diff options
Diffstat (limited to 'arch/x86/kvm/mmu')
| -rw-r--r-- | arch/x86/kvm/mmu/mmu.c | 1918 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/mmu_internal.h | 141 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/mmutrace.h | 3 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/page_track.c | 2 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/paging_tmpl.h | 122 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/spte.c | 219 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/spte.h | 141 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/tdp_iter.c | 10 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/tdp_iter.h | 57 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/tdp_mmu.c | 853 | ||||
| -rw-r--r-- | arch/x86/kvm/mmu/tdp_mmu.h | 60 |
11 files changed, 2094 insertions, 1432 deletions
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 992e651540e8..63bb77ee1bb1 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -179,7 +179,6 @@ struct kvm_shadow_walk_iterator { static struct kmem_cache *pte_list_desc_cache; struct kmem_cache *mmu_page_header_cache; -static struct percpu_counter kvm_total_used_mmu_pages; static void mmu_spte_set(u64 *sptep, u64 spte); @@ -336,16 +335,19 @@ static int is_cpuid_PSE36(void) #ifdef CONFIG_X86_64 static void __set_spte(u64 *sptep, u64 spte) { + KVM_MMU_WARN_ON(is_ept_ve_possible(spte)); WRITE_ONCE(*sptep, spte); } static void __update_clear_spte_fast(u64 *sptep, u64 spte) { + KVM_MMU_WARN_ON(is_ept_ve_possible(spte)); WRITE_ONCE(*sptep, spte); } static u64 __update_clear_spte_slow(u64 *sptep, u64 spte) { + KVM_MMU_WARN_ON(is_ept_ve_possible(spte)); return xchg(sptep, spte); } @@ -432,8 +434,8 @@ static u64 __update_clear_spte_slow(u64 *sptep, u64 spte) * The idea using the light way get the spte on x86_32 guest is from * gup_get_pte (mm/gup.c). * - * An spte tlb flush may be pending, because kvm_set_pte_rmap - * coalesces them and we are running out of the MMU lock. Therefore + * An spte tlb flush may be pending, because they are coalesced and + * we are running out of the MMU lock. Therefore * we need to protect against in-progress updates of the spte. * * Reading the spte while an update is in progress may get the old value @@ -482,11 +484,12 @@ static void mmu_spte_set(u64 *sptep, u64 new_spte) __set_spte(sptep, new_spte); } -/* - * Update the SPTE (excluding the PFN), but do not track changes in its - * accessed/dirty status. +/* Rules for using mmu_spte_update: + * Update the state bits, it means the mapped pfn is not changed. + * + * Returns true if the TLB needs to be flushed */ -static u64 mmu_spte_update_no_track(u64 *sptep, u64 new_spte) +static bool mmu_spte_update(u64 *sptep, u64 new_spte) { u64 old_spte = *sptep; @@ -495,61 +498,18 @@ static u64 mmu_spte_update_no_track(u64 *sptep, u64 new_spte) if (!is_shadow_present_pte(old_spte)) { mmu_spte_set(sptep, new_spte); - return old_spte; + return false; } - if (!spte_has_volatile_bits(old_spte)) + if (!spte_needs_atomic_update(old_spte)) __update_clear_spte_fast(sptep, new_spte); else old_spte = __update_clear_spte_slow(sptep, new_spte); - WARN_ON_ONCE(spte_to_pfn(old_spte) != spte_to_pfn(new_spte)); + WARN_ON_ONCE(!is_shadow_present_pte(old_spte) || + spte_to_pfn(old_spte) != spte_to_pfn(new_spte)); - return old_spte; -} - -/* Rules for using mmu_spte_update: - * Update the state bits, it means the mapped pfn is not changed. - * - * Whenever an MMU-writable SPTE is overwritten with a read-only SPTE, remote - * TLBs must be flushed. Otherwise rmap_write_protect will find a read-only - * spte, even though the writable spte might be cached on a CPU's TLB. - * - * Returns true if the TLB needs to be flushed - */ -static bool mmu_spte_update(u64 *sptep, u64 new_spte) -{ - bool flush = false; - u64 old_spte = mmu_spte_update_no_track(sptep, new_spte); - - if (!is_shadow_present_pte(old_spte)) - return false; - - /* - * For the spte updated out of mmu-lock is safe, since - * we always atomically update it, see the comments in - * spte_has_volatile_bits(). - */ - if (is_mmu_writable_spte(old_spte) && - !is_writable_pte(new_spte)) - flush = true; - - /* - * Flush TLB when accessed/dirty states are changed in the page tables, - * to guarantee consistency between TLB and page tables. - */ - - if (is_accessed_spte(old_spte) && !is_accessed_spte(new_spte)) { - flush = true; - kvm_set_pfn_accessed(spte_to_pfn(old_spte)); - } - - if (is_dirty_spte(old_spte) && !is_dirty_spte(new_spte)) { - flush = true; - kvm_set_pfn_dirty(spte_to_pfn(old_spte)); - } - - return flush; + return leaf_spte_change_needs_tlb_flush(old_spte, new_spte); } /* @@ -560,39 +520,19 @@ static bool mmu_spte_update(u64 *sptep, u64 new_spte) */ static u64 mmu_spte_clear_track_bits(struct kvm *kvm, u64 *sptep) { - kvm_pfn_t pfn; u64 old_spte = *sptep; int level = sptep_to_sp(sptep)->role.level; - struct page *page; if (!is_shadow_present_pte(old_spte) || - !spte_has_volatile_bits(old_spte)) - __update_clear_spte_fast(sptep, 0ull); + !spte_needs_atomic_update(old_spte)) + __update_clear_spte_fast(sptep, SHADOW_NONPRESENT_VALUE); else - old_spte = __update_clear_spte_slow(sptep, 0ull); + old_spte = __update_clear_spte_slow(sptep, SHADOW_NONPRESENT_VALUE); if (!is_shadow_present_pte(old_spte)) return old_spte; kvm_update_page_stats(kvm, level, -1); - - pfn = spte_to_pfn(old_spte); - - /* - * KVM doesn't hold a reference to any pages mapped into the guest, and - * instead uses the mmu_notifier to ensure that KVM unmaps any pages - * before they are reclaimed. Sanity check that, if the pfn is backed - * by a refcounted page, the refcount is elevated. - */ - page = kvm_pfn_to_refcounted_page(pfn); - WARN_ON_ONCE(page && !page_count(page)); - - if (is_accessed_spte(old_spte)) - kvm_set_pfn_accessed(pfn); - - if (is_dirty_spte(old_spte)) - kvm_set_pfn_dirty(pfn); - return old_spte; } @@ -603,7 +543,7 @@ static u64 mmu_spte_clear_track_bits(struct kvm *kvm, u64 *sptep) */ static void mmu_spte_clear_no_track(u64 *sptep) { - __update_clear_spte_fast(sptep, 0ull); + __update_clear_spte_fast(sptep, SHADOW_NONPRESENT_VALUE); } static u64 mmu_spte_get_lockless(u64 *sptep) @@ -611,32 +551,6 @@ static u64 mmu_spte_get_lockless(u64 *sptep) return __get_spte_lockless(sptep); } -/* Returns the Accessed status of the PTE and resets it at the same time. */ -static bool mmu_spte_age(u64 *sptep) -{ - u64 spte = mmu_spte_get_lockless(sptep); - - if (!is_accessed_spte(spte)) - return false; - - if (spte_ad_enabled(spte)) { - clear_bit((ffs(shadow_accessed_mask) - 1), - (unsigned long *)sptep); - } else { - /* - * Capture the dirty status of the page, so that it doesn't get - * lost when the SPTE is marked for access tracking. - */ - if (is_writable_pte(spte)) - kvm_set_pfn_dirty(spte_to_pfn(spte)); - - spte = mark_spte_for_access_track(spte); - mmu_spte_update_no_track(sptep, spte); - } - - return true; -} - static inline bool is_tdp_mmu_active(struct kvm_vcpu *vcpu) { return tdp_mmu_enabled && vcpu->arch.mmu->root_role.direct; @@ -685,6 +599,12 @@ static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu, bool maybe_indirect) 1 + PT64_ROOT_MAX_LEVEL + PTE_PREFETCH_NUM); if (r) return r; + if (kvm_has_mirrored_tdp(vcpu->kvm)) { + r = kvm_mmu_topup_memory_cache(&vcpu->arch.mmu_external_spt_cache, + PT64_ROOT_MAX_LEVEL); + if (r) + return r; + } r = kvm_mmu_topup_memory_cache(&vcpu->arch.mmu_shadow_page_cache, PT64_ROOT_MAX_LEVEL); if (r) @@ -704,6 +624,7 @@ static void mmu_free_memory_caches(struct kvm_vcpu *vcpu) kvm_mmu_free_memory_cache(&vcpu->arch.mmu_pte_list_desc_cache); kvm_mmu_free_memory_cache(&vcpu->arch.mmu_shadow_page_cache); kvm_mmu_free_memory_cache(&vcpu->arch.mmu_shadowed_info_cache); + kvm_mmu_free_memory_cache(&vcpu->arch.mmu_external_spt_cache); kvm_mmu_free_memory_cache(&vcpu->arch.mmu_page_header_cache); } @@ -719,7 +640,7 @@ static gfn_t kvm_mmu_page_get_gfn(struct kvm_mmu_page *sp, int index) if (sp->role.passthrough) return sp->gfn; - if (!sp->role.direct) + if (sp->shadowed_translation) return sp->shadowed_translation[index] >> PAGE_SHIFT; return sp->gfn + (index << ((sp->role.level - 1) * SPTE_LEVEL_BITS)); @@ -733,7 +654,7 @@ static gfn_t kvm_mmu_page_get_gfn(struct kvm_mmu_page *sp, int index) */ static u32 kvm_mmu_page_get_access(struct kvm_mmu_page *sp, int index) { - if (sp_has_gptes(sp)) + if (sp->shadowed_translation) return sp->shadowed_translation[index] & ACC_ALL; /* @@ -754,7 +675,7 @@ static u32 kvm_mmu_page_get_access(struct kvm_mmu_page *sp, int index) static void kvm_mmu_page_set_translation(struct kvm_mmu_page *sp, int index, gfn_t gfn, unsigned int access) { - if (sp_has_gptes(sp)) { + if (sp->shadowed_translation) { sp->shadowed_translation[index] = (gfn << PAGE_SHIFT) | access; return; } @@ -831,6 +752,15 @@ static void account_shadowed(struct kvm *kvm, struct kvm_mmu_page *sp) gfn_t gfn; kvm->arch.indirect_shadow_pages++; + /* + * Ensure indirect_shadow_pages is elevated prior to re-reading guest + * child PTEs in FNAME(gpte_changed), i.e. guarantee either in-flight + * emulated writes are visible before re-reading guest PTEs, or that + * an emulated write will see the elevated count and acquire mmu_lock + * to update SPTEs. Pairs with the smp_mb() in kvm_mmu_track_write(). + */ + smp_mb(); + gfn = sp->gfn; slots = kvm_memslots_for_spte_role(kvm, sp->role); slot = __gfn_to_memslot(slots, gfn); @@ -923,31 +853,173 @@ static struct kvm_memory_slot *gfn_to_memslot_dirty_bitmap(struct kvm_vcpu *vcpu * About rmap_head encoding: * * If the bit zero of rmap_head->val is clear, then it points to the only spte - * in this rmap chain. Otherwise, (rmap_head->val & ~1) points to a struct + * in this rmap chain. Otherwise, (rmap_head->val & ~3) points to a struct * pte_list_desc containing more mappings. */ +#define KVM_RMAP_MANY BIT(0) + +/* + * rmaps and PTE lists are mostly protected by mmu_lock (the shadow MMU always + * operates with mmu_lock held for write), but rmaps can be walked without + * holding mmu_lock so long as the caller can tolerate SPTEs in the rmap chain + * being zapped/dropped _while the rmap is locked_. + * + * Other than the KVM_RMAP_LOCKED flag, modifications to rmap entries must be + * done while holding mmu_lock for write. This allows a task walking rmaps + * without holding mmu_lock to concurrently walk the same entries as a task + * that is holding mmu_lock but _not_ the rmap lock. Neither task will modify + * the rmaps, thus the walks are stable. + * + * As alluded to above, SPTEs in rmaps are _not_ protected by KVM_RMAP_LOCKED, + * only the rmap chains themselves are protected. E.g. holding an rmap's lock + * ensures all "struct pte_list_desc" fields are stable. + */ +#define KVM_RMAP_LOCKED BIT(1) + +static unsigned long __kvm_rmap_lock(struct kvm_rmap_head *rmap_head) +{ + unsigned long old_val, new_val; + + lockdep_assert_preemption_disabled(); + + /* + * Elide the lock if the rmap is empty, as lockless walkers (read-only + * mode) don't need to (and can't) walk an empty rmap, nor can they add + * entries to the rmap. I.e. the only paths that process empty rmaps + * do so while holding mmu_lock for write, and are mutually exclusive. + */ + old_val = atomic_long_read(&rmap_head->val); + if (!old_val) + return 0; + + do { + /* + * If the rmap is locked, wait for it to be unlocked before + * trying acquire the lock, e.g. to avoid bouncing the cache + * line. + */ + while (old_val & KVM_RMAP_LOCKED) { + cpu_relax(); + old_val = atomic_long_read(&rmap_head->val); + } + + /* + * Recheck for an empty rmap, it may have been purged by the + * task that held the lock. + */ + if (!old_val) + return 0; + + new_val = old_val | KVM_RMAP_LOCKED; + /* + * Use try_cmpxchg_acquire() to prevent reads and writes to the rmap + * from being reordered outside of the critical section created by + * __kvm_rmap_lock(). + * + * Pairs with the atomic_long_set_release() in kvm_rmap_unlock(). + * + * For the !old_val case, no ordering is needed, as there is no rmap + * to walk. + */ + } while (!atomic_long_try_cmpxchg_acquire(&rmap_head->val, &old_val, new_val)); + + /* + * Return the old value, i.e. _without_ the LOCKED bit set. It's + * impossible for the return value to be 0 (see above), i.e. the read- + * only unlock flow can't get a false positive and fail to unlock. + */ + return old_val; +} + +static unsigned long kvm_rmap_lock(struct kvm *kvm, + struct kvm_rmap_head *rmap_head) +{ + lockdep_assert_held_write(&kvm->mmu_lock); + + return __kvm_rmap_lock(rmap_head); +} + +static void __kvm_rmap_unlock(struct kvm_rmap_head *rmap_head, + unsigned long val) +{ + KVM_MMU_WARN_ON(val & KVM_RMAP_LOCKED); + /* + * Ensure that all accesses to the rmap have completed before unlocking + * the rmap. + * + * Pairs with the atomic_long_try_cmpxchg_acquire() in __kvm_rmap_lock(). + */ + atomic_long_set_release(&rmap_head->val, val); +} + +static void kvm_rmap_unlock(struct kvm *kvm, + struct kvm_rmap_head *rmap_head, + unsigned long new_val) +{ + lockdep_assert_held_write(&kvm->mmu_lock); + + __kvm_rmap_unlock(rmap_head, new_val); +} + +static unsigned long kvm_rmap_get(struct kvm_rmap_head *rmap_head) +{ + return atomic_long_read(&rmap_head->val) & ~KVM_RMAP_LOCKED; +} + +/* + * If mmu_lock isn't held, rmaps can only be locked in read-only mode. The + * actual locking is the same, but the caller is disallowed from modifying the + * rmap, and so the unlock flow is a nop if the rmap is/was empty. + */ +static unsigned long kvm_rmap_lock_readonly(struct kvm_rmap_head *rmap_head) +{ + unsigned long rmap_val; + + preempt_disable(); + rmap_val = __kvm_rmap_lock(rmap_head); + + if (!rmap_val) + preempt_enable(); + + return rmap_val; +} + +static void kvm_rmap_unlock_readonly(struct kvm_rmap_head *rmap_head, + unsigned long old_val) +{ + if (!old_val) + return; + + KVM_MMU_WARN_ON(old_val != kvm_rmap_get(rmap_head)); + + __kvm_rmap_unlock(rmap_head, old_val); + preempt_enable(); +} /* * Returns the number of pointers in the rmap chain, not counting the new one. */ -static int pte_list_add(struct kvm_mmu_memory_cache *cache, u64 *spte, - struct kvm_rmap_head *rmap_head) +static int pte_list_add(struct kvm *kvm, struct kvm_mmu_memory_cache *cache, + u64 *spte, struct kvm_rmap_head *rmap_head) { + unsigned long old_val, new_val; struct pte_list_desc *desc; int count = 0; - if (!rmap_head->val) { - rmap_head->val = (unsigned long)spte; - } else if (!(rmap_head->val & 1)) { + old_val = kvm_rmap_lock(kvm, rmap_head); + + if (!old_val) { + new_val = (unsigned long)spte; + } else if (!(old_val & KVM_RMAP_MANY)) { desc = kvm_mmu_memory_cache_alloc(cache); - desc->sptes[0] = (u64 *)rmap_head->val; + desc->sptes[0] = (u64 *)old_val; desc->sptes[1] = spte; desc->spte_count = 2; desc->tail_count = 0; - rmap_head->val = (unsigned long)desc | 1; + new_val = (unsigned long)desc | KVM_RMAP_MANY; ++count; } else { - desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + desc = (struct pte_list_desc *)(old_val & ~KVM_RMAP_MANY); count = desc->tail_count + desc->spte_count; /* @@ -956,21 +1028,25 @@ static int pte_list_add(struct kvm_mmu_memory_cache *cache, u64 *spte, */ if (desc->spte_count == PTE_LIST_EXT) { desc = kvm_mmu_memory_cache_alloc(cache); - desc->more = (struct pte_list_desc *)(rmap_head->val & ~1ul); + desc->more = (struct pte_list_desc *)(old_val & ~KVM_RMAP_MANY); desc->spte_count = 0; desc->tail_count = count; - rmap_head->val = (unsigned long)desc | 1; + new_val = (unsigned long)desc | KVM_RMAP_MANY; + } else { + new_val = old_val; } desc->sptes[desc->spte_count++] = spte; } + + kvm_rmap_unlock(kvm, rmap_head, new_val); + return count; } -static void pte_list_desc_remove_entry(struct kvm *kvm, - struct kvm_rmap_head *rmap_head, +static void pte_list_desc_remove_entry(struct kvm *kvm, unsigned long *rmap_val, struct pte_list_desc *desc, int i) { - struct pte_list_desc *head_desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + struct pte_list_desc *head_desc = (struct pte_list_desc *)(*rmap_val & ~KVM_RMAP_MANY); int j = head_desc->spte_count - 1; /* @@ -997,9 +1073,9 @@ static void pte_list_desc_remove_entry(struct kvm *kvm, * head at the next descriptor, i.e. the new head. */ if (!head_desc->more) - rmap_head->val = 0; + *rmap_val = 0; else - rmap_head->val = (unsigned long)head_desc->more | 1; + *rmap_val = (unsigned long)head_desc->more | KVM_RMAP_MANY; mmu_free_pte_list_desc(head_desc); } @@ -1007,24 +1083,26 @@ static void pte_list_remove(struct kvm *kvm, u64 *spte, struct kvm_rmap_head *rmap_head) { struct pte_list_desc *desc; + unsigned long rmap_val; int i; - if (KVM_BUG_ON_DATA_CORRUPTION(!rmap_head->val, kvm)) - return; + rmap_val = kvm_rmap_lock(kvm, rmap_head); + if (KVM_BUG_ON_DATA_CORRUPTION(!rmap_val, kvm)) + goto out; - if (!(rmap_head->val & 1)) { - if (KVM_BUG_ON_DATA_CORRUPTION((u64 *)rmap_head->val != spte, kvm)) - return; + if (!(rmap_val & KVM_RMAP_MANY)) { + if (KVM_BUG_ON_DATA_CORRUPTION((u64 *)rmap_val != spte, kvm)) + goto out; - rmap_head->val = 0; + rmap_val = 0; } else { - desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + desc = (struct pte_list_desc *)(rmap_val & ~KVM_RMAP_MANY); while (desc) { for (i = 0; i < desc->spte_count; ++i) { if (desc->sptes[i] == spte) { - pte_list_desc_remove_entry(kvm, rmap_head, + pte_list_desc_remove_entry(kvm, &rmap_val, desc, i); - return; + goto out; } } desc = desc->more; @@ -1032,6 +1110,9 @@ static void pte_list_remove(struct kvm *kvm, u64 *spte, KVM_BUG_ON_DATA_CORRUPTION(true, kvm); } + +out: + kvm_rmap_unlock(kvm, rmap_head, rmap_val); } static void kvm_zap_one_rmap_spte(struct kvm *kvm, @@ -1046,17 +1127,19 @@ static bool kvm_zap_all_rmap_sptes(struct kvm *kvm, struct kvm_rmap_head *rmap_head) { struct pte_list_desc *desc, *next; + unsigned long rmap_val; int i; - if (!rmap_head->val) + rmap_val = kvm_rmap_lock(kvm, rmap_head); + if (!rmap_val) return false; - if (!(rmap_head->val & 1)) { - mmu_spte_clear_track_bits(kvm, (u64 *)rmap_head->val); + if (!(rmap_val & KVM_RMAP_MANY)) { + mmu_spte_clear_track_bits(kvm, (u64 *)rmap_val); goto out; } - desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + desc = (struct pte_list_desc *)(rmap_val & ~KVM_RMAP_MANY); for (; desc; desc = next) { for (i = 0; i < desc->spte_count; i++) @@ -1066,20 +1149,21 @@ static bool kvm_zap_all_rmap_sptes(struct kvm *kvm, } out: /* rmap_head is meaningless now, remember to reset it */ - rmap_head->val = 0; + kvm_rmap_unlock(kvm, rmap_head, 0); return true; } unsigned int pte_list_count(struct kvm_rmap_head *rmap_head) { + unsigned long rmap_val = kvm_rmap_get(rmap_head); struct pte_list_desc *desc; - if (!rmap_head->val) + if (!rmap_val) return 0; - else if (!(rmap_head->val & 1)) + else if (!(rmap_val & KVM_RMAP_MANY)) return 1; - desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + desc = (struct pte_list_desc *)(rmap_val & ~KVM_RMAP_MANY); return desc->tail_count + desc->spte_count; } @@ -1122,6 +1206,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) */ struct rmap_iterator { /* private fields */ + struct rmap_head *head; struct pte_list_desc *desc; /* holds the sptep if not NULL */ int pos; /* index of the sptep */ }; @@ -1136,23 +1221,19 @@ struct rmap_iterator { static u64 *rmap_get_first(struct kvm_rmap_head *rmap_head, struct rmap_iterator *iter) { - u64 *sptep; + unsigned long rmap_val = kvm_rmap_get(rmap_head); - if (!rmap_head->val) + if (!rmap_val) return NULL; - if (!(rmap_head->val & 1)) { + if (!(rmap_val & KVM_RMAP_MANY)) { iter->desc = NULL; - sptep = (u64 *)rmap_head->val; - goto out; + return (u64 *)rmap_val; } - iter->desc = (struct pte_list_desc *)(rmap_head->val & ~1ul); + iter->desc = (struct pte_list_desc *)(rmap_val & ~KVM_RMAP_MANY); iter->pos = 0; - sptep = iter->desc->sptes[iter->pos]; -out: - BUG_ON(!is_shadow_present_pte(*sptep)); - return sptep; + return iter->desc->sptes[iter->pos]; } /* @@ -1162,14 +1243,11 @@ out: */ static u64 *rmap_get_next(struct rmap_iterator *iter) { - u64 *sptep; - if (iter->desc) { if (iter->pos < PTE_LIST_EXT - 1) { ++iter->pos; - sptep = iter->desc->sptes[iter->pos]; - if (sptep) - goto out; + if (iter->desc->sptes[iter->pos]) + return iter->desc->sptes[iter->pos]; } iter->desc = iter->desc->more; @@ -1177,20 +1255,24 @@ static u64 *rmap_get_next(struct rmap_iterator *iter) if (iter->desc) { iter->pos = 0; /* desc->sptes[0] cannot be NULL */ - sptep = iter->desc->sptes[iter->pos]; - goto out; + return iter->desc->sptes[iter->pos]; } } return NULL; -out: - BUG_ON(!is_shadow_present_pte(*sptep)); - return sptep; } -#define for_each_rmap_spte(_rmap_head_, _iter_, _spte_) \ - for (_spte_ = rmap_get_first(_rmap_head_, _iter_); \ - _spte_; _spte_ = rmap_get_next(_iter_)) +#define __for_each_rmap_spte(_rmap_head_, _iter_, _sptep_) \ + for (_sptep_ = rmap_get_first(_rmap_head_, _iter_); \ + _sptep_; _sptep_ = rmap_get_next(_iter_)) + +#define for_each_rmap_spte(_rmap_head_, _iter_, _sptep_) \ + __for_each_rmap_spte(_rmap_head_, _iter_, _sptep_) \ + if (!WARN_ON_ONCE(!is_shadow_present_pte(*(_sptep_)))) \ + +#define for_each_rmap_spte_lockless(_rmap_head_, _iter_, _sptep_, _spte_) \ + __for_each_rmap_spte(_rmap_head_, _iter_, _sptep_) \ + if (is_shadow_present_pte(_spte_ = mmu_spte_get_lockless(sptep))) static void drop_spte(struct kvm *kvm, u64 *sptep) { @@ -1263,16 +1345,6 @@ static bool spte_clear_dirty(u64 *sptep) return mmu_spte_update(sptep, spte); } -static bool spte_wrprot_for_clear_dirty(u64 *sptep) -{ - bool was_writable = test_and_clear_bit(PT_WRITABLE_SHIFT, - (unsigned long *)sptep); - if (was_writable && !spte_ad_enabled(*sptep)) - kvm_set_pfn_dirty(spte_to_pfn(*sptep)); - - return was_writable; -} - /* * Gets the GFN ready for another round of dirty logging by clearing the * - D bit on ad-enabled SPTEs, and @@ -1286,24 +1358,17 @@ static bool __rmap_clear_dirty(struct kvm *kvm, struct kvm_rmap_head *rmap_head, struct rmap_iterator iter; bool flush = false; - for_each_rmap_spte(rmap_head, &iter, sptep) + for_each_rmap_spte(rmap_head, &iter, sptep) { if (spte_ad_need_write_protect(*sptep)) - flush |= spte_wrprot_for_clear_dirty(sptep); + flush |= test_and_clear_bit(PT_WRITABLE_SHIFT, + (unsigned long *)sptep); else flush |= spte_clear_dirty(sptep); + } return flush; } -/** - * kvm_mmu_write_protect_pt_masked - write protect selected PT level pages - * @kvm: kvm instance - * @slot: slot to protect - * @gfn_offset: start of the BITS_PER_LONG pages we care about - * @mask: indicates which pages we should protect - * - * Used when we do not need to care about huge page mappings. - */ static void kvm_mmu_write_protect_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn_offset, unsigned long mask) @@ -1327,16 +1392,6 @@ static void kvm_mmu_write_protect_pt_masked(struct kvm *kvm, } } -/** - * kvm_mmu_clear_dirty_pt_masked - clear MMU D-bit for PT level pages, or write - * protect the page if the D-bit isn't supported. - * @kvm: kvm instance - * @slot: slot to clear D-bit - * @gfn_offset: start of the BITS_PER_LONG pages we care about - * @mask: indicates which pages we should clear D-bit - * - * Used for PML to re-log the dirty GPAs after userspace querying dirty_bitmap. - */ static void kvm_mmu_clear_dirty_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn_offset, unsigned long mask) @@ -1360,24 +1415,16 @@ static void kvm_mmu_clear_dirty_pt_masked(struct kvm *kvm, } } -/** - * kvm_arch_mmu_enable_log_dirty_pt_masked - enable dirty logging for selected - * PT level pages. - * - * It calls kvm_mmu_write_protect_pt_masked to write protect selected pages to - * enable dirty logging for them. - * - * We need to care about huge page mappings: e.g. during dirty logging we may - * have such mappings. - */ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn_offset, unsigned long mask) { /* - * Huge pages are NOT write protected when we start dirty logging in - * initially-all-set mode; must write protect them here so that they - * are split to 4K on the first write. + * If the slot was assumed to be "initially all dirty", write-protect + * huge pages to ensure they are split to 4KiB on the first write (KVM + * dirty logs at 4KiB granularity). If eager page splitting is enabled, + * immediately try to split huge pages, e.g. so that vCPUs don't get + * saddled with the cost of splitting. * * The gfn_offset is guaranteed to be aligned to 64, but the base_gfn * of memslot has no such restriction, so the range can cross two large @@ -1399,7 +1446,16 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, PG_LEVEL_2M); } - /* Now handle 4K PTEs. */ + /* + * (Re)Enable dirty logging for all 4KiB SPTEs that map the GFNs in + * mask. If PML is enabled and the GFN doesn't need to be write- + * protected for other reasons, e.g. shadow paging, clear the Dirty bit. + * Otherwise clear the Writable bit. + * + * Note that kvm_mmu_clear_dirty_pt_masked() is called whenever PML is + * enabled but it chooses between clearing the Dirty bit and Writeable + * bit based on the context. + */ if (kvm_x86_ops.cpu_dirty_log_size) kvm_mmu_clear_dirty_pt_masked(kvm, slot, gfn_offset, mask); else @@ -1441,54 +1497,10 @@ static bool kvm_vcpu_write_protect_gfn(struct kvm_vcpu *vcpu, u64 gfn) return kvm_mmu_slot_gfn_write_protect(vcpu->kvm, slot, gfn, PG_LEVEL_4K); } -static bool __kvm_zap_rmap(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - const struct kvm_memory_slot *slot) -{ - return kvm_zap_all_rmap_sptes(kvm, rmap_head); -} - static bool kvm_zap_rmap(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - struct kvm_memory_slot *slot, gfn_t gfn, int level, - pte_t unused) -{ - return __kvm_zap_rmap(kvm, rmap_head, slot); -} - -static bool kvm_set_pte_rmap(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - struct kvm_memory_slot *slot, gfn_t gfn, int level, - pte_t pte) + const struct kvm_memory_slot *slot) { - u64 *sptep; - struct rmap_iterator iter; - bool need_flush = false; - u64 new_spte; - kvm_pfn_t new_pfn; - - WARN_ON_ONCE(pte_huge(pte)); - new_pfn = pte_pfn(pte); - -restart: - for_each_rmap_spte(rmap_head, &iter, sptep) { - need_flush = true; - - if (pte_write(pte)) { - kvm_zap_one_rmap_spte(kvm, rmap_head, sptep); - goto restart; - } else { - new_spte = kvm_mmu_changed_pte_notifier_make_spte( - *sptep, new_pfn); - - mmu_spte_clear_track_bits(kvm, sptep); - mmu_spte_set(sptep, new_spte); - } - } - - if (need_flush && kvm_available_flush_remote_tlbs_range()) { - kvm_flush_remote_tlbs_gfn(kvm, gfn, level); - return false; - } - - return need_flush; + return kvm_zap_all_rmap_sptes(kvm, rmap_head); } struct slot_rmap_walk_iterator { @@ -1539,9 +1551,9 @@ static bool slot_rmap_walk_okay(struct slot_rmap_walk_iterator *iterator) static void slot_rmap_walk_next(struct slot_rmap_walk_iterator *iterator) { while (++iterator->rmap <= iterator->end_rmap) { - iterator->gfn += (1UL << KVM_HPAGE_GFN_SHIFT(iterator->level)); + iterator->gfn += KVM_PAGES_PER_HPAGE(iterator->level); - if (iterator->rmap->val) + if (atomic_long_read(&iterator->rmap->val)) return; } @@ -1560,80 +1572,101 @@ static void slot_rmap_walk_next(struct slot_rmap_walk_iterator *iterator) slot_rmap_walk_okay(_iter_); \ slot_rmap_walk_next(_iter_)) -typedef bool (*rmap_handler_t)(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - struct kvm_memory_slot *slot, gfn_t gfn, - int level, pte_t pte); +/* The return value indicates if tlb flush on all vcpus is needed. */ +typedef bool (*slot_rmaps_handler) (struct kvm *kvm, + struct kvm_rmap_head *rmap_head, + const struct kvm_memory_slot *slot); -static __always_inline bool kvm_handle_gfn_range(struct kvm *kvm, - struct kvm_gfn_range *range, - rmap_handler_t handler) +static __always_inline bool __walk_slot_rmaps(struct kvm *kvm, + const struct kvm_memory_slot *slot, + slot_rmaps_handler fn, + int start_level, int end_level, + gfn_t start_gfn, gfn_t end_gfn, + bool can_yield, bool flush_on_yield, + bool flush) { struct slot_rmap_walk_iterator iterator; - bool ret = false; - - for_each_slot_rmap_range(range->slot, PG_LEVEL_4K, KVM_MAX_HUGEPAGE_LEVEL, - range->start, range->end - 1, &iterator) - ret |= handler(kvm, iterator.rmap, range->slot, iterator.gfn, - iterator.level, range->arg.pte); - - return ret; -} -bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) -{ - bool flush = false; + lockdep_assert_held_write(&kvm->mmu_lock); - if (kvm_memslots_have_rmaps(kvm)) - flush = kvm_handle_gfn_range(kvm, range, kvm_zap_rmap); + for_each_slot_rmap_range(slot, start_level, end_level, start_gfn, + end_gfn, &iterator) { + if (iterator.rmap) + flush |= fn(kvm, iterator.rmap, slot); - if (tdp_mmu_enabled) - flush = kvm_tdp_mmu_unmap_gfn_range(kvm, range, flush); + if (!can_yield) + continue; - if (kvm_x86_ops.set_apic_access_page_addr && - range->slot->id == APIC_ACCESS_PAGE_PRIVATE_MEMSLOT) - kvm_make_all_cpus_request(kvm, KVM_REQ_APIC_PAGE_RELOAD); + if (need_resched() || rwlock_needbreak(&kvm->mmu_lock)) { + if (flush && flush_on_yield) { + kvm_flush_remote_tlbs_range(kvm, start_gfn, + iterator.gfn - start_gfn + 1); + flush = false; + } + cond_resched_rwlock_write(&kvm->mmu_lock); + } + } return flush; } -bool kvm_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range) +static __always_inline bool walk_slot_rmaps(struct kvm *kvm, + const struct kvm_memory_slot *slot, + slot_rmaps_handler fn, + int start_level, int end_level, + bool flush_on_yield) { - bool flush = false; - - if (kvm_memslots_have_rmaps(kvm)) - flush = kvm_handle_gfn_range(kvm, range, kvm_set_pte_rmap); + return __walk_slot_rmaps(kvm, slot, fn, start_level, end_level, + slot->base_gfn, slot->base_gfn + slot->npages - 1, + true, flush_on_yield, false); +} - if (tdp_mmu_enabled) - flush |= kvm_tdp_mmu_set_spte_gfn(kvm, range); +static __always_inline bool walk_slot_rmaps_4k(struct kvm *kvm, + const struct kvm_memory_slot *slot, + slot_rmaps_handler fn, + bool flush_on_yield) +{ + return walk_slot_rmaps(kvm, slot, fn, PG_LEVEL_4K, PG_LEVEL_4K, flush_on_yield); +} - return flush; +static bool __kvm_rmap_zap_gfn_range(struct kvm *kvm, + const struct kvm_memory_slot *slot, + gfn_t start, gfn_t end, bool can_yield, + bool flush) +{ + return __walk_slot_rmaps(kvm, slot, kvm_zap_rmap, + PG_LEVEL_4K, KVM_MAX_HUGEPAGE_LEVEL, + start, end - 1, can_yield, true, flush); } -static bool kvm_age_rmap(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - struct kvm_memory_slot *slot, gfn_t gfn, int level, - pte_t unused) +bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) { - u64 *sptep; - struct rmap_iterator iter; - int young = 0; + bool flush = false; - for_each_rmap_spte(rmap_head, &iter, sptep) - young |= mmu_spte_age(sptep); + /* + * To prevent races with vCPUs faulting in a gfn using stale data, + * zapping a gfn range must be protected by mmu_invalidate_in_progress + * (and mmu_invalidate_seq). The only exception is memslot deletion; + * in that case, SRCU synchronization ensures that SPTEs are zapped + * after all vCPUs have unlocked SRCU, guaranteeing that vCPUs see the + * invalid slot. + */ + lockdep_assert_once(kvm->mmu_invalidate_in_progress || + lockdep_is_held(&kvm->slots_lock)); - return young; -} + if (kvm_memslots_have_rmaps(kvm)) + flush = __kvm_rmap_zap_gfn_range(kvm, range->slot, + range->start, range->end, + range->may_block, flush); -static bool kvm_test_age_rmap(struct kvm *kvm, struct kvm_rmap_head *rmap_head, - struct kvm_memory_slot *slot, gfn_t gfn, - int level, pte_t unused) -{ - u64 *sptep; - struct rmap_iterator iter; + if (tdp_mmu_enabled) + flush = kvm_tdp_mmu_unmap_gfn_range(kvm, range, flush); - for_each_rmap_spte(rmap_head, &iter, sptep) - if (is_accessed_spte(*sptep)) - return true; - return false; + if (kvm_x86_ops.set_apic_access_page_addr && + range->slot->id == APIC_ACCESS_PAGE_PRIVATE_MEMSLOT) + kvm_make_all_cpus_request(kvm, KVM_REQ_APIC_PAGE_RELOAD); + + return flush; } #define RMAP_RECYCLE_THRESHOLD 1000 @@ -1652,7 +1685,7 @@ static void __rmap_add(struct kvm *kvm, kvm_update_page_stats(kvm, sp->role.level, 1); rmap_head = gfn_to_rmap(gfn, sp->role.level, slot); - rmap_count = pte_list_add(cache, spte, rmap_head); + rmap_count = pte_list_add(kvm, cache, spte, rmap_head); if (rmap_count > kvm->stat.max_mmu_rmap_size) kvm->stat.max_mmu_rmap_size = rmap_count; @@ -1670,15 +1703,68 @@ static void rmap_add(struct kvm_vcpu *vcpu, const struct kvm_memory_slot *slot, __rmap_add(vcpu->kvm, cache, slot, spte, gfn, access); } -bool kvm_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range) +static bool kvm_rmap_age_gfn_range(struct kvm *kvm, + struct kvm_gfn_range *range, + bool test_only) { + struct kvm_rmap_head *rmap_head; + struct rmap_iterator iter; + unsigned long rmap_val; bool young = false; + u64 *sptep; + gfn_t gfn; + int level; + u64 spte; - if (kvm_memslots_have_rmaps(kvm)) - young = kvm_handle_gfn_range(kvm, range, kvm_age_rmap); + for (level = PG_LEVEL_4K; level <= KVM_MAX_HUGEPAGE_LEVEL; level++) { + for (gfn = range->start; gfn < range->end; + gfn += KVM_PAGES_PER_HPAGE(level)) { + rmap_head = gfn_to_rmap(gfn, level, range->slot); + rmap_val = kvm_rmap_lock_readonly(rmap_head); + + for_each_rmap_spte_lockless(rmap_head, &iter, sptep, spte) { + if (!is_accessed_spte(spte)) + continue; + + if (test_only) { + kvm_rmap_unlock_readonly(rmap_head, rmap_val); + return true; + } + + if (spte_ad_enabled(spte)) + clear_bit((ffs(shadow_accessed_mask) - 1), + (unsigned long *)sptep); + else + /* + * If the following cmpxchg fails, the + * spte is being concurrently modified + * and should most likely stay young. + */ + cmpxchg64(sptep, spte, + mark_spte_for_access_track(spte)); + young = true; + } + + kvm_rmap_unlock_readonly(rmap_head, rmap_val); + } + } + return young; +} + +static bool kvm_may_have_shadow_mmu_sptes(struct kvm *kvm) +{ + return !tdp_mmu_enabled || READ_ONCE(kvm->arch.indirect_shadow_pages); +} + +bool kvm_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range) +{ + bool young = false; if (tdp_mmu_enabled) - young |= kvm_tdp_mmu_age_gfn_range(kvm, range); + young = kvm_tdp_mmu_age_gfn_range(kvm, range); + + if (kvm_may_have_shadow_mmu_sptes(kvm)) + young |= kvm_rmap_age_gfn_range(kvm, range, false); return young; } @@ -1687,11 +1773,14 @@ bool kvm_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range) { bool young = false; - if (kvm_memslots_have_rmaps(kvm)) - young = kvm_handle_gfn_range(kvm, range, kvm_test_age_rmap); - if (tdp_mmu_enabled) - young |= kvm_tdp_mmu_test_age_gfn(kvm, range); + young = kvm_tdp_mmu_test_age_gfn(kvm, range); + + if (young) + return young; + + if (kvm_may_have_shadow_mmu_sptes(kvm)) + young |= kvm_rmap_age_gfn_range(kvm, range, true); return young; } @@ -1710,27 +1799,15 @@ static void kvm_mmu_check_sptes_at_free(struct kvm_mmu_page *sp) #endif } -/* - * This value is the sum of all of the kvm instances's - * kvm->arch.n_used_mmu_pages values. We need a global, - * aggregate version in order to make the slab shrinker - * faster - */ -static inline void kvm_mod_used_mmu_pages(struct kvm *kvm, long nr) -{ - kvm->arch.n_used_mmu_pages += nr; - percpu_counter_add(&kvm_total_used_mmu_pages, nr); -} - static void kvm_account_mmu_page(struct kvm *kvm, struct kvm_mmu_page *sp) { - kvm_mod_used_mmu_pages(kvm, +1); + kvm->arch.n_used_mmu_pages++; kvm_account_pgtable_pages((void *)sp->spt, +1); } static void kvm_unaccount_mmu_page(struct kvm *kvm, struct kvm_mmu_page *sp) { - kvm_mod_used_mmu_pages(kvm, -1); + kvm->arch.n_used_mmu_pages--; kvm_account_pgtable_pages((void *)sp->spt, -1); } @@ -1741,8 +1818,7 @@ static void kvm_mmu_free_shadow_page(struct kvm_mmu_page *sp) hlist_del(&sp->hash_link); list_del(&sp->link); free_page((unsigned long)sp->spt); - if (!sp->role.direct) - free_page((unsigned long)sp->shadowed_translation); + free_page((unsigned long)sp->shadowed_translation); kmem_cache_free(mmu_page_header_cache, sp); } @@ -1751,13 +1827,14 @@ static unsigned kvm_page_table_hashfn(gfn_t gfn) return hash_64(gfn, KVM_MMU_HASH_SHIFT); } -static void mmu_page_add_parent_pte(struct kvm_mmu_memory_cache *cache, +static void mmu_page_add_parent_pte(struct kvm *kvm, + struct kvm_mmu_memory_cache *cache, struct kvm_mmu_page *sp, u64 *parent_pte) { if (!parent_pte) return; - pte_list_add(cache, parent_pte, &sp->parent_ptes); + pte_list_add(kvm, cache, parent_pte, &sp->parent_ptes); } static void mmu_page_remove_parent_pte(struct kvm *kvm, struct kvm_mmu_page *sp, @@ -1950,7 +2027,8 @@ static bool kvm_sync_page_check(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) static int kvm_sync_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, int i) { - if (!sp->spt[i]) + /* sp->spt[i] has initial value of shadow page table allocation */ + if (sp->spt[i] == SHADOW_NONPRESENT_VALUE) return 0; return vcpu->arch.mmu->sync_spte(vcpu, sp, i); @@ -2243,7 +2321,7 @@ static struct kvm_mmu_page *kvm_mmu_alloc_shadow_page(struct kvm *kvm, sp = kvm_mmu_memory_cache_alloc(caches->page_header_cache); sp->spt = kvm_mmu_memory_cache_alloc(caches->shadow_page_cache); - if (!role.direct) + if (!role.direct && role.level <= KVM_MAX_HUGEPAGE_LEVEL) sp->shadowed_translation = kvm_mmu_memory_cache_alloc(caches->shadowed_info_cache); set_page_private(virt_to_page(sp->spt), (unsigned long)sp); @@ -2446,7 +2524,7 @@ static void __link_shadow_page(struct kvm *kvm, mmu_spte_set(sptep, spte); - mmu_page_add_parent_pte(cache, sp, sptep); + mmu_page_add_parent_pte(kvm, cache, sp, sptep); /* * The non-direct sub-pagetable must be updated before linking. For @@ -2510,11 +2588,12 @@ static int mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp, * avoids retaining a large number of stale nested SPs. */ if (tdp_enabled && invalid_list && - child->role.guest_mode && !child->parent_ptes.val) + child->role.guest_mode && + !atomic_long_read(&child->parent_ptes.val)) return kvm_mmu_prepare_zap_page(kvm, child, invalid_list); } - } else if (is_mmio_spte(pte)) { + } else if (is_mmio_spte(kvm, pte)) { mmu_spte_clear_no_track(spte); } return 0; @@ -2754,36 +2833,49 @@ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned long goal_nr_mmu_pages) write_unlock(&kvm->mmu_lock); } -int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) +bool __kvm_mmu_unprotect_gfn_and_retry(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, + bool always_retry) { - struct kvm_mmu_page *sp; + struct kvm *kvm = vcpu->kvm; LIST_HEAD(invalid_list); - int r; + struct kvm_mmu_page *sp; + gpa_t gpa = cr2_or_gpa; + bool r = false; + + /* + * Bail early if there aren't any write-protected shadow pages to avoid + * unnecessarily taking mmu_lock lock, e.g. if the gfn is write-tracked + * by a third party. Reading indirect_shadow_pages without holding + * mmu_lock is safe, as this is purely an optimization, i.e. a false + * positive is benign, and a false negative will simply result in KVM + * skipping the unprotect+retry path, which is also an optimization. + */ + if (!READ_ONCE(kvm->arch.indirect_shadow_pages)) + goto out; + + if (!vcpu->arch.mmu->root_role.direct) { + gpa = kvm_mmu_gva_to_gpa_write(vcpu, cr2_or_gpa, NULL); + if (gpa == INVALID_GPA) + goto out; + } - r = 0; write_lock(&kvm->mmu_lock); - for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) { - r = 1; + for_each_gfn_valid_sp_with_gptes(kvm, sp, gpa_to_gfn(gpa)) kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list); - } + + /* + * Snapshot the result before zapping, as zapping will remove all list + * entries, i.e. checking the list later would yield a false negative. + */ + r = !list_empty(&invalid_list); kvm_mmu_commit_zap_page(kvm, &invalid_list); write_unlock(&kvm->mmu_lock); - return r; -} - -static int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva) -{ - gpa_t gpa; - int r; - - if (vcpu->arch.mmu->root_role.direct) - return 0; - - gpa = kvm_mmu_gva_to_gpa_read(vcpu, gva, NULL); - - r = kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT); - +out: + if (r || always_retry) { + vcpu->arch.last_retry_eip = kvm_rip_read(vcpu); + vcpu->arch.last_retry_addr = cr2_or_gpa; + } return r; } @@ -2803,7 +2895,7 @@ static void kvm_unsync_page(struct kvm *kvm, struct kvm_mmu_page *sp) * be write-protected. */ int mmu_try_to_unsync_pages(struct kvm *kvm, const struct kvm_memory_slot *slot, - gfn_t gfn, bool can_unsync, bool prefetch) + gfn_t gfn, bool synchronizing, bool prefetch) { struct kvm_mmu_page *sp; bool locked = false; @@ -2818,12 +2910,12 @@ int mmu_try_to_unsync_pages(struct kvm *kvm, const struct kvm_memory_slot *slot, /* * The page is not write-tracked, mark existing shadow pages unsync - * unless KVM is synchronizing an unsync SP (can_unsync = false). In - * that case, KVM must complete emulation of the guest TLB flush before - * allowing shadow pages to become unsync (writable by the guest). + * unless KVM is synchronizing an unsync SP. In that case, KVM must + * complete emulation of the guest TLB flush before allowing shadow + * pages to become unsync (writable by the guest). */ for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) { - if (!can_unsync) + if (synchronizing) return -EPERM; if (sp->unsync) @@ -2927,6 +3019,9 @@ static int mmu_set_spte(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot, } if (is_shadow_present_pte(*sptep)) { + if (prefetch) + return RET_PF_SPURIOUS; + /* * If we overwrite a PTE page pointer with a 2MB PMD, unlink * the parent of the now unreachable PTE. @@ -2946,7 +3041,7 @@ static int mmu_set_spte(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot, } wrprot = make_spte(vcpu, sp, slot, pte_access, gfn, pfn, *sptep, prefetch, - true, host_writable, &spte); + false, host_writable, &spte); if (*sptep == spte) { ret = RET_PF_SPURIOUS; @@ -2955,10 +3050,8 @@ static int mmu_set_spte(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot, trace_kvm_mmu_set_spte(level, gfn, sptep); } - if (wrprot) { - if (write_fault) - ret = RET_PF_EMULATE; - } + if (wrprot && write_fault) + ret = RET_PF_WRITE_PROTECTED; if (flush) kvm_flush_remote_tlbs_gfn(vcpu->kvm, gfn, level); @@ -2974,32 +3067,51 @@ static int mmu_set_spte(struct kvm_vcpu *vcpu, struct kvm_memory_slot *slot, return ret; } -static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu, - struct kvm_mmu_page *sp, - u64 *start, u64 *end) +static bool kvm_mmu_prefetch_sptes(struct kvm_vcpu *vcpu, gfn_t gfn, u64 *sptep, + int nr_pages, unsigned int access) { struct page *pages[PTE_PREFETCH_NUM]; struct kvm_memory_slot *slot; - unsigned int access = sp->role.access; - int i, ret; - gfn_t gfn; + int i; + + if (WARN_ON_ONCE(nr_pages > PTE_PREFETCH_NUM)) + return false; - gfn = kvm_mmu_page_get_gfn(sp, spte_index(start)); slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, access & ACC_WRITE_MASK); if (!slot) - return -1; + return false; - ret = gfn_to_page_many_atomic(slot, gfn, pages, end - start); - if (ret <= 0) - return -1; + nr_pages = kvm_prefetch_pages(slot, gfn, pages, nr_pages); + if (nr_pages <= 0) + return false; - for (i = 0; i < ret; i++, gfn++, start++) { - mmu_set_spte(vcpu, slot, start, access, gfn, + for (i = 0; i < nr_pages; i++, gfn++, sptep++) { + mmu_set_spte(vcpu, slot, sptep, access, gfn, page_to_pfn(pages[i]), NULL); - put_page(pages[i]); + + /* + * KVM always prefetches writable pages from the primary MMU, + * and KVM can make its SPTE writable in the fast page handler, + * without notifying the primary MMU. Mark pages/folios dirty + * now to ensure file data is written back if it ends up being + * written by the guest. Because KVM's prefetching GUPs + * writable PTEs, the probability of unnecessary writeback is + * extremely low. + */ + kvm_release_page_dirty(pages[i]); } - return 0; + return true; +} + +static bool direct_pte_prefetch_many(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp, + u64 *start, u64 *end) +{ + gfn_t gfn = kvm_mmu_page_get_gfn(sp, spte_index(start)); + unsigned int access = sp->role.access; + + return kvm_mmu_prefetch_sptes(vcpu, gfn, start, end - start, access); } static void __direct_pte_prefetch(struct kvm_vcpu *vcpu, @@ -3017,8 +3129,9 @@ static void __direct_pte_prefetch(struct kvm_vcpu *vcpu, if (is_shadow_present_pte(*spte) || spte == sptep) { if (!start) continue; - if (direct_pte_prefetch_many(vcpu, sp, start, spte) < 0) + if (!direct_pte_prefetch_many(vcpu, sp, start, spte)) return; + start = NULL; } else if (!start) start = spte; @@ -3168,13 +3281,12 @@ static int __kvm_mmu_max_mapping_level(struct kvm *kvm, } int kvm_mmu_max_mapping_level(struct kvm *kvm, - const struct kvm_memory_slot *slot, gfn_t gfn, - int max_level) + const struct kvm_memory_slot *slot, gfn_t gfn) { bool is_private = kvm_slot_can_be_private(slot) && kvm_mem_is_private(kvm, gfn); - return __kvm_mmu_max_mapping_level(kvm, slot, gfn, max_level, is_private); + return __kvm_mmu_max_mapping_level(kvm, slot, gfn, PG_LEVEL_NUM, is_private); } void kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) @@ -3314,9 +3426,18 @@ static int kvm_handle_noslot_fault(struct kvm_vcpu *vcpu, { gva_t gva = fault->is_tdp ? 0 : fault->addr; + if (fault->is_private) { + kvm_mmu_prepare_memory_fault_exit(vcpu, fault); + return -EFAULT; + } + vcpu_cache_mmio_info(vcpu, gva, fault->gfn, access & shadow_mmio_access_mask); + fault->slot = NULL; + fault->pfn = KVM_PFN_NOSLOT; + fault->map_writable = false; + /* * If MMIO caching is disabled, emulate immediately without * touching the shadow page tables as attempting to install an @@ -3338,7 +3459,7 @@ static int kvm_handle_noslot_fault(struct kvm_vcpu *vcpu, return RET_PF_CONTINUE; } -static bool page_fault_can_be_fast(struct kvm_page_fault *fault) +static bool page_fault_can_be_fast(struct kvm *kvm, struct kvm_page_fault *fault) { /* * Page faults with reserved bits set, i.e. faults on MMIO SPTEs, only @@ -3350,6 +3471,26 @@ static bool page_fault_can_be_fast(struct kvm_page_fault *fault) return false; /* + * For hardware-protected VMs, certain conditions like attempting to + * perform a write to a page which is not in the state that the guest + * expects it to be in can result in a nested/extended #PF. In this + * case, the below code might misconstrue this situation as being the + * result of a write-protected access, and treat it as a spurious case + * rather than taking any action to satisfy the real source of the #PF + * such as generating a KVM_EXIT_MEMORY_FAULT. This can lead to the + * guest spinning on a #PF indefinitely, so don't attempt the fast path + * in this case. + * + * Note that the kvm_mem_is_private() check might race with an + * attribute update, but this will either result in the guest spinning + * on RET_PF_SPURIOUS until the update completes, or an actual spurious + * case might go down the slow path. Either case will resolve itself. + */ + if (kvm->arch.has_private_mem && + fault->is_private != kvm_mem_is_private(kvm, fault->gfn)) + return false; + + /* * #PF can be fast if: * * 1. The shadow page table entry is not present and A/D bits are @@ -3365,7 +3506,7 @@ static bool page_fault_can_be_fast(struct kvm_page_fault *fault) * by setting the Writable bit, which can be done out of mmu_lock. */ if (!fault->present) - return !kvm_ad_enabled(); + return !kvm_ad_enabled; /* * Note, instruction fetches and writes are mutually exclusive, ignore @@ -3392,7 +3533,7 @@ static bool fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, * harm. This also avoids the TLB flush needed after setting dirty bit * so non-PML cases won't be impacted. * - * Compare with set_spte where instead shadow_dirty_mask is set. + * Compare with make_spte() where instead shadow_dirty_mask is set. */ if (!try_cmpxchg64(sptep, &old_spte, new_spte)) return false; @@ -3403,18 +3544,6 @@ static bool fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, return true; } -static bool is_access_allowed(struct kvm_page_fault *fault, u64 spte) -{ - if (fault->exec) - return is_executable_pte(spte); - - if (fault->write) - return is_writable_pte(spte); - - /* Fault was on Read access */ - return spte & PT_PRESENT_MASK; -} - /* * Returns the last level spte pointer of the shadow page walk for the given * gpa, and sets *spte to the spte value. This spte may be non-preset. If no @@ -3449,7 +3578,7 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) u64 *sptep; uint retry_count = 0; - if (!page_fault_can_be_fast(fault)) + if (!page_fault_can_be_fast(vcpu->kvm, fault)) return ret; walk_shadow_page_lockless_begin(vcpu); @@ -3458,7 +3587,7 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) u64 new_spte; if (tdp_mmu_enabled) - sptep = kvm_tdp_mmu_fast_pf_get_last_sptep(vcpu, fault->addr, &spte); + sptep = kvm_tdp_mmu_fast_pf_get_last_sptep(vcpu, fault->gfn, &spte); else sptep = fast_pf_get_last_sptep(vcpu, fault->addr, &spte); @@ -3468,7 +3597,7 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) * available as the vCPU holds a reference to its root(s). */ if (WARN_ON_ONCE(!sptep)) - spte = REMOVED_SPTE; + spte = FROZEN_SPTE; if (!is_shadow_present_pte(spte)) break; @@ -3500,8 +3629,9 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) * uses A/D bits for non-nested MMUs. Thus, if A/D bits are * enabled, the SPTE can't be an access-tracked SPTE. */ - if (unlikely(!kvm_ad_enabled()) && is_access_track_spte(spte)) - new_spte = restore_acc_track_spte(new_spte); + if (unlikely(!kvm_ad_enabled) && is_access_track_spte(spte)) + new_spte = restore_acc_track_spte(new_spte) | + shadow_accessed_mask; /* * To keep things simple, only SPTEs that are MMU-writable can @@ -3706,8 +3836,13 @@ static int mmu_alloc_direct_roots(struct kvm_vcpu *vcpu) unsigned i; int r; - if (tdp_mmu_enabled) - return kvm_tdp_mmu_alloc_root(vcpu); + if (tdp_mmu_enabled) { + if (kvm_has_mirrored_tdp(vcpu->kvm) && + !VALID_PAGE(mmu->mirror_root_hpa)) + kvm_tdp_mmu_alloc_root(vcpu, true); + kvm_tdp_mmu_alloc_root(vcpu, false); + return 0; + } write_lock(&vcpu->kvm->mmu_lock); r = make_mmu_pages_available(vcpu); @@ -4134,23 +4269,31 @@ static int get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes, int *root_level return leaf; } -/* return true if reserved bit(s) are detected on a valid, non-MMIO SPTE. */ -static bool get_mmio_spte(struct kvm_vcpu *vcpu, u64 addr, u64 *sptep) +static int get_sptes_lockless(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes, + int *root_level) { - u64 sptes[PT64_ROOT_MAX_LEVEL + 1]; - struct rsvd_bits_validate *rsvd_check; - int root, leaf, level; - bool reserved = false; + int leaf; walk_shadow_page_lockless_begin(vcpu); if (is_tdp_mmu_active(vcpu)) - leaf = kvm_tdp_mmu_get_walk(vcpu, addr, sptes, &root); + leaf = kvm_tdp_mmu_get_walk(vcpu, addr, sptes, root_level); else - leaf = get_walk(vcpu, addr, sptes, &root); + leaf = get_walk(vcpu, addr, sptes, root_level); walk_shadow_page_lockless_end(vcpu); + return leaf; +} + +/* return true if reserved bit(s) are detected on a valid, non-MMIO SPTE. */ +static bool get_mmio_spte(struct kvm_vcpu *vcpu, u64 addr, u64 *sptep) +{ + u64 sptes[PT64_ROOT_MAX_LEVEL + 1]; + struct rsvd_bits_validate *rsvd_check; + int root, leaf, level; + bool reserved = false; + leaf = get_sptes_lockless(vcpu, addr, sptes, &root); if (unlikely(leaf < 0)) { *sptep = 0ull; return reserved; @@ -4196,7 +4339,7 @@ static int handle_mmio_page_fault(struct kvm_vcpu *vcpu, u64 addr, bool direct) if (WARN_ON_ONCE(reserved)) return -EINVAL; - if (is_mmio_spte(spte)) { + if (is_mmio_spte(vcpu->kvm, spte)) { gfn_t gfn = get_mmio_spte_gfn(spte); unsigned int access = get_mmio_spte_access(spte); @@ -4259,24 +4402,28 @@ static u32 alloc_apf_token(struct kvm_vcpu *vcpu) return (vcpu->arch.apf.id++ << 12) | vcpu->vcpu_id; } -static bool kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, - gfn_t gfn) +static bool kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault) { struct kvm_arch_async_pf arch; arch.token = alloc_apf_token(vcpu); - arch.gfn = gfn; + arch.gfn = fault->gfn; + arch.error_code = fault->error_code; arch.direct_map = vcpu->arch.mmu->root_role.direct; arch.cr3 = kvm_mmu_get_guest_pgd(vcpu, vcpu->arch.mmu); - return kvm_setup_async_pf(vcpu, cr2_or_gpa, - kvm_vcpu_gfn_to_hva(vcpu, gfn), &arch); + return kvm_setup_async_pf(vcpu, fault->addr, + kvm_vcpu_gfn_to_hva(vcpu, fault->gfn), &arch); } void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { int r; + if (WARN_ON_ONCE(work->arch.error_code & PFERR_PRIVATE_ACCESS)) + return; + if ((vcpu->arch.mmu->root_role.direct != work->arch.direct_map) || work->wakeup_all) return; @@ -4289,7 +4436,16 @@ void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) work->arch.cr3 != kvm_mmu_get_guest_pgd(vcpu, vcpu->arch.mmu)) return; - kvm_mmu_do_page_fault(vcpu, work->cr2_or_gpa, 0, true, NULL); + r = kvm_mmu_do_page_fault(vcpu, work->cr2_or_gpa, work->arch.error_code, + true, NULL, NULL); + + /* + * Account fixed page faults, otherwise they'll never be counted, but + * ignore stats for all other return times. Page-ready "faults" aren't + * truly spurious and never trigger emulation + */ + if (r == RET_PF_FIXED) + vcpu->stat.pf_fixed++; } static inline u8 kvm_max_level_for_order(int order) @@ -4309,16 +4465,34 @@ static inline u8 kvm_max_level_for_order(int order) return PG_LEVEL_4K; } -static void kvm_mmu_prepare_memory_fault_exit(struct kvm_vcpu *vcpu, - struct kvm_page_fault *fault) +static u8 kvm_max_private_mapping_level(struct kvm *kvm, kvm_pfn_t pfn, + u8 max_level, int gmem_order) { - kvm_prepare_memory_fault_exit(vcpu, fault->gfn << PAGE_SHIFT, - PAGE_SIZE, fault->write, fault->exec, - fault->is_private); + u8 req_max_level; + + if (max_level == PG_LEVEL_4K) + return PG_LEVEL_4K; + + max_level = min(kvm_max_level_for_order(gmem_order), max_level); + if (max_level == PG_LEVEL_4K) + return PG_LEVEL_4K; + + req_max_level = kvm_x86_call(private_max_mapping_level)(kvm, pfn); + if (req_max_level) + max_level = min(max_level, req_max_level); + + return max_level; +} + +static void kvm_mmu_finish_page_fault(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault, int r) +{ + kvm_release_faultin_page(vcpu->kvm, fault->refcounted_page, + r == RET_PF_RETRY, fault->map_writable); } -static int kvm_faultin_pfn_private(struct kvm_vcpu *vcpu, - struct kvm_page_fault *fault) +static int kvm_mmu_faultin_pfn_private(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault) { int max_order, r; @@ -4328,65 +4502,39 @@ static int kvm_faultin_pfn_private(struct kvm_vcpu *vcpu, } r = kvm_gmem_get_pfn(vcpu->kvm, fault->slot, fault->gfn, &fault->pfn, - &max_order); + &fault->refcounted_page, &max_order); if (r) { kvm_mmu_prepare_memory_fault_exit(vcpu, fault); return r; } - fault->max_level = min(kvm_max_level_for_order(max_order), - fault->max_level); fault->map_writable = !(fault->slot->flags & KVM_MEM_READONLY); + fault->max_level = kvm_max_private_mapping_level(vcpu->kvm, fault->pfn, + fault->max_level, max_order); return RET_PF_CONTINUE; } -static int __kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) +static int __kvm_mmu_faultin_pfn(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault) { - struct kvm_memory_slot *slot = fault->slot; - bool async; - - /* - * Retry the page fault if the gfn hit a memslot that is being deleted - * or moved. This ensures any existing SPTEs for the old memslot will - * be zapped before KVM inserts a new MMIO SPTE for the gfn. - */ - if (slot && (slot->flags & KVM_MEMSLOT_INVALID)) - return RET_PF_RETRY; - - if (!kvm_is_visible_memslot(slot)) { - /* Don't expose private memslots to L2. */ - if (is_guest_mode(vcpu)) { - fault->slot = NULL; - fault->pfn = KVM_PFN_NOSLOT; - fault->map_writable = false; - return RET_PF_CONTINUE; - } - /* - * If the APIC access page exists but is disabled, go directly - * to emulation without caching the MMIO access or creating a - * MMIO SPTE. That way the cache doesn't need to be purged - * when the AVIC is re-enabled. - */ - if (slot && slot->id == APIC_ACCESS_PAGE_PRIVATE_MEMSLOT && - !kvm_apicv_activated(vcpu->kvm)) - return RET_PF_EMULATE; - } - - if (fault->is_private != kvm_mem_is_private(vcpu->kvm, fault->gfn)) { - kvm_mmu_prepare_memory_fault_exit(vcpu, fault); - return -EFAULT; - } + unsigned int foll = fault->write ? FOLL_WRITE : 0; if (fault->is_private) - return kvm_faultin_pfn_private(vcpu, fault); + return kvm_mmu_faultin_pfn_private(vcpu, fault); - async = false; - fault->pfn = __gfn_to_pfn_memslot(slot, fault->gfn, false, false, &async, - fault->write, &fault->map_writable, - &fault->hva); - if (!async) - return RET_PF_CONTINUE; /* *pfn has correct page already */ + foll |= FOLL_NOWAIT; + fault->pfn = __kvm_faultin_pfn(fault->slot, fault->gfn, foll, + &fault->map_writable, &fault->refcounted_page); + + /* + * If resolving the page failed because I/O is needed to fault-in the + * page, then either set up an asynchronous #PF to do the I/O, or if + * doing an async #PF isn't possible, retry with I/O allowed. All + * other failures are terminal, i.e. retrying won't help. + */ + if (fault->pfn != KVM_PFN_ERR_NEEDS_IO) + return RET_PF_CONTINUE; if (!fault->prefetch && kvm_can_do_async_pf(vcpu)) { trace_kvm_try_async_get_page(fault->addr, fault->gfn); @@ -4394,7 +4542,7 @@ static int __kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault trace_kvm_async_pf_repeated_fault(fault->addr, fault->gfn); kvm_make_request(KVM_REQ_APF_HALT, vcpu); return RET_PF_RETRY; - } else if (kvm_arch_setup_async_pf(vcpu, fault->addr, fault->gfn)) { + } else if (kvm_arch_setup_async_pf(vcpu, fault)) { return RET_PF_RETRY; } } @@ -4404,21 +4552,79 @@ static int __kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault * to wait for IO. Note, gup always bails if it is unable to quickly * get a page and a fatal signal, i.e. SIGKILL, is pending. */ - fault->pfn = __gfn_to_pfn_memslot(slot, fault->gfn, false, true, NULL, - fault->write, &fault->map_writable, - &fault->hva); + foll |= FOLL_INTERRUPTIBLE; + foll &= ~FOLL_NOWAIT; + fault->pfn = __kvm_faultin_pfn(fault->slot, fault->gfn, foll, + &fault->map_writable, &fault->refcounted_page); + return RET_PF_CONTINUE; } -static int kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, - unsigned int access) +static int kvm_mmu_faultin_pfn(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault, unsigned int access) { + struct kvm_memory_slot *slot = fault->slot; + struct kvm *kvm = vcpu->kvm; int ret; + if (KVM_BUG_ON(kvm_is_gfn_alias(kvm, fault->gfn), kvm)) + return -EFAULT; + + /* + * Note that the mmu_invalidate_seq also serves to detect a concurrent + * change in attributes. is_page_fault_stale() will detect an + * invalidation relate to fault->fn and resume the guest without + * installing a mapping in the page tables. + */ fault->mmu_seq = vcpu->kvm->mmu_invalidate_seq; smp_rmb(); /* + * Now that we have a snapshot of mmu_invalidate_seq we can check for a + * private vs. shared mismatch. + */ + if (fault->is_private != kvm_mem_is_private(kvm, fault->gfn)) { + kvm_mmu_prepare_memory_fault_exit(vcpu, fault); + return -EFAULT; + } + + if (unlikely(!slot)) + return kvm_handle_noslot_fault(vcpu, fault, access); + + /* + * Retry the page fault if the gfn hit a memslot that is being deleted + * or moved. This ensures any existing SPTEs for the old memslot will + * be zapped before KVM inserts a new MMIO SPTE for the gfn. + */ + if (slot->flags & KVM_MEMSLOT_INVALID) + return RET_PF_RETRY; + + if (slot->id == APIC_ACCESS_PAGE_PRIVATE_MEMSLOT) { + /* + * Don't map L1's APIC access page into L2, KVM doesn't support + * using APICv/AVIC to accelerate L2 accesses to L1's APIC, + * i.e. the access needs to be emulated. Emulating access to + * L1's APIC is also correct if L1 is accelerating L2's own + * virtual APIC, but for some reason L1 also maps _L1's_ APIC + * into L2. Note, vcpu_is_mmio_gpa() always treats access to + * the APIC as MMIO. Allow an MMIO SPTE to be created, as KVM + * uses different roots for L1 vs. L2, i.e. there is no danger + * of breaking APICv/AVIC for L1. + */ + if (is_guest_mode(vcpu)) + return kvm_handle_noslot_fault(vcpu, fault, access); + + /* + * If the APIC access page exists but is disabled, go directly + * to emulation without caching the MMIO access or creating a + * MMIO SPTE. That way the cache doesn't need to be purged + * when the AVIC is re-enabled. + */ + if (!kvm_apicv_activated(vcpu->kvm)) + return RET_PF_EMULATE; + } + + /* * Check for a relevant mmu_notifier invalidation event before getting * the pfn from the primary MMU, and before acquiring mmu_lock. * @@ -4439,18 +4645,17 @@ static int kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, * *guaranteed* to need to retry, i.e. waiting until mmu_lock is held * to detect retry guarantees the worst case latency for the vCPU. */ - if (fault->slot && - mmu_invalidate_retry_gfn_unsafe(vcpu->kvm, fault->mmu_seq, fault->gfn)) + if (mmu_invalidate_retry_gfn_unsafe(kvm, fault->mmu_seq, fault->gfn)) return RET_PF_RETRY; - ret = __kvm_faultin_pfn(vcpu, fault); + ret = __kvm_mmu_faultin_pfn(vcpu, fault); if (ret != RET_PF_CONTINUE) return ret; if (unlikely(is_error_pfn(fault->pfn))) return kvm_handle_error_pfn(vcpu, fault); - if (unlikely(!fault->slot)) + if (WARN_ON_ONCE(!fault->slot || is_noslot_pfn(fault->pfn))) return kvm_handle_noslot_fault(vcpu, fault, access); /* @@ -4460,8 +4665,8 @@ static int kvm_faultin_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, * overall cost of failing to detect the invalidation until after * mmu_lock is acquired. */ - if (mmu_invalidate_retry_gfn_unsafe(vcpu->kvm, fault->mmu_seq, fault->gfn)) { - kvm_release_pfn_clean(fault->pfn); + if (mmu_invalidate_retry_gfn_unsafe(kvm, fault->mmu_seq, fault->gfn)) { + kvm_mmu_finish_page_fault(vcpu, fault, RET_PF_RETRY); return RET_PF_RETRY; } @@ -4510,7 +4715,7 @@ static int direct_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault return RET_PF_RETRY; if (page_fault_handle_page_track(vcpu, fault)) - return RET_PF_EMULATE; + return RET_PF_WRITE_PROTECTED; r = fast_page_fault(vcpu, fault); if (r != RET_PF_INVALID) @@ -4520,7 +4725,7 @@ static int direct_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault if (r) return r; - r = kvm_faultin_pfn(vcpu, fault, ACC_ALL); + r = kvm_mmu_faultin_pfn(vcpu, fault, ACC_ALL); if (r != RET_PF_CONTINUE) return r; @@ -4537,8 +4742,8 @@ static int direct_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault r = direct_map(vcpu, fault); out_unlock: + kvm_mmu_finish_page_fault(vcpu, fault, r); write_unlock(&vcpu->kvm->mmu_lock); - kvm_release_pfn_clean(fault->pfn); return r; } @@ -4561,13 +4766,24 @@ int kvm_handle_page_fault(struct kvm_vcpu *vcpu, u64 error_code, if (WARN_ON_ONCE(fault_address >> 32)) return -EFAULT; #endif + /* + * Legacy #PF exception only have a 32-bit error code. Simply drop the + * upper bits as KVM doesn't use them for #PF (because they are never + * set), and to ensure there are no collisions with KVM-defined bits. + */ + if (WARN_ON_ONCE(error_code >> 32)) + error_code = lower_32_bits(error_code); + + /* + * Restrict KVM-defined flags to bits 63:32 so that it's impossible for + * them to conflict with #PF error codes, which are limited to 32 bits. + */ + BUILD_BUG_ON(lower_32_bits(PFERR_SYNTHETIC_MASK)); vcpu->arch.l1tf_flush_l1d = true; if (!flags) { trace_kvm_page_fault(vcpu, fault_address, error_code); - if (kvm_event_needs_reinjection(vcpu)) - kvm_mmu_unprotect_page_virt(vcpu, fault_address); r = kvm_mmu_page_fault(vcpu, fault_address, error_code, insn, insn_len); } else if (flags & KVM_PV_REASON_PAGE_NOT_PRESENT) { @@ -4590,7 +4806,7 @@ static int kvm_tdp_mmu_page_fault(struct kvm_vcpu *vcpu, int r; if (page_fault_handle_page_track(vcpu, fault)) - return RET_PF_EMULATE; + return RET_PF_WRITE_PROTECTED; r = fast_page_fault(vcpu, fault); if (r != RET_PF_INVALID) @@ -4600,7 +4816,7 @@ static int kvm_tdp_mmu_page_fault(struct kvm_vcpu *vcpu, if (r) return r; - r = kvm_faultin_pfn(vcpu, fault, ACC_ALL); + r = kvm_mmu_faultin_pfn(vcpu, fault, ACC_ALL); if (r != RET_PF_CONTINUE) return r; @@ -4613,44 +4829,27 @@ static int kvm_tdp_mmu_page_fault(struct kvm_vcpu *vcpu, r = kvm_tdp_mmu_map(vcpu, fault); out_unlock: + kvm_mmu_finish_page_fault(vcpu, fault, r); read_unlock(&vcpu->kvm->mmu_lock); - kvm_release_pfn_clean(fault->pfn); return r; } #endif -bool __kvm_mmu_honors_guest_mtrrs(bool vm_has_noncoherent_dma) +bool kvm_mmu_may_ignore_guest_pat(void) { /* - * If host MTRRs are ignored (shadow_memtype_mask is non-zero), and the - * VM has non-coherent DMA (DMA doesn't snoop CPU caches), KVM's ABI is - * to honor the memtype from the guest's MTRRs so that guest accesses - * to memory that is DMA'd aren't cached against the guest's wishes. - * - * Note, KVM may still ultimately ignore guest MTRRs for certain PFNs, - * e.g. KVM will force UC memtype for host MMIO. + * When EPT is enabled (shadow_memtype_mask is non-zero), and the VM + * has non-coherent DMA (DMA doesn't snoop CPU caches), KVM's ABI is to + * honor the memtype from the guest's PAT so that guest accesses to + * memory that is DMA'd aren't cached against the guest's wishes. As a + * result, KVM _may_ ignore guest PAT, whereas without non-coherent DMA, + * KVM _always_ ignores guest PAT (when EPT is enabled). */ - return vm_has_noncoherent_dma && shadow_memtype_mask; + return shadow_memtype_mask; } int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) { - /* - * If the guest's MTRRs may be used to compute the "real" memtype, - * restrict the mapping level to ensure KVM uses a consistent memtype - * across the entire mapping. - */ - if (kvm_mmu_honors_guest_mtrrs(vcpu->kvm)) { - for ( ; fault->max_level > PG_LEVEL_4K; --fault->max_level) { - int page_num = KVM_PAGES_PER_HPAGE(fault->max_level); - gfn_t base = gfn_round_for_level(fault->gfn, - fault->max_level); - - if (kvm_mtrr_check_gfn_range_consistency(vcpu, base, page_num)) - break; - } - } - #ifdef CONFIG_X86_64 if (tdp_mmu_enabled) return kvm_tdp_mmu_page_fault(vcpu, fault); @@ -4659,6 +4858,85 @@ int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) return direct_page_fault(vcpu, fault); } +static int kvm_tdp_map_page(struct kvm_vcpu *vcpu, gpa_t gpa, u64 error_code, + u8 *level) +{ + int r; + + /* + * Restrict to TDP page fault, since that's the only case where the MMU + * is indexed by GPA. + */ + if (vcpu->arch.mmu->page_fault != kvm_tdp_page_fault) + return -EOPNOTSUPP; + + do { + if (signal_pending(current)) + return -EINTR; + cond_resched(); + r = kvm_mmu_do_page_fault(vcpu, gpa, error_code, true, NULL, level); + } while (r == RET_PF_RETRY); + + if (r < 0) + return r; + + switch (r) { + case RET_PF_FIXED: + case RET_PF_SPURIOUS: + case RET_PF_WRITE_PROTECTED: + return 0; + + case RET_PF_EMULATE: + return -ENOENT; + + case RET_PF_RETRY: + case RET_PF_CONTINUE: + case RET_PF_INVALID: + default: + WARN_ONCE(1, "could not fix page fault during prefault"); + return -EIO; + } +} + +long kvm_arch_vcpu_pre_fault_memory(struct kvm_vcpu *vcpu, + struct kvm_pre_fault_memory *range) +{ + u64 error_code = PFERR_GUEST_FINAL_MASK; + u8 level = PG_LEVEL_4K; + u64 end; + int r; + + if (!vcpu->kvm->arch.pre_fault_allowed) + return -EOPNOTSUPP; + + /* + * reload is efficient when called repeatedly, so we can do it on + * every iteration. + */ + r = kvm_mmu_reload(vcpu); + if (r) + return r; + + if (kvm_arch_has_private_mem(vcpu->kvm) && + kvm_mem_is_private(vcpu->kvm, gpa_to_gfn(range->gpa))) + error_code |= PFERR_PRIVATE_ACCESS; + + /* + * Shadow paging uses GVA for kvm page fault, so restrict to + * two-dimensional paging. + */ + r = kvm_tdp_map_page(vcpu, range->gpa, error_code, &level); + if (r < 0) + return r; + + /* + * If the mapping that covers range->gpa can use a huge page, it + * may start below it or end after range->gpa + range->size. + */ + end = (range->gpa & KVM_HPAGE_MASK(level)) + KVM_HPAGE_SIZE(level); + return min(range->size, end - range->gpa); +} + static void nonpaging_init_context(struct kvm_mmu *context) { context->page_fault = nonpaging_page_fault; @@ -4812,7 +5090,7 @@ EXPORT_SYMBOL_GPL(kvm_mmu_new_pgd); static bool sync_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, gfn_t gfn, unsigned int access) { - if (unlikely(is_mmio_spte(*sptep))) { + if (unlikely(is_mmio_spte(vcpu->kvm, *sptep))) { if (gfn != get_mmio_spte_gfn(*sptep)) { mmu_spte_clear_no_track(sptep); return true; @@ -4933,9 +5211,9 @@ static void reset_guest_rsvds_bits_mask(struct kvm_vcpu *vcpu, __reset_rsvds_bits_mask(&context->guest_rsvd_check, vcpu->arch.reserved_gpa_bits, context->cpu_role.base.level, is_efer_nx(context), - guest_can_use(vcpu, X86_FEATURE_GBPAGES), + guest_cpu_cap_has(vcpu, X86_FEATURE_GBPAGES), is_cr4_pse(context), - guest_cpuid_is_amd_or_hygon(vcpu)); + guest_cpuid_is_amd_compatible(vcpu)); } static void __reset_rsvds_bits_mask_ept(struct rsvd_bits_validate *rsvd_check, @@ -4986,7 +5264,7 @@ static void reset_rsvds_bits_mask_ept(struct kvm_vcpu *vcpu, static inline u64 reserved_hpa_bits(void) { - return rsvd_bits(shadow_phys_bits, 63); + return rsvd_bits(kvm_host.maxphyaddr, 63); } /* @@ -5010,7 +5288,7 @@ static void reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, __reset_rsvds_bits_mask(shadow_zero_check, reserved_hpa_bits(), context->root_role.level, context->root_role.efer_nx, - guest_can_use(vcpu, X86_FEATURE_GBPAGES), + guest_cpu_cap_has(vcpu, X86_FEATURE_GBPAGES), is_pse, is_amd); if (!shadow_me_mask) @@ -5322,6 +5600,11 @@ static inline int kvm_mmu_get_tdp_level(struct kvm_vcpu *vcpu) return max_tdp_level; } +u8 kvm_mmu_get_max_tdp_level(void) +{ + return tdp_root_level ? tdp_root_level : max_tdp_level; +} + static union kvm_mmu_page_role kvm_calc_tdp_mmu_root_page_role(struct kvm_vcpu *vcpu, union kvm_cpu_role cpu_role) @@ -5333,7 +5616,7 @@ kvm_calc_tdp_mmu_root_page_role(struct kvm_vcpu *vcpu, role.efer_nx = true; role.smm = cpu_role.base.smm; role.guest_mode = cpu_role.base.guest_mode; - role.ad_disabled = !kvm_ad_enabled(); + role.ad_disabled = !kvm_ad_enabled; role.level = kvm_mmu_get_tdp_level(vcpu); role.direct = true; role.has_4_byte_gpte = false; @@ -5430,7 +5713,7 @@ void kvm_init_shadow_npt_mmu(struct kvm_vcpu *vcpu, unsigned long cr0, union kvm_mmu_page_role root_role; /* NPT requires CR0.PG=1. */ - WARN_ON_ONCE(cpu_role.base.direct); + WARN_ON_ONCE(cpu_role.base.direct || !cpu_role.base.guest_mode); root_role = cpu_role.base; root_role.level = kvm_mmu_get_tdp_level(vcpu); @@ -5576,9 +5859,9 @@ void kvm_mmu_after_set_cpuid(struct kvm_vcpu *vcpu) * that problem is swept under the rug; KVM's CPUID API is horrific and * it's all but impossible to solve it without introducing a new API. */ - vcpu->arch.root_mmu.root_role.word = 0; - vcpu->arch.guest_mmu.root_role.word = 0; - vcpu->arch.nested_mmu.root_role.word = 0; + vcpu->arch.root_mmu.root_role.invalid = 1; + vcpu->arch.guest_mmu.root_role.invalid = 1; + vcpu->arch.nested_mmu.root_role.invalid = 1; vcpu->arch.root_mmu.cpu_role.ext.valid = 0; vcpu->arch.guest_mmu.cpu_role.ext.valid = 0; vcpu->arch.nested_mmu.cpu_role.ext.valid = 0; @@ -5626,7 +5909,7 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu) * stale entries. Flushing on alloc also allows KVM to skip the TLB * flush when freeing a root (see kvm_tdp_mmu_put_root()). */ - static_call(kvm_x86_flush_tlb_current)(vcpu); + kvm_x86_call(flush_tlb_current)(vcpu); out: return r; } @@ -5802,10 +6085,15 @@ void kvm_mmu_track_write(struct kvm_vcpu *vcpu, gpa_t gpa, const u8 *new, bool flush = false; /* - * If we don't have indirect shadow pages, it means no page is - * write-protected, so we can exit simply. + * When emulating guest writes, ensure the written value is visible to + * any task that is handling page faults before checking whether or not + * KVM is shadowing a guest PTE. This ensures either KVM will create + * the correct SPTE in the page fault handler, or this task will see + * a non-zero indirect_shadow_pages. Pairs with the smp_mb() in + * account_shadowed(). */ - if (!READ_ONCE(vcpu->kvm->arch.indirect_shadow_pages)) + smp_mb(); + if (!vcpu->kvm->arch.indirect_shadow_pages) return; write_lock(&vcpu->kvm->mmu_lock); @@ -5840,78 +6128,195 @@ void kvm_mmu_track_write(struct kvm_vcpu *vcpu, gpa_t gpa, const u8 *new, write_unlock(&vcpu->kvm->mmu_lock); } -int noinline kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u64 error_code, - void *insn, int insn_len) +static bool is_write_to_guest_page_table(u64 error_code) +{ + const u64 mask = PFERR_GUEST_PAGE_MASK | PFERR_WRITE_MASK | PFERR_PRESENT_MASK; + + return (error_code & mask) == mask; +} + +static int kvm_mmu_write_protect_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, + u64 error_code, int *emulation_type) { - int r, emulation_type = EMULTYPE_PF; bool direct = vcpu->arch.mmu->root_role.direct; /* - * IMPLICIT_ACCESS is a KVM-defined flag used to correctly perform SMAP - * checks when emulating instructions that triggers implicit access. - * WARN if hardware generates a fault with an error code that collides - * with the KVM-defined value. Clear the flag and continue on, i.e. - * don't terminate the VM, as KVM can't possibly be relying on a flag - * that KVM doesn't know about. + * Do not try to unprotect and retry if the vCPU re-faulted on the same + * RIP with the same address that was previously unprotected, as doing + * so will likely put the vCPU into an infinite. E.g. if the vCPU uses + * a non-page-table modifying instruction on the PDE that points to the + * instruction, then unprotecting the gfn will unmap the instruction's + * code, i.e. make it impossible for the instruction to ever complete. + */ + if (vcpu->arch.last_retry_eip == kvm_rip_read(vcpu) && + vcpu->arch.last_retry_addr == cr2_or_gpa) + return RET_PF_EMULATE; + + /* + * Reset the unprotect+retry values that guard against infinite loops. + * The values will be refreshed if KVM explicitly unprotects a gfn and + * retries, in all other cases it's safe to retry in the future even if + * the next page fault happens on the same RIP+address. + */ + vcpu->arch.last_retry_eip = 0; + vcpu->arch.last_retry_addr = 0; + + /* + * It should be impossible to reach this point with an MMIO cache hit, + * as RET_PF_WRITE_PROTECTED is returned if and only if there's a valid, + * writable memslot, and creating a memslot should invalidate the MMIO + * cache by way of changing the memslot generation. WARN and disallow + * retry if MMIO is detected, as retrying MMIO emulation is pointless + * and could put the vCPU into an infinite loop because the processor + * will keep faulting on the non-existent MMIO address. */ - if (WARN_ON_ONCE(error_code & PFERR_IMPLICIT_ACCESS)) - error_code &= ~PFERR_IMPLICIT_ACCESS; + if (WARN_ON_ONCE(mmio_info_in_cache(vcpu, cr2_or_gpa, direct))) + return RET_PF_EMULATE; + + /* + * Before emulating the instruction, check to see if the access was due + * to a read-only violation while the CPU was walking non-nested NPT + * page tables, i.e. for a direct MMU, for _guest_ page tables in L1. + * If L1 is sharing (a subset of) its page tables with L2, e.g. by + * having nCR3 share lower level page tables with hCR3, then when KVM + * (L0) write-protects the nested NPTs, i.e. npt12 entries, KVM is also + * unknowingly write-protecting L1's guest page tables, which KVM isn't + * shadowing. + * + * Because the CPU (by default) walks NPT page tables using a write + * access (to ensure the CPU can do A/D updates), page walks in L1 can + * trigger write faults for the above case even when L1 isn't modifying + * PTEs. As a result, KVM will unnecessarily emulate (or at least, try + * to emulate) an excessive number of L1 instructions; because L1's MMU + * isn't shadowed by KVM, there is no need to write-protect L1's gPTEs + * and thus no need to emulate in order to guarantee forward progress. + * + * Try to unprotect the gfn, i.e. zap any shadow pages, so that L1 can + * proceed without triggering emulation. If one or more shadow pages + * was zapped, skip emulation and resume L1 to let it natively execute + * the instruction. If no shadow pages were zapped, then the write- + * fault is due to something else entirely, i.e. KVM needs to emulate, + * as resuming the guest will put it into an infinite loop. + * + * Note, this code also applies to Intel CPUs, even though it is *very* + * unlikely that an L1 will share its page tables (IA32/PAE/paging64 + * format) with L2's page tables (EPT format). + * + * For indirect MMUs, i.e. if KVM is shadowing the current MMU, try to + * unprotect the gfn and retry if an event is awaiting reinjection. If + * KVM emulates multiple instructions before completing event injection, + * the event could be delayed beyond what is architecturally allowed, + * e.g. KVM could inject an IRQ after the TPR has been raised. + */ + if (((direct && is_write_to_guest_page_table(error_code)) || + (!direct && kvm_event_needs_reinjection(vcpu))) && + kvm_mmu_unprotect_gfn_and_retry(vcpu, cr2_or_gpa)) + return RET_PF_RETRY; + + /* + * The gfn is write-protected, but if KVM detects its emulating an + * instruction that is unlikely to be used to modify page tables, or if + * emulation fails, KVM can try to unprotect the gfn and let the CPU + * re-execute the instruction that caused the page fault. Do not allow + * retrying an instruction from a nested guest as KVM is only explicitly + * shadowing L1's page tables, i.e. unprotecting something for L1 isn't + * going to magically fix whatever issue caused L2 to fail. + */ + if (!is_guest_mode(vcpu)) + *emulation_type |= EMULTYPE_ALLOW_RETRY_PF; + + return RET_PF_EMULATE; +} + +int noinline kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, u64 error_code, + void *insn, int insn_len) +{ + int r, emulation_type = EMULTYPE_PF; + bool direct = vcpu->arch.mmu->root_role.direct; if (WARN_ON_ONCE(!VALID_PAGE(vcpu->arch.mmu->root.hpa))) return RET_PF_RETRY; + /* + * Except for reserved faults (emulated MMIO is shared-only), set the + * PFERR_PRIVATE_ACCESS flag for software-protected VMs based on the gfn's + * current attributes, which are the source of truth for such VMs. Note, + * this wrong for nested MMUs as the GPA is an L2 GPA, but KVM doesn't + * currently supported nested virtualization (among many other things) + * for software-protected VMs. + */ + if (IS_ENABLED(CONFIG_KVM_SW_PROTECTED_VM) && + !(error_code & PFERR_RSVD_MASK) && + vcpu->kvm->arch.vm_type == KVM_X86_SW_PROTECTED_VM && + kvm_mem_is_private(vcpu->kvm, gpa_to_gfn(cr2_or_gpa))) + error_code |= PFERR_PRIVATE_ACCESS; + r = RET_PF_INVALID; if (unlikely(error_code & PFERR_RSVD_MASK)) { + if (WARN_ON_ONCE(error_code & PFERR_PRIVATE_ACCESS)) + return -EFAULT; + r = handle_mmio_page_fault(vcpu, cr2_or_gpa, direct); if (r == RET_PF_EMULATE) goto emulate; } if (r == RET_PF_INVALID) { - r = kvm_mmu_do_page_fault(vcpu, cr2_or_gpa, - lower_32_bits(error_code), false, - &emulation_type); + vcpu->stat.pf_taken++; + + r = kvm_mmu_do_page_fault(vcpu, cr2_or_gpa, error_code, false, + &emulation_type, NULL); if (KVM_BUG_ON(r == RET_PF_INVALID, vcpu->kvm)) return -EIO; } if (r < 0) return r; - if (r != RET_PF_EMULATE) - return 1; + + if (r == RET_PF_WRITE_PROTECTED) + r = kvm_mmu_write_protect_fault(vcpu, cr2_or_gpa, error_code, + &emulation_type); + + if (r == RET_PF_FIXED) + vcpu->stat.pf_fixed++; + else if (r == RET_PF_EMULATE) + vcpu->stat.pf_emulate++; + else if (r == RET_PF_SPURIOUS) + vcpu->stat.pf_spurious++; /* - * Before emulating the instruction, check if the error code - * was due to a RO violation while translating the guest page. - * This can occur when using nested virtualization with nested - * paging in both guests. If true, we simply unprotect the page - * and resume the guest. + * None of handle_mmio_page_fault(), kvm_mmu_do_page_fault(), or + * kvm_mmu_write_protect_fault() return RET_PF_CONTINUE. + * kvm_mmu_do_page_fault() only uses RET_PF_CONTINUE internally to + * indicate continuing the page fault handling until to the final + * page table mapping phase. */ - if (vcpu->arch.mmu->root_role.direct && - (error_code & PFERR_NESTED_GUEST_PAGE) == PFERR_NESTED_GUEST_PAGE) { - kvm_mmu_unprotect_page(vcpu->kvm, gpa_to_gfn(cr2_or_gpa)); - return 1; - } + WARN_ON_ONCE(r == RET_PF_CONTINUE); + if (r != RET_PF_EMULATE) + return r; - /* - * vcpu->arch.mmu.page_fault returned RET_PF_EMULATE, but we can still - * optimistically try to just unprotect the page and let the processor - * re-execute the instruction that caused the page fault. Do not allow - * retrying MMIO emulation, as it's not only pointless but could also - * cause us to enter an infinite loop because the processor will keep - * faulting on the non-existent MMIO address. Retrying an instruction - * from a nested guest is also pointless and dangerous as we are only - * explicitly shadowing L1's page tables, i.e. unprotecting something - * for L1 isn't going to magically fix whatever issue cause L2 to fail. - */ - if (!mmio_info_in_cache(vcpu, cr2_or_gpa, direct) && !is_guest_mode(vcpu)) - emulation_type |= EMULTYPE_ALLOW_RETRY_PF; emulate: return x86_emulate_instruction(vcpu, cr2_or_gpa, emulation_type, insn, insn_len); } EXPORT_SYMBOL_GPL(kvm_mmu_page_fault); +void kvm_mmu_print_sptes(struct kvm_vcpu *vcpu, gpa_t gpa, const char *msg) +{ + u64 sptes[PT64_ROOT_MAX_LEVEL + 1]; + int root_level, leaf, level; + + leaf = get_sptes_lockless(vcpu, gpa, sptes, &root_level); + if (unlikely(leaf < 0)) + return; + + pr_err("%s %llx", msg, gpa); + for (level = root_level; level >= leaf; level--) + pr_cont(", spte[%d] = 0x%llx", level, sptes[level]); + pr_cont("\n"); +} +EXPORT_SYMBOL_GPL(kvm_mmu_print_sptes); + static void __kvm_mmu_invalidate_addr(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, u64 addr, hpa_t root_hpa) { @@ -5959,10 +6364,10 @@ void kvm_mmu_invalidate_addr(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, /* It's actually a GPA for vcpu->arch.guest_mmu. */ if (mmu != &vcpu->arch.guest_mmu) { /* INVLPG on a non-canonical address is a NOP according to the SDM. */ - if (is_noncanonical_address(addr, vcpu)) + if (is_noncanonical_invlpg_address(addr, vcpu)) return; - static_call(kvm_x86_flush_tlb_gva)(vcpu, addr); + kvm_x86_call(flush_tlb_gva)(vcpu, addr); } if (!mmu->sync_spte) @@ -6048,59 +6453,6 @@ void kvm_configure_mmu(bool enable_tdp, int tdp_forced_root_level, } EXPORT_SYMBOL_GPL(kvm_configure_mmu); -/* The return value indicates if tlb flush on all vcpus is needed. */ -typedef bool (*slot_rmaps_handler) (struct kvm *kvm, - struct kvm_rmap_head *rmap_head, - const struct kvm_memory_slot *slot); - -static __always_inline bool __walk_slot_rmaps(struct kvm *kvm, - const struct kvm_memory_slot *slot, - slot_rmaps_handler fn, - int start_level, int end_level, - gfn_t start_gfn, gfn_t end_gfn, - bool flush_on_yield, bool flush) -{ - struct slot_rmap_walk_iterator iterator; - - lockdep_assert_held_write(&kvm->mmu_lock); - - for_each_slot_rmap_range(slot, start_level, end_level, start_gfn, - end_gfn, &iterator) { - if (iterator.rmap) - flush |= fn(kvm, iterator.rmap, slot); - - if (need_resched() || rwlock_needbreak(&kvm->mmu_lock)) { - if (flush && flush_on_yield) { - kvm_flush_remote_tlbs_range(kvm, start_gfn, - iterator.gfn - start_gfn + 1); - flush = false; - } - cond_resched_rwlock_write(&kvm->mmu_lock); - } - } - - return flush; -} - -static __always_inline bool walk_slot_rmaps(struct kvm *kvm, - const struct kvm_memory_slot *slot, - slot_rmaps_handler fn, - int start_level, int end_level, - bool flush_on_yield) -{ - return __walk_slot_rmaps(kvm, slot, fn, start_level, end_level, - slot->base_gfn, slot->base_gfn + slot->npages - 1, - flush_on_yield, false); -} - -static __always_inline bool walk_slot_rmaps_4k(struct kvm *kvm, - const struct kvm_memory_slot *slot, - slot_rmaps_handler fn, - bool flush_on_yield) -{ - return walk_slot_rmaps(kvm, slot, fn, PG_LEVEL_4K, PG_LEVEL_4K, flush_on_yield); -} - static void free_mmu_pages(struct kvm_mmu *mmu) { if (!tdp_enabled && mmu->pae_root) @@ -6117,6 +6469,7 @@ static int __kvm_mmu_create(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu) mmu->root.hpa = INVALID_PAGE; mmu->root.pgd = 0; + mmu->mirror_root_hpa = INVALID_PAGE; for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++) mmu->prev_roots[i] = KVM_MMU_ROOT_INFO_INVALID; @@ -6173,7 +6526,10 @@ int kvm_mmu_create(struct kvm_vcpu *vcpu) vcpu->arch.mmu_page_header_cache.kmem_cache = mmu_page_header_cache; vcpu->arch.mmu_page_header_cache.gfp_zero = __GFP_ZERO; - vcpu->arch.mmu_shadow_page_cache.gfp_zero = __GFP_ZERO; + vcpu->arch.mmu_shadow_page_cache.init_value = + SHADOW_NONPRESENT_VALUE; + if (!vcpu->arch.mmu_shadow_page_cache.init_value) + vcpu->arch.mmu_shadow_page_cache.gfp_zero = __GFP_ZERO; vcpu->arch.mmu = &vcpu->arch.root_mmu; vcpu->arch.walk_mmu = &vcpu->arch.root_mmu; @@ -6197,8 +6553,11 @@ static void kvm_zap_obsolete_pages(struct kvm *kvm) { struct kvm_mmu_page *sp, *node; int nr_zapped, batch = 0; + LIST_HEAD(invalid_list); bool unstable; + lockdep_assert_held(&kvm->slots_lock); + restart: list_for_each_entry_safe_reverse(sp, node, &kvm->arch.active_mmu_pages, link) { @@ -6230,7 +6589,7 @@ restart: } unstable = __kvm_mmu_prepare_zap_page(kvm, sp, - &kvm->arch.zapped_obsolete_pages, &nr_zapped); + &invalid_list, &nr_zapped); batch += nr_zapped; if (unstable) @@ -6246,7 +6605,7 @@ restart: * kvm_mmu_load()), and the reload in the caller ensure no vCPUs are * running with an obsolete MMU. */ - kvm_mmu_commit_zap_page(kvm, &kvm->arch.zapped_obsolete_pages); + kvm_mmu_commit_zap_page(kvm, &invalid_list); } /* @@ -6280,8 +6639,13 @@ static void kvm_mmu_zap_all_fast(struct kvm *kvm) * write and in the same critical section as making the reload request, * e.g. before kvm_zap_obsolete_pages() could drop mmu_lock and yield. */ - if (tdp_mmu_enabled) - kvm_tdp_mmu_invalidate_all_roots(kvm); + if (tdp_mmu_enabled) { + /* + * External page tables don't support fast zapping, therefore + * their mirrors must be invalidated separately by the caller. + */ + kvm_tdp_mmu_invalidate_roots(kvm, KVM_DIRECT_ROOTS); + } /* * Notify all vcpus to reload its shadow page table and flush TLB. @@ -6306,18 +6670,13 @@ static void kvm_mmu_zap_all_fast(struct kvm *kvm) * lead to use-after-free. */ if (tdp_mmu_enabled) - kvm_tdp_mmu_zap_invalidated_roots(kvm); -} - -static bool kvm_has_zapped_obsolete_pages(struct kvm *kvm) -{ - return unlikely(!list_empty_careful(&kvm->arch.zapped_obsolete_pages)); + kvm_tdp_mmu_zap_invalidated_roots(kvm, true); } void kvm_mmu_init_vm(struct kvm *kvm) { + kvm->arch.shadow_mmio_value = shadow_mmio_value; INIT_LIST_HEAD(&kvm->arch.active_mmu_pages); - INIT_LIST_HEAD(&kvm->arch.zapped_obsolete_pages); INIT_LIST_HEAD(&kvm->arch.possible_nx_huge_pages); spin_lock_init(&kvm->arch.mmu_unsync_pages_lock); @@ -6370,9 +6729,8 @@ static bool kvm_rmap_zap_gfn_range(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_e if (WARN_ON_ONCE(start >= end)) continue; - flush = __walk_slot_rmaps(kvm, memslot, __kvm_zap_rmap, - PG_LEVEL_4K, KVM_MAX_HUGEPAGE_LEVEL, - start, end - 1, true, flush); + flush = __kvm_rmap_zap_gfn_range(kvm, memslot, start, + end, true, flush); } } @@ -6552,7 +6910,7 @@ static void shadow_mmu_split_huge_page(struct kvm *kvm, continue; } - spte = make_huge_page_split_spte(kvm, huge_spte, sp->role, index); + spte = make_small_spte(kvm, huge_spte, sp->role, index); mmu_spte_set(sptep, spte); __rmap_add(kvm, cache, slot, sptep, gfn, sp->role.access); } @@ -6660,7 +7018,7 @@ static void kvm_shadow_mmu_try_split_huge_pages(struct kvm *kvm, */ for (level = KVM_MAX_HUGEPAGE_LEVEL; level > target_level; level--) __walk_slot_rmaps(kvm, slot, shadow_mmu_try_split_huge_pages, - level, level, start, end - 1, true, false); + level, level, start, end - 1, true, true, false); } /* Must be called with the mmu_lock held in write-mode. */ @@ -6735,8 +7093,7 @@ restart: * mapping if the indirect sp has level = 1. */ if (sp->role.direct && - sp->role.level < kvm_mmu_max_mapping_level(kvm, slot, sp->gfn, - PG_LEVEL_NUM)) { + sp->role.level < kvm_mmu_max_mapping_level(kvm, slot, sp->gfn)) { kvm_zap_one_rmap_spte(kvm, rmap_head, sptep); if (kvm_available_flush_remote_tlbs_range()) @@ -6750,6 +7107,7 @@ restart: return need_tlb_flush; } +EXPORT_SYMBOL_GPL(kvm_zap_gfn_range); static void kvm_rmap_zap_collapsible_sptes(struct kvm *kvm, const struct kvm_memory_slot *slot) @@ -6763,8 +7121,8 @@ static void kvm_rmap_zap_collapsible_sptes(struct kvm *kvm, kvm_flush_remote_tlbs_memslot(kvm, slot); } -void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm, - const struct kvm_memory_slot *slot) +void kvm_mmu_recover_huge_pages(struct kvm *kvm, + const struct kvm_memory_slot *slot) { if (kvm_memslots_have_rmaps(kvm)) { write_lock(&kvm->mmu_lock); @@ -6774,7 +7132,7 @@ void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm, if (tdp_mmu_enabled) { read_lock(&kvm->mmu_lock); - kvm_tdp_mmu_zap_collapsible_sptes(kvm, slot); + kvm_tdp_mmu_recover_huge_pages(kvm, slot); read_unlock(&kvm->mmu_lock); } } @@ -6838,10 +7196,70 @@ void kvm_arch_flush_shadow_all(struct kvm *kvm) kvm_mmu_zap_all(kvm); } +static void kvm_mmu_zap_memslot_pages_and_flush(struct kvm *kvm, + struct kvm_memory_slot *slot, + bool flush) +{ + LIST_HEAD(invalid_list); + unsigned long i; + + if (list_empty(&kvm->arch.active_mmu_pages)) + goto out_flush; + + /* + * Since accounting information is stored in struct kvm_arch_memory_slot, + * all MMU pages that are shadowing guest PTEs must be zapped before the + * memslot is deleted, as freeing such pages after the memslot is freed + * will result in use-after-free, e.g. in unaccount_shadowed(). + */ + for (i = 0; i < slot->npages; i++) { + struct kvm_mmu_page *sp; + gfn_t gfn = slot->base_gfn + i; + + for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) + kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list); + + if (need_resched() || rwlock_needbreak(&kvm->mmu_lock)) { + kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush); + flush = false; + cond_resched_rwlock_write(&kvm->mmu_lock); + } + } + +out_flush: + kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush); +} + +static void kvm_mmu_zap_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot) +{ + struct kvm_gfn_range range = { + .slot = slot, + .start = slot->base_gfn, + .end = slot->base_gfn + slot->npages, + .may_block = true, + }; + bool flush; + + write_lock(&kvm->mmu_lock); + flush = kvm_unmap_gfn_range(kvm, &range); + kvm_mmu_zap_memslot_pages_and_flush(kvm, slot, flush); + write_unlock(&kvm->mmu_lock); +} + +static inline bool kvm_memslot_flush_zap_all(struct kvm *kvm) +{ + return kvm->arch.vm_type == KVM_X86_DEFAULT_VM && + kvm_check_has_quirk(kvm, KVM_X86_QUIRK_SLOT_ZAP_ALL); +} + void kvm_arch_flush_shadow_memslot(struct kvm *kvm, struct kvm_memory_slot *slot) { - kvm_mmu_zap_all_fast(kvm); + if (kvm_memslot_flush_zap_all(kvm)) + kvm_mmu_zap_all_fast(kvm); + else + kvm_mmu_zap_memslot(kvm, slot); } void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, u64 gen) @@ -6869,77 +7287,23 @@ void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, u64 gen) } } -static unsigned long mmu_shrink_scan(struct shrinker *shrink, - struct shrink_control *sc) +static void mmu_destroy_caches(void) { - struct kvm *kvm; - int nr_to_scan = sc->nr_to_scan; - unsigned long freed = 0; - - mutex_lock(&kvm_lock); - - list_for_each_entry(kvm, &vm_list, vm_list) { - int idx; - LIST_HEAD(invalid_list); - - /* - * Never scan more than sc->nr_to_scan VM instances. - * Will not hit this condition practically since we do not try - * to shrink more than one VM and it is very unlikely to see - * !n_used_mmu_pages so many times. - */ - if (!nr_to_scan--) - break; - /* - * n_used_mmu_pages is accessed without holding kvm->mmu_lock - * here. We may skip a VM instance errorneosly, but we do not - * want to shrink a VM that only started to populate its MMU - * anyway. - */ - if (!kvm->arch.n_used_mmu_pages && - !kvm_has_zapped_obsolete_pages(kvm)) - continue; - - idx = srcu_read_lock(&kvm->srcu); - write_lock(&kvm->mmu_lock); - - if (kvm_has_zapped_obsolete_pages(kvm)) { - kvm_mmu_commit_zap_page(kvm, - &kvm->arch.zapped_obsolete_pages); - goto unlock; - } - - freed = kvm_mmu_zap_oldest_mmu_pages(kvm, sc->nr_to_scan); - -unlock: - write_unlock(&kvm->mmu_lock); - srcu_read_unlock(&kvm->srcu, idx); - - /* - * unfair on small ones - * per-vm shrinkers cry out - * sadness comes quickly - */ - list_move_tail(&kvm->vm_list, &vm_list); - break; - } - - mutex_unlock(&kvm_lock); - return freed; + kmem_cache_destroy(pte_list_desc_cache); + kmem_cache_destroy(mmu_page_header_cache); } -static unsigned long mmu_shrink_count(struct shrinker *shrink, - struct shrink_control *sc) +static void kvm_wake_nx_recovery_thread(struct kvm *kvm) { - return percpu_counter_read_positive(&kvm_total_used_mmu_pages); -} - -static struct shrinker *mmu_shrinker; + /* + * The NX recovery thread is spawned on-demand at the first KVM_RUN and + * may not be valid even though the VM is globally visible. Do nothing, + * as such a VM can't have any possible NX huge pages. + */ + struct vhost_task *nx_thread = READ_ONCE(kvm->arch.nx_huge_page_recovery_thread); -static void mmu_destroy_caches(void) -{ - kmem_cache_destroy(pte_list_desc_cache); - kmem_cache_destroy(mmu_page_header_cache); + if (nx_thread) + vhost_task_wake(nx_thread); } static int get_nx_huge_pages(char *buffer, const struct kernel_param *kp) @@ -7002,7 +7366,7 @@ static int set_nx_huge_pages(const char *val, const struct kernel_param *kp) kvm_mmu_zap_all_fast(kvm); mutex_unlock(&kvm->slots_lock); - wake_up_process(kvm->arch.nx_huge_page_recovery_thread); + kvm_wake_nx_recovery_thread(kvm); } mutex_unlock(&kvm_lock); } @@ -7062,23 +7426,8 @@ int kvm_mmu_vendor_module_init(void) if (!mmu_page_header_cache) goto out; - if (percpu_counter_init(&kvm_total_used_mmu_pages, 0, GFP_KERNEL)) - goto out; - - mmu_shrinker = shrinker_alloc(0, "x86-mmu"); - if (!mmu_shrinker) - goto out_shrinker; - - mmu_shrinker->count_objects = mmu_shrink_count; - mmu_shrinker->scan_objects = mmu_shrink_scan; - mmu_shrinker->seeks = DEFAULT_SEEKS * 10; - - shrinker_register(mmu_shrinker); - return 0; -out_shrinker: - percpu_counter_destroy(&kvm_total_used_mmu_pages); out: mmu_destroy_caches(); return ret; @@ -7087,6 +7436,12 @@ out: void kvm_mmu_destroy(struct kvm_vcpu *vcpu) { kvm_mmu_unload(vcpu); + if (tdp_mmu_enabled) { + read_lock(&vcpu->kvm->mmu_lock); + mmu_free_root_page(vcpu->kvm, &vcpu->arch.mmu->mirror_root_hpa, + NULL); + read_unlock(&vcpu->kvm->mmu_lock); + } free_mmu_pages(&vcpu->arch.root_mmu); free_mmu_pages(&vcpu->arch.guest_mmu); mmu_free_memory_caches(vcpu); @@ -7095,8 +7450,6 @@ void kvm_mmu_destroy(struct kvm_vcpu *vcpu) void kvm_mmu_vendor_module_exit(void) { mmu_destroy_caches(); - percpu_counter_destroy(&kvm_total_used_mmu_pages); - shrinker_free(mmu_shrinker); } /* @@ -7148,7 +7501,7 @@ static int set_nx_huge_pages_recovery_param(const char *val, const struct kernel mutex_lock(&kvm_lock); list_for_each_entry(kvm, &vm_list, vm_list) - wake_up_process(kvm->arch.nx_huge_page_recovery_thread); + kvm_wake_nx_recovery_thread(kvm); mutex_unlock(&kvm_lock); } @@ -7251,62 +7604,68 @@ static void kvm_recover_nx_huge_pages(struct kvm *kvm) srcu_read_unlock(&kvm->srcu, rcu_idx); } -static long get_nx_huge_page_recovery_timeout(u64 start_time) +static void kvm_nx_huge_page_recovery_worker_kill(void *data) +{ +} + +static bool kvm_nx_huge_page_recovery_worker(void *data) { + struct kvm *kvm = data; bool enabled; uint period; + long remaining_time; enabled = calc_nx_huge_pages_recovery_period(&period); + if (!enabled) + return false; + + remaining_time = kvm->arch.nx_huge_page_last + msecs_to_jiffies(period) + - get_jiffies_64(); + if (remaining_time > 0) { + schedule_timeout(remaining_time); + /* check for signals and come back */ + return true; + } - return enabled ? start_time + msecs_to_jiffies(period) - get_jiffies_64() - : MAX_SCHEDULE_TIMEOUT; + __set_current_state(TASK_RUNNING); + kvm_recover_nx_huge_pages(kvm); + kvm->arch.nx_huge_page_last = get_jiffies_64(); + return true; } -static int kvm_nx_huge_page_recovery_worker(struct kvm *kvm, uintptr_t data) +static int kvm_mmu_start_lpage_recovery(struct once *once) { - u64 start_time; - long remaining_time; + struct kvm_arch *ka = container_of(once, struct kvm_arch, nx_once); + struct kvm *kvm = container_of(ka, struct kvm, arch); + struct vhost_task *nx_thread; - while (true) { - start_time = get_jiffies_64(); - remaining_time = get_nx_huge_page_recovery_timeout(start_time); + kvm->arch.nx_huge_page_last = get_jiffies_64(); + nx_thread = vhost_task_create(kvm_nx_huge_page_recovery_worker, + kvm_nx_huge_page_recovery_worker_kill, + kvm, "kvm-nx-lpage-recovery"); - set_current_state(TASK_INTERRUPTIBLE); - while (!kthread_should_stop() && remaining_time > 0) { - schedule_timeout(remaining_time); - remaining_time = get_nx_huge_page_recovery_timeout(start_time); - set_current_state(TASK_INTERRUPTIBLE); - } - - set_current_state(TASK_RUNNING); + if (IS_ERR(nx_thread)) + return PTR_ERR(nx_thread); - if (kthread_should_stop()) - return 0; + vhost_task_start(nx_thread); - kvm_recover_nx_huge_pages(kvm); - } + /* Make the task visible only once it is fully started. */ + WRITE_ONCE(kvm->arch.nx_huge_page_recovery_thread, nx_thread); + return 0; } int kvm_mmu_post_init_vm(struct kvm *kvm) { - int err; - if (nx_hugepage_mitigation_hard_disabled) return 0; - err = kvm_vm_create_worker_thread(kvm, kvm_nx_huge_page_recovery_worker, 0, - "kvm-nx-lpage-recovery", - &kvm->arch.nx_huge_page_recovery_thread); - if (!err) - kthread_unpark(kvm->arch.nx_huge_page_recovery_thread); - - return err; + return call_once(&kvm->arch.nx_once, kvm_mmu_start_lpage_recovery); } void kvm_mmu_pre_destroy_vm(struct kvm *kvm) { if (kvm->arch.nx_huge_page_recovery_thread) - kthread_stop(kvm->arch.nx_huge_page_recovery_thread); + vhost_task_stop(kvm->arch.nx_huge_page_recovery_thread); } #ifdef CONFIG_KVM_GENERIC_MEMORY_ATTRIBUTES @@ -7327,6 +7686,12 @@ bool kvm_arch_pre_set_memory_attributes(struct kvm *kvm, if (WARN_ON_ONCE(!kvm_arch_has_private_mem(kvm))) return false; + /* Unmap the old attribute page. */ + if (range->arg.attributes & KVM_MEMORY_ATTRIBUTE_PRIVATE) + range->attr_filter = KVM_FILTER_SHARED; + else + range->attr_filter = KVM_FILTER_PRIVATE; + return kvm_unmap_gfn_range(kvm, range); } @@ -7355,7 +7720,7 @@ static bool hugepage_has_attrs(struct kvm *kvm, struct kvm_memory_slot *slot, const unsigned long end = start + KVM_PAGES_PER_HPAGE(level); if (level == PG_LEVEL_2M) - return kvm_range_has_memory_attributes(kvm, start, end, attrs); + return kvm_range_has_memory_attributes(kvm, start, end, ~0, attrs); for (gfn = start; gfn < end; gfn += KVM_PAGES_PER_HPAGE(level - 1)) { if (hugepage_test_mixed(slot, gfn, level - 1) || @@ -7399,7 +7764,8 @@ bool kvm_arch_post_set_memory_attributes(struct kvm *kvm, * by the memslot, KVM can't use a hugepage due to the * misaligned address regardless of memory attributes. */ - if (gfn >= slot->base_gfn) { + if (gfn >= slot->base_gfn && + gfn + nr_pages <= slot->base_gfn + slot->npages) { if (hugepage_has_attrs(kvm, slot, gfn, level, attrs)) hugepage_clear_mixed(slot, gfn, level); else diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h index 5390a591a571..75f00598289d 100644 --- a/arch/x86/kvm/mmu/mmu_internal.h +++ b/arch/x86/kvm/mmu/mmu_internal.h @@ -6,6 +6,8 @@ #include <linux/kvm_host.h> #include <asm/kvm_host.h> +#include "mmu.h" + #ifdef CONFIG_KVM_PROVE_MMU #define KVM_MMU_WARN_ON(x) WARN_ON_ONCE(x) #else @@ -101,7 +103,22 @@ struct kvm_mmu_page { int root_count; refcount_t tdp_mmu_root_count; }; - unsigned int unsync_children; + union { + /* These two members aren't used for TDP MMU */ + struct { + unsigned int unsync_children; + /* + * Number of writes since the last time traversal + * visited this page. + */ + atomic_t write_flooding_count; + }; + /* + * Page table page of external PT. + * Passed to TDX module, not accessed by KVM. + */ + void *external_spt; + }; union { struct kvm_rmap_head parent_ptes; /* rmap pointers to parent sptes */ tdp_ptep_t ptep; @@ -124,9 +141,6 @@ struct kvm_mmu_page { int clear_spte_count; #endif - /* Number of writes since the last time traversal visited this page. */ - atomic_t write_flooding_count; - #ifdef CONFIG_X86_64 /* Used for freeing the page asynchronously if it is a TDP MMU page. */ struct rcu_head rcu_head; @@ -145,6 +159,34 @@ static inline int kvm_mmu_page_as_id(struct kvm_mmu_page *sp) return kvm_mmu_role_as_id(sp->role); } +static inline bool is_mirror_sp(const struct kvm_mmu_page *sp) +{ + return sp->role.is_mirror; +} + +static inline void kvm_mmu_alloc_external_spt(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + /* + * external_spt is allocated for TDX module to hold private EPT mappings, + * TDX module will initialize the page by itself. + * Therefore, KVM does not need to initialize or access external_spt. + * KVM only interacts with sp->spt for private EPT operations. + */ + sp->external_spt = kvm_mmu_memory_cache_alloc(&vcpu->arch.mmu_external_spt_cache); +} + +static inline gfn_t kvm_gfn_root_bits(const struct kvm *kvm, const struct kvm_mmu_page *root) +{ + /* + * Since mirror SPs are used only for TDX, which maps private memory + * at its "natural" GFN, no mask needs to be applied to them - and, dually, + * we expect that the bits is only used for the shared PT. + */ + if (is_mirror_sp(root)) + return 0; + return kvm_gfn_direct_bits(kvm); +} + static inline bool kvm_mmu_page_ad_need_write_protect(struct kvm_mmu_page *sp) { /* @@ -164,7 +206,7 @@ static inline gfn_t gfn_round_for_level(gfn_t gfn, int level) } int mmu_try_to_unsync_pages(struct kvm *kvm, const struct kvm_memory_slot *slot, - gfn_t gfn, bool can_unsync, bool prefetch); + gfn_t gfn, bool synchronizing, bool prefetch); void kvm_mmu_gfn_disallow_lpage(const struct kvm_memory_slot *slot, gfn_t gfn); void kvm_mmu_gfn_allow_lpage(const struct kvm_memory_slot *slot, gfn_t gfn); @@ -190,7 +232,7 @@ static inline bool is_nx_huge_page_enabled(struct kvm *kvm) struct kvm_page_fault { /* arguments to kvm_mmu_do_page_fault. */ const gpa_t addr; - const u32 error_code; + const u64 error_code; const bool prefetch; /* Derived from error_code. */ @@ -229,16 +271,21 @@ struct kvm_page_fault { */ u8 goal_level; - /* Shifted addr, or result of guest page table walk if addr is a gva. */ + /* + * Shifted addr, or result of guest page table walk if addr is a gva. In + * the case of VM where memslot's can be mapped at multiple GPA aliases + * (i.e. TDX), the gfn field does not contain the bit that selects between + * the aliases (i.e. the shared bit for TDX). + */ gfn_t gfn; /* The memslot containing gfn. May be NULL. */ struct kvm_memory_slot *slot; - /* Outputs of kvm_faultin_pfn. */ + /* Outputs of kvm_mmu_faultin_pfn(). */ unsigned long mmu_seq; kvm_pfn_t pfn; - hva_t hva; + struct page *refcounted_page; bool map_writable; /* @@ -258,6 +305,8 @@ int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault); * RET_PF_CONTINUE: So far, so good, keep handling the page fault. * RET_PF_RETRY: let CPU fault again on the address. * RET_PF_EMULATE: mmio page fault, emulate the instruction directly. + * RET_PF_WRITE_PROTECTED: the gfn is write-protected, either unprotected the + * gfn and retry, or emulate the instruction directly. * RET_PF_INVALID: the spte is invalid, let the real page fault path update it. * RET_PF_FIXED: The faulting entry has been fixed. * RET_PF_SPURIOUS: The faulting entry was already fixed, e.g. by another vCPU. @@ -266,21 +315,37 @@ int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault); * tracepoints via TRACE_DEFINE_ENUM() in mmutrace.h * * Note, all values must be greater than or equal to zero so as not to encroach - * on -errno return values. Somewhat arbitrarily use '0' for CONTINUE, which - * will allow for efficient machine code when checking for CONTINUE, e.g. - * "TEST %rax, %rax, JNZ", as all "stop!" values are non-zero. + * on -errno return values. */ enum { RET_PF_CONTINUE = 0, RET_PF_RETRY, RET_PF_EMULATE, + RET_PF_WRITE_PROTECTED, RET_PF_INVALID, RET_PF_FIXED, RET_PF_SPURIOUS, }; +/* + * Define RET_PF_CONTINUE as 0 to allow for + * - efficient machine code when checking for CONTINUE, e.g. + * "TEST %rax, %rax, JNZ", as all "stop!" values are non-zero, + * - kvm_mmu_do_page_fault() to return other RET_PF_* as a positive value. + */ +static_assert(RET_PF_CONTINUE == 0); + +static inline void kvm_mmu_prepare_memory_fault_exit(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault) +{ + kvm_prepare_memory_fault_exit(vcpu, fault->gfn << PAGE_SHIFT, + PAGE_SIZE, fault->write, fault->exec, + fault->is_private); +} + static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, - u32 err, bool prefetch, int *emulation_type) + u64 err, bool prefetch, + int *emulation_type, u8 *level) { struct kvm_page_fault fault = { .addr = cr2_or_gpa, @@ -298,55 +363,55 @@ static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa, .max_level = KVM_MAX_HUGEPAGE_LEVEL, .req_level = PG_LEVEL_4K, .goal_level = PG_LEVEL_4K, - .is_private = kvm_mem_is_private(vcpu->kvm, cr2_or_gpa >> PAGE_SHIFT), + .is_private = err & PFERR_PRIVATE_ACCESS, + + .pfn = KVM_PFN_ERR_FAULT, }; int r; if (vcpu->arch.mmu->root_role.direct) { - fault.gfn = fault.addr >> PAGE_SHIFT; + /* + * Things like memslots don't understand the concept of a shared + * bit. Strip it so that the GFN can be used like normal, and the + * fault.addr can be used when the shared bit is needed. + */ + fault.gfn = gpa_to_gfn(fault.addr) & ~kvm_gfn_direct_bits(vcpu->kvm); fault.slot = kvm_vcpu_gfn_to_memslot(vcpu, fault.gfn); } /* - * Async #PF "faults", a.k.a. prefetch faults, are not faults from the - * guest perspective and have already been counted at the time of the - * original fault. + * With retpoline being active an indirect call is rather expensive, + * so do a direct call in the most common case. */ - if (!prefetch) - vcpu->stat.pf_taken++; - if (IS_ENABLED(CONFIG_MITIGATION_RETPOLINE) && fault.is_tdp) r = kvm_tdp_page_fault(vcpu, &fault); else r = vcpu->arch.mmu->page_fault(vcpu, &fault); + /* + * Not sure what's happening, but punt to userspace and hope that + * they can fix it by changing memory to shared, or they can + * provide a better error. + */ + if (r == RET_PF_EMULATE && fault.is_private) { + pr_warn_ratelimited("kvm: unexpected emulation request on private memory\n"); + kvm_mmu_prepare_memory_fault_exit(vcpu, &fault); + return -EFAULT; + } + if (fault.write_fault_to_shadow_pgtable && emulation_type) *emulation_type |= EMULTYPE_WRITE_PF_TO_SP; + if (level) + *level = fault.goal_level; - /* - * Similar to above, prefetch faults aren't truly spurious, and the - * async #PF path doesn't do emulation. Do count faults that are fixed - * by the async #PF handler though, otherwise they'll never be counted. - */ - if (r == RET_PF_FIXED) - vcpu->stat.pf_fixed++; - else if (prefetch) - ; - else if (r == RET_PF_EMULATE) - vcpu->stat.pf_emulate++; - else if (r == RET_PF_SPURIOUS) - vcpu->stat.pf_spurious++; return r; } int kvm_mmu_max_mapping_level(struct kvm *kvm, - const struct kvm_memory_slot *slot, gfn_t gfn, - int max_level); + const struct kvm_memory_slot *slot, gfn_t gfn); void kvm_mmu_hugepage_adjust(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault); void disallowed_hugepage_adjust(struct kvm_page_fault *fault, u64 spte, int cur_level); -void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc); - void track_possible_nx_huge_page(struct kvm *kvm, struct kvm_mmu_page *sp); void untrack_possible_nx_huge_page(struct kvm *kvm, struct kvm_mmu_page *sp); diff --git a/arch/x86/kvm/mmu/mmutrace.h b/arch/x86/kvm/mmu/mmutrace.h index ae86820cef69..f35a830ce469 100644 --- a/arch/x86/kvm/mmu/mmutrace.h +++ b/arch/x86/kvm/mmu/mmutrace.h @@ -57,6 +57,7 @@ TRACE_DEFINE_ENUM(RET_PF_CONTINUE); TRACE_DEFINE_ENUM(RET_PF_RETRY); TRACE_DEFINE_ENUM(RET_PF_EMULATE); +TRACE_DEFINE_ENUM(RET_PF_WRITE_PROTECTED); TRACE_DEFINE_ENUM(RET_PF_INVALID); TRACE_DEFINE_ENUM(RET_PF_FIXED); TRACE_DEFINE_ENUM(RET_PF_SPURIOUS); @@ -260,7 +261,7 @@ TRACE_EVENT( TP_STRUCT__entry( __field(int, vcpu_id) __field(gpa_t, cr2_or_gpa) - __field(u32, error_code) + __field(u64, error_code) __field(u64 *, sptep) __field(u64, old_spte) __field(u64, new_spte) diff --git a/arch/x86/kvm/mmu/page_track.c b/arch/x86/kvm/mmu/page_track.c index f6448284c18e..561c331fd6ec 100644 --- a/arch/x86/kvm/mmu/page_track.c +++ b/arch/x86/kvm/mmu/page_track.c @@ -41,7 +41,7 @@ bool kvm_page_track_write_tracking_enabled(struct kvm *kvm) void kvm_page_track_free_memslot(struct kvm_memory_slot *slot) { - kvfree(slot->arch.gfn_write_track); + vfree(slot->arch.gfn_write_track); slot->arch.gfn_write_track = NULL; } diff --git a/arch/x86/kvm/mmu/paging_tmpl.h b/arch/x86/kvm/mmu/paging_tmpl.h index 4d4e98fe4f35..68e323568e95 100644 --- a/arch/x86/kvm/mmu/paging_tmpl.h +++ b/arch/x86/kvm/mmu/paging_tmpl.h @@ -497,21 +497,20 @@ error: * The other bits are set to 0. */ if (!(errcode & PFERR_RSVD_MASK)) { - vcpu->arch.exit_qualification &= (EPT_VIOLATION_GVA_IS_VALID | - EPT_VIOLATION_GVA_TRANSLATED); + walker->fault.exit_qualification = 0; + if (write_fault) - vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_WRITE; + walker->fault.exit_qualification |= EPT_VIOLATION_ACC_WRITE; if (user_fault) - vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_READ; + walker->fault.exit_qualification |= EPT_VIOLATION_ACC_READ; if (fetch_fault) - vcpu->arch.exit_qualification |= EPT_VIOLATION_ACC_INSTR; + walker->fault.exit_qualification |= EPT_VIOLATION_ACC_INSTR; /* * Note, pte_access holds the raw RWX bits from the EPTE, not * ACC_*_MASK flags! */ - vcpu->arch.exit_qualification |= (pte_access & VMX_EPT_RWX_MASK) << - EPT_VIOLATION_RWX_SHIFT; + walker->fault.exit_qualification |= EPT_VIOLATION_RWX_TO_PROT(pte_access); } #endif walker->fault.address = addr; @@ -533,10 +532,8 @@ static bool FNAME(prefetch_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, u64 *spte, pt_element_t gpte) { - struct kvm_memory_slot *slot; unsigned pte_access; gfn_t gfn; - kvm_pfn_t pfn; if (FNAME(prefetch_invalid_gpte)(vcpu, sp, spte, gpte)) return false; @@ -545,17 +542,7 @@ FNAME(prefetch_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, pte_access = sp->role.access & FNAME(gpte_access)(gpte); FNAME(protect_clean_gpte)(vcpu->arch.mmu, &pte_access, gpte); - slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, pte_access & ACC_WRITE_MASK); - if (!slot) - return false; - - pfn = gfn_to_pfn_memslot_atomic(slot, gfn); - if (is_error_pfn(pfn)) - return false; - - mmu_set_spte(vcpu, slot, spte, pte_access, gfn, pfn, NULL); - kvm_release_pfn_clean(pfn); - return true; + return kvm_mmu_prefetch_sptes(vcpu, gfn, spte, 1, pte_access); } static bool FNAME(gpte_changed)(struct kvm_vcpu *vcpu, @@ -646,10 +633,10 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, * really care if it changes underneath us after this point). */ if (FNAME(gpte_changed)(vcpu, gw, top_level)) - goto out_gpte_changed; + return RET_PF_RETRY; if (WARN_ON_ONCE(!VALID_PAGE(vcpu->arch.mmu->root.hpa))) - goto out_gpte_changed; + return RET_PF_RETRY; /* * Load a new root and retry the faulting instruction in the extremely @@ -659,7 +646,7 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, */ if (unlikely(kvm_mmu_is_dummy_root(vcpu->arch.mmu->root.hpa))) { kvm_make_request(KVM_REQ_MMU_FREE_OBSOLETE_ROOTS, vcpu); - goto out_gpte_changed; + return RET_PF_RETRY; } for_each_shadow_entry(vcpu, fault->addr, it) { @@ -674,34 +661,38 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, sp = kvm_mmu_get_child_sp(vcpu, it.sptep, table_gfn, false, access); - if (sp != ERR_PTR(-EEXIST)) { - /* - * We must synchronize the pagetable before linking it - * because the guest doesn't need to flush tlb when - * the gpte is changed from non-present to present. - * Otherwise, the guest may use the wrong mapping. - * - * For PG_LEVEL_4K, kvm_mmu_get_page() has already - * synchronized it transiently via kvm_sync_page(). - * - * For higher level pagetable, we synchronize it via - * the slower mmu_sync_children(). If it needs to - * break, some progress has been made; return - * RET_PF_RETRY and retry on the next #PF. - * KVM_REQ_MMU_SYNC is not necessary but it - * expedites the process. - */ - if (sp->unsync_children && - mmu_sync_children(vcpu, sp, false)) - return RET_PF_RETRY; - } + /* + * Synchronize the new page before linking it, as the CPU (KVM) + * is architecturally disallowed from inserting non-present + * entries into the TLB, i.e. the guest isn't required to flush + * the TLB when changing the gPTE from non-present to present. + * + * For PG_LEVEL_4K, kvm_mmu_find_shadow_page() has already + * synchronized the page via kvm_sync_page(). + * + * For higher level pages, which cannot be unsync themselves + * but can have unsync children, synchronize via the slower + * mmu_sync_children(). If KVM needs to drop mmu_lock due to + * contention or to reschedule, instruct the caller to retry + * the #PF (mmu_sync_children() ensures forward progress will + * be made). + */ + if (sp != ERR_PTR(-EEXIST) && sp->unsync_children && + mmu_sync_children(vcpu, sp, false)) + return RET_PF_RETRY; /* - * Verify that the gpte in the page we've just write - * protected is still there. + * Verify that the gpte in the page, which is now either + * write-protected or unsync, wasn't modified between the fault + * and acquiring mmu_lock. This needs to be done even when + * reusing an existing shadow page to ensure the information + * gathered by the walker matches the information stored in the + * shadow page (which could have been modified by a different + * vCPU even if the page was already linked). Holding mmu_lock + * prevents the shadow page from changing after this point. */ if (FNAME(gpte_changed)(vcpu, gw, it.level - 1)) - goto out_gpte_changed; + return RET_PF_RETRY; if (sp != ERR_PTR(-EEXIST)) link_shadow_page(vcpu, it.sptep, sp); @@ -755,9 +746,6 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault, FNAME(pte_prefetch)(vcpu, gw, it.sptep); return ret; - -out_gpte_changed: - return RET_PF_RETRY; } /* @@ -805,14 +793,14 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault if (page_fault_handle_page_track(vcpu, fault)) { shadow_page_table_clear_flood(vcpu, fault->addr); - return RET_PF_EMULATE; + return RET_PF_WRITE_PROTECTED; } r = mmu_topup_memory_caches(vcpu, true); if (r) return r; - r = kvm_faultin_pfn(vcpu, fault, walker.pte_access); + r = kvm_mmu_faultin_pfn(vcpu, fault, walker.pte_access); if (r != RET_PF_CONTINUE) return r; @@ -847,8 +835,8 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault r = FNAME(fetch)(vcpu, fault, &walker); out_unlock: + kvm_mmu_finish_page_fault(vcpu, fault, r); write_unlock(&vcpu->kvm->mmu_lock); - kvm_release_pfn_clean(fault->pfn); return r; } @@ -891,9 +879,9 @@ static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, /* * Using the information in sp->shadowed_translation (kvm_mmu_page_get_gfn()) is - * safe because: - * - The spte has a reference to the struct page, so the pfn for a given gfn - * can't change unless all sptes pointing to it are nuked first. + * safe because SPTEs are protected by mmu_notifiers and memslot generations, so + * the pfn for a given gfn can't change unless all SPTEs pointing to the gfn are + * nuked first. * * Returns * < 0: failed to sync spte @@ -911,7 +899,8 @@ static int FNAME(sync_spte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, int gpa_t pte_gpa; gfn_t gfn; - if (WARN_ON_ONCE(!sp->spt[i])) + if (WARN_ON_ONCE(sp->spt[i] == SHADOW_NONPRESENT_VALUE || + !sp->shadowed_translation)) return 0; first_pte_gpa = FNAME(get_level1_sp_gpa)(sp); @@ -933,13 +922,13 @@ static int FNAME(sync_spte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, int return 0; /* - * Drop the SPTE if the new protections would result in a RWX=0 - * SPTE or if the gfn is changing. The RWX=0 case only affects - * EPT with execute-only support, i.e. EPT without an effective - * "present" bit, as all other paging modes will create a - * read-only SPTE if pte_access is zero. + * Drop the SPTE if the new protections result in no effective + * "present" bit or if the gfn is changing. The former case + * only affects EPT with execute-only support with pte_access==0; + * all other paging modes will create a read-only SPTE if + * pte_access is zero. */ - if ((!pte_access && !shadow_present_mask) || + if ((pte_access | shadow_present_mask) == SHADOW_NONPRESENT_VALUE || gfn != kvm_mmu_page_get_gfn(sp, i)) { drop_spte(vcpu->kvm, &sp->spt[i]); return 1; @@ -961,9 +950,14 @@ static int FNAME(sync_spte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, int host_writable = spte & shadow_host_writable_mask; slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn); make_spte(vcpu, sp, slot, pte_access, gfn, - spte_to_pfn(spte), spte, true, false, + spte_to_pfn(spte), spte, true, true, host_writable, &spte); + /* + * There is no need to mark the pfn dirty, as the new protections must + * be a subset of the old protections, i.e. synchronizing a SPTE cannot + * change the SPTE from read-only to writable. + */ return mmu_spte_update(sptep, spte); } diff --git a/arch/x86/kvm/mmu/spte.c b/arch/x86/kvm/mmu/spte.c index 4a599130e9c9..0f9f47b4ab0e 100644 --- a/arch/x86/kvm/mmu/spte.c +++ b/arch/x86/kvm/mmu/spte.c @@ -24,6 +24,8 @@ static bool __ro_after_init allow_mmio_caching; module_param_named(mmio_caching, enable_mmio_caching, bool, 0444); EXPORT_SYMBOL_GPL(enable_mmio_caching); +bool __read_mostly kvm_ad_enabled; + u64 __read_mostly shadow_host_writable_mask; u64 __read_mostly shadow_mmu_writable_mask; u64 __read_mostly shadow_nx_mask; @@ -43,7 +45,25 @@ u64 __read_mostly shadow_acc_track_mask; u64 __read_mostly shadow_nonpresent_or_rsvd_mask; u64 __read_mostly shadow_nonpresent_or_rsvd_lower_gfn_mask; -u8 __read_mostly shadow_phys_bits; +static u8 __init kvm_get_host_maxphyaddr(void) +{ + /* + * boot_cpu_data.x86_phys_bits is reduced when MKTME or SME are detected + * in CPU detection code, but the processor treats those reduced bits as + * 'keyID' thus they are not reserved bits. Therefore KVM needs to look at + * the physical address bits reported by CPUID, i.e. the raw MAXPHYADDR, + * when reasoning about CPU behavior with respect to MAXPHYADDR. + */ + if (likely(boot_cpu_data.extended_cpuid_level >= 0x80000008)) + return cpuid_eax(0x80000008) & 0xff; + + /* + * Quite weird to have VMX or SVM but not MAXPHYADDR; probably a VM with + * custom CPUID. Proceed with whatever the kernel found since these features + * aren't virtualizable (SME/SEV also require CPUIDs higher than 0x80000008). + */ + return boot_cpu_data.x86_phys_bits; +} void __init kvm_mmu_spte_module_init(void) { @@ -55,6 +75,8 @@ void __init kvm_mmu_spte_module_init(void) * will change when the vendor module is (re)loaded. */ allow_mmio_caching = enable_mmio_caching; + + kvm_host.maxphyaddr = kvm_get_host_maxphyaddr(); } static u64 generation_mmio_spte_mask(u64 gen) @@ -74,10 +96,10 @@ u64 make_mmio_spte(struct kvm_vcpu *vcpu, u64 gfn, unsigned int access) u64 spte = generation_mmio_spte_mask(gen); u64 gpa = gfn << PAGE_SHIFT; - WARN_ON_ONCE(!shadow_mmio_value); + WARN_ON_ONCE(!vcpu->kvm->arch.shadow_mmio_value); access &= shadow_mmio_access_mask; - spte |= shadow_mmio_value | access; + spte |= vcpu->kvm->arch.shadow_mmio_value | access; spte |= gpa | shadow_nonpresent_or_rsvd_mask; spte |= (gpa & shadow_nonpresent_or_rsvd_mask) << SHADOW_NONPRESENT_OR_RSVD_MASK_LEN; @@ -107,59 +129,60 @@ static bool kvm_is_mmio_pfn(kvm_pfn_t pfn) } /* - * Returns true if the SPTE has bits that may be set without holding mmu_lock. - * The caller is responsible for checking if the SPTE is shadow-present, and - * for determining whether or not the caller cares about non-leaf SPTEs. + * Returns true if the SPTE needs to be updated atomically due to having bits + * that may be changed without holding mmu_lock, and for which KVM must not + * lose information. E.g. KVM must not drop Dirty bit information. The caller + * is responsible for checking if the SPTE is shadow-present, and for + * determining whether or not the caller cares about non-leaf SPTEs. */ -bool spte_has_volatile_bits(u64 spte) +bool spte_needs_atomic_update(u64 spte) { - /* - * Always atomically update spte if it can be updated - * out of mmu-lock, it can ensure dirty bit is not lost, - * also, it can help us to get a stable is_writable_pte() - * to ensure tlb flush is not missed. - */ + /* SPTEs can be made Writable bit by KVM's fast page fault handler. */ if (!is_writable_pte(spte) && is_mmu_writable_spte(spte)) return true; - if (is_access_track_spte(spte)) + /* + * A/D-disabled SPTEs can be access-tracked by aging, and access-tracked + * SPTEs can be restored by KVM's fast page fault handler. + */ + if (!spte_ad_enabled(spte)) return true; - if (spte_ad_enabled(spte)) { - if (!(spte & shadow_accessed_mask) || - (is_writable_pte(spte) && !(spte & shadow_dirty_mask))) - return true; - } - - return false; + /* + * Dirty and Accessed bits can be set by the CPU. Ignore the Accessed + * bit, as KVM tolerates false negatives/positives, e.g. KVM doesn't + * invalidate TLBs when aging SPTEs, and so it's safe to clobber the + * Accessed bit (and rare in practice). + */ + return is_writable_pte(spte) && !(spte & shadow_dirty_mask); } bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, const struct kvm_memory_slot *slot, unsigned int pte_access, gfn_t gfn, kvm_pfn_t pfn, - u64 old_spte, bool prefetch, bool can_unsync, + u64 old_spte, bool prefetch, bool synchronizing, bool host_writable, u64 *new_spte) { int level = sp->role.level; u64 spte = SPTE_MMU_PRESENT_MASK; bool wrprot = false; - WARN_ON_ONCE(!pte_access && !shadow_present_mask); + /* + * For the EPT case, shadow_present_mask has no RWX bits set if + * exec-only page table entries are supported. In that case, + * ACC_USER_MASK and shadow_user_mask are used to represent + * read access. See FNAME(gpte_access) in paging_tmpl.h. + */ + WARN_ON_ONCE((pte_access | shadow_present_mask) == SHADOW_NONPRESENT_VALUE); if (sp->role.ad_disabled) spte |= SPTE_TDP_AD_DISABLED; else if (kvm_mmu_page_ad_need_write_protect(sp)) spte |= SPTE_TDP_AD_WRPROT_ONLY; - /* - * For the EPT case, shadow_present_mask is 0 if hardware - * supports exec-only page table entries. In that case, - * ACC_USER_MASK and shadow_user_mask are used to represent - * read access. See FNAME(gpte_access) in paging_tmpl.h. - */ spte |= shadow_present_mask; - if (!prefetch) - spte |= spte_shadow_accessed_mask(spte); + if (!prefetch || synchronizing) + spte |= shadow_accessed_mask; /* * For simplicity, enforce the NX huge page mitigation even if not @@ -190,8 +213,8 @@ bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, spte |= PT_PAGE_SIZE_MASK; if (shadow_memtype_mask) - spte |= static_call(kvm_x86_get_mt_mask)(vcpu, gfn, - kvm_is_mmio_pfn(pfn)); + spte |= kvm_x86_call(get_mt_mask)(vcpu, gfn, + kvm_is_mmio_pfn(pfn)); if (host_writable) spte |= shadow_host_writable_mask; else @@ -203,41 +226,39 @@ bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, spte |= (u64)pfn << PAGE_SHIFT; if (pte_access & ACC_WRITE_MASK) { - spte |= PT_WRITABLE_MASK | shadow_mmu_writable_mask; - - /* - * Optimization: for pte sync, if spte was writable the hash - * lookup is unnecessary (and expensive). Write protection - * is responsibility of kvm_mmu_get_page / kvm_mmu_sync_roots. - * Same reasoning can be applied to dirty page accounting. - */ - if (is_writable_pte(old_spte)) - goto out; - /* * Unsync shadow pages that are reachable by the new, writable * SPTE. Write-protect the SPTE if the page can't be unsync'd, * e.g. it's write-tracked (upper-level SPs) or has one or more * shadow pages and unsync'ing pages is not allowed. + * + * When overwriting an existing leaf SPTE, and the old SPTE was + * writable, skip trying to unsync shadow pages as any relevant + * shadow pages must already be unsync, i.e. the hash lookup is + * unnecessary (and expensive). Note, this relies on KVM not + * changing PFNs without first zapping the old SPTE, which is + * guaranteed by both the shadow MMU and the TDP MMU. */ - if (mmu_try_to_unsync_pages(vcpu->kvm, slot, gfn, can_unsync, prefetch)) { + if ((!is_last_spte(old_spte, level) || !is_writable_pte(old_spte)) && + mmu_try_to_unsync_pages(vcpu->kvm, slot, gfn, synchronizing, prefetch)) wrprot = true; - pte_access &= ~ACC_WRITE_MASK; - spte &= ~(PT_WRITABLE_MASK | shadow_mmu_writable_mask); - } + else + spte |= PT_WRITABLE_MASK | shadow_mmu_writable_mask | + shadow_dirty_mask; } - if (pte_access & ACC_WRITE_MASK) - spte |= spte_shadow_dirty_mask(spte); - -out: - if (prefetch) + if (prefetch && !synchronizing) spte = mark_spte_for_access_track(spte); WARN_ONCE(is_rsvd_spte(&vcpu->arch.mmu->shadow_zero_check, spte, level), "spte = 0x%llx, level = %d, rsvd bits = 0x%llx", spte, level, get_rsvd_bits(&vcpu->arch.mmu->shadow_zero_check, spte, level)); + /* + * Mark the memslot dirty *after* modifying it for access tracking. + * Unlike folios, memslots can be safely marked dirty out of mmu_lock, + * i.e. in the fast page fault handler. + */ if ((spte & PT_WRITABLE_MASK) && kvm_slot_dirty_track_enabled(slot)) { /* Enforced by kvm_mmu_hugepage_adjust. */ WARN_ON_ONCE(level > PG_LEVEL_4K); @@ -248,15 +269,15 @@ out: return wrprot; } -static u64 make_spte_executable(u64 spte) +static u64 modify_spte_protections(u64 spte, u64 set, u64 clear) { bool is_access_track = is_access_track_spte(spte); if (is_access_track) spte = restore_acc_track_spte(spte); - spte &= ~shadow_nx_mask; - spte |= shadow_x_mask; + KVM_MMU_WARN_ON(set & clear); + spte = (spte | set) & ~clear; if (is_access_track) spte = mark_spte_for_access_track(spte); @@ -264,6 +285,16 @@ static u64 make_spte_executable(u64 spte) return spte; } +static u64 make_spte_executable(u64 spte) +{ + return modify_spte_protections(spte, shadow_x_mask, shadow_nx_mask); +} + +static u64 make_spte_nonexecutable(u64 spte) +{ + return modify_spte_protections(spte, shadow_nx_mask, shadow_x_mask); +} + /* * Construct an SPTE that maps a sub-page of the given huge page SPTE where * `index` identifies which sub-page. @@ -271,18 +302,12 @@ static u64 make_spte_executable(u64 spte) * This is used during huge page splitting to build the SPTEs that make up the * new page table. */ -u64 make_huge_page_split_spte(struct kvm *kvm, u64 huge_spte, union kvm_mmu_page_role role, - int index) +u64 make_small_spte(struct kvm *kvm, u64 huge_spte, + union kvm_mmu_page_role role, int index) { - u64 child_spte; - - if (WARN_ON_ONCE(!is_shadow_present_pte(huge_spte))) - return 0; + u64 child_spte = huge_spte; - if (WARN_ON_ONCE(!is_large_pte(huge_spte))) - return 0; - - child_spte = huge_spte; + KVM_BUG_ON(!is_shadow_present_pte(huge_spte) || !is_large_pte(huge_spte), kvm); /* * The child_spte already has the base address of the huge page being @@ -306,6 +331,26 @@ u64 make_huge_page_split_spte(struct kvm *kvm, u64 huge_spte, union kvm_mmu_page return child_spte; } +u64 make_huge_spte(struct kvm *kvm, u64 small_spte, int level) +{ + u64 huge_spte; + + KVM_BUG_ON(!is_shadow_present_pte(small_spte) || level == PG_LEVEL_4K, kvm); + + huge_spte = small_spte | PT_PAGE_SIZE_MASK; + + /* + * huge_spte already has the address of the sub-page being collapsed + * from small_spte, so just clear the lower address bits to create the + * huge page address. + */ + huge_spte &= KVM_HPAGE_MASK(level) | ~PAGE_MASK; + + if (is_nx_huge_page_enabled(kvm)) + huge_spte = make_spte_nonexecutable(huge_spte); + + return huge_spte; +} u64 make_nonleaf_spte(u64 *child_pt, bool ad_disabled) { @@ -322,22 +367,6 @@ u64 make_nonleaf_spte(u64 *child_pt, bool ad_disabled) return spte; } -u64 kvm_mmu_changed_pte_notifier_make_spte(u64 old_spte, kvm_pfn_t new_pfn) -{ - u64 new_spte; - - new_spte = old_spte & ~SPTE_BASE_ADDR_MASK; - new_spte |= (u64)new_pfn << PAGE_SHIFT; - - new_spte &= ~PT_WRITABLE_MASK; - new_spte &= ~shadow_host_writable_mask; - new_spte &= ~shadow_mmu_writable_mask; - - new_spte = mark_spte_for_access_track(new_spte); - - return new_spte; -} - u64 mark_spte_for_access_track(u64 spte) { if (spte_ad_enabled(spte)) @@ -354,7 +383,7 @@ u64 mark_spte_for_access_track(u64 spte) spte |= (spte & SHADOW_ACC_TRACK_SAVED_BITS_MASK) << SHADOW_ACC_TRACK_SAVED_BITS_SHIFT; - spte &= ~shadow_acc_track_mask; + spte &= ~(shadow_acc_track_mask | shadow_accessed_mask); return spte; } @@ -393,13 +422,13 @@ void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 mmio_mask, u64 access_mask) mmio_value = 0; /* - * The masked MMIO value must obviously match itself and a removed SPTE - * must not get a false positive. Removed SPTEs and MMIO SPTEs should - * never collide as MMIO must set some RWX bits, and removed SPTEs must + * The masked MMIO value must obviously match itself and a frozen SPTE + * must not get a false positive. Frozen SPTEs and MMIO SPTEs should + * never collide as MMIO must set some RWX bits, and frozen SPTEs must * not set any RWX bits. */ if (WARN_ON((mmio_value & mmio_mask) != mmio_value) || - WARN_ON(mmio_value && (REMOVED_SPTE & mmio_mask) == mmio_value)) + WARN_ON(mmio_value && (FROZEN_SPTE & mmio_mask) == mmio_value)) mmio_value = 0; if (!mmio_value) @@ -424,12 +453,16 @@ EXPORT_SYMBOL_GPL(kvm_mmu_set_me_spte_mask); void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only) { + kvm_ad_enabled = has_ad_bits; + shadow_user_mask = VMX_EPT_READABLE_MASK; - shadow_accessed_mask = has_ad_bits ? VMX_EPT_ACCESS_BIT : 0ull; - shadow_dirty_mask = has_ad_bits ? VMX_EPT_DIRTY_BIT : 0ull; + shadow_accessed_mask = VMX_EPT_ACCESS_BIT; + shadow_dirty_mask = VMX_EPT_DIRTY_BIT; shadow_nx_mask = 0ull; shadow_x_mask = VMX_EPT_EXECUTABLE_MASK; - shadow_present_mask = has_exec_only ? 0ull : VMX_EPT_READABLE_MASK; + /* VMX_EPT_SUPPRESS_VE_BIT is needed for W or X violation. */ + shadow_present_mask = + (has_exec_only ? 0ull : VMX_EPT_READABLE_MASK) | VMX_EPT_SUPPRESS_VE_BIT; /* * EPT overrides the host MTRRs, and so KVM must program the desired * memtype directly into the SPTEs. Note, this mask is just the mask @@ -446,7 +479,7 @@ void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only) * of an EPT paging-structure entry is 110b (write/execute). */ kvm_mmu_set_mmio_spte_mask(VMX_EPT_MISCONFIG_WX_VALUE, - VMX_EPT_RWX_MASK, 0); + VMX_EPT_RWX_MASK | VMX_EPT_SUPPRESS_VE_BIT, 0); } EXPORT_SYMBOL_GPL(kvm_mmu_set_ept_masks); @@ -455,7 +488,7 @@ void kvm_mmu_reset_all_pte_masks(void) u8 low_phys_bits; u64 mask; - shadow_phys_bits = kvm_get_shadow_phys_bits(); + kvm_ad_enabled = true; /* * If the CPU has 46 or less physical address bits, then set an @@ -508,7 +541,7 @@ void kvm_mmu_reset_all_pte_masks(void) * 52-bit physical addresses then there are no reserved PA bits in the * PTEs and so the reserved PA approach must be disabled. */ - if (shadow_phys_bits < 52) + if (kvm_host.maxphyaddr < 52) mask = BIT_ULL(51) | PT_PRESENT_MASK; else mask = 0; diff --git a/arch/x86/kvm/mmu/spte.h b/arch/x86/kvm/mmu/spte.h index a129951c9a88..79cdceba9857 100644 --- a/arch/x86/kvm/mmu/spte.h +++ b/arch/x86/kvm/mmu/spte.h @@ -3,6 +3,8 @@ #ifndef KVM_X86_MMU_SPTE_H #define KVM_X86_MMU_SPTE_H +#include <asm/vmx.h> + #include "mmu.h" #include "mmu_internal.h" @@ -149,6 +151,31 @@ static_assert(MMIO_SPTE_GEN_LOW_BITS == 8 && MMIO_SPTE_GEN_HIGH_BITS == 11); #define MMIO_SPTE_GEN_MASK GENMASK_ULL(MMIO_SPTE_GEN_LOW_BITS + MMIO_SPTE_GEN_HIGH_BITS - 1, 0) +/* + * Non-present SPTE value needs to set bit 63 for TDX, in order to suppress + * #VE and get EPT violations on non-present PTEs. We can use the + * same value also without TDX for both VMX and SVM: + * + * For SVM NPT, for non-present spte (bit 0 = 0), other bits are ignored. + * For VMX EPT, bit 63 is ignored if #VE is disabled. (EPT_VIOLATION_VE=0) + * bit 63 is #VE suppress if #VE is enabled. (EPT_VIOLATION_VE=1) + */ +#ifdef CONFIG_X86_64 +#define SHADOW_NONPRESENT_VALUE BIT_ULL(63) +static_assert(!(SHADOW_NONPRESENT_VALUE & SPTE_MMU_PRESENT_MASK)); +#else +#define SHADOW_NONPRESENT_VALUE 0ULL +#endif + + +/* + * True if A/D bits are supported in hardware and are enabled by KVM. When + * enabled, KVM uses A/D bits for all non-nested MMUs. Because L1 can disable + * A/D bits in EPTP12, SP and SPTE variants are needed to handle the scenario + * where KVM is using A/D bits for L1, but not L2. + */ +extern bool __read_mostly kvm_ad_enabled; + extern u64 __read_mostly shadow_host_writable_mask; extern u64 __read_mostly shadow_mmu_writable_mask; extern u64 __read_mostly shadow_nx_mask; @@ -184,24 +211,24 @@ extern u64 __read_mostly shadow_nonpresent_or_rsvd_mask; /* * If a thread running without exclusive control of the MMU lock must perform a - * multi-part operation on an SPTE, it can set the SPTE to REMOVED_SPTE as a + * multi-part operation on an SPTE, it can set the SPTE to FROZEN_SPTE as a * non-present intermediate value. Other threads which encounter this value * should not modify the SPTE. * * Use a semi-arbitrary value that doesn't set RWX bits, i.e. is not-present on * both AMD and Intel CPUs, and doesn't set PFN bits, i.e. doesn't create a L1TF - * vulnerability. Use only low bits to avoid 64-bit immediates. + * vulnerability. * * Only used by the TDP MMU. */ -#define REMOVED_SPTE 0x5a0ULL +#define FROZEN_SPTE (SHADOW_NONPRESENT_VALUE | 0x5a0ULL) -/* Removed SPTEs must not be misconstrued as shadow present PTEs. */ -static_assert(!(REMOVED_SPTE & SPTE_MMU_PRESENT_MASK)); +/* Frozen SPTEs must not be misconstrued as shadow present PTEs. */ +static_assert(!(FROZEN_SPTE & SPTE_MMU_PRESENT_MASK)); -static inline bool is_removed_spte(u64 spte) +static inline bool is_frozen_spte(u64 spte) { - return spte == REMOVED_SPTE; + return spte == FROZEN_SPTE; } /* Get an SPTE's index into its parent's page table (and the spt array). */ @@ -249,9 +276,14 @@ static inline struct kvm_mmu_page *root_to_sp(hpa_t root) return spte_to_child_sp(root); } -static inline bool is_mmio_spte(u64 spte) +static inline bool is_mirror_sptep(tdp_ptep_t sptep) { - return (spte & shadow_mmio_mask) == shadow_mmio_value && + return is_mirror_sp(sptep_to_sp(rcu_dereference(sptep))); +} + +static inline bool is_mmio_spte(struct kvm *kvm, u64 spte) +{ + return (spte & shadow_mmio_mask) == kvm->arch.shadow_mmio_value && likely(enable_mmio_caching); } @@ -260,15 +292,11 @@ static inline bool is_shadow_present_pte(u64 pte) return !!(pte & SPTE_MMU_PRESENT_MASK); } -/* - * Returns true if A/D bits are supported in hardware and are enabled by KVM. - * When enabled, KVM uses A/D bits for all non-nested MMUs. Because L1 can - * disable A/D bits in EPTP12, SP and SPTE variants are needed to handle the - * scenario where KVM is using A/D bits for L1, but not L2. - */ -static inline bool kvm_ad_enabled(void) +static inline bool is_ept_ve_possible(u64 spte) { - return !!shadow_accessed_mask; + return (shadow_present_mask & VMX_EPT_SUPPRESS_VE_BIT) && + !(spte & VMX_EPT_SUPPRESS_VE_BIT) && + (spte & VMX_EPT_RWX_MASK) != VMX_EPT_MISCONFIG_WX_VALUE; } static inline bool sp_ad_disabled(struct kvm_mmu_page *sp) @@ -293,18 +321,6 @@ static inline bool spte_ad_need_write_protect(u64 spte) return (spte & SPTE_TDP_AD_MASK) != SPTE_TDP_AD_ENABLED; } -static inline u64 spte_shadow_accessed_mask(u64 spte) -{ - KVM_MMU_WARN_ON(!is_shadow_present_pte(spte)); - return spte_ad_enabled(spte) ? shadow_accessed_mask : 0; -} - -static inline u64 spte_shadow_dirty_mask(u64 spte) -{ - KVM_MMU_WARN_ON(!is_shadow_present_pte(spte)); - return spte_ad_enabled(spte) ? shadow_dirty_mask : 0; -} - static inline bool is_access_track_spte(u64 spte) { return !spte_ad_enabled(spte) && (spte & shadow_acc_track_mask) == 0; @@ -332,17 +348,7 @@ static inline kvm_pfn_t spte_to_pfn(u64 pte) static inline bool is_accessed_spte(u64 spte) { - u64 accessed_mask = spte_shadow_accessed_mask(spte); - - return accessed_mask ? spte & accessed_mask - : !is_access_track_spte(spte); -} - -static inline bool is_dirty_spte(u64 spte) -{ - u64 dirty_mask = spte_shadow_dirty_mask(spte); - - return dirty_mask ? spte & dirty_mask : spte & PT_WRITABLE_MASK; + return spte & shadow_accessed_mask; } static inline u64 get_rsvd_bits(struct rsvd_bits_validate *rsvd_check, u64 pte, @@ -460,6 +466,50 @@ static inline bool is_mmu_writable_spte(u64 spte) return spte & shadow_mmu_writable_mask; } +/* + * Returns true if the access indicated by @fault is allowed by the existing + * SPTE protections. Note, the caller is responsible for checking that the + * SPTE is a shadow-present, leaf SPTE (either before or after). + */ +static inline bool is_access_allowed(struct kvm_page_fault *fault, u64 spte) +{ + if (fault->exec) + return is_executable_pte(spte); + + if (fault->write) + return is_writable_pte(spte); + + /* Fault was on Read access */ + return spte & PT_PRESENT_MASK; +} + +/* + * If the MMU-writable flag is cleared, i.e. the SPTE is write-protected for + * write-tracking, remote TLBs must be flushed, even if the SPTE was read-only, + * as KVM allows stale Writable TLB entries to exist. When dirty logging, KVM + * flushes TLBs based on whether or not dirty bitmap/ring entries were reaped, + * not whether or not SPTEs were modified, i.e. only the write-tracking case + * needs to flush at the time the SPTEs is modified, before dropping mmu_lock. + * + * Don't flush if the Accessed bit is cleared, as access tracking tolerates + * false negatives, e.g. KVM x86 omits TLB flushes even when aging SPTEs for a + * mmu_notifier.clear_flush_young() event. + * + * Lastly, don't flush if the Dirty bit is cleared, as KVM unconditionally + * flushes when enabling dirty logging (see kvm_mmu_slot_apply_flags()), and + * when clearing dirty logs, KVM flushes based on whether or not dirty entries + * were reaped from the bitmap/ring, not whether or not dirty SPTEs were found. + * + * Note, this logic only applies to shadow-present leaf SPTEs. The caller is + * responsible for checking that the old SPTE is shadow-present, and is also + * responsible for determining whether or not a TLB flush is required when + * modifying a shadow-present non-leaf SPTE. + */ +static inline bool leaf_spte_change_needs_tlb_flush(u64 old_spte, u64 new_spte) +{ + return is_mmu_writable_spte(old_spte) && !is_mmu_writable_spte(new_spte); +} + static inline u64 get_mmio_spte_generation(u64 spte) { u64 gen; @@ -469,15 +519,16 @@ static inline u64 get_mmio_spte_generation(u64 spte) return gen; } -bool spte_has_volatile_bits(u64 spte); +bool spte_needs_atomic_update(u64 spte); bool make_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, const struct kvm_memory_slot *slot, unsigned int pte_access, gfn_t gfn, kvm_pfn_t pfn, - u64 old_spte, bool prefetch, bool can_unsync, + u64 old_spte, bool prefetch, bool synchronizing, bool host_writable, u64 *new_spte); -u64 make_huge_page_split_spte(struct kvm *kvm, u64 huge_spte, - union kvm_mmu_page_role role, int index); +u64 make_small_spte(struct kvm *kvm, u64 huge_spte, + union kvm_mmu_page_role role, int index); +u64 make_huge_spte(struct kvm *kvm, u64 small_spte, int level); u64 make_nonleaf_spte(u64 *child_pt, bool ad_disabled); u64 make_mmio_spte(struct kvm_vcpu *vcpu, u64 gfn, unsigned int access); u64 mark_spte_for_access_track(u64 spte); @@ -496,8 +547,6 @@ static inline u64 restore_acc_track_spte(u64 spte) return spte; } -u64 kvm_mmu_changed_pte_notifier_make_spte(u64 old_spte, kvm_pfn_t new_pfn); - void __init kvm_mmu_spte_module_init(void); void kvm_mmu_reset_all_pte_masks(void); diff --git a/arch/x86/kvm/mmu/tdp_iter.c b/arch/x86/kvm/mmu/tdp_iter.c index 04c247bfe318..9e17bfa80901 100644 --- a/arch/x86/kvm/mmu/tdp_iter.c +++ b/arch/x86/kvm/mmu/tdp_iter.c @@ -12,7 +12,7 @@ static void tdp_iter_refresh_sptep(struct tdp_iter *iter) { iter->sptep = iter->pt_path[iter->level - 1] + - SPTE_INDEX(iter->gfn << PAGE_SHIFT, iter->level); + SPTE_INDEX((iter->gfn | iter->gfn_bits) << PAGE_SHIFT, iter->level); iter->old_spte = kvm_tdp_mmu_read_spte(iter->sptep); } @@ -37,15 +37,17 @@ void tdp_iter_restart(struct tdp_iter *iter) * rooted at root_pt, starting with the walk to translate next_last_level_gfn. */ void tdp_iter_start(struct tdp_iter *iter, struct kvm_mmu_page *root, - int min_level, gfn_t next_last_level_gfn) + int min_level, gfn_t next_last_level_gfn, gfn_t gfn_bits) { if (WARN_ON_ONCE(!root || (root->role.level < 1) || - (root->role.level > PT64_ROOT_MAX_LEVEL))) { + (root->role.level > PT64_ROOT_MAX_LEVEL) || + (gfn_bits && next_last_level_gfn >= gfn_bits))) { iter->valid = false; return; } iter->next_last_level_gfn = next_last_level_gfn; + iter->gfn_bits = gfn_bits; iter->root_level = root->role.level; iter->min_level = min_level; iter->pt_path[iter->root_level - 1] = (tdp_ptep_t)root->spt; @@ -113,7 +115,7 @@ static bool try_step_side(struct tdp_iter *iter) * Check if the iterator is already at the end of the current page * table. */ - if (SPTE_INDEX(iter->gfn << PAGE_SHIFT, iter->level) == + if (SPTE_INDEX((iter->gfn | iter->gfn_bits) << PAGE_SHIFT, iter->level) == (SPTE_ENT_PER_PAGE - 1)) return false; diff --git a/arch/x86/kvm/mmu/tdp_iter.h b/arch/x86/kvm/mmu/tdp_iter.h index fae559559a80..364c5da6c499 100644 --- a/arch/x86/kvm/mmu/tdp_iter.h +++ b/arch/x86/kvm/mmu/tdp_iter.h @@ -21,37 +21,39 @@ static inline u64 kvm_tdp_mmu_read_spte(tdp_ptep_t sptep) static inline u64 kvm_tdp_mmu_write_spte_atomic(tdp_ptep_t sptep, u64 new_spte) { + KVM_MMU_WARN_ON(is_ept_ve_possible(new_spte)); return xchg(rcu_dereference(sptep), new_spte); } +static inline u64 tdp_mmu_clear_spte_bits_atomic(tdp_ptep_t sptep, u64 mask) +{ + atomic64_t *sptep_atomic = (atomic64_t *)rcu_dereference(sptep); + + return (u64)atomic64_fetch_and(~mask, sptep_atomic); +} + static inline void __kvm_tdp_mmu_write_spte(tdp_ptep_t sptep, u64 new_spte) { + KVM_MMU_WARN_ON(is_ept_ve_possible(new_spte)); WRITE_ONCE(*rcu_dereference(sptep), new_spte); } /* - * SPTEs must be modified atomically if they are shadow-present, leaf - * SPTEs, and have volatile bits, i.e. has bits that can be set outside - * of mmu_lock. The Writable bit can be set by KVM's fast page fault - * handler, and Accessed and Dirty bits can be set by the CPU. - * - * Note, non-leaf SPTEs do have Accessed bits and those bits are - * technically volatile, but KVM doesn't consume the Accessed bit of - * non-leaf SPTEs, i.e. KVM doesn't care if it clobbers the bit. This - * logic needs to be reassessed if KVM were to use non-leaf Accessed - * bits, e.g. to skip stepping down into child SPTEs when aging SPTEs. + * SPTEs must be modified atomically if they are shadow-present, leaf SPTEs, + * and have volatile bits (bits that can be set outside of mmu_lock) that + * must not be clobbered. */ -static inline bool kvm_tdp_mmu_spte_need_atomic_write(u64 old_spte, int level) +static inline bool kvm_tdp_mmu_spte_need_atomic_update(u64 old_spte, int level) { return is_shadow_present_pte(old_spte) && is_last_spte(old_spte, level) && - spte_has_volatile_bits(old_spte); + spte_needs_atomic_update(old_spte); } static inline u64 kvm_tdp_mmu_write_spte(tdp_ptep_t sptep, u64 old_spte, u64 new_spte, int level) { - if (kvm_tdp_mmu_spte_need_atomic_write(old_spte, level)) + if (kvm_tdp_mmu_spte_need_atomic_update(old_spte, level)) return kvm_tdp_mmu_write_spte_atomic(sptep, new_spte); __kvm_tdp_mmu_write_spte(sptep, new_spte); @@ -61,12 +63,8 @@ static inline u64 kvm_tdp_mmu_write_spte(tdp_ptep_t sptep, u64 old_spte, static inline u64 tdp_mmu_clear_spte_bits(tdp_ptep_t sptep, u64 old_spte, u64 mask, int level) { - atomic64_t *sptep_atomic; - - if (kvm_tdp_mmu_spte_need_atomic_write(old_spte, level)) { - sptep_atomic = (atomic64_t *)rcu_dereference(sptep); - return (u64)atomic64_fetch_and(~mask, sptep_atomic); - } + if (kvm_tdp_mmu_spte_need_atomic_update(old_spte, level)) + return tdp_mmu_clear_spte_bits_atomic(sptep, mask); __kvm_tdp_mmu_write_spte(sptep, old_spte & ~mask); return old_spte; @@ -91,8 +89,10 @@ struct tdp_iter { tdp_ptep_t pt_path[PT64_ROOT_MAX_LEVEL]; /* A pointer to the current SPTE */ tdp_ptep_t sptep; - /* The lowest GFN mapped by the current SPTE */ + /* The lowest GFN (mask bits excluded) mapped by the current SPTE */ gfn_t gfn; + /* Mask applied to convert the GFN to the mapping GPA */ + gfn_t gfn_bits; /* The level of the root page given to the iterator */ int root_level; /* The lowest level the iterator should traverse to */ @@ -120,18 +120,23 @@ struct tdp_iter { * Iterates over every SPTE mapping the GFN range [start, end) in a * preorder traversal. */ -#define for_each_tdp_pte_min_level(iter, root, min_level, start, end) \ - for (tdp_iter_start(&iter, root, min_level, start); \ - iter.valid && iter.gfn < end; \ +#define for_each_tdp_pte_min_level(iter, kvm, root, min_level, start, end) \ + for (tdp_iter_start(&iter, root, min_level, start, kvm_gfn_root_bits(kvm, root)); \ + iter.valid && iter.gfn < end; \ tdp_iter_next(&iter)) -#define for_each_tdp_pte(iter, root, start, end) \ - for_each_tdp_pte_min_level(iter, root, PG_LEVEL_4K, start, end) +#define for_each_tdp_pte_min_level_all(iter, root, min_level) \ + for (tdp_iter_start(&iter, root, min_level, 0, 0); \ + iter.valid && iter.gfn < tdp_mmu_max_gfn_exclusive(); \ + tdp_iter_next(&iter)) + +#define for_each_tdp_pte(iter, kvm, root, start, end) \ + for_each_tdp_pte_min_level(iter, kvm, root, PG_LEVEL_4K, start, end) tdp_ptep_t spte_to_child_pt(u64 pte, int level); void tdp_iter_start(struct tdp_iter *iter, struct kvm_mmu_page *root, - int min_level, gfn_t next_last_level_gfn); + int min_level, gfn_t next_last_level_gfn, gfn_t gfn_bits); void tdp_iter_next(struct tdp_iter *iter); void tdp_iter_restart(struct tdp_iter *iter); diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c index d078157e62aa..7cc0564f5f97 100644 --- a/arch/x86/kvm/mmu/tdp_mmu.c +++ b/arch/x86/kvm/mmu/tdp_mmu.c @@ -37,8 +37,8 @@ void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm) * for zapping and thus puts the TDP MMU's reference to each root, i.e. * ultimately frees all roots. */ - kvm_tdp_mmu_invalidate_all_roots(kvm); - kvm_tdp_mmu_zap_invalidated_roots(kvm); + kvm_tdp_mmu_invalidate_roots(kvm, KVM_VALID_ROOTS); + kvm_tdp_mmu_zap_invalidated_roots(kvm, false); WARN_ON(atomic64_read(&kvm->arch.tdp_mmu_pages)); WARN_ON(!list_empty(&kvm->arch.tdp_mmu_roots)); @@ -53,6 +53,7 @@ void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm) static void tdp_mmu_free_sp(struct kvm_mmu_page *sp) { + free_page((unsigned long)sp->external_spt); free_page((unsigned long)sp->spt); kmem_cache_free(mmu_page_header_cache, sp); } @@ -91,19 +92,33 @@ void kvm_tdp_mmu_put_root(struct kvm *kvm, struct kvm_mmu_page *root) call_rcu(&root->rcu_head, tdp_mmu_free_sp_rcu_callback); } +static bool tdp_mmu_root_match(struct kvm_mmu_page *root, + enum kvm_tdp_mmu_root_types types) +{ + if (WARN_ON_ONCE(!(types & KVM_VALID_ROOTS))) + return false; + + if (root->role.invalid && !(types & KVM_INVALID_ROOTS)) + return false; + + if (likely(!is_mirror_sp(root))) + return types & KVM_DIRECT_ROOTS; + return types & KVM_MIRROR_ROOTS; +} + /* * Returns the next root after @prev_root (or the first root if @prev_root is - * NULL). A reference to the returned root is acquired, and the reference to - * @prev_root is released (the caller obviously must hold a reference to - * @prev_root if it's non-NULL). + * NULL) that matches with @types. A reference to the returned root is + * acquired, and the reference to @prev_root is released (the caller obviously + * must hold a reference to @prev_root if it's non-NULL). * - * If @only_valid is true, invalid roots are skipped. + * Roots that doesn't match with @types are skipped. * * Returns NULL if the end of tdp_mmu_roots was reached. */ static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm, struct kvm_mmu_page *prev_root, - bool only_valid) + enum kvm_tdp_mmu_root_types types) { struct kvm_mmu_page *next_root; @@ -124,7 +139,7 @@ static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm, typeof(*next_root), link); while (next_root) { - if ((!only_valid || !next_root->role.invalid) && + if (tdp_mmu_root_match(next_root, types) && kvm_tdp_mmu_get_root(next_root)) break; @@ -149,20 +164,20 @@ static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm, * If shared is set, this function is operating under the MMU lock in read * mode. */ -#define __for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _only_valid) \ - for (_root = tdp_mmu_next_root(_kvm, NULL, _only_valid); \ +#define __for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _types) \ + for (_root = tdp_mmu_next_root(_kvm, NULL, _types); \ ({ lockdep_assert_held(&(_kvm)->mmu_lock); }), _root; \ - _root = tdp_mmu_next_root(_kvm, _root, _only_valid)) \ + _root = tdp_mmu_next_root(_kvm, _root, _types)) \ if (_as_id >= 0 && kvm_mmu_page_as_id(_root) != _as_id) { \ } else #define for_each_valid_tdp_mmu_root_yield_safe(_kvm, _root, _as_id) \ - __for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, true) + __for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, KVM_VALID_ROOTS) #define for_each_tdp_mmu_root_yield_safe(_kvm, _root) \ - for (_root = tdp_mmu_next_root(_kvm, NULL, false); \ + for (_root = tdp_mmu_next_root(_kvm, NULL, KVM_ALL_ROOTS); \ ({ lockdep_assert_held(&(_kvm)->mmu_lock); }), _root; \ - _root = tdp_mmu_next_root(_kvm, _root, false)) + _root = tdp_mmu_next_root(_kvm, _root, KVM_ALL_ROOTS)) /* * Iterate over all TDP MMU roots. Requires that mmu_lock be held for write, @@ -171,18 +186,28 @@ static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm, * Holding mmu_lock for write obviates the need for RCU protection as the list * is guaranteed to be stable. */ -#define __for_each_tdp_mmu_root(_kvm, _root, _as_id, _only_valid) \ +#define __for_each_tdp_mmu_root(_kvm, _root, _as_id, _types) \ list_for_each_entry(_root, &_kvm->arch.tdp_mmu_roots, link) \ if (kvm_lockdep_assert_mmu_lock_held(_kvm, false) && \ ((_as_id >= 0 && kvm_mmu_page_as_id(_root) != _as_id) || \ - ((_only_valid) && (_root)->role.invalid))) { \ + !tdp_mmu_root_match((_root), (_types)))) { \ } else -#define for_each_tdp_mmu_root(_kvm, _root, _as_id) \ - __for_each_tdp_mmu_root(_kvm, _root, _as_id, false) +/* + * Iterate over all TDP MMU roots in an RCU read-side critical section. + * It is safe to iterate over the SPTEs under the root, but their values will + * be unstable, so all writes must be atomic. As this routine is meant to be + * used without holding the mmu_lock at all, any bits that are flipped must + * be reflected in kvm_tdp_mmu_spte_need_atomic_write(). + */ +#define for_each_tdp_mmu_root_rcu(_kvm, _root, _as_id, _types) \ + list_for_each_entry_rcu(_root, &_kvm->arch.tdp_mmu_roots, link) \ + if ((_as_id >= 0 && kvm_mmu_page_as_id(_root) != _as_id) || \ + !tdp_mmu_root_match((_root), (_types))) { \ + } else #define for_each_valid_tdp_mmu_root(_kvm, _root, _as_id) \ - __for_each_tdp_mmu_root(_kvm, _root, _as_id, true) + __for_each_tdp_mmu_root(_kvm, _root, _as_id, KVM_VALID_ROOTS) static struct kvm_mmu_page *tdp_mmu_alloc_sp(struct kvm_vcpu *vcpu) { @@ -223,7 +248,7 @@ static void tdp_mmu_init_child_sp(struct kvm_mmu_page *child_sp, tdp_mmu_init_sp(child_sp, iter->sptep, iter->gfn, role); } -int kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu) +void kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu, bool mirror) { struct kvm_mmu *mmu = vcpu->arch.mmu; union kvm_mmu_page_role role = mmu->root_role; @@ -231,6 +256,9 @@ int kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu) struct kvm *kvm = vcpu->kvm; struct kvm_mmu_page *root; + if (mirror) + role.is_mirror = true; + /* * Check for an existing root before acquiring the pages lock to avoid * unnecessary serialization if multiple vCPUs are loading a new root. @@ -282,9 +310,12 @@ out_read_unlock: * and actually consuming the root if it's invalidated after dropping * mmu_lock, and the root can't be freed as this vCPU holds a reference. */ - mmu->root.hpa = __pa(root->spt); - mmu->root.pgd = 0; - return 0; + if (mirror) { + mmu->mirror_root_hpa = __pa(root->spt); + } else { + mmu->root.hpa = __pa(root->spt); + mmu->root.pgd = 0; + } } static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn, @@ -322,6 +353,29 @@ static void tdp_mmu_unlink_sp(struct kvm *kvm, struct kvm_mmu_page *sp) spin_unlock(&kvm->arch.tdp_mmu_pages_lock); } +static void remove_external_spte(struct kvm *kvm, gfn_t gfn, u64 old_spte, + int level) +{ + kvm_pfn_t old_pfn = spte_to_pfn(old_spte); + int ret; + + /* + * External (TDX) SPTEs are limited to PG_LEVEL_4K, and external + * PTs are removed in a special order, involving free_external_spt(). + * But remove_external_spte() will be called on non-leaf PTEs via + * __tdp_mmu_zap_root(), so avoid the error the former would return + * in this case. + */ + if (!is_last_spte(old_spte, level)) + return; + + /* Zapping leaf spte is allowed only when write lock is held. */ + lockdep_assert_held_write(&kvm->mmu_lock); + /* Because write lock is held, operation should success. */ + ret = static_call(kvm_x86_remove_external_spte)(kvm, gfn, level, old_pfn); + KVM_BUG_ON(ret, kvm); +} + /** * handle_removed_pt() - handle a page table removed from the TDP structure * @@ -359,14 +413,14 @@ static void handle_removed_pt(struct kvm *kvm, tdp_ptep_t pt, bool shared) /* * Set the SPTE to a nonpresent value that other * threads will not overwrite. If the SPTE was - * already marked as removed then another thread + * already marked as frozen then another thread * handling a page fault could overwrite it, so * set the SPTE until it is set from some other - * value to the removed SPTE value. + * value to the frozen SPTE value. */ for (;;) { - old_spte = kvm_tdp_mmu_write_spte_atomic(sptep, REMOVED_SPTE); - if (!is_removed_spte(old_spte)) + old_spte = kvm_tdp_mmu_write_spte_atomic(sptep, FROZEN_SPTE); + if (!is_frozen_spte(old_spte)) break; cpu_relax(); } @@ -397,11 +451,11 @@ static void handle_removed_pt(struct kvm *kvm, tdp_ptep_t pt, bool shared) * No retry is needed in the atomic update path as the * sole concern is dropping a Dirty bit, i.e. no other * task can zap/remove the SPTE as mmu_lock is held for - * write. Marking the SPTE as a removed SPTE is not + * write. Marking the SPTE as a frozen SPTE is not * strictly necessary for the same reason, but using - * the remove SPTE value keeps the shared/exclusive + * the frozen SPTE value keeps the shared/exclusive * paths consistent and allows the handle_changed_spte() - * call below to hardcode the new value to REMOVED_SPTE. + * call below to hardcode the new value to FROZEN_SPTE. * * Note, even though dropping a Dirty bit is the only * scenario where a non-atomic update could result in a @@ -413,15 +467,85 @@ static void handle_removed_pt(struct kvm *kvm, tdp_ptep_t pt, bool shared) * it here. */ old_spte = kvm_tdp_mmu_write_spte(sptep, old_spte, - REMOVED_SPTE, level); + FROZEN_SPTE, level); } handle_changed_spte(kvm, kvm_mmu_page_as_id(sp), gfn, - old_spte, REMOVED_SPTE, level, shared); + old_spte, FROZEN_SPTE, level, shared); + + if (is_mirror_sp(sp)) { + KVM_BUG_ON(shared, kvm); + remove_external_spte(kvm, gfn, old_spte, level); + } + } + + if (is_mirror_sp(sp) && + WARN_ON(static_call(kvm_x86_free_external_spt)(kvm, base_gfn, sp->role.level, + sp->external_spt))) { + /* + * Failed to free page table page in mirror page table and + * there is nothing to do further. + * Intentionally leak the page to prevent the kernel from + * accessing the encrypted page. + */ + sp->external_spt = NULL; } call_rcu(&sp->rcu_head, tdp_mmu_free_sp_rcu_callback); } +static void *get_external_spt(gfn_t gfn, u64 new_spte, int level) +{ + if (is_shadow_present_pte(new_spte) && !is_last_spte(new_spte, level)) { + struct kvm_mmu_page *sp = spte_to_child_sp(new_spte); + + WARN_ON_ONCE(sp->role.level + 1 != level); + WARN_ON_ONCE(sp->gfn != gfn); + return sp->external_spt; + } + + return NULL; +} + +static int __must_check set_external_spte_present(struct kvm *kvm, tdp_ptep_t sptep, + gfn_t gfn, u64 old_spte, + u64 new_spte, int level) +{ + bool was_present = is_shadow_present_pte(old_spte); + bool is_present = is_shadow_present_pte(new_spte); + bool is_leaf = is_present && is_last_spte(new_spte, level); + kvm_pfn_t new_pfn = spte_to_pfn(new_spte); + int ret = 0; + + KVM_BUG_ON(was_present, kvm); + + lockdep_assert_held(&kvm->mmu_lock); + /* + * We need to lock out other updates to the SPTE until the external + * page table has been modified. Use FROZEN_SPTE similar to + * the zapping case. + */ + if (!try_cmpxchg64(rcu_dereference(sptep), &old_spte, FROZEN_SPTE)) + return -EBUSY; + + /* + * Use different call to either set up middle level + * external page table, or leaf. + */ + if (is_leaf) { + ret = static_call(kvm_x86_set_external_spte)(kvm, gfn, level, new_pfn); + } else { + void *external_spt = get_external_spt(gfn, new_spte, level); + + KVM_BUG_ON(!external_spt, kvm); + ret = static_call(kvm_x86_link_external_spt)(kvm, gfn, level, external_spt); + } + if (ret) + __kvm_tdp_mmu_write_spte(sptep, old_spte); + else + __kvm_tdp_mmu_write_spte(sptep, new_spte); + return ret; +} + /** * handle_changed_spte - handle bookkeeping associated with an SPTE change * @kvm: kvm instance @@ -490,19 +614,19 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn, */ if (!was_present && !is_present) { /* - * If this change does not involve a MMIO SPTE or removed SPTE, + * If this change does not involve a MMIO SPTE or frozen SPTE, * it is unexpected. Log the change, though it should not * impact the guest since both the former and current SPTEs * are nonpresent. */ - if (WARN_ON_ONCE(!is_mmio_spte(old_spte) && - !is_mmio_spte(new_spte) && - !is_removed_spte(new_spte))) + if (WARN_ON_ONCE(!is_mmio_spte(kvm, old_spte) && + !is_mmio_spte(kvm, new_spte) && + !is_frozen_spte(new_spte))) pr_err("Unexpected SPTE change! Nonpresent SPTEs\n" "should not be replaced with another,\n" "different nonpresent SPTE, unless one or both\n" "are MMIO SPTEs, or the new SPTE is\n" - "a temporary removed SPTE.\n" + "a temporary frozen SPTE.\n" "as_id: %d gfn: %llx old_spte: %llx new_spte: %llx level: %d", as_id, gfn, old_spte, new_spte, level); return; @@ -511,10 +635,6 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn, if (is_leaf != was_leaf) kvm_update_page_stats(kvm, level, is_leaf ? 1 : -1); - if (was_leaf && is_dirty_spte(old_spte) && - (!is_present || !is_dirty_spte(new_spte) || pfn_changed)) - kvm_set_pfn_dirty(spte_to_pfn(old_spte)); - /* * Recursively handle child PTs if the change removed a subtree from * the paging structure. Note the WARN on the PFN changing without the @@ -524,10 +644,50 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn, if (was_present && !was_leaf && (is_leaf || !is_present || WARN_ON_ONCE(pfn_changed))) handle_removed_pt(kvm, spte_to_child_pt(old_spte, level), shared); +} + +static inline int __must_check __tdp_mmu_set_spte_atomic(struct kvm *kvm, + struct tdp_iter *iter, + u64 new_spte) +{ + /* + * The caller is responsible for ensuring the old SPTE is not a FROZEN + * SPTE. KVM should never attempt to zap or manipulate a FROZEN SPTE, + * and pre-checking before inserting a new SPTE is advantageous as it + * avoids unnecessary work. + */ + WARN_ON_ONCE(iter->yielded || is_frozen_spte(iter->old_spte)); + + if (is_mirror_sptep(iter->sptep) && !is_frozen_spte(new_spte)) { + int ret; - if (was_leaf && is_accessed_spte(old_spte) && - (!is_present || !is_accessed_spte(new_spte) || pfn_changed)) - kvm_set_pfn_accessed(spte_to_pfn(old_spte)); + /* + * Users of atomic zapping don't operate on mirror roots, + * so don't handle it and bug the VM if it's seen. + */ + if (KVM_BUG_ON(!is_shadow_present_pte(new_spte), kvm)) + return -EBUSY; + + ret = set_external_spte_present(kvm, iter->sptep, iter->gfn, + iter->old_spte, new_spte, iter->level); + if (ret) + return ret; + } else { + u64 *sptep = rcu_dereference(iter->sptep); + + /* + * Note, fast_pf_fix_direct_spte() can also modify TDP MMU SPTEs + * and does not hold the mmu_lock. On failure, i.e. if a + * different logical CPU modified the SPTE, try_cmpxchg64() + * updates iter->old_spte with the current value, so the caller + * operates on fresh data, e.g. if it retries + * tdp_mmu_set_spte_atomic() + */ + if (!try_cmpxchg64(sptep, &iter->old_spte, new_spte)) + return -EBUSY; + } + + return 0; } /* @@ -547,68 +707,24 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn, * no side-effects other than setting iter->old_spte to the last * known value of the spte. */ -static inline int tdp_mmu_set_spte_atomic(struct kvm *kvm, - struct tdp_iter *iter, - u64 new_spte) +static inline int __must_check tdp_mmu_set_spte_atomic(struct kvm *kvm, + struct tdp_iter *iter, + u64 new_spte) { - u64 *sptep = rcu_dereference(iter->sptep); - - /* - * The caller is responsible for ensuring the old SPTE is not a REMOVED - * SPTE. KVM should never attempt to zap or manipulate a REMOVED SPTE, - * and pre-checking before inserting a new SPTE is advantageous as it - * avoids unnecessary work. - */ - WARN_ON_ONCE(iter->yielded || is_removed_spte(iter->old_spte)); + int ret; lockdep_assert_held_read(&kvm->mmu_lock); - /* - * Note, fast_pf_fix_direct_spte() can also modify TDP MMU SPTEs and - * does not hold the mmu_lock. On failure, i.e. if a different logical - * CPU modified the SPTE, try_cmpxchg64() updates iter->old_spte with - * the current value, so the caller operates on fresh data, e.g. if it - * retries tdp_mmu_set_spte_atomic() - */ - if (!try_cmpxchg64(sptep, &iter->old_spte, new_spte)) - return -EBUSY; - - handle_changed_spte(kvm, iter->as_id, iter->gfn, iter->old_spte, - new_spte, iter->level, true); - - return 0; -} - -static inline int tdp_mmu_zap_spte_atomic(struct kvm *kvm, - struct tdp_iter *iter) -{ - int ret; - - /* - * Freeze the SPTE by setting it to a special, - * non-present value. This will stop other threads from - * immediately installing a present entry in its place - * before the TLBs are flushed. - */ - ret = tdp_mmu_set_spte_atomic(kvm, iter, REMOVED_SPTE); + ret = __tdp_mmu_set_spte_atomic(kvm, iter, new_spte); if (ret) return ret; - kvm_flush_remote_tlbs_gfn(kvm, iter->gfn, iter->level); - - /* - * No other thread can overwrite the removed SPTE as they must either - * wait on the MMU lock or use tdp_mmu_set_spte_atomic() which will not - * overwrite the special removed SPTE value. No bookkeeping is needed - * here since the SPTE is going from non-present to non-present. Use - * the raw write helper to avoid an unnecessary check on volatile bits. - */ - __kvm_tdp_mmu_write_spte(iter->sptep, 0); + handle_changed_spte(kvm, iter->as_id, iter->gfn, iter->old_spte, + new_spte, iter->level, true); return 0; } - /* * tdp_mmu_set_spte - Set a TDP MMU SPTE and handle the associated bookkeeping * @kvm: KVM instance @@ -629,16 +745,26 @@ static u64 tdp_mmu_set_spte(struct kvm *kvm, int as_id, tdp_ptep_t sptep, /* * No thread should be using this function to set SPTEs to or from the - * temporary removed SPTE value. + * temporary frozen SPTE value. * If operating under the MMU lock in read mode, tdp_mmu_set_spte_atomic * should be used. If operating under the MMU lock in write mode, the - * use of the removed SPTE should not be necessary. + * use of the frozen SPTE should not be necessary. */ - WARN_ON_ONCE(is_removed_spte(old_spte) || is_removed_spte(new_spte)); + WARN_ON_ONCE(is_frozen_spte(old_spte) || is_frozen_spte(new_spte)); old_spte = kvm_tdp_mmu_write_spte(sptep, old_spte, new_spte, level); handle_changed_spte(kvm, as_id, gfn, old_spte, new_spte, level, false); + + /* + * Users that do non-atomic setting of PTEs don't operate on mirror + * roots, so don't handle it and bug the VM if it's seen. + */ + if (is_mirror_sptep(sptep)) { + KVM_BUG_ON(is_shadow_present_pte(new_spte), kvm); + remove_external_spte(kvm, gfn, old_spte, level); + } + return old_spte; } @@ -651,18 +777,25 @@ static inline void tdp_mmu_iter_set_spte(struct kvm *kvm, struct tdp_iter *iter, iter->gfn, iter->level); } -#define tdp_root_for_each_pte(_iter, _root, _start, _end) \ - for_each_tdp_pte(_iter, _root, _start, _end) +#define tdp_root_for_each_pte(_iter, _kvm, _root, _start, _end) \ + for_each_tdp_pte(_iter, _kvm, _root, _start, _end) -#define tdp_root_for_each_leaf_pte(_iter, _root, _start, _end) \ - tdp_root_for_each_pte(_iter, _root, _start, _end) \ +#define tdp_root_for_each_leaf_pte(_iter, _kvm, _root, _start, _end) \ + tdp_root_for_each_pte(_iter, _kvm, _root, _start, _end) \ if (!is_shadow_present_pte(_iter.old_spte) || \ !is_last_spte(_iter.old_spte, _iter.level)) \ continue; \ else -#define tdp_mmu_for_each_pte(_iter, _mmu, _start, _end) \ - for_each_tdp_pte(_iter, root_to_sp(_mmu->root.hpa), _start, _end) +static inline bool __must_check tdp_mmu_iter_need_resched(struct kvm *kvm, + struct tdp_iter *iter) +{ + if (!need_resched() && !rwlock_needbreak(&kvm->mmu_lock)) + return false; + + /* Ensure forward progress has been made before yielding. */ + return iter->next_last_level_gfn != iter->yielded_gfn; +} /* * Yield if the MMU lock is contended or this thread needs to return control @@ -682,31 +815,27 @@ static inline bool __must_check tdp_mmu_iter_cond_resched(struct kvm *kvm, struct tdp_iter *iter, bool flush, bool shared) { - WARN_ON_ONCE(iter->yielded); + KVM_MMU_WARN_ON(iter->yielded); - /* Ensure forward progress has been made before yielding. */ - if (iter->next_last_level_gfn == iter->yielded_gfn) + if (!tdp_mmu_iter_need_resched(kvm, iter)) return false; - if (need_resched() || rwlock_needbreak(&kvm->mmu_lock)) { - if (flush) - kvm_flush_remote_tlbs(kvm); - - rcu_read_unlock(); + if (flush) + kvm_flush_remote_tlbs(kvm); - if (shared) - cond_resched_rwlock_read(&kvm->mmu_lock); - else - cond_resched_rwlock_write(&kvm->mmu_lock); + rcu_read_unlock(); - rcu_read_lock(); + if (shared) + cond_resched_rwlock_read(&kvm->mmu_lock); + else + cond_resched_rwlock_write(&kvm->mmu_lock); - WARN_ON_ONCE(iter->gfn > iter->next_last_level_gfn); + rcu_read_lock(); - iter->yielded = true; - } + WARN_ON_ONCE(iter->gfn > iter->next_last_level_gfn); - return iter->yielded; + iter->yielded = true; + return true; } static inline gfn_t tdp_mmu_max_gfn_exclusive(void) @@ -725,10 +854,7 @@ static void __tdp_mmu_zap_root(struct kvm *kvm, struct kvm_mmu_page *root, { struct tdp_iter iter; - gfn_t end = tdp_mmu_max_gfn_exclusive(); - gfn_t start = 0; - - for_each_tdp_pte_min_level(iter, root, zap_level, start, end) { + for_each_tdp_pte_min_level_all(iter, root, zap_level) { retry: if (tdp_mmu_iter_cond_resched(kvm, &iter, false, shared)) continue; @@ -740,8 +866,8 @@ retry: continue; if (!shared) - tdp_mmu_iter_set_spte(kvm, &iter, 0); - else if (tdp_mmu_set_spte_atomic(kvm, &iter, 0)) + tdp_mmu_iter_set_spte(kvm, &iter, SHADOW_NONPRESENT_VALUE); + else if (tdp_mmu_set_spte_atomic(kvm, &iter, SHADOW_NONPRESENT_VALUE)) goto retry; } } @@ -808,8 +934,8 @@ bool kvm_tdp_mmu_zap_sp(struct kvm *kvm, struct kvm_mmu_page *sp) if (WARN_ON_ONCE(!is_shadow_present_pte(old_spte))) return false; - tdp_mmu_set_spte(kvm, kvm_mmu_page_as_id(sp), sp->ptep, old_spte, 0, - sp->gfn, sp->role.level + 1); + tdp_mmu_set_spte(kvm, kvm_mmu_page_as_id(sp), sp->ptep, old_spte, + SHADOW_NONPRESENT_VALUE, sp->gfn, sp->role.level + 1); return true; } @@ -832,7 +958,7 @@ static bool tdp_mmu_zap_leafs(struct kvm *kvm, struct kvm_mmu_page *root, rcu_read_lock(); - for_each_tdp_pte_min_level(iter, root, PG_LEVEL_4K, start, end) { + for_each_tdp_pte_min_level(iter, kvm, root, PG_LEVEL_4K, start, end) { if (can_yield && tdp_mmu_iter_cond_resched(kvm, &iter, flush, false)) { flush = false; @@ -843,7 +969,7 @@ static bool tdp_mmu_zap_leafs(struct kvm *kvm, struct kvm_mmu_page *root, !is_last_spte(iter.old_spte, iter.level)) continue; - tdp_mmu_iter_set_spte(kvm, &iter, 0); + tdp_mmu_iter_set_spte(kvm, &iter, SHADOW_NONPRESENT_VALUE); /* * Zappings SPTEs in invalid roots doesn't require a TLB flush, @@ -883,19 +1009,21 @@ void kvm_tdp_mmu_zap_all(struct kvm *kvm) struct kvm_mmu_page *root; /* - * Zap all roots, including invalid roots, as all SPTEs must be dropped - * before returning to the caller. Zap directly even if the root is - * also being zapped by a worker. Walking zapped top-level SPTEs isn't - * all that expensive and mmu_lock is already held, which means the - * worker has yielded, i.e. flushing the work instead of zapping here - * isn't guaranteed to be any faster. + * Zap all direct roots, including invalid direct roots, as all direct + * SPTEs must be dropped before returning to the caller. For TDX, mirror + * roots don't need handling in response to the mmu notifier (the caller). + * + * Zap directly even if the root is also being zapped by a concurrent + * "fast zap". Walking zapped top-level SPTEs isn't all that expensive + * and mmu_lock is already held, which means the other thread has yielded. * * A TLB flush is unnecessary, KVM zaps everything if and only the VM * is being destroyed or the userspace VMM has exited. In both cases, * KVM_RUN is unreachable, i.e. no vCPUs will ever service the request. */ lockdep_assert_held_write(&kvm->mmu_lock); - for_each_tdp_mmu_root_yield_safe(kvm, root) + __for_each_tdp_mmu_root_yield_safe(kvm, root, -1, + KVM_DIRECT_ROOTS | KVM_INVALID_ROOTS) tdp_mmu_zap_root(kvm, root, false); } @@ -903,11 +1031,14 @@ void kvm_tdp_mmu_zap_all(struct kvm *kvm) * Zap all invalidated roots to ensure all SPTEs are dropped before the "fast * zap" completes. */ -void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm) +void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm, bool shared) { struct kvm_mmu_page *root; - read_lock(&kvm->mmu_lock); + if (shared) + read_lock(&kvm->mmu_lock); + else + write_lock(&kvm->mmu_lock); for_each_tdp_mmu_root_yield_safe(kvm, root) { if (!root->tdp_mmu_scheduled_root_to_zap) @@ -925,7 +1056,7 @@ void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm) * that may be zapped, as such entries are associated with the * ASID on both VMX and SVM. */ - tdp_mmu_zap_root(kvm, root, true); + tdp_mmu_zap_root(kvm, root, shared); /* * The referenced needs to be put *after* zapping the root, as @@ -935,7 +1066,10 @@ void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm) kvm_tdp_mmu_put_root(kvm, root); } - read_unlock(&kvm->mmu_lock); + if (shared) + read_unlock(&kvm->mmu_lock); + else + write_unlock(&kvm->mmu_lock); } /* @@ -948,11 +1082,19 @@ void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm) * Note, kvm_tdp_mmu_zap_invalidated_roots() is gifted the TDP MMU's reference. * See kvm_tdp_mmu_alloc_root(). */ -void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm) +void kvm_tdp_mmu_invalidate_roots(struct kvm *kvm, + enum kvm_tdp_mmu_root_types root_types) { struct kvm_mmu_page *root; /* + * Invalidating invalid roots doesn't make sense, prevent developers from + * having to think about it. + */ + if (WARN_ON_ONCE(root_types & KVM_INVALID_ROOTS)) + root_types &= ~KVM_INVALID_ROOTS; + + /* * mmu_lock must be held for write to ensure that a root doesn't become * invalid while there are active readers (invalidating a root while * there are active readers may or may not be problematic in practice, @@ -973,6 +1115,9 @@ void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm) * or get/put references to roots. */ list_for_each_entry(root, &kvm->arch.tdp_mmu_roots, link) { + if (!tdp_mmu_root_match(root, root_types)) + continue; + /* * Note, invalid roots can outlive a memslot update! Invalid * roots must be *zapped* before the memslot update completes, @@ -1002,19 +1147,28 @@ static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu, if (WARN_ON_ONCE(sp->role.level != fault->goal_level)) return RET_PF_RETRY; + if (fault->prefetch && is_shadow_present_pte(iter->old_spte)) + return RET_PF_SPURIOUS; + + if (is_shadow_present_pte(iter->old_spte) && + is_access_allowed(fault, iter->old_spte) && + is_last_spte(iter->old_spte, iter->level)) + return RET_PF_SPURIOUS; + if (unlikely(!fault->slot)) new_spte = make_mmio_spte(vcpu, iter->gfn, ACC_ALL); else wrprot = make_spte(vcpu, sp, fault->slot, ACC_ALL, iter->gfn, - fault->pfn, iter->old_spte, fault->prefetch, true, - fault->map_writable, &new_spte); + fault->pfn, iter->old_spte, fault->prefetch, + false, fault->map_writable, &new_spte); if (new_spte == iter->old_spte) ret = RET_PF_SPURIOUS; else if (tdp_mmu_set_spte_atomic(vcpu->kvm, iter, new_spte)) return RET_PF_RETRY; else if (is_shadow_present_pte(iter->old_spte) && - !is_last_spte(iter->old_spte, iter->level)) + (!is_last_spte(iter->old_spte, iter->level) || + WARN_ON_ONCE(leaf_spte_change_needs_tlb_flush(iter->old_spte, new_spte)))) kvm_flush_remote_tlbs_gfn(vcpu->kvm, iter->gfn, iter->level); /* @@ -1022,13 +1176,11 @@ static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu, * protected, emulation is needed. If the emulation was skipped, * the vCPU would have the same fault again. */ - if (wrprot) { - if (fault->write) - ret = RET_PF_EMULATE; - } + if (wrprot && fault->write) + ret = RET_PF_WRITE_PROTECTED; /* If a MMIO SPTE is installed, the MMIO will need to be emulated. */ - if (unlikely(is_mmio_spte(new_spte))) { + if (unlikely(is_mmio_spte(vcpu->kvm, new_spte))) { vcpu->stat.pf_mmio_spte_created++; trace_mark_mmio_spte(rcu_dereference(iter->sptep), iter->gfn, new_spte); @@ -1056,7 +1208,7 @@ static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu, static int tdp_mmu_link_sp(struct kvm *kvm, struct tdp_iter *iter, struct kvm_mmu_page *sp, bool shared) { - u64 spte = make_nonleaf_spte(sp->spt, !kvm_ad_enabled()); + u64 spte = make_nonleaf_spte(sp->spt, !kvm_ad_enabled); int ret = 0; if (shared) { @@ -1081,7 +1233,7 @@ static int tdp_mmu_split_huge_page(struct kvm *kvm, struct tdp_iter *iter, */ int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) { - struct kvm_mmu *mmu = vcpu->arch.mmu; + struct kvm_mmu_page *root = tdp_mmu_get_root_for_fault(vcpu, fault); struct kvm *kvm = vcpu->kvm; struct tdp_iter iter; struct kvm_mmu_page *sp; @@ -1093,7 +1245,7 @@ int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) rcu_read_lock(); - tdp_mmu_for_each_pte(iter, mmu, fault->gfn, fault->gfn + 1) { + for_each_tdp_pte(iter, kvm, root, fault->gfn, fault->gfn + 1) { int r; if (fault->nx_huge_page_workaround_enabled) @@ -1103,7 +1255,7 @@ int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) * If SPTE has been frozen by another thread, just give up and * retry, avoiding unnecessary page table allocation and free. */ - if (is_removed_spte(iter.old_spte)) + if (is_frozen_spte(iter.old_spte)) goto retry; if (iter.level == fault->goal_level) @@ -1120,13 +1272,18 @@ int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault) */ sp = tdp_mmu_alloc_sp(vcpu); tdp_mmu_init_child_sp(sp, &iter); + if (is_mirror_sp(sp)) + kvm_mmu_alloc_external_spt(vcpu, sp); sp->nx_huge_page_disallowed = fault->huge_page_disallowed; - if (is_shadow_present_pte(iter.old_spte)) + if (is_shadow_present_pte(iter.old_spte)) { + /* Don't support large page for mirrored roots (TDX) */ + KVM_BUG_ON(is_mirror_sptep(iter.sptep), vcpu->kvm); r = tdp_mmu_split_huge_page(kvm, &iter, sp, true); - else + } else { r = tdp_mmu_link_sp(kvm, &iter, sp, true); + } /* * Force the guest to retry if installing an upper level SPTE @@ -1161,45 +1318,22 @@ retry: return ret; } +/* Used by mmu notifier via kvm_unmap_gfn_range() */ bool kvm_tdp_mmu_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range, bool flush) { + enum kvm_tdp_mmu_root_types types; struct kvm_mmu_page *root; - __for_each_tdp_mmu_root_yield_safe(kvm, root, range->slot->as_id, false) + types = kvm_gfn_range_filter_to_root_types(kvm, range->attr_filter) | KVM_INVALID_ROOTS; + + __for_each_tdp_mmu_root_yield_safe(kvm, root, range->slot->as_id, types) flush = tdp_mmu_zap_leafs(kvm, root, range->start, range->end, range->may_block, flush); return flush; } -typedef bool (*tdp_handler_t)(struct kvm *kvm, struct tdp_iter *iter, - struct kvm_gfn_range *range); - -static __always_inline bool kvm_tdp_mmu_handle_gfn(struct kvm *kvm, - struct kvm_gfn_range *range, - tdp_handler_t handler) -{ - struct kvm_mmu_page *root; - struct tdp_iter iter; - bool ret = false; - - /* - * Don't support rescheduling, none of the MMU notifiers that funnel - * into this helper allow blocking; it'd be dead, wasteful code. - */ - for_each_tdp_mmu_root(kvm, root, range->slot->as_id) { - rcu_read_lock(); - - tdp_root_for_each_leaf_pte(iter, root, range->start, range->end) - ret |= handler(kvm, &iter, range); - - rcu_read_unlock(); - } - - return ret; -} - /* * Mark the SPTEs range of GFNs [start, end) unaccessed and return non-zero * if any of the GFNs in the range have been accessed. @@ -1208,100 +1342,72 @@ static __always_inline bool kvm_tdp_mmu_handle_gfn(struct kvm *kvm, * from the clear_young() or clear_flush_young() notifier, which uses the * return value to determine if the page has been accessed. */ -static bool age_gfn_range(struct kvm *kvm, struct tdp_iter *iter, - struct kvm_gfn_range *range) +static void kvm_tdp_mmu_age_spte(struct kvm *kvm, struct tdp_iter *iter) { u64 new_spte; - /* If we have a non-accessed entry we don't need to change the pte. */ - if (!is_accessed_spte(iter->old_spte)) - return false; - if (spte_ad_enabled(iter->old_spte)) { - iter->old_spte = tdp_mmu_clear_spte_bits(iter->sptep, - iter->old_spte, - shadow_accessed_mask, - iter->level); + iter->old_spte = tdp_mmu_clear_spte_bits_atomic(iter->sptep, + shadow_accessed_mask); new_spte = iter->old_spte & ~shadow_accessed_mask; } else { + new_spte = mark_spte_for_access_track(iter->old_spte); /* - * Capture the dirty status of the page, so that it doesn't get - * lost when the SPTE is marked for access tracking. + * It is safe for the following cmpxchg to fail. Leave the + * Accessed bit set, as the spte is most likely young anyway. */ - if (is_writable_pte(iter->old_spte)) - kvm_set_pfn_dirty(spte_to_pfn(iter->old_spte)); - - new_spte = mark_spte_for_access_track(iter->old_spte); - iter->old_spte = kvm_tdp_mmu_write_spte(iter->sptep, - iter->old_spte, new_spte, - iter->level); + if (__tdp_mmu_set_spte_atomic(kvm, iter, new_spte)) + return; } trace_kvm_tdp_mmu_spte_changed(iter->as_id, iter->gfn, iter->level, iter->old_spte, new_spte); - return true; } -bool kvm_tdp_mmu_age_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) +static bool __kvm_tdp_mmu_age_gfn_range(struct kvm *kvm, + struct kvm_gfn_range *range, + bool test_only) { - return kvm_tdp_mmu_handle_gfn(kvm, range, age_gfn_range); -} - -static bool test_age_gfn(struct kvm *kvm, struct tdp_iter *iter, - struct kvm_gfn_range *range) -{ - return is_accessed_spte(iter->old_spte); -} - -bool kvm_tdp_mmu_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range) -{ - return kvm_tdp_mmu_handle_gfn(kvm, range, test_age_gfn); -} - -static bool set_spte_gfn(struct kvm *kvm, struct tdp_iter *iter, - struct kvm_gfn_range *range) -{ - u64 new_spte; - - /* Huge pages aren't expected to be modified without first being zapped. */ - WARN_ON_ONCE(pte_huge(range->arg.pte) || range->start + 1 != range->end); + enum kvm_tdp_mmu_root_types types; + struct kvm_mmu_page *root; + struct tdp_iter iter; + bool ret = false; - if (iter->level != PG_LEVEL_4K || - !is_shadow_present_pte(iter->old_spte)) - return false; + types = kvm_gfn_range_filter_to_root_types(kvm, range->attr_filter); /* - * Note, when changing a read-only SPTE, it's not strictly necessary to - * zero the SPTE before setting the new PFN, but doing so preserves the - * invariant that the PFN of a present * leaf SPTE can never change. - * See handle_changed_spte(). + * Don't support rescheduling, none of the MMU notifiers that funnel + * into this helper allow blocking; it'd be dead, wasteful code. Note, + * this helper must NOT be used to unmap GFNs, as it processes only + * valid roots! */ - tdp_mmu_iter_set_spte(kvm, iter, 0); + WARN_ON(types & ~KVM_VALID_ROOTS); + + guard(rcu)(); + for_each_tdp_mmu_root_rcu(kvm, root, range->slot->as_id, types) { + tdp_root_for_each_leaf_pte(iter, kvm, root, range->start, range->end) { + if (!is_accessed_spte(iter.old_spte)) + continue; - if (!pte_write(range->arg.pte)) { - new_spte = kvm_mmu_changed_pte_notifier_make_spte(iter->old_spte, - pte_pfn(range->arg.pte)); + if (test_only) + return true; - tdp_mmu_iter_set_spte(kvm, iter, new_spte); + ret = true; + kvm_tdp_mmu_age_spte(kvm, &iter); + } } - return true; + return ret; } -/* - * Handle the changed_pte MMU notifier for the TDP MMU. - * data is a pointer to the new pte_t mapping the HVA specified by the MMU - * notifier. - * Returns non-zero if a flush is needed before releasing the MMU lock. - */ -bool kvm_tdp_mmu_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range) +bool kvm_tdp_mmu_age_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) { - /* - * No need to handle the remote TLB flush under RCU protection, the - * target SPTE _must_ be a leaf SPTE, i.e. cannot result in freeing a - * shadow page. See the WARN on pfn_changed in handle_changed_spte(). - */ - return kvm_tdp_mmu_handle_gfn(kvm, range, set_spte_gfn); + return __kvm_tdp_mmu_age_gfn_range(kvm, range, false); +} + +bool kvm_tdp_mmu_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range) +{ + return __kvm_tdp_mmu_age_gfn_range(kvm, range, true); } /* @@ -1320,7 +1426,7 @@ static bool wrprot_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root, BUG_ON(min_level > KVM_MAX_HUGEPAGE_LEVEL); - for_each_tdp_pte_min_level(iter, root, min_level, start, end) { + for_each_tdp_pte_min_level(iter, kvm, root, min_level, start, end) { retry: if (tdp_mmu_iter_cond_resched(kvm, &iter, false, true)) continue; @@ -1362,17 +1468,15 @@ bool kvm_tdp_mmu_wrprot_slot(struct kvm *kvm, return spte_set; } -static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp) +static struct kvm_mmu_page *tdp_mmu_alloc_sp_for_split(void) { struct kvm_mmu_page *sp; - gfp |= __GFP_ZERO; - - sp = kmem_cache_alloc(mmu_page_header_cache, gfp); + sp = kmem_cache_zalloc(mmu_page_header_cache, GFP_KERNEL_ACCOUNT); if (!sp) return NULL; - sp->spt = (void *)__get_free_page(gfp); + sp->spt = (void *)get_zeroed_page(GFP_KERNEL_ACCOUNT); if (!sp->spt) { kmem_cache_free(mmu_page_header_cache, sp); return NULL; @@ -1381,47 +1485,6 @@ static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp) return sp; } -static struct kvm_mmu_page *tdp_mmu_alloc_sp_for_split(struct kvm *kvm, - struct tdp_iter *iter, - bool shared) -{ - struct kvm_mmu_page *sp; - - kvm_lockdep_assert_mmu_lock_held(kvm, shared); - - /* - * Since we are allocating while under the MMU lock we have to be - * careful about GFP flags. Use GFP_NOWAIT to avoid blocking on direct - * reclaim and to avoid making any filesystem callbacks (which can end - * up invoking KVM MMU notifiers, resulting in a deadlock). - * - * If this allocation fails we drop the lock and retry with reclaim - * allowed. - */ - sp = __tdp_mmu_alloc_sp_for_split(GFP_NOWAIT | __GFP_ACCOUNT); - if (sp) - return sp; - - rcu_read_unlock(); - - if (shared) - read_unlock(&kvm->mmu_lock); - else - write_unlock(&kvm->mmu_lock); - - iter->yielded = true; - sp = __tdp_mmu_alloc_sp_for_split(GFP_KERNEL_ACCOUNT); - - if (shared) - read_lock(&kvm->mmu_lock); - else - write_lock(&kvm->mmu_lock); - - rcu_read_lock(); - - return sp; -} - /* Note, the caller is responsible for initializing @sp. */ static int tdp_mmu_split_huge_page(struct kvm *kvm, struct tdp_iter *iter, struct kvm_mmu_page *sp, bool shared) @@ -1435,7 +1498,7 @@ static int tdp_mmu_split_huge_page(struct kvm *kvm, struct tdp_iter *iter, * not been linked in yet and thus is not reachable from any other CPU. */ for (i = 0; i < SPTE_ENT_PER_PAGE; i++) - sp->spt[i] = make_huge_page_split_spte(kvm, huge_spte, sp->role, i); + sp->spt[i] = make_small_spte(kvm, huge_spte, sp->role, i); /* * Replace the huge spte with a pointer to the populated lower level @@ -1468,7 +1531,6 @@ static int tdp_mmu_split_huge_pages_root(struct kvm *kvm, { struct kvm_mmu_page *sp = NULL; struct tdp_iter iter; - int ret = 0; rcu_read_lock(); @@ -1483,7 +1545,7 @@ static int tdp_mmu_split_huge_pages_root(struct kvm *kvm, * level above the target level (e.g. splitting a 1GB to 512 2MB pages, * and then splitting each of those to 512 4KB pages). */ - for_each_tdp_pte_min_level(iter, root, target_level + 1, start, end) { + for_each_tdp_pte_min_level(iter, kvm, root, target_level + 1, start, end) { retry: if (tdp_mmu_iter_cond_resched(kvm, &iter, false, shared)) continue; @@ -1492,17 +1554,31 @@ retry: continue; if (!sp) { - sp = tdp_mmu_alloc_sp_for_split(kvm, &iter, shared); + rcu_read_unlock(); + + if (shared) + read_unlock(&kvm->mmu_lock); + else + write_unlock(&kvm->mmu_lock); + + sp = tdp_mmu_alloc_sp_for_split(); + + if (shared) + read_lock(&kvm->mmu_lock); + else + write_lock(&kvm->mmu_lock); + if (!sp) { - ret = -ENOMEM; trace_kvm_mmu_split_huge_page(iter.gfn, iter.old_spte, - iter.level, ret); - break; + iter.level, -ENOMEM); + return -ENOMEM; } - if (iter.yielded) - continue; + rcu_read_lock(); + + iter.yielded = true; + continue; } tdp_mmu_init_child_sp(sp, &iter); @@ -1523,7 +1599,7 @@ retry: if (sp) tdp_mmu_free_sp(sp); - return ret; + return 0; } @@ -1548,23 +1624,26 @@ void kvm_tdp_mmu_try_split_huge_pages(struct kvm *kvm, } } -/* - * Clear the dirty status of all the SPTEs mapping GFNs in the memslot. If - * AD bits are enabled, this will involve clearing the dirty bit on each SPTE. - * If AD bits are not enabled, this will require clearing the writable bit on - * each SPTE. Returns true if an SPTE has been changed and the TLBs need to - * be flushed. - */ -static bool clear_dirty_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root, - gfn_t start, gfn_t end) +static bool tdp_mmu_need_write_protect(struct kvm_mmu_page *sp) +{ + /* + * All TDP MMU shadow pages share the same role as their root, aside + * from level, so it is valid to key off any shadow page to determine if + * write protection is needed for an entire tree. + */ + return kvm_mmu_page_ad_need_write_protect(sp) || !kvm_ad_enabled; +} + +static void clear_dirty_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root, + gfn_t start, gfn_t end) { - u64 dbit = kvm_ad_enabled() ? shadow_dirty_mask : PT_WRITABLE_MASK; + const u64 dbit = tdp_mmu_need_write_protect(root) ? PT_WRITABLE_MASK : + shadow_dirty_mask; struct tdp_iter iter; - bool spte_set = false; rcu_read_lock(); - tdp_root_for_each_pte(iter, root, start, end) { + tdp_root_for_each_pte(iter, kvm, root, start, end) { retry: if (!is_shadow_present_pte(iter.old_spte) || !is_last_spte(iter.old_spte, iter.level)) @@ -1573,7 +1652,7 @@ retry: if (tdp_mmu_iter_cond_resched(kvm, &iter, false, true)) continue; - KVM_MMU_WARN_ON(kvm_ad_enabled() && + KVM_MMU_WARN_ON(dbit == shadow_dirty_mask && spte_ad_need_write_protect(iter.old_spte)); if (!(iter.old_spte & dbit)) @@ -1581,59 +1660,43 @@ retry: if (tdp_mmu_set_spte_atomic(kvm, &iter, iter.old_spte & ~dbit)) goto retry; - - spte_set = true; } rcu_read_unlock(); - return spte_set; } /* - * Clear the dirty status of all the SPTEs mapping GFNs in the memslot. If - * AD bits are enabled, this will involve clearing the dirty bit on each SPTE. - * If AD bits are not enabled, this will require clearing the writable bit on - * each SPTE. Returns true if an SPTE has been changed and the TLBs need to - * be flushed. + * Clear the dirty status (D-bit or W-bit) of all the SPTEs mapping GFNs in the + * memslot. */ -bool kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm, +void kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm, const struct kvm_memory_slot *slot) { struct kvm_mmu_page *root; - bool spte_set = false; lockdep_assert_held_read(&kvm->mmu_lock); for_each_valid_tdp_mmu_root_yield_safe(kvm, root, slot->as_id) - spte_set |= clear_dirty_gfn_range(kvm, root, slot->base_gfn, - slot->base_gfn + slot->npages); - - return spte_set; + clear_dirty_gfn_range(kvm, root, slot->base_gfn, + slot->base_gfn + slot->npages); } -/* - * Clears the dirty status of all the 4k SPTEs mapping GFNs for which a bit is - * set in mask, starting at gfn. The given memslot is expected to contain all - * the GFNs represented by set bits in the mask. If AD bits are enabled, - * clearing the dirty status will involve clearing the dirty bit on each SPTE - * or, if AD bits are not enabled, clearing the writable bit on each SPTE. - */ static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root, gfn_t gfn, unsigned long mask, bool wrprot) { - u64 dbit = (wrprot || !kvm_ad_enabled()) ? PT_WRITABLE_MASK : - shadow_dirty_mask; + const u64 dbit = (wrprot || tdp_mmu_need_write_protect(root)) ? PT_WRITABLE_MASK : + shadow_dirty_mask; struct tdp_iter iter; lockdep_assert_held_write(&kvm->mmu_lock); rcu_read_lock(); - tdp_root_for_each_leaf_pte(iter, root, gfn + __ffs(mask), + tdp_root_for_each_leaf_pte(iter, kvm, root, gfn + __ffs(mask), gfn + BITS_PER_LONG) { if (!mask) break; - KVM_MMU_WARN_ON(kvm_ad_enabled() && + KVM_MMU_WARN_ON(dbit == shadow_dirty_mask && spte_ad_need_write_protect(iter.old_spte)); if (iter.level > PG_LEVEL_4K || @@ -1652,18 +1715,15 @@ static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root, trace_kvm_tdp_mmu_spte_changed(iter.as_id, iter.gfn, iter.level, iter.old_spte, iter.old_spte & ~dbit); - kvm_set_pfn_dirty(spte_to_pfn(iter.old_spte)); } rcu_read_unlock(); } /* - * Clears the dirty status of all the 4k SPTEs mapping GFNs for which a bit is - * set in mask, starting at gfn. The given memslot is expected to contain all - * the GFNs represented by set bits in the mask. If AD bits are enabled, - * clearing the dirty status will involve clearing the dirty bit on each SPTE - * or, if AD bits are not enabled, clearing the writable bit on each SPTE. + * Clear the dirty status (D-bit or W-bit) of all the 4k SPTEs mapping GFNs for + * which a bit is set in mask, starting at gfn. The given memslot is expected to + * contain all the GFNs represented by set bits in the mask. */ void kvm_tdp_mmu_clear_dirty_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, @@ -1676,21 +1736,55 @@ void kvm_tdp_mmu_clear_dirty_pt_masked(struct kvm *kvm, clear_dirty_pt_masked(kvm, root, gfn, mask, wrprot); } -static void zap_collapsible_spte_range(struct kvm *kvm, - struct kvm_mmu_page *root, - const struct kvm_memory_slot *slot) +static int tdp_mmu_make_huge_spte(struct kvm *kvm, + struct tdp_iter *parent, + u64 *huge_spte) +{ + struct kvm_mmu_page *root = spte_to_child_sp(parent->old_spte); + gfn_t start = parent->gfn; + gfn_t end = start + KVM_PAGES_PER_HPAGE(parent->level); + struct tdp_iter iter; + + tdp_root_for_each_leaf_pte(iter, kvm, root, start, end) { + /* + * Use the parent iterator when checking for forward progress so + * that KVM doesn't get stuck continuously trying to yield (i.e. + * returning -EAGAIN here and then failing the forward progress + * check in the caller ad nauseam). + */ + if (tdp_mmu_iter_need_resched(kvm, parent)) + return -EAGAIN; + + *huge_spte = make_huge_spte(kvm, iter.old_spte, parent->level); + return 0; + } + + return -ENOENT; +} + +static void recover_huge_pages_range(struct kvm *kvm, + struct kvm_mmu_page *root, + const struct kvm_memory_slot *slot) { gfn_t start = slot->base_gfn; gfn_t end = start + slot->npages; struct tdp_iter iter; int max_mapping_level; + bool flush = false; + u64 huge_spte; + int r; + + if (WARN_ON_ONCE(kvm_slot_dirty_track_enabled(slot))) + return; rcu_read_lock(); - for_each_tdp_pte_min_level(iter, root, PG_LEVEL_2M, start, end) { + for_each_tdp_pte_min_level(iter, kvm, root, PG_LEVEL_2M, start, end) { retry: - if (tdp_mmu_iter_cond_resched(kvm, &iter, false, true)) + if (tdp_mmu_iter_cond_resched(kvm, &iter, flush, true)) { + flush = false; continue; + } if (iter.level > KVM_MAX_HUGEPAGE_LEVEL || !is_shadow_present_pte(iter.old_spte)) @@ -1714,31 +1808,40 @@ retry: if (iter.gfn < start || iter.gfn >= end) continue; - max_mapping_level = kvm_mmu_max_mapping_level(kvm, slot, - iter.gfn, PG_LEVEL_NUM); + max_mapping_level = kvm_mmu_max_mapping_level(kvm, slot, iter.gfn); if (max_mapping_level < iter.level) continue; - /* Note, a successful atomic zap also does a remote TLB flush. */ - if (tdp_mmu_zap_spte_atomic(kvm, &iter)) + r = tdp_mmu_make_huge_spte(kvm, &iter, &huge_spte); + if (r == -EAGAIN) goto retry; + else if (r) + continue; + + if (tdp_mmu_set_spte_atomic(kvm, &iter, huge_spte)) + goto retry; + + flush = true; } + if (flush) + kvm_flush_remote_tlbs_memslot(kvm, slot); + rcu_read_unlock(); } /* - * Zap non-leaf SPTEs (and free their associated page tables) which could - * be replaced by huge pages, for GFNs within the slot. + * Recover huge page mappings within the slot by replacing non-leaf SPTEs with + * huge SPTEs where possible. */ -void kvm_tdp_mmu_zap_collapsible_sptes(struct kvm *kvm, - const struct kvm_memory_slot *slot) +void kvm_tdp_mmu_recover_huge_pages(struct kvm *kvm, + const struct kvm_memory_slot *slot) { struct kvm_mmu_page *root; lockdep_assert_held_read(&kvm->mmu_lock); for_each_valid_tdp_mmu_root_yield_safe(kvm, root, slot->as_id) - zap_collapsible_spte_range(kvm, root, slot); + recover_huge_pages_range(kvm, root, slot); } /* @@ -1757,7 +1860,7 @@ static bool write_protect_gfn(struct kvm *kvm, struct kvm_mmu_page *root, rcu_read_lock(); - for_each_tdp_pte_min_level(iter, root, min_level, gfn, gfn + 1) { + for_each_tdp_pte_min_level(iter, kvm, root, min_level, gfn, gfn + 1) { if (!is_shadow_present_pte(iter.old_spte) || !is_last_spte(iter.old_spte, iter.level)) continue; @@ -1805,14 +1908,14 @@ bool kvm_tdp_mmu_write_protect_gfn(struct kvm *kvm, int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes, int *root_level) { + struct kvm_mmu_page *root = root_to_sp(vcpu->arch.mmu->root.hpa); struct tdp_iter iter; - struct kvm_mmu *mmu = vcpu->arch.mmu; gfn_t gfn = addr >> PAGE_SHIFT; int leaf = -1; *root_level = vcpu->arch.mmu->root_role.level; - tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) { + for_each_tdp_pte(iter, vcpu->kvm, root, gfn, gfn + 1) { leaf = iter.level; sptes[leaf] = iter.old_spte; } @@ -1831,15 +1934,15 @@ int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes, * * WARNING: This function is only intended to be called during fast_page_fault. */ -u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, u64 addr, +u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, gfn_t gfn, u64 *spte) { + /* Fast pf is not supported for mirrored roots */ + struct kvm_mmu_page *root = tdp_mmu_get_root(vcpu, KVM_DIRECT_ROOTS); struct tdp_iter iter; - struct kvm_mmu *mmu = vcpu->arch.mmu; - gfn_t gfn = addr >> PAGE_SHIFT; tdp_ptep_t sptep = NULL; - tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) { + for_each_tdp_pte(iter, vcpu->kvm, root, gfn, gfn + 1) { *spte = iter.old_spte; sptep = iter.sptep; } diff --git a/arch/x86/kvm/mmu/tdp_mmu.h b/arch/x86/kvm/mmu/tdp_mmu.h index 6e1ea04ca885..52acf99d40a0 100644 --- a/arch/x86/kvm/mmu/tdp_mmu.h +++ b/arch/x86/kvm/mmu/tdp_mmu.h @@ -10,7 +10,7 @@ void kvm_mmu_init_tdp_mmu(struct kvm *kvm); void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm); -int kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu); +void kvm_tdp_mmu_alloc_root(struct kvm_vcpu *vcpu, bool private); __must_check static inline bool kvm_tdp_mmu_get_root(struct kvm_mmu_page *root) { @@ -19,11 +19,56 @@ __must_check static inline bool kvm_tdp_mmu_get_root(struct kvm_mmu_page *root) void kvm_tdp_mmu_put_root(struct kvm *kvm, struct kvm_mmu_page *root); +enum kvm_tdp_mmu_root_types { + KVM_INVALID_ROOTS = BIT(0), + KVM_DIRECT_ROOTS = BIT(1), + KVM_MIRROR_ROOTS = BIT(2), + KVM_VALID_ROOTS = KVM_DIRECT_ROOTS | KVM_MIRROR_ROOTS, + KVM_ALL_ROOTS = KVM_VALID_ROOTS | KVM_INVALID_ROOTS, +}; + +static inline enum kvm_tdp_mmu_root_types kvm_gfn_range_filter_to_root_types(struct kvm *kvm, + enum kvm_gfn_range_filter process) +{ + enum kvm_tdp_mmu_root_types ret = 0; + + if (!kvm_has_mirrored_tdp(kvm)) + return KVM_DIRECT_ROOTS; + + if (process & KVM_FILTER_PRIVATE) + ret |= KVM_MIRROR_ROOTS; + if (process & KVM_FILTER_SHARED) + ret |= KVM_DIRECT_ROOTS; + + WARN_ON_ONCE(!ret); + + return ret; +} + +static inline struct kvm_mmu_page *tdp_mmu_get_root_for_fault(struct kvm_vcpu *vcpu, + struct kvm_page_fault *fault) +{ + if (unlikely(!kvm_is_addr_direct(vcpu->kvm, fault->addr))) + return root_to_sp(vcpu->arch.mmu->mirror_root_hpa); + + return root_to_sp(vcpu->arch.mmu->root.hpa); +} + +static inline struct kvm_mmu_page *tdp_mmu_get_root(struct kvm_vcpu *vcpu, + enum kvm_tdp_mmu_root_types type) +{ + if (unlikely(type == KVM_MIRROR_ROOTS)) + return root_to_sp(vcpu->arch.mmu->mirror_root_hpa); + + return root_to_sp(vcpu->arch.mmu->root.hpa); +} + bool kvm_tdp_mmu_zap_leafs(struct kvm *kvm, gfn_t start, gfn_t end, bool flush); bool kvm_tdp_mmu_zap_sp(struct kvm *kvm, struct kvm_mmu_page *sp); void kvm_tdp_mmu_zap_all(struct kvm *kvm); -void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm); -void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm); +void kvm_tdp_mmu_invalidate_roots(struct kvm *kvm, + enum kvm_tdp_mmu_root_types root_types); +void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm, bool shared); int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault); @@ -31,18 +76,17 @@ bool kvm_tdp_mmu_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range, bool flush); bool kvm_tdp_mmu_age_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range); bool kvm_tdp_mmu_test_age_gfn(struct kvm *kvm, struct kvm_gfn_range *range); -bool kvm_tdp_mmu_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range); bool kvm_tdp_mmu_wrprot_slot(struct kvm *kvm, const struct kvm_memory_slot *slot, int min_level); -bool kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm, +void kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm, const struct kvm_memory_slot *slot); void kvm_tdp_mmu_clear_dirty_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn, unsigned long mask, bool wrprot); -void kvm_tdp_mmu_zap_collapsible_sptes(struct kvm *kvm, - const struct kvm_memory_slot *slot); +void kvm_tdp_mmu_recover_huge_pages(struct kvm *kvm, + const struct kvm_memory_slot *slot); bool kvm_tdp_mmu_write_protect_gfn(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn, @@ -65,7 +109,7 @@ static inline void kvm_tdp_mmu_walk_lockless_end(void) int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes, int *root_level); -u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, u64 addr, +u64 *kvm_tdp_mmu_fast_pf_get_last_sptep(struct kvm_vcpu *vcpu, gfn_t gfn, u64 *spte); #ifdef CONFIG_X86_64 |