kvm_host.h 36.2 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 21 22

#include <linux/kvm.h>
#include <linux/kvm_para.h>
23
#include <linux/kvm_types.h>
24
#include <linux/perf_event.h>
25 26
#include <linux/pvclock_gtod.h>
#include <linux/clocksource.h>
F
Feng Wu 已提交
27
#include <linux/irqbypass.h>
28
#include <linux/hyperv.h>
29

30
#include <asm/pvclock-abi.h>
31
#include <asm/desc.h>
S
Sheng Yang 已提交
32
#include <asm/mtrr.h>
33
#include <asm/msr-index.h>
34
#include <asm/asm.h>
35

36
#define KVM_MAX_VCPUS 255
37
#define KVM_SOFT_MAX_VCPUS 160
38
#define KVM_USER_MEM_SLOTS 509
39 40
/* memory slots that are not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 3
41
#define KVM_MEM_SLOTS_NUM (KVM_USER_MEM_SLOTS + KVM_PRIVATE_MEM_SLOTS)
42

43
#define KVM_PIO_PAGE_OFFSET 1
44
#define KVM_COALESCED_MMIO_PAGE_OFFSET 2
45
#define KVM_HALT_POLL_NS_DEFAULT 500000
46

47 48
#define KVM_IRQCHIP_NUM_PINS  KVM_IOAPIC_NUM_PINS

49 50 51 52 53
#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))

N
Nadav Amit 已提交
54
#define CR3_L_MODE_RESERVED_BITS 0xFFFFFF0000000000ULL
55
#define CR3_PCID_INVD		 BIT_64(63)
56 57 58
#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     \
59
			  | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR | X86_CR4_PCIDE \
60
			  | X86_CR4_OSXSAVE | X86_CR4_SMEP | X86_CR4_FSGSBASE \
61
			  | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE | X86_CR4_SMAP))
62 63 64 65

#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR)


66 67

#define INVALID_PAGE (~(hpa_t)0)
68 69
#define VALID_PAGE(x) ((x) != INVALID_PAGE)

70 71
#define UNMAPPED_GVA (~(gpa_t)0)

72
/* KVM Hugepage definitions for x86 */
73
#define KVM_NR_PAGE_SIZES	3
74 75
#define KVM_HPAGE_GFN_SHIFT(x)	(((x) - 1) * 9)
#define KVM_HPAGE_SHIFT(x)	(PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x))
76 77 78
#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
79

80 81 82 83 84 85 86
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));
}

87 88
#define KVM_PERMILLE_MMU_PAGES 20
#define KVM_MIN_ALLOC_MMU_PAGES 64
89 90
#define KVM_MMU_HASH_SHIFT 10
#define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT)
91 92
#define KVM_MIN_FREE_MMU_PAGES 5
#define KVM_REFILL_PAGES 25
93
#define KVM_MAX_CPUID_ENTRIES 80
S
Sheng Yang 已提交
94
#define KVM_NR_FIXED_MTRR_REGION 88
95
#define KVM_NR_VAR_MTRR 8
96

97 98
#define ASYNC_PF_PER_VCPU 64

99
enum kvm_reg {
100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
	VCPU_REGS_RAX = 0,
	VCPU_REGS_RCX = 1,
	VCPU_REGS_RDX = 2,
	VCPU_REGS_RBX = 3,
	VCPU_REGS_RSP = 4,
	VCPU_REGS_RBP = 5,
	VCPU_REGS_RSI = 6,
	VCPU_REGS_RDI = 7,
#ifdef CONFIG_X86_64
	VCPU_REGS_R8 = 8,
	VCPU_REGS_R9 = 9,
	VCPU_REGS_R10 = 10,
	VCPU_REGS_R11 = 11,
	VCPU_REGS_R12 = 12,
	VCPU_REGS_R13 = 13,
	VCPU_REGS_R14 = 14,
	VCPU_REGS_R15 = 15,
#endif
118
	VCPU_REGS_RIP,
119 120 121
	NR_VCPU_REGS
};

A
Avi Kivity 已提交
122 123
enum kvm_reg_ex {
	VCPU_EXREG_PDPTR = NR_VCPU_REGS,
124
	VCPU_EXREG_CR3,
A
Avi Kivity 已提交
125
	VCPU_EXREG_RFLAGS,
A
Avi Kivity 已提交
126
	VCPU_EXREG_SEGMENTS,
A
Avi Kivity 已提交
127 128
};

129
enum {
130
	VCPU_SREG_ES,
131
	VCPU_SREG_CS,
132
	VCPU_SREG_SS,
133 134 135 136 137 138 139
	VCPU_SREG_DS,
	VCPU_SREG_FS,
	VCPU_SREG_GS,
	VCPU_SREG_TR,
	VCPU_SREG_LDTR,
};

140
#include <asm/kvm_emulate.h>
141

142 143
#define KVM_NR_MEM_OBJS 40

144 145 146 147
#define KVM_NR_DB_REGS	4

#define DR6_BD		(1 << 13)
#define DR6_BS		(1 << 14)
148 149 150 151
#define DR6_RTM		(1 << 16)
#define DR6_FIXED_1	0xfffe0ff0
#define DR6_INIT	0xffff0ff0
#define DR6_VOLATILE	0x0001e00f
152 153 154 155 156

#define DR7_BP_EN_MASK	0x000000ff
#define DR7_GE		(1 << 9)
#define DR7_GD		(1 << 13)
#define DR7_FIXED_1	0x00000400
157
#define DR7_VOLATILE	0xffff2bff
158

159 160 161 162 163 164 165 166 167 168 169 170
#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

#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)

171 172
/* apic attention bits */
#define KVM_APIC_CHECK_VAPIC	0
173 174 175 176 177 178 179
/*
 * 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
180

F
Feng Wu 已提交
181 182
struct kvm_kernel_irq_routing_entry;

183 184 185 186 187 188 189 190 191 192 193 194
/*
 * 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];
};

union kvm_mmu_page_role {
	unsigned word;
	struct {
195
		unsigned level:4;
196
		unsigned cr4_pae:1;
197
		unsigned quadrant:2;
198
		unsigned direct:1;
199
		unsigned access:3;
200
		unsigned invalid:1;
201
		unsigned nxe:1;
202
		unsigned cr0_wp:1;
203
		unsigned smep_andnot_wp:1;
204
		unsigned smap_andnot_wp:1;
205 206 207 208 209 210 211 212 213
		unsigned :8;

		/*
		 * 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;
214 215 216
	};
};

217 218 219 220
struct kvm_rmap_head {
	unsigned long val;
};

221 222 223 224 225 226 227 228 229 230 231 232 233 234
struct kvm_mmu_page {
	struct list_head link;
	struct hlist_node hash_link;

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

	u64 *spt;
	/* hold the gfn of each spte inside spt */
	gfn_t *gfns;
235
	bool unsync;
236
	int root_count;          /* Currently serving as active root */
237
	unsigned int unsync_children;
238
	struct kvm_rmap_head parent_ptes; /* rmap pointers to parent sptes */
239 240

	/* The page is obsolete if mmu_valid_gen != kvm->arch.mmu_valid_gen.  */
241
	unsigned long mmu_valid_gen;
242

243
	DECLARE_BITMAP(unsync_child_bitmap, 512);
244 245

#ifdef CONFIG_X86_32
246 247 248 249
	/*
	 * Used out of the mmu-lock to avoid reading spte values while an
	 * update is in progress; see the comments in __get_spte_lockless().
	 */
250 251 252
	int clear_spte_count;
#endif

253
	/* Number of writes since the last time traversal visited this page.  */
254
	int write_flooding_count;
255 256
};

257 258 259 260 261 262 263
struct kvm_pio_request {
	unsigned long count;
	int in;
	int port;
	int size;
};

264 265 266 267 268
struct rsvd_bits_validate {
	u64 rsvd_bits_mask[2][4];
	u64 bad_mt_xwr;
};

269 270 271 272 273 274
/*
 * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
 * 32-bit).  The kvm_mmu structure abstracts the details of the current mmu
 * mode.
 */
struct kvm_mmu {
275
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
276
	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
277
	u64 (*get_pdptr)(struct kvm_vcpu *vcpu, int index);
278 279
	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
			  bool prefault);
280 281
	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
				  struct x86_exception *fault);
282
	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
283
			    struct x86_exception *exception);
284 285
	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
			       struct x86_exception *exception);
286
	int (*sync_page)(struct kvm_vcpu *vcpu,
287
			 struct kvm_mmu_page *sp);
M
Marcelo Tosatti 已提交
288
	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva);
289
	void (*update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
290
			   u64 *spte, const void *pte);
291 292 293
	hpa_t root_hpa;
	int root_level;
	int shadow_root_level;
294
	union kvm_mmu_page_role base_role;
295
	bool direct_map;
296

297 298 299 300 301 302 303
	/*
	 * Bitmap; bit set = permission fault
	 * Byte index: page fault error code [4:1]
	 * Bit index: pte permissions in ACC_* format
	 */
	u8 permissions[16];

304
	u64 *pae_root;
305
	u64 *lm_root;
306 307 308 309 310 311 312 313

	/*
	 * 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;

314
	struct rsvd_bits_validate guest_rsvd_check;
315

A
Avi Kivity 已提交
316 317 318 319 320 321 322
	/*
	 * Bitmap: bit set = last pte in walk
	 * index[0:1]: level (zero-based)
	 * index[2]: pte.ps
	 */
	u8 last_pte_bitmap;

323 324
	bool nx;

325
	u64 pdptrs[4]; /* pae */
326 327
};

328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
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;
352
	u64 reserved_bits;
353
	u8 version;
354 355
	struct kvm_pmc gp_counters[INTEL_PMC_MAX_GENERIC];
	struct kvm_pmc fixed_counters[INTEL_PMC_MAX_FIXED];
356 357 358 359
	struct irq_work irq_work;
	u64 reprogram_pmi;
};

360 361
struct kvm_pmu_ops;

362 363
enum {
	KVM_DEBUGREG_BP_ENABLED = 1,
364
	KVM_DEBUGREG_WONT_EXIT = 2,
365
	KVM_DEBUGREG_RELOAD = 4,
366 367
};

368 369 370
struct kvm_mtrr_range {
	u64 base;
	u64 mask;
X
Xiao Guangrong 已提交
371
	struct list_head node;
372 373
};

374
struct kvm_mtrr {
375
	struct kvm_mtrr_range var_ranges[KVM_NR_VAR_MTRR];
376
	mtrr_type fixed_ranges[KVM_NR_FIXED_MTRR_REGION];
377
	u64 deftype;
X
Xiao Guangrong 已提交
378 379

	struct list_head head;
380 381
};

382 383 384 385 386 387 388 389 390 391 392 393 394
/* 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;
};

395 396 397
/* Hyper-V per vcpu emulation context */
struct kvm_vcpu_hv {
	u64 hv_vapic;
398
	s64 runtime_offset;
399
	struct kvm_vcpu_hv_synic synic;
A
Andrey Smetanin 已提交
400
	struct kvm_hyperv_exit exit;
401 402
};

403
struct kvm_vcpu_arch {
404 405 406 407 408 409 410
	/*
	 * 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;
411 412

	unsigned long cr0;
413
	unsigned long cr0_guest_owned_bits;
414 415 416
	unsigned long cr2;
	unsigned long cr3;
	unsigned long cr4;
417
	unsigned long cr4_guest_owned_bits;
418
	unsigned long cr8;
419
	u32 hflags;
420
	u64 efer;
421 422
	u64 apic_base;
	struct kvm_lapic *apic;    /* kernel irqchip context */
423
	bool apicv_active;
424
	DECLARE_BITMAP(ioapic_handled_vectors, 256);
425
	unsigned long apic_attention;
426
	int32_t apic_arb_prio;
427 428
	int mp_state;
	u64 ia32_misc_enable_msr;
P
Paolo Bonzini 已提交
429
	u64 smbase;
430
	bool tpr_access_reporting;
W
Wanpeng Li 已提交
431
	u64 ia32_xss;
432

433 434 435 436 437 438 439
	/*
	 * 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.
	 */
440
	struct kvm_mmu mmu;
441

442 443 444 445 446 447 448 449 450 451
	/*
	 * 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;

452 453 454 455 456 457
	/*
	 * Pointer to the mmu context currently used for
	 * gva_to_gpa translations.
	 */
	struct kvm_mmu *walk_mmu;

458
	struct kvm_mmu_memory_cache mmu_pte_list_desc_cache;
459 460 461
	struct kvm_mmu_memory_cache mmu_page_cache;
	struct kvm_mmu_memory_cache mmu_page_header_cache;

S
Sheng Yang 已提交
462
	struct fpu guest_fpu;
463
	bool eager_fpu;
464
	u64 xcr0;
465
	u64 guest_supported_xcr0;
466
	u32 guest_xstate_size;
467 468 469 470

	struct kvm_pio_request pio;
	void *pio_data;

471 472
	u8 event_exit_inst_len;

473 474 475
	struct kvm_queued_exception {
		bool pending;
		bool has_error_code;
476
		bool reinject;
477 478 479 480
		u8 nr;
		u32 error_code;
	} exception;

A
Avi Kivity 已提交
481 482
	struct kvm_queued_interrupt {
		bool pending;
483
		bool soft;
A
Avi Kivity 已提交
484 485 486
		u8 nr;
	} interrupt;

487 488 489
	int halt_request; /* real mode on Intel only */

	int cpuid_nent;
490
	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
491 492 493

	int maxphyaddr;

494 495 496
	/* emulate context */

	struct x86_emulate_ctxt emulate_ctxt;
497 498
	bool emulate_regs_need_sync_to_vcpu;
	bool emulate_regs_need_sync_from_vcpu;
499
	int (*complete_userspace_io)(struct kvm_vcpu *vcpu);
500 501

	gpa_t time;
502
	struct pvclock_vcpu_time_info hv_clock;
Z
Zachary Amsden 已提交
503
	unsigned int hw_tsc_khz;
504 505
	struct gfn_to_hva_cache pv_time;
	bool pv_time_enabled;
506 507
	/* set guest stopped flag in pvclock flags field */
	bool pvclock_set_guest_stopped_request;
G
Glauber Costa 已提交
508 509 510 511 512 513 514 515 516

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

517
	u64 last_guest_tsc;
518
	u64 last_host_tsc;
519
	u64 tsc_offset_adjustment;
520 521
	u64 this_tsc_nsec;
	u64 this_tsc_write;
T
Tomasz Grabiec 已提交
522
	u64 this_tsc_generation;
Z
Zachary Amsden 已提交
523
	bool tsc_catchup;
524 525 526 527
	bool tsc_always_catchup;
	s8 virtual_tsc_shift;
	u32 virtual_tsc_mult;
	u32 virtual_tsc_khz;
W
Will Auld 已提交
528
	s64 ia32_tsc_adjust_msr;
529
	u64 tsc_scaling_ratio;
530

A
Avi Kivity 已提交
531 532 533
	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 */
534
	bool smi_pending;    /* SMI queued after currently running handler */
A
Avi Kivity 已提交
535

536
	struct kvm_mtrr mtrr_state;
537
	u64 pat;
538

539
	unsigned switch_db_regs;
540 541 542 543
	unsigned long db[KVM_NR_DB_REGS];
	unsigned long dr6;
	unsigned long dr7;
	unsigned long eff_db[KVM_NR_DB_REGS];
544
	unsigned long guest_debug_dr7;
H
Huang Ying 已提交
545 546 547 548 549

	u64 mcg_cap;
	u64 mcg_status;
	u64 mcg_ctl;
	u64 *mce_banks;
550

551 552 553 554
	/* Cache MMIO info */
	u64 mmio_gva;
	unsigned access;
	gfn_t mmio_gfn;
555
	u64 mmio_gen;
556

557 558
	struct kvm_pmu pmu;

559 560
	/* used for guest single stepping over the given code position */
	unsigned long singlestep_rip;
J
Jan Kiszka 已提交
561

562
	struct kvm_vcpu_hv hyperv;
563 564

	cpumask_var_t wbinvd_dirty_mask;
565

566 567 568
	unsigned long last_retry_eip;
	unsigned long last_retry_addr;

569 570 571
	struct {
		bool halted;
		gfn_t gfns[roundup_pow_of_two(ASYNC_PF_PER_VCPU)];
572 573
		struct gfn_to_hva_cache data;
		u64 msr_val;
574
		u32 id;
575
		bool send_user_only;
576
	} apf;
577 578 579 580 581 582

	/* OSVW MSRs (AMD only) */
	struct {
		u64 length;
		u64 status;
	} osvw;
583 584 585 586 587

	struct {
		u64 msr_val;
		struct gfn_to_hva_cache data;
	} pv_eoi;
588 589 590 591 592 593 594

	/*
	 * 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;
595 596 597

	/* set at EPT violation at this point */
	unsigned long exit_qualification;
598 599 600 601 602

	/* pv related host specific info */
	struct {
		bool pv_unhalted;
	} pv;
603 604

	int pending_ioapic_eoi;
605
	int pending_external_vector;
606 607
};

608 609 610 611 612
struct kvm_lpage_info {
	int write_count;
};

struct kvm_arch_memory_slot {
613
	struct kvm_rmap_head *rmap[KVM_NR_PAGE_SIZES];
614 615 616
	struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1];
};

617 618 619 620 621 622 623 624 625 626 627
/*
 * 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

628 629
struct kvm_apic_map {
	struct rcu_head rcu;
630
	u8 mode;
631 632 633 634 635
	struct kvm_lapic *phys_map[256];
	/* first index is cluster id second is cpu id in a cluster */
	struct kvm_lapic *logical_map[16][16];
};

636 637 638 639 640
/* Hyper-V emulation context */
struct kvm_hv {
	u64 hv_guest_os_id;
	u64 hv_hypercall;
	u64 hv_tsc_page;
641 642 643 644

	/* Hyper-v based guest crash (NT kernel bugcheck) parameters */
	u64 hv_crash_param[HV_X64_MSR_CRASH_PARAMS];
	u64 hv_crash_ctl;
645 646
};

647
struct kvm_arch {
648
	unsigned int n_used_mmu_pages;
649
	unsigned int n_requested_mmu_pages;
650
	unsigned int n_max_mmu_pages;
651
	unsigned int indirect_shadow_pages;
652
	unsigned long mmu_valid_gen;
653 654 655 656 657
	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
	/*
	 * Hash table of struct kvm_mmu_page.
	 */
	struct list_head active_mmu_pages;
658 659
	struct list_head zapped_obsolete_pages;

B
Ben-Ami Yassour 已提交
660
	struct list_head assigned_dev_head;
J
Joerg Roedel 已提交
661
	struct iommu_domain *iommu_domain;
662
	bool iommu_noncoherent;
663 664
#define __KVM_HAVE_ARCH_NONCOHERENT_DMA
	atomic_t noncoherent_dma_count;
665 666
#define __KVM_HAVE_ARCH_ASSIGNED_DEVICE
	atomic_t assigned_device_count;
667 668
	struct kvm_pic *vpic;
	struct kvm_ioapic *vioapic;
S
Sheng Yang 已提交
669
	struct kvm_pit *vpit;
670
	atomic_t vapics_in_nmi_mode;
671 672
	struct mutex apic_map_lock;
	struct kvm_apic_map *apic_map;
673 674

	unsigned int tss_addr;
675
	bool apic_access_page_done;
676 677

	gpa_t wall_clock;
678 679

	bool ept_identity_pagetable_done;
680
	gpa_t ept_identity_map_addr;
681 682

	unsigned long irq_sources_bitmap;
683
	s64 kvmclock_offset;
684
	raw_spinlock_t tsc_write_lock;
Z
Zachary Amsden 已提交
685 686
	u64 last_tsc_nsec;
	u64 last_tsc_write;
687
	u32 last_tsc_khz;
688 689 690
	u64 cur_tsc_nsec;
	u64 cur_tsc_write;
	u64 cur_tsc_offset;
T
Tomasz Grabiec 已提交
691
	u64 cur_tsc_generation;
692
	int nr_vcpus_matched_tsc;
E
Ed Swierk 已提交
693

694 695 696 697
	spinlock_t pvclock_gtod_sync_lock;
	bool use_master_clock;
	u64 master_kernel_ns;
	cycle_t master_cycle_now;
698
	struct delayed_work kvmclock_update_work;
699
	struct delayed_work kvmclock_sync_work;
700

E
Ed Swierk 已提交
701
	struct kvm_xen_hvm_config xen_hvm_config;
702

703 704 705
	/* reads protected by irq_srcu, writes by irq_lock */
	struct hlist_head mask_notifier_list;

706
	struct kvm_hv hyperv;
707 708 709 710

	#ifdef CONFIG_KVM_MMU_AUDIT
	int audit_point;
	#endif
711 712

	bool boot_vcpu_runs_old_kvmclock;
713
	u32 bsp_vcpu_id;
714 715

	u64 disabled_quirks;
716 717

	bool irqchip_split;
718
	u8 nr_reserved_ioapic_pins;
719 720
};

721 722 723 724 725 726 727
struct kvm_vm_stat {
	u32 mmu_shadow_zapped;
	u32 mmu_pte_write;
	u32 mmu_pte_updated;
	u32 mmu_pde_zapped;
	u32 mmu_flooded;
	u32 mmu_recycled;
A
Avi Kivity 已提交
728
	u32 mmu_cache_miss;
729
	u32 mmu_unsync;
730
	u32 remote_tlb_flush;
M
Marcelo Tosatti 已提交
731
	u32 lpages;
732 733
};

734 735 736 737 738 739 740 741 742 743 744
struct kvm_vcpu_stat {
	u32 pf_fixed;
	u32 pf_guest;
	u32 tlb_flush;
	u32 invlpg;

	u32 exits;
	u32 io_exits;
	u32 mmio_exits;
	u32 signal_exits;
	u32 irq_window_exits;
745
	u32 nmi_window_exits;
746
	u32 halt_exits;
747
	u32 halt_successful_poll;
748
	u32 halt_attempted_poll;
749 750 751 752 753 754 755 756
	u32 halt_wakeup;
	u32 request_irq_exits;
	u32 irq_exits;
	u32 host_state_reload;
	u32 efer_reload;
	u32 fpu_reload;
	u32 insn_emulation;
	u32 insn_emulation_fail;
A
Amit Shah 已提交
757
	u32 hypercalls;
758
	u32 irq_injections;
759
	u32 nmi_injections;
760
};
761

762 763
struct x86_instruction_info;

764 765 766 767 768 769
struct msr_data {
	bool host_initiated;
	u32 index;
	u64 data;
};

P
Paolo Bonzini 已提交
770 771
struct kvm_lapic_irq {
	u32 vector;
772 773 774 775
	u16 delivery_mode;
	u16 dest_mode;
	bool level;
	u16 trig_mode;
P
Paolo Bonzini 已提交
776 777
	u32 shorthand;
	u32 dest_id;
778
	bool msi_redir_hint;
P
Paolo Bonzini 已提交
779 780
};

781 782 783
struct kvm_x86_ops {
	int (*cpu_has_kvm_support)(void);          /* __init */
	int (*disabled_by_bios)(void);             /* __init */
784 785
	int (*hardware_enable)(void);
	void (*hardware_disable)(void);
786 787 788
	void (*check_processor_compatibility)(void *rtn);
	int (*hardware_setup)(void);               /* __init */
	void (*hardware_unsetup)(void);            /* __exit */
789
	bool (*cpu_has_accelerated_tpr)(void);
790
	bool (*cpu_has_high_real_mode_segbase)(void);
791
	void (*cpuid_update)(struct kvm_vcpu *vcpu);
792 793 794 795

	/* Create, but do not attach this VCPU */
	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
	void (*vcpu_free)(struct kvm_vcpu *vcpu);
796
	void (*vcpu_reset)(struct kvm_vcpu *vcpu, bool init_event);
797 798 799 800 801

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

802
	void (*update_bp_intercept)(struct kvm_vcpu *vcpu);
803
	int (*get_msr)(struct kvm_vcpu *vcpu, struct msr_data *msr);
804
	int (*set_msr)(struct kvm_vcpu *vcpu, struct msr_data *msr);
805 806 807
	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
	void (*get_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
808
	int (*get_cpl)(struct kvm_vcpu *vcpu);
809 810 811
	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);
812
	void (*decache_cr0_guest_bits)(struct kvm_vcpu *vcpu);
813
	void (*decache_cr3)(struct kvm_vcpu *vcpu);
814 815 816
	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);
817
	int (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
818
	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
819 820 821 822
	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
823 824
	u64 (*get_dr6)(struct kvm_vcpu *vcpu);
	void (*set_dr6)(struct kvm_vcpu *vcpu, unsigned long value);
825
	void (*sync_dirty_debug_regs)(struct kvm_vcpu *vcpu);
826
	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
827
	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
828 829
	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
830
	void (*fpu_activate)(struct kvm_vcpu *vcpu);
831
	void (*fpu_deactivate)(struct kvm_vcpu *vcpu);
832 833 834

	void (*tlb_flush)(struct kvm_vcpu *vcpu);

A
Avi Kivity 已提交
835 836
	void (*run)(struct kvm_vcpu *vcpu);
	int (*handle_exit)(struct kvm_vcpu *vcpu);
837
	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
838
	void (*set_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
839
	u32 (*get_interrupt_shadow)(struct kvm_vcpu *vcpu);
840 841
	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
				unsigned char *hypercall_addr);
842
	void (*set_irq)(struct kvm_vcpu *vcpu);
843
	void (*set_nmi)(struct kvm_vcpu *vcpu);
844
	void (*queue_exception)(struct kvm_vcpu *vcpu, unsigned nr,
845 846
				bool has_error_code, u32 error_code,
				bool reinject);
A
Avi Kivity 已提交
847
	void (*cancel_injection)(struct kvm_vcpu *vcpu);
848
	int (*interrupt_allowed)(struct kvm_vcpu *vcpu);
849
	int (*nmi_allowed)(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
850 851
	bool (*get_nmi_mask)(struct kvm_vcpu *vcpu);
	void (*set_nmi_mask)(struct kvm_vcpu *vcpu, bool masked);
852 853
	void (*enable_nmi_window)(struct kvm_vcpu *vcpu);
	void (*enable_irq_window)(struct kvm_vcpu *vcpu);
854
	void (*update_cr8_intercept)(struct kvm_vcpu *vcpu, int tpr, int irr);
855 856
	bool (*get_enable_apicv)(void);
	void (*refresh_apicv_exec_ctrl)(struct kvm_vcpu *vcpu);
857 858
	void (*hwapic_irr_update)(struct kvm_vcpu *vcpu, int max_irr);
	void (*hwapic_isr_update)(struct kvm *kvm, int isr);
859
	void (*load_eoi_exitmap)(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap);
860
	void (*set_virtual_x2apic_mode)(struct kvm_vcpu *vcpu, bool set);
861
	void (*set_apic_access_page_addr)(struct kvm_vcpu *vcpu, hpa_t hpa);
862 863
	void (*deliver_posted_interrupt)(struct kvm_vcpu *vcpu, int vector);
	void (*sync_pir_to_irr)(struct kvm_vcpu *vcpu);
864
	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
865
	int (*get_tdp_level)(void);
866
	u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio);
867
	int (*get_lpage_level)(void);
868
	bool (*rdtscp_supported)(void);
869
	bool (*invpcid_supported)(void);
870
	void (*adjust_tsc_offset_guest)(struct kvm_vcpu *vcpu, s64 adjustment);
871

872 873
	void (*set_tdp_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);

874 875
	void (*set_supported_cpuid)(u32 func, struct kvm_cpuid_entry2 *entry);

876 877
	bool (*has_wbinvd_exit)(void);

W
Will Auld 已提交
878
	u64 (*read_tsc_offset)(struct kvm_vcpu *vcpu);
879 880
	void (*write_tsc_offset)(struct kvm_vcpu *vcpu, u64 offset);

881
	u64 (*read_l1_tsc)(struct kvm_vcpu *vcpu, u64 host_tsc);
882

883
	void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2);
884 885 886 887

	int (*check_intercept)(struct kvm_vcpu *vcpu,
			       struct x86_instruction_info *info,
			       enum x86_intercept_stage stage);
888
	void (*handle_external_intr)(struct kvm_vcpu *vcpu);
889
	bool (*mpx_supported)(void);
890
	bool (*xsaves_supported)(void);
891 892

	int (*check_nested_events)(struct kvm_vcpu *vcpu, bool external_intr);
893 894

	void (*sched_in)(struct kvm_vcpu *kvm, int cpu);
895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919

	/*
	 * 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);
920 921
	/* pmu operations of sub-arch */
	const struct kvm_pmu_ops *pmu_ops;
922

923 924 925 926 927 928 929 930 931 932 933
	/*
	 * 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);
934 935
	int (*update_pi_irte)(struct kvm *kvm, unsigned int host_irq,
			      uint32_t guest_irq, bool set);
936 937
};

938
struct kvm_arch_async_pf {
939
	u32 token;
940
	gfn_t gfn;
X
Xiao Guangrong 已提交
941
	unsigned long cr3;
942
	bool direct_map;
943 944
};

945 946
extern struct kvm_x86_ops *kvm_x86_ops;

947 948 949 950 951
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);
952
void kvm_mmu_setup(struct kvm_vcpu *vcpu);
S
Sheng Yang 已提交
953
void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
954
		u64 dirty_mask, u64 nx_mask, u64 x_mask);
955

956
void kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
957 958
void kvm_mmu_slot_remove_write_access(struct kvm *kvm,
				      struct kvm_memory_slot *memslot);
959
void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm,
960
				   const struct kvm_memory_slot *memslot);
961 962 963 964 965 966 967 968 969
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);
970
void kvm_mmu_zap_all(struct kvm *kvm);
971
void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, struct kvm_memslots *slots);
972
unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
973 974
void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);

975
int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3);
976

977
int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
978
			  const void *val, int bytes);
979

980 981 982 983 984 985 986 987 988 989 990 991 992
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);

993
extern bool tdp_enabled;
994

995 996
u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu);

997 998 999 1000
/* 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;
1001 1002 1003 1004
/* 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;
1005

1006
enum emulation_result {
P
Paolo Bonzini 已提交
1007 1008
	EMULATE_DONE,         /* no further processing */
	EMULATE_USER_EXIT,    /* kvm_run ready for userspace exit */
1009 1010 1011
	EMULATE_FAIL,         /* can't emulate this instruction */
};

1012 1013
#define EMULTYPE_NO_DECODE	    (1 << 0)
#define EMULTYPE_TRAP_UD	    (1 << 1)
1014
#define EMULTYPE_SKIP		    (1 << 2)
1015
#define EMULTYPE_RETRY		    (1 << 3)
1016
#define EMULTYPE_NO_REEXECUTE	    (1 << 4)
1017 1018
int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2,
			    int emulation_type, void *insn, int insn_len);
1019 1020 1021 1022

static inline int emulate_instruction(struct kvm_vcpu *vcpu,
			int emulation_type)
{
1023
	return x86_emulate_instruction(vcpu, 0, emulation_type, NULL, 0);
1024 1025
}

1026
void kvm_enable_efer_bits(u64);
1027
bool kvm_valid_efer(struct kvm_vcpu *vcpu, u64 efer);
1028
int kvm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr);
1029
int kvm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr);
1030 1031 1032

struct x86_emulate_ctxt;

1033
int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port);
1034 1035
void kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
int kvm_emulate_halt(struct kvm_vcpu *vcpu);
1036
int kvm_vcpu_halt(struct kvm_vcpu *vcpu);
1037
int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu);
1038

1039
void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg);
1040
int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg);
1041
void kvm_vcpu_deliver_sipi_vector(struct kvm_vcpu *vcpu, u8 vector);
1042

1043 1044
int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int idt_index,
		    int reason, bool has_error_code, u32 error_code);
1045

1046
int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
1047
int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3);
1048
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4);
A
Andre Przywara 已提交
1049
int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8);
1050 1051
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);
1052 1053
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu);
void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
1054
void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
1055
int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr);
1056

1057
int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr);
1058
int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr);
1059

1060 1061
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
1062
bool kvm_rdpmc(struct kvm_vcpu *vcpu);
1063

1064 1065
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
1066 1067
void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
1068
void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
1069 1070 1071
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);
1072
bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
1073
bool kvm_require_dr(struct kvm_vcpu *vcpu, int dr);
1074

1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088
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);
}

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);
1089

1090 1091
void kvm_inject_nmi(struct kvm_vcpu *vcpu);

1092
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
1093
		       const u8 *new, int bytes);
1094
int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn);
1095 1096 1097 1098
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);
1099
void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
1100 1101
gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
			   struct x86_exception *exception);
1102 1103 1104 1105 1106 1107 1108 1109
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);
1110

1111 1112
void kvm_vcpu_deactivate_apicv(struct kvm_vcpu *vcpu);

1113 1114
int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);

1115 1116
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code,
		       void *insn, int insn_len);
M
Marcelo Tosatti 已提交
1117
void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva);
1118
void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu);
1119

1120
void kvm_enable_tdp(void);
1121
void kvm_disable_tdp(void);
1122

1123 1124
static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access,
				  struct x86_exception *exception)
1125 1126 1127 1128
{
	return gpa;
}

1129 1130 1131 1132 1133 1134 1135
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);
}

1136
static inline u16 kvm_read_ldt(void)
1137 1138 1139 1140 1141 1142
{
	u16 ldt;
	asm("sldt %0" : "=g"(ldt));
	return ldt;
}

1143
static inline void kvm_load_ldt(u16 sel)
1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
{
	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 */
}

1163 1164 1165 1166 1167
static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
{
	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
}

1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
static inline u64 get_canonical(u64 la)
{
	return ((int64_t)la << 16) >> 16;
}

static inline bool is_noncanonical_address(u64 la)
{
#ifdef CONFIG_X86_64
	return get_canonical(la) != la;
#else
	return false;
#endif
}

1182 1183 1184 1185
#define TSS_IOPB_BASE_OFFSET 0x66
#define TSS_BASE_SIZE 0x68
#define TSS_IOPB_SIZE (65536 / 8)
#define TSS_REDIRECTION_SIZE (256 / 8)
1186 1187
#define RMODE_TSS_SIZE							\
	(TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
1188

1189 1190 1191 1192 1193 1194 1195
enum {
	TASK_SWITCH_CALL = 0,
	TASK_SWITCH_IRET = 1,
	TASK_SWITCH_JMP = 2,
	TASK_SWITCH_GATE = 3,
};

1196
#define HF_GIF_MASK		(1 << 0)
A
Alexander Graf 已提交
1197 1198
#define HF_HIF_MASK		(1 << 1)
#define HF_VINTR_MASK		(1 << 2)
1199
#define HF_NMI_MASK		(1 << 3)
1200
#define HF_IRET_MASK		(1 << 4)
1201
#define HF_GUEST_MASK		(1 << 5) /* VCPU is in guest-mode */
1202 1203
#define HF_SMM_MASK		(1 << 6)
#define HF_SMM_INSIDE_NMI_MASK	(1 << 7)
1204

1205 1206 1207 1208 1209
#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)
1210

1211 1212 1213 1214 1215
/*
 * 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.
 */
1216
asmlinkage void kvm_spurious_fault(void);
1217

1218
#define ____kvm_handle_fault_on_reboot(insn, cleanup_insn)	\
1219
	"666: " insn "\n\t" \
1220
	"668: \n\t"                           \
1221
	".pushsection .fixup, \"ax\" \n" \
1222
	"667: \n\t" \
1223
	cleanup_insn "\n\t"		      \
1224 1225
	"cmpb $0, kvm_rebooting \n\t"	      \
	"jne 668b \n\t"      		      \
1226
	__ASM_SIZE(push) " $666b \n\t"	      \
1227
	"call kvm_spurious_fault \n\t"	      \
1228
	".popsection \n\t" \
1229
	_ASM_EXTABLE(666b, 667b)
1230

1231 1232 1233
#define __kvm_handle_fault_on_reboot(insn)		\
	____kvm_handle_fault_on_reboot(insn, "")

1234 1235
#define KVM_ARCH_WANT_MMU_NOTIFIER
int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
1236
int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end);
A
Andres Lagar-Cavilla 已提交
1237
int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
A
Andrea Arcangeli 已提交
1238
int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
1239
void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
1240
int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v);
1241 1242
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
1243
int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
1244
void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event);
1245
void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu);
1246 1247
void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
					   unsigned long address);
1248

A
Avi Kivity 已提交
1249
void kvm_define_shared_msr(unsigned index, u32 msr);
1250
int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
A
Avi Kivity 已提交
1251

1252
u64 kvm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc);
1253
u64 kvm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc);
1254

1255
unsigned long kvm_get_linear_rip(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
1256 1257
bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip);

1258 1259 1260 1261
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
1262 1263
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
			       struct kvm_async_pf *work);
1264
bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu);
1265 1266
extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn);

1267 1268
void kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err);

1269 1270
int kvm_is_in_guest(void);

1271 1272
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);
1273 1274
bool kvm_vcpu_is_reset_bsp(struct kvm_vcpu *vcpu);
bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu);
1275

1276 1277 1278
bool kvm_intr_is_single_vcpu(struct kvm *kvm, struct kvm_lapic_irq *irq,
			     struct kvm_vcpu **dest_vcpu);

F
Feng Wu 已提交
1279 1280
void kvm_set_msi_irq(struct kvm_kernel_irq_routing_entry *e,
		     struct kvm_lapic_irq *irq);
P
Paolo Bonzini 已提交
1281

1282 1283 1284
static inline void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu) {}
static inline void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu) {}

H
H. Peter Anvin 已提交
1285
#endif /* _ASM_X86_KVM_HOST_H */