
|
3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 |
goto out; } bytes -= towrite; data += towrite; addr += towrite; } out: return r; } |
6a4d75506
|
3434 |
EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system); |
77c2002e7
|
3435 |
|
af7cc7d1e
|
3436 3437 3438 3439 3440 |
static int vcpu_mmio_gva_to_gpa(struct kvm_vcpu *vcpu, unsigned long gva, gpa_t *gpa, struct x86_exception *exception, bool write) { u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; |
bebb106a5
|
3441 3442 3443 3444 3445 |
if (vcpu_match_mmio_gva(vcpu, gva) && check_write_user_access(vcpu, write, access, vcpu->arch.access)) { *gpa = vcpu->arch.mmio_gfn << PAGE_SHIFT | (gva & (PAGE_SIZE - 1)); |
4f0226482
|
3446 |
trace_vcpu_match_mmio(gva, *gpa, write, false); |
bebb106a5
|
3447 3448 |
return 1; } |
af7cc7d1e
|
3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 |
if (write) access |= PFERR_WRITE_MASK; *gpa = vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, access, exception); if (*gpa == UNMAPPED_GVA) return -1; /* For APIC access vmexit */ if ((*gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) return 1; |
4f0226482
|
3460 3461 |
if (vcpu_match_mmio_gpa(vcpu, *gpa)) { trace_vcpu_match_mmio(gva, *gpa, write, true); |
bebb106a5
|
3462 |
return 1; |
4f0226482
|
3463 |
} |
bebb106a5
|
3464 |
|
af7cc7d1e
|
3465 3466 |
return 0; } |
3200f405a
|
3467 |
int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, |
bcc55cba9
|
3468 |
const void *val, int bytes) |
bbd9b64e3
|
3469 3470 3471 3472 |
{ int ret; ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes); |
9f8112859
|
3473 |
if (ret < 0) |
bbd9b64e3
|
3474 |
return 0; |
f57f2ef58
|
3475 |
kvm_mmu_pte_write(vcpu, gpa, val, bytes); |
bbd9b64e3
|
3476 3477 |
return 1; } |
77d197b2c
|
3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 |
struct read_write_emulator_ops { int (*read_write_prepare)(struct kvm_vcpu *vcpu, void *val, int bytes); int (*read_write_emulate)(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes); int (*read_write_mmio)(struct kvm_vcpu *vcpu, gpa_t gpa, int bytes, void *val); int (*read_write_exit_mmio)(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes); bool write; }; static int read_prepare(struct kvm_vcpu *vcpu, void *val, int bytes) { if (vcpu->mmio_read_completed) { memcpy(val, vcpu->mmio_data, bytes); trace_kvm_mmio(KVM_TRACE_MMIO_READ, bytes, vcpu->mmio_phys_addr, *(u64 *)val); vcpu->mmio_read_completed = 0; return 1; } return 0; } static int read_emulate(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes) { return !kvm_read_guest(vcpu->kvm, gpa, val, bytes); } static int write_emulate(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes) { return emulator_write_phys(vcpu, gpa, val, bytes); } static int write_mmio(struct kvm_vcpu *vcpu, gpa_t gpa, int bytes, void *val) { trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, bytes, gpa, *(u64 *)val); return vcpu_mmio_write(vcpu, gpa, bytes, val); } static int read_exit_mmio(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes) { trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, bytes, gpa, 0); return X86EMUL_IO_NEEDED; } static int write_exit_mmio(struct kvm_vcpu *vcpu, gpa_t gpa, void *val, int bytes) { memcpy(vcpu->mmio_data, val, bytes); memcpy(vcpu->run->mmio.data, vcpu->mmio_data, 8); return X86EMUL_CONTINUE; } static struct read_write_emulator_ops read_emultor = { .read_write_prepare = read_prepare, .read_write_emulate = read_emulate, .read_write_mmio = vcpu_mmio_read, .read_write_exit_mmio = read_exit_mmio, }; static struct read_write_emulator_ops write_emultor = { .read_write_emulate = write_emulate, .read_write_mmio = write_mmio, .read_write_exit_mmio = write_exit_mmio, .write = true, }; |
22388a3c8
|
3549 3550 3551 3552 3553 |
static int emulator_read_write_onepage(unsigned long addr, void *val, unsigned int bytes, struct x86_exception *exception, struct kvm_vcpu *vcpu, struct read_write_emulator_ops *ops) |
bbd9b64e3
|
3554 |
{ |
af7cc7d1e
|
3555 3556 |
gpa_t gpa; int handled, ret; |
22388a3c8
|
3557 3558 3559 3560 3561 |
bool write = ops->write; if (ops->read_write_prepare && ops->read_write_prepare(vcpu, val, bytes)) return X86EMUL_CONTINUE; |
10589a469
|
3562 |
|
22388a3c8
|
3563 |
ret = vcpu_mmio_gva_to_gpa(vcpu, addr, &gpa, exception, write); |
bbd9b64e3
|
3564 |
|
af7cc7d1e
|
3565 |
if (ret < 0) |
bbd9b64e3
|
3566 |
return X86EMUL_PROPAGATE_FAULT; |
bbd9b64e3
|
3567 3568 |
/* For APIC access vmexit */ |
af7cc7d1e
|
3569 |
if (ret) |
bbd9b64e3
|
3570 |
goto mmio; |
22388a3c8
|
3571 |
if (ops->read_write_emulate(vcpu, gpa, val, bytes)) |
bbd9b64e3
|
3572 3573 3574 3575 3576 3577 |
return X86EMUL_CONTINUE; mmio: /* * Is this MMIO handled locally? */ |
22388a3c8
|
3578 |
handled = ops->read_write_mmio(vcpu, gpa, bytes, val); |
70252a105
|
3579 |
if (handled == bytes) |
bbd9b64e3
|
3580 |
return X86EMUL_CONTINUE; |
bbd9b64e3
|
3581 |
|
70252a105
|
3582 3583 3584 |
gpa += handled; bytes -= handled; val += handled; |
bbd9b64e3
|
3585 |
vcpu->mmio_needed = 1; |
411c35b7e
|
3586 3587 |
vcpu->run->exit_reason = KVM_EXIT_MMIO; vcpu->run->mmio.phys_addr = vcpu->mmio_phys_addr = gpa; |
cef4dea07
|
3588 3589 |
vcpu->mmio_size = bytes; vcpu->run->mmio.len = min(vcpu->mmio_size, 8); |
22388a3c8
|
3590 |
vcpu->run->mmio.is_write = vcpu->mmio_is_write = write; |
cef4dea07
|
3591 |
vcpu->mmio_index = 0; |
bbd9b64e3
|
3592 |
|
22388a3c8
|
3593 |
return ops->read_write_exit_mmio(vcpu, gpa, val, bytes); |
bbd9b64e3
|
3594 |
} |
22388a3c8
|
3595 3596 3597 3598 |
int emulator_read_write(struct x86_emulate_ctxt *ctxt, unsigned long addr, void *val, unsigned int bytes, struct x86_exception *exception, struct read_write_emulator_ops *ops) |
bbd9b64e3
|
3599 |
{ |
0f65dd70a
|
3600 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
bbd9b64e3
|
3601 3602 3603 3604 3605 |
/* Crossing a page boundary? */ if (((addr + bytes - 1) ^ addr) & PAGE_MASK) { int rc, now; now = -addr & ~PAGE_MASK; |
22388a3c8
|
3606 3607 |
rc = emulator_read_write_onepage(addr, val, now, exception, vcpu, ops); |
bbd9b64e3
|
3608 3609 3610 3611 3612 3613 |
if (rc != X86EMUL_CONTINUE) return rc; addr += now; val += now; bytes -= now; } |
22388a3c8
|
3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 |
return emulator_read_write_onepage(addr, val, bytes, exception, vcpu, ops); } static int emulator_read_emulated(struct x86_emulate_ctxt *ctxt, unsigned long addr, void *val, unsigned int bytes, struct x86_exception *exception) { return emulator_read_write(ctxt, addr, val, bytes, exception, &read_emultor); } int emulator_write_emulated(struct x86_emulate_ctxt *ctxt, unsigned long addr, const void *val, unsigned int bytes, struct x86_exception *exception) { return emulator_read_write(ctxt, addr, (void *)val, bytes, exception, &write_emultor); |
bbd9b64e3
|
3637 |
} |
bbd9b64e3
|
3638 |
|
daea3e73c
|
3639 3640 3641 3642 3643 3644 3645 |
#define CMPXCHG_TYPE(t, ptr, old, new) \ (cmpxchg((t *)(ptr), *(t *)(old), *(t *)(new)) == *(t *)(old)) #ifdef CONFIG_X86_64 # define CMPXCHG64(ptr, old, new) CMPXCHG_TYPE(u64, ptr, old, new) #else # define CMPXCHG64(ptr, old, new) \ |
9749a6c0f
|
3646 |
(cmpxchg64((u64 *)(ptr), *(u64 *)(old), *(u64 *)(new)) == *(u64 *)(old)) |
daea3e73c
|
3647 |
#endif |
0f65dd70a
|
3648 3649 |
static int emulator_cmpxchg_emulated(struct x86_emulate_ctxt *ctxt, unsigned long addr, |
bbd9b64e3
|
3650 3651 3652 |
const void *old, const void *new, unsigned int bytes, |
0f65dd70a
|
3653 |
struct x86_exception *exception) |
bbd9b64e3
|
3654 |
{ |
0f65dd70a
|
3655 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
daea3e73c
|
3656 3657 3658 3659 |
gpa_t gpa; struct page *page; char *kaddr; bool exchanged; |
2bacc55c7
|
3660 |
|
daea3e73c
|
3661 3662 3663 |
/* guests cmpxchg8b have to be emulated atomically */ if (bytes > 8 || (bytes & (bytes - 1))) goto emul_write; |
10589a469
|
3664 |
|
daea3e73c
|
3665 |
gpa = kvm_mmu_gva_to_gpa_write(vcpu, addr, NULL); |
2bacc55c7
|
3666 |
|
daea3e73c
|
3667 3668 3669 |
if (gpa == UNMAPPED_GVA || (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) goto emul_write; |
2bacc55c7
|
3670 |
|
daea3e73c
|
3671 3672 |
if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK)) goto emul_write; |
72dc67a69
|
3673 |
|
daea3e73c
|
3674 |
page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT); |
c19b8bd60
|
3675 3676 3677 3678 |
if (is_error_page(page)) { kvm_release_page_clean(page); goto emul_write; } |
72dc67a69
|
3679 |
|
daea3e73c
|
3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 |
kaddr = kmap_atomic(page, KM_USER0); kaddr += offset_in_page(gpa); switch (bytes) { case 1: exchanged = CMPXCHG_TYPE(u8, kaddr, old, new); break; case 2: exchanged = CMPXCHG_TYPE(u16, kaddr, old, new); break; case 4: exchanged = CMPXCHG_TYPE(u32, kaddr, old, new); break; case 8: exchanged = CMPXCHG64(kaddr, old, new); break; default: BUG(); |
2bacc55c7
|
3697 |
} |
daea3e73c
|
3698 3699 3700 3701 3702 |
kunmap_atomic(kaddr, KM_USER0); kvm_release_page_dirty(page); if (!exchanged) return X86EMUL_CMPXCHG_FAILED; |
f57f2ef58
|
3703 |
kvm_mmu_pte_write(vcpu, gpa, new, bytes); |
8f6abd06f
|
3704 3705 |
return X86EMUL_CONTINUE; |
4a5f48f66
|
3706 |
|
3200f405a
|
3707 |
emul_write: |
daea3e73c
|
3708 3709 |
printk_once(KERN_WARNING "kvm: emulating exchange as write "); |
2bacc55c7
|
3710 |
|
0f65dd70a
|
3711 |
return emulator_write_emulated(ctxt, addr, new, bytes, exception); |
bbd9b64e3
|
3712 |
} |
cf8f70bfe
|
3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 |
static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) { /* TODO: String I/O for in kernel device */ int r; if (vcpu->arch.pio.in) r = kvm_io_bus_read(vcpu->kvm, KVM_PIO_BUS, vcpu->arch.pio.port, vcpu->arch.pio.size, pd); else r = kvm_io_bus_write(vcpu->kvm, KVM_PIO_BUS, vcpu->arch.pio.port, vcpu->arch.pio.size, pd); return r; } |
6f6fbe98c
|
3727 3728 3729 |
static int emulator_pio_in_out(struct kvm_vcpu *vcpu, int size, unsigned short port, void *val, unsigned int count, bool in) |
cf8f70bfe
|
3730 |
{ |
6f6fbe98c
|
3731 |
trace_kvm_pio(!in, port, size, count); |
cf8f70bfe
|
3732 3733 |
vcpu->arch.pio.port = port; |
6f6fbe98c
|
3734 |
vcpu->arch.pio.in = in; |
7972995b0
|
3735 |
vcpu->arch.pio.count = count; |
cf8f70bfe
|
3736 3737 3738 |
vcpu->arch.pio.size = size; if (!kernel_pio(vcpu, vcpu->arch.pio_data)) { |
7972995b0
|
3739 |
vcpu->arch.pio.count = 0; |
cf8f70bfe
|
3740 3741 3742 3743 |
return 1; } vcpu->run->exit_reason = KVM_EXIT_IO; |
6f6fbe98c
|
3744 |
vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT; |
cf8f70bfe
|
3745 3746 3747 3748 3749 3750 3751 |
vcpu->run->io.size = size; vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE; vcpu->run->io.count = count; vcpu->run->io.port = port; return 0; } |
6f6fbe98c
|
3752 3753 3754 |
static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt, int size, unsigned short port, void *val, unsigned int count) |
cf8f70bfe
|
3755 |
{ |
ca1d4a9e7
|
3756 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
6f6fbe98c
|
3757 |
int ret; |
ca1d4a9e7
|
3758 |
|
6f6fbe98c
|
3759 3760 |
if (vcpu->arch.pio.count) goto data_avail; |
cf8f70bfe
|
3761 |
|
6f6fbe98c
|
3762 3763 3764 3765 |
ret = emulator_pio_in_out(vcpu, size, port, val, count, true); if (ret) { data_avail: memcpy(val, vcpu->arch.pio_data, size * count); |
7972995b0
|
3766 |
vcpu->arch.pio.count = 0; |
cf8f70bfe
|
3767 3768 |
return 1; } |
cf8f70bfe
|
3769 3770 |
return 0; } |
6f6fbe98c
|
3771 3772 3773 3774 3775 3776 3777 3778 3779 |
static int emulator_pio_out_emulated(struct x86_emulate_ctxt *ctxt, int size, unsigned short port, const void *val, unsigned int count) { struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); memcpy(vcpu->arch.pio_data, val, size * count); return emulator_pio_in_out(vcpu, size, port, (void *)val, count, false); } |
bbd9b64e3
|
3780 3781 3782 3783 |
static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) { return kvm_x86_ops->get_segment_base(vcpu, seg); } |
3cb16fe78
|
3784 |
static void emulator_invlpg(struct x86_emulate_ctxt *ctxt, ulong address) |
bbd9b64e3
|
3785 |
{ |
3cb16fe78
|
3786 |
kvm_mmu_invlpg(emul_to_vcpu(ctxt), address); |
bbd9b64e3
|
3787 |
} |
f5f48ee15
|
3788 3789 3790 3791 3792 3793 |
int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) { if (!need_emulate_wbinvd(vcpu)) return X86EMUL_CONTINUE; if (kvm_x86_ops->has_wbinvd_exit()) { |
2eec73437
|
3794 3795 3796 |
int cpu = get_cpu(); cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask); |
f5f48ee15
|
3797 3798 |
smp_call_function_many(vcpu->arch.wbinvd_dirty_mask, wbinvd_ipi, NULL, 1); |
2eec73437
|
3799 |
put_cpu(); |
f5f48ee15
|
3800 |
cpumask_clear(vcpu->arch.wbinvd_dirty_mask); |
2eec73437
|
3801 3802 |
} else wbinvd(); |
f5f48ee15
|
3803 3804 3805 |
return X86EMUL_CONTINUE; } EXPORT_SYMBOL_GPL(kvm_emulate_wbinvd); |
bcaf5cc54
|
3806 3807 3808 3809 |
static void emulator_wbinvd(struct x86_emulate_ctxt *ctxt) { kvm_emulate_wbinvd(emul_to_vcpu(ctxt)); } |
717746e38
|
3810 |
int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long *dest) |
bbd9b64e3
|
3811 |
{ |
717746e38
|
3812 |
return _kvm_get_dr(emul_to_vcpu(ctxt), dr, dest); |
bbd9b64e3
|
3813 |
} |
717746e38
|
3814 |
int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) |
bbd9b64e3
|
3815 |
{ |
338dbc978
|
3816 |
|
717746e38
|
3817 |
return __kvm_set_dr(emul_to_vcpu(ctxt), dr, value); |
bbd9b64e3
|
3818 |
} |
52a466173
|
3819 |
static u64 mk_cr_64(u64 curr_cr, u32 new_val) |
5fdbf9765
|
3820 |
{ |
52a466173
|
3821 |
return (curr_cr & ~((1ULL << 32) - 1)) | new_val; |
5fdbf9765
|
3822 |
} |
717746e38
|
3823 |
static unsigned long emulator_get_cr(struct x86_emulate_ctxt *ctxt, int cr) |
bbd9b64e3
|
3824 |
{ |
717746e38
|
3825 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
52a466173
|
3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 |
unsigned long value; switch (cr) { case 0: value = kvm_read_cr0(vcpu); break; case 2: value = vcpu->arch.cr2; break; case 3: |
9f8fe5043
|
3836 |
value = kvm_read_cr3(vcpu); |
52a466173
|
3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 |
break; case 4: value = kvm_read_cr4(vcpu); break; case 8: value = kvm_get_cr8(vcpu); break; default: vcpu_printf(vcpu, "%s: unexpected cr %u ", __func__, cr); return 0; } return value; } |
717746e38
|
3852 |
static int emulator_set_cr(struct x86_emulate_ctxt *ctxt, int cr, ulong val) |
52a466173
|
3853 |
{ |
717746e38
|
3854 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
0f12244fe
|
3855 |
int res = 0; |
52a466173
|
3856 3857 |
switch (cr) { case 0: |
49a9b07ed
|
3858 |
res = kvm_set_cr0(vcpu, mk_cr_64(kvm_read_cr0(vcpu), val)); |
52a466173
|
3859 3860 3861 3862 3863 |
break; case 2: vcpu->arch.cr2 = val; break; case 3: |
2390218b6
|
3864 |
res = kvm_set_cr3(vcpu, val); |
52a466173
|
3865 3866 |
break; case 4: |
a83b29c6a
|
3867 |
res = kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); |
52a466173
|
3868 3869 |
break; case 8: |
eea1cff9a
|
3870 |
res = kvm_set_cr8(vcpu, val); |
52a466173
|
3871 3872 3873 3874 |
break; default: vcpu_printf(vcpu, "%s: unexpected cr %u ", __func__, cr); |
0f12244fe
|
3875 |
res = -1; |
52a466173
|
3876 |
} |
0f12244fe
|
3877 3878 |
return res; |
52a466173
|
3879 |
} |
717746e38
|
3880 |
static int emulator_get_cpl(struct x86_emulate_ctxt *ctxt) |
9c5372445
|
3881 |
{ |
717746e38
|
3882 |
return kvm_x86_ops->get_cpl(emul_to_vcpu(ctxt)); |
9c5372445
|
3883 |
} |
4bff1e86a
|
3884 |
static void emulator_get_gdt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) |
2dafc6c23
|
3885 |
{ |
4bff1e86a
|
3886 |
kvm_x86_ops->get_gdt(emul_to_vcpu(ctxt), dt); |
2dafc6c23
|
3887 |
} |
4bff1e86a
|
3888 |
static void emulator_get_idt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) |
160ce1f1a
|
3889 |
{ |
4bff1e86a
|
3890 |
kvm_x86_ops->get_idt(emul_to_vcpu(ctxt), dt); |
160ce1f1a
|
3891 |
} |
1ac9d0cfb
|
3892 3893 3894 3895 3896 3897 3898 3899 3900 |
static void emulator_set_gdt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) { kvm_x86_ops->set_gdt(emul_to_vcpu(ctxt), dt); } static void emulator_set_idt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) { kvm_x86_ops->set_idt(emul_to_vcpu(ctxt), dt); } |
4bff1e86a
|
3901 3902 |
static unsigned long emulator_get_cached_segment_base( struct x86_emulate_ctxt *ctxt, int seg) |
5951c4423
|
3903 |
{ |
4bff1e86a
|
3904 |
return get_segment_base(emul_to_vcpu(ctxt), seg); |
5951c4423
|
3905 |
} |
1aa366163
|
3906 3907 3908 |
static bool emulator_get_segment(struct x86_emulate_ctxt *ctxt, u16 *selector, struct desc_struct *desc, u32 *base3, int seg) |
2dafc6c23
|
3909 3910 |
{ struct kvm_segment var; |
4bff1e86a
|
3911 |
kvm_get_segment(emul_to_vcpu(ctxt), &var, seg); |
1aa366163
|
3912 |
*selector = var.selector; |
2dafc6c23
|
3913 3914 3915 3916 3917 3918 3919 3920 |
if (var.unusable) return false; if (var.g) var.limit >>= 12; set_desc_limit(desc, var.limit); set_desc_base(desc, (unsigned long)var.base); |
5601d05b8
|
3921 3922 3923 3924 |
#ifdef CONFIG_X86_64 if (base3) *base3 = var.base >> 32; #endif |
2dafc6c23
|
3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 |
desc->type = var.type; desc->s = var.s; desc->dpl = var.dpl; desc->p = var.present; desc->avl = var.avl; desc->l = var.l; desc->d = var.db; desc->g = var.g; return true; } |
1aa366163
|
3936 3937 3938 |
static void emulator_set_segment(struct x86_emulate_ctxt *ctxt, u16 selector, struct desc_struct *desc, u32 base3, int seg) |
2dafc6c23
|
3939 |
{ |
4bff1e86a
|
3940 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
2dafc6c23
|
3941 |
struct kvm_segment var; |
1aa366163
|
3942 |
var.selector = selector; |
2dafc6c23
|
3943 |
var.base = get_desc_base(desc); |
5601d05b8
|
3944 3945 3946 |
#ifdef CONFIG_X86_64 var.base |= ((u64)base3) << 32; #endif |
2dafc6c23
|
3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 |
var.limit = get_desc_limit(desc); if (desc->g) var.limit = (var.limit << 12) | 0xfff; var.type = desc->type; var.present = desc->p; var.dpl = desc->dpl; var.db = desc->d; var.s = desc->s; var.l = desc->l; var.g = desc->g; var.avl = desc->avl; var.present = desc->p; var.unusable = !var.present; var.padding = 0; kvm_set_segment(vcpu, &var, seg); return; } |
717746e38
|
3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 |
static int emulator_get_msr(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 *pdata) { return kvm_get_msr(emul_to_vcpu(ctxt), msr_index, pdata); } static int emulator_set_msr(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 data) { return kvm_set_msr(emul_to_vcpu(ctxt), msr_index, data); } |
222d21aa0
|
3976 3977 3978 3979 3980 |
static int emulator_read_pmc(struct x86_emulate_ctxt *ctxt, u32 pmc, u64 *pdata) { return kvm_pmu_read_pmc(emul_to_vcpu(ctxt), pmc, pdata); } |
6c3287f7c
|
3981 3982 3983 3984 |
static void emulator_halt(struct x86_emulate_ctxt *ctxt) { emul_to_vcpu(ctxt)->arch.halt_request = 1; } |
5037f6f32
|
3985 3986 3987 |
static void emulator_get_fpu(struct x86_emulate_ctxt *ctxt) { preempt_disable(); |
5197b808a
|
3988 |
kvm_load_guest_fpu(emul_to_vcpu(ctxt)); |
5037f6f32
|
3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 |
/* * CR0.TS may reference the host fpu state, not the guest fpu state, * so it may be clear at this point. */ clts(); } static void emulator_put_fpu(struct x86_emulate_ctxt *ctxt) { preempt_enable(); } |
2953538eb
|
4000 |
static int emulator_intercept(struct x86_emulate_ctxt *ctxt, |
8a76d7f25
|
4001 |
struct x86_instruction_info *info, |
c4f035c60
|
4002 4003 |
enum x86_intercept_stage stage) { |
2953538eb
|
4004 |
return kvm_x86_ops->check_intercept(emul_to_vcpu(ctxt), info, stage); |
c4f035c60
|
4005 |
} |
14af3f3c5
|
4006 |
static struct x86_emulate_ops emulate_ops = { |
1871c6020
|
4007 |
.read_std = kvm_read_guest_virt_system, |
2dafc6c23
|
4008 |
.write_std = kvm_write_guest_virt_system, |
1871c6020
|
4009 |
.fetch = kvm_fetch_guest_virt, |
bbd9b64e3
|
4010 4011 4012 |
.read_emulated = emulator_read_emulated, .write_emulated = emulator_write_emulated, .cmpxchg_emulated = emulator_cmpxchg_emulated, |
3cb16fe78
|
4013 |
.invlpg = emulator_invlpg, |
cf8f70bfe
|
4014 4015 |
.pio_in_emulated = emulator_pio_in_emulated, .pio_out_emulated = emulator_pio_out_emulated, |
1aa366163
|
4016 4017 |
.get_segment = emulator_get_segment, .set_segment = emulator_set_segment, |
5951c4423
|
4018 |
.get_cached_segment_base = emulator_get_cached_segment_base, |
2dafc6c23
|
4019 |
.get_gdt = emulator_get_gdt, |
160ce1f1a
|
4020 |
.get_idt = emulator_get_idt, |
1ac9d0cfb
|
4021 4022 |
.set_gdt = emulator_set_gdt, .set_idt = emulator_set_idt, |
52a466173
|
4023 4024 |
.get_cr = emulator_get_cr, .set_cr = emulator_set_cr, |
9c5372445
|
4025 |
.cpl = emulator_get_cpl, |
35aa5375d
|
4026 4027 |
.get_dr = emulator_get_dr, .set_dr = emulator_set_dr, |
717746e38
|
4028 4029 |
.set_msr = emulator_set_msr, .get_msr = emulator_get_msr, |
222d21aa0
|
4030 |
.read_pmc = emulator_read_pmc, |
6c3287f7c
|
4031 |
.halt = emulator_halt, |
bcaf5cc54
|
4032 |
.wbinvd = emulator_wbinvd, |
d6aa10003
|
4033 |
.fix_hypercall = emulator_fix_hypercall, |
5037f6f32
|
4034 4035 |
.get_fpu = emulator_get_fpu, .put_fpu = emulator_put_fpu, |
c4f035c60
|
4036 |
.intercept = emulator_intercept, |
bbd9b64e3
|
4037 |
}; |
5fdbf9765
|
4038 4039 4040 4041 4042 4043 4044 |
static void cache_all_regs(struct kvm_vcpu *vcpu) { kvm_register_read(vcpu, VCPU_REGS_RAX); kvm_register_read(vcpu, VCPU_REGS_RSP); kvm_register_read(vcpu, VCPU_REGS_RIP); vcpu->arch.regs_dirty = ~0; } |
95cb22953
|
4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 |
static void toggle_interruptibility(struct kvm_vcpu *vcpu, u32 mask) { u32 int_shadow = kvm_x86_ops->get_interrupt_shadow(vcpu, mask); /* * an sti; sti; sequence only disable interrupts for the first * instruction. So, if the last instruction, be it emulated or * not, left the system with the INT_STI flag enabled, it * means that the last instruction is an sti. We should not * leave the flag on in this case. The same goes for mov ss */ if (!(int_shadow & mask)) kvm_x86_ops->set_interrupt_shadow(vcpu, mask); } |
54b8486f4
|
4058 4059 4060 |
static void inject_emulated_exception(struct kvm_vcpu *vcpu) { struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; |
da9cb575b
|
4061 |
if (ctxt->exception.vector == PF_VECTOR) |
6389ee946
|
4062 |
kvm_propagate_fault(vcpu, &ctxt->exception); |
da9cb575b
|
4063 4064 4065 |
else if (ctxt->exception.error_code_valid) kvm_queue_exception_e(vcpu, ctxt->exception.vector, ctxt->exception.error_code); |
54b8486f4
|
4066 |
else |
da9cb575b
|
4067 |
kvm_queue_exception(vcpu, ctxt->exception.vector); |
54b8486f4
|
4068 |
} |
9dac77fa4
|
4069 |
static void init_decode_cache(struct x86_emulate_ctxt *ctxt, |
b5c9ff731
|
4070 4071 |
const unsigned long *regs) { |
9dac77fa4
|
4072 4073 4074 |
memset(&ctxt->twobyte, 0, (void *)&ctxt->regs - (void *)&ctxt->twobyte); memcpy(ctxt->regs, regs, sizeof(ctxt->regs)); |
b5c9ff731
|
4075 |
|
9dac77fa4
|
4076 4077 4078 4079 4080 4081 |
ctxt->fetch.start = 0; ctxt->fetch.end = 0; ctxt->io_read.pos = 0; ctxt->io_read.end = 0; ctxt->mem_read.pos = 0; ctxt->mem_read.end = 0; |
b5c9ff731
|
4082 |
} |
8ec4722dd
|
4083 4084 |
static void init_emulate_ctxt(struct kvm_vcpu *vcpu) { |
adf52235b
|
4085 |
struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; |
8ec4722dd
|
4086 |
int cs_db, cs_l; |
2aab2c5b2
|
4087 4088 4089 4090 4091 4092 |
/* * TODO: fix emulate.c to use guest_read/write_register * instead of direct ->regs accesses, can save hundred cycles * on Intel for instructions that don't read/change RSP, for * for example. */ |
8ec4722dd
|
4093 4094 4095 |
cache_all_regs(vcpu); kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); |
adf52235b
|
4096 4097 4098 4099 4100 4101 4102 4103 |
ctxt->eflags = kvm_get_rflags(vcpu); ctxt->eip = kvm_rip_read(vcpu); ctxt->mode = (!is_protmode(vcpu)) ? X86EMUL_MODE_REAL : (ctxt->eflags & X86_EFLAGS_VM) ? X86EMUL_MODE_VM86 : cs_l ? X86EMUL_MODE_PROT64 : cs_db ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16; ctxt->guest_mode = is_guest_mode(vcpu); |
9dac77fa4
|
4104 |
init_decode_cache(ctxt, vcpu->arch.regs); |
7ae441eac
|
4105 |
vcpu->arch.emulate_regs_need_sync_from_vcpu = false; |
8ec4722dd
|
4106 |
} |
71f9833bb
|
4107 |
int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq, int inc_eip) |
63995653a
|
4108 |
{ |
9d74191ab
|
4109 |
struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; |
63995653a
|
4110 4111 4112 |
int ret; init_emulate_ctxt(vcpu); |
9dac77fa4
|
4113 4114 4115 |
ctxt->op_bytes = 2; ctxt->ad_bytes = 2; ctxt->_eip = ctxt->eip + inc_eip; |
9d74191ab
|
4116 |
ret = emulate_int_real(ctxt, irq); |
63995653a
|
4117 4118 4119 |
if (ret != X86EMUL_CONTINUE) return EMULATE_FAIL; |
9dac77fa4
|
4120 4121 |
ctxt->eip = ctxt->_eip; memcpy(vcpu->arch.regs, ctxt->regs, sizeof ctxt->regs); |
9d74191ab
|
4122 4123 |
kvm_rip_write(vcpu, ctxt->eip); kvm_set_rflags(vcpu, ctxt->eflags); |
63995653a
|
4124 4125 |
if (irq == NMI_VECTOR) |
7460fb4a3
|
4126 |
vcpu->arch.nmi_pending = 0; |
63995653a
|
4127 4128 4129 4130 4131 4132 |
else vcpu->arch.interrupt.pending = false; return EMULATE_DONE; } EXPORT_SYMBOL_GPL(kvm_inject_realmode_interrupt); |
6d77dbfc8
|
4133 4134 |
static int handle_emulation_failure(struct kvm_vcpu *vcpu) { |
fc3a9157d
|
4135 |
int r = EMULATE_DONE; |
6d77dbfc8
|
4136 4137 |
++vcpu->stat.insn_emulation_fail; trace_kvm_emulate_insn_failed(vcpu); |
fc3a9157d
|
4138 4139 4140 4141 4142 4143 |
if (!is_guest_mode(vcpu)) { vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; vcpu->run->internal.ndata = 0; r = EMULATE_FAIL; } |
6d77dbfc8
|
4144 |
kvm_queue_exception(vcpu, UD_VECTOR); |
fc3a9157d
|
4145 4146 |
return r; |
6d77dbfc8
|
4147 |
} |
a6f177efa
|
4148 4149 4150 |
static bool reexecute_instruction(struct kvm_vcpu *vcpu, gva_t gva) { gpa_t gpa; |
68be08034
|
4151 4152 |
if (tdp_enabled) return false; |
a6f177efa
|
4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 |
/* * if emulation was due to access to shadowed page table * and it failed try to unshadow page and re-entetr the * guest to let CPU execute the instruction. */ if (kvm_mmu_unprotect_page_virt(vcpu, gva)) return true; gpa = kvm_mmu_gva_to_gpa_system(vcpu, gva, NULL); if (gpa == UNMAPPED_GVA) return true; /* let cpu generate fault */ if (!kvm_is_error_hva(gfn_to_hva(vcpu->kvm, gpa >> PAGE_SHIFT))) return true; return false; } |
1cb3f3ae5
|
4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 |
static bool retry_instruction(struct x86_emulate_ctxt *ctxt, unsigned long cr2, int emulation_type) { struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); unsigned long last_retry_eip, last_retry_addr, gpa = cr2; last_retry_eip = vcpu->arch.last_retry_eip; last_retry_addr = vcpu->arch.last_retry_addr; /* * If the emulation is caused by #PF and it is non-page_table * writing instruction, it means the VM-EXIT is caused by shadow * page protected, we can zap the shadow page and retry this * instruction directly. * * Note: if the guest uses a non-page-table modifying instruction * on the PDE that points to the instruction, then we will unmap * the instruction and go to an infinite loop. So, we cache the * last retried eip and the last fault address, if we meet the eip * and the address again, we can break out of the potential infinite * loop. */ vcpu->arch.last_retry_eip = vcpu->arch.last_retry_addr = 0; if (!(emulation_type & EMULTYPE_RETRY)) return false; if (x86_page_table_writing_insn(ctxt)) return false; if (ctxt->eip == last_retry_eip && last_retry_addr == cr2) return false; vcpu->arch.last_retry_eip = ctxt->eip; vcpu->arch.last_retry_addr = cr2; if (!vcpu->arch.mmu.direct_map) gpa = kvm_mmu_gva_to_gpa_write(vcpu, cr2, NULL); kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT); return true; } |
51d8b6619
|
4214 4215 |
int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, |
dc25e89e0
|
4216 4217 4218 |
int emulation_type, void *insn, int insn_len) |
bbd9b64e3
|
4219 |
{ |
95cb22953
|
4220 |
int r; |
9d74191ab
|
4221 |
struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; |
7ae441eac
|
4222 |
bool writeback = true; |
bbd9b64e3
|
4223 |
|
26eef70c3
|
4224 |
kvm_clear_exception_queue(vcpu); |
8d7d81025
|
4225 |
|
571008dac
|
4226 |
if (!(emulation_type & EMULTYPE_NO_DECODE)) { |
8ec4722dd
|
4227 |
init_emulate_ctxt(vcpu); |
9d74191ab
|
4228 4229 4230 |
ctxt->interruptibility = 0; ctxt->have_exception = false; ctxt->perm_ok = false; |
bbd9b64e3
|
4231 |
|
9d74191ab
|
4232 |
ctxt->only_vendor_specific_insn |
4005996e4
|
4233 |
= emulation_type & EMULTYPE_TRAP_UD; |
9d74191ab
|
4234 |
r = x86_decode_insn(ctxt, insn, insn_len); |
bbd9b64e3
|
4235 |
|
e46479f85
|
4236 |
trace_kvm_emulate_insn_start(vcpu); |
f2b5756bb
|
4237 |
++vcpu->stat.insn_emulation; |
1d2887e2d
|
4238 |
if (r != EMULATION_OK) { |
4005996e4
|
4239 4240 |
if (emulation_type & EMULTYPE_TRAP_UD) return EMULATE_FAIL; |
a6f177efa
|
4241 |
if (reexecute_instruction(vcpu, cr2)) |
bbd9b64e3
|
4242 |
return EMULATE_DONE; |
6d77dbfc8
|
4243 4244 4245 |
if (emulation_type & EMULTYPE_SKIP) return EMULATE_FAIL; return handle_emulation_failure(vcpu); |
bbd9b64e3
|
4246 4247 |
} } |
ba8afb6b0
|
4248 |
if (emulation_type & EMULTYPE_SKIP) { |
9dac77fa4
|
4249 |
kvm_rip_write(vcpu, ctxt->_eip); |
ba8afb6b0
|
4250 4251 |
return EMULATE_DONE; } |
1cb3f3ae5
|
4252 4253 |
if (retry_instruction(ctxt, cr2, emulation_type)) return EMULATE_DONE; |
7ae441eac
|
4254 |
/* this is needed for vmware backdoor interface to work since it |
4d2179e1e
|
4255 |
changes registers values during IO operation */ |
7ae441eac
|
4256 4257 |
if (vcpu->arch.emulate_regs_need_sync_from_vcpu) { vcpu->arch.emulate_regs_need_sync_from_vcpu = false; |
9dac77fa4
|
4258 |
memcpy(ctxt->regs, vcpu->arch.regs, sizeof ctxt->regs); |
7ae441eac
|
4259 |
} |
4d2179e1e
|
4260 |
|
5cd21917d
|
4261 |
restart: |
9d74191ab
|
4262 |
r = x86_emulate_insn(ctxt); |
bbd9b64e3
|
4263 |
|
775fde864
|
4264 4265 |
if (r == EMULATION_INTERCEPTED) return EMULATE_DONE; |
d2ddd1c48
|
4266 |
if (r == EMULATION_FAILED) { |
a6f177efa
|
4267 |
if (reexecute_instruction(vcpu, cr2)) |
c3cd7ffaf
|
4268 |
return EMULATE_DONE; |
6d77dbfc8
|
4269 |
return handle_emulation_failure(vcpu); |
bbd9b64e3
|
4270 |
} |
9d74191ab
|
4271 |
if (ctxt->have_exception) { |
54b8486f4
|
4272 |
inject_emulated_exception(vcpu); |
d2ddd1c48
|
4273 4274 |
r = EMULATE_DONE; } else if (vcpu->arch.pio.count) { |
3457e4192
|
4275 4276 |
if (!vcpu->arch.pio.in) vcpu->arch.pio.count = 0; |
7ae441eac
|
4277 4278 |
else writeback = false; |
e85d28f8e
|
4279 |
r = EMULATE_DO_MMIO; |
7ae441eac
|
4280 4281 4282 |
} else if (vcpu->mmio_needed) { if (!vcpu->mmio_is_write) writeback = false; |
e85d28f8e
|
4283 |
r = EMULATE_DO_MMIO; |
7ae441eac
|
4284 |
} else if (r == EMULATION_RESTART) |
5cd21917d
|
4285 |
goto restart; |
d2ddd1c48
|
4286 4287 |
else r = EMULATE_DONE; |
f850e2e60
|
4288 |
|
7ae441eac
|
4289 |
if (writeback) { |
9d74191ab
|
4290 4291 |
toggle_interruptibility(vcpu, ctxt->interruptibility); kvm_set_rflags(vcpu, ctxt->eflags); |
7ae441eac
|
4292 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
9dac77fa4
|
4293 |
memcpy(vcpu->arch.regs, ctxt->regs, sizeof ctxt->regs); |
7ae441eac
|
4294 |
vcpu->arch.emulate_regs_need_sync_to_vcpu = false; |
9d74191ab
|
4295 |
kvm_rip_write(vcpu, ctxt->eip); |
7ae441eac
|
4296 4297 |
} else vcpu->arch.emulate_regs_need_sync_to_vcpu = true; |
e85d28f8e
|
4298 4299 |
return r; |
de7d789ac
|
4300 |
} |
51d8b6619
|
4301 |
EXPORT_SYMBOL_GPL(x86_emulate_instruction); |
de7d789ac
|
4302 |
|
cf8f70bfe
|
4303 |
int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port) |
de7d789ac
|
4304 |
{ |
cf8f70bfe
|
4305 |
unsigned long val = kvm_register_read(vcpu, VCPU_REGS_RAX); |
ca1d4a9e7
|
4306 4307 |
int ret = emulator_pio_out_emulated(&vcpu->arch.emulate_ctxt, size, port, &val, 1); |
cf8f70bfe
|
4308 |
/* do not return to emulator after return from userspace */ |
7972995b0
|
4309 |
vcpu->arch.pio.count = 0; |
de7d789ac
|
4310 4311 |
return ret; } |
cf8f70bfe
|
4312 |
EXPORT_SYMBOL_GPL(kvm_fast_pio_out); |
de7d789ac
|
4313 |
|
8cfdc0008
|
4314 4315 |
static void tsc_bad(void *info) { |
0a3aee0da
|
4316 |
__this_cpu_write(cpu_tsc_khz, 0); |
8cfdc0008
|
4317 4318 4319 |
} static void tsc_khz_changed(void *data) |
c80766040
|
4320 |
{ |
8cfdc0008
|
4321 4322 4323 4324 4325 4326 4327 4328 4329 |
struct cpufreq_freqs *freq = data; unsigned long khz = 0; if (data) khz = freq->new; else if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) khz = cpufreq_quick_get(raw_smp_processor_id()); if (!khz) khz = tsc_khz; |
0a3aee0da
|
4330 |
__this_cpu_write(cpu_tsc_khz, khz); |
c80766040
|
4331 |
} |
c80766040
|
4332 4333 4334 4335 4336 4337 4338 |
static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) { struct cpufreq_freqs *freq = data; struct kvm *kvm; struct kvm_vcpu *vcpu; int i, send_ipi = 0; |
8cfdc0008
|
4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 |
/* * We allow guests to temporarily run on slowing clocks, * provided we notify them after, or to run on accelerating * clocks, provided we notify them before. Thus time never * goes backwards. * * However, we have a problem. We can't atomically update * the frequency of a given CPU from this function; it is * merely a notifier, which can be called from any CPU. * Changing the TSC frequency at arbitrary points in time * requires a recomputation of local variables related to * the TSC for each VCPU. We must flag these local variables * to be updated and be sure the update takes place with the * new frequency before any guests proceed. * * Unfortunately, the combination of hotplug CPU and frequency * change creates an intractable locking scenario; the order * of when these callouts happen is undefined with respect to * CPU hotplug, and they can race with each other. As such, * merely setting per_cpu(cpu_tsc_khz) = X during a hotadd is * undefined; you can actually have a CPU frequency change take * place in between the computation of X and the setting of the * variable. To protect against this problem, all updates of * the per_cpu tsc_khz variable are done in an interrupt * protected IPI, and all callers wishing to update the value * must wait for a synchronous IPI to complete (which is trivial * if the caller is on the CPU already). This establishes the * necessary total order on variable updates. * * Note that because a guest time update may take place * anytime after the setting of the VCPU's request bit, the * correct TSC value must be set before the request. However, * to ensure the update actually makes it to any guest which * starts running in hardware virtualization between the set * and the acquisition of the spinlock, we must also ping the * CPU after setting the request bit. * */ |
c80766040
|
4377 4378 4379 4380 |
if (val == CPUFREQ_PRECHANGE && freq->old > freq->new) return 0; if (val == CPUFREQ_POSTCHANGE && freq->old < freq->new) return 0; |
8cfdc0008
|
4381 4382 |
smp_call_function_single(freq->cpu, tsc_khz_changed, freq, 1); |
c80766040
|
4383 |
|
e935b8372
|
4384 |
raw_spin_lock(&kvm_lock); |
c80766040
|
4385 |
list_for_each_entry(kvm, &vm_list, vm_list) { |
988a2cae6
|
4386 |
kvm_for_each_vcpu(i, vcpu, kvm) { |
c80766040
|
4387 4388 |
if (vcpu->cpu != freq->cpu) continue; |
c285545f8
|
4389 |
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); |
c80766040
|
4390 |
if (vcpu->cpu != smp_processor_id()) |
8cfdc0008
|
4391 |
send_ipi = 1; |
c80766040
|
4392 4393 |
} } |
e935b8372
|
4394 |
raw_spin_unlock(&kvm_lock); |
c80766040
|
4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 |
if (freq->old < freq->new && send_ipi) { /* * We upscale the frequency. Must make the guest * doesn't see old kvmclock values while running with * the new frequency, otherwise we risk the guest sees * time go backwards. * * In case we update the frequency for another cpu * (which might be in guest context) send an interrupt * to kick the cpu out of guest context. Next time * guest context is entered kvmclock will be updated, * so the guest will not see stale values. */ |
8cfdc0008
|
4409 |
smp_call_function_single(freq->cpu, tsc_khz_changed, freq, 1); |
c80766040
|
4410 4411 4412 4413 4414 |
} return 0; } static struct notifier_block kvmclock_cpufreq_notifier_block = { |
8cfdc0008
|
4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 |
.notifier_call = kvmclock_cpufreq_notifier }; static int kvmclock_cpu_notifier(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; switch (action) { case CPU_ONLINE: case CPU_DOWN_FAILED: smp_call_function_single(cpu, tsc_khz_changed, NULL, 1); break; case CPU_DOWN_PREPARE: smp_call_function_single(cpu, tsc_bad, NULL, 1); break; } return NOTIFY_OK; } static struct notifier_block kvmclock_cpu_notifier_block = { .notifier_call = kvmclock_cpu_notifier, .priority = -INT_MAX |
c80766040
|
4438 |
}; |
b820cc0ca
|
4439 4440 4441 |
static void kvm_timer_init(void) { int cpu; |
c285545f8
|
4442 |
max_tsc_khz = tsc_khz; |
8cfdc0008
|
4443 |
register_hotcpu_notifier(&kvmclock_cpu_notifier_block); |
b820cc0ca
|
4444 |
if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) { |
c285545f8
|
4445 4446 4447 |
#ifdef CONFIG_CPU_FREQ struct cpufreq_policy policy; memset(&policy, 0, sizeof(policy)); |
3e26f2309
|
4448 4449 |
cpu = get_cpu(); cpufreq_get_policy(&policy, cpu); |
c285545f8
|
4450 4451 |
if (policy.cpuinfo.max_freq) max_tsc_khz = policy.cpuinfo.max_freq; |
3e26f2309
|
4452 |
put_cpu(); |
c285545f8
|
4453 |
#endif |
b820cc0ca
|
4454 4455 4456 |
cpufreq_register_notifier(&kvmclock_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); } |
c285545f8
|
4457 4458 |
pr_debug("kvm: max_tsc_khz = %ld ", max_tsc_khz); |
8cfdc0008
|
4459 4460 |
for_each_online_cpu(cpu) smp_call_function_single(cpu, tsc_khz_changed, NULL, 1); |
b820cc0ca
|
4461 |
} |
ff9d07a0e
|
4462 |
static DEFINE_PER_CPU(struct kvm_vcpu *, current_vcpu); |
f5132b013
|
4463 |
int kvm_is_in_guest(void) |
ff9d07a0e
|
4464 |
{ |
086c98550
|
4465 |
return __this_cpu_read(current_vcpu) != NULL; |
ff9d07a0e
|
4466 4467 4468 4469 4470 |
} static int kvm_is_user_mode(void) { int user_mode = 3; |
dcf46b944
|
4471 |
|
086c98550
|
4472 4473 |
if (__this_cpu_read(current_vcpu)) user_mode = kvm_x86_ops->get_cpl(__this_cpu_read(current_vcpu)); |
dcf46b944
|
4474 |
|
ff9d07a0e
|
4475 4476 4477 4478 4479 4480 |
return user_mode != 0; } static unsigned long kvm_get_guest_ip(void) { unsigned long ip = 0; |
dcf46b944
|
4481 |
|
086c98550
|
4482 4483 |
if (__this_cpu_read(current_vcpu)) ip = kvm_rip_read(__this_cpu_read(current_vcpu)); |
dcf46b944
|
4484 |
|
ff9d07a0e
|
4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 |
return ip; } static struct perf_guest_info_callbacks kvm_guest_cbs = { .is_in_guest = kvm_is_in_guest, .is_user_mode = kvm_is_user_mode, .get_guest_ip = kvm_get_guest_ip, }; void kvm_before_handle_nmi(struct kvm_vcpu *vcpu) { |
086c98550
|
4496 |
__this_cpu_write(current_vcpu, vcpu); |
ff9d07a0e
|
4497 4498 4499 4500 4501 |
} EXPORT_SYMBOL_GPL(kvm_before_handle_nmi); void kvm_after_handle_nmi(struct kvm_vcpu *vcpu) { |
086c98550
|
4502 |
__this_cpu_write(current_vcpu, NULL); |
ff9d07a0e
|
4503 4504 |
} EXPORT_SYMBOL_GPL(kvm_after_handle_nmi); |
ce88decff
|
4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 |
static void kvm_set_mmio_spte_mask(void) { u64 mask; int maxphyaddr = boot_cpu_data.x86_phys_bits; /* * Set the reserved bits and the present bit of an paging-structure * entry to generate page fault with PFER.RSV = 1. */ mask = ((1ull << (62 - maxphyaddr + 1)) - 1) << maxphyaddr; mask |= 1ull; #ifdef CONFIG_X86_64 /* * If reserved bit is not supported, clear the present bit to disable * mmio page fault. */ if (maxphyaddr == 52) mask &= ~1ull; #endif kvm_mmu_set_mmio_spte_mask(mask); } |
f8c16bbaa
|
4528 |
int kvm_arch_init(void *opaque) |
043405e10
|
4529 |
{ |
b820cc0ca
|
4530 |
int r; |
f8c16bbaa
|
4531 |
struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque; |
f8c16bbaa
|
4532 4533 4534 |
if (kvm_x86_ops) { printk(KERN_ERR "kvm: already loaded the other module "); |
56c6d28a9
|
4535 4536 |
r = -EEXIST; goto out; |
f8c16bbaa
|
4537 4538 4539 4540 4541 |
} if (!ops->cpu_has_kvm_support()) { printk(KERN_ERR "kvm: no hardware support "); |
56c6d28a9
|
4542 4543 |
r = -EOPNOTSUPP; goto out; |
f8c16bbaa
|
4544 4545 4546 4547 |
} if (ops->disabled_by_bios()) { printk(KERN_ERR "kvm: disabled by bios "); |
56c6d28a9
|
4548 4549 |
r = -EOPNOTSUPP; goto out; |
f8c16bbaa
|
4550 |
} |
97db56ce6
|
4551 4552 4553 |
r = kvm_mmu_module_init(); if (r) goto out; |
ce88decff
|
4554 |
kvm_set_mmio_spte_mask(); |
97db56ce6
|
4555 |
kvm_init_msr_list(); |
f8c16bbaa
|
4556 |
kvm_x86_ops = ops; |
7b52345e2
|
4557 |
kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, |
4b12f0de3
|
4558 |
PT_DIRTY_MASK, PT64_NX_MASK, 0); |
c80766040
|
4559 |
|
b820cc0ca
|
4560 |
kvm_timer_init(); |
c80766040
|
4561 |
|
ff9d07a0e
|
4562 |
perf_register_guest_info_callbacks(&kvm_guest_cbs); |
2acf923e3
|
4563 4564 |
if (cpu_has_xsave) host_xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); |
f8c16bbaa
|
4565 |
return 0; |
56c6d28a9
|
4566 4567 |
out: |
56c6d28a9
|
4568 |
return r; |
043405e10
|
4569 |
} |
8776e5194
|
4570 |
|
f8c16bbaa
|
4571 4572 |
void kvm_arch_exit(void) { |
ff9d07a0e
|
4573 |
perf_unregister_guest_info_callbacks(&kvm_guest_cbs); |
888d256e9
|
4574 4575 4576 |
if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) cpufreq_unregister_notifier(&kvmclock_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); |
8cfdc0008
|
4577 |
unregister_hotcpu_notifier(&kvmclock_cpu_notifier_block); |
f8c16bbaa
|
4578 |
kvm_x86_ops = NULL; |
56c6d28a9
|
4579 4580 |
kvm_mmu_module_exit(); } |
f8c16bbaa
|
4581 |
|
8776e5194
|
4582 4583 4584 4585 |
int kvm_emulate_halt(struct kvm_vcpu *vcpu) { ++vcpu->stat.halt_exits; if (irqchip_in_kernel(vcpu->kvm)) { |
a45352908
|
4586 |
vcpu->arch.mp_state = KVM_MP_STATE_HALTED; |
8776e5194
|
4587 4588 4589 4590 4591 4592 4593 |
return 1; } else { vcpu->run->exit_reason = KVM_EXIT_HLT; return 0; } } EXPORT_SYMBOL_GPL(kvm_emulate_halt); |
55cd8e5a4
|
4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 |
int kvm_hv_hypercall(struct kvm_vcpu *vcpu) { u64 param, ingpa, outgpa, ret; uint16_t code, rep_idx, rep_cnt, res = HV_STATUS_SUCCESS, rep_done = 0; bool fast, longmode; int cs_db, cs_l; /* * hypercall generates UD from non zero cpl and real mode * per HYPER-V spec */ |
3eeb3288b
|
4605 |
if (kvm_x86_ops->get_cpl(vcpu) != 0 || !is_protmode(vcpu)) { |
55cd8e5a4
|
4606 4607 4608 4609 4610 4611 4612 4613 |
kvm_queue_exception(vcpu, UD_VECTOR); return 0; } kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); longmode = is_long_mode(vcpu) && cs_l == 1; if (!longmode) { |
ccd469362
|
4614 4615 4616 4617 4618 4619 |
param = ((u64)kvm_register_read(vcpu, VCPU_REGS_RDX) << 32) | (kvm_register_read(vcpu, VCPU_REGS_RAX) & 0xffffffff); ingpa = ((u64)kvm_register_read(vcpu, VCPU_REGS_RBX) << 32) | (kvm_register_read(vcpu, VCPU_REGS_RCX) & 0xffffffff); outgpa = ((u64)kvm_register_read(vcpu, VCPU_REGS_RDI) << 32) | (kvm_register_read(vcpu, VCPU_REGS_RSI) & 0xffffffff); |
55cd8e5a4
|
4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 |
} #ifdef CONFIG_X86_64 else { param = kvm_register_read(vcpu, VCPU_REGS_RCX); ingpa = kvm_register_read(vcpu, VCPU_REGS_RDX); outgpa = kvm_register_read(vcpu, VCPU_REGS_R8); } #endif code = param & 0xffff; fast = (param >> 16) & 0x1; rep_cnt = (param >> 32) & 0xfff; rep_idx = (param >> 48) & 0xfff; trace_kvm_hv_hypercall(code, fast, rep_cnt, rep_idx, ingpa, outgpa); |
c25bc1638
|
4635 4636 4637 4638 4639 4640 4641 4642 |
switch (code) { case HV_X64_HV_NOTIFY_LONG_SPIN_WAIT: kvm_vcpu_on_spin(vcpu); break; default: res = HV_STATUS_INVALID_HYPERCALL_CODE; break; } |
55cd8e5a4
|
4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 |
ret = res | (((u64)rep_done & 0xfff) << 32); if (longmode) { kvm_register_write(vcpu, VCPU_REGS_RAX, ret); } else { kvm_register_write(vcpu, VCPU_REGS_RDX, ret >> 32); kvm_register_write(vcpu, VCPU_REGS_RAX, ret & 0xffffffff); } return 1; } |
8776e5194
|
4654 4655 4656 |
int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) { unsigned long nr, a0, a1, a2, a3, ret; |
2f333bcb4
|
4657 |
int r = 1; |
8776e5194
|
4658 |
|
55cd8e5a4
|
4659 4660 |
if (kvm_hv_hypercall_enabled(vcpu->kvm)) return kvm_hv_hypercall(vcpu); |
5fdbf9765
|
4661 4662 4663 4664 4665 |
nr = kvm_register_read(vcpu, VCPU_REGS_RAX); a0 = kvm_register_read(vcpu, VCPU_REGS_RBX); a1 = kvm_register_read(vcpu, VCPU_REGS_RCX); a2 = kvm_register_read(vcpu, VCPU_REGS_RDX); a3 = kvm_register_read(vcpu, VCPU_REGS_RSI); |
8776e5194
|
4666 |
|
229456fc3
|
4667 |
trace_kvm_hypercall(nr, a0, a1, a2, a3); |
2714d1d3d
|
4668 |
|
8776e5194
|
4669 4670 4671 4672 4673 4674 4675 |
if (!is_long_mode(vcpu)) { nr &= 0xFFFFFFFF; a0 &= 0xFFFFFFFF; a1 &= 0xFFFFFFFF; a2 &= 0xFFFFFFFF; a3 &= 0xFFFFFFFF; } |
07708c4af
|
4676 4677 4678 4679 |
if (kvm_x86_ops->get_cpl(vcpu) != 0) { ret = -KVM_EPERM; goto out; } |
8776e5194
|
4680 |
switch (nr) { |
b93463aa5
|
4681 4682 4683 |
case KVM_HC_VAPIC_POLL_IRQ: ret = 0; break; |
8776e5194
|
4684 4685 4686 4687 |
default: ret = -KVM_ENOSYS; break; } |
07708c4af
|
4688 |
out: |
5fdbf9765
|
4689 |
kvm_register_write(vcpu, VCPU_REGS_RAX, ret); |
f11c3a8d8
|
4690 |
++vcpu->stat.hypercalls; |
2f333bcb4
|
4691 |
return r; |
8776e5194
|
4692 4693 |
} EXPORT_SYMBOL_GPL(kvm_emulate_hypercall); |
d6aa10003
|
4694 |
int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt) |
8776e5194
|
4695 |
{ |
d6aa10003
|
4696 |
struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); |
8776e5194
|
4697 |
char instruction[3]; |
5fdbf9765
|
4698 |
unsigned long rip = kvm_rip_read(vcpu); |
8776e5194
|
4699 |
|
8776e5194
|
4700 4701 4702 4703 4704 4705 |
/* * Blow out the MMU to ensure that no other VCPU has an active mapping * to ensure that the updated hypercall appears atomically across all * VCPUs. */ kvm_mmu_zap_all(vcpu->kvm); |
8776e5194
|
4706 |
kvm_x86_ops->patch_hypercall(vcpu, instruction); |
8776e5194
|
4707 |
|
9d74191ab
|
4708 |
return emulator_write_emulated(ctxt, rip, instruction, 3, NULL); |
8776e5194
|
4709 |
} |
d07520607
|
4710 |
/* |
b6c7a5dcc
|
4711 4712 4713 4714 4715 |
* Check if userspace requested an interrupt window, and that the * interrupt window is open. * * No need to exit to userspace if we already have an interrupt queued. */ |
851ba6922
|
4716 |
static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu) |
b6c7a5dcc
|
4717 |
{ |
8061823a2
|
4718 |
return (!irqchip_in_kernel(vcpu->kvm) && !kvm_cpu_has_interrupt(vcpu) && |
851ba6922
|
4719 |
vcpu->run->request_interrupt_window && |
5df566464
|
4720 |
kvm_arch_interrupt_allowed(vcpu)); |
b6c7a5dcc
|
4721 |
} |
851ba6922
|
4722 |
static void post_kvm_run_save(struct kvm_vcpu *vcpu) |
b6c7a5dcc
|
4723 |
{ |
851ba6922
|
4724 |
struct kvm_run *kvm_run = vcpu->run; |
91586a3b7
|
4725 |
kvm_run->if_flag = (kvm_get_rflags(vcpu) & X86_EFLAGS_IF) != 0; |
2d3ad1f40
|
4726 |
kvm_run->cr8 = kvm_get_cr8(vcpu); |
b6c7a5dcc
|
4727 |
kvm_run->apic_base = kvm_get_apic_base(vcpu); |
4531220b7
|
4728 |
if (irqchip_in_kernel(vcpu->kvm)) |
b6c7a5dcc
|
4729 |
kvm_run->ready_for_interrupt_injection = 1; |
4531220b7
|
4730 |
else |
b6c7a5dcc
|
4731 |
kvm_run->ready_for_interrupt_injection = |
fa9726b07
|
4732 4733 4734 |
kvm_arch_interrupt_allowed(vcpu) && !kvm_cpu_has_interrupt(vcpu) && !kvm_event_needs_reinjection(vcpu); |
b6c7a5dcc
|
4735 |
} |
b93463aa5
|
4736 4737 4738 4739 4740 4741 4742 4743 4744 |
static void vapic_enter(struct kvm_vcpu *vcpu) { struct kvm_lapic *apic = vcpu->arch.apic; struct page *page; if (!apic || !apic->vapic_addr) return; page = gfn_to_page(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT); |
72dc67a69
|
4745 4746 |
vcpu->arch.apic->vapic_page = page; |
b93463aa5
|
4747 4748 4749 4750 4751 |
} static void vapic_exit(struct kvm_vcpu *vcpu) { struct kvm_lapic *apic = vcpu->arch.apic; |
f656ce018
|
4752 |
int idx; |
b93463aa5
|
4753 4754 4755 |
if (!apic || !apic->vapic_addr) return; |
f656ce018
|
4756 |
idx = srcu_read_lock(&vcpu->kvm->srcu); |
b93463aa5
|
4757 4758 |
kvm_release_page_dirty(apic->vapic_page); mark_page_dirty(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT); |
f656ce018
|
4759 |
srcu_read_unlock(&vcpu->kvm->srcu, idx); |
b93463aa5
|
4760 |
} |
95ba82731
|
4761 4762 4763 4764 4765 4766 |
static void update_cr8_intercept(struct kvm_vcpu *vcpu) { int max_irr, tpr; if (!kvm_x86_ops->update_cr8_intercept) return; |
88c808fd4
|
4767 4768 |
if (!vcpu->arch.apic) return; |
8db3baa2d
|
4769 4770 4771 4772 |
if (!vcpu->arch.apic->vapic_addr) max_irr = kvm_lapic_find_highest_irr(vcpu); else max_irr = -1; |
95ba82731
|
4773 4774 4775 4776 4777 4778 4779 4780 |
if (max_irr != -1) max_irr >>= 4; tpr = kvm_lapic_get_cr8(vcpu); kvm_x86_ops->update_cr8_intercept(vcpu, tpr, max_irr); } |
851ba6922
|
4781 |
static void inject_pending_event(struct kvm_vcpu *vcpu) |
95ba82731
|
4782 4783 |
{ /* try to reinject previous events if any */ |
b59bb7bdf
|
4784 |
if (vcpu->arch.exception.pending) { |
5c1c85d08
|
4785 4786 4787 |
trace_kvm_inj_exception(vcpu->arch.exception.nr, vcpu->arch.exception.has_error_code, vcpu->arch.exception.error_code); |
b59bb7bdf
|
4788 4789 |
kvm_x86_ops->queue_exception(vcpu, vcpu->arch.exception.nr, vcpu->arch.exception.has_error_code, |
ce7ddec4b
|
4790 4791 |
vcpu->arch.exception.error_code, vcpu->arch.exception.reinject); |
b59bb7bdf
|
4792 4793 |
return; } |
95ba82731
|
4794 4795 4796 4797 4798 4799 |
if (vcpu->arch.nmi_injected) { kvm_x86_ops->set_nmi(vcpu); return; } if (vcpu->arch.interrupt.pending) { |
66fd3f7f9
|
4800 |
kvm_x86_ops->set_irq(vcpu); |
95ba82731
|
4801 4802 4803 4804 4805 4806 |
return; } /* try to inject new event if pending */ if (vcpu->arch.nmi_pending) { if (kvm_x86_ops->nmi_allowed(vcpu)) { |
7460fb4a3
|
4807 |
--vcpu->arch.nmi_pending; |
95ba82731
|
4808 4809 4810 4811 4812 |
vcpu->arch.nmi_injected = true; kvm_x86_ops->set_nmi(vcpu); } } else if (kvm_cpu_has_interrupt(vcpu)) { if (kvm_x86_ops->interrupt_allowed(vcpu)) { |
66fd3f7f9
|
4813 4814 4815 |
kvm_queue_interrupt(vcpu, kvm_cpu_get_interrupt(vcpu), false); kvm_x86_ops->set_irq(vcpu); |
95ba82731
|
4816 4817 4818 |
} } } |
2acf923e3
|
4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 |
static void kvm_load_guest_xcr0(struct kvm_vcpu *vcpu) { if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE) && !vcpu->guest_xcr0_loaded) { /* kvm_set_xcr() also depends on this */ xsetbv(XCR_XFEATURE_ENABLED_MASK, vcpu->arch.xcr0); vcpu->guest_xcr0_loaded = 1; } } static void kvm_put_guest_xcr0(struct kvm_vcpu *vcpu) { if (vcpu->guest_xcr0_loaded) { if (vcpu->arch.xcr0 != host_xcr0) xsetbv(XCR_XFEATURE_ENABLED_MASK, host_xcr0); vcpu->guest_xcr0_loaded = 0; } } |
7460fb4a3
|
4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 |
static void process_nmi(struct kvm_vcpu *vcpu) { unsigned limit = 2; /* * x86 is limited to one NMI running, and one NMI pending after it. * If an NMI is already in progress, limit further NMIs to just one. * Otherwise, allow two (and we'll inject the first one immediately). */ if (kvm_x86_ops->get_nmi_mask(vcpu) || vcpu->arch.nmi_injected) limit = 1; vcpu->arch.nmi_pending += atomic_xchg(&vcpu->arch.nmi_queued, 0); vcpu->arch.nmi_pending = min(vcpu->arch.nmi_pending, limit); kvm_make_request(KVM_REQ_EVENT, vcpu); } |
851ba6922
|
4853 |
static int vcpu_enter_guest(struct kvm_vcpu *vcpu) |
b6c7a5dcc
|
4854 4855 |
{ int r; |
6a8b1d131
|
4856 |
bool req_int_win = !irqchip_in_kernel(vcpu->kvm) && |
851ba6922
|
4857 |
vcpu->run->request_interrupt_window; |
d6185f20a
|
4858 |
bool req_immediate_exit = 0; |
b6c7a5dcc
|
4859 |
|
3e0075094
|
4860 |
if (vcpu->requests) { |
a8eeb04a4
|
4861 |
if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) |
2e53d63ac
|
4862 |
kvm_mmu_unload(vcpu); |
a8eeb04a4
|
4863 |
if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu)) |
2f5997140
|
4864 |
__kvm_migrate_timers(vcpu); |
34c238a1d
|
4865 4866 |
if (kvm_check_request(KVM_REQ_CLOCK_UPDATE, vcpu)) { r = kvm_guest_time_update(vcpu); |
8cfdc0008
|
4867 4868 4869 |
if (unlikely(r)) goto out; } |
a8eeb04a4
|
4870 |
if (kvm_check_request(KVM_REQ_MMU_SYNC, vcpu)) |
4731d4c7a
|
4871 |
kvm_mmu_sync_roots(vcpu); |
a8eeb04a4
|
4872 |
if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) |
d4acf7e7a
|
4873 |
kvm_x86_ops->tlb_flush(vcpu); |
a8eeb04a4
|
4874 |
if (kvm_check_request(KVM_REQ_REPORT_TPR_ACCESS, vcpu)) { |
851ba6922
|
4875 |
vcpu->run->exit_reason = KVM_EXIT_TPR_ACCESS; |
b93463aa5
|
4876 4877 4878 |
r = 0; goto out; } |
a8eeb04a4
|
4879 |
if (kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu)) { |
851ba6922
|
4880 |
vcpu->run->exit_reason = KVM_EXIT_SHUTDOWN; |
71c4dfafc
|
4881 4882 4883 |
r = 0; goto out; } |
a8eeb04a4
|
4884 |
if (kvm_check_request(KVM_REQ_DEACTIVATE_FPU, vcpu)) { |
02daab21d
|
4885 4886 4887 |
vcpu->fpu_active = 0; kvm_x86_ops->fpu_deactivate(vcpu); } |
af585b921
|
4888 4889 4890 4891 4892 4893 |
if (kvm_check_request(KVM_REQ_APF_HALT, vcpu)) { /* Page is swapped out. Do synthetic halt */ vcpu->arch.apf.halted = true; r = 1; goto out; } |
c9aaa8957
|
4894 4895 |
if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu)) record_steal_time(vcpu); |
7460fb4a3
|
4896 4897 |
if (kvm_check_request(KVM_REQ_NMI, vcpu)) process_nmi(vcpu); |
d6185f20a
|
4898 4899 |
req_immediate_exit = kvm_check_request(KVM_REQ_IMMEDIATE_EXIT, vcpu); |
f5132b013
|
4900 4901 4902 4903 |
if (kvm_check_request(KVM_REQ_PMU, vcpu)) kvm_handle_pmu_event(vcpu); if (kvm_check_request(KVM_REQ_PMI, vcpu)) kvm_deliver_pmi(vcpu); |
2f52d58c9
|
4904 |
} |
b93463aa5
|
4905 |
|
3e0075094
|
4906 4907 4908 |
r = kvm_mmu_reload(vcpu); if (unlikely(r)) goto out; |
b463a6f74
|
4909 4910 4911 4912 |
if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) { inject_pending_event(vcpu); /* enable NMI/IRQ window open exits if needed */ |
7460fb4a3
|
4913 |
if (vcpu->arch.nmi_pending) |
b463a6f74
|
4914 4915 4916 4917 4918 4919 4920 4921 4922 |
kvm_x86_ops->enable_nmi_window(vcpu); else if (kvm_cpu_has_interrupt(vcpu) || req_int_win) kvm_x86_ops->enable_irq_window(vcpu); if (kvm_lapic_enabled(vcpu)) { update_cr8_intercept(vcpu); kvm_lapic_sync_to_vapic(vcpu); } } |
b6c7a5dcc
|
4923 4924 4925 |
preempt_disable(); kvm_x86_ops->prepare_guest_switch(vcpu); |
2608d7a12
|
4926 4927 |
if (vcpu->fpu_active) kvm_load_guest_fpu(vcpu); |
2acf923e3
|
4928 |
kvm_load_guest_xcr0(vcpu); |
b6c7a5dcc
|
4929 |
|
6b7e2d099
|
4930 4931 4932 4933 4934 4935 |
vcpu->mode = IN_GUEST_MODE; /* We should set ->mode before check ->requests, * see the comment in make_all_cpus_request. */ smp_mb(); |
b6c7a5dcc
|
4936 |
|
d94e1dc9a
|
4937 |
local_irq_disable(); |
32f884006
|
4938 |
|
6b7e2d099
|
4939 |
if (vcpu->mode == EXITING_GUEST_MODE || vcpu->requests |
d94e1dc9a
|
4940 |
|| need_resched() || signal_pending(current)) { |
6b7e2d099
|
4941 |
vcpu->mode = OUTSIDE_GUEST_MODE; |
d94e1dc9a
|
4942 |
smp_wmb(); |
6c1428012
|
4943 4944 |
local_irq_enable(); preempt_enable(); |
b463a6f74
|
4945 |
kvm_x86_ops->cancel_injection(vcpu); |
6c1428012
|
4946 4947 4948 |
r = 1; goto out; } |
f656ce018
|
4949 |
srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); |
3200f405a
|
4950 |
|
d6185f20a
|
4951 4952 |
if (req_immediate_exit) smp_send_reschedule(vcpu->cpu); |
b6c7a5dcc
|
4953 |
kvm_guest_enter(); |
42dbaa5a0
|
4954 |
if (unlikely(vcpu->arch.switch_db_regs)) { |
42dbaa5a0
|
4955 4956 4957 4958 4959 4960 |
set_debugreg(0, 7); set_debugreg(vcpu->arch.eff_db[0], 0); set_debugreg(vcpu->arch.eff_db[1], 1); set_debugreg(vcpu->arch.eff_db[2], 2); set_debugreg(vcpu->arch.eff_db[3], 3); } |
b6c7a5dcc
|
4961 |
|
229456fc3
|
4962 |
trace_kvm_entry(vcpu->vcpu_id); |
851ba6922
|
4963 |
kvm_x86_ops->run(vcpu); |
b6c7a5dcc
|
4964 |
|
24f1e32c6
|
4965 4966 4967 4968 4969 4970 4971 |
/* * If the guest has used debug registers, at least dr7 * will be disabled while returning to the host. * If we don't have active breakpoints in the host, we don't * care about the messed up debug address registers. But if * we have some of them active, restore the old state. */ |
59d8eb53e
|
4972 |
if (hw_breakpoint_active()) |
24f1e32c6
|
4973 |
hw_breakpoint_restore(); |
42dbaa5a0
|
4974 |
|
d5c1785d2
|
4975 |
vcpu->arch.last_guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu); |
1d5f066e0
|
4976 |
|
6b7e2d099
|
4977 |
vcpu->mode = OUTSIDE_GUEST_MODE; |
d94e1dc9a
|
4978 |
smp_wmb(); |
b6c7a5dcc
|
4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 |
local_irq_enable(); ++vcpu->stat.exits; /* * We must have an instruction between local_irq_enable() and * kvm_guest_exit(), so the timer interrupt isn't delayed by * the interrupt shadow. The stat.exits increment will do nicely. * But we need to prevent reordering, hence this barrier(): */ barrier(); kvm_guest_exit(); preempt_enable(); |
f656ce018
|
4994 |
vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); |
3200f405a
|
4995 |
|
b6c7a5dcc
|
4996 4997 4998 4999 |
/* * Profile KVM exit RIPs: */ if (unlikely(prof_on == KVM_PROFILING)) { |
5fdbf9765
|
5000 5001 |
unsigned long rip = kvm_rip_read(vcpu); profile_hit(KVM_PROFILING, (void *)rip); |
b6c7a5dcc
|
5002 |
} |
298101da2
|
5003 |
|
b93463aa5
|
5004 |
kvm_lapic_sync_from_vapic(vcpu); |
851ba6922
|
5005 |
r = kvm_x86_ops->handle_exit(vcpu); |
d76901750
|
5006 5007 5008 |
out: return r; } |
b6c7a5dcc
|
5009 |
|
09cec7548
|
5010 |
|
851ba6922
|
5011 |
static int __vcpu_run(struct kvm_vcpu *vcpu) |
d76901750
|
5012 5013 |
{ int r; |
f656ce018
|
5014 |
struct kvm *kvm = vcpu->kvm; |
d76901750
|
5015 5016 |
if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED)) { |
1b10bf31a
|
5017 5018 5019 |
pr_debug("vcpu %d received sipi with vector # %x ", vcpu->vcpu_id, vcpu->arch.sipi_vector); |
d76901750
|
5020 |
kvm_lapic_reset(vcpu); |
5f179287f
|
5021 |
r = kvm_arch_vcpu_reset(vcpu); |
d76901750
|
5022 5023 5024 |
if (r) return r; vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; |
b6c7a5dcc
|
5025 |
} |
f656ce018
|
5026 |
vcpu->srcu_idx = srcu_read_lock(&kvm->srcu); |
d76901750
|
5027 5028 5029 5030 |
vapic_enter(vcpu); r = 1; while (r > 0) { |
af585b921
|
5031 5032 |
if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && !vcpu->arch.apf.halted) |
851ba6922
|
5033 |
r = vcpu_enter_guest(vcpu); |
d76901750
|
5034 |
else { |
f656ce018
|
5035 |
srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx); |
d76901750
|
5036 |
kvm_vcpu_block(vcpu); |
f656ce018
|
5037 |
vcpu->srcu_idx = srcu_read_lock(&kvm->srcu); |
a8eeb04a4
|
5038 |
if (kvm_check_request(KVM_REQ_UNHALT, vcpu)) |
09cec7548
|
5039 5040 5041 |
{ switch(vcpu->arch.mp_state) { case KVM_MP_STATE_HALTED: |
d76901750
|
5042 |
vcpu->arch.mp_state = |
09cec7548
|
5043 5044 |
KVM_MP_STATE_RUNNABLE; case KVM_MP_STATE_RUNNABLE: |
af585b921
|
5045 |
vcpu->arch.apf.halted = false; |
09cec7548
|
5046 5047 5048 5049 5050 5051 5052 |
break; case KVM_MP_STATE_SIPI_RECEIVED: default: r = -EINTR; break; } } |
d76901750
|
5053 |
} |
09cec7548
|
5054 5055 5056 5057 5058 5059 |
if (r <= 0) break; clear_bit(KVM_REQ_PENDING_TIMER, &vcpu->requests); if (kvm_cpu_has_pending_timer(vcpu)) kvm_inject_pending_timer_irqs(vcpu); |
851ba6922
|
5060 |
if (dm_request_for_irq_injection(vcpu)) { |
09cec7548
|
5061 |
r = -EINTR; |
851ba6922
|
5062 |
vcpu->run->exit_reason = KVM_EXIT_INTR; |
09cec7548
|
5063 5064 |
++vcpu->stat.request_irq_exits; } |
af585b921
|
5065 5066 |
kvm_check_async_pf_completion(vcpu); |
09cec7548
|
5067 5068 |
if (signal_pending(current)) { r = -EINTR; |
851ba6922
|
5069 |
vcpu->run->exit_reason = KVM_EXIT_INTR; |
09cec7548
|
5070 5071 5072 |
++vcpu->stat.signal_exits; } if (need_resched()) { |
f656ce018
|
5073 |
srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx); |
09cec7548
|
5074 |
kvm_resched(vcpu); |
f656ce018
|
5075 |
vcpu->srcu_idx = srcu_read_lock(&kvm->srcu); |
d76901750
|
5076 |
} |
b6c7a5dcc
|
5077 |
} |
f656ce018
|
5078 |
srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx); |
b6c7a5dcc
|
5079 |
|
b93463aa5
|
5080 |
vapic_exit(vcpu); |
b6c7a5dcc
|
5081 5082 |
return r; } |
5287f194b
|
5083 5084 5085 5086 5087 5088 5089 5090 5091 |
static int complete_mmio(struct kvm_vcpu *vcpu) { struct kvm_run *run = vcpu->run; int r; if (!(vcpu->arch.pio.count || vcpu->mmio_needed)) return 1; if (vcpu->mmio_needed) { |
5287f194b
|
5092 |
vcpu->mmio_needed = 0; |
cef4dea07
|
5093 |
if (!vcpu->mmio_is_write) |
0004c7c25
|
5094 5095 |
memcpy(vcpu->mmio_data + vcpu->mmio_index, run->mmio.data, 8); |
cef4dea07
|
5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 |
vcpu->mmio_index += 8; if (vcpu->mmio_index < vcpu->mmio_size) { run->exit_reason = KVM_EXIT_MMIO; run->mmio.phys_addr = vcpu->mmio_phys_addr + vcpu->mmio_index; memcpy(run->mmio.data, vcpu->mmio_data + vcpu->mmio_index, 8); run->mmio.len = min(vcpu->mmio_size - vcpu->mmio_index, 8); run->mmio.is_write = vcpu->mmio_is_write; vcpu->mmio_needed = 1; return 0; } if (vcpu->mmio_is_write) return 1; vcpu->mmio_read_completed = 1; |
5287f194b
|
5109 5110 5111 5112 5113 5114 5115 5116 |
} vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); r = emulate_instruction(vcpu, EMULTYPE_NO_DECODE); srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); if (r != EMULATE_DONE) return 0; return 1; } |
b6c7a5dcc
|
5117 5118 5119 5120 |
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { int r; sigset_t sigsaved; |
e5c301428
|
5121 5122 |
if (!tsk_used_math(current) && init_fpu(current)) return -ENOMEM; |
ac9f6dc0d
|
5123 5124 |
if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); |
a45352908
|
5125 |
if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_UNINITIALIZED)) { |
b6c7a5dcc
|
5126 |
kvm_vcpu_block(vcpu); |
d76901750
|
5127 |
clear_bit(KVM_REQ_UNHALT, &vcpu->requests); |
ac9f6dc0d
|
5128 5129 |
r = -EAGAIN; goto out; |
b6c7a5dcc
|
5130 |
} |
b6c7a5dcc
|
5131 |
/* re-sync apic's tpr */ |
eea1cff9a
|
5132 5133 5134 5135 5136 5137 |
if (!irqchip_in_kernel(vcpu->kvm)) { if (kvm_set_cr8(vcpu, kvm_run->cr8) != 0) { r = -EINVAL; goto out; } } |
b6c7a5dcc
|
5138 |
|
5287f194b
|
5139 5140 5141 |
r = complete_mmio(vcpu); if (r <= 0) goto out; |
851ba6922
|
5142 |
r = __vcpu_run(vcpu); |
b6c7a5dcc
|
5143 5144 |
out: |
f1d86e469
|
5145 |
post_kvm_run_save(vcpu); |
b6c7a5dcc
|
5146 5147 |
if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &sigsaved, NULL); |
b6c7a5dcc
|
5148 5149 5150 5151 5152 |
return r; } int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { |
7ae441eac
|
5153 5154 5155 5156 5157 5158 5159 5160 |
if (vcpu->arch.emulate_regs_need_sync_to_vcpu) { /* * We are here if userspace calls get_regs() in the middle of * instruction emulation. Registers state needs to be copied * back from emulation context to vcpu. Usrapace shouldn't do * that usually, but some bad designed PV devices (vmware * backdoor interface) need this to work */ |
9dac77fa4
|
5161 5162 |
struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; memcpy(vcpu->arch.regs, ctxt->regs, sizeof ctxt->regs); |
7ae441eac
|
5163 5164 |
vcpu->arch.emulate_regs_need_sync_to_vcpu = false; } |
5fdbf9765
|
5165 5166 5167 5168 5169 5170 5171 5172 |
regs->rax = kvm_register_read(vcpu, VCPU_REGS_RAX); regs->rbx = kvm_register_read(vcpu, VCPU_REGS_RBX); regs->rcx = kvm_register_read(vcpu, VCPU_REGS_RCX); regs->rdx = kvm_register_read(vcpu, VCPU_REGS_RDX); regs->rsi = kvm_register_read(vcpu, VCPU_REGS_RSI); regs->rdi = kvm_register_read(vcpu, VCPU_REGS_RDI); regs->rsp = kvm_register_read(vcpu, VCPU_REGS_RSP); regs->rbp = kvm_register_read(vcpu, VCPU_REGS_RBP); |
b6c7a5dcc
|
5173 |
#ifdef CONFIG_X86_64 |
5fdbf9765
|
5174 5175 5176 5177 5178 5179 5180 5181 |
regs->r8 = kvm_register_read(vcpu, VCPU_REGS_R8); regs->r9 = kvm_register_read(vcpu, VCPU_REGS_R9); regs->r10 = kvm_register_read(vcpu, VCPU_REGS_R10); regs->r11 = kvm_register_read(vcpu, VCPU_REGS_R11); regs->r12 = kvm_register_read(vcpu, VCPU_REGS_R12); regs->r13 = kvm_register_read(vcpu, VCPU_REGS_R13); regs->r14 = kvm_register_read(vcpu, VCPU_REGS_R14); regs->r15 = kvm_register_read(vcpu, VCPU_REGS_R15); |
b6c7a5dcc
|
5182 |
#endif |
5fdbf9765
|
5183 |
regs->rip = kvm_rip_read(vcpu); |
91586a3b7
|
5184 |
regs->rflags = kvm_get_rflags(vcpu); |
b6c7a5dcc
|
5185 |
|
b6c7a5dcc
|
5186 5187 5188 5189 5190 |
return 0; } int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { |
7ae441eac
|
5191 5192 |
vcpu->arch.emulate_regs_need_sync_from_vcpu = true; vcpu->arch.emulate_regs_need_sync_to_vcpu = false; |
5fdbf9765
|
5193 5194 5195 5196 5197 5198 5199 5200 |
kvm_register_write(vcpu, VCPU_REGS_RAX, regs->rax); kvm_register_write(vcpu, VCPU_REGS_RBX, regs->rbx); kvm_register_write(vcpu, VCPU_REGS_RCX, regs->rcx); kvm_register_write(vcpu, VCPU_REGS_RDX, regs->rdx); kvm_register_write(vcpu, VCPU_REGS_RSI, regs->rsi); kvm_register_write(vcpu, VCPU_REGS_RDI, regs->rdi); kvm_register_write(vcpu, VCPU_REGS_RSP, regs->rsp); kvm_register_write(vcpu, VCPU_REGS_RBP, regs->rbp); |
b6c7a5dcc
|
5201 |
#ifdef CONFIG_X86_64 |
5fdbf9765
|
5202 5203 5204 5205 5206 5207 5208 5209 |
kvm_register_write(vcpu, VCPU_REGS_R8, regs->r8); kvm_register_write(vcpu, VCPU_REGS_R9, regs->r9); kvm_register_write(vcpu, VCPU_REGS_R10, regs->r10); kvm_register_write(vcpu, VCPU_REGS_R11, regs->r11); kvm_register_write(vcpu, VCPU_REGS_R12, regs->r12); kvm_register_write(vcpu, VCPU_REGS_R13, regs->r13); kvm_register_write(vcpu, VCPU_REGS_R14, regs->r14); kvm_register_write(vcpu, VCPU_REGS_R15, regs->r15); |
b6c7a5dcc
|
5210 |
#endif |
5fdbf9765
|
5211 |
kvm_rip_write(vcpu, regs->rip); |
91586a3b7
|
5212 |
kvm_set_rflags(vcpu, regs->rflags); |
b6c7a5dcc
|
5213 |
|
b4f14abd9
|
5214 |
vcpu->arch.exception.pending = false; |
3842d135f
|
5215 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
b6c7a5dcc
|
5216 5217 |
return 0; } |
b6c7a5dcc
|
5218 5219 5220 |
void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) { struct kvm_segment cs; |
3e6e0aab1
|
5221 |
kvm_get_segment(vcpu, &cs, VCPU_SREG_CS); |
b6c7a5dcc
|
5222 5223 5224 5225 5226 5227 5228 5229 |
*db = cs.db; *l = cs.l; } EXPORT_SYMBOL_GPL(kvm_get_cs_db_l_bits); int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs) { |
89a27f4d0
|
5230 |
struct desc_ptr dt; |
b6c7a5dcc
|
5231 |
|
3e6e0aab1
|
5232 5233 5234 5235 5236 5237 |
kvm_get_segment(vcpu, &sregs->cs, VCPU_SREG_CS); kvm_get_segment(vcpu, &sregs->ds, VCPU_SREG_DS); kvm_get_segment(vcpu, &sregs->es, VCPU_SREG_ES); kvm_get_segment(vcpu, &sregs->fs, VCPU_SREG_FS); kvm_get_segment(vcpu, &sregs->gs, VCPU_SREG_GS); kvm_get_segment(vcpu, &sregs->ss, VCPU_SREG_SS); |
b6c7a5dcc
|
5238 |
|
3e6e0aab1
|
5239 5240 |
kvm_get_segment(vcpu, &sregs->tr, VCPU_SREG_TR); kvm_get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); |
b6c7a5dcc
|
5241 5242 |
kvm_x86_ops->get_idt(vcpu, &dt); |
89a27f4d0
|
5243 5244 |
sregs->idt.limit = dt.size; sregs->idt.base = dt.address; |
b6c7a5dcc
|
5245 |
kvm_x86_ops->get_gdt(vcpu, &dt); |
89a27f4d0
|
5246 5247 |
sregs->gdt.limit = dt.size; sregs->gdt.base = dt.address; |
b6c7a5dcc
|
5248 |
|
4d4ec0874
|
5249 |
sregs->cr0 = kvm_read_cr0(vcpu); |
ad312c7c7
|
5250 |
sregs->cr2 = vcpu->arch.cr2; |
9f8fe5043
|
5251 |
sregs->cr3 = kvm_read_cr3(vcpu); |
fc78f5193
|
5252 |
sregs->cr4 = kvm_read_cr4(vcpu); |
2d3ad1f40
|
5253 |
sregs->cr8 = kvm_get_cr8(vcpu); |
f6801dff2
|
5254 |
sregs->efer = vcpu->arch.efer; |
b6c7a5dcc
|
5255 |
sregs->apic_base = kvm_get_apic_base(vcpu); |
923c61bbc
|
5256 |
memset(sregs->interrupt_bitmap, 0, sizeof sregs->interrupt_bitmap); |
b6c7a5dcc
|
5257 |
|
36752c9b9
|
5258 |
if (vcpu->arch.interrupt.pending && !vcpu->arch.interrupt.soft) |
14d0bc1f7
|
5259 5260 |
set_bit(vcpu->arch.interrupt.nr, (unsigned long *)sregs->interrupt_bitmap); |
16d7a1911
|
5261 |
|
b6c7a5dcc
|
5262 5263 |
return 0; } |
62d9f0dbc
|
5264 5265 5266 |
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, struct kvm_mp_state *mp_state) { |
62d9f0dbc
|
5267 |
mp_state->mp_state = vcpu->arch.mp_state; |
62d9f0dbc
|
5268 5269 5270 5271 5272 5273 |
return 0; } int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, struct kvm_mp_state *mp_state) { |
62d9f0dbc
|
5274 |
vcpu->arch.mp_state = mp_state->mp_state; |
3842d135f
|
5275 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
62d9f0dbc
|
5276 5277 |
return 0; } |
e269fb218
|
5278 5279 |
int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason, bool has_error_code, u32 error_code) |
b6c7a5dcc
|
5280 |
{ |
9d74191ab
|
5281 |
struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; |
8ec4722dd
|
5282 |
int ret; |
e01c24261
|
5283 |
|
8ec4722dd
|
5284 |
init_emulate_ctxt(vcpu); |
c697518a8
|
5285 |
|
9d74191ab
|
5286 5287 |
ret = emulator_task_switch(ctxt, tss_selector, reason, has_error_code, error_code); |
c697518a8
|
5288 |
|
c697518a8
|
5289 |
if (ret) |
19d044372
|
5290 |
return EMULATE_FAIL; |
37817f298
|
5291 |
|
9dac77fa4
|
5292 |
memcpy(vcpu->arch.regs, ctxt->regs, sizeof ctxt->regs); |
9d74191ab
|
5293 5294 |
kvm_rip_write(vcpu, ctxt->eip); kvm_set_rflags(vcpu, ctxt->eflags); |
3842d135f
|
5295 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
19d044372
|
5296 |
return EMULATE_DONE; |
37817f298
|
5297 5298 |
} EXPORT_SYMBOL_GPL(kvm_task_switch); |
b6c7a5dcc
|
5299 5300 5301 5302 |
int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs) { int mmu_reset_needed = 0; |
63f42e023
|
5303 |
int pending_vec, max_bits, idx; |
89a27f4d0
|
5304 |
struct desc_ptr dt; |
b6c7a5dcc
|
5305 |
|
89a27f4d0
|
5306 5307 |
dt.size = sregs->idt.limit; dt.address = sregs->idt.base; |
b6c7a5dcc
|
5308 |
kvm_x86_ops->set_idt(vcpu, &dt); |
89a27f4d0
|
5309 5310 |
dt.size = sregs->gdt.limit; dt.address = sregs->gdt.base; |
b6c7a5dcc
|
5311 |
kvm_x86_ops->set_gdt(vcpu, &dt); |
ad312c7c7
|
5312 |
vcpu->arch.cr2 = sregs->cr2; |
9f8fe5043
|
5313 |
mmu_reset_needed |= kvm_read_cr3(vcpu) != sregs->cr3; |
dc7e795e3
|
5314 |
vcpu->arch.cr3 = sregs->cr3; |
aff48baa3
|
5315 |
__set_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail); |
b6c7a5dcc
|
5316 |
|
2d3ad1f40
|
5317 |
kvm_set_cr8(vcpu, sregs->cr8); |
b6c7a5dcc
|
5318 |
|
f6801dff2
|
5319 |
mmu_reset_needed |= vcpu->arch.efer != sregs->efer; |
b6c7a5dcc
|
5320 |
kvm_x86_ops->set_efer(vcpu, sregs->efer); |
b6c7a5dcc
|
5321 |
kvm_set_apic_base(vcpu, sregs->apic_base); |
4d4ec0874
|
5322 |
mmu_reset_needed |= kvm_read_cr0(vcpu) != sregs->cr0; |
b6c7a5dcc
|
5323 |
kvm_x86_ops->set_cr0(vcpu, sregs->cr0); |
d73061638
|
5324 |
vcpu->arch.cr0 = sregs->cr0; |
b6c7a5dcc
|
5325 |
|
fc78f5193
|
5326 |
mmu_reset_needed |= kvm_read_cr4(vcpu) != sregs->cr4; |
b6c7a5dcc
|
5327 |
kvm_x86_ops->set_cr4(vcpu, sregs->cr4); |
3ea3aa8cf
|
5328 |
if (sregs->cr4 & X86_CR4_OSXSAVE) |
00b27a3ef
|
5329 |
kvm_update_cpuid(vcpu); |
63f42e023
|
5330 5331 |
idx = srcu_read_lock(&vcpu->kvm->srcu); |
7c93be44a
|
5332 |
if (!is_long_mode(vcpu) && is_pae(vcpu)) { |
9f8fe5043
|
5333 |
load_pdptrs(vcpu, vcpu->arch.walk_mmu, kvm_read_cr3(vcpu)); |
7c93be44a
|
5334 5335 |
mmu_reset_needed = 1; } |
63f42e023
|
5336 |
srcu_read_unlock(&vcpu->kvm->srcu, idx); |
b6c7a5dcc
|
5337 5338 5339 |
if (mmu_reset_needed) kvm_mmu_reset_context(vcpu); |
923c61bbc
|
5340 5341 5342 5343 |
max_bits = (sizeof sregs->interrupt_bitmap) << 3; pending_vec = find_first_bit( (const unsigned long *)sregs->interrupt_bitmap, max_bits); if (pending_vec < max_bits) { |
66fd3f7f9
|
5344 |
kvm_queue_interrupt(vcpu, pending_vec, false); |
923c61bbc
|
5345 5346 |
pr_debug("Set back pending irq %d ", pending_vec); |
b6c7a5dcc
|
5347 |
} |
3e6e0aab1
|
5348 5349 5350 5351 5352 5353 |
kvm_set_segment(vcpu, &sregs->cs, VCPU_SREG_CS); kvm_set_segment(vcpu, &sregs->ds, VCPU_SREG_DS); kvm_set_segment(vcpu, &sregs->es, VCPU_SREG_ES); kvm_set_segment(vcpu, &sregs->fs, VCPU_SREG_FS); kvm_set_segment(vcpu, &sregs->gs, VCPU_SREG_GS); kvm_set_segment(vcpu, &sregs->ss, VCPU_SREG_SS); |
b6c7a5dcc
|
5354 |
|
3e6e0aab1
|
5355 5356 |
kvm_set_segment(vcpu, &sregs->tr, VCPU_SREG_TR); kvm_set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); |
b6c7a5dcc
|
5357 |
|
5f0269f5d
|
5358 |
update_cr8_intercept(vcpu); |
9c3e4aab5
|
5359 |
/* Older userspace won't unhalt the vcpu on reset. */ |
c5af89b68
|
5360 |
if (kvm_vcpu_is_bsp(vcpu) && kvm_rip_read(vcpu) == 0xfff0 && |
9c3e4aab5
|
5361 |
sregs->cs.selector == 0xf000 && sregs->cs.base == 0xffff0000 && |
3eeb3288b
|
5362 |
!is_protmode(vcpu)) |
9c3e4aab5
|
5363 |
vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; |
3842d135f
|
5364 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
b6c7a5dcc
|
5365 5366 |
return 0; } |
d0bfb940e
|
5367 5368 |
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) |
b6c7a5dcc
|
5369 |
{ |
355be0b93
|
5370 |
unsigned long rflags; |
ae675ef01
|
5371 |
int i, r; |
b6c7a5dcc
|
5372 |
|
4f926bf29
|
5373 5374 5375 |
if (dbg->control & (KVM_GUESTDBG_INJECT_DB | KVM_GUESTDBG_INJECT_BP)) { r = -EBUSY; if (vcpu->arch.exception.pending) |
2122ff5ea
|
5376 |
goto out; |
4f926bf29
|
5377 5378 5379 5380 5381 |
if (dbg->control & KVM_GUESTDBG_INJECT_DB) kvm_queue_exception(vcpu, DB_VECTOR); else kvm_queue_exception(vcpu, BP_VECTOR); } |
91586a3b7
|
5382 5383 5384 5385 5386 |
/* * Read rflags as long as potentially injected trace flags are still * filtered out. */ rflags = kvm_get_rflags(vcpu); |
355be0b93
|
5387 5388 5389 5390 5391 5392 |
vcpu->guest_debug = dbg->control; if (!(vcpu->guest_debug & KVM_GUESTDBG_ENABLE)) vcpu->guest_debug = 0; if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { |
ae675ef01
|
5393 5394 5395 5396 5397 5398 5399 5400 5401 |
for (i = 0; i < KVM_NR_DB_REGS; ++i) vcpu->arch.eff_db[i] = dbg->arch.debugreg[i]; vcpu->arch.switch_db_regs = (dbg->arch.debugreg[7] & DR7_BP_EN_MASK); } else { for (i = 0; i < KVM_NR_DB_REGS; i++) vcpu->arch.eff_db[i] = vcpu->arch.db[i]; vcpu->arch.switch_db_regs = (vcpu->arch.dr7 & DR7_BP_EN_MASK); } |
f92653eeb
|
5402 5403 5404 |
if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) vcpu->arch.singlestep_rip = kvm_rip_read(vcpu) + get_segment_base(vcpu, VCPU_SREG_CS); |
94fe45da4
|
5405 |
|
91586a3b7
|
5406 5407 5408 5409 5410 |
/* * Trigger an rflags update that will inject or remove the trace * flags. */ kvm_set_rflags(vcpu, rflags); |
b6c7a5dcc
|
5411 |
|
355be0b93
|
5412 |
kvm_x86_ops->set_guest_debug(vcpu, dbg); |
b6c7a5dcc
|
5413 |
|
4f926bf29
|
5414 |
r = 0; |
d0bfb940e
|
5415 |
|
2122ff5ea
|
5416 |
out: |
b6c7a5dcc
|
5417 5418 5419 5420 5421 |
return r; } /* |
8b0067913
|
5422 5423 5424 5425 5426 5427 5428 |
* Translate a guest virtual address to a guest physical address. */ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, struct kvm_translation *tr) { unsigned long vaddr = tr->linear_address; gpa_t gpa; |
f656ce018
|
5429 |
int idx; |
8b0067913
|
5430 |
|
f656ce018
|
5431 |
idx = srcu_read_lock(&vcpu->kvm->srcu); |
1871c6020
|
5432 |
gpa = kvm_mmu_gva_to_gpa_system(vcpu, vaddr, NULL); |
f656ce018
|
5433 |
srcu_read_unlock(&vcpu->kvm->srcu, idx); |
8b0067913
|
5434 5435 5436 5437 |
tr->physical_address = gpa; tr->valid = gpa != UNMAPPED_GVA; tr->writeable = 1; tr->usermode = 0; |
8b0067913
|
5438 5439 5440 |
return 0; } |
d07520607
|
5441 5442 |
int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) { |
98918833a
|
5443 5444 |
struct i387_fxsave_struct *fxsave = &vcpu->arch.guest_fpu.state->fxsave; |
d07520607
|
5445 |
|
d07520607
|
5446 5447 5448 5449 5450 5451 5452 5453 |
memcpy(fpu->fpr, fxsave->st_space, 128); fpu->fcw = fxsave->cwd; fpu->fsw = fxsave->swd; fpu->ftwx = fxsave->twd; fpu->last_opcode = fxsave->fop; fpu->last_ip = fxsave->rip; fpu->last_dp = fxsave->rdp; memcpy(fpu->xmm, fxsave->xmm_space, sizeof fxsave->xmm_space); |
d07520607
|
5454 5455 5456 5457 5458 |
return 0; } int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) { |
98918833a
|
5459 5460 |
struct i387_fxsave_struct *fxsave = &vcpu->arch.guest_fpu.state->fxsave; |
d07520607
|
5461 |
|
d07520607
|
5462 5463 5464 5465 5466 5467 5468 5469 |
memcpy(fxsave->st_space, fpu->fpr, 128); fxsave->cwd = fpu->fcw; fxsave->swd = fpu->fsw; fxsave->twd = fpu->ftwx; fxsave->fop = fpu->last_opcode; fxsave->rip = fpu->last_ip; fxsave->rdp = fpu->last_dp; memcpy(fxsave->xmm_space, fpu->xmm, sizeof fxsave->xmm_space); |
d07520607
|
5470 5471 |
return 0; } |
10ab25cd6
|
5472 |
int fx_init(struct kvm_vcpu *vcpu) |
d07520607
|
5473 |
{ |
10ab25cd6
|
5474 5475 5476 5477 5478 |
int err; err = fpu_alloc(&vcpu->arch.guest_fpu); if (err) return err; |
98918833a
|
5479 |
fpu_finit(&vcpu->arch.guest_fpu); |
d07520607
|
5480 |
|
2acf923e3
|
5481 5482 5483 5484 |
/* * Ensure guest xcr0 is valid for loading */ vcpu->arch.xcr0 = XSTATE_FP; |
ad312c7c7
|
5485 |
vcpu->arch.cr0 |= X86_CR0_ET; |
10ab25cd6
|
5486 5487 |
return 0; |
d07520607
|
5488 5489 |
} EXPORT_SYMBOL_GPL(fx_init); |
98918833a
|
5490 5491 5492 5493 |
static void fx_free(struct kvm_vcpu *vcpu) { fpu_free(&vcpu->arch.guest_fpu); } |
d07520607
|
5494 5495 |
void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) { |
2608d7a12
|
5496 |
if (vcpu->guest_fpu_loaded) |
d07520607
|
5497 |
return; |
2acf923e3
|
5498 5499 5500 5501 5502 5503 |
/* * Restore all possible states in the guest, * and assume host would use all available bits. * Guest xcr0 would be loaded later. */ kvm_put_guest_xcr0(vcpu); |
d07520607
|
5504 |
vcpu->guest_fpu_loaded = 1; |
7cf30855e
|
5505 |
unlazy_fpu(current); |
98918833a
|
5506 |
fpu_restore_checking(&vcpu->arch.guest_fpu); |
0c04851c0
|
5507 |
trace_kvm_fpu(1); |
d07520607
|
5508 |
} |
d07520607
|
5509 5510 5511 |
void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) { |
2acf923e3
|
5512 |
kvm_put_guest_xcr0(vcpu); |
d07520607
|
5513 5514 5515 5516 |
if (!vcpu->guest_fpu_loaded) return; vcpu->guest_fpu_loaded = 0; |
98918833a
|
5517 |
fpu_save_init(&vcpu->arch.guest_fpu); |
f096ed858
|
5518 |
++vcpu->stat.fpu_reload; |
a8eeb04a4
|
5519 |
kvm_make_request(KVM_REQ_DEACTIVATE_FPU, vcpu); |
0c04851c0
|
5520 |
trace_kvm_fpu(0); |
d07520607
|
5521 |
} |
e9b11c175
|
5522 5523 5524 |
void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) { |
12f9a48f7
|
5525 |
kvmclock_reset(vcpu); |
7f1ea2089
|
5526 |
|
f5f48ee15
|
5527 |
free_cpumask_var(vcpu->arch.wbinvd_dirty_mask); |
98918833a
|
5528 |
fx_free(vcpu); |
e9b11c175
|
5529 5530 5531 5532 5533 5534 |
kvm_x86_ops->vcpu_free(vcpu); } struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) { |
6755bae8e
|
5535 5536 5537 5538 5539 |
if (check_tsc_unstable() && atomic_read(&kvm->online_vcpus) != 0) printk_once(KERN_WARNING "kvm: SMP vm created on host with unstable TSC; " "guest TSC will not be reliable "); |
26e5215fd
|
5540 5541 |
return kvm_x86_ops->vcpu_create(kvm, id); } |
e9b11c175
|
5542 |
|
26e5215fd
|
5543 5544 5545 |
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) { int r; |
e9b11c175
|
5546 |
|
0bed3b568
|
5547 |
vcpu->arch.mtrr_state.have_fixed = 1; |
e9b11c175
|
5548 5549 5550 5551 5552 |
vcpu_load(vcpu); r = kvm_arch_vcpu_reset(vcpu); if (r == 0) r = kvm_mmu_setup(vcpu); vcpu_put(vcpu); |
e9b11c175
|
5553 |
|
26e5215fd
|
5554 |
return r; |
e9b11c175
|
5555 |
} |
d40ccc624
|
5556 |
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) |
e9b11c175
|
5557 |
{ |
344d9588a
|
5558 |
vcpu->arch.apf.msr_val = 0; |
e9b11c175
|
5559 5560 5561 |
vcpu_load(vcpu); kvm_mmu_unload(vcpu); vcpu_put(vcpu); |
98918833a
|
5562 |
fx_free(vcpu); |
e9b11c175
|
5563 5564 5565 5566 5567 |
kvm_x86_ops->vcpu_free(vcpu); } int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) { |
7460fb4a3
|
5568 5569 |
atomic_set(&vcpu->arch.nmi_queued, 0); vcpu->arch.nmi_pending = 0; |
448fa4a9c
|
5570 |
vcpu->arch.nmi_injected = false; |
42dbaa5a0
|
5571 5572 5573 5574 |
vcpu->arch.switch_db_regs = 0; memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db)); vcpu->arch.dr6 = DR6_FIXED_1; vcpu->arch.dr7 = DR7_FIXED_1; |
3842d135f
|
5575 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
344d9588a
|
5576 |
vcpu->arch.apf.msr_val = 0; |
c9aaa8957
|
5577 |
vcpu->arch.st.msr_val = 0; |
3842d135f
|
5578 |
|
12f9a48f7
|
5579 |
kvmclock_reset(vcpu); |
af585b921
|
5580 5581 5582 |
kvm_clear_async_pf_completion_queue(vcpu); kvm_async_pf_hash_reset(vcpu); vcpu->arch.apf.halted = false; |
3842d135f
|
5583 |
|
f5132b013
|
5584 |
kvm_pmu_reset(vcpu); |
e9b11c175
|
5585 5586 |
return kvm_x86_ops->vcpu_reset(vcpu); } |
10474ae89
|
5587 |
int kvm_arch_hardware_enable(void *garbage) |
e9b11c175
|
5588 |
{ |
ca84d1a24
|
5589 5590 5591 |
struct kvm *kvm; struct kvm_vcpu *vcpu; int i; |
18863bdd6
|
5592 5593 |
kvm_shared_msr_cpu_online(); |
ca84d1a24
|
5594 5595 5596 |
list_for_each_entry(kvm, &vm_list, vm_list) kvm_for_each_vcpu(i, vcpu, kvm) if (vcpu->cpu == smp_processor_id()) |
c285545f8
|
5597 |
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); |
10474ae89
|
5598 |
return kvm_x86_ops->hardware_enable(garbage); |
e9b11c175
|
5599 5600 5601 5602 5603 |
} void kvm_arch_hardware_disable(void *garbage) { kvm_x86_ops->hardware_disable(garbage); |
3548bab50
|
5604 |
drop_user_return_notifiers(garbage); |
e9b11c175
|
5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 |
} int kvm_arch_hardware_setup(void) { return kvm_x86_ops->hardware_setup(); } void kvm_arch_hardware_unsetup(void) { kvm_x86_ops->hardware_unsetup(); } void kvm_arch_check_processor_compat(void *rtn) { kvm_x86_ops->check_processor_compatibility(rtn); } int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) { struct page *page; struct kvm *kvm; int r; BUG_ON(vcpu->kvm == NULL); kvm = vcpu->kvm; |
9aabc88fc
|
5630 |
vcpu->arch.emulate_ctxt.ops = &emulate_ops; |
c5af89b68
|
5631 |
if (!irqchip_in_kernel(kvm) || kvm_vcpu_is_bsp(vcpu)) |
a45352908
|
5632 |
vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; |
e9b11c175
|
5633 |
else |
a45352908
|
5634 |
vcpu->arch.mp_state = KVM_MP_STATE_UNINITIALIZED; |
e9b11c175
|
5635 5636 5637 5638 5639 5640 |
page = alloc_page(GFP_KERNEL | __GFP_ZERO); if (!page) { r = -ENOMEM; goto fail; } |
ad312c7c7
|
5641 |
vcpu->arch.pio_data = page_address(page); |
e9b11c175
|
5642 |
|
1e993611d
|
5643 |
kvm_init_tsc_catchup(vcpu, max_tsc_khz); |
c285545f8
|
5644 |
|
e9b11c175
|
5645 5646 5647 5648 5649 5650 5651 5652 5653 |
r = kvm_mmu_create(vcpu); if (r < 0) goto fail_free_pio_data; if (irqchip_in_kernel(kvm)) { r = kvm_create_lapic(vcpu); if (r < 0) goto fail_mmu_destroy; } |
890ca9aef
|
5654 5655 5656 5657 |
vcpu->arch.mce_banks = kzalloc(KVM_MAX_MCE_BANKS * sizeof(u64) * 4, GFP_KERNEL); if (!vcpu->arch.mce_banks) { r = -ENOMEM; |
443c39bc9
|
5658 |
goto fail_free_lapic; |
890ca9aef
|
5659 5660 |
} vcpu->arch.mcg_cap = KVM_MAX_MCE_BANKS; |
f5f48ee15
|
5661 5662 |
if (!zalloc_cpumask_var(&vcpu->arch.wbinvd_dirty_mask, GFP_KERNEL)) goto fail_free_mce_banks; |
af585b921
|
5663 |
kvm_async_pf_hash_reset(vcpu); |
f5132b013
|
5664 |
kvm_pmu_init(vcpu); |
af585b921
|
5665 |
|
e9b11c175
|
5666 |
return 0; |
f5f48ee15
|
5667 5668 |
fail_free_mce_banks: kfree(vcpu->arch.mce_banks); |
443c39bc9
|
5669 5670 |
fail_free_lapic: kvm_free_lapic(vcpu); |
e9b11c175
|
5671 5672 5673 |
fail_mmu_destroy: kvm_mmu_destroy(vcpu); fail_free_pio_data: |
ad312c7c7
|
5674 |
free_page((unsigned long)vcpu->arch.pio_data); |
e9b11c175
|
5675 5676 5677 5678 5679 5680 |
fail: return r; } void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) { |
f656ce018
|
5681 |
int idx; |
f5132b013
|
5682 |
kvm_pmu_destroy(vcpu); |
36cb93fd6
|
5683 |
kfree(vcpu->arch.mce_banks); |
e9b11c175
|
5684 |
kvm_free_lapic(vcpu); |
f656ce018
|
5685 |
idx = srcu_read_lock(&vcpu->kvm->srcu); |
e9b11c175
|
5686 |
kvm_mmu_destroy(vcpu); |
f656ce018
|
5687 |
srcu_read_unlock(&vcpu->kvm->srcu, idx); |
ad312c7c7
|
5688 |
free_page((unsigned long)vcpu->arch.pio_data); |
e9b11c175
|
5689 |
} |
d19a9cd27
|
5690 |
|
d89f5eff7
|
5691 |
int kvm_arch_init_vm(struct kvm *kvm) |
d19a9cd27
|
5692 |
{ |
f05e70ac0
|
5693 |
INIT_LIST_HEAD(&kvm->arch.active_mmu_pages); |
4d5c5d0fe
|
5694 |
INIT_LIST_HEAD(&kvm->arch.assigned_dev_head); |
d19a9cd27
|
5695 |
|
5550af4df
|
5696 5697 |
/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */ set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap); |
038f8c110
|
5698 |
raw_spin_lock_init(&kvm->arch.tsc_write_lock); |
53f658b3c
|
5699 |
|
d89f5eff7
|
5700 |
return 0; |
d19a9cd27
|
5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 |
} static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu) { vcpu_load(vcpu); kvm_mmu_unload(vcpu); vcpu_put(vcpu); } static void kvm_free_vcpus(struct kvm *kvm) { unsigned int i; |
988a2cae6
|
5713 |
struct kvm_vcpu *vcpu; |
d19a9cd27
|
5714 5715 5716 5717 |
/* * Unpin any mmu pages first. */ |
af585b921
|
5718 5719 |
kvm_for_each_vcpu(i, vcpu, kvm) { kvm_clear_async_pf_completion_queue(vcpu); |
988a2cae6
|
5720 |
kvm_unload_vcpu_mmu(vcpu); |
af585b921
|
5721 |
} |
988a2cae6
|
5722 5723 5724 5725 5726 5727 |
kvm_for_each_vcpu(i, vcpu, kvm) kvm_arch_vcpu_free(vcpu); mutex_lock(&kvm->lock); for (i = 0; i < atomic_read(&kvm->online_vcpus); i++) kvm->vcpus[i] = NULL; |
d19a9cd27
|
5728 |
|
988a2cae6
|
5729 5730 |
atomic_set(&kvm->online_vcpus, 0); mutex_unlock(&kvm->lock); |
d19a9cd27
|
5731 |
} |
ad8ba2cd4
|
5732 5733 |
void kvm_arch_sync_events(struct kvm *kvm) { |
ba4cef31d
|
5734 |
kvm_free_all_assigned_devices(kvm); |
aea924f60
|
5735 |
kvm_free_pit(kvm); |
ad8ba2cd4
|
5736 |
} |
d19a9cd27
|
5737 5738 |
void kvm_arch_destroy_vm(struct kvm *kvm) { |
6eb55818c
|
5739 |
kvm_iommu_unmap_guest(kvm); |
d7deeeb02
|
5740 5741 |
kfree(kvm->arch.vpic); kfree(kvm->arch.vioapic); |
d19a9cd27
|
5742 |
kvm_free_vcpus(kvm); |
3d45830c2
|
5743 5744 |
if (kvm->arch.apic_access_page) put_page(kvm->arch.apic_access_page); |
b7ebfb050
|
5745 5746 |
if (kvm->arch.ept_identity_pagetable) put_page(kvm->arch.ept_identity_pagetable); |
d19a9cd27
|
5747 |
} |
0de10343b
|
5748 |
|
f7784b8ec
|
5749 5750 |
int kvm_arch_prepare_memory_region(struct kvm *kvm, struct kvm_memory_slot *memslot, |
0de10343b
|
5751 |
struct kvm_memory_slot old, |
f7784b8ec
|
5752 |
struct kvm_userspace_memory_region *mem, |
0de10343b
|
5753 5754 |
int user_alloc) { |
f7784b8ec
|
5755 |
int npages = memslot->npages; |
7ac77099c
|
5756 5757 5758 5759 5760 |
int map_flags = MAP_PRIVATE | MAP_ANONYMOUS; /* Prevent internal slot pages from being moved by fork()/COW. */ if (memslot->id >= KVM_MEMORY_SLOTS) map_flags = MAP_SHARED | MAP_ANONYMOUS; |
0de10343b
|
5761 5762 5763 5764 5765 5766 |
/*To keep backward compatibility with older userspace, *x86 needs to hanlde !user_alloc case. */ if (!user_alloc) { if (npages && !old.rmap) { |
604b38ac0
|
5767 |
unsigned long userspace_addr; |
72dc67a69
|
5768 |
down_write(¤t->mm->mmap_sem); |
604b38ac0
|
5769 5770 5771 |
userspace_addr = do_mmap(NULL, 0, npages * PAGE_SIZE, PROT_READ | PROT_WRITE, |
7ac77099c
|
5772 |
map_flags, |
604b38ac0
|
5773 |
0); |
72dc67a69
|
5774 |
up_write(¤t->mm->mmap_sem); |
0de10343b
|
5775 |
|
604b38ac0
|
5776 5777 |
if (IS_ERR((void *)userspace_addr)) return PTR_ERR((void *)userspace_addr); |
604b38ac0
|
5778 |
memslot->userspace_addr = userspace_addr; |
0de10343b
|
5779 5780 |
} } |
f7784b8ec
|
5781 5782 5783 5784 5785 5786 5787 5788 5789 |
return 0; } void kvm_arch_commit_memory_region(struct kvm *kvm, struct kvm_userspace_memory_region *mem, struct kvm_memory_slot old, int user_alloc) { |
48c0e4e90
|
5790 |
int nr_mmu_pages = 0, npages = mem->memory_size >> PAGE_SHIFT; |
f7784b8ec
|
5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 |
if (!user_alloc && !old.user_alloc && old.rmap && !npages) { int ret; down_write(¤t->mm->mmap_sem); ret = do_munmap(current->mm, old.userspace_addr, old.npages * PAGE_SIZE); up_write(¤t->mm->mmap_sem); if (ret < 0) printk(KERN_WARNING "kvm_vm_ioctl_set_memory_region: " "failed to munmap memory "); } |
48c0e4e90
|
5805 5806 |
if (!kvm->arch.n_requested_mmu_pages) nr_mmu_pages = kvm_mmu_calculate_mmu_pages(kvm); |
7c8a83b75
|
5807 |
spin_lock(&kvm->mmu_lock); |
48c0e4e90
|
5808 |
if (nr_mmu_pages) |
0de10343b
|
5809 |
kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages); |
0de10343b
|
5810 |
kvm_mmu_slot_remove_write_access(kvm, mem->slot); |
7c8a83b75
|
5811 |
spin_unlock(&kvm->mmu_lock); |
0de10343b
|
5812 |
} |
1d737c8a6
|
5813 |
|
34d4cb8fc
|
5814 5815 5816 |
void kvm_arch_flush_shadow(struct kvm *kvm) { kvm_mmu_zap_all(kvm); |
8986ecc0e
|
5817 |
kvm_reload_remote_mmus(kvm); |
34d4cb8fc
|
5818 |
} |
1d737c8a6
|
5819 5820 |
int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { |
af585b921
|
5821 5822 5823 |
return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && !vcpu->arch.apf.halted) || !list_empty_careful(&vcpu->async_pf.done) |
a1b37100d
|
5824 |
|| vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED |
7460fb4a3
|
5825 |
|| atomic_read(&vcpu->arch.nmi_queued) || |
a1b37100d
|
5826 5827 |
(kvm_arch_interrupt_allowed(vcpu) && kvm_cpu_has_interrupt(vcpu)); |
1d737c8a6
|
5828 |
} |
5736199af
|
5829 |
|
5736199af
|
5830 5831 |
void kvm_vcpu_kick(struct kvm_vcpu *vcpu) { |
32f884006
|
5832 5833 |
int me; int cpu = vcpu->cpu; |
5736199af
|
5834 5835 5836 5837 5838 |
if (waitqueue_active(&vcpu->wq)) { wake_up_interruptible(&vcpu->wq); ++vcpu->stat.halt_wakeup; } |
32f884006
|
5839 5840 5841 |
me = get_cpu(); if (cpu != me && (unsigned)cpu < nr_cpu_ids && cpu_online(cpu)) |
6b7e2d099
|
5842 |
if (kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE) |
32f884006
|
5843 |
smp_send_reschedule(cpu); |
e9571ed54
|
5844 |
put_cpu(); |
5736199af
|
5845 |
} |
78646121e
|
5846 5847 5848 5849 5850 |
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu) { return kvm_x86_ops->interrupt_allowed(vcpu); } |
229456fc3
|
5851 |
|
f92653eeb
|
5852 5853 5854 5855 5856 5857 5858 5859 |
bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip) { unsigned long current_rip = kvm_rip_read(vcpu) + get_segment_base(vcpu, VCPU_SREG_CS); return current_rip == linear_rip; } EXPORT_SYMBOL_GPL(kvm_is_linear_rip); |
94fe45da4
|
5860 5861 5862 5863 5864 5865 |
unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu) { unsigned long rflags; rflags = kvm_x86_ops->get_rflags(vcpu); if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) |
c310bac5a
|
5866 |
rflags &= ~X86_EFLAGS_TF; |
94fe45da4
|
5867 5868 5869 5870 5871 5872 5873 |
return rflags; } EXPORT_SYMBOL_GPL(kvm_get_rflags); void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) { if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP && |
f92653eeb
|
5874 |
kvm_is_linear_rip(vcpu, vcpu->arch.singlestep_rip)) |
c310bac5a
|
5875 |
rflags |= X86_EFLAGS_TF; |
94fe45da4
|
5876 |
kvm_x86_ops->set_rflags(vcpu, rflags); |
3842d135f
|
5877 |
kvm_make_request(KVM_REQ_EVENT, vcpu); |
94fe45da4
|
5878 5879 |
} EXPORT_SYMBOL_GPL(kvm_set_rflags); |
56028d086
|
5880 5881 5882 |
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { int r; |
fb67e14fc
|
5883 |
if ((vcpu->arch.mmu.direct_map != work->arch.direct_map) || |
c4806acdc
|
5884 |
is_error_page(work->page)) |
56028d086
|
5885 5886 5887 5888 5889 |
return; r = kvm_mmu_reload(vcpu); if (unlikely(r)) return; |
fb67e14fc
|
5890 5891 5892 |
if (!vcpu->arch.mmu.direct_map && work->arch.cr3 != vcpu->arch.mmu.get_cr3(vcpu)) return; |
56028d086
|
5893 5894 |
vcpu->arch.mmu.page_fault(vcpu, work->gva, 0, true); } |
af585b921
|
5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 |
static inline u32 kvm_async_pf_hash_fn(gfn_t gfn) { return hash_32(gfn & 0xffffffff, order_base_2(ASYNC_PF_PER_VCPU)); } static inline u32 kvm_async_pf_next_probe(u32 key) { return (key + 1) & (roundup_pow_of_two(ASYNC_PF_PER_VCPU) - 1); } static void kvm_add_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) { u32 key = kvm_async_pf_hash_fn(gfn); while (vcpu->arch.apf.gfns[key] != ~0) key = kvm_async_pf_next_probe(key); vcpu->arch.apf.gfns[key] = gfn; } static u32 kvm_async_pf_gfn_slot(struct kvm_vcpu *vcpu, gfn_t gfn) { int i; u32 key = kvm_async_pf_hash_fn(gfn); for (i = 0; i < roundup_pow_of_two(ASYNC_PF_PER_VCPU) && |
c7d28c240
|
5921 5922 |
(vcpu->arch.apf.gfns[key] != gfn && vcpu->arch.apf.gfns[key] != ~0); i++) |
af585b921
|
5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 |
key = kvm_async_pf_next_probe(key); return key; } bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) { return vcpu->arch.apf.gfns[kvm_async_pf_gfn_slot(vcpu, gfn)] == gfn; } static void kvm_del_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) { u32 i, j, k; i = j = kvm_async_pf_gfn_slot(vcpu, gfn); while (true) { vcpu->arch.apf.gfns[i] = ~0; do { j = kvm_async_pf_next_probe(j); if (vcpu->arch.apf.gfns[j] == ~0) return; k = kvm_async_pf_hash_fn(vcpu->arch.apf.gfns[j]); /* * k lies cyclically in ]i,j] * | i.k.j | * |....j i.k.| or |.k..j i...| */ } while ((i <= j) ? (i < k && k <= j) : (i < k || k <= j)); vcpu->arch.apf.gfns[i] = vcpu->arch.apf.gfns[j]; i = j; } } |
7c90705bf
|
5955 5956 5957 5958 5959 5960 |
static int apf_put_user(struct kvm_vcpu *vcpu, u32 val) { return kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apf.data, &val, sizeof(val)); } |
af585b921
|
5961 5962 5963 |
void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { |
6389ee946
|
5964 |
struct x86_exception fault; |
7c90705bf
|
5965 |
trace_kvm_async_pf_not_present(work->arch.token, work->gva); |
af585b921
|
5966 |
kvm_add_async_pf_gfn(vcpu, work->arch.gfn); |
7c90705bf
|
5967 5968 |
if (!(vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED) || |
fc5f06fac
|
5969 5970 |
(vcpu->arch.apf.send_user_only && kvm_x86_ops->get_cpl(vcpu) == 0)) |
7c90705bf
|
5971 5972 |
kvm_make_request(KVM_REQ_APF_HALT, vcpu); else if (!apf_put_user(vcpu, KVM_PV_REASON_PAGE_NOT_PRESENT)) { |
6389ee946
|
5973 5974 5975 5976 5977 5978 |
fault.vector = PF_VECTOR; fault.error_code_valid = true; fault.error_code = 0; fault.nested_page_fault = false; fault.address = work->arch.token; kvm_inject_page_fault(vcpu, &fault); |
7c90705bf
|
5979 |
} |
af585b921
|
5980 5981 5982 5983 5984 |
} void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { |
6389ee946
|
5985 |
struct x86_exception fault; |
7c90705bf
|
5986 5987 5988 5989 5990 5991 5992 5993 |
trace_kvm_async_pf_ready(work->arch.token, work->gva); if (is_error_page(work->page)) work->arch.token = ~0; /* broadcast wakeup */ else kvm_del_async_pf_gfn(vcpu, work->arch.gfn); if ((vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED) && !apf_put_user(vcpu, KVM_PV_REASON_PAGE_READY)) { |
6389ee946
|
5994 5995 5996 5997 5998 5999 |
fault.vector = PF_VECTOR; fault.error_code_valid = true; fault.error_code = 0; fault.nested_page_fault = false; fault.address = work->arch.token; kvm_inject_page_fault(vcpu, &fault); |
7c90705bf
|
6000 |
} |
e6d53e3b0
|
6001 |
vcpu->arch.apf.halted = false; |
7c90705bf
|
6002 6003 6004 6005 6006 6007 6008 6009 6010 |
} bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu) { if (!(vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED)) return true; else return !kvm_event_needs_reinjection(vcpu) && kvm_x86_ops->interrupt_allowed(vcpu); |
af585b921
|
6011 |
} |
229456fc3
|
6012 6013 6014 6015 6016 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_inj_virq); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_page_fault); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_msr); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_cr); |
0ac406de8
|
6017 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmrun); |
d8cabddf7
|
6018 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmexit); |
17897f366
|
6019 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmexit_inject); |
236649de3
|
6020 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intr_vmexit); |
ec1ff7908
|
6021 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga); |
532a46b98
|
6022 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit); |
2e554e8d6
|
6023 |
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts); |