#include "vmx.h"
#include "kvm.h"
+#include "x86.h"
#include <linux/types.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/module.h>
+#include <linux/swap.h>
#include <asm/page.h>
#include <asm/cmpxchg.h>
+#include <asm/io.h>
#undef MMU_DEBUG
#define PT_PAGE_SIZE_MASK (1ULL << 7)
#define PT_PAT_MASK (1ULL << 7)
#define PT_GLOBAL_MASK (1ULL << 8)
-#define PT64_NX_MASK (1ULL << 63)
+#define PT64_NX_SHIFT 63
+#define PT64_NX_MASK (1ULL << PT64_NX_SHIFT)
#define PT_PAT_SHIFT 7
#define PT_DIR_PAT_SHIFT 12
#define PT32_DIR_PSE36_SIZE 4
#define PT32_DIR_PSE36_SHIFT 13
-#define PT32_DIR_PSE36_MASK (((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT)
+#define PT32_DIR_PSE36_MASK \
+ (((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT)
#define PT_FIRST_AVAIL_BITS_SHIFT 9
#define PT64_LEVEL_BITS 9
#define PT64_LEVEL_SHIFT(level) \
- ( PAGE_SHIFT + (level - 1) * PT64_LEVEL_BITS )
+ (PAGE_SHIFT + (level - 1) * PT64_LEVEL_BITS)
#define PT64_LEVEL_MASK(level) \
(((1ULL << PT64_LEVEL_BITS) - 1) << PT64_LEVEL_SHIFT(level))
#define PT32_LEVEL_BITS 10
#define PT32_LEVEL_SHIFT(level) \
- ( PAGE_SHIFT + (level - 1) * PT32_LEVEL_BITS )
+ (PAGE_SHIFT + (level - 1) * PT32_LEVEL_BITS)
#define PT32_LEVEL_MASK(level) \
(((1ULL << PT32_LEVEL_BITS) - 1) << PT32_LEVEL_SHIFT(level))
#define PT32_DIR_BASE_ADDR_MASK \
(PAGE_MASK & ~((1ULL << (PAGE_SHIFT + PT32_LEVEL_BITS)) - 1))
+#define PT64_PERM_MASK (PT_PRESENT_MASK | PT_WRITABLE_MASK | PT_USER_MASK \
+ | PT64_NX_MASK)
#define PFERR_PRESENT_MASK (1U << 0)
#define PFERR_WRITE_MASK (1U << 1)
#define RMAP_EXT 4
+#define ACC_EXEC_MASK 1
+#define ACC_WRITE_MASK PT_WRITABLE_MASK
+#define ACC_USER_MASK PT_USER_MASK
+#define ACC_ALL (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK)
+
struct kvm_rmap_desc {
u64 *shadow_ptes[RMAP_EXT];
struct kvm_rmap_desc *more;
static struct kmem_cache *rmap_desc_cache;
static struct kmem_cache *mmu_page_header_cache;
+static u64 __read_mostly shadow_trap_nonpresent_pte;
+static u64 __read_mostly shadow_notrap_nonpresent_pte;
+
+void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte)
+{
+ shadow_trap_nonpresent_pte = trap_pte;
+ shadow_notrap_nonpresent_pte = notrap_pte;
+}
+EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes);
+
static int is_write_protection(struct kvm_vcpu *vcpu)
{
- return vcpu->cr0 & CR0_WP_MASK;
+ return vcpu->arch.cr0 & X86_CR0_WP;
}
static int is_cpuid_PSE36(void)
static int is_nx(struct kvm_vcpu *vcpu)
{
- return vcpu->shadow_efer & EFER_NX;
+ 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)
+{
+ pte &= ~PT_SHADOW_IO_MARK;
+ return pte != shadow_trap_nonpresent_pte
+ && pte != shadow_notrap_nonpresent_pte;
+}
+
static int is_writeble_pte(unsigned long pte)
{
return pte & PT_WRITABLE_MASK;
}
+static int is_dirty_pte(unsigned long pte)
+{
+ return pte & PT_DIRTY_MASK;
+}
+
static int is_io_pte(unsigned long pte)
{
return pte & PT_SHADOW_IO_MARK;
static int is_rmap_pte(u64 pte)
{
- return (pte & (PT_WRITABLE_MASK | PT_PRESENT_MASK))
- == (PT_WRITABLE_MASK | PT_PRESENT_MASK);
+ return pte != shadow_trap_nonpresent_pte
+ && pte != shadow_notrap_nonpresent_pte;
+}
+
+static gfn_t pse36_gfn_delta(u32 gpte)
+{
+ int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
+
+ return (gpte & PT32_DIR_PSE36_MASK) << shift;
}
static void set_shadow_pte(u64 *sptep, u64 spte)
}
static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache,
- struct kmem_cache *base_cache, int min,
- gfp_t gfp_flags)
+ struct kmem_cache *base_cache, int min)
{
void *obj;
if (cache->nobjs >= min)
return 0;
while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
- obj = kmem_cache_zalloc(base_cache, gfp_flags);
+ obj = kmem_cache_zalloc(base_cache, GFP_KERNEL);
if (!obj)
return -ENOMEM;
cache->objects[cache->nobjs++] = obj;
}
static int mmu_topup_memory_cache_page(struct kvm_mmu_memory_cache *cache,
- int min, gfp_t gfp_flags)
+ int min)
{
struct page *page;
if (cache->nobjs >= min)
return 0;
while (cache->nobjs < ARRAY_SIZE(cache->objects)) {
- page = alloc_page(gfp_flags);
+ page = alloc_page(GFP_KERNEL);
if (!page)
return -ENOMEM;
set_page_private(page, 0);
free_page((unsigned long)mc->objects[--mc->nobjs]);
}
-static int __mmu_topup_memory_caches(struct kvm_vcpu *vcpu, gfp_t gfp_flags)
+static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
{
int r;
- r = mmu_topup_memory_cache(&vcpu->mmu_pte_chain_cache,
- pte_chain_cache, 4, gfp_flags);
+ kvm_mmu_free_some_pages(vcpu);
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_pte_chain_cache,
+ pte_chain_cache, 4);
if (r)
goto out;
- r = mmu_topup_memory_cache(&vcpu->mmu_rmap_desc_cache,
- rmap_desc_cache, 1, gfp_flags);
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache,
+ rmap_desc_cache, 1);
if (r)
goto out;
- r = mmu_topup_memory_cache_page(&vcpu->mmu_page_cache, 4, gfp_flags);
+ r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8);
if (r)
goto out;
- r = mmu_topup_memory_cache(&vcpu->mmu_page_header_cache,
- mmu_page_header_cache, 4, gfp_flags);
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_page_header_cache,
+ mmu_page_header_cache, 4);
out:
return r;
}
-static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
-{
- int r;
-
- r = __mmu_topup_memory_caches(vcpu, GFP_NOWAIT);
- kvm_mmu_free_some_pages(vcpu);
- if (r < 0) {
- spin_unlock(&vcpu->kvm->lock);
- kvm_arch_ops->vcpu_put(vcpu);
- r = __mmu_topup_memory_caches(vcpu, GFP_KERNEL);
- kvm_arch_ops->vcpu_load(vcpu);
- spin_lock(&vcpu->kvm->lock);
- kvm_mmu_free_some_pages(vcpu);
- }
- return r;
-}
-
static void mmu_free_memory_caches(struct kvm_vcpu *vcpu)
{
- mmu_free_memory_cache(&vcpu->mmu_pte_chain_cache);
- mmu_free_memory_cache(&vcpu->mmu_rmap_desc_cache);
- mmu_free_memory_cache_page(&vcpu->mmu_page_cache);
- mmu_free_memory_cache(&vcpu->mmu_page_header_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_pte_chain_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_rmap_desc_cache);
+ mmu_free_memory_cache_page(&vcpu->arch.mmu_page_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_page_header_cache);
}
static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
static struct kvm_pte_chain *mmu_alloc_pte_chain(struct kvm_vcpu *vcpu)
{
- return mmu_memory_cache_alloc(&vcpu->mmu_pte_chain_cache,
+ return mmu_memory_cache_alloc(&vcpu->arch.mmu_pte_chain_cache,
sizeof(struct kvm_pte_chain));
}
static struct kvm_rmap_desc *mmu_alloc_rmap_desc(struct kvm_vcpu *vcpu)
{
- return mmu_memory_cache_alloc(&vcpu->mmu_rmap_desc_cache,
+ return mmu_memory_cache_alloc(&vcpu->arch.mmu_rmap_desc_cache,
sizeof(struct kvm_rmap_desc));
}
kfree(rd);
}
+/*
+ * Take gfn and return the reverse mapping to it.
+ * Note: gfn must be unaliased before this function get called
+ */
+
+static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn)
+{
+ struct kvm_memory_slot *slot;
+
+ slot = gfn_to_memslot(kvm, gfn);
+ return &slot->rmap[gfn - slot->base_gfn];
+}
+
/*
* Reverse mapping data structures:
*
- * If page->private bit zero is zero, then page->private points to the
- * shadow page table entry that points to page_address(page).
+ * If rmapp bit zero is zero, then rmapp point to the shadw page table entry
+ * that points to page_address(page).
*
- * If page->private bit zero is one, (then page->private & ~1) points
- * to a struct kvm_rmap_desc containing more mappings.
+ * If rmapp bit zero is one, (then rmap & ~1) points to a struct kvm_rmap_desc
+ * containing more mappings.
*/
-static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte)
+static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
{
- struct page *page;
+ struct kvm_mmu_page *sp;
struct kvm_rmap_desc *desc;
+ unsigned long *rmapp;
int i;
if (!is_rmap_pte(*spte))
return;
- page = pfn_to_page((*spte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT);
- if (!page_private(page)) {
+ gfn = unalias_gfn(vcpu->kvm, gfn);
+ sp = page_header(__pa(spte));
+ sp->gfns[spte - sp->spt] = gfn;
+ rmapp = gfn_to_rmap(vcpu->kvm, gfn);
+ if (!*rmapp) {
rmap_printk("rmap_add: %p %llx 0->1\n", spte, *spte);
- set_page_private(page,(unsigned long)spte);
- } else if (!(page_private(page) & 1)) {
+ *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 *)page_private(page);
+ desc->shadow_ptes[0] = (u64 *)*rmapp;
desc->shadow_ptes[1] = spte;
- set_page_private(page,(unsigned long)desc | 1);
+ *rmapp = (unsigned long)desc | 1;
} else {
rmap_printk("rmap_add: %p %llx many->many\n", spte, *spte);
- desc = (struct kvm_rmap_desc *)(page_private(page) & ~1ul);
+ desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
while (desc->shadow_ptes[RMAP_EXT-1] && desc->more)
desc = desc->more;
if (desc->shadow_ptes[RMAP_EXT-1]) {
}
}
-static void rmap_desc_remove_entry(struct page *page,
+static void rmap_desc_remove_entry(unsigned long *rmapp,
struct kvm_rmap_desc *desc,
int i,
struct kvm_rmap_desc *prev_desc)
if (j != 0)
return;
if (!prev_desc && !desc->more)
- set_page_private(page,(unsigned long)desc->shadow_ptes[0]);
+ *rmapp = (unsigned long)desc->shadow_ptes[0];
else
if (prev_desc)
prev_desc->more = desc->more;
else
- set_page_private(page,(unsigned long)desc->more | 1);
+ *rmapp = (unsigned long)desc->more | 1;
mmu_free_rmap_desc(desc);
}
-static void rmap_remove(u64 *spte)
+static void rmap_remove(struct kvm *kvm, u64 *spte)
{
- struct page *page;
struct kvm_rmap_desc *desc;
struct kvm_rmap_desc *prev_desc;
+ struct kvm_mmu_page *sp;
+ struct page *page;
+ unsigned long *rmapp;
int i;
if (!is_rmap_pte(*spte))
return;
+ sp = page_header(__pa(spte));
page = pfn_to_page((*spte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT);
- if (!page_private(page)) {
+ mark_page_accessed(page);
+ if (is_writeble_pte(*spte))
+ kvm_release_page_dirty(page);
+ else
+ kvm_release_page_clean(page);
+ rmapp = gfn_to_rmap(kvm, sp->gfns[spte - sp->spt]);
+ if (!*rmapp) {
printk(KERN_ERR "rmap_remove: %p %llx 0->BUG\n", spte, *spte);
BUG();
- } else if (!(page_private(page) & 1)) {
+ } else if (!(*rmapp & 1)) {
rmap_printk("rmap_remove: %p %llx 1->0\n", spte, *spte);
- if ((u64 *)page_private(page) != spte) {
+ if ((u64 *)*rmapp != spte) {
printk(KERN_ERR "rmap_remove: %p %llx 1->BUG\n",
spte, *spte);
BUG();
}
- set_page_private(page,0);
+ *rmapp = 0;
} else {
rmap_printk("rmap_remove: %p %llx many->many\n", spte, *spte);
- desc = (struct kvm_rmap_desc *)(page_private(page) & ~1ul);
+ 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) {
- rmap_desc_remove_entry(page,
+ rmap_desc_remove_entry(rmapp,
desc, i,
prev_desc);
return;
}
}
-static void rmap_write_protect(struct kvm_vcpu *vcpu, u64 gfn)
+static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte)
{
- struct kvm *kvm = vcpu->kvm;
- struct page *page;
struct kvm_rmap_desc *desc;
+ struct kvm_rmap_desc *prev_desc;
+ u64 *prev_spte;
+ int i;
+
+ if (!*rmapp)
+ return NULL;
+ else if (!(*rmapp & 1)) {
+ if (!spte)
+ return (u64 *)*rmapp;
+ return NULL;
+ }
+ desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
+ prev_desc = NULL;
+ prev_spte = NULL;
+ while (desc) {
+ for (i = 0; i < RMAP_EXT && desc->shadow_ptes[i]; ++i) {
+ if (prev_spte == spte)
+ return desc->shadow_ptes[i];
+ prev_spte = desc->shadow_ptes[i];
+ }
+ desc = desc->more;
+ }
+ return NULL;
+}
+
+static void rmap_write_protect(struct kvm *kvm, u64 gfn)
+{
+ unsigned long *rmapp;
u64 *spte;
- page = gfn_to_page(kvm, gfn);
- BUG_ON(!page);
+ gfn = unalias_gfn(kvm, gfn);
+ rmapp = gfn_to_rmap(kvm, gfn);
- while (page_private(page)) {
- if (!(page_private(page) & 1))
- spte = (u64 *)page_private(page);
- else {
- desc = (struct kvm_rmap_desc *)(page_private(page) & ~1ul);
- spte = desc->shadow_ptes[0];
- }
+ spte = rmap_next(kvm, rmapp, NULL);
+ while (spte) {
BUG_ON(!spte);
- BUG_ON((*spte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT
- != page_to_pfn(page));
BUG_ON(!(*spte & PT_PRESENT_MASK));
- BUG_ON(!(*spte & PT_WRITABLE_MASK));
rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte);
- rmap_remove(spte);
- set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK);
- kvm_flush_remote_tlbs(vcpu->kvm);
+ if (is_writeble_pte(*spte))
+ set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK);
+ kvm_flush_remote_tlbs(kvm);
+ spte = rmap_next(kvm, rmapp, spte);
}
}
u64 *end;
for (pos = spt, end = pos + PAGE_SIZE / sizeof(u64); pos != end; pos++)
- if (*pos != 0) {
+ if ((*pos & ~PT_SHADOW_IO_MARK) != shadow_trap_nonpresent_pte) {
printk(KERN_ERR "%s: %p %llx\n", __FUNCTION__,
pos, *pos);
return 0;
}
#endif
-static void kvm_mmu_free_page(struct kvm *kvm,
- struct kvm_mmu_page *page_head)
+static void kvm_mmu_free_page(struct kvm *kvm, struct kvm_mmu_page *sp)
{
- ASSERT(is_empty_shadow_page(page_head->spt));
- list_del(&page_head->link);
- __free_page(virt_to_page(page_head->spt));
- kfree(page_head);
+ ASSERT(is_empty_shadow_page(sp->spt));
+ list_del(&sp->link);
+ __free_page(virt_to_page(sp->spt));
+ __free_page(virt_to_page(sp->gfns));
+ kfree(sp);
++kvm->n_free_mmu_pages;
}
static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
u64 *parent_pte)
{
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
if (!vcpu->kvm->n_free_mmu_pages)
return NULL;
- page = mmu_memory_cache_alloc(&vcpu->mmu_page_header_cache,
- sizeof *page);
- page->spt = mmu_memory_cache_alloc(&vcpu->mmu_page_cache, PAGE_SIZE);
- set_page_private(virt_to_page(page->spt), (unsigned long)page);
- list_add(&page->link, &vcpu->kvm->active_mmu_pages);
- ASSERT(is_empty_shadow_page(page->spt));
- page->slot_bitmap = 0;
- page->multimapped = 0;
- page->parent_pte = parent_pte;
+ sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
+ sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
+ 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->active_mmu_pages);
+ ASSERT(is_empty_shadow_page(sp->spt));
+ sp->slot_bitmap = 0;
+ sp->multimapped = 0;
+ sp->parent_pte = parent_pte;
--vcpu->kvm->n_free_mmu_pages;
- return page;
+ return sp;
}
static void mmu_page_add_parent_pte(struct kvm_vcpu *vcpu,
- struct kvm_mmu_page *page, u64 *parent_pte)
+ struct kvm_mmu_page *sp, u64 *parent_pte)
{
struct kvm_pte_chain *pte_chain;
struct hlist_node *node;
if (!parent_pte)
return;
- if (!page->multimapped) {
- u64 *old = page->parent_pte;
+ if (!sp->multimapped) {
+ u64 *old = sp->parent_pte;
if (!old) {
- page->parent_pte = parent_pte;
+ sp->parent_pte = parent_pte;
return;
}
- page->multimapped = 1;
+ sp->multimapped = 1;
pte_chain = mmu_alloc_pte_chain(vcpu);
- INIT_HLIST_HEAD(&page->parent_ptes);
- hlist_add_head(&pte_chain->link, &page->parent_ptes);
+ INIT_HLIST_HEAD(&sp->parent_ptes);
+ hlist_add_head(&pte_chain->link, &sp->parent_ptes);
pte_chain->parent_ptes[0] = old;
}
- hlist_for_each_entry(pte_chain, node, &page->parent_ptes, link) {
+ hlist_for_each_entry(pte_chain, node, &sp->parent_ptes, link) {
if (pte_chain->parent_ptes[NR_PTE_CHAIN_ENTRIES-1])
continue;
for (i = 0; i < NR_PTE_CHAIN_ENTRIES; ++i)
}
pte_chain = mmu_alloc_pte_chain(vcpu);
BUG_ON(!pte_chain);
- hlist_add_head(&pte_chain->link, &page->parent_ptes);
+ hlist_add_head(&pte_chain->link, &sp->parent_ptes);
pte_chain->parent_ptes[0] = parent_pte;
}
-static void mmu_page_remove_parent_pte(struct kvm_mmu_page *page,
+static void mmu_page_remove_parent_pte(struct kvm_mmu_page *sp,
u64 *parent_pte)
{
struct kvm_pte_chain *pte_chain;
struct hlist_node *node;
int i;
- if (!page->multimapped) {
- BUG_ON(page->parent_pte != parent_pte);
- page->parent_pte = NULL;
+ if (!sp->multimapped) {
+ BUG_ON(sp->parent_pte != parent_pte);
+ sp->parent_pte = NULL;
return;
}
- hlist_for_each_entry(pte_chain, node, &page->parent_ptes, link)
+ 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;
if (i == 0) {
hlist_del(&pte_chain->link);
mmu_free_pte_chain(pte_chain);
- if (hlist_empty(&page->parent_ptes)) {
- page->multimapped = 0;
- page->parent_pte = NULL;
+ if (hlist_empty(&sp->parent_ptes)) {
+ sp->multimapped = 0;
+ sp->parent_pte = NULL;
}
}
return;
BUG();
}
-static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm_vcpu *vcpu,
- gfn_t gfn)
+static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
{
unsigned index;
struct hlist_head *bucket;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
struct hlist_node *node;
pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &vcpu->kvm->mmu_page_hash[index];
- hlist_for_each_entry(page, node, bucket, hash_link)
- if (page->gfn == gfn && !page->role.metaphysical) {
+ bucket = &kvm->mmu_page_hash[index];
+ hlist_for_each_entry(sp, node, bucket, hash_link)
+ if (sp->gfn == gfn && !sp->role.metaphysical) {
pgprintk("%s: found role %x\n",
- __FUNCTION__, page->role.word);
- return page;
+ __FUNCTION__, sp->role.word);
+ return sp;
}
return NULL;
}
gva_t gaddr,
unsigned level,
int metaphysical,
- unsigned hugepage_access,
- u64 *parent_pte)
+ unsigned access,
+ u64 *parent_pte,
+ bool *new_page)
{
union kvm_mmu_page_role role;
unsigned index;
unsigned quadrant;
struct hlist_head *bucket;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
struct hlist_node *node;
role.word = 0;
- role.glevels = vcpu->mmu.root_level;
+ role.glevels = vcpu->arch.mmu.root_level;
role.level = level;
role.metaphysical = metaphysical;
- role.hugepage_access = hugepage_access;
- if (vcpu->mmu.root_level <= PT32_ROOT_LEVEL) {
+ 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;
gfn, role.word);
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
bucket = &vcpu->kvm->mmu_page_hash[index];
- hlist_for_each_entry(page, node, bucket, hash_link)
- if (page->gfn == gfn && page->role.word == role.word) {
- mmu_page_add_parent_pte(vcpu, page, parent_pte);
+ hlist_for_each_entry(sp, node, bucket, hash_link)
+ if (sp->gfn == gfn && sp->role.word == role.word) {
+ mmu_page_add_parent_pte(vcpu, sp, parent_pte);
pgprintk("%s: found\n", __FUNCTION__);
- return page;
+ return sp;
}
- page = kvm_mmu_alloc_page(vcpu, parent_pte);
- if (!page)
- return page;
+ sp = kvm_mmu_alloc_page(vcpu, parent_pte);
+ if (!sp)
+ return sp;
pgprintk("%s: adding gfn %lx role %x\n", __FUNCTION__, gfn, role.word);
- page->gfn = gfn;
- page->role = role;
- hlist_add_head(&page->hash_link, bucket);
+ sp->gfn = gfn;
+ sp->role = role;
+ hlist_add_head(&sp->hash_link, bucket);
+ vcpu->arch.mmu.prefetch_page(vcpu, sp);
if (!metaphysical)
- rmap_write_protect(vcpu, gfn);
- return page;
+ rmap_write_protect(vcpu->kvm, gfn);
+ if (new_page)
+ *new_page = 1;
+ return sp;
}
static void kvm_mmu_page_unlink_children(struct kvm *kvm,
- struct kvm_mmu_page *page)
+ struct kvm_mmu_page *sp)
{
unsigned i;
u64 *pt;
u64 ent;
- pt = page->spt;
+ pt = sp->spt;
- if (page->role.level == PT_PAGE_TABLE_LEVEL) {
+ if (sp->role.level == PT_PAGE_TABLE_LEVEL) {
for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
- if (pt[i] & PT_PRESENT_MASK)
- rmap_remove(&pt[i]);
- pt[i] = 0;
+ if (is_shadow_present_pte(pt[i]))
+ rmap_remove(kvm, &pt[i]);
+ pt[i] = shadow_trap_nonpresent_pte;
}
kvm_flush_remote_tlbs(kvm);
return;
for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
ent = pt[i];
- pt[i] = 0;
- if (!(ent & PT_PRESENT_MASK))
+ pt[i] = shadow_trap_nonpresent_pte;
+ if (!is_shadow_present_pte(ent))
continue;
ent &= PT64_BASE_ADDR_MASK;
mmu_page_remove_parent_pte(page_header(ent), &pt[i]);
kvm_flush_remote_tlbs(kvm);
}
-static void kvm_mmu_put_page(struct kvm_mmu_page *page,
- u64 *parent_pte)
+static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte)
{
- mmu_page_remove_parent_pte(page, parent_pte);
+ mmu_page_remove_parent_pte(sp, parent_pte);
+}
+
+static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm)
+{
+ int i;
+
+ for (i = 0; i < KVM_MAX_VCPUS; ++i)
+ if (kvm->vcpus[i])
+ kvm->vcpus[i]->arch.last_pte_updated = NULL;
}
-static void kvm_mmu_zap_page(struct kvm *kvm,
- struct kvm_mmu_page *page)
+static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
{
u64 *parent_pte;
- while (page->multimapped || page->parent_pte) {
- if (!page->multimapped)
- parent_pte = page->parent_pte;
+ ++kvm->stat.mmu_shadow_zapped;
+ while (sp->multimapped || sp->parent_pte) {
+ if (!sp->multimapped)
+ parent_pte = sp->parent_pte;
else {
struct kvm_pte_chain *chain;
- chain = container_of(page->parent_ptes.first,
+ chain = container_of(sp->parent_ptes.first,
struct kvm_pte_chain, link);
parent_pte = chain->parent_ptes[0];
}
BUG_ON(!parent_pte);
- kvm_mmu_put_page(page, parent_pte);
- set_shadow_pte(parent_pte, 0);
+ kvm_mmu_put_page(sp, parent_pte);
+ set_shadow_pte(parent_pte, shadow_trap_nonpresent_pte);
}
- kvm_mmu_page_unlink_children(kvm, page);
- if (!page->root_count) {
- hlist_del(&page->hash_link);
- kvm_mmu_free_page(kvm, page);
+ kvm_mmu_page_unlink_children(kvm, sp);
+ if (!sp->root_count) {
+ hlist_del(&sp->hash_link);
+ kvm_mmu_free_page(kvm, sp);
} else
- list_move(&page->link, &kvm->active_mmu_pages);
+ list_move(&sp->link, &kvm->active_mmu_pages);
+ kvm_mmu_reset_last_pte_updated(kvm);
}
-static int kvm_mmu_unprotect_page(struct kvm_vcpu *vcpu, gfn_t gfn)
+/*
+ * Changing the number of mmu pages allocated to the vm
+ * Note: if kvm_nr_mmu_pages is too small, you will get dead lock
+ */
+void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_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->n_alloc_mmu_pages - kvm->n_free_mmu_pages) >
+ kvm_nr_mmu_pages) {
+ int n_used_mmu_pages = kvm->n_alloc_mmu_pages
+ - kvm->n_free_mmu_pages;
+
+ while (n_used_mmu_pages > kvm_nr_mmu_pages) {
+ struct kvm_mmu_page *page;
+
+ page = container_of(kvm->active_mmu_pages.prev,
+ struct kvm_mmu_page, link);
+ kvm_mmu_zap_page(kvm, page);
+ n_used_mmu_pages--;
+ }
+ kvm->n_free_mmu_pages = 0;
+ }
+ else
+ kvm->n_free_mmu_pages += kvm_nr_mmu_pages
+ - kvm->n_alloc_mmu_pages;
+
+ kvm->n_alloc_mmu_pages = kvm_nr_mmu_pages;
+}
+
+static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
{
unsigned index;
struct hlist_head *bucket;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
struct hlist_node *node, *n;
int r;
pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
r = 0;
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &vcpu->kvm->mmu_page_hash[index];
- hlist_for_each_entry_safe(page, node, n, bucket, hash_link)
- if (page->gfn == gfn && !page->role.metaphysical) {
+ bucket = &kvm->mmu_page_hash[index];
+ hlist_for_each_entry_safe(sp, node, n, bucket, hash_link)
+ if (sp->gfn == gfn && !sp->role.metaphysical) {
pgprintk("%s: gfn %lx role %x\n", __FUNCTION__, gfn,
- page->role.word);
- kvm_mmu_zap_page(vcpu->kvm, page);
+ sp->role.word);
+ kvm_mmu_zap_page(kvm, sp);
r = 1;
}
return r;
}
-static void mmu_unshadow(struct kvm_vcpu *vcpu, gfn_t gfn)
+static void mmu_unshadow(struct kvm *kvm, gfn_t gfn)
{
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
- while ((page = kvm_mmu_lookup_page(vcpu, gfn)) != NULL) {
- pgprintk("%s: zap %lx %x\n",
- __FUNCTION__, gfn, page->role.word);
- kvm_mmu_zap_page(vcpu->kvm, page);
+ while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) {
+ pgprintk("%s: zap %lx %x\n", __FUNCTION__, gfn, sp->role.word);
+ kvm_mmu_zap_page(kvm, sp);
}
}
-static void page_header_update_slot(struct kvm *kvm, void *pte, gpa_t gpa)
+static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn)
{
- int slot = memslot_id(kvm, gfn_to_memslot(kvm, gpa >> PAGE_SHIFT));
- struct kvm_mmu_page *page_head = page_header(__pa(pte));
+ int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn));
+ struct kvm_mmu_page *sp = page_header(__pa(pte));
- __set_bit(slot, &page_head->slot_bitmap);
+ __set_bit(slot, &sp->slot_bitmap);
}
-hpa_t safe_gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa)
+struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
{
- hpa_t hpa = gpa_to_hpa(vcpu, gpa);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
- return is_error_hpa(hpa) ? bad_page_address | (gpa & ~PAGE_MASK): hpa;
+ if (gpa == UNMAPPED_GVA)
+ return NULL;
+ return gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
}
-hpa_t gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa)
+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, gfn_t gfn)
{
+ u64 spte;
+ int was_rmapped = is_rmap_pte(*shadow_pte);
struct page *page;
- ASSERT((gpa & HPA_ERR_MASK) == 0);
- page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
- if (!page)
- return gpa | HPA_ERR_MASK;
- return ((hpa_t)page_to_pfn(page) << PAGE_SHIFT)
- | (gpa & (PAGE_SIZE-1));
-}
+ pgprintk("%s: spte %llx access %x write_fault %d"
+ " user_fault %d gfn %lx\n",
+ __FUNCTION__, *shadow_pte, pt_access,
+ write_fault, user_fault, gfn);
-hpa_t gva_to_hpa(struct kvm_vcpu *vcpu, gva_t gva)
-{
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva);
+ /*
+ * We don't set the accessed bit, since we sometimes want to see
+ * whether the guest actually used the pte (in order to detect
+ * demand paging).
+ */
+ spte = PT_PRESENT_MASK | PT_DIRTY_MASK;
+ if (!dirty)
+ pte_access &= ~ACC_WRITE_MASK;
+ if (!(pte_access & ACC_EXEC_MASK))
+ spte |= PT64_NX_MASK;
+
+ page = gfn_to_page(vcpu->kvm, gfn);
+
+ spte |= PT_PRESENT_MASK;
+ if (pte_access & ACC_USER_MASK)
+ spte |= PT_USER_MASK;
+
+ if (is_error_page(page)) {
+ set_shadow_pte(shadow_pte,
+ shadow_trap_nonpresent_pte | PT_SHADOW_IO_MARK);
+ kvm_release_page_clean(page);
+ return;
+ }
- if (gpa == UNMAPPED_GVA)
- return UNMAPPED_GVA;
- return gpa_to_hpa(vcpu, gpa);
-}
+ spte |= page_to_phys(page);
-struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
-{
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva);
+ if ((pte_access & ACC_WRITE_MASK)
+ || (write_fault && !is_write_protection(vcpu) && !user_fault)) {
+ struct kvm_mmu_page *shadow;
- if (gpa == UNMAPPED_GVA)
- return NULL;
- return pfn_to_page(gpa_to_hpa(vcpu, gpa) >> PAGE_SHIFT);
+ spte |= PT_WRITABLE_MASK;
+ if (user_fault) {
+ mmu_unshadow(vcpu->kvm, gfn);
+ goto unshadowed;
+ }
+
+ shadow = kvm_mmu_lookup_page(vcpu->kvm, gfn);
+ if (shadow) {
+ pgprintk("%s: found shadow page for %lx, marking ro\n",
+ __FUNCTION__, gfn);
+ pte_access &= ~ACC_WRITE_MASK;
+ if (is_writeble_pte(spte)) {
+ spte &= ~PT_WRITABLE_MASK;
+ kvm_x86_ops->tlb_flush(vcpu);
+ }
+ if (write_fault)
+ *ptwrite = 1;
+ }
+ }
+
+unshadowed:
+
+ if (pte_access & ACC_WRITE_MASK)
+ mark_page_dirty(vcpu->kvm, gfn);
+
+ pgprintk("%s: setting spte %llx\n", __FUNCTION__, spte);
+ set_shadow_pte(shadow_pte, spte);
+ page_header_update_slot(vcpu->kvm, shadow_pte, gfn);
+ if (!was_rmapped) {
+ rmap_add(vcpu, shadow_pte, gfn);
+ if (!is_rmap_pte(*shadow_pte))
+ kvm_release_page_clean(page);
+ }
+ else
+ kvm_release_page_clean(page);
+ if (!ptwrite || !*ptwrite)
+ vcpu->arch.last_pte_updated = shadow_pte;
}
static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
{
}
-static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, hpa_t p)
+static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
{
int level = PT32E_ROOT_LEVEL;
- hpa_t table_addr = vcpu->mmu.root_hpa;
+ hpa_t table_addr = vcpu->arch.mmu.root_hpa;
+ int pt_write = 0;
for (; ; level--) {
u32 index = PT64_INDEX(v, level);
u64 *table;
- u64 pte;
ASSERT(VALID_PAGE(table_addr));
table = __va(table_addr);
if (level == 1) {
- pte = table[index];
- if (is_present_pte(pte) && is_writeble_pte(pte))
- return 0;
- mark_page_dirty(vcpu->kvm, v >> PAGE_SHIFT);
- page_header_update_slot(vcpu->kvm, table, v);
- table[index] = p | PT_PRESENT_MASK | PT_WRITABLE_MASK |
- PT_USER_MASK;
- rmap_add(vcpu, &table[index]);
- return 0;
+ mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL,
+ 0, write, 1, &pt_write, gfn);
+ return pt_write || is_io_pte(table[index]);
}
- if (table[index] == 0) {
+ if (table[index] == shadow_trap_nonpresent_pte) {
struct kvm_mmu_page *new_table;
gfn_t pseudo_gfn;
>> PAGE_SHIFT;
new_table = kvm_mmu_get_page(vcpu, pseudo_gfn,
v, level - 1,
- 1, 0, &table[index]);
+ 1, ACC_ALL, &table[index],
+ NULL);
if (!new_table) {
pgprintk("nonpaging_map: ENOMEM\n");
return -ENOMEM;
}
}
+static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu,
+ struct kvm_mmu_page *sp)
+{
+ int i;
+
+ for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
+ sp->spt[i] = shadow_trap_nonpresent_pte;
+}
+
static void mmu_free_roots(struct kvm_vcpu *vcpu)
{
int i;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
- if (!VALID_PAGE(vcpu->mmu.root_hpa))
+ if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
return;
#ifdef CONFIG_X86_64
- if (vcpu->mmu.shadow_root_level == PT64_ROOT_LEVEL) {
- hpa_t root = vcpu->mmu.root_hpa;
+ if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+ hpa_t root = vcpu->arch.mmu.root_hpa;
- page = page_header(root);
- --page->root_count;
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ sp = page_header(root);
+ --sp->root_count;
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
return;
}
#endif
for (i = 0; i < 4; ++i) {
- hpa_t root = vcpu->mmu.pae_root[i];
+ hpa_t root = vcpu->arch.mmu.pae_root[i];
if (root) {
root &= PT64_BASE_ADDR_MASK;
- page = page_header(root);
- --page->root_count;
+ sp = page_header(root);
+ --sp->root_count;
}
- vcpu->mmu.pae_root[i] = INVALID_PAGE;
+ vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
}
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
{
int i;
gfn_t root_gfn;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
- root_gfn = vcpu->cr3 >> PAGE_SHIFT;
+ root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT;
#ifdef CONFIG_X86_64
- if (vcpu->mmu.shadow_root_level == PT64_ROOT_LEVEL) {
- hpa_t root = vcpu->mmu.root_hpa;
+ if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+ hpa_t root = vcpu->arch.mmu.root_hpa;
ASSERT(!VALID_PAGE(root));
- page = kvm_mmu_get_page(vcpu, root_gfn, 0,
- PT64_ROOT_LEVEL, 0, 0, NULL);
- root = __pa(page->spt);
- ++page->root_count;
- vcpu->mmu.root_hpa = root;
+ sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
+ PT64_ROOT_LEVEL, 0, ACC_ALL, NULL, NULL);
+ root = __pa(sp->spt);
+ ++sp->root_count;
+ vcpu->arch.mmu.root_hpa = root;
return;
}
#endif
for (i = 0; i < 4; ++i) {
- hpa_t root = vcpu->mmu.pae_root[i];
+ hpa_t root = vcpu->arch.mmu.pae_root[i];
ASSERT(!VALID_PAGE(root));
- if (vcpu->mmu.root_level == PT32E_ROOT_LEVEL) {
- if (!is_present_pte(vcpu->pdptrs[i])) {
- vcpu->mmu.pae_root[i] = 0;
+ if (vcpu->arch.mmu.root_level == PT32E_ROOT_LEVEL) {
+ if (!is_present_pte(vcpu->arch.pdptrs[i])) {
+ vcpu->arch.mmu.pae_root[i] = 0;
continue;
}
- root_gfn = vcpu->pdptrs[i] >> PAGE_SHIFT;
- } else if (vcpu->mmu.root_level == 0)
+ root_gfn = vcpu->arch.pdptrs[i] >> PAGE_SHIFT;
+ } else if (vcpu->arch.mmu.root_level == 0)
root_gfn = 0;
- page = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
- PT32_ROOT_LEVEL, !is_paging(vcpu),
- 0, NULL);
- root = __pa(page->spt);
- ++page->root_count;
- vcpu->mmu.pae_root[i] = root | PT_PRESENT_MASK;
+ sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
+ PT32_ROOT_LEVEL, !is_paging(vcpu),
+ ACC_ALL, NULL, NULL);
+ root = __pa(sp->spt);
+ ++sp->root_count;
+ vcpu->arch.mmu.pae_root[i] = root | PT_PRESENT_MASK;
}
- vcpu->mmu.root_hpa = __pa(vcpu->mmu.pae_root);
+ vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root);
}
static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr)
}
static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
- u32 error_code)
+ u32 error_code)
{
- gpa_t addr = gva;
- hpa_t paddr;
+ gfn_t gfn;
int r;
+ pgprintk("%s: gva %lx error %x\n", __FUNCTION__, gva, error_code);
r = mmu_topup_memory_caches(vcpu);
if (r)
return r;
ASSERT(vcpu);
- ASSERT(VALID_PAGE(vcpu->mmu.root_hpa));
-
+ ASSERT(VALID_PAGE(vcpu->arch.mmu.root_hpa));
- paddr = gpa_to_hpa(vcpu , addr & PT64_BASE_ADDR_MASK);
+ gfn = gva >> PAGE_SHIFT;
- if (is_error_hpa(paddr))
- return 1;
-
- return nonpaging_map(vcpu, addr & PAGE_MASK, paddr);
+ return nonpaging_map(vcpu, gva & PAGE_MASK,
+ error_code & PFERR_WRITE_MASK, gfn);
}
static void nonpaging_free(struct kvm_vcpu *vcpu)
static int nonpaging_init_context(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
context->new_cr3 = nonpaging_new_cr3;
context->page_fault = nonpaging_page_fault;
context->gva_to_gpa = nonpaging_gva_to_gpa;
context->free = nonpaging_free;
+ context->prefetch_page = nonpaging_prefetch_page;
context->root_level = 0;
context->shadow_root_level = PT32E_ROOT_LEVEL;
context->root_hpa = INVALID_PAGE;
return 0;
}
-static void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
+void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
{
++vcpu->stat.tlb_flush;
- kvm_arch_ops->tlb_flush(vcpu);
+ kvm_x86_ops->tlb_flush(vcpu);
}
static void paging_new_cr3(struct kvm_vcpu *vcpu)
u64 addr,
u32 err_code)
{
- kvm_arch_ops->inject_page_fault(vcpu, addr, err_code);
+ kvm_inject_page_fault(vcpu, addr, err_code);
}
static void paging_free(struct kvm_vcpu *vcpu)
static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
ASSERT(is_pae(vcpu));
context->new_cr3 = paging_new_cr3;
context->page_fault = paging64_page_fault;
context->gva_to_gpa = paging64_gva_to_gpa;
+ context->prefetch_page = paging64_prefetch_page;
context->free = paging_free;
context->root_level = level;
context->shadow_root_level = level;
static int paging32_init_context(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
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->root_level = PT32_ROOT_LEVEL;
context->shadow_root_level = PT32E_ROOT_LEVEL;
context->root_hpa = INVALID_PAGE;
static int init_kvm_mmu(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
if (!is_paging(vcpu))
return nonpaging_init_context(vcpu);
static void destroy_kvm_mmu(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- if (VALID_PAGE(vcpu->mmu.root_hpa)) {
- vcpu->mmu.free(vcpu);
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ if (VALID_PAGE(vcpu->arch.mmu.root_hpa)) {
+ vcpu->arch.mmu.free(vcpu);
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
}
destroy_kvm_mmu(vcpu);
return init_kvm_mmu(vcpu);
}
+EXPORT_SYMBOL_GPL(kvm_mmu_reset_context);
int kvm_mmu_load(struct kvm_vcpu *vcpu)
{
int r;
- spin_lock(&vcpu->kvm->lock);
+ mutex_lock(&vcpu->kvm->lock);
r = mmu_topup_memory_caches(vcpu);
if (r)
goto out;
mmu_alloc_roots(vcpu);
- kvm_arch_ops->set_cr3(vcpu, vcpu->mmu.root_hpa);
+ kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
kvm_mmu_flush_tlb(vcpu);
out:
- spin_unlock(&vcpu->kvm->lock);
+ mutex_unlock(&vcpu->kvm->lock);
return r;
}
EXPORT_SYMBOL_GPL(kvm_mmu_load);
}
static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu,
- struct kvm_mmu_page *page,
+ struct kvm_mmu_page *sp,
u64 *spte)
{
u64 pte;
struct kvm_mmu_page *child;
pte = *spte;
- if (is_present_pte(pte)) {
- if (page->role.level == PT_PAGE_TABLE_LEVEL)
- rmap_remove(spte);
+ if (is_shadow_present_pte(pte)) {
+ if (sp->role.level == PT_PAGE_TABLE_LEVEL)
+ rmap_remove(vcpu->kvm, spte);
else {
child = page_header(pte & PT64_BASE_ADDR_MASK);
mmu_page_remove_parent_pte(child, spte);
}
}
- *spte = 0;
- kvm_flush_remote_tlbs(vcpu->kvm);
+ set_shadow_pte(spte, shadow_trap_nonpresent_pte);
}
static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu,
- struct kvm_mmu_page *page,
+ struct kvm_mmu_page *sp,
u64 *spte,
- const void *new, int bytes)
+ const void *new, int bytes,
+ int offset_in_pte)
{
- if (page->role.level != PT_PAGE_TABLE_LEVEL)
+ if (sp->role.level != PT_PAGE_TABLE_LEVEL) {
+ ++vcpu->kvm->stat.mmu_pde_zapped;
return;
+ }
- if (page->role.glevels == PT32_ROOT_LEVEL)
- paging32_update_pte(vcpu, page, spte, new, bytes);
+ ++vcpu->kvm->stat.mmu_pte_updated;
+ if (sp->role.glevels == PT32_ROOT_LEVEL)
+ paging32_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
else
- paging64_update_pte(vcpu, page, spte, new, bytes);
+ paging64_update_pte(vcpu, sp, spte, new, bytes, offset_in_pte);
+}
+
+static bool need_remote_flush(u64 old, u64 new)
+{
+ if (!is_shadow_present_pte(old))
+ return false;
+ if (!is_shadow_present_pte(new))
+ return true;
+ if ((old ^ new) & PT64_BASE_ADDR_MASK)
+ return true;
+ old ^= PT64_NX_MASK;
+ new ^= PT64_NX_MASK;
+ return (old & ~new & PT64_PERM_MASK) != 0;
+}
+
+static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, u64 old, u64 new)
+{
+ if (need_remote_flush(old, new))
+ kvm_flush_remote_tlbs(vcpu->kvm);
+ else
+ kvm_mmu_flush_tlb(vcpu);
+}
+
+static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
+{
+ u64 *spte = vcpu->arch.last_pte_updated;
+
+ return !!(spte && (*spte & PT_ACCESSED_MASK));
}
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
- const u8 *old, const u8 *new, int bytes)
+ const u8 *new, int bytes)
{
gfn_t gfn = gpa >> PAGE_SHIFT;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
struct hlist_node *node, *n;
struct hlist_head *bucket;
unsigned index;
+ u64 entry;
u64 *spte;
unsigned offset = offset_in_page(gpa);
unsigned pte_size;
int npte;
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
- if (gfn == vcpu->last_pt_write_gfn) {
- ++vcpu->last_pt_write_count;
- if (vcpu->last_pt_write_count >= 3)
+ ++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->last_pt_write_gfn = gfn;
- vcpu->last_pt_write_count = 1;
+ 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) % KVM_NUM_MMU_PAGES;
bucket = &vcpu->kvm->mmu_page_hash[index];
- hlist_for_each_entry_safe(page, node, n, bucket, hash_link) {
- if (page->gfn != gfn || page->role.metaphysical)
+ hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) {
+ if (sp->gfn != gfn || sp->role.metaphysical)
continue;
- pte_size = page->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
+ pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1);
misaligned |= bytes < 4;
if (misaligned || flooded) {
* page.
*/
pgprintk("misaligned: gpa %llx bytes %d role %x\n",
- gpa, bytes, page->role.word);
- kvm_mmu_zap_page(vcpu->kvm, page);
+ gpa, bytes, sp->role.word);
+ kvm_mmu_zap_page(vcpu->kvm, sp);
+ ++vcpu->kvm->stat.mmu_flooded;
continue;
}
page_offset = offset;
- level = page->role.level;
+ level = sp->role.level;
npte = 1;
- if (page->role.glevels == PT32_ROOT_LEVEL) {
+ if (sp->role.glevels == PT32_ROOT_LEVEL) {
page_offset <<= 1; /* 32->64 */
/*
* A 32-bit pde maps 4MB while the shadow pdes map
}
quadrant = page_offset >> PAGE_SHIFT;
page_offset &= ~PAGE_MASK;
- if (quadrant != page->role.quadrant)
+ if (quadrant != sp->role.quadrant)
continue;
}
- spte = &page->spt[page_offset / sizeof(*spte)];
+ spte = &sp->spt[page_offset / sizeof(*spte)];
while (npte--) {
- mmu_pte_write_zap_pte(vcpu, page, spte);
- mmu_pte_write_new_pte(vcpu, page, spte, new, bytes);
+ entry = *spte;
+ mmu_pte_write_zap_pte(vcpu, sp, spte);
+ mmu_pte_write_new_pte(vcpu, sp, spte, new, bytes,
+ page_offset & (pte_size - 1));
+ mmu_pte_write_flush_tlb(vcpu, entry, *spte);
++spte;
}
}
+ kvm_mmu_audit(vcpu, "post pte write");
}
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
{
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
- return kvm_mmu_unprotect_page(vcpu, gpa >> PAGE_SHIFT);
+ return kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
}
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
while (vcpu->kvm->n_free_mmu_pages < KVM_REFILL_PAGES) {
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
+
+ sp = container_of(vcpu->kvm->active_mmu_pages.prev,
+ struct kvm_mmu_page, link);
+ kvm_mmu_zap_page(vcpu->kvm, sp);
+ ++vcpu->kvm->stat.mmu_recycled;
+ }
+}
+
+int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
+{
+ int r;
+ enum emulation_result er;
+
+ mutex_lock(&vcpu->kvm->lock);
+ r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code);
+ if (r < 0)
+ goto out;
- page = container_of(vcpu->kvm->active_mmu_pages.prev,
- struct kvm_mmu_page, link);
- kvm_mmu_zap_page(vcpu->kvm, page);
+ if (!r) {
+ r = 1;
+ goto out;
+ }
+
+ r = mmu_topup_memory_caches(vcpu);
+ if (r)
+ goto out;
+
+ er = emulate_instruction(vcpu, vcpu->run, cr2, error_code, 0);
+ mutex_unlock(&vcpu->kvm->lock);
+
+ switch (er) {
+ case EMULATE_DONE:
+ return 1;
+ case EMULATE_DO_MMIO:
+ ++vcpu->stat.mmio_exits;
+ return 0;
+ case EMULATE_FAIL:
+ kvm_report_emulation_failure(vcpu, "pagetable");
+ return 1;
+ default:
+ BUG();
}
+out:
+ mutex_unlock(&vcpu->kvm->lock);
+ return r;
}
+EXPORT_SYMBOL_GPL(kvm_mmu_page_fault);
static void free_mmu_pages(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
while (!list_empty(&vcpu->kvm->active_mmu_pages)) {
- page = container_of(vcpu->kvm->active_mmu_pages.next,
- struct kvm_mmu_page, link);
- kvm_mmu_zap_page(vcpu->kvm, page);
+ sp = container_of(vcpu->kvm->active_mmu_pages.next,
+ struct kvm_mmu_page, link);
+ kvm_mmu_zap_page(vcpu->kvm, sp);
}
- free_page((unsigned long)vcpu->mmu.pae_root);
+ free_page((unsigned long)vcpu->arch.mmu.pae_root);
}
static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
ASSERT(vcpu);
- vcpu->kvm->n_free_mmu_pages = KVM_NUM_MMU_PAGES;
-
+ if (vcpu->kvm->n_requested_mmu_pages)
+ vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_requested_mmu_pages;
+ else
+ vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_alloc_mmu_pages;
/*
* 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
page = alloc_page(GFP_KERNEL | __GFP_DMA32);
if (!page)
goto error_1;
- vcpu->mmu.pae_root = page_address(page);
+ vcpu->arch.mmu.pae_root = page_address(page);
for (i = 0; i < 4; ++i)
- vcpu->mmu.pae_root[i] = INVALID_PAGE;
+ vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
return 0;
int kvm_mmu_create(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
return alloc_mmu_pages(vcpu);
}
int kvm_mmu_setup(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
return init_kvm_mmu(vcpu);
}
void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
{
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
- list_for_each_entry(page, &kvm->active_mmu_pages, link) {
+ list_for_each_entry(sp, &kvm->active_mmu_pages, link) {
int i;
u64 *pt;
- if (!test_bit(slot, &page->slot_bitmap))
+ if (!test_bit(slot, &sp->slot_bitmap))
continue;
- pt = page->spt;
+ pt = sp->spt;
for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
/* avoid RMW */
- if (pt[i] & PT_WRITABLE_MASK) {
- rmap_remove(&pt[i]);
+ if (pt[i] & PT_WRITABLE_MASK)
pt[i] &= ~PT_WRITABLE_MASK;
- }
}
}
void kvm_mmu_zap_all(struct kvm *kvm)
{
- struct kvm_mmu_page *page, *node;
+ struct kvm_mmu_page *sp, *node;
- list_for_each_entry_safe(page, node, &kvm->active_mmu_pages, link)
- kvm_mmu_zap_page(kvm, page);
+ list_for_each_entry_safe(sp, node, &kvm->active_mmu_pages, link)
+ kvm_mmu_zap_page(kvm, sp);
kvm_flush_remote_tlbs(kvm);
}
return -ENOMEM;
}
+/*
+ * Caculate mmu pages needed for kvm.
+ */
+unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm)
+{
+ int i;
+ unsigned int nr_mmu_pages;
+ unsigned int nr_pages = 0;
+
+ for (i = 0; i < kvm->nmemslots; i++)
+ nr_pages += kvm->memslots[i].npages;
+
+ nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000;
+ nr_mmu_pages = max(nr_mmu_pages,
+ (unsigned int) KVM_MIN_ALLOC_MMU_PAGES);
+
+ return nr_mmu_pages;
+}
+
#ifdef AUDIT
static const char *audit_msg;
for (i = 0; i < PT64_ENT_PER_PAGE; ++i, va += va_delta) {
u64 ent = pt[i];
- if (!(ent & PT_PRESENT_MASK))
+ if (ent == shadow_trap_nonpresent_pte)
continue;
va = canonicalize(va);
- if (level > 1)
+ 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);
+
audit_mappings_page(vcpu, ent, va, level - 1);
- else {
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, va);
- hpa_t hpa = gpa_to_hpa(vcpu, gpa);
+ } else {
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);
+ struct page *page = gpa_to_page(vcpu, gpa);
+ hpa_t hpa = page_to_phys(page);
- if ((ent & PT_PRESENT_MASK)
+ if (is_shadow_present_pte(ent)
&& (ent & PT64_BASE_ADDR_MASK) != hpa)
- printk(KERN_ERR "audit error: (%s) levels %d"
- " gva %lx gpa %llx hpa %llx ent %llx\n",
- audit_msg, vcpu->mmu.root_level,
- va, gpa, hpa, ent);
+ printk(KERN_ERR "xx audit error: (%s) levels %d"
+ " gva %lx gpa %llx hpa %llx ent %llx %d\n",
+ audit_msg, vcpu->arch.mmu.root_level,
+ va, gpa, hpa, ent,
+ is_shadow_present_pte(ent));
+ else if (ent == shadow_notrap_nonpresent_pte
+ && !is_error_hpa(hpa))
+ printk(KERN_ERR "audit: (%s) notrap shadow,"
+ " valid guest gva %lx\n", audit_msg, va);
+ kvm_release_page_clean(page);
+
}
}
}
{
unsigned i;
- if (vcpu->mmu.root_level == 4)
- audit_mappings_page(vcpu, vcpu->mmu.root_hpa, 0, 4);
+ if (vcpu->arch.mmu.root_level == 4)
+ audit_mappings_page(vcpu, vcpu->arch.mmu.root_hpa, 0, 4);
else
for (i = 0; i < 4; ++i)
- if (vcpu->mmu.pae_root[i] & PT_PRESENT_MASK)
+ if (vcpu->arch.mmu.pae_root[i] & PT_PRESENT_MASK)
audit_mappings_page(vcpu,
- vcpu->mmu.pae_root[i],
+ vcpu->arch.mmu.pae_root[i],
i << 30,
2);
}
struct kvm_rmap_desc *d;
for (j = 0; j < m->npages; ++j) {
- struct page *page = m->phys_mem[j];
+ unsigned long *rmapp = &m->rmap[j];
- if (!page->private)
+ if (!*rmapp)
continue;
- if (!(page->private & 1)) {
+ if (!(*rmapp & 1)) {
++nmaps;
continue;
}
- d = (struct kvm_rmap_desc *)(page->private & ~1ul);
+ d = (struct kvm_rmap_desc *)(*rmapp & ~1ul);
while (d) {
for (k = 0; k < RMAP_EXT; ++k)
if (d->shadow_ptes[k])
static int count_writable_mappings(struct kvm_vcpu *vcpu)
{
int nmaps = 0;
- struct kvm_mmu_page *page;
+ struct kvm_mmu_page *sp;
int i;
- list_for_each_entry(page, &vcpu->kvm->active_mmu_pages, link) {
- u64 *pt = page->spt;
+ list_for_each_entry(sp, &vcpu->kvm->active_mmu_pages, link) {
+ u64 *pt = sp->spt;
- if (page->role.level != PT_PAGE_TABLE_LEVEL)
+ if (sp->role.level != PT_PAGE_TABLE_LEVEL)
continue;
for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
static void audit_write_protection(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu_page *page;
-
- list_for_each_entry(page, &vcpu->kvm->active_mmu_pages, link) {
- hfn_t hfn;
- struct page *pg;
+ struct kvm_mmu_page *sp;
+ struct kvm_memory_slot *slot;
+ unsigned long *rmapp;
+ gfn_t gfn;
- if (page->role.metaphysical)
+ list_for_each_entry(sp, &vcpu->kvm->active_mmu_pages, link) {
+ if (sp->role.metaphysical)
continue;
- hfn = gpa_to_hpa(vcpu, (gpa_t)page->gfn << PAGE_SHIFT)
- >> PAGE_SHIFT;
- pg = pfn_to_page(hfn);
- if (pg->private)
+ slot = gfn_to_memslot(vcpu->kvm, sp->gfn);
+ gfn = unalias_gfn(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",
- __FUNCTION__, audit_msg, page->gfn,
- page->role.word);
+ __FUNCTION__, audit_msg, sp->gfn,
+ sp->role.word);
}
}
git.openpandora.org / pandora-kernel.git / blobdiff