1/* SPDX-License-Identifier: GPL-2.0 */
2#ifndef __KVM_X86_MMU_INTERNAL_H
3#define __KVM_X86_MMU_INTERNAL_H
4
5#include <linux/types.h>
6#include <linux/kvm_host.h>
7#include <asm/kvm_host.h>
8
9#undef MMU_DEBUG
10
11#ifdef MMU_DEBUG
12extern bool dbg;
13
14#define pgprintk(x...) do { if (dbg) printk(x); } while (0)
15#define rmap_printk(x...) do { if (dbg) printk(x); } while (0)
16#define MMU_WARN_ON(x) WARN_ON(x)
17#else
18#define pgprintk(x...) do { } while (0)
19#define rmap_printk(x...) do { } while (0)
20#define MMU_WARN_ON(x) do { } while (0)
21#endif
22
23struct kvm_mmu_page {
24	struct list_head link;
25	struct hlist_node hash_link;
26	struct list_head lpage_disallowed_link;
27
28	bool unsync;
29	u8 mmu_valid_gen;
30	bool mmio_cached;
31	bool lpage_disallowed; /* Can't be replaced by an equiv large page */
32
33	/*
34	 * The following two entries are used to key the shadow page in the
35	 * hash table.
36	 */
37	union kvm_mmu_page_role role;
38	gfn_t gfn;
39
40	u64 *spt;
41	/* hold the gfn of each spte inside spt */
42	gfn_t *gfns;
43	int root_count;          /* Currently serving as active root */
44	unsigned int unsync_children;
45	struct kvm_rmap_head parent_ptes; /* rmap pointers to parent sptes */
46	DECLARE_BITMAP(unsync_child_bitmap, 512);
47
48#ifdef CONFIG_X86_32
49	/*
50	 * Used out of the mmu-lock to avoid reading spte values while an
51	 * update is in progress; see the comments in __get_spte_lockless().
52	 */
53	int clear_spte_count;
54#endif
55
56	/* Number of writes since the last time traversal visited this page.  */
57	atomic_t write_flooding_count;
58
59	bool tdp_mmu_page;
60};
61
62extern struct kmem_cache *mmu_page_header_cache;
63
64static inline struct kvm_mmu_page *to_shadow_page(hpa_t shadow_page)
65{
66	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);
67
68	return (struct kvm_mmu_page *)page_private(page);
69}
70
71static inline struct kvm_mmu_page *sptep_to_sp(u64 *sptep)
72{
73	return to_shadow_page(__pa(sptep));
74}
75
76static inline bool kvm_vcpu_ad_need_write_protect(struct kvm_vcpu *vcpu)
77{
78	/*
79	 * When using the EPT page-modification log, the GPAs in the log
80	 * would come from L2 rather than L1.  Therefore, we need to rely
81	 * on write protection to record dirty pages.  This also bypasses
82	 * PML, since writes now result in a vmexit.
83	 */
84	return vcpu->arch.mmu == &vcpu->arch.guest_mmu;
85}
86
87bool is_nx_huge_page_enabled(void);
88bool mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn,
89			    bool can_unsync);
90
91void kvm_mmu_gfn_disallow_lpage(struct kvm_memory_slot *slot, gfn_t gfn);
92void kvm_mmu_gfn_allow_lpage(struct kvm_memory_slot *slot, gfn_t gfn);
93bool kvm_mmu_slot_gfn_write_protect(struct kvm *kvm,
94				    struct kvm_memory_slot *slot, u64 gfn);
95void kvm_flush_remote_tlbs_with_address(struct kvm *kvm,
96					u64 start_gfn, u64 pages);
97
98static inline void kvm_mmu_get_root(struct kvm *kvm, struct kvm_mmu_page *sp)
99{
100	BUG_ON(!sp->root_count);
101	lockdep_assert_held(&kvm->mmu_lock);
102
103	++sp->root_count;
104}
105
106static inline bool kvm_mmu_put_root(struct kvm *kvm, struct kvm_mmu_page *sp)
107{
108	lockdep_assert_held(&kvm->mmu_lock);
109	--sp->root_count;
110
111	return !sp->root_count;
112}
113
114/*
115 * Return values of handle_mmio_page_fault, mmu.page_fault, and fast_page_fault().
116 *
117 * RET_PF_RETRY: let CPU fault again on the address.
118 * RET_PF_EMULATE: mmio page fault, emulate the instruction directly.
119 * RET_PF_INVALID: the spte is invalid, let the real page fault path update it.
120 * RET_PF_FIXED: The faulting entry has been fixed.
121 * RET_PF_SPURIOUS: The faulting entry was already fixed, e.g. by another vCPU.
122 */
123enum {
124	RET_PF_RETRY = 0,
125	RET_PF_EMULATE,
126	RET_PF_INVALID,
127	RET_PF_FIXED,
128	RET_PF_SPURIOUS,
129};
130
131/* Bits which may be returned by set_spte() */
132#define SET_SPTE_WRITE_PROTECTED_PT	BIT(0)
133#define SET_SPTE_NEED_REMOTE_TLB_FLUSH	BIT(1)
134#define SET_SPTE_SPURIOUS		BIT(2)
135
136int kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, gfn_t gfn,
137			    int max_level, kvm_pfn_t *pfnp,
138			    bool huge_page_disallowed, int *req_level);
139void disallowed_hugepage_adjust(u64 spte, gfn_t gfn, int cur_level,
140				kvm_pfn_t *pfnp, int *goal_levelp);
141
142bool is_nx_huge_page_enabled(void);
143
144void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc);
145
146void account_huge_nx_page(struct kvm *kvm, struct kvm_mmu_page *sp);
147void unaccount_huge_nx_page(struct kvm *kvm, struct kvm_mmu_page *sp);
148
149#endif /* __KVM_X86_MMU_INTERNAL_H */
150