kvm_host.h 46.2 KB
Newer Older
1
/*
2 3 4 5 6 7 8 9 10
 * Kernel-based Virtual Machine driver for Linux
 *
 * This header defines architecture specific interfaces, x86 version
 *
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 *
 */

H
H. Peter Anvin 已提交
11 12
#ifndef _ASM_X86_KVM_HOST_H
#define _ASM_X86_KVM_HOST_H
13

14 15
#include <linux/types.h>
#include <linux/mm.h>
16
#include <linux/mmu_notifier.h>
17
#include <linux/tracepoint.h>
18
#include <linux/cpumask.h>
19
#include <linux/irq_work.h>
20
#include <linux/irq.h>
21 22 23

#include <linux/kvm.h>
#include <linux/kvm_para.h>
24
#include <linux/kvm_types.h>
25
#include <linux/perf_event.h>
26 27
#include <linux/pvclock_gtod.h>
#include <linux/clocksource.h>
F
Feng Wu 已提交
28
#include <linux/irqbypass.h>
29
#include <linux/hyperv.h>
30

31
#include <asm/apic.h>
32
#include <asm/pvclock-abi.h>
33
#include <asm/desc.h>
S
Sheng Yang 已提交
34
#include <asm/mtrr.h>
35
#include <asm/msr-index.h>
36
#include <asm/asm.h>
37
#include <asm/kvm_page_track.h>
38
#include <asm/kvm_vcpu_regs.h>
39
#include <asm/hyperv-tlfs.h>
40

41
#define KVM_MAX_VCPUS 288
42
#define KVM_SOFT_MAX_VCPUS 240
43
#define KVM_MAX_VCPU_ID 1023
44
#define KVM_USER_MEM_SLOTS 509
45 46
/* memory slots that are not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 3
47
#define KVM_MEM_SLOTS_NUM (KVM_USER_MEM_SLOTS + KVM_PRIVATE_MEM_SLOTS)
48

49
#define KVM_HALT_POLL_NS_DEFAULT 200000
50

51 52
#define KVM_IRQCHIP_NUM_PINS  KVM_IOAPIC_NUM_PINS

53
/* x86-specific vcpu->requests bit members */
54 55 56 57 58
#define KVM_REQ_MIGRATE_TIMER		KVM_ARCH_REQ(0)
#define KVM_REQ_REPORT_TPR_ACCESS	KVM_ARCH_REQ(1)
#define KVM_REQ_TRIPLE_FAULT		KVM_ARCH_REQ(2)
#define KVM_REQ_MMU_SYNC		KVM_ARCH_REQ(3)
#define KVM_REQ_CLOCK_UPDATE		KVM_ARCH_REQ(4)
59
#define KVM_REQ_LOAD_CR3		KVM_ARCH_REQ(5)
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
#define KVM_REQ_EVENT			KVM_ARCH_REQ(6)
#define KVM_REQ_APF_HALT		KVM_ARCH_REQ(7)
#define KVM_REQ_STEAL_UPDATE		KVM_ARCH_REQ(8)
#define KVM_REQ_NMI			KVM_ARCH_REQ(9)
#define KVM_REQ_PMU			KVM_ARCH_REQ(10)
#define KVM_REQ_PMI			KVM_ARCH_REQ(11)
#define KVM_REQ_SMI			KVM_ARCH_REQ(12)
#define KVM_REQ_MASTERCLOCK_UPDATE	KVM_ARCH_REQ(13)
#define KVM_REQ_MCLOCK_INPROGRESS \
	KVM_ARCH_REQ_FLAGS(14, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
#define KVM_REQ_SCAN_IOAPIC \
	KVM_ARCH_REQ_FLAGS(15, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
#define KVM_REQ_GLOBAL_CLOCK_UPDATE	KVM_ARCH_REQ(16)
#define KVM_REQ_APIC_PAGE_RELOAD \
	KVM_ARCH_REQ_FLAGS(17, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
#define KVM_REQ_HV_CRASH		KVM_ARCH_REQ(18)
#define KVM_REQ_IOAPIC_EOI_EXIT		KVM_ARCH_REQ(19)
#define KVM_REQ_HV_RESET		KVM_ARCH_REQ(20)
#define KVM_REQ_HV_EXIT			KVM_ARCH_REQ(21)
#define KVM_REQ_HV_STIMER		KVM_ARCH_REQ(22)
80
#define KVM_REQ_LOAD_EOI_EXITMAP	KVM_ARCH_REQ(23)
81
#define KVM_REQ_GET_VMCS12_PAGES	KVM_ARCH_REQ(24)
82

83 84 85 86 87 88 89 90
#define CR0_RESERVED_BITS                                               \
	(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
			  | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \
			  | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG))

#define CR4_RESERVED_BITS                                               \
	(~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\
			  | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE     \
91
			  | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR | X86_CR4_PCIDE \
92
			  | X86_CR4_OSXSAVE | X86_CR4_SMEP | X86_CR4_FSGSBASE \
93
			  | X86_CR4_OSXMMEXCPT | X86_CR4_LA57 | X86_CR4_VMXE \
P
Paolo Bonzini 已提交
94
			  | X86_CR4_SMAP | X86_CR4_PKE | X86_CR4_UMIP))
95 96 97 98

#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR)


99 100

#define INVALID_PAGE (~(hpa_t)0)
101 102
#define VALID_PAGE(x) ((x) != INVALID_PAGE)

103 104
#define UNMAPPED_GVA (~(gpa_t)0)

105
/* KVM Hugepage definitions for x86 */
106 107 108 109 110 111 112 113 114
enum {
	PT_PAGE_TABLE_LEVEL   = 1,
	PT_DIRECTORY_LEVEL    = 2,
	PT_PDPE_LEVEL         = 3,
	/* set max level to the biggest one */
	PT_MAX_HUGEPAGE_LEVEL = PT_PDPE_LEVEL,
};
#define KVM_NR_PAGE_SIZES	(PT_MAX_HUGEPAGE_LEVEL - \
				 PT_PAGE_TABLE_LEVEL + 1)
115 116
#define KVM_HPAGE_GFN_SHIFT(x)	(((x) - 1) * 9)
#define KVM_HPAGE_SHIFT(x)	(PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x))
117 118 119
#define KVM_HPAGE_SIZE(x)	(1UL << KVM_HPAGE_SHIFT(x))
#define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
#define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
M
Marcelo Tosatti 已提交
120

121 122 123 124 125 126 127
static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
{
	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
}

128
#define KVM_PERMILLE_MMU_PAGES 20
129
#define KVM_MIN_ALLOC_MMU_PAGES 64UL
130
#define KVM_MMU_HASH_SHIFT 12
131
#define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT)
132 133
#define KVM_MIN_FREE_MMU_PAGES 5
#define KVM_REFILL_PAGES 25
134
#define KVM_MAX_CPUID_ENTRIES 80
S
Sheng Yang 已提交
135
#define KVM_NR_FIXED_MTRR_REGION 88
136
#define KVM_NR_VAR_MTRR 8
137

138 139
#define ASYNC_PF_PER_VCPU 64

140
enum kvm_reg {
141 142 143 144 145 146 147 148
	VCPU_REGS_RAX = __VCPU_REGS_RAX,
	VCPU_REGS_RCX = __VCPU_REGS_RCX,
	VCPU_REGS_RDX = __VCPU_REGS_RDX,
	VCPU_REGS_RBX = __VCPU_REGS_RBX,
	VCPU_REGS_RSP = __VCPU_REGS_RSP,
	VCPU_REGS_RBP = __VCPU_REGS_RBP,
	VCPU_REGS_RSI = __VCPU_REGS_RSI,
	VCPU_REGS_RDI = __VCPU_REGS_RDI,
149
#ifdef CONFIG_X86_64
150 151 152 153 154 155 156 157
	VCPU_REGS_R8  = __VCPU_REGS_R8,
	VCPU_REGS_R9  = __VCPU_REGS_R9,
	VCPU_REGS_R10 = __VCPU_REGS_R10,
	VCPU_REGS_R11 = __VCPU_REGS_R11,
	VCPU_REGS_R12 = __VCPU_REGS_R12,
	VCPU_REGS_R13 = __VCPU_REGS_R13,
	VCPU_REGS_R14 = __VCPU_REGS_R14,
	VCPU_REGS_R15 = __VCPU_REGS_R15,
158
#endif
159
	VCPU_REGS_RIP,
160 161 162
	NR_VCPU_REGS
};

A
Avi Kivity 已提交
163 164
enum kvm_reg_ex {
	VCPU_EXREG_PDPTR = NR_VCPU_REGS,
165
	VCPU_EXREG_CR3,
A
Avi Kivity 已提交
166
	VCPU_EXREG_RFLAGS,
A
Avi Kivity 已提交
167
	VCPU_EXREG_SEGMENTS,
A
Avi Kivity 已提交
168 169
};

170
enum {
171
	VCPU_SREG_ES,
172
	VCPU_SREG_CS,
173
	VCPU_SREG_SS,
174 175 176 177 178 179 180
	VCPU_SREG_DS,
	VCPU_SREG_FS,
	VCPU_SREG_GS,
	VCPU_SREG_TR,
	VCPU_SREG_LDTR,
};

181
#include <asm/kvm_emulate.h>
182

183 184
#define KVM_NR_MEM_OBJS 40

185 186 187 188
#define KVM_NR_DB_REGS	4

#define DR6_BD		(1 << 13)
#define DR6_BS		(1 << 14)
189
#define DR6_BT		(1 << 15)
190 191 192 193
#define DR6_RTM		(1 << 16)
#define DR6_FIXED_1	0xfffe0ff0
#define DR6_INIT	0xffff0ff0
#define DR6_VOLATILE	0x0001e00f
194 195 196 197 198

#define DR7_BP_EN_MASK	0x000000ff
#define DR7_GE		(1 << 9)
#define DR7_GD		(1 << 13)
#define DR7_FIXED_1	0x00000400
199
#define DR7_VOLATILE	0xffff2bff
200

201 202 203 204 205
#define PFERR_PRESENT_BIT 0
#define PFERR_WRITE_BIT 1
#define PFERR_USER_BIT 2
#define PFERR_RSVD_BIT 3
#define PFERR_FETCH_BIT 4
206
#define PFERR_PK_BIT 5
207 208
#define PFERR_GUEST_FINAL_BIT 32
#define PFERR_GUEST_PAGE_BIT 33
209 210 211 212 213 214

#define PFERR_PRESENT_MASK (1U << PFERR_PRESENT_BIT)
#define PFERR_WRITE_MASK (1U << PFERR_WRITE_BIT)
#define PFERR_USER_MASK (1U << PFERR_USER_BIT)
#define PFERR_RSVD_MASK (1U << PFERR_RSVD_BIT)
#define PFERR_FETCH_MASK (1U << PFERR_FETCH_BIT)
215
#define PFERR_PK_MASK (1U << PFERR_PK_BIT)
216 217 218 219 220 221
#define PFERR_GUEST_FINAL_MASK (1ULL << PFERR_GUEST_FINAL_BIT)
#define PFERR_GUEST_PAGE_MASK (1ULL << PFERR_GUEST_PAGE_BIT)

#define PFERR_NESTED_GUEST_PAGE (PFERR_GUEST_PAGE_MASK |	\
				 PFERR_WRITE_MASK |		\
				 PFERR_PRESENT_MASK)
222

223 224 225 226 227 228 229
/*
 * The mask used to denote special SPTEs, which can be either MMIO SPTEs or
 * Access Tracking SPTEs. We use bit 62 instead of bit 63 to avoid conflicting
 * with the SVE bit in EPT PTEs.
 */
#define SPTE_SPECIAL_MASK (1ULL << 62)

230 231
/* apic attention bits */
#define KVM_APIC_CHECK_VAPIC	0
232 233 234 235 236 237 238
/*
 * The following bit is set with PV-EOI, unset on EOI.
 * We detect PV-EOI changes by guest by comparing
 * this bit with PV-EOI in guest memory.
 * See the implementation in apic_update_pv_eoi.
 */
#define KVM_APIC_PV_EOI_PENDING	1
239

F
Feng Wu 已提交
240 241
struct kvm_kernel_irq_routing_entry;

242 243 244 245 246 247 248 249 250
/*
 * We don't want allocation failures within the mmu code, so we preallocate
 * enough memory for a single page fault in a cache.
 */
struct kvm_mmu_memory_cache {
	int nobjs;
	void *objects[KVM_NR_MEM_OBJS];
};

251 252 253 254 255
/*
 * the pages used as guest page table on soft mmu are tracked by
 * kvm_memory_slot.arch.gfn_track which is 16 bits, so the role bits used
 * by indirect shadow page can not be more than 15 bits.
 *
256
 * Currently, we used 14 bits that are @level, @gpte_is_8_bytes, @quadrant, @access,
257 258
 * @nxe, @cr0_wp, @smep_andnot_wp and @smap_andnot_wp.
 */
259
union kvm_mmu_page_role {
260
	u32 word;
261
	struct {
262
		unsigned level:4;
263
		unsigned gpte_is_8_bytes:1;
264
		unsigned quadrant:2;
265
		unsigned direct:1;
266
		unsigned access:3;
267
		unsigned invalid:1;
268
		unsigned nxe:1;
269
		unsigned cr0_wp:1;
270
		unsigned smep_andnot_wp:1;
271
		unsigned smap_andnot_wp:1;
272
		unsigned ad_disabled:1;
273 274
		unsigned guest_mode:1;
		unsigned :6;
275 276 277 278 279 280 281 282

		/*
		 * This is left at the top of the word so that
		 * kvm_memslots_for_spte_role can extract it with a
		 * simple shift.  While there is room, give it a whole
		 * byte so it is also faster to load it from memory.
		 */
		unsigned smm:8;
283 284 285
	};
};

286
union kvm_mmu_extended_role {
287 288 289 290 291 292
/*
 * This structure complements kvm_mmu_page_role caching everything needed for
 * MMU configuration. If nothing in both these structures changed, MMU
 * re-configuration can be skipped. @valid bit is set on first usage so we don't
 * treat all-zero structure as valid data.
 */
293
	u32 word;
294 295 296
	struct {
		unsigned int valid:1;
		unsigned int execonly:1;
297
		unsigned int cr0_pg:1;
298
		unsigned int cr4_pae:1;
299 300 301 302
		unsigned int cr4_pse:1;
		unsigned int cr4_pke:1;
		unsigned int cr4_smap:1;
		unsigned int cr4_smep:1;
303
		unsigned int cr4_la57:1;
304
		unsigned int maxphyaddr:6;
305
	};
306 307 308 309 310 311 312 313 314 315
};

union kvm_mmu_role {
	u64 as_u64;
	struct {
		union kvm_mmu_page_role base;
		union kvm_mmu_extended_role ext;
	};
};

316 317 318 319
struct kvm_rmap_head {
	unsigned long val;
};

320 321 322
struct kvm_mmu_page {
	struct list_head link;
	struct hlist_node hash_link;
323
	bool unsync;
324
	bool mmio_cached;
325 326 327 328 329 330

	/*
	 * The following two entries are used to key the shadow page in the
	 * hash table.
	 */
	union kvm_mmu_page_role role;
331
	gfn_t gfn;
332 333 334 335

	u64 *spt;
	/* hold the gfn of each spte inside spt */
	gfn_t *gfns;
336
	int root_count;          /* Currently serving as active root */
337
	unsigned int unsync_children;
338
	struct kvm_rmap_head parent_ptes; /* rmap pointers to parent sptes */
339
	DECLARE_BITMAP(unsync_child_bitmap, 512);
340 341

#ifdef CONFIG_X86_32
342 343 344 345
	/*
	 * Used out of the mmu-lock to avoid reading spte values while an
	 * update is in progress; see the comments in __get_spte_lockless().
	 */
346 347 348
	int clear_spte_count;
#endif

349
	/* Number of writes since the last time traversal visited this page.  */
350
	atomic_t write_flooding_count;
351 352
};

353
struct kvm_pio_request {
354
	unsigned long linear_rip;
355 356 357 358 359 360
	unsigned long count;
	int in;
	int port;
	int size;
};

361
#define PT64_ROOT_MAX_LEVEL 5
362

363
struct rsvd_bits_validate {
364
	u64 rsvd_bits_mask[2][PT64_ROOT_MAX_LEVEL];
365 366 367
	u64 bad_mt_xwr;
};

368 369 370 371 372 373 374 375
struct kvm_mmu_root_info {
	gpa_t cr3;
	hpa_t hpa;
};

#define KVM_MMU_ROOT_INFO_INVALID \
	((struct kvm_mmu_root_info) { .cr3 = INVALID_PAGE, .hpa = INVALID_PAGE })

376 377
#define KVM_MMU_NUM_PREV_ROOTS 3

378
/*
379 380 381
 * x86 supports 4 paging modes (5-level 64-bit, 4-level 64-bit, 3-level 32-bit,
 * and 2-level 32-bit).  The kvm_mmu structure abstracts the details of the
 * current mmu mode.
382 383
 */
struct kvm_mmu {
384
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
385
	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
386
	u64 (*get_pdptr)(struct kvm_vcpu *vcpu, int index);
387 388
	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
			  bool prefault);
389 390
	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
				  struct x86_exception *fault);
391
	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
392
			    struct x86_exception *exception);
393 394
	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
			       struct x86_exception *exception);
395
	int (*sync_page)(struct kvm_vcpu *vcpu,
396
			 struct kvm_mmu_page *sp);
397
	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva, hpa_t root_hpa);
398
	void (*update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
399
			   u64 *spte, const void *pte);
400
	hpa_t root_hpa;
401
	gpa_t root_cr3;
402
	union kvm_mmu_role mmu_role;
403 404 405
	u8 root_level;
	u8 shadow_root_level;
	u8 ept_ad;
406
	bool direct_map;
407
	struct kvm_mmu_root_info prev_roots[KVM_MMU_NUM_PREV_ROOTS];
408

409 410 411 412 413 414 415
	/*
	 * Bitmap; bit set = permission fault
	 * Byte index: page fault error code [4:1]
	 * Bit index: pte permissions in ACC_* format
	 */
	u8 permissions[16];

416 417 418 419 420 421 422 423
	/*
	* The pkru_mask indicates if protection key checks are needed.  It
	* consists of 16 domains indexed by page fault error code bits [4:1],
	* with PFEC.RSVD replaced by ACC_USER_MASK from the page tables.
	* Each domain has 2 bits which are ANDed with AD and WD from PKRU.
	*/
	u32 pkru_mask;

424
	u64 *pae_root;
425
	u64 *lm_root;
426 427 428 429 430 431 432 433

	/*
	 * check zero bits on shadow page table entries, these
	 * bits include not only hardware reserved bits but also
	 * the bits spte never used.
	 */
	struct rsvd_bits_validate shadow_zero_check;

434
	struct rsvd_bits_validate guest_rsvd_check;
435

436 437
	/* Can have large pages at levels 2..last_nonleaf_level-1. */
	u8 last_nonleaf_level;
A
Avi Kivity 已提交
438

439 440
	bool nx;

441
	u64 pdptrs[4]; /* pae */
442 443
};

444 445 446 447 448
struct kvm_tlb_range {
	u64 start_gfn;
	u64 pages;
};

449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472
enum pmc_type {
	KVM_PMC_GP = 0,
	KVM_PMC_FIXED,
};

struct kvm_pmc {
	enum pmc_type type;
	u8 idx;
	u64 counter;
	u64 eventsel;
	struct perf_event *perf_event;
	struct kvm_vcpu *vcpu;
};

struct kvm_pmu {
	unsigned nr_arch_gp_counters;
	unsigned nr_arch_fixed_counters;
	unsigned available_event_types;
	u64 fixed_ctr_ctrl;
	u64 global_ctrl;
	u64 global_status;
	u64 global_ovf_ctrl;
	u64 counter_bitmask[2];
	u64 global_ctrl_mask;
473
	u64 global_ovf_ctrl_mask;
474
	u64 reserved_bits;
475
	u8 version;
476 477
	struct kvm_pmc gp_counters[INTEL_PMC_MAX_GENERIC];
	struct kvm_pmc fixed_counters[INTEL_PMC_MAX_FIXED];
478 479 480 481
	struct irq_work irq_work;
	u64 reprogram_pmi;
};

482 483
struct kvm_pmu_ops;

484 485
enum {
	KVM_DEBUGREG_BP_ENABLED = 1,
486
	KVM_DEBUGREG_WONT_EXIT = 2,
487
	KVM_DEBUGREG_RELOAD = 4,
488 489
};

490 491 492
struct kvm_mtrr_range {
	u64 base;
	u64 mask;
X
Xiao Guangrong 已提交
493
	struct list_head node;
494 495
};

496
struct kvm_mtrr {
497
	struct kvm_mtrr_range var_ranges[KVM_NR_VAR_MTRR];
498
	mtrr_type fixed_ranges[KVM_NR_FIXED_MTRR_REGION];
499
	u64 deftype;
X
Xiao Guangrong 已提交
500 501

	struct list_head head;
502 503
};

A
Andrey Smetanin 已提交
504 505 506 507
/* Hyper-V SynIC timer */
struct kvm_vcpu_hv_stimer {
	struct hrtimer timer;
	int index;
508
	union hv_stimer_config config;
A
Andrey Smetanin 已提交
509 510 511 512 513 514
	u64 count;
	u64 exp_time;
	struct hv_message msg;
	bool msg_pending;
};

515 516 517 518 519 520 521 522 523 524 525
/* Hyper-V synthetic interrupt controller (SynIC)*/
struct kvm_vcpu_hv_synic {
	u64 version;
	u64 control;
	u64 msg_page;
	u64 evt_page;
	atomic64_t sint[HV_SYNIC_SINT_COUNT];
	atomic_t sint_to_gsi[HV_SYNIC_SINT_COUNT];
	DECLARE_BITMAP(auto_eoi_bitmap, 256);
	DECLARE_BITMAP(vec_bitmap, 256);
	bool active;
526
	bool dont_zero_synic_pages;
527 528
};

529 530
/* Hyper-V per vcpu emulation context */
struct kvm_vcpu_hv {
531
	u32 vp_index;
532
	u64 hv_vapic;
533
	s64 runtime_offset;
534
	struct kvm_vcpu_hv_synic synic;
A
Andrey Smetanin 已提交
535
	struct kvm_hyperv_exit exit;
A
Andrey Smetanin 已提交
536 537
	struct kvm_vcpu_hv_stimer stimer[HV_SYNIC_STIMER_COUNT];
	DECLARE_BITMAP(stimer_pending_bitmap, HV_SYNIC_STIMER_COUNT);
538
	cpumask_t tlb_flush;
539 540
};

541
struct kvm_vcpu_arch {
542 543 544 545 546 547 548
	/*
	 * rip and regs accesses must go through
	 * kvm_{register,rip}_{read,write} functions.
	 */
	unsigned long regs[NR_VCPU_REGS];
	u32 regs_avail;
	u32 regs_dirty;
549 550

	unsigned long cr0;
551
	unsigned long cr0_guest_owned_bits;
552 553 554
	unsigned long cr2;
	unsigned long cr3;
	unsigned long cr4;
555
	unsigned long cr4_guest_owned_bits;
556
	unsigned long cr8;
557
	u32 pkru;
558
	u32 hflags;
559
	u64 efer;
560 561
	u64 apic_base;
	struct kvm_lapic *apic;    /* kernel irqchip context */
562
	bool apicv_active;
563
	bool load_eoi_exitmap_pending;
564
	DECLARE_BITMAP(ioapic_handled_vectors, 256);
565
	unsigned long apic_attention;
566
	int32_t apic_arb_prio;
567 568
	int mp_state;
	u64 ia32_misc_enable_msr;
P
Paolo Bonzini 已提交
569
	u64 smbase;
570
	u64 smi_count;
571
	bool tpr_access_reporting;
W
Wanpeng Li 已提交
572
	u64 ia32_xss;
573
	u64 microcode_version;
574
	u64 arch_capabilities;
575

576 577 578 579 580 581 582
	/*
	 * Paging state of the vcpu
	 *
	 * If the vcpu runs in guest mode with two level paging this still saves
	 * the paging mode of the l1 guest. This context is always used to
	 * handle faults.
	 */
583 584 585 586
	struct kvm_mmu *mmu;

	/* Non-nested MMU for L1 */
	struct kvm_mmu root_mmu;
587

588 589 590
	/* L1 MMU when running nested */
	struct kvm_mmu guest_mmu;

591 592 593 594 595 596 597 598 599 600
	/*
	 * Paging state of an L2 guest (used for nested npt)
	 *
	 * This context will save all necessary information to walk page tables
	 * of the an L2 guest. This context is only initialized for page table
	 * walking and not for faulting since we never handle l2 page faults on
	 * the host.
	 */
	struct kvm_mmu nested_mmu;

601 602 603 604 605 606
	/*
	 * Pointer to the mmu context currently used for
	 * gva_to_gpa translations.
	 */
	struct kvm_mmu *walk_mmu;

607
	struct kvm_mmu_memory_cache mmu_pte_list_desc_cache;
608 609 610
	struct kvm_mmu_memory_cache mmu_page_cache;
	struct kvm_mmu_memory_cache mmu_page_header_cache;

611 612
	/*
	 * QEMU userspace and the guest each have their own FPU state.
613 614 615
	 * In vcpu_run, we switch between the user, maintained in the
	 * task_struct struct, and guest FPU contexts. While running a VCPU,
	 * the VCPU thread will have the guest FPU context.
616 617 618 619 620 621
	 *
	 * Note that while the PKRU state lives inside the fpu registers,
	 * it is switched out separately at VMENTER and VMEXIT time. The
	 * "guest_fpu" state here contains the guest FPU context, with the
	 * host PRKU bits.
	 */
622
	struct fpu *guest_fpu;
623

624
	u64 xcr0;
625
	u64 guest_supported_xcr0;
626
	u32 guest_xstate_size;
627 628 629 630

	struct kvm_pio_request pio;
	void *pio_data;

631 632
	u8 event_exit_inst_len;

633 634
	struct kvm_queued_exception {
		bool pending;
635
		bool injected;
636 637 638
		bool has_error_code;
		u8 nr;
		u32 error_code;
639 640
		unsigned long payload;
		bool has_payload;
641
		u8 nested_apf;
642 643
	} exception;

A
Avi Kivity 已提交
644
	struct kvm_queued_interrupt {
645
		bool injected;
646
		bool soft;
A
Avi Kivity 已提交
647 648 649
		u8 nr;
	} interrupt;

650 651 652
	int halt_request; /* real mode on Intel only */

	int cpuid_nent;
653
	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
654 655 656

	int maxphyaddr;

657 658 659
	/* emulate context */

	struct x86_emulate_ctxt emulate_ctxt;
660 661
	bool emulate_regs_need_sync_to_vcpu;
	bool emulate_regs_need_sync_from_vcpu;
662
	int (*complete_userspace_io)(struct kvm_vcpu *vcpu);
663 664

	gpa_t time;
665
	struct pvclock_vcpu_time_info hv_clock;
Z
Zachary Amsden 已提交
666
	unsigned int hw_tsc_khz;
667 668
	struct gfn_to_hva_cache pv_time;
	bool pv_time_enabled;
669 670
	/* set guest stopped flag in pvclock flags field */
	bool pvclock_set_guest_stopped_request;
G
Glauber Costa 已提交
671 672 673 674 675 676 677 678

	struct {
		u64 msr_val;
		u64 last_steal;
		struct gfn_to_hva_cache stime;
		struct kvm_steal_time steal;
	} st;

679
	u64 tsc_offset;
680
	u64 last_guest_tsc;
681
	u64 last_host_tsc;
682
	u64 tsc_offset_adjustment;
683 684
	u64 this_tsc_nsec;
	u64 this_tsc_write;
T
Tomasz Grabiec 已提交
685
	u64 this_tsc_generation;
Z
Zachary Amsden 已提交
686
	bool tsc_catchup;
687 688 689 690
	bool tsc_always_catchup;
	s8 virtual_tsc_shift;
	u32 virtual_tsc_mult;
	u32 virtual_tsc_khz;
W
Will Auld 已提交
691
	s64 ia32_tsc_adjust_msr;
692
	u64 tsc_scaling_ratio;
693

A
Avi Kivity 已提交
694 695 696
	atomic_t nmi_queued;  /* unprocessed asynchronous NMIs */
	unsigned nmi_pending; /* NMI queued after currently running handler */
	bool nmi_injected;    /* Trying to inject an NMI this entry */
697
	bool smi_pending;    /* SMI queued after currently running handler */
A
Avi Kivity 已提交
698

699
	struct kvm_mtrr mtrr_state;
700
	u64 pat;
701

702
	unsigned switch_db_regs;
703 704 705 706
	unsigned long db[KVM_NR_DB_REGS];
	unsigned long dr6;
	unsigned long dr7;
	unsigned long eff_db[KVM_NR_DB_REGS];
707
	unsigned long guest_debug_dr7;
K
Kyle Huey 已提交
708 709
	u64 msr_platform_info;
	u64 msr_misc_features_enables;
H
Huang Ying 已提交
710 711 712 713

	u64 mcg_cap;
	u64 mcg_status;
	u64 mcg_ctl;
714
	u64 mcg_ext_ctl;
H
Huang Ying 已提交
715
	u64 *mce_banks;
716

717 718 719 720
	/* Cache MMIO info */
	u64 mmio_gva;
	unsigned access;
	gfn_t mmio_gfn;
721
	u64 mmio_gen;
722

723 724
	struct kvm_pmu pmu;

725 726
	/* used for guest single stepping over the given code position */
	unsigned long singlestep_rip;
J
Jan Kiszka 已提交
727

728
	struct kvm_vcpu_hv hyperv;
729 730

	cpumask_var_t wbinvd_dirty_mask;
731

732 733 734
	unsigned long last_retry_eip;
	unsigned long last_retry_addr;

735 736 737
	struct {
		bool halted;
		gfn_t gfns[roundup_pow_of_two(ASYNC_PF_PER_VCPU)];
738 739
		struct gfn_to_hva_cache data;
		u64 msr_val;
740
		u32 id;
741
		bool send_user_only;
742
		u32 host_apf_reason;
743
		unsigned long nested_apf_token;
744
		bool delivery_as_pf_vmexit;
745
	} apf;
746 747 748 749 750 751

	/* OSVW MSRs (AMD only) */
	struct {
		u64 length;
		u64 status;
	} osvw;
752 753 754 755 756

	struct {
		u64 msr_val;
		struct gfn_to_hva_cache data;
	} pv_eoi;
757

758 759
	u64 msr_kvm_poll_control;

760 761 762 763 764 765
	/*
	 * Indicate whether the access faults on its page table in guest
	 * which is set when fix page fault and used to detect unhandeable
	 * instruction.
	 */
	bool write_fault_to_shadow_pgtable;
766 767 768

	/* set at EPT violation at this point */
	unsigned long exit_qualification;
769 770 771 772 773

	/* pv related host specific info */
	struct {
		bool pv_unhalted;
	} pv;
774 775

	int pending_ioapic_eoi;
776
	int pending_external_vector;
777

778
	/* GPA available */
779
	bool gpa_available;
780
	gpa_t gpa_val;
781 782 783

	/* be preempted when it's in kernel-mode(cpl=0) */
	bool preempted_in_kernel;
P
Paolo Bonzini 已提交
784 785 786

	/* Flush the L1 Data cache for L1TF mitigation on VMENTER */
	bool l1tf_flush_l1d;
787 788 789

	/* AMD MSRC001_0015 Hardware Configuration */
	u64 msr_hwcr;
790 791
};

792
struct kvm_lpage_info {
793
	int disallow_lpage;
794 795 796
};

struct kvm_arch_memory_slot {
797
	struct kvm_rmap_head *rmap[KVM_NR_PAGE_SIZES];
798
	struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1];
799
	unsigned short *gfn_track[KVM_PAGE_TRACK_MAX];
800 801
};

802 803 804 805 806 807 808 809 810 811 812
/*
 * We use as the mode the number of bits allocated in the LDR for the
 * logical processor ID.  It happens that these are all powers of two.
 * This makes it is very easy to detect cases where the APICs are
 * configured for multiple modes; in that case, we cannot use the map and
 * hence cannot use kvm_irq_delivery_to_apic_fast either.
 */
#define KVM_APIC_MODE_XAPIC_CLUSTER          4
#define KVM_APIC_MODE_XAPIC_FLAT             8
#define KVM_APIC_MODE_X2APIC                16

813 814
struct kvm_apic_map {
	struct rcu_head rcu;
815
	u8 mode;
R
Radim Krčmář 已提交
816
	u32 max_apic_id;
817 818 819 820
	union {
		struct kvm_lapic *xapic_flat_map[8];
		struct kvm_lapic *xapic_cluster_map[16][4];
	};
R
Radim Krčmář 已提交
821
	struct kvm_lapic *phys_map[];
822 823
};

824 825
/* Hyper-V emulation context */
struct kvm_hv {
826
	struct mutex hv_lock;
827 828 829
	u64 hv_guest_os_id;
	u64 hv_hypercall;
	u64 hv_tsc_page;
830 831 832 833

	/* Hyper-v based guest crash (NT kernel bugcheck) parameters */
	u64 hv_crash_param[HV_X64_MSR_CRASH_PARAMS];
	u64 hv_crash_ctl;
P
Paolo Bonzini 已提交
834 835

	HV_REFERENCE_TSC_PAGE tsc_ref;
836 837

	struct idr conn_to_evt;
838 839 840 841

	u64 hv_reenlightenment_control;
	u64 hv_tsc_emulation_control;
	u64 hv_tsc_emulation_status;
842 843 844

	/* How many vCPUs have VP index != vCPU index */
	atomic_t num_mismatched_vp_indexes;
845 846
};

847 848 849 850 851 852
enum kvm_irqchip_mode {
	KVM_IRQCHIP_NONE,
	KVM_IRQCHIP_KERNEL,       /* created with KVM_CREATE_IRQCHIP */
	KVM_IRQCHIP_SPLIT,        /* created with KVM_CAP_SPLIT_IRQCHIP */
};

853
struct kvm_arch {
854 855 856
	unsigned long n_used_mmu_pages;
	unsigned long n_requested_mmu_pages;
	unsigned long n_max_mmu_pages;
857
	unsigned int indirect_shadow_pages;
858 859 860 861 862
	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
	/*
	 * Hash table of struct kvm_mmu_page.
	 */
	struct list_head active_mmu_pages;
863
	struct kvm_page_track_notifier_node mmu_sp_tracker;
864
	struct kvm_page_track_notifier_head track_notifier_head;
865

B
Ben-Ami Yassour 已提交
866
	struct list_head assigned_dev_head;
J
Joerg Roedel 已提交
867
	struct iommu_domain *iommu_domain;
868
	bool iommu_noncoherent;
869 870
#define __KVM_HAVE_ARCH_NONCOHERENT_DMA
	atomic_t noncoherent_dma_count;
871 872
#define __KVM_HAVE_ARCH_ASSIGNED_DEVICE
	atomic_t assigned_device_count;
873 874
	struct kvm_pic *vpic;
	struct kvm_ioapic *vioapic;
S
Sheng Yang 已提交
875
	struct kvm_pit *vpit;
876
	atomic_t vapics_in_nmi_mode;
877 878
	struct mutex apic_map_lock;
	struct kvm_apic_map *apic_map;
879

880
	bool apic_access_page_done;
881 882

	gpa_t wall_clock;
883

884
	bool mwait_in_guest;
885
	bool hlt_in_guest;
886
	bool pause_in_guest;
887
	bool cstate_in_guest;
888

889
	unsigned long irq_sources_bitmap;
890
	s64 kvmclock_offset;
891
	raw_spinlock_t tsc_write_lock;
Z
Zachary Amsden 已提交
892 893
	u64 last_tsc_nsec;
	u64 last_tsc_write;
894
	u32 last_tsc_khz;
895 896 897
	u64 cur_tsc_nsec;
	u64 cur_tsc_write;
	u64 cur_tsc_offset;
T
Tomasz Grabiec 已提交
898
	u64 cur_tsc_generation;
899
	int nr_vcpus_matched_tsc;
E
Ed Swierk 已提交
900

901 902 903
	spinlock_t pvclock_gtod_sync_lock;
	bool use_master_clock;
	u64 master_kernel_ns;
904
	u64 master_cycle_now;
905
	struct delayed_work kvmclock_update_work;
906
	struct delayed_work kvmclock_sync_work;
907

E
Ed Swierk 已提交
908
	struct kvm_xen_hvm_config xen_hvm_config;
909

910 911 912
	/* reads protected by irq_srcu, writes by irq_lock */
	struct hlist_head mask_notifier_list;

913
	struct kvm_hv hyperv;
914 915 916 917

	#ifdef CONFIG_KVM_MMU_AUDIT
	int audit_point;
	#endif
918

919
	bool backwards_tsc_observed;
920
	bool boot_vcpu_runs_old_kvmclock;
921
	u32 bsp_vcpu_id;
922 923

	u64 disabled_quirks;
924

925
	enum kvm_irqchip_mode irqchip_mode;
926
	u8 nr_reserved_ioapic_pins;
927 928

	bool disabled_lapic_found;
929

930
	bool x2apic_format;
931
	bool x2apic_broadcast_quirk_disabled;
932 933

	bool guest_can_read_msr_platform_info;
934
	bool exception_payload_enabled;
935 936
};

937
struct kvm_vm_stat {
938 939 940 941 942 943 944 945 946 947
	ulong mmu_shadow_zapped;
	ulong mmu_pte_write;
	ulong mmu_pte_updated;
	ulong mmu_pde_zapped;
	ulong mmu_flooded;
	ulong mmu_recycled;
	ulong mmu_cache_miss;
	ulong mmu_unsync;
	ulong remote_tlb_flush;
	ulong lpages;
948
	ulong max_mmu_page_hash_collisions;
949 950
};

951
struct kvm_vcpu_stat {
952 953 954 955 956 957 958 959 960 961 962
	u64 pf_fixed;
	u64 pf_guest;
	u64 tlb_flush;
	u64 invlpg;

	u64 exits;
	u64 io_exits;
	u64 mmio_exits;
	u64 signal_exits;
	u64 irq_window_exits;
	u64 nmi_window_exits;
P
Paolo Bonzini 已提交
963
	u64 l1d_flush;
964 965 966 967 968 969 970 971 972 973 974 975 976 977
	u64 halt_exits;
	u64 halt_successful_poll;
	u64 halt_attempted_poll;
	u64 halt_poll_invalid;
	u64 halt_wakeup;
	u64 request_irq_exits;
	u64 irq_exits;
	u64 host_state_reload;
	u64 fpu_reload;
	u64 insn_emulation;
	u64 insn_emulation_fail;
	u64 hypercalls;
	u64 irq_injections;
	u64 nmi_injections;
978
	u64 req_event;
979
};
980

981 982
struct x86_instruction_info;

983 984 985 986 987 988
struct msr_data {
	bool host_initiated;
	u32 index;
	u64 data;
};

P
Paolo Bonzini 已提交
989 990
struct kvm_lapic_irq {
	u32 vector;
991 992 993 994
	u16 delivery_mode;
	u16 dest_mode;
	bool level;
	u16 trig_mode;
P
Paolo Bonzini 已提交
995 996
	u32 shorthand;
	u32 dest_id;
997
	bool msi_redir_hint;
P
Paolo Bonzini 已提交
998 999
};

1000 1001 1002
struct kvm_x86_ops {
	int (*cpu_has_kvm_support)(void);          /* __init */
	int (*disabled_by_bios)(void);             /* __init */
1003 1004
	int (*hardware_enable)(void);
	void (*hardware_disable)(void);
1005
	int (*check_processor_compatibility)(void);/* __init */
1006 1007
	int (*hardware_setup)(void);               /* __init */
	void (*hardware_unsetup)(void);            /* __exit */
1008
	bool (*cpu_has_accelerated_tpr)(void);
1009
	bool (*has_emulated_msr)(int index);
1010
	void (*cpuid_update)(struct kvm_vcpu *vcpu);
1011

1012 1013
	struct kvm *(*vm_alloc)(void);
	void (*vm_free)(struct kvm *);
1014 1015 1016
	int (*vm_init)(struct kvm *kvm);
	void (*vm_destroy)(struct kvm *kvm);

1017 1018 1019
	/* Create, but do not attach this VCPU */
	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
	void (*vcpu_free)(struct kvm_vcpu *vcpu);
1020
	void (*vcpu_reset)(struct kvm_vcpu *vcpu, bool init_event);
1021 1022 1023 1024 1025

	void (*prepare_guest_switch)(struct kvm_vcpu *vcpu);
	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
	void (*vcpu_put)(struct kvm_vcpu *vcpu);

1026
	void (*update_bp_intercept)(struct kvm_vcpu *vcpu);
1027
	int (*get_msr)(struct kvm_vcpu *vcpu, struct msr_data *msr);
1028
	int (*set_msr)(struct kvm_vcpu *vcpu, struct msr_data *msr);
1029 1030 1031
	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
	void (*get_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
1032
	int (*get_cpl)(struct kvm_vcpu *vcpu);
1033 1034 1035
	void (*set_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
	void (*get_cs_db_l_bits)(struct kvm_vcpu *vcpu, int *db, int *l);
1036
	void (*decache_cr0_guest_bits)(struct kvm_vcpu *vcpu);
1037
	void (*decache_cr3)(struct kvm_vcpu *vcpu);
1038 1039 1040
	void (*decache_cr4_guest_bits)(struct kvm_vcpu *vcpu);
	void (*set_cr0)(struct kvm_vcpu *vcpu, unsigned long cr0);
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);
1041
	int (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
1042
	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
1043 1044 1045 1046
	void (*get_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*set_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*get_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*set_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
J
Jan Kiszka 已提交
1047 1048
	u64 (*get_dr6)(struct kvm_vcpu *vcpu);
	void (*set_dr6)(struct kvm_vcpu *vcpu, unsigned long value);
1049
	void (*sync_dirty_debug_regs)(struct kvm_vcpu *vcpu);
1050
	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
1051
	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
1052 1053 1054
	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);

1055
	void (*tlb_flush)(struct kvm_vcpu *vcpu, bool invalidate_gpa);
1056
	int  (*tlb_remote_flush)(struct kvm *kvm);
1057 1058
	int  (*tlb_remote_flush_with_range)(struct kvm *kvm,
			struct kvm_tlb_range *range);
1059

1060 1061 1062 1063 1064 1065 1066
	/*
	 * Flush any TLB entries associated with the given GVA.
	 * Does not need to flush GPA->HPA mappings.
	 * Can potentially get non-canonical addresses through INVLPGs, which
	 * the implementation may choose to ignore if appropriate.
	 */
	void (*tlb_flush_gva)(struct kvm_vcpu *vcpu, gva_t addr);
1067

A
Avi Kivity 已提交
1068 1069
	void (*run)(struct kvm_vcpu *vcpu);
	int (*handle_exit)(struct kvm_vcpu *vcpu);
1070
	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
1071
	void (*set_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
1072
	u32 (*get_interrupt_shadow)(struct kvm_vcpu *vcpu);
1073 1074
	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
				unsigned char *hypercall_addr);
1075
	void (*set_irq)(struct kvm_vcpu *vcpu);
1076
	void (*set_nmi)(struct kvm_vcpu *vcpu);
1077
	void (*queue_exception)(struct kvm_vcpu *vcpu);
A
Avi Kivity 已提交
1078
	void (*cancel_injection)(struct kvm_vcpu *vcpu);
1079
	int (*interrupt_allowed)(struct kvm_vcpu *vcpu);
1080
	int (*nmi_allowed)(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
1081 1082
	bool (*get_nmi_mask)(struct kvm_vcpu *vcpu);
	void (*set_nmi_mask)(struct kvm_vcpu *vcpu, bool masked);
1083 1084
	void (*enable_nmi_window)(struct kvm_vcpu *vcpu);
	void (*enable_irq_window)(struct kvm_vcpu *vcpu);
1085
	void (*update_cr8_intercept)(struct kvm_vcpu *vcpu, int tpr, int irr);
1086
	bool (*get_enable_apicv)(struct kvm_vcpu *vcpu);
1087
	void (*refresh_apicv_exec_ctrl)(struct kvm_vcpu *vcpu);
1088
	void (*hwapic_irr_update)(struct kvm_vcpu *vcpu, int max_irr);
1089
	void (*hwapic_isr_update)(struct kvm_vcpu *vcpu, int isr);
1090
	bool (*guest_apic_has_interrupt)(struct kvm_vcpu *vcpu);
1091
	void (*load_eoi_exitmap)(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap);
1092
	void (*set_virtual_apic_mode)(struct kvm_vcpu *vcpu);
1093
	void (*set_apic_access_page_addr)(struct kvm_vcpu *vcpu, hpa_t hpa);
1094
	void (*deliver_posted_interrupt)(struct kvm_vcpu *vcpu, int vector);
1095
	int (*sync_pir_to_irr)(struct kvm_vcpu *vcpu);
1096
	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
1097
	int (*set_identity_map_addr)(struct kvm *kvm, u64 ident_addr);
1098
	int (*get_tdp_level)(struct kvm_vcpu *vcpu);
1099
	u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio);
1100
	int (*get_lpage_level)(void);
1101
	bool (*rdtscp_supported)(void);
1102
	bool (*invpcid_supported)(void);
1103

1104 1105
	void (*set_tdp_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);

1106 1107
	void (*set_supported_cpuid)(u32 func, struct kvm_cpuid_entry2 *entry);

1108 1109
	bool (*has_wbinvd_exit)(void);

1110
	u64 (*read_l1_tsc_offset)(struct kvm_vcpu *vcpu);
1111 1112
	/* Returns actual tsc_offset set in active VMCS */
	u64 (*write_l1_tsc_offset)(struct kvm_vcpu *vcpu, u64 offset);
1113

1114
	void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2);
1115 1116 1117 1118

	int (*check_intercept)(struct kvm_vcpu *vcpu,
			       struct x86_instruction_info *info,
			       enum x86_intercept_stage stage);
1119
	void (*handle_external_intr)(struct kvm_vcpu *vcpu);
1120
	bool (*mpx_supported)(void);
1121
	bool (*xsaves_supported)(void);
1122
	bool (*umip_emulated)(void);
1123
	bool (*pt_supported)(void);
1124 1125

	int (*check_nested_events)(struct kvm_vcpu *vcpu, bool external_intr);
1126
	void (*request_immediate_exit)(struct kvm_vcpu *vcpu);
1127 1128

	void (*sched_in)(struct kvm_vcpu *kvm, int cpu);
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

	/*
	 * Arch-specific dirty logging hooks. These hooks are only supposed to
	 * be valid if the specific arch has hardware-accelerated dirty logging
	 * mechanism. Currently only for PML on VMX.
	 *
	 *  - slot_enable_log_dirty:
	 *	called when enabling log dirty mode for the slot.
	 *  - slot_disable_log_dirty:
	 *	called when disabling log dirty mode for the slot.
	 *	also called when slot is created with log dirty disabled.
	 *  - flush_log_dirty:
	 *	called before reporting dirty_bitmap to userspace.
	 *  - enable_log_dirty_pt_masked:
	 *	called when reenabling log dirty for the GFNs in the mask after
	 *	corresponding bits are cleared in slot->dirty_bitmap.
	 */
	void (*slot_enable_log_dirty)(struct kvm *kvm,
				      struct kvm_memory_slot *slot);
	void (*slot_disable_log_dirty)(struct kvm *kvm,
				       struct kvm_memory_slot *slot);
	void (*flush_log_dirty)(struct kvm *kvm);
	void (*enable_log_dirty_pt_masked)(struct kvm *kvm,
					   struct kvm_memory_slot *slot,
					   gfn_t offset, unsigned long mask);
1154 1155
	int (*write_log_dirty)(struct kvm_vcpu *vcpu);

1156 1157
	/* pmu operations of sub-arch */
	const struct kvm_pmu_ops *pmu_ops;
1158

1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
	/*
	 * Architecture specific hooks for vCPU blocking due to
	 * HLT instruction.
	 * Returns for .pre_block():
	 *    - 0 means continue to block the vCPU.
	 *    - 1 means we cannot block the vCPU since some event
	 *        happens during this period, such as, 'ON' bit in
	 *        posted-interrupts descriptor is set.
	 */
	int (*pre_block)(struct kvm_vcpu *vcpu);
	void (*post_block)(struct kvm_vcpu *vcpu);
1170 1171 1172 1173

	void (*vcpu_blocking)(struct kvm_vcpu *vcpu);
	void (*vcpu_unblocking)(struct kvm_vcpu *vcpu);

1174 1175
	int (*update_pi_irte)(struct kvm *kvm, unsigned int host_irq,
			      uint32_t guest_irq, bool set);
1176
	void (*apicv_post_state_restore)(struct kvm_vcpu *vcpu);
1177

1178 1179
	int (*set_hv_timer)(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc,
			    bool *expired);
1180
	void (*cancel_hv_timer)(struct kvm_vcpu *vcpu);
1181 1182

	void (*setup_mce)(struct kvm_vcpu *vcpu);
1183

1184 1185 1186 1187 1188 1189
	int (*get_nested_state)(struct kvm_vcpu *vcpu,
				struct kvm_nested_state __user *user_kvm_nested_state,
				unsigned user_data_size);
	int (*set_nested_state)(struct kvm_vcpu *vcpu,
				struct kvm_nested_state __user *user_kvm_nested_state,
				struct kvm_nested_state *kvm_state);
1190 1191
	void (*get_vmcs12_pages)(struct kvm_vcpu *vcpu);

1192
	int (*smi_allowed)(struct kvm_vcpu *vcpu);
1193
	int (*pre_enter_smm)(struct kvm_vcpu *vcpu, char *smstate);
1194
	int (*pre_leave_smm)(struct kvm_vcpu *vcpu, const char *smstate);
1195
	int (*enable_smi_window)(struct kvm_vcpu *vcpu);
1196 1197

	int (*mem_enc_op)(struct kvm *kvm, void __user *argp);
1198 1199
	int (*mem_enc_reg_region)(struct kvm *kvm, struct kvm_enc_region *argp);
	int (*mem_enc_unreg_region)(struct kvm *kvm, struct kvm_enc_region *argp);
1200 1201

	int (*get_msr_feature)(struct kvm_msr_entry *entry);
1202 1203 1204

	int (*nested_enable_evmcs)(struct kvm_vcpu *vcpu,
				   uint16_t *vmcs_version);
1205
	uint16_t (*nested_get_evmcs_version)(struct kvm_vcpu *vcpu);
1206 1207

	bool (*need_emulation_on_page_fault)(struct kvm_vcpu *vcpu);
1208 1209
};

1210
struct kvm_arch_async_pf {
1211
	u32 token;
1212
	gfn_t gfn;
X
Xiao Guangrong 已提交
1213
	unsigned long cr3;
1214
	bool direct_map;
1215 1216
};

1217
extern struct kvm_x86_ops *kvm_x86_ops;
1218
extern struct kmem_cache *x86_fpu_cache;
1219

1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230
#define __KVM_HAVE_ARCH_VM_ALLOC
static inline struct kvm *kvm_arch_alloc_vm(void)
{
	return kvm_x86_ops->vm_alloc();
}

static inline void kvm_arch_free_vm(struct kvm *kvm)
{
	return kvm_x86_ops->vm_free(kvm);
}

1231 1232 1233 1234 1235 1236 1237 1238 1239 1240
#define __KVM_HAVE_ARCH_FLUSH_REMOTE_TLB
static inline int kvm_arch_flush_remote_tlb(struct kvm *kvm)
{
	if (kvm_x86_ops->tlb_remote_flush &&
	    !kvm_x86_ops->tlb_remote_flush(kvm))
		return 0;
	else
		return -ENOTSUPP;
}

1241 1242 1243 1244 1245
int kvm_mmu_module_init(void);
void kvm_mmu_module_exit(void);

void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
int kvm_mmu_create(struct kvm_vcpu *vcpu);
1246 1247
void kvm_mmu_init_vm(struct kvm *kvm);
void kvm_mmu_uninit_vm(struct kvm *kvm);
S
Sheng Yang 已提交
1248
void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
1249
		u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask,
1250
		u64 acc_track_mask, u64 me_mask);
1251

1252
void kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
1253 1254
void kvm_mmu_slot_remove_write_access(struct kvm *kvm,
				      struct kvm_memory_slot *memslot);
1255
void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm,
1256
				   const struct kvm_memory_slot *memslot);
1257 1258 1259 1260 1261 1262 1263 1264 1265
void kvm_mmu_slot_leaf_clear_dirty(struct kvm *kvm,
				   struct kvm_memory_slot *memslot);
void kvm_mmu_slot_largepage_remove_write_access(struct kvm *kvm,
					struct kvm_memory_slot *memslot);
void kvm_mmu_slot_set_dirty(struct kvm *kvm,
			    struct kvm_memory_slot *memslot);
void kvm_mmu_clear_dirty_pt_masked(struct kvm *kvm,
				   struct kvm_memory_slot *slot,
				   gfn_t gfn_offset, unsigned long mask);
1266
void kvm_mmu_zap_all(struct kvm *kvm);
1267
void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, u64 gen);
1268 1269
unsigned long kvm_mmu_calculate_default_mmu_pages(struct kvm *kvm);
void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned long kvm_nr_mmu_pages);
1270

1271
int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3);
1272
bool pdptrs_changed(struct kvm_vcpu *vcpu);
1273

1274
int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
1275
			  const void *val, int bytes);
1276

1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
struct kvm_irq_mask_notifier {
	void (*func)(struct kvm_irq_mask_notifier *kimn, bool masked);
	int irq;
	struct hlist_node link;
};

void kvm_register_irq_mask_notifier(struct kvm *kvm, int irq,
				    struct kvm_irq_mask_notifier *kimn);
void kvm_unregister_irq_mask_notifier(struct kvm *kvm, int irq,
				      struct kvm_irq_mask_notifier *kimn);
void kvm_fire_mask_notifiers(struct kvm *kvm, unsigned irqchip, unsigned pin,
			     bool mask);

1290
extern bool tdp_enabled;
1291

1292 1293
u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu);

1294 1295 1296 1297
/* control of guest tsc rate supported? */
extern bool kvm_has_tsc_control;
/* maximum supported tsc_khz for guests */
extern u32  kvm_max_guest_tsc_khz;
1298 1299 1300 1301
/* number of bits of the fractional part of the TSC scaling ratio */
extern u8   kvm_tsc_scaling_ratio_frac_bits;
/* maximum allowed value of TSC scaling ratio */
extern u64  kvm_max_tsc_scaling_ratio;
1302 1303
/* 1ull << kvm_tsc_scaling_ratio_frac_bits */
extern u64  kvm_default_tsc_scaling_ratio;
1304

1305
extern u64 kvm_mce_cap_supported;
1306

1307
enum emulation_result {
P
Paolo Bonzini 已提交
1308 1309
	EMULATE_DONE,         /* no further processing */
	EMULATE_USER_EXIT,    /* kvm_run ready for userspace exit */
1310 1311 1312
	EMULATE_FAIL,         /* can't emulate this instruction */
};

1313 1314
#define EMULTYPE_NO_DECODE	    (1 << 0)
#define EMULTYPE_TRAP_UD	    (1 << 1)
1315
#define EMULTYPE_SKIP		    (1 << 2)
1316 1317 1318
#define EMULTYPE_ALLOW_RETRY	    (1 << 3)
#define EMULTYPE_NO_UD_ON_FAIL	    (1 << 4)
#define EMULTYPE_VMWARE		    (1 << 5)
1319 1320 1321
int kvm_emulate_instruction(struct kvm_vcpu *vcpu, int emulation_type);
int kvm_emulate_instruction_from_buffer(struct kvm_vcpu *vcpu,
					void *insn, int insn_len);
1322

1323
void kvm_enable_efer_bits(u64);
1324
bool kvm_valid_efer(struct kvm_vcpu *vcpu, u64 efer);
1325
int kvm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr);
1326
int kvm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr);
1327 1328 1329

struct x86_emulate_ctxt;

1330
int kvm_fast_pio(struct kvm_vcpu *vcpu, int size, unsigned short port, int in);
1331
int kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
1332
int kvm_emulate_halt(struct kvm_vcpu *vcpu);
1333
int kvm_vcpu_halt(struct kvm_vcpu *vcpu);
1334
int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu);
1335

1336
void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg);
1337
int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg);
1338
void kvm_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector);
1339

1340 1341
int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int idt_index,
		    int reason, bool has_error_code, u32 error_code);
1342

1343
int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
1344
int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3);
1345
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4);
A
Andre Przywara 已提交
1346
int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8);
1347 1348
int kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val);
int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val);
1349 1350
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu);
void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
1351
void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
1352
int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr);
1353

1354
int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr);
1355
int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr);
1356

1357 1358
unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu);
void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);
A
Avi Kivity 已提交
1359
bool kvm_rdpmc(struct kvm_vcpu *vcpu);
1360

1361 1362
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
1363 1364
void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
1365
void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
1366 1367 1368
int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
			    gfn_t gfn, void *data, int offset, int len,
			    u32 access);
1369
bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
1370
bool kvm_require_dr(struct kvm_vcpu *vcpu, int dr);
1371

1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383
static inline int __kvm_irq_line_state(unsigned long *irq_state,
				       int irq_source_id, int level)
{
	/* Logical OR for level trig interrupt */
	if (level)
		__set_bit(irq_source_id, irq_state);
	else
		__clear_bit(irq_source_id, irq_state);

	return !!(*irq_state);
}

1384 1385 1386
#define KVM_MMU_ROOT_CURRENT		BIT(0)
#define KVM_MMU_ROOT_PREVIOUS(i)	BIT(1+i)
#define KVM_MMU_ROOTS_ALL		(~0UL)
1387

1388 1389
int kvm_pic_set_irq(struct kvm_pic *pic, int irq, int irq_source_id, int level);
void kvm_pic_clear_all(struct kvm_pic *pic, int irq_source_id);
1390

1391 1392
void kvm_inject_nmi(struct kvm_vcpu *vcpu);

1393
int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn);
1394 1395 1396 1397
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
int kvm_mmu_load(struct kvm_vcpu *vcpu);
void kvm_mmu_unload(struct kvm_vcpu *vcpu);
1398
void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
1399 1400
void kvm_mmu_free_roots(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
			ulong roots_to_free);
1401 1402
gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
			   struct x86_exception *exception);
1403 1404 1405 1406 1407 1408 1409 1410
gpa_t kvm_mmu_gva_to_gpa_read(struct kvm_vcpu *vcpu, gva_t gva,
			      struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_fetch(struct kvm_vcpu *vcpu, gva_t gva,
			       struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_write(struct kvm_vcpu *vcpu, gva_t gva,
			       struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_system(struct kvm_vcpu *vcpu, gva_t gva,
				struct x86_exception *exception);
1411

1412 1413
void kvm_vcpu_deactivate_apicv(struct kvm_vcpu *vcpu);

1414 1415
int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);

1416
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u64 error_code,
1417
		       void *insn, int insn_len);
M
Marcelo Tosatti 已提交
1418
void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva);
1419
void kvm_mmu_invpcid_gva(struct kvm_vcpu *vcpu, gva_t gva, unsigned long pcid);
1420
void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu, gpa_t new_cr3, bool skip_tlb_flush);
1421

1422
void kvm_enable_tdp(void);
1423
void kvm_disable_tdp(void);
1424

1425 1426
static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
				  struct x86_exception *exception)
1427 1428 1429 1430
{
	return gpa;
}

1431 1432 1433 1434 1435 1436 1437
static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
{
	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);

	return (struct kvm_mmu_page *)page_private(page);
}

1438
static inline u16 kvm_read_ldt(void)
1439 1440 1441 1442 1443 1444
{
	u16 ldt;
	asm("sldt %0" : "=g"(ldt));
	return ldt;
}

1445
static inline void kvm_load_ldt(u16 sel)
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464
{
	asm("lldt %0" : : "rm"(sel));
}

#ifdef CONFIG_X86_64
static inline unsigned long read_msr(unsigned long msr)
{
	u64 value;

	rdmsrl(msr, value);
	return value;
}
#endif

static inline u32 get_rdx_init_val(void)
{
	return 0x600; /* P6 family */
}

1465 1466 1467 1468 1469
static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
{
	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
}

1470 1471 1472 1473
#define TSS_IOPB_BASE_OFFSET 0x66
#define TSS_BASE_SIZE 0x68
#define TSS_IOPB_SIZE (65536 / 8)
#define TSS_REDIRECTION_SIZE (256 / 8)
1474 1475
#define RMODE_TSS_SIZE							\
	(TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
1476

1477 1478 1479 1480 1481 1482 1483
enum {
	TASK_SWITCH_CALL = 0,
	TASK_SWITCH_IRET = 1,
	TASK_SWITCH_JMP = 2,
	TASK_SWITCH_GATE = 3,
};

1484
#define HF_GIF_MASK		(1 << 0)
A
Alexander Graf 已提交
1485 1486
#define HF_HIF_MASK		(1 << 1)
#define HF_VINTR_MASK		(1 << 2)
1487
#define HF_NMI_MASK		(1 << 3)
1488
#define HF_IRET_MASK		(1 << 4)
1489
#define HF_GUEST_MASK		(1 << 5) /* VCPU is in guest-mode */
1490 1491
#define HF_SMM_MASK		(1 << 6)
#define HF_SMM_INSIDE_NMI_MASK	(1 << 7)
1492

1493 1494 1495 1496 1497
#define __KVM_VCPU_MULTIPLE_ADDRESS_SPACE
#define KVM_ADDRESS_SPACE_NUM 2

#define kvm_arch_vcpu_memslots_id(vcpu) ((vcpu)->arch.hflags & HF_SMM_MASK ? 1 : 0)
#define kvm_memslots_for_spte_role(kvm, role) __kvm_memslots(kvm, (role).smm)
1498

1499 1500 1501 1502 1503
/*
 * Hardware virtualization extension instructions may fault if a
 * reboot turns off virtualization while processes are running.
 * Trap the fault and ignore the instruction if that happens.
 */
1504
asmlinkage void kvm_spurious_fault(void);
1505

1506
#define ____kvm_handle_fault_on_reboot(insn, cleanup_insn)	\
1507
	"666: " insn "\n\t" \
1508
	"668: \n\t"                           \
1509
	".pushsection .fixup, \"ax\" \n" \
1510
	"667: \n\t" \
1511
	cleanup_insn "\n\t"		      \
1512 1513
	"cmpb $0, kvm_rebooting \n\t"	      \
	"jne 668b \n\t"      		      \
1514
	__ASM_SIZE(push) " $666b \n\t"	      \
1515
	"jmp kvm_spurious_fault \n\t"	      \
1516
	".popsection \n\t" \
1517
	_ASM_EXTABLE(666b, 667b)
1518

1519 1520 1521
#define __kvm_handle_fault_on_reboot(insn)		\
	____kvm_handle_fault_on_reboot(insn, "")

1522
#define KVM_ARCH_WANT_MMU_NOTIFIER
1523
int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end);
A
Andres Lagar-Cavilla 已提交
1524
int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
A
Andrea Arcangeli 已提交
1525
int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
1526
int kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
1527
int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v);
1528 1529
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
1530
int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
1531
void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event);
1532
void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu);
1533

1534
int kvm_pv_send_ipi(struct kvm *kvm, unsigned long ipi_bitmap_low,
1535
		    unsigned long ipi_bitmap_high, u32 min,
1536 1537
		    unsigned long icr, int op_64_bit);

A
Avi Kivity 已提交
1538
void kvm_define_shared_msr(unsigned index, u32 msr);
1539
int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
A
Avi Kivity 已提交
1540

1541
u64 kvm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc);
1542
u64 kvm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc);
1543

1544
unsigned long kvm_get_linear_rip(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
1545 1546
bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip);

1547 1548 1549
void kvm_make_mclock_inprogress_request(struct kvm *kvm);
void kvm_make_scan_ioapic_request(struct kvm *kvm);

1550 1551 1552 1553
void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu,
				     struct kvm_async_pf *work);
void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
				 struct kvm_async_pf *work);
G
Gleb Natapov 已提交
1554 1555
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
			       struct kvm_async_pf *work);
1556
bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu);
1557 1558
extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn);

1559 1560
int kvm_skip_emulated_instruction(struct kvm_vcpu *vcpu);
int kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err);
1561
void __kvm_request_immediate_exit(struct kvm_vcpu *vcpu);
1562

1563 1564
int kvm_is_in_guest(void);

1565 1566
int __x86_set_memory_region(struct kvm *kvm, int id, gpa_t gpa, u32 size);
int x86_set_memory_region(struct kvm *kvm, int id, gpa_t gpa, u32 size);
1567 1568
bool kvm_vcpu_is_reset_bsp(struct kvm_vcpu *vcpu);
bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu);
1569

1570 1571 1572
bool kvm_intr_is_single_vcpu(struct kvm *kvm, struct kvm_lapic_irq *irq,
			     struct kvm_vcpu **dest_vcpu);

1573
void kvm_set_msi_irq(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e,
F
Feng Wu 已提交
1574
		     struct kvm_lapic_irq *irq);
P
Paolo Bonzini 已提交
1575

1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587
static inline void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
{
	if (kvm_x86_ops->vcpu_blocking)
		kvm_x86_ops->vcpu_blocking(vcpu);
}

static inline void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu)
{
	if (kvm_x86_ops->vcpu_unblocking)
		kvm_x86_ops->vcpu_unblocking(vcpu);
}

1588
static inline void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu) {}
1589

1590 1591 1592
static inline int kvm_cpu_get_apicid(int mps_cpu)
{
#ifdef CONFIG_X86_LOCAL_APIC
1593
	return default_cpu_present_to_apicid(mps_cpu);
1594 1595 1596 1597 1598 1599
#else
	WARN_ON_ONCE(1);
	return BAD_APICID;
#endif
}

1600 1601 1602
#define put_smstate(type, buf, offset, val)                      \
	*(type *)((buf) + (offset) - 0x7e00) = val

1603 1604 1605
#define GET_SMSTATE(type, buf, offset)		\
	(*(type *)((buf) + (offset) - 0x7e00))

H
H. Peter Anvin 已提交
1606
#endif /* _ASM_X86_KVM_HOST_H */