kvm_host.h 23.6 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 20 21

#include <linux/kvm.h>
#include <linux/kvm_para.h>
22
#include <linux/kvm_types.h>
23

24
#include <asm/pvclock-abi.h>
25
#include <asm/desc.h>
S
Sheng Yang 已提交
26
#include <asm/mtrr.h>
27
#include <asm/msr-index.h>
28

A
Avi Kivity 已提交
29
#define KVM_MAX_VCPUS 64
30 31 32 33 34
#define KVM_MEMORY_SLOTS 32
/* memory slots that does not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 4

#define KVM_PIO_PAGE_OFFSET 1
35
#define KVM_COALESCED_MMIO_PAGE_OFFSET 2
36

37 38
#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
39 40
#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS |	\
				  0xFFFFFF0000000000ULL)
41 42

#define INVALID_PAGE (~(hpa_t)0)
43 44
#define VALID_PAGE(x) ((x) != INVALID_PAGE)

45 46
#define UNMAPPED_GVA (~(gpa_t)0)

47
/* KVM Hugepage definitions for x86 */
48
#define KVM_NR_PAGE_SIZES	3
49 50
#define KVM_HPAGE_GFN_SHIFT(x)	(((x) - 1) * 9)
#define KVM_HPAGE_SHIFT(x)	(PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x))
51 52 53
#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
54

55
#define DE_VECTOR 0
J
Jan Kiszka 已提交
56
#define DB_VECTOR 1
57 58 59
#define BP_VECTOR 3
#define OF_VECTOR 4
#define BR_VECTOR 5
60 61 62 63 64 65 66 67
#define UD_VECTOR 6
#define NM_VECTOR 7
#define DF_VECTOR 8
#define TS_VECTOR 10
#define NP_VECTOR 11
#define SS_VECTOR 12
#define GP_VECTOR 13
#define PF_VECTOR 14
68
#define MF_VECTOR 16
69
#define MC_VECTOR 18
70 71 72 73 74 75

#define SELECTOR_TI_MASK (1 << 2)
#define SELECTOR_RPL_MASK 0x03

#define IOPL_SHIFT 12

76 77
#define KVM_PERMILLE_MMU_PAGES 20
#define KVM_MIN_ALLOC_MMU_PAGES 64
78 79
#define KVM_MMU_HASH_SHIFT 10
#define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT)
80 81
#define KVM_MIN_FREE_MMU_PAGES 5
#define KVM_REFILL_PAGES 25
82
#define KVM_MAX_CPUID_ENTRIES 80
S
Sheng Yang 已提交
83
#define KVM_NR_FIXED_MTRR_REGION 88
A
Avi Kivity 已提交
84
#define KVM_NR_VAR_MTRR 8
85

86 87
#define ASYNC_PF_PER_VCPU 64

88 89 90
extern spinlock_t kvm_lock;
extern struct list_head vm_list;

91 92
struct kvm_vcpu;
struct kvm;
93
struct kvm_async_pf;
94

95
enum kvm_reg {
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
	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
114
	VCPU_REGS_RIP,
115 116 117
	NR_VCPU_REGS
};

A
Avi Kivity 已提交
118 119
enum kvm_reg_ex {
	VCPU_EXREG_PDPTR = NR_VCPU_REGS,
120
	VCPU_EXREG_CR3,
A
Avi Kivity 已提交
121 122
};

123
enum {
124
	VCPU_SREG_ES,
125
	VCPU_SREG_CS,
126
	VCPU_SREG_SS,
127 128 129 130 131 132 133
	VCPU_SREG_DS,
	VCPU_SREG_FS,
	VCPU_SREG_GS,
	VCPU_SREG_TR,
	VCPU_SREG_LDTR,
};

134
#include <asm/kvm_emulate.h>
135

136 137
#define KVM_NR_MEM_OBJS 40

138 139 140 141 142 143 144 145 146 147 148 149 150
#define KVM_NR_DB_REGS	4

#define DR6_BD		(1 << 13)
#define DR6_BS		(1 << 14)
#define DR6_FIXED_1	0xffff0ff0
#define DR6_VOLATILE	0x0000e00f

#define DR7_BP_EN_MASK	0x000000ff
#define DR7_GE		(1 << 9)
#define DR7_GD		(1 << 13)
#define DR7_FIXED_1	0x00000400
#define DR7_VOLATILE	0xffff23ff

151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
/*
 * 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];
};

#define NR_PTE_CHAIN_ENTRIES 5

struct kvm_pte_chain {
	u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
	struct hlist_node link;
};

/*
 * kvm_mmu_page_role, below, is defined as:
 *
 *   bits 0:3 - total guest paging levels (2-4, or zero for real mode)
 *   bits 4:7 - page table level for this shadow (1-4)
 *   bits 8:9 - page table quadrant for 2-level guests
173 174
 *   bit   16 - direct mapping of virtual to physical mapping at gfn
 *              used for real mode and two-dimensional paging
175 176 177 178 179
 *   bits 17:19 - common access permissions for all ptes in this shadow page
 */
union kvm_mmu_page_role {
	unsigned word;
	struct {
180
		unsigned level:4;
181
		unsigned cr4_pae:1;
182 183
		unsigned quadrant:2;
		unsigned pad_for_nice_hex_output:6;
184
		unsigned direct:1;
185
		unsigned access:3;
186
		unsigned invalid:1;
187
		unsigned nxe:1;
188
		unsigned cr0_wp:1;
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205
	};
};

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;
206 207 208 209 210
	/*
	 * One bit set per slot which has memory
	 * in this shadow page.
	 */
	DECLARE_BITMAP(slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS);
211
	bool multimapped;         /* More than one parent_pte? */
212
	bool unsync;
213
	int root_count;          /* Currently serving as active root */
214
	unsigned int unsync_children;
215 216 217 218
	union {
		u64 *parent_pte;               /* !multimapped */
		struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
	};
219
	DECLARE_BITMAP(unsync_child_bitmap, 512);
220 221
};

222 223 224 225 226 227 228
struct kvm_pv_mmu_op_buffer {
	void *ptr;
	unsigned len;
	unsigned processed;
	char buf[512] __aligned(sizeof(long));
};

229 230 231 232 233 234 235
struct kvm_pio_request {
	unsigned long count;
	int in;
	int port;
	int size;
};

236 237 238 239 240 241 242
/*
 * 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 {
	void (*new_cr3)(struct kvm_vcpu *vcpu);
243
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
244
	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
245 246
	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
			  bool prefault);
247 248
	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
				  struct x86_exception *fault);
249
	void (*free)(struct kvm_vcpu *vcpu);
250
	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
251
			    struct x86_exception *exception);
252
	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access);
253 254
	void (*prefetch_page)(struct kvm_vcpu *vcpu,
			      struct kvm_mmu_page *page);
255
	int (*sync_page)(struct kvm_vcpu *vcpu,
256
			 struct kvm_mmu_page *sp);
M
Marcelo Tosatti 已提交
257
	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva);
258 259 260
	hpa_t root_hpa;
	int root_level;
	int shadow_root_level;
261
	union kvm_mmu_page_role base_role;
262
	bool direct_map;
263 264

	u64 *pae_root;
265
	u64 *lm_root;
266
	u64 rsvd_bits_mask[2][4];
267

268 269
	bool nx;

270
	u64 pdptrs[4]; /* pae */
271 272
};

273
struct kvm_vcpu_arch {
274 275 276 277 278 279 280
	/*
	 * 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;
281 282

	unsigned long cr0;
283
	unsigned long cr0_guest_owned_bits;
284 285 286
	unsigned long cr2;
	unsigned long cr3;
	unsigned long cr4;
287
	unsigned long cr4_guest_owned_bits;
288
	unsigned long cr8;
289
	u32 hflags;
290
	u64 efer;
291 292
	u64 apic_base;
	struct kvm_lapic *apic;    /* kernel irqchip context */
293
	int32_t apic_arb_prio;
294 295 296
	int mp_state;
	int sipi_vector;
	u64 ia32_misc_enable_msr;
297
	bool tpr_access_reporting;
298

299 300 301 302 303 304 305
	/*
	 * 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.
	 */
306
	struct kvm_mmu mmu;
307

308 309 310 311 312 313 314 315 316 317
	/*
	 * 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;

318 319 320 321 322 323
	/*
	 * Pointer to the mmu context currently used for
	 * gva_to_gpa translations.
	 */
	struct kvm_mmu *walk_mmu;

324 325 326
	/* only needed in kvm_pv_mmu_op() path, but it's hot so
	 * put it here to avoid allocation */
	struct kvm_pv_mmu_op_buffer mmu_op_buffer;
327 328 329 330 331 332 333 334 335

	struct kvm_mmu_memory_cache mmu_pte_chain_cache;
	struct kvm_mmu_memory_cache mmu_rmap_desc_cache;
	struct kvm_mmu_memory_cache mmu_page_cache;
	struct kvm_mmu_memory_cache mmu_page_header_cache;

	gfn_t last_pt_write_gfn;
	int   last_pt_write_count;
	u64  *last_pte_updated;
336
	gfn_t last_pte_gfn;
337

338
	struct {
339 340
		gfn_t gfn;	/* presumed gfn during guest pte update */
		pfn_t pfn;	/* pfn corresponding to that gfn */
341
		unsigned long mmu_seq;
342 343
	} update_pte;

S
Sheng Yang 已提交
344
	struct fpu guest_fpu;
345
	u64 xcr0;
346 347 348 349 350

	gva_t mmio_fault_cr2;
	struct kvm_pio_request pio;
	void *pio_data;

351 352
	u8 event_exit_inst_len;

353 354 355
	struct kvm_queued_exception {
		bool pending;
		bool has_error_code;
356
		bool reinject;
357 358 359 360
		u8 nr;
		u32 error_code;
	} exception;

A
Avi Kivity 已提交
361 362
	struct kvm_queued_interrupt {
		bool pending;
363
		bool soft;
A
Avi Kivity 已提交
364 365 366
		u8 nr;
	} interrupt;

367 368 369
	int halt_request; /* real mode on Intel only */

	int cpuid_nent;
370
	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
371 372 373
	/* emulate context */

	struct x86_emulate_ctxt emulate_ctxt;
374 375

	gpa_t time;
376
	struct pvclock_vcpu_time_info hv_clock;
Z
Zachary Amsden 已提交
377
	unsigned int hw_tsc_khz;
378 379
	unsigned int time_offset;
	struct page *time_page;
Z
Zachary Amsden 已提交
380
	u64 last_host_tsc;
381 382
	u64 last_guest_tsc;
	u64 last_kernel_ns;
Z
Zachary Amsden 已提交
383 384 385
	u64 last_tsc_nsec;
	u64 last_tsc_write;
	bool tsc_catchup;
386 387

	bool nmi_pending;
388
	bool nmi_injected;
A
Avi Kivity 已提交
389

S
Sheng Yang 已提交
390 391
	struct mtrr_state_type mtrr_state;
	u32 pat;
392 393 394 395 396 397

	int switch_db_regs;
	unsigned long db[KVM_NR_DB_REGS];
	unsigned long dr6;
	unsigned long dr7;
	unsigned long eff_db[KVM_NR_DB_REGS];
H
Huang Ying 已提交
398 399 400 401 402

	u64 mcg_cap;
	u64 mcg_status;
	u64 mcg_ctl;
	u64 *mce_banks;
403 404 405

	/* used for guest single stepping over the given code position */
	unsigned long singlestep_rip;
J
Jan Kiszka 已提交
406

G
Gleb Natapov 已提交
407 408
	/* fields used by HYPER-V emulation */
	u64 hv_vapic;
409 410

	cpumask_var_t wbinvd_dirty_mask;
411 412 413 414

	struct {
		bool halted;
		gfn_t gfns[roundup_pow_of_two(ASYNC_PF_PER_VCPU)];
415 416
		struct gfn_to_hva_cache data;
		u64 msr_val;
417
		u32 id;
418
		bool send_user_only;
419
	} apf;
420 421
};

422
struct kvm_arch {
423
	unsigned int n_used_mmu_pages;
424
	unsigned int n_requested_mmu_pages;
425
	unsigned int n_max_mmu_pages;
426
	atomic_t invlpg_counter;
427 428 429 430 431
	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
	/*
	 * Hash table of struct kvm_mmu_page.
	 */
	struct list_head active_mmu_pages;
B
Ben-Ami Yassour 已提交
432
	struct list_head assigned_dev_head;
J
Joerg Roedel 已提交
433
	struct iommu_domain *iommu_domain;
434
	int iommu_flags;
435 436
	struct kvm_pic *vpic;
	struct kvm_ioapic *vioapic;
S
Sheng Yang 已提交
437
	struct kvm_pit *vpit;
438
	int vapics_in_nmi_mode;
439 440 441

	unsigned int tss_addr;
	struct page *apic_access_page;
442 443

	gpa_t wall_clock;
444 445 446

	struct page *ept_identity_pagetable;
	bool ept_identity_pagetable_done;
447
	gpa_t ept_identity_map_addr;
448 449

	unsigned long irq_sources_bitmap;
450
	s64 kvmclock_offset;
451
	spinlock_t tsc_write_lock;
Z
Zachary Amsden 已提交
452 453 454
	u64 last_tsc_nsec;
	u64 last_tsc_offset;
	u64 last_tsc_write;
Z
Zachary Amsden 已提交
455 456 457
	u32 virtual_tsc_khz;
	u32 virtual_tsc_mult;
	s8 virtual_tsc_shift;
E
Ed Swierk 已提交
458 459

	struct kvm_xen_hvm_config xen_hvm_config;
460 461 462 463

	/* fields used by HYPER-V emulation */
	u64 hv_guest_os_id;
	u64 hv_hypercall;
464 465 466 467

	#ifdef CONFIG_KVM_MMU_AUDIT
	int audit_point;
	#endif
468 469
};

470 471 472 473 474 475 476
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
477
	u32 mmu_cache_miss;
478
	u32 mmu_unsync;
479
	u32 remote_tlb_flush;
M
Marcelo Tosatti 已提交
480
	u32 lpages;
481 482
};

483 484 485 486 487 488 489 490 491 492 493
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;
494
	u32 nmi_window_exits;
495 496 497 498 499 500 501 502 503
	u32 halt_exits;
	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
504
	u32 hypercalls;
505
	u32 irq_injections;
506
	u32 nmi_injections;
507
};
508

509 510 511
struct kvm_x86_ops {
	int (*cpu_has_kvm_support)(void);          /* __init */
	int (*disabled_by_bios)(void);             /* __init */
512
	int (*hardware_enable)(void *dummy);
513 514 515 516
	void (*hardware_disable)(void *dummy);
	void (*check_processor_compatibility)(void *rtn);
	int (*hardware_setup)(void);               /* __init */
	void (*hardware_unsetup)(void);            /* __exit */
517
	bool (*cpu_has_accelerated_tpr)(void);
518
	void (*cpuid_update)(struct kvm_vcpu *vcpu);
519 520 521 522 523 524 525 526 527 528

	/* Create, but do not attach this VCPU */
	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
	void (*vcpu_free)(struct kvm_vcpu *vcpu);
	int (*vcpu_reset)(struct kvm_vcpu *vcpu);

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

529 530
	void (*set_guest_debug)(struct kvm_vcpu *vcpu,
				struct kvm_guest_debug *dbg);
531 532 533 534 535
	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
	void (*get_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
536
	int (*get_cpl)(struct kvm_vcpu *vcpu);
537 538 539
	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);
540
	void (*decache_cr0_guest_bits)(struct kvm_vcpu *vcpu);
541
	void (*decache_cr3)(struct kvm_vcpu *vcpu);
542 543 544 545 546
	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);
	void (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
547 548 549 550
	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);
551
	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
552
	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
553 554
	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
A
Avi Kivity 已提交
555
	void (*fpu_activate)(struct kvm_vcpu *vcpu);
556
	void (*fpu_deactivate)(struct kvm_vcpu *vcpu);
557 558 559

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

A
Avi Kivity 已提交
560 561
	void (*run)(struct kvm_vcpu *vcpu);
	int (*handle_exit)(struct kvm_vcpu *vcpu);
562
	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
563 564
	void (*set_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
	u32 (*get_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
565 566
	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
				unsigned char *hypercall_addr);
567
	void (*set_irq)(struct kvm_vcpu *vcpu);
568
	void (*set_nmi)(struct kvm_vcpu *vcpu);
569
	void (*queue_exception)(struct kvm_vcpu *vcpu, unsigned nr,
570 571
				bool has_error_code, u32 error_code,
				bool reinject);
A
Avi Kivity 已提交
572
	void (*cancel_injection)(struct kvm_vcpu *vcpu);
573
	int (*interrupt_allowed)(struct kvm_vcpu *vcpu);
574
	int (*nmi_allowed)(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
575 576
	bool (*get_nmi_mask)(struct kvm_vcpu *vcpu);
	void (*set_nmi_mask)(struct kvm_vcpu *vcpu, bool masked);
577 578 579
	void (*enable_nmi_window)(struct kvm_vcpu *vcpu);
	void (*enable_irq_window)(struct kvm_vcpu *vcpu);
	void (*update_cr8_intercept)(struct kvm_vcpu *vcpu, int tpr, int irr);
580
	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
581
	int (*get_tdp_level)(void);
582
	u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio);
583
	int (*get_lpage_level)(void);
584
	bool (*rdtscp_supported)(void);
Z
Zachary Amsden 已提交
585
	void (*adjust_tsc_offset)(struct kvm_vcpu *vcpu, s64 adjustment);
586

587 588
	void (*set_tdp_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);

589 590
	void (*set_supported_cpuid)(u32 func, struct kvm_cpuid_entry2 *entry);

591 592
	bool (*has_wbinvd_exit)(void);

593 594
	void (*write_tsc_offset)(struct kvm_vcpu *vcpu, u64 offset);

595
	void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2);
596
	const struct trace_print_flags *exit_reasons_str;
597 598
};

599
struct kvm_arch_async_pf {
600
	u32 token;
601
	gfn_t gfn;
X
Xiao Guangrong 已提交
602
	unsigned long cr3;
603
	bool direct_map;
604 605
};

606 607
extern struct kvm_x86_ops *kvm_x86_ops;

608 609 610 611 612 613 614
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);
int kvm_mmu_setup(struct kvm_vcpu *vcpu);
void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
S
Sheng Yang 已提交
615
void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
616
		u64 dirty_mask, u64 nx_mask, u64 x_mask);
617 618 619 620

int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
void kvm_mmu_zap_all(struct kvm *kvm);
621
unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
622 623
void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);

624
int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3);
625

626
int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
627
			  const void *val, int bytes);
628 629
int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes,
		  gpa_t addr, unsigned long *ret);
630
u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn);
631 632

extern bool tdp_enabled;
633

634 635 636 637 638 639
enum emulation_result {
	EMULATE_DONE,       /* no further processing */
	EMULATE_DO_MMIO,      /* kvm_run filled with mmio request */
	EMULATE_FAIL,         /* can't emulate this instruction */
};

640 641
#define EMULTYPE_NO_DECODE	    (1 << 0)
#define EMULTYPE_TRAP_UD	    (1 << 1)
642
#define EMULTYPE_SKIP		    (1 << 2)
643 644
int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2,
			    int emulation_type, void *insn, int insn_len);
645 646 647 648

static inline int emulate_instruction(struct kvm_vcpu *vcpu,
			int emulation_type)
{
649
	return x86_emulate_instruction(vcpu, 0, emulation_type, NULL, 0);
650 651
}

652 653 654
void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);

655
void kvm_enable_efer_bits(u64);
656 657 658 659 660
int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *data);
int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);

struct x86_emulate_ctxt;

661
int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port);
662 663 664 665
void kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
int kvm_emulate_halt(struct kvm_vcpu *vcpu);
int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address);
int emulate_clts(struct kvm_vcpu *vcpu);
666
int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu);
667

668
void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg);
669
int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg);
670

671 672
int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason,
		    bool has_error_code, u32 error_code);
673

674
int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
675
int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3);
676
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4);
A
Andre Przywara 已提交
677
int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8);
678 679
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);
680 681
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu);
void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
682
void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
683
int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr);
684 685 686 687

int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata);
int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data);

688 689 690
unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu);
void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);

691 692
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
693 694
void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
695
void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
696 697 698
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);
699
void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
700
bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
701

702
int kvm_pic_set_irq(void *opaque, int irq, int level);
703

704 705
void kvm_inject_nmi(struct kvm_vcpu *vcpu);

706
int fx_init(struct kvm_vcpu *vcpu);
707

708
void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
709
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
710 711
		       const u8 *new, int bytes,
		       bool guest_initiated);
712 713 714 715
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);
716
void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
717 718 719 720 721 722 723 724
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);
725 726 727 728 729

int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);

int kvm_fix_hypercall(struct kvm_vcpu *vcpu);

730 731
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
732
void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva);
733

734
void kvm_enable_tdp(void);
735
void kvm_disable_tdp(void);
736

737
int complete_pio(struct kvm_vcpu *vcpu);
738
bool kvm_check_iopl(struct kvm_vcpu *vcpu);
739 740 741 742 743 744 745 746

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);
}

747
static inline u16 kvm_read_ldt(void)
748 749 750 751 752 753
{
	u16 ldt;
	asm("sldt %0" : "=g"(ldt));
	return ldt;
}

754
static inline void kvm_load_ldt(u16 sel)
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773
{
	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 */
}

774 775 776 777 778
static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
{
	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
}

779 780 781 782
#define TSS_IOPB_BASE_OFFSET 0x66
#define TSS_BASE_SIZE 0x68
#define TSS_IOPB_SIZE (65536 / 8)
#define TSS_REDIRECTION_SIZE (256 / 8)
783 784
#define RMODE_TSS_SIZE							\
	(TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
785

786 787 788 789 790 791 792
enum {
	TASK_SWITCH_CALL = 0,
	TASK_SWITCH_IRET = 1,
	TASK_SWITCH_JMP = 2,
	TASK_SWITCH_GATE = 3,
};

793
#define HF_GIF_MASK		(1 << 0)
A
Alexander Graf 已提交
794 795
#define HF_HIF_MASK		(1 << 1)
#define HF_VINTR_MASK		(1 << 2)
796
#define HF_NMI_MASK		(1 << 3)
797
#define HF_IRET_MASK		(1 << 4)
798
#define HF_GUEST_MASK		(1 << 5) /* VCPU is in guest-mode */
799

800 801 802 803 804
/*
 * 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.
 */
805 806
asmlinkage void kvm_spurious_fault(void);
extern bool kvm_rebooting;
807 808 809

#define __kvm_handle_fault_on_reboot(insn) \
	"666: " insn "\n\t" \
810
	"668: \n\t"                           \
811
	".pushsection .fixup, \"ax\" \n" \
812
	"667: \n\t" \
813 814
	"cmpb $0, kvm_rebooting \n\t"	      \
	"jne 668b \n\t"      		      \
815
	__ASM_SIZE(push) " $666b \n\t"	      \
816
	"call kvm_spurious_fault \n\t"	      \
817 818
	".popsection \n\t" \
	".pushsection __ex_table, \"a\" \n\t" \
819
	_ASM_PTR " 666b, 667b \n\t" \
820 821
	".popsection"

822 823 824
#define KVM_ARCH_WANT_MMU_NOTIFIER
int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
int kvm_age_hva(struct kvm *kvm, unsigned long hva);
A
Andrea Arcangeli 已提交
825
int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
826
void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
827
int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
828 829
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
830
int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
831

A
Avi Kivity 已提交
832
void kvm_define_shared_msr(unsigned index, u32 msr);
833
void kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
A
Avi Kivity 已提交
834

J
Jan Kiszka 已提交
835 836
bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip);

837 838 839 840
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
841 842
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
			       struct kvm_async_pf *work);
843
bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu);
844 845
extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn);

846 847
void kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err);

H
H. Peter Anvin 已提交
848
#endif /* _ASM_X86_KVM_HOST_H */