X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=arch%2Fx86%2Fkvm%2Fmmu.c;h=1249c12e1d5c50febda0129ccecbfac20e43217a;hb=7e4e4056f72da51c5dede48515df0ecd20eaf8ca;hp=a1ca4ff9c11630263618cc395bf41570d3dc777f;hpb=140754bc80e1cdbf2d14cdb10d900da1f7718e7b;p=safe%2Fjmp%2Flinux-2.6 diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index a1ca4ff..1249c12 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -17,8 +17,8 @@ * */ -#include "vmx.h" #include "mmu.h" +#include "kvm_cache_regs.h" #include #include @@ -33,6 +33,7 @@ #include #include #include +#include /* * When setting this variable to true it enables Two-Dimensional-Paging @@ -70,6 +71,9 @@ static int dbg = 0; module_param(dbg, bool, 0644); #endif +static int oos_shadow = 1; +module_param(oos_shadow, bool, 0644); + #ifndef MMU_DEBUG #define ASSERT(x) do { } while (0) #else @@ -104,6 +108,9 @@ module_param(dbg, bool, 0644); #define PT32_LEVEL_MASK(level) \ (((1ULL << PT32_LEVEL_BITS) - 1) << PT32_LEVEL_SHIFT(level)) +#define PT32_LVL_OFFSET_MASK(level) \ + (PT32_BASE_ADDR_MASK & ((1ULL << (PAGE_SHIFT + (((level) - 1) \ + * PT32_LEVEL_BITS))) - 1)) #define PT32_INDEX(address, level)\ (((address) >> PT32_LEVEL_SHIFT(level)) & ((1 << PT32_LEVEL_BITS) - 1)) @@ -112,10 +119,19 @@ module_param(dbg, bool, 0644); #define PT64_BASE_ADDR_MASK (((1ULL << 52) - 1) & ~(u64)(PAGE_SIZE-1)) #define PT64_DIR_BASE_ADDR_MASK \ (PT64_BASE_ADDR_MASK & ~((1ULL << (PAGE_SHIFT + PT64_LEVEL_BITS)) - 1)) +#define PT64_LVL_ADDR_MASK(level) \ + (PT64_BASE_ADDR_MASK & ~((1ULL << (PAGE_SHIFT + (((level) - 1) \ + * PT64_LEVEL_BITS))) - 1)) +#define PT64_LVL_OFFSET_MASK(level) \ + (PT64_BASE_ADDR_MASK & ((1ULL << (PAGE_SHIFT + (((level) - 1) \ + * PT64_LEVEL_BITS))) - 1)) #define PT32_BASE_ADDR_MASK PAGE_MASK #define PT32_DIR_BASE_ADDR_MASK \ (PAGE_MASK & ~((1ULL << (PAGE_SHIFT + PT32_LEVEL_BITS)) - 1)) +#define PT32_LVL_ADDR_MASK(level) \ + (PAGE_MASK & ~((1ULL << (PAGE_SHIFT + (((level) - 1) \ + * PT32_LEVEL_BITS))) - 1)) #define PT64_PERM_MASK (PT_PRESENT_MASK | PT_WRITABLE_MASK | PT_USER_MASK \ | PT64_NX_MASK) @@ -123,8 +139,10 @@ module_param(dbg, bool, 0644); #define PFERR_PRESENT_MASK (1U << 0) #define PFERR_WRITE_MASK (1U << 1) #define PFERR_USER_MASK (1U << 2) +#define PFERR_RSVD_MASK (1U << 3) #define PFERR_FETCH_MASK (1U << 4) +#define PT_PDPE_LEVEL 3 #define PT_DIRECTORY_LEVEL 2 #define PT_PAGE_TABLE_LEVEL 1 @@ -135,18 +153,36 @@ module_param(dbg, bool, 0644); #define ACC_USER_MASK PT_USER_MASK #define ACC_ALL (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK) +#define CREATE_TRACE_POINTS +#include "mmutrace.h" + #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) struct kvm_rmap_desc { - u64 *shadow_ptes[RMAP_EXT]; + u64 *sptes[RMAP_EXT]; struct kvm_rmap_desc *more; }; -struct kvm_shadow_walk { - int (*entry)(struct kvm_shadow_walk *walk, struct kvm_vcpu *vcpu, - gva_t addr, u64 *spte, int level); +struct kvm_shadow_walk_iterator { + u64 addr; + hpa_t shadow_addr; + int level; + u64 *sptep; + unsigned index; }; +#define for_each_shadow_entry(_vcpu, _addr, _walker) \ + for (shadow_walk_init(&(_walker), _vcpu, _addr); \ + shadow_walk_okay(&(_walker)); \ + shadow_walk_next(&(_walker))) + + +struct kvm_unsync_walk { + int (*entry) (struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk); +}; + +typedef int (*mmu_parent_walk_fn) (struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp); + static struct kmem_cache *pte_chain_cache; static struct kmem_cache *rmap_desc_cache; static struct kmem_cache *mmu_page_header_cache; @@ -160,6 +196,11 @@ static u64 __read_mostly shadow_user_mask; static u64 __read_mostly shadow_accessed_mask; static u64 __read_mostly shadow_dirty_mask; +static inline u64 rsvd_bits(int s, int e) +{ + return ((1ULL << (e - s + 1)) - 1) << s; +} + void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte) { shadow_trap_nonpresent_pte = trap_pte; @@ -199,11 +240,6 @@ static int is_nx(struct kvm_vcpu *vcpu) return vcpu->arch.shadow_efer & EFER_NX; } -static int is_present_pte(unsigned long pte) -{ - return pte & PT_PRESENT_MASK; -} - static int is_shadow_present_pte(u64 pte) { return pte != shadow_trap_nonpresent_pte @@ -220,16 +256,25 @@ static int is_writeble_pte(unsigned long pte) return pte & PT_WRITABLE_MASK; } -static int is_dirty_pte(unsigned long pte) +static int is_dirty_gpte(unsigned long pte) { - return pte & shadow_dirty_mask; + return pte & PT_DIRTY_MASK; } -static int is_rmap_pte(u64 pte) +static int is_rmap_spte(u64 pte) { return is_shadow_present_pte(pte); } +static int is_last_spte(u64 pte, int level) +{ + if (level == PT_PAGE_TABLE_LEVEL) + return 1; + if (is_large_pte(pte)) + return 1; + return 0; +} + static pfn_t spte_to_pfn(u64 pte) { return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT; @@ -242,7 +287,7 @@ static gfn_t pse36_gfn_delta(u32 gpte) return (gpte & PT32_DIR_PSE36_MASK) << shift; } -static void set_shadow_pte(u64 *sptep, u64 spte) +static void __set_spte(u64 *sptep, u64 spte) { #ifdef CONFIG_X86_64 set_64bit((unsigned long *)sptep, spte); @@ -305,7 +350,7 @@ static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu) if (r) goto out; r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache, - rmap_desc_cache, 1); + rmap_desc_cache, 4); if (r) goto out; r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8); @@ -332,7 +377,6 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc, BUG_ON(!mc->nobjs); p = mc->objects[--mc->nobjs]; - memset(p, 0, size); return p; } @@ -362,76 +406,120 @@ static void mmu_free_rmap_desc(struct kvm_rmap_desc *rd) * Return the pointer to the largepage write count for a given * gfn, handling slots that are not large page aligned. */ -static int *slot_largepage_idx(gfn_t gfn, struct kvm_memory_slot *slot) +static int *slot_largepage_idx(gfn_t gfn, + struct kvm_memory_slot *slot, + int level) { unsigned long idx; - idx = (gfn / KVM_PAGES_PER_HPAGE) - - (slot->base_gfn / KVM_PAGES_PER_HPAGE); - return &slot->lpage_info[idx].write_count; + idx = (gfn / KVM_PAGES_PER_HPAGE(level)) - + (slot->base_gfn / KVM_PAGES_PER_HPAGE(level)); + return &slot->lpage_info[level - 2][idx].write_count; } static void account_shadowed(struct kvm *kvm, gfn_t gfn) { + struct kvm_memory_slot *slot; int *write_count; + int i; + + gfn = unalias_gfn(kvm, gfn); - write_count = slot_largepage_idx(gfn, gfn_to_memslot(kvm, gfn)); - *write_count += 1; + slot = gfn_to_memslot_unaliased(kvm, gfn); + for (i = PT_DIRECTORY_LEVEL; + i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { + write_count = slot_largepage_idx(gfn, slot, i); + *write_count += 1; + } } static void unaccount_shadowed(struct kvm *kvm, gfn_t gfn) { + struct kvm_memory_slot *slot; int *write_count; + int i; - write_count = slot_largepage_idx(gfn, gfn_to_memslot(kvm, gfn)); - *write_count -= 1; - WARN_ON(*write_count < 0); + gfn = unalias_gfn(kvm, gfn); + for (i = PT_DIRECTORY_LEVEL; + i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { + slot = gfn_to_memslot_unaliased(kvm, gfn); + write_count = slot_largepage_idx(gfn, slot, i); + *write_count -= 1; + WARN_ON(*write_count < 0); + } } -static int has_wrprotected_page(struct kvm *kvm, gfn_t gfn) +static int has_wrprotected_page(struct kvm *kvm, + gfn_t gfn, + int level) { - struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn); + struct kvm_memory_slot *slot; int *largepage_idx; + gfn = unalias_gfn(kvm, gfn); + slot = gfn_to_memslot_unaliased(kvm, gfn); if (slot) { - largepage_idx = slot_largepage_idx(gfn, slot); + largepage_idx = slot_largepage_idx(gfn, slot, level); return *largepage_idx; } return 1; } -static int host_largepage_backed(struct kvm *kvm, gfn_t gfn) +static int host_mapping_level(struct kvm *kvm, gfn_t gfn) { + unsigned long page_size = PAGE_SIZE; struct vm_area_struct *vma; unsigned long addr; + int i, ret = 0; addr = gfn_to_hva(kvm, gfn); if (kvm_is_error_hva(addr)) - return 0; + return page_size; + down_read(¤t->mm->mmap_sem); vma = find_vma(current->mm, addr); - if (vma && is_vm_hugetlb_page(vma)) - return 1; + if (!vma) + goto out; - return 0; + page_size = vma_kernel_pagesize(vma); + +out: + up_read(¤t->mm->mmap_sem); + + for (i = PT_PAGE_TABLE_LEVEL; + i < (PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES); ++i) { + if (page_size >= KVM_HPAGE_SIZE(i)) + ret = i; + else + break; + } + + return ret; } -static int is_largepage_backed(struct kvm_vcpu *vcpu, gfn_t large_gfn) +static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn) { struct kvm_memory_slot *slot; - - if (has_wrprotected_page(vcpu->kvm, large_gfn)) - return 0; - - if (!host_largepage_backed(vcpu->kvm, large_gfn)) - return 0; + int host_level; + int level = PT_PAGE_TABLE_LEVEL; slot = gfn_to_memslot(vcpu->kvm, large_gfn); if (slot && slot->dirty_bitmap) - return 0; + return PT_PAGE_TABLE_LEVEL; - return 1; + host_level = host_mapping_level(vcpu->kvm, large_gfn); + + if (host_level == PT_PAGE_TABLE_LEVEL) + return host_level; + + for (level = PT_DIRECTORY_LEVEL; level <= host_level; ++level) { + + if (has_wrprotected_page(vcpu->kvm, large_gfn, level)) + break; + } + + return level - 1; } /* @@ -439,19 +527,19 @@ static int is_largepage_backed(struct kvm_vcpu *vcpu, gfn_t large_gfn) * Note: gfn must be unaliased before this function get called */ -static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int lpage) +static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int level) { struct kvm_memory_slot *slot; unsigned long idx; slot = gfn_to_memslot(kvm, gfn); - if (!lpage) + if (likely(level == PT_PAGE_TABLE_LEVEL)) return &slot->rmap[gfn - slot->base_gfn]; - idx = (gfn / KVM_PAGES_PER_HPAGE) - - (slot->base_gfn / KVM_PAGES_PER_HPAGE); + idx = (gfn / KVM_PAGES_PER_HPAGE(level)) - + (slot->base_gfn / KVM_PAGES_PER_HPAGE(level)); - return &slot->lpage_info[idx].rmap_pde; + return &slot->lpage_info[level - 2][idx].rmap_pde; } /* @@ -462,42 +550,49 @@ static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int lpage) * * If rmapp bit zero is one, (then rmap & ~1) points to a struct kvm_rmap_desc * containing more mappings. + * + * Returns the number of rmap entries before the spte was added or zero if + * the spte was not added. + * */ -static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn, int lpage) +static int rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) { struct kvm_mmu_page *sp; struct kvm_rmap_desc *desc; unsigned long *rmapp; - int i; + int i, count = 0; - if (!is_rmap_pte(*spte)) - return; + if (!is_rmap_spte(*spte)) + return count; gfn = unalias_gfn(vcpu->kvm, gfn); sp = page_header(__pa(spte)); sp->gfns[spte - sp->spt] = gfn; - rmapp = gfn_to_rmap(vcpu->kvm, gfn, lpage); + rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level); if (!*rmapp) { rmap_printk("rmap_add: %p %llx 0->1\n", spte, *spte); *rmapp = (unsigned long)spte; } else if (!(*rmapp & 1)) { rmap_printk("rmap_add: %p %llx 1->many\n", spte, *spte); desc = mmu_alloc_rmap_desc(vcpu); - desc->shadow_ptes[0] = (u64 *)*rmapp; - desc->shadow_ptes[1] = spte; + desc->sptes[0] = (u64 *)*rmapp; + desc->sptes[1] = spte; *rmapp = (unsigned long)desc | 1; } else { rmap_printk("rmap_add: %p %llx many->many\n", spte, *spte); desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul); - while (desc->shadow_ptes[RMAP_EXT-1] && desc->more) + while (desc->sptes[RMAP_EXT-1] && desc->more) { desc = desc->more; - if (desc->shadow_ptes[RMAP_EXT-1]) { + count += RMAP_EXT; + } + if (desc->sptes[RMAP_EXT-1]) { desc->more = mmu_alloc_rmap_desc(vcpu); desc = desc->more; } - for (i = 0; desc->shadow_ptes[i]; ++i) + for (i = 0; desc->sptes[i]; ++i) ; - desc->shadow_ptes[i] = spte; + desc->sptes[i] = spte; } + return count; } static void rmap_desc_remove_entry(unsigned long *rmapp, @@ -507,14 +602,14 @@ static void rmap_desc_remove_entry(unsigned long *rmapp, { int j; - for (j = RMAP_EXT - 1; !desc->shadow_ptes[j] && j > i; --j) + for (j = RMAP_EXT - 1; !desc->sptes[j] && j > i; --j) ; - desc->shadow_ptes[i] = desc->shadow_ptes[j]; - desc->shadow_ptes[j] = NULL; + desc->sptes[i] = desc->sptes[j]; + desc->sptes[j] = NULL; if (j != 0) return; if (!prev_desc && !desc->more) - *rmapp = (unsigned long)desc->shadow_ptes[0]; + *rmapp = (unsigned long)desc->sptes[0]; else if (prev_desc) prev_desc->more = desc->more; @@ -532,7 +627,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) unsigned long *rmapp; int i; - if (!is_rmap_pte(*spte)) + if (!is_rmap_spte(*spte)) return; sp = page_header(__pa(spte)); pfn = spte_to_pfn(*spte); @@ -542,7 +637,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) kvm_release_pfn_dirty(pfn); else kvm_release_pfn_clean(pfn); - rmapp = gfn_to_rmap(kvm, sp->gfns[spte - sp->spt], is_large_pte(*spte)); + rmapp = gfn_to_rmap(kvm, sp->gfns[spte - sp->spt], sp->role.level); if (!*rmapp) { printk(KERN_ERR "rmap_remove: %p %llx 0->BUG\n", spte, *spte); BUG(); @@ -559,8 +654,8 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul); prev_desc = NULL; while (desc) { - for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i) - if (desc->shadow_ptes[i] == spte) { + for (i = 0; i < RMAP_EXT && desc->sptes[i]; ++i) + if (desc->sptes[i] == spte) { rmap_desc_remove_entry(rmapp, desc, i, prev_desc); @@ -591,24 +686,24 @@ static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte) prev_desc = NULL; prev_spte = NULL; while (desc) { - for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i) { + for (i = 0; i < RMAP_EXT && desc->sptes[i]; ++i) { if (prev_spte == spte) - return desc->shadow_ptes[i]; - prev_spte = desc->shadow_ptes[i]; + return desc->sptes[i]; + prev_spte = desc->sptes[i]; } desc = desc->more; } return NULL; } -static void rmap_write_protect(struct kvm *kvm, u64 gfn) +static int rmap_write_protect(struct kvm *kvm, u64 gfn) { unsigned long *rmapp; u64 *spte; - int write_protected = 0; + int i, write_protected = 0; gfn = unalias_gfn(kvm, gfn); - rmapp = gfn_to_rmap(kvm, gfn, 0); + rmapp = gfn_to_rmap(kvm, gfn, PT_PAGE_TABLE_LEVEL); spte = rmap_next(kvm, rmapp, NULL); while (spte) { @@ -616,7 +711,7 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn) BUG_ON(!(*spte & PT_PRESENT_MASK)); rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte); if (is_writeble_pte(*spte)) { - set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK); + __set_spte(spte, *spte & ~PT_WRITABLE_MASK); write_protected = 1; } spte = rmap_next(kvm, rmapp, spte); @@ -630,27 +725,27 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn) } /* check for huge page mappings */ - rmapp = gfn_to_rmap(kvm, gfn, 1); - spte = rmap_next(kvm, rmapp, NULL); - while (spte) { - BUG_ON(!spte); - BUG_ON(!(*spte & PT_PRESENT_MASK)); - BUG_ON((*spte & (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)) != (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)); - pgprintk("rmap_write_protect(large): spte %p %llx %lld\n", spte, *spte, gfn); - if (is_writeble_pte(*spte)) { - rmap_remove(kvm, spte); - --kvm->stat.lpages; - set_shadow_pte(spte, shadow_trap_nonpresent_pte); - spte = NULL; - write_protected = 1; + for (i = PT_DIRECTORY_LEVEL; + i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { + rmapp = gfn_to_rmap(kvm, gfn, i); + spte = rmap_next(kvm, rmapp, NULL); + while (spte) { + BUG_ON(!spte); + BUG_ON(!(*spte & PT_PRESENT_MASK)); + BUG_ON((*spte & (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)) != (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)); + pgprintk("rmap_write_protect(large): spte %p %llx %lld\n", spte, *spte, gfn); + if (is_writeble_pte(*spte)) { + rmap_remove(kvm, spte); + --kvm->stat.lpages; + __set_spte(spte, shadow_trap_nonpresent_pte); + spte = NULL; + write_protected = 1; + } + spte = rmap_next(kvm, rmapp, spte); } - spte = rmap_next(kvm, rmapp, spte); } - if (write_protected) - kvm_flush_remote_tlbs(kvm); - - account_shadowed(kvm, gfn); + return write_protected; } static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp) @@ -662,7 +757,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp) BUG_ON(!(*spte & PT_PRESENT_MASK)); rmap_printk("kvm_rmap_unmap_hva: spte %p %llx\n", spte, *spte); rmap_remove(kvm, spte); - set_shadow_pte(spte, shadow_trap_nonpresent_pte); + __set_spte(spte, shadow_trap_nonpresent_pte); need_tlb_flush = 1; } return need_tlb_flush; @@ -671,7 +766,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp) static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, int (*handler)(struct kvm *kvm, unsigned long *rmapp)) { - int i; + int i, j; int retval = 0; /* @@ -690,11 +785,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, end = start + (memslot->npages << PAGE_SHIFT); if (hva >= start && hva < end) { gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT; + retval |= handler(kvm, &memslot->rmap[gfn_offset]); - retval |= handler(kvm, - &memslot->lpage_info[ - gfn_offset / - KVM_PAGES_PER_HPAGE].rmap_pde); + + for (j = 0; j < KVM_NR_PAGE_SIZES - 1; ++j) { + int idx = gfn_offset; + idx /= KVM_PAGES_PER_HPAGE(PT_DIRECTORY_LEVEL + j); + retval |= handler(kvm, + &memslot->lpage_info[j][idx].rmap_pde); + } } } @@ -730,6 +829,22 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp) return young; } +#define RMAP_RECYCLE_THRESHOLD 1000 + +static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) +{ + unsigned long *rmapp; + struct kvm_mmu_page *sp; + + sp = page_header(__pa(spte)); + + gfn = unalias_gfn(vcpu->kvm, gfn); + rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level); + + kvm_unmap_rmapp(vcpu->kvm, rmapp); + kvm_flush_remote_tlbs(vcpu->kvm); +} + int kvm_age_hva(struct kvm *kvm, unsigned long hva) { return kvm_handle_hva(kvm, hva, kvm_age_rmapp); @@ -776,8 +891,8 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu, sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE); set_page_private(virt_to_page(sp->spt), (unsigned long)sp); list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages); - ASSERT(is_empty_shadow_page(sp->spt)); - sp->slot_bitmap = 0; + INIT_LIST_HEAD(&sp->oos_link); + bitmap_zero(sp->slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS); sp->multimapped = 0; sp->parent_pte = parent_pte; --vcpu->kvm->arch.n_free_mmu_pages; @@ -859,6 +974,77 @@ static void mmu_page_remove_parent_pte(struct kvm_mmu_page *sp, BUG(); } + +static void mmu_parent_walk(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, + mmu_parent_walk_fn fn) +{ + struct kvm_pte_chain *pte_chain; + struct hlist_node *node; + struct kvm_mmu_page *parent_sp; + int i; + + if (!sp->multimapped && sp->parent_pte) { + parent_sp = page_header(__pa(sp->parent_pte)); + fn(vcpu, parent_sp); + mmu_parent_walk(vcpu, parent_sp, fn); + return; + } + hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) + for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) { + if (!pte_chain->parent_ptes[i]) + break; + parent_sp = page_header(__pa(pte_chain->parent_ptes[i])); + fn(vcpu, parent_sp); + mmu_parent_walk(vcpu, parent_sp, fn); + } +} + +static void kvm_mmu_update_unsync_bitmap(u64 *spte) +{ + unsigned int index; + struct kvm_mmu_page *sp = page_header(__pa(spte)); + + index = spte - sp->spt; + if (!__test_and_set_bit(index, sp->unsync_child_bitmap)) + sp->unsync_children++; + WARN_ON(!sp->unsync_children); +} + +static void kvm_mmu_update_parents_unsync(struct kvm_mmu_page *sp) +{ + struct kvm_pte_chain *pte_chain; + struct hlist_node *node; + int i; + + if (!sp->parent_pte) + return; + + if (!sp->multimapped) { + kvm_mmu_update_unsync_bitmap(sp->parent_pte); + return; + } + + hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) + for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i) { + if (!pte_chain->parent_ptes[i]) + break; + kvm_mmu_update_unsync_bitmap(pte_chain->parent_ptes[i]); + } +} + +static int unsync_walk_fn(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + kvm_mmu_update_parents_unsync(sp); + return 1; +} + +static void kvm_mmu_mark_parents_unsync(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + mmu_parent_walk(vcpu, sp, unsync_walk_fn); + kvm_mmu_update_parents_unsync(sp); +} + static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { @@ -868,6 +1054,96 @@ static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, sp->spt[i] = shadow_trap_nonpresent_pte; } +static int nonpaging_sync_page(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *sp) +{ + return 1; +} + +static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ +} + +#define KVM_PAGE_ARRAY_NR 16 + +struct kvm_mmu_pages { + struct mmu_page_and_offset { + struct kvm_mmu_page *sp; + unsigned int idx; + } page[KVM_PAGE_ARRAY_NR]; + unsigned int nr; +}; + +#define for_each_unsync_children(bitmap, idx) \ + for (idx = find_first_bit(bitmap, 512); \ + idx < 512; \ + idx = find_next_bit(bitmap, 512, idx+1)) + +static int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp, + int idx) +{ + int i; + + if (sp->unsync) + for (i=0; i < pvec->nr; i++) + if (pvec->page[i].sp == sp) + return 0; + + pvec->page[pvec->nr].sp = sp; + pvec->page[pvec->nr].idx = idx; + pvec->nr++; + return (pvec->nr == KVM_PAGE_ARRAY_NR); +} + +static int __mmu_unsync_walk(struct kvm_mmu_page *sp, + struct kvm_mmu_pages *pvec) +{ + int i, ret, nr_unsync_leaf = 0; + + for_each_unsync_children(sp->unsync_child_bitmap, i) { + u64 ent = sp->spt[i]; + + if (is_shadow_present_pte(ent) && !is_large_pte(ent)) { + struct kvm_mmu_page *child; + child = page_header(ent & PT64_BASE_ADDR_MASK); + + if (child->unsync_children) { + if (mmu_pages_add(pvec, child, i)) + return -ENOSPC; + + ret = __mmu_unsync_walk(child, pvec); + if (!ret) + __clear_bit(i, sp->unsync_child_bitmap); + else if (ret > 0) + nr_unsync_leaf += ret; + else + return ret; + } + + if (child->unsync) { + nr_unsync_leaf++; + if (mmu_pages_add(pvec, child, i)) + return -ENOSPC; + } + } + } + + if (find_first_bit(sp->unsync_child_bitmap, 512) == 512) + sp->unsync_children = 0; + + return nr_unsync_leaf; +} + +static int mmu_unsync_walk(struct kvm_mmu_page *sp, + struct kvm_mmu_pages *pvec) +{ + if (!sp->unsync_children) + return 0; + + mmu_pages_add(pvec, sp, 0); + return __mmu_unsync_walk(sp, pvec); +} + static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) { unsigned index; @@ -879,7 +1155,7 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) index = kvm_page_table_hashfn(gfn); bucket = &kvm->arch.mmu_page_hash[index]; hlist_for_each_entry(sp, node, bucket, hash_link) - if (sp->gfn == gfn && !sp->role.metaphysical + if (sp->gfn == gfn && !sp->role.direct && !sp->role.invalid) { pgprintk("%s: found role %x\n", __func__, sp->role.word); @@ -888,11 +1164,126 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) return NULL; } +static void kvm_unlink_unsync_page(struct kvm *kvm, struct kvm_mmu_page *sp) +{ + WARN_ON(!sp->unsync); + sp->unsync = 0; + --kvm->stat.mmu_unsync; +} + +static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp); + +static int kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + if (sp->role.glevels != vcpu->arch.mmu.root_level) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + trace_kvm_mmu_sync_page(sp); + if (rmap_write_protect(vcpu->kvm, sp->gfn)) + kvm_flush_remote_tlbs(vcpu->kvm); + kvm_unlink_unsync_page(vcpu->kvm, sp); + if (vcpu->arch.mmu.sync_page(vcpu, sp)) { + kvm_mmu_zap_page(vcpu->kvm, sp); + return 1; + } + + kvm_mmu_flush_tlb(vcpu); + return 0; +} + +struct mmu_page_path { + struct kvm_mmu_page *parent[PT64_ROOT_LEVEL-1]; + unsigned int idx[PT64_ROOT_LEVEL-1]; +}; + +#define for_each_sp(pvec, sp, parents, i) \ + for (i = mmu_pages_next(&pvec, &parents, -1), \ + sp = pvec.page[i].sp; \ + i < pvec.nr && ({ sp = pvec.page[i].sp; 1;}); \ + i = mmu_pages_next(&pvec, &parents, i)) + +static int mmu_pages_next(struct kvm_mmu_pages *pvec, + struct mmu_page_path *parents, + int i) +{ + int n; + + for (n = i+1; n < pvec->nr; n++) { + struct kvm_mmu_page *sp = pvec->page[n].sp; + + if (sp->role.level == PT_PAGE_TABLE_LEVEL) { + parents->idx[0] = pvec->page[n].idx; + return n; + } + + parents->parent[sp->role.level-2] = sp; + parents->idx[sp->role.level-1] = pvec->page[n].idx; + } + + return n; +} + +static void mmu_pages_clear_parents(struct mmu_page_path *parents) +{ + struct kvm_mmu_page *sp; + unsigned int level = 0; + + do { + unsigned int idx = parents->idx[level]; + + sp = parents->parent[level]; + if (!sp) + return; + + --sp->unsync_children; + WARN_ON((int)sp->unsync_children < 0); + __clear_bit(idx, sp->unsync_child_bitmap); + level++; + } while (level < PT64_ROOT_LEVEL-1 && !sp->unsync_children); +} + +static void kvm_mmu_pages_init(struct kvm_mmu_page *parent, + struct mmu_page_path *parents, + struct kvm_mmu_pages *pvec) +{ + parents->parent[parent->role.level-1] = NULL; + pvec->nr = 0; +} + +static void mmu_sync_children(struct kvm_vcpu *vcpu, + struct kvm_mmu_page *parent) +{ + int i; + struct kvm_mmu_page *sp; + struct mmu_page_path parents; + struct kvm_mmu_pages pages; + + kvm_mmu_pages_init(parent, &parents, &pages); + while (mmu_unsync_walk(parent, &pages)) { + int protected = 0; + + for_each_sp(pages, sp, parents, i) + protected |= rmap_write_protect(vcpu->kvm, sp->gfn); + + if (protected) + kvm_flush_remote_tlbs(vcpu->kvm); + + for_each_sp(pages, sp, parents, i) { + kvm_sync_page(vcpu, sp); + mmu_pages_clear_parents(&parents); + } + cond_resched_lock(&vcpu->kvm->mmu_lock); + kvm_mmu_pages_init(parent, &parents, &pages); + } +} + static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr, unsigned level, - int metaphysical, + int direct, unsigned access, u64 *parent_pte) { @@ -901,72 +1292,90 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, unsigned quadrant; struct hlist_head *bucket; struct kvm_mmu_page *sp; - struct hlist_node *node; + struct hlist_node *node, *tmp; - role.word = 0; - role.glevels = vcpu->arch.mmu.root_level; + role = vcpu->arch.mmu.base_role; role.level = level; - role.metaphysical = metaphysical; + role.direct = direct; role.access = access; if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) { quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level)); quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1; role.quadrant = quadrant; } - pgprintk("%s: looking gfn %lx role %x\n", __func__, - gfn, role.word); index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; - hlist_for_each_entry(sp, node, bucket, hash_link) - if (sp->gfn == gfn && sp->role.word == role.word) { + hlist_for_each_entry_safe(sp, node, tmp, bucket, hash_link) + if (sp->gfn == gfn) { + if (sp->unsync) + if (kvm_sync_page(vcpu, sp)) + continue; + + if (sp->role.word != role.word) + continue; + mmu_page_add_parent_pte(vcpu, sp, parent_pte); - pgprintk("%s: found\n", __func__); + if (sp->unsync_children) { + set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); + kvm_mmu_mark_parents_unsync(vcpu, sp); + } + trace_kvm_mmu_get_page(sp, false); return sp; } ++vcpu->kvm->stat.mmu_cache_miss; sp = kvm_mmu_alloc_page(vcpu, parent_pte); if (!sp) return sp; - pgprintk("%s: adding gfn %lx role %x\n", __func__, gfn, role.word); sp->gfn = gfn; sp->role = role; hlist_add_head(&sp->hash_link, bucket); - if (!metaphysical) - rmap_write_protect(vcpu->kvm, gfn); + if (!direct) { + if (rmap_write_protect(vcpu->kvm, gfn)) + kvm_flush_remote_tlbs(vcpu->kvm); + account_shadowed(vcpu->kvm, gfn); + } if (shadow_trap_nonpresent_pte != shadow_notrap_nonpresent_pte) vcpu->arch.mmu.prefetch_page(vcpu, sp); else nonpaging_prefetch_page(vcpu, sp); + trace_kvm_mmu_get_page(sp, true); return sp; } -static int walk_shadow(struct kvm_shadow_walk *walker, - struct kvm_vcpu *vcpu, gva_t addr) +static void shadow_walk_init(struct kvm_shadow_walk_iterator *iterator, + struct kvm_vcpu *vcpu, u64 addr) { - hpa_t shadow_addr; - int level; - int r; - u64 *sptep; - unsigned index; - - shadow_addr = vcpu->arch.mmu.root_hpa; - level = vcpu->arch.mmu.shadow_root_level; - if (level == PT32E_ROOT_LEVEL) { - shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; - shadow_addr &= PT64_BASE_ADDR_MASK; - --level; + iterator->addr = addr; + iterator->shadow_addr = vcpu->arch.mmu.root_hpa; + iterator->level = vcpu->arch.mmu.shadow_root_level; + if (iterator->level == PT32E_ROOT_LEVEL) { + iterator->shadow_addr + = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; + iterator->shadow_addr &= PT64_BASE_ADDR_MASK; + --iterator->level; + if (!iterator->shadow_addr) + iterator->level = 0; } +} - while (level >= PT_PAGE_TABLE_LEVEL) { - index = SHADOW_PT_INDEX(addr, level); - sptep = ((u64 *)__va(shadow_addr)) + index; - r = walker->entry(walker, vcpu, addr, sptep, level); - if (r) - return r; - shadow_addr = *sptep & PT64_BASE_ADDR_MASK; - --level; - } - return 0; +static bool shadow_walk_okay(struct kvm_shadow_walk_iterator *iterator) +{ + if (iterator->level < PT_PAGE_TABLE_LEVEL) + return false; + + if (iterator->level == PT_PAGE_TABLE_LEVEL) + if (is_large_pte(*iterator->sptep)) + return false; + + iterator->index = SHADOW_PT_INDEX(iterator->addr, iterator->level); + iterator->sptep = ((u64 *)__va(iterator->shadow_addr)) + iterator->index; + return true; +} + +static void shadow_walk_next(struct kvm_shadow_walk_iterator *iterator) +{ + iterator->shadow_addr = *iterator->sptep & PT64_BASE_ADDR_MASK; + --iterator->level; } static void kvm_mmu_page_unlink_children(struct kvm *kvm, @@ -978,25 +1387,17 @@ static void kvm_mmu_page_unlink_children(struct kvm *kvm, pt = sp->spt; - if (sp->role.level == PT_PAGE_TABLE_LEVEL) { - for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { - if (is_shadow_present_pte(pt[i])) - rmap_remove(kvm, &pt[i]); - pt[i] = shadow_trap_nonpresent_pte; - } - return; - } - for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { ent = pt[i]; if (is_shadow_present_pte(ent)) { - if (!is_large_pte(ent)) { + if (!is_last_spte(ent, sp->role.level)) { ent &= PT64_BASE_ADDR_MASK; mmu_page_remove_parent_pte(page_header(ent), &pt[i]); } else { - --kvm->stat.lpages; + if (is_large_pte(ent)) + --kvm->stat.lpages; rmap_remove(kvm, &pt[i]); } } @@ -1012,10 +1413,10 @@ static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte) static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm) { int i; + struct kvm_vcpu *vcpu; - for (i = 0; i < KVM_MAX_VCPUS; ++i) - if (kvm->vcpus[i]) - kvm->vcpus[i]->arch.last_pte_updated = NULL; + kvm_for_each_vcpu(i, vcpu, kvm) + vcpu->arch.last_pte_updated = NULL; } static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp) @@ -1034,18 +1435,49 @@ static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp) } BUG_ON(!parent_pte); kvm_mmu_put_page(sp, parent_pte); - set_shadow_pte(parent_pte, shadow_trap_nonpresent_pte); + __set_spte(parent_pte, shadow_trap_nonpresent_pte); + } +} + +static int mmu_zap_unsync_children(struct kvm *kvm, + struct kvm_mmu_page *parent) +{ + int i, zapped = 0; + struct mmu_page_path parents; + struct kvm_mmu_pages pages; + + if (parent->role.level == PT_PAGE_TABLE_LEVEL) + return 0; + + kvm_mmu_pages_init(parent, &parents, &pages); + while (mmu_unsync_walk(parent, &pages)) { + struct kvm_mmu_page *sp; + + for_each_sp(pages, sp, parents, i) { + kvm_mmu_zap_page(kvm, sp); + mmu_pages_clear_parents(&parents); + } + zapped += pages.nr; + kvm_mmu_pages_init(parent, &parents, &pages); } + + return zapped; } -static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) +static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) { + int ret; + + trace_kvm_mmu_zap_page(sp); ++kvm->stat.mmu_shadow_zapped; + ret = mmu_zap_unsync_children(kvm, sp); kvm_mmu_page_unlink_children(kvm, sp); kvm_mmu_unlink_parents(kvm, sp); kvm_flush_remote_tlbs(kvm); - if (!sp->role.invalid && !sp->role.metaphysical) + if (!sp->role.invalid && !sp->role.direct) unaccount_shadowed(kvm, sp->gfn); + if (sp->unsync) + kvm_unlink_unsync_page(kvm, sp); if (!sp->root_count) { hlist_del(&sp->hash_link); kvm_mmu_free_page(kvm, sp); @@ -1055,6 +1487,7 @@ static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) kvm_reload_remote_mmus(kvm); } kvm_mmu_reset_last_pte_updated(kvm); + return ret; } /* @@ -1063,24 +1496,25 @@ static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) */ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages) { + int used_pages; + + used_pages = kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages; + used_pages = max(0, used_pages); + /* * If we set the number of mmu pages to be smaller be than the * number of actived pages , we must to free some mmu pages before we * change the value */ - if ((kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages) > - kvm_nr_mmu_pages) { - int n_used_mmu_pages = kvm->arch.n_alloc_mmu_pages - - kvm->arch.n_free_mmu_pages; - - while (n_used_mmu_pages > kvm_nr_mmu_pages) { + if (used_pages > kvm_nr_mmu_pages) { + while (used_pages > kvm_nr_mmu_pages) { struct kvm_mmu_page *page; page = container_of(kvm->arch.active_mmu_pages.prev, struct kvm_mmu_page, link); kvm_mmu_zap_page(kvm, page); - n_used_mmu_pages--; + used_pages--; } kvm->arch.n_free_mmu_pages = 0; } @@ -1104,22 +1538,32 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) index = kvm_page_table_hashfn(gfn); bucket = &kvm->arch.mmu_page_hash[index]; hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) - if (sp->gfn == gfn && !sp->role.metaphysical) { + if (sp->gfn == gfn && !sp->role.direct) { pgprintk("%s: gfn %lx role %x\n", __func__, gfn, sp->role.word); - kvm_mmu_zap_page(kvm, sp); r = 1; + if (kvm_mmu_zap_page(kvm, sp)) + n = bucket->first; } return r; } static void mmu_unshadow(struct kvm *kvm, gfn_t gfn) { + unsigned index; + struct hlist_head *bucket; struct kvm_mmu_page *sp; + struct hlist_node *node, *nn; - while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) { - pgprintk("%s: zap %lx %x\n", __func__, gfn, sp->role.word); - kvm_mmu_zap_page(kvm, sp); + index = kvm_page_table_hashfn(gfn); + bucket = &kvm->arch.mmu_page_hash[index]; + hlist_for_each_entry_safe(sp, node, nn, bucket, hash_link) { + if (sp->gfn == gfn && !sp->role.direct + && !sp->role.invalid) { + pgprintk("%s: zap %lx %x\n", + __func__, gfn, sp->role.word); + kvm_mmu_zap_page(kvm, sp); + } } } @@ -1128,62 +1572,194 @@ static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn) int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn)); struct kvm_mmu_page *sp = page_header(__pa(pte)); - __set_bit(slot, &sp->slot_bitmap); + __set_bit(slot, sp->slot_bitmap); +} + +static void mmu_convert_notrap(struct kvm_mmu_page *sp) +{ + int i; + u64 *pt = sp->spt; + + if (shadow_trap_nonpresent_pte == shadow_notrap_nonpresent_pte) + return; + + for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { + if (pt[i] == shadow_notrap_nonpresent_pte) + __set_spte(&pt[i], shadow_trap_nonpresent_pte); + } +} + +struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) +{ + struct page *page; + + gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva); + + if (gpa == UNMAPPED_GVA) + return NULL; + + page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT); + + return page; +} + +/* + * The function is based on mtrr_type_lookup() in + * arch/x86/kernel/cpu/mtrr/generic.c + */ +static int get_mtrr_type(struct mtrr_state_type *mtrr_state, + u64 start, u64 end) +{ + int i; + u64 base, mask; + u8 prev_match, curr_match; + int num_var_ranges = KVM_NR_VAR_MTRR; + + if (!mtrr_state->enabled) + return 0xFF; + + /* Make end inclusive end, instead of exclusive */ + end--; + + /* Look in fixed ranges. Just return the type as per start */ + if (mtrr_state->have_fixed && (start < 0x100000)) { + int idx; + + if (start < 0x80000) { + idx = 0; + idx += (start >> 16); + return mtrr_state->fixed_ranges[idx]; + } else if (start < 0xC0000) { + idx = 1 * 8; + idx += ((start - 0x80000) >> 14); + return mtrr_state->fixed_ranges[idx]; + } else if (start < 0x1000000) { + idx = 3 * 8; + idx += ((start - 0xC0000) >> 12); + return mtrr_state->fixed_ranges[idx]; + } + } + + /* + * Look in variable ranges + * Look of multiple ranges matching this address and pick type + * as per MTRR precedence + */ + if (!(mtrr_state->enabled & 2)) + return mtrr_state->def_type; + + prev_match = 0xFF; + for (i = 0; i < num_var_ranges; ++i) { + unsigned short start_state, end_state; + + if (!(mtrr_state->var_ranges[i].mask_lo & (1 << 11))) + continue; + + base = (((u64)mtrr_state->var_ranges[i].base_hi) << 32) + + (mtrr_state->var_ranges[i].base_lo & PAGE_MASK); + mask = (((u64)mtrr_state->var_ranges[i].mask_hi) << 32) + + (mtrr_state->var_ranges[i].mask_lo & PAGE_MASK); + + start_state = ((start & mask) == (base & mask)); + end_state = ((end & mask) == (base & mask)); + if (start_state != end_state) + return 0xFE; + + if ((start & mask) != (base & mask)) + continue; + + curr_match = mtrr_state->var_ranges[i].base_lo & 0xff; + if (prev_match == 0xFF) { + prev_match = curr_match; + continue; + } + + if (prev_match == MTRR_TYPE_UNCACHABLE || + curr_match == MTRR_TYPE_UNCACHABLE) + return MTRR_TYPE_UNCACHABLE; + + if ((prev_match == MTRR_TYPE_WRBACK && + curr_match == MTRR_TYPE_WRTHROUGH) || + (prev_match == MTRR_TYPE_WRTHROUGH && + curr_match == MTRR_TYPE_WRBACK)) { + prev_match = MTRR_TYPE_WRTHROUGH; + curr_match = MTRR_TYPE_WRTHROUGH; + } + + if (prev_match != curr_match) + return MTRR_TYPE_UNCACHABLE; + } + + if (prev_match != 0xFF) + return prev_match; + + return mtrr_state->def_type; +} + +u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + u8 mtrr; + + mtrr = get_mtrr_type(&vcpu->arch.mtrr_state, gfn << PAGE_SHIFT, + (gfn << PAGE_SHIFT) + PAGE_SIZE); + if (mtrr == 0xfe || mtrr == 0xff) + mtrr = MTRR_TYPE_WRBACK; + return mtrr; +} +EXPORT_SYMBOL_GPL(kvm_get_guest_memory_type); + +static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +{ + unsigned index; + struct hlist_head *bucket; + struct kvm_mmu_page *s; + struct hlist_node *node, *n; + + trace_kvm_mmu_unsync_page(sp); + index = kvm_page_table_hashfn(sp->gfn); + bucket = &vcpu->kvm->arch.mmu_page_hash[index]; + /* don't unsync if pagetable is shadowed with multiple roles */ + hlist_for_each_entry_safe(s, node, n, bucket, hash_link) { + if (s->gfn != sp->gfn || s->role.direct) + continue; + if (s->role.word != sp->role.word) + return 1; + } + ++vcpu->kvm->stat.mmu_unsync; + sp->unsync = 1; + + kvm_mmu_mark_parents_unsync(vcpu, sp); + + mmu_convert_notrap(sp); + return 0; } -struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva) +static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn, + bool can_unsync) { - struct page *page; - - gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva); - - if (gpa == UNMAPPED_GVA) - return NULL; + struct kvm_mmu_page *shadow; - down_read(¤t->mm->mmap_sem); - page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT); - up_read(¤t->mm->mmap_sem); - - return page; + shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); + if (shadow) { + if (shadow->role.level != PT_PAGE_TABLE_LEVEL) + return 1; + if (shadow->unsync) + return 0; + if (can_unsync && oos_shadow) + return kvm_unsync_page(vcpu, shadow); + return 1; + } + return 0; } -static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, - unsigned pt_access, unsigned pte_access, - int user_fault, int write_fault, int dirty, - int *ptwrite, int largepage, gfn_t gfn, - pfn_t pfn, bool speculative) +static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, + unsigned pte_access, int user_fault, + int write_fault, int dirty, int level, + gfn_t gfn, pfn_t pfn, bool speculative, + bool can_unsync) { u64 spte; - int was_rmapped = 0; - int was_writeble = is_writeble_pte(*shadow_pte); - - pgprintk("%s: spte %llx access %x write_fault %d" - " user_fault %d gfn %lx\n", - __func__, *shadow_pte, pt_access, - write_fault, user_fault, gfn); - - if (is_rmap_pte(*shadow_pte)) { - /* - * If we overwrite a PTE page pointer with a 2MB PMD, unlink - * the parent of the now unreachable PTE. - */ - if (largepage && !is_large_pte(*shadow_pte)) { - struct kvm_mmu_page *child; - u64 pte = *shadow_pte; - - child = page_header(pte & PT64_BASE_ADDR_MASK); - mmu_page_remove_parent_pte(child, shadow_pte); - } else if (pfn != spte_to_pfn(*shadow_pte)) { - pgprintk("hfn old %lx new %lx\n", - spte_to_pfn(*shadow_pte), pfn); - rmap_remove(vcpu->kvm, shadow_pte); - } else { - if (largepage) - was_rmapped = is_large_pte(*shadow_pte); - else - was_rmapped = 1; - } - } + int ret = 0; /* * We don't set the accessed bit, since we sometimes want to see @@ -1192,7 +1768,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, */ spte = shadow_base_present_pte | shadow_dirty_mask; if (!speculative) - pte_access |= PT_ACCESSED_MASK; + spte |= shadow_accessed_mask; if (!dirty) pte_access &= ~ACC_WRITE_MASK; if (pte_access & ACC_EXEC_MASK) @@ -1201,49 +1777,110 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, spte |= shadow_nx_mask; if (pte_access & ACC_USER_MASK) spte |= shadow_user_mask; - if (largepage) + if (level > PT_PAGE_TABLE_LEVEL) spte |= PT_PAGE_SIZE_MASK; + if (tdp_enabled) + spte |= kvm_x86_ops->get_mt_mask(vcpu, gfn, + kvm_is_mmio_pfn(pfn)); spte |= (u64)pfn << PAGE_SHIFT; if ((pte_access & ACC_WRITE_MASK) || (write_fault && !is_write_protection(vcpu) && !user_fault)) { - struct kvm_mmu_page *shadow; + + if (level > PT_PAGE_TABLE_LEVEL && + has_wrprotected_page(vcpu->kvm, gfn, level)) { + ret = 1; + spte = shadow_trap_nonpresent_pte; + goto set_pte; + } spte |= PT_WRITABLE_MASK; - shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn); - if (shadow || - (largepage && has_wrprotected_page(vcpu->kvm, gfn))) { + /* + * Optimization: for pte sync, if spte was writable the hash + * lookup is unnecessary (and expensive). Write protection + * is responsibility of mmu_get_page / kvm_sync_page. + * Same reasoning can be applied to dirty page accounting. + */ + if (!can_unsync && is_writeble_pte(*sptep)) + goto set_pte; + + if (mmu_need_write_protect(vcpu, gfn, can_unsync)) { pgprintk("%s: found shadow page for %lx, marking ro\n", __func__, gfn); + ret = 1; pte_access &= ~ACC_WRITE_MASK; - if (is_writeble_pte(spte)) { + if (is_writeble_pte(spte)) spte &= ~PT_WRITABLE_MASK; - kvm_x86_ops->tlb_flush(vcpu); - } - if (write_fault) - *ptwrite = 1; } } if (pte_access & ACC_WRITE_MASK) mark_page_dirty(vcpu->kvm, gfn); - pgprintk("%s: setting spte %llx\n", __func__, spte); +set_pte: + __set_spte(sptep, spte); + return ret; +} + +static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, + unsigned pt_access, unsigned pte_access, + int user_fault, int write_fault, int dirty, + int *ptwrite, int level, gfn_t gfn, + pfn_t pfn, bool speculative) +{ + int was_rmapped = 0; + int was_writeble = is_writeble_pte(*sptep); + int rmap_count; + + pgprintk("%s: spte %llx access %x write_fault %d" + " user_fault %d gfn %lx\n", + __func__, *sptep, pt_access, + write_fault, user_fault, gfn); + + if (is_rmap_spte(*sptep)) { + /* + * If we overwrite a PTE page pointer with a 2MB PMD, unlink + * the parent of the now unreachable PTE. + */ + if (level > PT_PAGE_TABLE_LEVEL && + !is_large_pte(*sptep)) { + struct kvm_mmu_page *child; + u64 pte = *sptep; + + child = page_header(pte & PT64_BASE_ADDR_MASK); + mmu_page_remove_parent_pte(child, sptep); + } else if (pfn != spte_to_pfn(*sptep)) { + pgprintk("hfn old %lx new %lx\n", + spte_to_pfn(*sptep), pfn); + rmap_remove(vcpu->kvm, sptep); + } else + was_rmapped = 1; + } + + if (set_spte(vcpu, sptep, pte_access, user_fault, write_fault, + dirty, level, gfn, pfn, speculative, true)) { + if (write_fault) + *ptwrite = 1; + kvm_x86_ops->tlb_flush(vcpu); + } + + pgprintk("%s: setting spte %llx\n", __func__, *sptep); pgprintk("instantiating %s PTE (%s) at %ld (%llx) addr %p\n", - (spte&PT_PAGE_SIZE_MASK)? "2MB" : "4kB", - (spte&PT_WRITABLE_MASK)?"RW":"R", gfn, spte, shadow_pte); - set_shadow_pte(shadow_pte, spte); - if (!was_rmapped && (spte & PT_PAGE_SIZE_MASK) - && (spte & PT_PRESENT_MASK)) + is_large_pte(*sptep)? "2MB" : "4kB", + *sptep & PT_PRESENT_MASK ?"RW":"R", gfn, + *sptep, sptep); + if (!was_rmapped && is_large_pte(*sptep)) ++vcpu->kvm->stat.lpages; - page_header_update_slot(vcpu->kvm, shadow_pte, gfn); + page_header_update_slot(vcpu->kvm, sptep, gfn); if (!was_rmapped) { - rmap_add(vcpu, shadow_pte, gfn, largepage); - if (!is_rmap_pte(*shadow_pte)) + rmap_count = rmap_add(vcpu, sptep, gfn); + if (!is_rmap_spte(*sptep)) kvm_release_pfn_clean(pfn); + if (rmap_count > RMAP_RECYCLE_THRESHOLD) + rmap_recycle(vcpu, sptep, gfn); } else { if (was_writeble) kvm_release_pfn_dirty(pfn); @@ -1251,7 +1888,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, kvm_release_pfn_clean(pfn); } if (speculative) { - vcpu->arch.last_pte_updated = shadow_pte; + vcpu->arch.last_pte_updated = sptep; vcpu->arch.last_pte_gfn = gfn; } } @@ -1260,85 +1897,64 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) { } -struct direct_shadow_walk { - struct kvm_shadow_walk walker; - pfn_t pfn; - int write; - int largepage; - int pt_write; -}; - -static int direct_map_entry(struct kvm_shadow_walk *_walk, - struct kvm_vcpu *vcpu, - gva_t addr, u64 *sptep, int level) +static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, + int level, gfn_t gfn, pfn_t pfn) { - struct direct_shadow_walk *walk = - container_of(_walk, struct direct_shadow_walk, walker); + struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; + int pt_write = 0; gfn_t pseudo_gfn; - gfn_t gfn = addr >> PAGE_SHIFT; - - if (level == PT_PAGE_TABLE_LEVEL - || (walk->largepage && level == PT_DIRECTORY_LEVEL)) { - mmu_set_spte(vcpu, sptep, ACC_ALL, ACC_ALL, - 0, walk->write, 1, &walk->pt_write, - walk->largepage, gfn, walk->pfn, false); - return 1; - } - if (*sptep == shadow_trap_nonpresent_pte) { - pseudo_gfn = (addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT; - sp = kvm_mmu_get_page(vcpu, pseudo_gfn, addr, level - 1, - 1, ACC_ALL, sptep); - if (!sp) { - pgprintk("nonpaging_map: ENOMEM\n"); - kvm_release_pfn_clean(walk->pfn); - return -ENOMEM; + for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) { + if (iterator.level == level) { + mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL, + 0, write, 1, &pt_write, + level, gfn, pfn, false); + ++vcpu->stat.pf_fixed; + break; } - set_shadow_pte(sptep, - __pa(sp->spt) - | PT_PRESENT_MASK | PT_WRITABLE_MASK - | shadow_user_mask | shadow_x_mask); - } - return 0; -} + if (*iterator.sptep == shadow_trap_nonpresent_pte) { + pseudo_gfn = (iterator.addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT; + sp = kvm_mmu_get_page(vcpu, pseudo_gfn, iterator.addr, + iterator.level - 1, + 1, ACC_ALL, iterator.sptep); + if (!sp) { + pgprintk("nonpaging_map: ENOMEM\n"); + kvm_release_pfn_clean(pfn); + return -ENOMEM; + } -static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, - int largepage, gfn_t gfn, pfn_t pfn) -{ - int r; - struct direct_shadow_walk walker = { - .walker = { .entry = direct_map_entry, }, - .pfn = pfn, - .largepage = largepage, - .write = write, - .pt_write = 0, - }; - - r = walk_shadow(&walker.walker, vcpu, (gva_t)gfn << PAGE_SHIFT); - if (r < 0) - return r; - return walker.pt_write; + __set_spte(iterator.sptep, + __pa(sp->spt) + | PT_PRESENT_MASK | PT_WRITABLE_MASK + | shadow_user_mask | shadow_x_mask); + } + } + return pt_write; } static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) { int r; - int largepage = 0; + int level; pfn_t pfn; unsigned long mmu_seq; - down_read(¤t->mm->mmap_sem); - if (is_largepage_backed(vcpu, gfn & ~(KVM_PAGES_PER_HPAGE-1))) { - gfn &= ~(KVM_PAGES_PER_HPAGE-1); - largepage = 1; - } + level = mapping_level(vcpu, gfn); + + /* + * This path builds a PAE pagetable - so we can map 2mb pages at + * maximum. Therefore check if the level is larger than that. + */ + if (level > PT_DIRECTORY_LEVEL) + level = PT_DIRECTORY_LEVEL; + + gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); /* mmio */ if (is_error_pfn(pfn)) { @@ -1350,7 +1966,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); - r = __direct_map(vcpu, v, write, largepage, gfn, pfn); + r = __direct_map(vcpu, v, write, level, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); @@ -1398,12 +2014,25 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu) vcpu->arch.mmu.root_hpa = INVALID_PAGE; } -static void mmu_alloc_roots(struct kvm_vcpu *vcpu) +static int mmu_check_root(struct kvm_vcpu *vcpu, gfn_t root_gfn) +{ + int ret = 0; + + if (!kvm_is_visible_gfn(vcpu->kvm, root_gfn)) { + set_bit(KVM_REQ_TRIPLE_FAULT, &vcpu->requests); + ret = 1; + } + + return ret; +} + +static int mmu_alloc_roots(struct kvm_vcpu *vcpu) { int i; gfn_t root_gfn; struct kvm_mmu_page *sp; - int metaphysical = 0; + int direct = 0; + u64 pdptr; root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT; @@ -1412,38 +2041,75 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) ASSERT(!VALID_PAGE(root)); if (tdp_enabled) - metaphysical = 1; + direct = 1; + if (mmu_check_root(vcpu, root_gfn)) + return 1; sp = kvm_mmu_get_page(vcpu, root_gfn, 0, - PT64_ROOT_LEVEL, metaphysical, + PT64_ROOT_LEVEL, direct, ACC_ALL, NULL); root = __pa(sp->spt); ++sp->root_count; vcpu->arch.mmu.root_hpa = root; - return; + return 0; } - metaphysical = !is_paging(vcpu); + direct = !is_paging(vcpu); if (tdp_enabled) - metaphysical = 1; + direct = 1; for (i = 0; i < 4; ++i) { hpa_t root = vcpu->arch.mmu.pae_root[i]; ASSERT(!VALID_PAGE(root)); if (vcpu->arch.mmu.root_level == PT32E_ROOT_LEVEL) { - if (!is_present_pte(vcpu->arch.pdptrs[i])) { + pdptr = kvm_pdptr_read(vcpu, i); + if (!is_present_gpte(pdptr)) { vcpu->arch.mmu.pae_root[i] = 0; continue; } - root_gfn = vcpu->arch.pdptrs[i] >> PAGE_SHIFT; + root_gfn = pdptr >> PAGE_SHIFT; } else if (vcpu->arch.mmu.root_level == 0) root_gfn = 0; + if (mmu_check_root(vcpu, root_gfn)) + return 1; sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30, - PT32_ROOT_LEVEL, metaphysical, + PT32_ROOT_LEVEL, direct, ACC_ALL, NULL); root = __pa(sp->spt); ++sp->root_count; vcpu->arch.mmu.pae_root[i] = root | PT_PRESENT_MASK; } vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root); + return 0; +} + +static void mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_mmu_page *sp; + + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return; + if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { + hpa_t root = vcpu->arch.mmu.root_hpa; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + return; + } + for (i = 0; i < 4; ++i) { + hpa_t root = vcpu->arch.mmu.pae_root[i]; + + if (root && VALID_PAGE(root)) { + root &= PT64_BASE_ADDR_MASK; + sp = page_header(root); + mmu_sync_children(vcpu, sp); + } + } +} + +void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu) +{ + spin_lock(&vcpu->kvm->mmu_lock); + mmu_sync_roots(vcpu); + spin_unlock(&vcpu->kvm->mmu_lock); } static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr) @@ -1476,7 +2142,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, { pfn_t pfn; int r; - int largepage = 0; + int level; gfn_t gfn = gpa >> PAGE_SHIFT; unsigned long mmu_seq; @@ -1487,15 +2153,13 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, if (r) return r; - down_read(¤t->mm->mmap_sem); - if (is_largepage_backed(vcpu, gfn & ~(KVM_PAGES_PER_HPAGE-1))) { - gfn &= ~(KVM_PAGES_PER_HPAGE-1); - largepage = 1; - } + level = mapping_level(vcpu, gfn); + + gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); + mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); if (is_error_pfn(pfn)) { kvm_release_pfn_clean(pfn); return 1; @@ -1505,7 +2169,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, goto out_unlock; kvm_mmu_free_some_pages(vcpu); r = __direct_map(vcpu, gpa, error_code & PFERR_WRITE_MASK, - largepage, gfn, pfn); + level, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); return r; @@ -1530,6 +2194,8 @@ static int nonpaging_init_context(struct kvm_vcpu *vcpu) context->gva_to_gpa = nonpaging_gva_to_gpa; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->root_level = 0; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1560,6 +2226,14 @@ static void paging_free(struct kvm_vcpu *vcpu) nonpaging_free(vcpu); } +static bool is_rsvd_bits_set(struct kvm_vcpu *vcpu, u64 gpte, int level) +{ + int bit7; + + bit7 = (gpte >> 7) & 1; + return (gpte & vcpu->arch.mmu.rsvd_bits_mask[bit7][level-1]) != 0; +} + #define PTTYPE 64 #include "paging_tmpl.h" #undef PTTYPE @@ -1568,6 +2242,61 @@ static void paging_free(struct kvm_vcpu *vcpu) #include "paging_tmpl.h" #undef PTTYPE +static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level) +{ + struct kvm_mmu *context = &vcpu->arch.mmu; + int maxphyaddr = cpuid_maxphyaddr(vcpu); + u64 exb_bit_rsvd = 0; + + if (!is_nx(vcpu)) + exb_bit_rsvd = rsvd_bits(63, 63); + switch (level) { + case PT32_ROOT_LEVEL: + /* no rsvd bits for 2 level 4K page table entries */ + context->rsvd_bits_mask[0][1] = 0; + context->rsvd_bits_mask[0][0] = 0; + if (is_cpuid_PSE36()) + /* 36bits PSE 4MB page */ + context->rsvd_bits_mask[1][1] = rsvd_bits(17, 21); + else + /* 32 bits PSE 4MB page */ + context->rsvd_bits_mask[1][1] = rsvd_bits(13, 21); + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; + break; + case PT32E_ROOT_LEVEL: + context->rsvd_bits_mask[0][2] = + rsvd_bits(maxphyaddr, 63) | + rsvd_bits(7, 8) | rsvd_bits(1, 2); /* PDPTE */ + context->rsvd_bits_mask[0][1] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 62); /* PDE */ + context->rsvd_bits_mask[0][0] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 62); /* PTE */ + context->rsvd_bits_mask[1][1] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 62) | + rsvd_bits(13, 20); /* large page */ + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; + break; + case PT64_ROOT_LEVEL: + context->rsvd_bits_mask[0][3] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51) | rsvd_bits(7, 8); + context->rsvd_bits_mask[0][2] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51) | rsvd_bits(7, 8); + context->rsvd_bits_mask[0][1] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51); + context->rsvd_bits_mask[0][0] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51); + context->rsvd_bits_mask[1][3] = context->rsvd_bits_mask[0][3]; + context->rsvd_bits_mask[1][2] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51) | + rsvd_bits(13, 29); + context->rsvd_bits_mask[1][1] = exb_bit_rsvd | + rsvd_bits(maxphyaddr, 51) | + rsvd_bits(13, 20); /* large page */ + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; + break; + } +} + static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) { struct kvm_mmu *context = &vcpu->arch.mmu; @@ -1577,6 +2306,8 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) context->page_fault = paging64_page_fault; context->gva_to_gpa = paging64_gva_to_gpa; context->prefetch_page = paging64_prefetch_page; + context->sync_page = paging64_sync_page; + context->invlpg = paging64_invlpg; context->free = paging_free; context->root_level = level; context->shadow_root_level = level; @@ -1586,6 +2317,7 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level) static int paging64_init_context(struct kvm_vcpu *vcpu) { + reset_rsvds_bits_mask(vcpu, PT64_ROOT_LEVEL); return paging64_init_context_common(vcpu, PT64_ROOT_LEVEL); } @@ -1593,11 +2325,14 @@ static int paging32_init_context(struct kvm_vcpu *vcpu) { struct kvm_mmu *context = &vcpu->arch.mmu; + reset_rsvds_bits_mask(vcpu, PT32_ROOT_LEVEL); context->new_cr3 = paging_new_cr3; context->page_fault = paging32_page_fault; context->gva_to_gpa = paging32_gva_to_gpa; context->free = paging_free; context->prefetch_page = paging32_prefetch_page; + context->sync_page = paging32_sync_page; + context->invlpg = paging32_invlpg; context->root_level = PT32_ROOT_LEVEL; context->shadow_root_level = PT32E_ROOT_LEVEL; context->root_hpa = INVALID_PAGE; @@ -1606,6 +2341,7 @@ static int paging32_init_context(struct kvm_vcpu *vcpu) static int paging32E_init_context(struct kvm_vcpu *vcpu) { + reset_rsvds_bits_mask(vcpu, PT32E_ROOT_LEVEL); return paging64_init_context_common(vcpu, PT32E_ROOT_LEVEL); } @@ -1617,6 +2353,8 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->page_fault = tdp_page_fault; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; + context->sync_page = nonpaging_sync_page; + context->invlpg = nonpaging_invlpg; context->shadow_root_level = kvm_x86_ops->get_tdp_level(); context->root_hpa = INVALID_PAGE; @@ -1624,12 +2362,15 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->gva_to_gpa = nonpaging_gva_to_gpa; context->root_level = 0; } else if (is_long_mode(vcpu)) { + reset_rsvds_bits_mask(vcpu, PT64_ROOT_LEVEL); context->gva_to_gpa = paging64_gva_to_gpa; context->root_level = PT64_ROOT_LEVEL; } else if (is_pae(vcpu)) { + reset_rsvds_bits_mask(vcpu, PT32E_ROOT_LEVEL); context->gva_to_gpa = paging64_gva_to_gpa; context->root_level = PT32E_ROOT_LEVEL; } else { + reset_rsvds_bits_mask(vcpu, PT32_ROOT_LEVEL); context->gva_to_gpa = paging32_gva_to_gpa; context->root_level = PT32_ROOT_LEVEL; } @@ -1639,17 +2380,23 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) static int init_kvm_softmmu(struct kvm_vcpu *vcpu) { + int r; + ASSERT(vcpu); ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa)); if (!is_paging(vcpu)) - return nonpaging_init_context(vcpu); + r = nonpaging_init_context(vcpu); else if (is_long_mode(vcpu)) - return paging64_init_context(vcpu); + r = paging64_init_context(vcpu); else if (is_pae(vcpu)) - return paging32E_init_context(vcpu); + r = paging32E_init_context(vcpu); else - return paging32_init_context(vcpu); + r = paging32_init_context(vcpu); + + vcpu->arch.mmu.base_role.glevels = vcpu->arch.mmu.root_level; + + return r; } static int init_kvm_mmu(struct kvm_vcpu *vcpu) @@ -1687,10 +2434,13 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu) goto out; spin_lock(&vcpu->kvm->mmu_lock); kvm_mmu_free_some_pages(vcpu); - mmu_alloc_roots(vcpu); + r = mmu_alloc_roots(vcpu); + mmu_sync_roots(vcpu); spin_unlock(&vcpu->kvm->mmu_lock); + if (r) + goto out; + /* set_cr3() should ensure TLB has been flushed */ kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa); - kvm_mmu_flush_tlb(vcpu); out: return r; } @@ -1710,15 +2460,14 @@ static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu, pte = *spte; if (is_shadow_present_pte(pte)) { - if (sp->role.level == PT_PAGE_TABLE_LEVEL || - is_large_pte(pte)) + if (is_last_spte(pte, sp->role.level)) rmap_remove(vcpu->kvm, spte); else { child = page_header(pte & PT64_BASE_ADDR_MASK); mmu_page_remove_parent_pte(child, spte); } } - set_shadow_pte(spte, shadow_trap_nonpresent_pte); + __set_spte(spte, shadow_trap_nonpresent_pte); if (is_large_pte(pte)) --vcpu->kvm->stat.lpages; } @@ -1729,11 +2478,8 @@ static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu, const void *new) { if (sp->role.level != PT_PAGE_TABLE_LEVEL) { - if (!vcpu->arch.update_pte.largepage || - sp->role.glevels == PT32_ROOT_LEVEL) { - ++vcpu->kvm->stat.mmu_pde_zapped; - return; - } + ++vcpu->kvm->stat.mmu_pde_zapped; + return; } ++vcpu->kvm->stat.mmu_pte_updated; @@ -1779,8 +2525,6 @@ static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, u64 gpte = 0; pfn_t pfn; - vcpu->arch.update_pte.largepage = 0; - if (bytes != 4 && bytes != 8) return; @@ -1804,19 +2548,13 @@ static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, if ((bytes == 4) && (gpa % 4 == 0)) memcpy((void *)&gpte, new, 4); } - if (!is_present_pte(gpte)) + if (!is_present_gpte(gpte)) return; gfn = (gpte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT; - down_read(¤t->mm->mmap_sem); - if (is_large_pte(gpte) && is_largepage_backed(vcpu, gfn)) { - gfn &= ~(KVM_PAGES_PER_HPAGE-1); - vcpu->arch.update_pte.largepage = 1; - } vcpu->arch.update_pte.mmu_seq = vcpu->kvm->mmu_notifier_seq; - /* implicit mb(), we'll read before PT lock is unlocked */ + smp_rmb(); pfn = gfn_to_pfn(vcpu->kvm, gfn); - up_read(¤t->mm->mmap_sem); if (is_error_pfn(pfn)) { kvm_release_pfn_clean(pfn); @@ -1839,7 +2577,8 @@ static void kvm_mmu_access_page(struct kvm_vcpu *vcpu, gfn_t gfn) } void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, - const u8 *new, int bytes) + const u8 *new, int bytes, + bool guest_initiated) { gfn_t gfn = gpa >> PAGE_SHIFT; struct kvm_mmu_page *sp; @@ -1865,20 +2604,22 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, kvm_mmu_free_some_pages(vcpu); ++vcpu->kvm->stat.mmu_pte_write; kvm_mmu_audit(vcpu, "pre pte write"); - if (gfn == vcpu->arch.last_pt_write_gfn - && !last_updated_pte_accessed(vcpu)) { - ++vcpu->arch.last_pt_write_count; - if (vcpu->arch.last_pt_write_count >= 3) - flooded = 1; - } else { - vcpu->arch.last_pt_write_gfn = gfn; - vcpu->arch.last_pt_write_count = 1; - vcpu->arch.last_pte_updated = NULL; + if (guest_initiated) { + if (gfn == vcpu->arch.last_pt_write_gfn + && !last_updated_pte_accessed(vcpu)) { + ++vcpu->arch.last_pt_write_count; + if (vcpu->arch.last_pt_write_count >= 3) + flooded = 1; + } else { + vcpu->arch.last_pt_write_gfn = gfn; + vcpu->arch.last_pt_write_count = 1; + vcpu->arch.last_pte_updated = NULL; + } } index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) { - if (sp->gfn != gfn || sp->role.metaphysical || sp->role.invalid) + if (sp->gfn != gfn || sp->role.direct || sp->role.invalid) continue; pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8; misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1); @@ -1896,7 +2637,8 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, */ pgprintk("misaligned: gpa %llx bytes %d role %x\n", gpa, bytes, sp->role.word); - kvm_mmu_zap_page(vcpu->kvm, sp); + if (kvm_mmu_zap_page(vcpu->kvm, sp)) + n = bucket->first; ++vcpu->kvm->stat.mmu_flooded; continue; } @@ -2000,8 +2742,9 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code) ++vcpu->stat.mmio_exits; return 0; case EMULATE_FAIL: - kvm_report_emulation_failure(vcpu, "pagetable"); - return 1; + vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; + return 0; default: BUG(); } @@ -2010,6 +2753,14 @@ out: } EXPORT_SYMBOL_GPL(kvm_mmu_page_fault); +void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva) +{ + vcpu->arch.mmu.invlpg(vcpu, gva); + kvm_mmu_flush_tlb(vcpu); + ++vcpu->stat.invlpg; +} +EXPORT_SYMBOL_GPL(kvm_mmu_invlpg); + void kvm_enable_tdp(void) { tdp_enabled = true; @@ -2024,14 +2775,6 @@ EXPORT_SYMBOL_GPL(kvm_disable_tdp); static void free_mmu_pages(struct kvm_vcpu *vcpu) { - struct kvm_mmu_page *sp; - - while (!list_empty(&vcpu->kvm->arch.active_mmu_pages)) { - sp = container_of(vcpu->kvm->arch.active_mmu_pages.next, - struct kvm_mmu_page, link); - kvm_mmu_zap_page(vcpu->kvm, sp); - cond_resched(); - } free_page((unsigned long)vcpu->arch.mmu.pae_root); } @@ -2042,12 +2785,14 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu) ASSERT(vcpu); + spin_lock(&vcpu->kvm->mmu_lock); if (vcpu->kvm->arch.n_requested_mmu_pages) vcpu->kvm->arch.n_free_mmu_pages = vcpu->kvm->arch.n_requested_mmu_pages; else vcpu->kvm->arch.n_free_mmu_pages = vcpu->kvm->arch.n_alloc_mmu_pages; + spin_unlock(&vcpu->kvm->mmu_lock); /* * When emulating 32-bit mode, cr3 is only 32 bits even on x86_64. * Therefore we need to allocate shadow page tables in the first @@ -2100,7 +2845,7 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) int i; u64 *pt; - if (!test_bit(slot, &sp->slot_bitmap)) + if (!test_bit(slot, sp->slot_bitmap)) continue; pt = sp->spt; @@ -2109,6 +2854,7 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) if (pt[i] & PT_WRITABLE_MASK) pt[i] &= ~PT_WRITABLE_MASK; } + kvm_flush_remote_tlbs(kvm); } void kvm_mmu_zap_all(struct kvm *kvm) @@ -2117,7 +2863,9 @@ void kvm_mmu_zap_all(struct kvm *kvm) spin_lock(&kvm->mmu_lock); list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link) - kvm_mmu_zap_page(kvm, sp); + if (kvm_mmu_zap_page(kvm, sp)) + node = container_of(kvm->arch.active_mmu_pages.next, + struct kvm_mmu_page, link); spin_unlock(&kvm->mmu_lock); kvm_flush_remote_tlbs(kvm); @@ -2278,7 +3026,7 @@ static int kvm_pv_mmu_write(struct kvm_vcpu *vcpu, static int kvm_pv_mmu_flush_tlb(struct kvm_vcpu *vcpu) { - kvm_x86_ops->tlb_flush(vcpu); + kvm_set_cr3(vcpu, vcpu->arch.cr3); return 1; } @@ -2356,6 +3104,24 @@ out: return r; } +int kvm_mmu_get_spte_hierarchy(struct kvm_vcpu *vcpu, u64 addr, u64 sptes[4]) +{ + struct kvm_shadow_walk_iterator iterator; + int nr_sptes = 0; + + spin_lock(&vcpu->kvm->mmu_lock); + for_each_shadow_entry(vcpu, addr, iterator) { + sptes[iterator.level-1] = *iterator.sptep; + nr_sptes++; + if (!is_shadow_present_pte(*iterator.sptep)) + break; + } + spin_unlock(&vcpu->kvm->mmu_lock); + + return nr_sptes; +} +EXPORT_SYMBOL_GPL(kvm_mmu_get_spte_hierarchy); + #ifdef AUDIT static const char *audit_msg; @@ -2368,6 +3134,54 @@ static gva_t canonicalize(gva_t gva) return gva; } + +typedef void (*inspect_spte_fn) (struct kvm *kvm, struct kvm_mmu_page *sp, + u64 *sptep); + +static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp, + inspect_spte_fn fn) +{ + int i; + + for (i = 0; i < PT64_ENT_PER_PAGE; ++i) { + u64 ent = sp->spt[i]; + + if (is_shadow_present_pte(ent)) { + if (!is_last_spte(ent, sp->role.level)) { + struct kvm_mmu_page *child; + child = page_header(ent & PT64_BASE_ADDR_MASK); + __mmu_spte_walk(kvm, child, fn); + } else + fn(kvm, sp, &sp->spt[i]); + } + } +} + +static void mmu_spte_walk(struct kvm_vcpu *vcpu, inspect_spte_fn fn) +{ + int i; + struct kvm_mmu_page *sp; + + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return; + if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { + hpa_t root = vcpu->arch.mmu.root_hpa; + sp = page_header(root); + __mmu_spte_walk(vcpu->kvm, sp, fn); + return; + } + for (i = 0; i < 4; ++i) { + hpa_t root = vcpu->arch.mmu.pae_root[i]; + + if (root && VALID_PAGE(root)) { + root &= PT64_BASE_ADDR_MASK; + sp = page_header(root); + __mmu_spte_walk(vcpu->kvm, sp, fn); + } + } + return; +} + static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte, gva_t va, int level) { @@ -2382,17 +3196,18 @@ static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte, continue; va = canonicalize(va); - if (level > 1) { - if (ent == shadow_notrap_nonpresent_pte) - printk(KERN_ERR "audit: (%s) nontrapping pte" - " in nonleaf level: levels %d gva %lx" - " level %d pte %llx\n", audit_msg, - vcpu->arch.mmu.root_level, va, level, ent); - + if (is_shadow_present_pte(ent) && !is_last_spte(ent, level)) audit_mappings_page(vcpu, ent, va, level - 1); - } else { + else { gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va); - hpa_t hpa = (hpa_t)gpa_to_pfn(vcpu, gpa) << PAGE_SHIFT; + gfn_t gfn = gpa >> PAGE_SHIFT; + pfn_t pfn = gfn_to_pfn(vcpu->kvm, gfn); + hpa_t hpa = (hpa_t)pfn << PAGE_SHIFT; + + if (is_error_pfn(pfn)) { + kvm_release_pfn_clean(pfn); + continue; + } if (is_shadow_present_pte(ent) && (ent & PT64_BASE_ADDR_MASK) != hpa) @@ -2447,7 +3262,7 @@ static int count_rmaps(struct kvm_vcpu *vcpu) d = (struct kvm_rmap_desc *)(*rmapp & ~1ul); while (d) { for (k = 0; k < RMAP_EXT; ++k) - if (d->shadow_ptes[k]) + if (d->sptes[k]) ++nmaps; else break; @@ -2458,9 +3273,48 @@ static int count_rmaps(struct kvm_vcpu *vcpu) return nmaps; } -static int count_writable_mappings(struct kvm_vcpu *vcpu) +void inspect_spte_has_rmap(struct kvm *kvm, struct kvm_mmu_page *sp, u64 *sptep) +{ + unsigned long *rmapp; + struct kvm_mmu_page *rev_sp; + gfn_t gfn; + + if (*sptep & PT_WRITABLE_MASK) { + rev_sp = page_header(__pa(sptep)); + gfn = rev_sp->gfns[sptep - rev_sp->spt]; + + if (!gfn_to_memslot(kvm, gfn)) { + if (!printk_ratelimit()) + return; + printk(KERN_ERR "%s: no memslot for gfn %ld\n", + audit_msg, gfn); + printk(KERN_ERR "%s: index %ld of sp (gfn=%lx)\n", + audit_msg, sptep - rev_sp->spt, + rev_sp->gfn); + dump_stack(); + return; + } + + rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], + is_large_pte(*sptep)); + if (!*rmapp) { + if (!printk_ratelimit()) + return; + printk(KERN_ERR "%s: no rmap for writable spte %llx\n", + audit_msg, *sptep); + dump_stack(); + } + } + +} + +void audit_writable_sptes_have_rmaps(struct kvm_vcpu *vcpu) +{ + mmu_spte_walk(vcpu, inspect_spte_has_rmap); +} + +static void check_writable_mappings_rmap(struct kvm_vcpu *vcpu) { - int nmaps = 0; struct kvm_mmu_page *sp; int i; @@ -2477,20 +3331,16 @@ static int count_writable_mappings(struct kvm_vcpu *vcpu) continue; if (!(ent & PT_WRITABLE_MASK)) continue; - ++nmaps; + inspect_spte_has_rmap(vcpu->kvm, sp, &pt[i]); } } - return nmaps; + return; } static void audit_rmap(struct kvm_vcpu *vcpu) { - int n_rmap = count_rmaps(vcpu); - int n_actual = count_writable_mappings(vcpu); - - if (n_rmap != n_actual) - printk(KERN_ERR "%s: (%s) rmap %d actual %d\n", - __func__, audit_msg, n_rmap, n_actual); + check_writable_mappings_rmap(vcpu); + count_rmaps(vcpu); } static void audit_write_protection(struct kvm_vcpu *vcpu) @@ -2498,20 +3348,28 @@ static void audit_write_protection(struct kvm_vcpu *vcpu) struct kvm_mmu_page *sp; struct kvm_memory_slot *slot; unsigned long *rmapp; + u64 *spte; gfn_t gfn; list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) { - if (sp->role.metaphysical) + if (sp->role.direct) + continue; + if (sp->unsync) continue; - slot = gfn_to_memslot(vcpu->kvm, sp->gfn); gfn = unalias_gfn(vcpu->kvm, sp->gfn); + slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn); rmapp = &slot->rmap[gfn - slot->base_gfn]; - if (*rmapp) - printk(KERN_ERR "%s: (%s) shadow page has writable" - " mappings: gfn %lx role %x\n", + + spte = rmap_next(vcpu->kvm, rmapp, NULL); + while (spte) { + if (*spte & PT_WRITABLE_MASK) + printk(KERN_ERR "%s: (%s) shadow page has " + "writable mappings: gfn %lx role %x\n", __func__, audit_msg, sp->gfn, sp->role.word); + spte = rmap_next(vcpu->kvm, rmapp, spte); + } } } @@ -2523,7 +3381,9 @@ static void kvm_mmu_audit(struct kvm_vcpu *vcpu, const char *msg) audit_msg = msg; audit_rmap(vcpu); audit_write_protection(vcpu); - audit_mappings(vcpu); + if (strcmp("pre pte write", audit_msg) != 0) + audit_mappings(vcpu); + audit_writable_sptes_have_rmaps(vcpu); dbg = olddbg; }