[PATCH 3/8] KVM: MMU: do not add a nonpresent spte to rmaps of itschild
From: Xiao Guangrong
Date: Fri Dec 16 2011 - 05:15:44 EST
Set the spte before adding it to the rmap of its child so that all parent
spte are valid when propagate unsync bit from a usnync page / children page
And this feature is needed by the later patch
Signed-off-by: Xiao Guangrong <xiaoguangrong@xxxxxxxxxxxxxxxxxx>
---
arch/x86/kvm/mmu.c | 74 +++++++++++++++----------------------------
arch/x86/kvm/mmutrace.h | 2 +-
arch/x86/kvm/paging_tmpl.h | 14 +++-----
3 files changed, 32 insertions(+), 58 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index a2d28aa..89202f4 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -1321,12 +1321,14 @@ static unsigned kvm_page_table_hashfn(gfn_t gfn)
return gfn & ((1 << KVM_MMU_HASH_SHIFT) - 1);
}
-static void mmu_page_add_parent_pte(struct kvm_vcpu *vcpu,
- struct kvm_mmu_page *sp, u64 *parent_pte)
+static void mmu_page_add_set_parent_pte(struct kvm_vcpu *vcpu,
+ struct kvm_mmu_page *sp,
+ u64 *parent_pte)
{
if (!parent_pte)
return;
+ mmu_spte_set(parent_pte, __pa(sp->spt) | SHADOW_PAGE_TABLE);
pte_list_add(vcpu, parent_pte, &sp->parent_ptes);
}
@@ -1357,7 +1359,7 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
bitmap_zero(sp->slot_bitmap, KVM_MEM_SLOTS_NUM);
sp->parent_ptes = 0;
- mmu_page_add_parent_pte(vcpu, sp, parent_pte);
+ mmu_page_add_set_parent_pte(vcpu, sp, parent_pte);
kvm_mod_used_mmu_pages(vcpu->kvm, +1);
return sp;
}
@@ -1690,13 +1692,10 @@ static void clear_sp_write_flooding_count(u64 *spte)
__clear_sp_write_flooding_count(sp);
}
-static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
- gfn_t gfn,
- gva_t gaddr,
- unsigned level,
- int direct,
- unsigned access,
- u64 *parent_pte)
+static struct kvm_mmu_page *
+kvm_mmu_get_set_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr,
+ unsigned level, int direct, unsigned access,
+ u64 *parent_pte)
{
union kvm_mmu_page_role role;
unsigned quadrant;
@@ -1726,7 +1725,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
if (sp_is_unsync(sp) && kvm_sync_page_transient(vcpu, sp))
break;
- mmu_page_add_parent_pte(vcpu, sp, parent_pte);
+ mmu_page_add_set_parent_pte(vcpu, sp, parent_pte);
if (sp_unsync_children_num(sp)) {
kvm_make_request(KVM_REQ_MMU_SYNC, vcpu);
kvm_mmu_mark_parents_unsync(sp);
@@ -1734,7 +1733,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
kvm_mmu_mark_parents_unsync(sp);
__clear_sp_write_flooding_count(sp);
- trace_kvm_mmu_get_page(sp, false);
+ trace_kvm_mmu_get_set_page(sp, false);
return sp;
}
++vcpu->kvm->stat.mmu_cache_miss;
@@ -1754,7 +1753,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
account_shadowed(vcpu->kvm, gfn);
}
init_shadow_page_table(sp);
- trace_kvm_mmu_get_page(sp, true);
+ trace_kvm_mmu_get_set_page(sp, true);
return sp;
}
@@ -1807,14 +1806,6 @@ static void shadow_walk_next(struct kvm_shadow_walk_iterator *iterator)
return __shadow_walk_next(iterator, *iterator->sptep);
}
-static void link_shadow_page(u64 *sptep, struct kvm_mmu_page *sp)
-{
- u64 spte;
-
- spte = __pa(sp->spt) | SHADOW_PAGE_TABLE;
- mmu_spte_set(sptep, spte);
-}
-
static void drop_large_spte(struct kvm_vcpu *vcpu, u64 *sptep)
{
if (is_large_pte(*sptep)) {
@@ -1879,11 +1870,6 @@ static void kvm_mmu_page_unlink_children(struct kvm *kvm,
mmu_page_zap_pte(kvm, sp, sp->spt + i);
}
-static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte)
-{
- mmu_page_remove_parent_pte(sp, parent_pte);
-}
-
static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp)
{
u64 *parent_pte;
@@ -2468,7 +2454,6 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
bool prefault)
{
struct kvm_shadow_walk_iterator iterator;
- struct kvm_mmu_page *sp;
int emulate = 0;
gfn_t pseudo_gfn;
@@ -2489,16 +2474,9 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
base_addr &= PT64_LVL_ADDR_MASK(iterator.level);
pseudo_gfn = base_addr >> 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;
- }
-
- link_shadow_page(iterator.sptep, sp);
+ kvm_mmu_get_set_page(vcpu, pseudo_gfn, iterator.addr,
+ iterator.level - 1,
+ 1, ACC_ALL, iterator.sptep);
}
}
return emulate;
@@ -2713,8 +2691,8 @@ static int mmu_alloc_direct_roots(struct kvm_vcpu *vcpu)
if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
spin_lock(&vcpu->kvm->mmu_lock);
kvm_mmu_free_some_pages(vcpu);
- sp = kvm_mmu_get_page(vcpu, 0, 0, PT64_ROOT_LEVEL,
- 1, ACC_ALL, NULL);
+ sp = kvm_mmu_get_set_page(vcpu, 0, 0, PT64_ROOT_LEVEL,
+ 1, ACC_ALL, NULL);
++sp->root_count;
spin_unlock(&vcpu->kvm->mmu_lock);
vcpu->arch.mmu.root_hpa = __pa(sp->spt);
@@ -2725,10 +2703,10 @@ static int mmu_alloc_direct_roots(struct kvm_vcpu *vcpu)
ASSERT(!VALID_PAGE(root));
spin_lock(&vcpu->kvm->mmu_lock);
kvm_mmu_free_some_pages(vcpu);
- sp = kvm_mmu_get_page(vcpu, i << (30 - PAGE_SHIFT),
- i << 30,
- PT32_ROOT_LEVEL, 1, ACC_ALL,
- NULL);
+ sp = kvm_mmu_get_set_page(vcpu, i << (30 - PAGE_SHIFT),
+ i << 30,
+ PT32_ROOT_LEVEL, 1, ACC_ALL,
+ NULL);
root = __pa(sp->spt);
++sp->root_count;
spin_unlock(&vcpu->kvm->mmu_lock);
@@ -2764,8 +2742,8 @@ static int mmu_alloc_shadow_roots(struct kvm_vcpu *vcpu)
spin_lock(&vcpu->kvm->mmu_lock);
kvm_mmu_free_some_pages(vcpu);
- sp = kvm_mmu_get_page(vcpu, root_gfn, 0, PT64_ROOT_LEVEL,
- 0, ACC_ALL, NULL);
+ sp = kvm_mmu_get_set_page(vcpu, root_gfn, 0, PT64_ROOT_LEVEL,
+ 0, ACC_ALL, NULL);
root = __pa(sp->spt);
++sp->root_count;
spin_unlock(&vcpu->kvm->mmu_lock);
@@ -2798,9 +2776,9 @@ static int mmu_alloc_shadow_roots(struct kvm_vcpu *vcpu)
}
spin_lock(&vcpu->kvm->mmu_lock);
kvm_mmu_free_some_pages(vcpu);
- sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
- PT32_ROOT_LEVEL, 0,
- ACC_ALL, NULL);
+ sp = kvm_mmu_get_set_page(vcpu, root_gfn, i << 30,
+ PT32_ROOT_LEVEL, 0,
+ ACC_ALL, NULL);
root = __pa(sp->spt);
++sp->root_count;
spin_unlock(&vcpu->kvm->mmu_lock);
diff --git a/arch/x86/kvm/mmutrace.h b/arch/x86/kvm/mmutrace.h
index 7fe9562..f100078 100644
--- a/arch/x86/kvm/mmutrace.h
+++ b/arch/x86/kvm/mmutrace.h
@@ -145,7 +145,7 @@ TRACE_EVENT(
);
TRACE_EVENT(
- kvm_mmu_get_page,
+ kvm_mmu_get_set_page,
TP_PROTO(struct kvm_mmu_page *sp, bool created),
TP_ARGS(sp, created),
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 7dacc80..c79c503 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -503,8 +503,8 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
sp = NULL;
if (!is_shadow_present_pte(*it.sptep)) {
table_gfn = gw->table_gfn[it.level - 2];
- sp = kvm_mmu_get_page(vcpu, table_gfn, addr, it.level-1,
- false, access, it.sptep);
+ sp = kvm_mmu_get_set_page(vcpu, table_gfn, addr,
+ it.level - 1, false, access, it.sptep);
}
/*
@@ -513,9 +513,6 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
*/
if (FNAME(gpte_changed)(vcpu, gw, it.level - 1))
goto out_gpte_changed;
-
- if (sp)
- link_shadow_page(it.sptep, sp);
}
for (;
@@ -533,9 +530,8 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
direct_gfn = gw->gfn & ~(KVM_PAGES_PER_HPAGE(it.level) - 1);
- sp = kvm_mmu_get_page(vcpu, direct_gfn, addr, it.level-1,
- true, direct_access, it.sptep);
- link_shadow_page(it.sptep, sp);
+ kvm_mmu_get_set_page(vcpu, direct_gfn, addr, it.level - 1,
+ true, direct_access, it.sptep);
}
clear_sp_write_flooding_count(it.sptep);
@@ -548,7 +544,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
out_gpte_changed:
if (sp)
- kvm_mmu_put_page(sp, it.sptep);
+ drop_parent_pte(sp, it.sptep);
kvm_release_pfn_clean(pfn);
return NULL;
}
--
1.7.7.4
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/