Skip to contentSkip to conversation

ksm support for kvm

PATCH In review Izik Eidus 3 participants 14 messages Sep 10, 2009 – Sep 14, 2009

Subsystems (3)
lore
Share & tools
lore

Izik Eidus

author
#0 lore raw

cover ksm support for kvm

Hi,

The following seires add ksm support to the kvm mmu.

Thanks.

author
#1 lore raw

PATCH 1/3 kvm: dont hold pagecount reference for mapped sptes pages

When using mmu notifiers, we are allowed to remove the page count reference tooken by get_user_pages to a specific page that is mapped inside the shadow page tables.

This is needed so we can balance the pagecount against mapcount checking.

(Right now kvm increase the pagecount and does not increase the mapcount when mapping page into shadow page table entry, so when comparing pagecount against mapcount, you have no reliable result.)

---
 arch/x86/kvm/mmu.c |    7 ++-----
 1 files changed, 2 insertions(+), 5 deletions(-)
arch/x86/kvm/mmu.c +2 -5
index f76d086..62d2f86 100644
@@ -634,9 +634,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte)
 	if (*spte & shadow_accessed_mask)
 		kvm_set_pfn_accessed(pfn);
 	if (is_writeble_pte(*spte))
-		kvm_release_pfn_dirty(pfn);
-	else
-		kvm_release_pfn_clean(pfn);
+		kvm_set_pfn_dirty(pfn);
 	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);
@@ -1877,8 +1875,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	page_header_update_slot(vcpu->kvm, sptep, gfn);
 	if (!was_rmapped) {
 		rmap_count = rmap_add(vcpu, sptep, gfn);
-		if (!is_rmap_spte(*sptep))
-			kvm_release_pfn_clean(pfn);
+		kvm_release_pfn_clean(pfn);
 		if (rmap_count > RMAP_RECYCLE_THRESHOLD)
 			rmap_recycle(vcpu, sptep, gfn);
 	} else {
Signature · 1 line
1.5.6.5
author
#2 lore raw

PATCH 2/3 add SPTE_HOST_WRITEABLE flag to the shadow ptes

this flag notify that the host physical page we are pointing to from the spte is write protected, and therefore we cant change its access to be write unless we run get_user_pages(write = 1).

(this is needed for change_pte support in kvm)

---
 arch/x86/kvm/mmu.c         |   15 +++++++++++----
 arch/x86/kvm/paging_tmpl.h |   18 +++++++++++++++---
 2 files changed, 26 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu.c +11 -4
index 62d2f86..a7151b8 100644
@@ -156,6 +156,8 @@ module_param(oos_shadow, bool, 0644);
 #define CREATE_TRACE_POINTS
 #include "mmutrace.h"
 
+#define SPTE_HOST_WRITEABLE (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
+
 #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
 
 struct kvm_rmap_desc {
@@ -1754,7 +1756,7 @@ 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)
+		    bool can_unsync, bool reset_host_protection)
 {
 	u64 spte;
 	int ret = 0;
@@ -1781,6 +1783,9 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 		spte |= kvm_x86_ops->get_mt_mask(vcpu, gfn,
 			kvm_is_mmio_pfn(pfn));
 
+	if (reset_host_protection)
+		spte |= SPTE_HOST_WRITEABLE;
+
 	spte |= (u64)pfn << PAGE_SHIFT;
 
 	if ((pte_access & ACC_WRITE_MASK)
@@ -1826,7 +1831,8 @@ 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)
+			 pfn_t pfn, bool speculative,
+			 bool reset_host_protection)
 {
 	int was_rmapped = 0;
 	int was_writeble = is_writeble_pte(*sptep);
@@ -1858,7 +1864,8 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	}
 
 	if (set_spte(vcpu, sptep, pte_access, user_fault, write_fault,
-		      dirty, level, gfn, pfn, speculative, true)) {
+		      dirty, level, gfn, pfn, speculative, true,
+		      reset_host_protection)) {
 		if (write_fault)
 			*ptwrite = 1;
 		kvm_x86_ops->tlb_flush(vcpu);
@@ -1906,7 +1913,7 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
 		if (iterator.level == level) {
 			mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL,
 				     0, write, 1, &pt_write,
-				     level, gfn, pfn, false);
+				     level, gfn, pfn, false, true);
 			++vcpu->stat.pf_fixed;
 			break;
 		}
arch/x86/kvm/paging_tmpl.h +15 -3
index d2fec9c..c9256ee 100644
@@ -273,9 +273,13 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 	if (mmu_notifier_retry(vcpu, vcpu->arch.update_pte.mmu_seq))
 		return;
 	kvm_get_pfn(pfn);
+ 	/*
+ 	 * we call mmu_set_spte() with reset_host_protection = true beacuse that
+ 	 * vcpu->arch.update_pte.pfn was fetched from get_user_pages(write = 1).
+ 	 */
 	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
 		     gpte & PT_DIRTY_MASK, NULL, PT_PAGE_TABLE_LEVEL,
-		     gpte_to_gfn(gpte), pfn, true);
+		     gpte_to_gfn(gpte), pfn, true, true);
 }
 
 /*
@@ -308,7 +312,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 				     user_fault, write_fault,
 				     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
 				     ptwrite, level,
-				     gw->gfn, pfn, false);
+				     gw->gfn, pfn, false, true);
 			break;
 		}
 
@@ -558,6 +562,7 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
 	int i, offset, nr_present;
+        bool reset_host_protection;
 
 	offset = nr_present = 0;
 
@@ -595,9 +600,16 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 
 		nr_present++;
 		pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte);
+		if (!(sp->spt[i] & SPTE_HOST_WRITEABLE)) {
+ 			pte_access &= ~PT_WRITABLE_MASK;
+			reset_host_protection = 0;
+		} else {
+			reset_host_protection = 1;
+		}
 		set_spte(vcpu, &sp->spt[i], pte_access, 0, 0,
 			 is_dirty_gpte(gpte), PT_PAGE_TABLE_LEVEL, gfn,
-			 spte_to_pfn(sp->spt[i]), true, false);
+			 spte_to_pfn(sp->spt[i]), true, false,
+			 reset_host_protection);
 	}
 
 	return !nr_present;
Signature · 1 line
1.5.6.5
author
#3 lore raw

PATCH 3/3 add support for change_pte mmu notifiers

this is needed for kvm if it want ksm to directly map pages into its shadow page tables.

---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |   70 ++++++++++++++++++++++++++++++++++----
 virt/kvm/kvm_main.c             |   14 ++++++++
 3 files changed, 77 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 6046e6f..594d131 100644
@@ -797,6 +797,7 @@ asmlinkage void kvm_handle_fault_on_reboot(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
arch/x86/kvm/mmu.c +62 -8

119 lines not on this page · raw message

virt/kvm/kvm_main.c +14 -0
index cb7966f..c9a88af 100644
@@ -240,6 +240,19 @@ static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn,
 
 }
 
+static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn,
+					struct mm_struct *mm,
+					unsigned long address,
+					pte_t pte)
+{
+	struct kvm *kvm = mmu_notifier_to_kvm(mn);
+
+	spin_lock(&kvm->mmu_lock);
+	kvm->mmu_notifier_seq++;
+	kvm_set_spte_hva(kvm, address, pte);
+	spin_unlock(&kvm->mmu_lock);
+}
+
 static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 						    struct mm_struct *mm,
 						    unsigned long start,
@@ -319,6 +332,7 @@ static const struct mmu_notifier_ops kvm_mmu_notifier_ops = {
 	.invalidate_range_start	= kvm_mmu_notifier_invalidate_range_start,
 	.invalidate_range_end	= kvm_mmu_notifier_invalidate_range_end,
 	.clear_flush_young	= kvm_mmu_notifier_clear_flush_young,
+	.change_pte		= kvm_mmu_notifier_change_pte,
 	.release		= kvm_mmu_notifier_release,
 };
 #endif /* CONFIG_MMU_NOTIFIER && KVM_ARCH_WANT_MMU_NOTIFIER */
Signature · 1 line
1.5.6.5
#4 lore raw

Re: [PATCH 3/3] add support for change_pte mmu notifiers

Quoted 103 lines · Izik Eidus

this is needed for kvm if it want ksm to directly map pages into its shadow page tables.

---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |   70 ++++++++++++++++++++++++++++++++++----
 virt/kvm/kvm_main.c             |   14 ++++++++
 3 files changed, 77 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 6046e6f..594d131 100644
@@ -797,6 +797,7 @@ asmlinkage void kvm_handle_fault_on_reboot(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
arch/x86/kvm/mmu.c +48 -3
index a7151b8..3fd19f2 100644
@@ -282,6 +282,11 @@ static pfn_t spte_to_pfn(u64 pte)
 	return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
 }
 
+static pte_t ptep_val(pte_t *ptep)
+{
+	return *ptep;
+}
+
 static gfn_t pse36_gfn_delta(u32 gpte)
 {
 	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
@@ -748,7 +753,8 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 	return write_protected;
 }
 
-static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
+static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			   unsigned long data)
 {
 	u64 *spte;
 	int need_tlb_flush = 0;
@@ -763,8 +769,48 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
 	return need_tlb_flush;
 }
 
+static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			     unsigned long data)
+{
+	int need_flush = 0;
+	u64 *spte, new_spte;
+	pte_t *ptep = (pte_t *)data;
+	pfn_t new_pfn;
+
+	new_pfn = pte_pfn(ptep_val(ptep));
+	spte = rmap_next(kvm, rmapp, NULL);
+	while (spte) {
+		BUG_ON(!is_shadow_present_pte(*spte));
+		rmap_printk("kvm_set_pte_rmapp: spte %p %llx\n", spte, *spte);
+		need_flush = 1;
+		if (pte_write(ptep_val(ptep))) {
+			rmap_remove(kvm, spte);
+			__set_spte(spte, shadow_trap_nonpresent_pte);
+			spte = rmap_next(kvm, rmapp, NULL);
+		} else {
+			new_spte = *spte &~ (PT64_BASE_ADDR_MASK);
+			new_spte |= new_pfn << PAGE_SHIFT;
+
+			if (!pte_write(ptep_val(ptep))) {
+				new_spte &= ~PT_WRITABLE_MASK;
+				new_spte &= ~SPTE_HOST_WRITEABLE;
+				if (is_writeble_pte(*spte))
+					kvm_set_pfn_dirty(spte_to_pfn(*spte));
+			}
+			__set_spte(spte, new_spte);
+			spte = rmap_next(kvm, rmapp, spte);
+		}
+	}
+	if (need_flush)
+		kvm_flush_remote_tlbs(kvm);
+
+	return 0;
+}
+
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp))
+			  unsigned long data,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 unsigned long data))
 {
 	int i, j;
 	int retval = 0;
@@ -786,13 +832,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 		if (hva >= start && hva < end) {
 			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
 
-			retval |= handler(kvm, &memslot->rmap[gfn_offset]);

arch/x86/kvm/mmu.c static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, · line 843

Quoted 10 lines
arch/x86/kvm/mmu.c +4 -1
+			retval |= handler(kvm, &memslot->rmap[gfn_offset],
+					  data);
 
 			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);
+					&memslot->lpage_info[j][idx].rmap_pde,
+					data);

If change_pte is called to modify a largepage pte, and the shadow has that largepage mapped with 4k sptes, you'll set the wrong pfn. That is, the patch does not attempt to handle different page sizes properly.

So either disable change_pte updates to non-4k host vmas (making it explictly it does not handle different pagesizes), or handle largepages properly, although i don't see any use for change_pte or largepage mappings?

Quoted 77 lines
arch/x86/kvm/mmu.c +10 -4
 			}
 		}
 	}
@@ -802,10 +850,16 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
 {
-	return kvm_handle_hva(kvm, hva, kvm_unmap_rmapp);
+	return kvm_handle_hva(kvm, hva, 0, kvm_unmap_rmapp);
+}
+
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
+{
+	kvm_handle_hva(kvm, hva, (unsigned long)&pte, kvm_set_pte_rmapp);
 }
 
-static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp)
+static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			 unsigned long data)
 {
 	u64 *spte;
 	int young = 0;
@@ -841,13 +895,13 @@ static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
 	gfn = unalias_gfn(vcpu->kvm, gfn);
 	rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level);
 
-	kvm_unmap_rmapp(vcpu->kvm, rmapp);
+	kvm_unmap_rmapp(vcpu->kvm, rmapp, 0);
 	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);
+	return kvm_handle_hva(kvm, hva, 0, kvm_age_rmapp);
 }
 
 #ifdef MMU_DEBUG
virt/kvm/kvm_main.c +14 -0
index cb7966f..c9a88af 100644
@@ -240,6 +240,19 @@ static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn,
 
 }
 
+static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn,
+					struct mm_struct *mm,
+					unsigned long address,
+					pte_t pte)
+{
+	struct kvm *kvm = mmu_notifier_to_kvm(mn);
+
+	spin_lock(&kvm->mmu_lock);
+	kvm->mmu_notifier_seq++;
+	kvm_set_spte_hva(kvm, address, pte);
+	spin_unlock(&kvm->mmu_lock);
+}
+
 static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 						    struct mm_struct *mm,
 						    unsigned long start,
@@ -319,6 +332,7 @@ static const struct mmu_notifier_ops kvm_mmu_notifier_ops = {
 	.invalidate_range_start	= kvm_mmu_notifier_invalidate_range_start,
 	.invalidate_range_end	= kvm_mmu_notifier_invalidate_range_end,
 	.clear_flush_young	= kvm_mmu_notifier_clear_flush_young,
+	.change_pte		= kvm_mmu_notifier_change_pte,
 	.release		= kvm_mmu_notifier_release,
 };
 #endif /* CONFIG_MMU_NOTIFIER && KVM_ARCH_WANT_MMU_NOTIFIER */
Signature · 6 lines
1.5.6.5

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#5 lore raw

Re: [PATCH 3/3] add support for change_pte mmu notifiers

Quoted 121 lines · Marcelo Tosatti
Quoted 104 lines · Izik Eidus

this is needed for kvm if it want ksm to directly map pages into its shadow page tables.

---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |   70 ++++++++++++++++++++++++++++++++++----
 virt/kvm/kvm_main.c             |   14 ++++++++
 3 files changed, 77 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 6046e6f..594d131 100644
@@ -797,6 +797,7 @@ asmlinkage void kvm_handle_fault_on_reboot(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
arch/x86/kvm/mmu.c +49 -3
index a7151b8..3fd19f2 100644
@@ -282,6 +282,11 @@ static pfn_t spte_to_pfn(u64 pte)
 	return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
 }
 
+static pte_t ptep_val(pte_t *ptep)
+{
+	return *ptep;
+}
+
 static gfn_t pse36_gfn_delta(u32 gpte)
 {
 	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
@@ -748,7 +753,8 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 	return write_protected;
 }
 
-static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
+static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			   unsigned long data)
 {
 	u64 *spte;
 	int need_tlb_flush = 0;
@@ -763,8 +769,48 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
 	return need_tlb_flush;
 }
 
+static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			     unsigned long data)
+{
+	int need_flush = 0;
+	u64 *spte, new_spte;
+	pte_t *ptep = (pte_t *)data;
+	pfn_t new_pfn;
+
+	new_pfn = pte_pfn(ptep_val(ptep));
+	spte = rmap_next(kvm, rmapp, NULL);
+	while (spte) {
+		BUG_ON(!is_shadow_present_pte(*spte));
+		rmap_printk("kvm_set_pte_rmapp: spte %p %llx\n", spte, *spte);
+		need_flush = 1;
+		if (pte_write(ptep_val(ptep))) {
+			rmap_remove(kvm, spte);
+			__set_spte(spte, shadow_trap_nonpresent_pte);
+			spte = rmap_next(kvm, rmapp, NULL);
+		} else {
+			new_spte = *spte &~ (PT64_BASE_ADDR_MASK);
+			new_spte |= new_pfn << PAGE_SHIFT;
+
+			if (!pte_write(ptep_val(ptep))) {
+				new_spte &= ~PT_WRITABLE_MASK;
+				new_spte &= ~SPTE_HOST_WRITEABLE;
+				if (is_writeble_pte(*spte))
+					kvm_set_pfn_dirty(spte_to_pfn(*spte));
+			}
+			__set_spte(spte, new_spte);
+			spte = rmap_next(kvm, rmapp, spte);
+		}
+	}
+	if (need_flush)
+		kvm_flush_remote_tlbs(kvm);
+
+	return 0;
+}
+
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp))
+			  unsigned long data,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 unsigned long data))
 {
 	int i, j;
 	int retval = 0;
@@ -786,13 +832,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 		if (hva >= start && hva < end) {
 			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
 
-			retval |= handler(kvm, &memslot->rmap[gfn_offset]);
+			retval |= handler(kvm, &memslot->rmap[gfn_offset],

arch/x86/kvm/mmu.c static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, · line 844

Quoted 10 lines
arch/x86/kvm/mmu.c +3 -1
+					  data);
 
 			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);
+					&memslot->lpage_info[j][idx].rmap_pde,
+					data);
 

If change_pte is called to modify a largepage pte, and the shadow has that largepage mapped with 4k sptes, you'll set the wrong pfn. That is, the patch does not attempt to handle different page sizes properly.

Quoted 5 lines

So either disable change_pte updates to non-4k host vmas (making it explictly it does not handle different pagesizes), or handle largepages properly, although i don't see any use for change_pte or largepage mappings?

change_pte doesn't get called on 4k pages...
So adding commet to this function saying it is working just on 4k pages would be enough ?

#6 lore raw

Re: [PATCH 3/3] add support for change_pte mmu notifiers

Quoted 123 lines · Izik Eidus
Quoted 116 lines · Marcelo Tosatti
Quoted 99 lines · Izik Eidus

this is needed for kvm if it want ksm to directly map pages into its shadow page tables.

---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |   70 ++++++++++++++++++++++++++++++++++----
 virt/kvm/kvm_main.c             |   14 ++++++++
 3 files changed, 77 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 6046e6f..594d131 100644
@@ -797,6 +797,7 @@ asmlinkage void kvm_handle_fault_on_reboot(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
arch/x86/kvm/mmu.c +46 -1
index a7151b8..3fd19f2 100644
@@ -282,6 +282,11 @@ static pfn_t spte_to_pfn(u64 pte)
 	return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
 }
 +static pte_t ptep_val(pte_t *ptep)
+{
+	return *ptep;
+}
+
 static gfn_t pse36_gfn_delta(u32 gpte)
 {
 	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
@@ -748,7 +753,8 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 	return write_protected;
 }
 -static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
+static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			   unsigned long data)
 {
 	u64 *spte;
 	int need_tlb_flush = 0;
@@ -763,8 +769,48 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
 	return need_tlb_flush;
 }
 +static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			     unsigned long data)
+{
+	int need_flush = 0;
+	u64 *spte, new_spte;
+	pte_t *ptep = (pte_t *)data;
+	pfn_t new_pfn;
+
+	new_pfn = pte_pfn(ptep_val(ptep));
+	spte = rmap_next(kvm, rmapp, NULL);
+	while (spte) {
+		BUG_ON(!is_shadow_present_pte(*spte));
+		rmap_printk("kvm_set_pte_rmapp: spte %p %llx\n", spte, *spte);
+		need_flush = 1;
+		if (pte_write(ptep_val(ptep))) {
+			rmap_remove(kvm, spte);
+			__set_spte(spte, shadow_trap_nonpresent_pte);
+			spte = rmap_next(kvm, rmapp, NULL);
+		} else {
+			new_spte = *spte &~ (PT64_BASE_ADDR_MASK);
+			new_spte |= new_pfn << PAGE_SHIFT;
+
+			if (!pte_write(ptep_val(ptep))) {
+				new_spte &= ~PT_WRITABLE_MASK;
+				new_spte &= ~SPTE_HOST_WRITEABLE;
+				if (is_writeble_pte(*spte))
+					kvm_set_pfn_dirty(spte_to_pfn(*spte));
+			}
+			__set_spte(spte, new_spte);
+			spte = rmap_next(kvm, rmapp, spte);
+		}
+	}
+	if (need_flush)
+		kvm_flush_remote_tlbs(kvm);
+
+	return 0;
+}
+
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp))
+			  unsigned long data,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 unsigned long data))
 {
 	int i, j;
 	int retval = 0;
@@ -786,13 +832,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 		if (hva >= start && hva < end) {
 			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
 -			retval |= handler(kvm, &memslot->rmap[gfn_offset]);

arch/x86/kvm/mmu.c static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, · line 843

Quoted 10 lines
arch/x86/kvm/mmu.c +4 -1
+			retval |= handler(kvm, &memslot->rmap[gfn_offset],
+					  data);
  			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);
+					&memslot->lpage_info[j][idx].rmap_pde,
+					data);
 

If change_pte is called to modify a largepage pte, and the shadow has that largepage mapped with 4k sptes, you'll set the wrong pfn. That is, the patch does not attempt to handle different page sizes properly.

Quoted 5 lines

So either disable change_pte updates to non-4k host vmas (making it explictly it does not handle different pagesizes), or handle largepages properly, although i don't see any use for change_pte or largepage mappings?

Quoted 3 lines

change_pte doesn't get called on 4k pages...
So adding commet to this function saying it is working just on 4k pages would be enough ?

It would be better to fail/WARN on non-4k host ptes (can't it?), but if that is not possible i think a comment would be enough.

author
#7 lore raw

Re: [PATCH 3/3] add support for change_pte mmu notifiers

Quoted 127 lines · Marcelo Tosatti
Quoted 125 lines · Izik Eidus
Quoted 123 lines · Marcelo Tosatti
Quoted 100 lines · Izik Eidus

this is needed for kvm if it want ksm to directly map pages into its shadow page tables.

---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |   70 ++++++++++++++++++++++++++++++++++----
 virt/kvm/kvm_main.c             |   14 ++++++++
 3 files changed, 77 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
index 6046e6f..594d131 100644
@@ -797,6 +797,7 @@ asmlinkage void kvm_handle_fault_on_reboot(void);
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
 int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
 int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
arch/x86/kvm/mmu.c +47 -1
index a7151b8..3fd19f2 100644
@@ -282,6 +282,11 @@ static pfn_t spte_to_pfn(u64 pte)
 	return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
 }
 +static pte_t ptep_val(pte_t *ptep)
+{
+	return *ptep;
+}
+
 static gfn_t pse36_gfn_delta(u32 gpte)
 {
 	int shift = 32 - PT32_DIR_PSE36_SHIFT - PAGE_SHIFT;
@@ -748,7 +753,8 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn)
 	return write_protected;
 }
 -static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
+static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			   unsigned long data)
 {
 	u64 *spte;
 	int need_tlb_flush = 0;
@@ -763,8 +769,48 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp)
 	return need_tlb_flush;
 }
 +static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
+			     unsigned long data)
+{
+	int need_flush = 0;
+	u64 *spte, new_spte;
+	pte_t *ptep = (pte_t *)data;
+	pfn_t new_pfn;
+
+	new_pfn = pte_pfn(ptep_val(ptep));
+	spte = rmap_next(kvm, rmapp, NULL);
+	while (spte) {
+		BUG_ON(!is_shadow_present_pte(*spte));
+		rmap_printk("kvm_set_pte_rmapp: spte %p %llx\n", spte, *spte);
+		need_flush = 1;
+		if (pte_write(ptep_val(ptep))) {
+			rmap_remove(kvm, spte);
+			__set_spte(spte, shadow_trap_nonpresent_pte);
+			spte = rmap_next(kvm, rmapp, NULL);
+		} else {
+			new_spte = *spte &~ (PT64_BASE_ADDR_MASK);
+			new_spte |= new_pfn << PAGE_SHIFT;
+
+			if (!pte_write(ptep_val(ptep))) {
+				new_spte &= ~PT_WRITABLE_MASK;
+				new_spte &= ~SPTE_HOST_WRITEABLE;
+				if (is_writeble_pte(*spte))
+					kvm_set_pfn_dirty(spte_to_pfn(*spte));
+			}
+			__set_spte(spte, new_spte);
+			spte = rmap_next(kvm, rmapp, spte);
+		}
+	}
+	if (need_flush)
+		kvm_flush_remote_tlbs(kvm);
+
+	return 0;
+}
+
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp))
+			  unsigned long data,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 unsigned long data))
 {
 	int i, j;
 	int retval = 0;
@@ -786,13 +832,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 		if (hva >= start && hva < end) {
 			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
 -			retval |= handler(kvm, &memslot->rmap[gfn_offset]);
+			retval |= handler(kvm, &memslot->rmap[gfn_offset],

arch/x86/kvm/mmu.c static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, · line 844

Quoted 10 lines
arch/x86/kvm/mmu.c +3 -1
+					  data);
  			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);
+					&memslot->lpage_info[j][idx].rmap_pde,
+					data);
 
 

If change_pte is called to modify a largepage pte, and the shadow has that largepage mapped with 4k sptes, you'll set the wrong pfn. That is, the patch does not attempt to handle different page sizes properly.

So either disable change_pte updates to non-4k host vmas (making it explictly it does not handle different pagesizes), or handle largepages properly, although i don't see any use for change_pte or largepage mappings?

Quoted 8 lines
Quoted 3 lines

change_pte doesn't get called on 4k pages...
So adding commet to this function saying it is working just on 4k pages would be enough ?

It would be better to fail/WARN on non-4k host ptes (can't it?), but if that is not possible i think a comment would be enough.

Ok so you want is_large_pte() check on data right? (and WARN() there..)

#8 lore raw

Re: [PATCH 3/3] add support for change_pte mmu notifiers

Quoted 14 lines · Izik Eidus
Quoted 14 lines
Quoted 13 lines
Quoted 9 lines

If change_pte is called to modify a largepage pte, and the shadow has that largepage mapped with 4k sptes, you'll set the wrong pfn. That is, the patch does not attempt to handle different page sizes properly.

So either disable change_pte updates to non-4k host vmas (making it explictly it does not handle different pagesizes), or handle largepages properly, although i don't see any use for change_pte or largepage mappings?

change_pte doesn't get called on 4k pages...
So adding commet to this function saying it is working just on 4k pages would be enough ?

Quoted 5 lines
Quoted 4 lines

It would be better to fail/WARN on non-4k host ptes (can't it?), but if that is not possible i think a comment would be enough.

Ok so you want is_large_pte() check on data right? (and WARN() there..)

pte_huge(pte), yeah.

#9 lore raw

Re: [PATCH 2/3] add SPTE_HOST_WRITEABLE flag to the shadow ptes

In reply to Izik Eidus
Quoted 26 lines · Izik Eidus

this flag notify that the host physical page we are pointing to from the spte is write protected, and therefore we cant change its access to be write unless we run get_user_pages(write = 1).

(this is needed for change_pte support in kvm)

---
 arch/x86/kvm/mmu.c         |   15 +++++++++++----
 arch/x86/kvm/paging_tmpl.h |   18 +++++++++++++++---
 2 files changed, 26 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu.c +2 -0
index 62d2f86..a7151b8 100644
@@ -156,6 +156,8 @@ module_param(oos_shadow, bool, 0644);
 #define CREATE_TRACE_POINTS
 #include "mmutrace.h"
 
+#define SPTE_HOST_WRITEABLE (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
+
 #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
 
 struct kvm_rmap_desc {

arch/x86/kvm/mmu.c static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, · line 1759

Quoted 5 lines
arch/x86/kvm/mmu.c +1 -1
@@ -1754,7 +1756,7 @@ 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)
+		    bool can_unsync, bool reset_host_protection)

bool host_pte_writeable ?

Quoted 71 lines
arch/x86/kvm/mmu.c +8 -3
 {
 	u64 spte;
 	int ret = 0;
@@ -1781,6 +1783,9 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 		spte |= kvm_x86_ops->get_mt_mask(vcpu, gfn,
 			kvm_is_mmio_pfn(pfn));
 
+	if (reset_host_protection)
+		spte |= SPTE_HOST_WRITEABLE;
+
 	spte |= (u64)pfn << PAGE_SHIFT;
 
 	if ((pte_access & ACC_WRITE_MASK)
@@ -1826,7 +1831,8 @@ 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)
+			 pfn_t pfn, bool speculative,
+			 bool reset_host_protection)
 {
 	int was_rmapped = 0;
 	int was_writeble = is_writeble_pte(*sptep);
@@ -1858,7 +1864,8 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	}
 
 	if (set_spte(vcpu, sptep, pte_access, user_fault, write_fault,
-		      dirty, level, gfn, pfn, speculative, true)) {
+		      dirty, level, gfn, pfn, speculative, true,
+		      reset_host_protection)) {
 		if (write_fault)
 			*ptwrite = 1;
 		kvm_x86_ops->tlb_flush(vcpu);
@@ -1906,7 +1913,7 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
 		if (iterator.level == level) {
 			mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL,
 				     0, write, 1, &pt_write,
-				     level, gfn, pfn, false);
+				     level, gfn, pfn, false, true);
 			++vcpu->stat.pf_fixed;
 			break;
 		}
arch/x86/kvm/paging_tmpl.h +6 -2
index d2fec9c..c9256ee 100644
@@ -273,9 +273,13 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 	if (mmu_notifier_retry(vcpu, vcpu->arch.update_pte.mmu_seq))
 		return;
 	kvm_get_pfn(pfn);
+ 	/*
+ 	 * we call mmu_set_spte() with reset_host_protection = true beacuse that
+ 	 * vcpu->arch.update_pte.pfn was fetched from get_user_pages(write = 1).
+ 	 */
 	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
 		     gpte & PT_DIRTY_MASK, NULL, PT_PAGE_TABLE_LEVEL,
-		     gpte_to_gfn(gpte), pfn, true);
+		     gpte_to_gfn(gpte), pfn, true, true);
 }
 
 /*
@@ -308,7 +312,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 				     user_fault, write_fault,
 				     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
 				     ptwrite, level,
-				     gw->gfn, pfn, false);
+				     gw->gfn, pfn, false, true);
 			break;
 		}
 

arch/x86/kvm/paging_tmpl.h static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, · line 565

Quoted 4 lines
arch/x86/kvm/paging_tmpl.h +1 -0
@@ -558,6 +562,7 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
 	int i, offset, nr_present;
+        bool reset_host_protection;

Tab damaged.

Quoted 10 lines
arch/x86/kvm/paging_tmpl.h +4 -0
 	offset = nr_present = 0;
 
@@ -595,9 +600,16 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 
 		nr_present++;
 		pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte);
+		if (!(sp->spt[i] & SPTE_HOST_WRITEABLE)) {
+ 			pte_access &= ~PT_WRITABLE_MASK;
+			reset_host_protection = 0;
+		} else {

arch/x86/kvm/paging_tmpl.h static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) · line 615

Quoted 10 lines
arch/x86/kvm/paging_tmpl.h +4 -1
+			reset_host_protection = 1;
+		}
 		set_spte(vcpu, &sp->spt[i], pte_access, 0, 0,
 			 is_dirty_gpte(gpte), PT_PAGE_TABLE_LEVEL, gfn,
-			 spte_to_pfn(sp->spt[i]), true, false);
+			 spte_to_pfn(sp->spt[i]), true, false,
+			 reset_host_protection);
 	}
 
 	return !nr_present;

Looks good otherwise.

Quoted 7 lines
Signature · 6 lines
1.5.6.5

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#10 lore raw

Re: [PATCH 2/3] add SPTE_HOST_WRITEABLE flag to the shadow ptes

Quoted 26 lines · Marcelo Tosatti
Quoted 24 lines · Izik Eidus

this flag notify that the host physical page we are pointing to from the spte is write protected, and therefore we cant change its access to be write unless we run get_user_pages(write = 1).

(this is needed for change_pte support in kvm)

---
 arch/x86/kvm/mmu.c         |   15 +++++++++++----
 arch/x86/kvm/paging_tmpl.h |   18 +++++++++++++++---
 2 files changed, 26 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu.c +2 -0
index 62d2f86..a7151b8 100644
@@ -156,6 +156,8 @@ module_param(oos_shadow, bool, 0644);
 #define CREATE_TRACE_POINTS
 #include "mmutrace.h"
 
+#define SPTE_HOST_WRITEABLE (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
+
 #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
 
Quoted 12 lines

arch/x86/kvm/mmu.c static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, · line 1760

Quoted 8 lines
arch/x86/kvm/mmu.c +1 -1
 struct kvm_rmap_desc {
@@ -1754,7 +1756,7 @@ 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)
+		    bool can_unsync, bool reset_host_protection)
 

bool host_pte_writeable ?

Sure.

#11 lore raw

Re: [PATCH 2/3] add SPTE_HOST_WRITEABLE flag to the shadow ptes

In reply to Izik Eidus
Quoted 13 lines · Izik Eidus

this flag notify that the host physical page we are pointing to from the spte is write protected, and therefore we cant change its access to be write unless we run get_user_pages(write = 1).

(this is needed for change_pte support in kvm)

arch/x86/kvm/mmu.c +0 -0
index 62d2f86..a7151b8 100644

arch/x86/kvm/mmu.c module_param(oos_shadow, bool, 0644); · line 161

Quoted 6 lines
arch/x86/kvm/mmu.c +2 -0
@@ -156,6 +156,8 @@ module_param(oos_shadow, bool, 0644);
  #define CREATE_TRACE_POINTS
  #include "mmutrace.h"
 
+#define SPTE_HOST_WRITEABLE (1ULL<<  PT_FIRST_AVAIL_BITS_SHIFT)
+
 

Luckilly, this bit is available on EPT too.

arch/x86/kvm/mmu.c static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) · line 605

Quoted 8 lines
arch/x86/kvm/mmu.c +2 -0
  #define SHADOW_PT_INDEX(addr, level) PT64_I
@@ -595,9 +600,16 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 
  		nr_present++;
  		pte_access = sp->role.access&  FNAME(gpte_access)(vcpu, gpte);
+		if (!(sp->spt[i]&  SPTE_HOST_WRITEABLE)) {
+ 			pte_access&= ~PT_WRITABLE_MASK;
 

pte_access uses ACC_ masks, not PT_ masks.

Signature · 2 lines
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
#12 lore raw

Re: [PATCH 2/3] add SPTE_HOST_WRITEABLE flag to the shadow ptes

In reply to Izik Eidus
Quoted 114 lines · Izik Eidus

this flag notify that the host physical page we are pointing to from the spte is write protected, and therefore we cant change its access to be write unless we run get_user_pages(write = 1).

(this is needed for change_pte support in kvm)

---
 arch/x86/kvm/mmu.c         |   15 +++++++++++----
 arch/x86/kvm/paging_tmpl.h |   18 +++++++++++++++---
 2 files changed, 26 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu.c +11 -4
index 62d2f86..a7151b8 100644
@@ -156,6 +156,8 @@ module_param(oos_shadow, bool, 0644);
 #define CREATE_TRACE_POINTS
 #include "mmutrace.h"
 
+#define SPTE_HOST_WRITEABLE (1ULL << PT_FIRST_AVAIL_BITS_SHIFT)
+
 #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level)
 
 struct kvm_rmap_desc {
@@ -1754,7 +1756,7 @@ 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)
+		    bool can_unsync, bool reset_host_protection)
 {
 	u64 spte;
 	int ret = 0;
@@ -1781,6 +1783,9 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 		spte |= kvm_x86_ops->get_mt_mask(vcpu, gfn,
 			kvm_is_mmio_pfn(pfn));
 
+	if (reset_host_protection)
+		spte |= SPTE_HOST_WRITEABLE;
+
 	spte |= (u64)pfn << PAGE_SHIFT;
 
 	if ((pte_access & ACC_WRITE_MASK)
@@ -1826,7 +1831,8 @@ 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)
+			 pfn_t pfn, bool speculative,
+			 bool reset_host_protection)
 {
 	int was_rmapped = 0;
 	int was_writeble = is_writeble_pte(*sptep);
@@ -1858,7 +1864,8 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	}
 
 	if (set_spte(vcpu, sptep, pte_access, user_fault, write_fault,
-		      dirty, level, gfn, pfn, speculative, true)) {
+		      dirty, level, gfn, pfn, speculative, true,
+		      reset_host_protection)) {
 		if (write_fault)
 			*ptwrite = 1;
 		kvm_x86_ops->tlb_flush(vcpu);
@@ -1906,7 +1913,7 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
 		if (iterator.level == level) {
 			mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL,
 				     0, write, 1, &pt_write,
-				     level, gfn, pfn, false);
+				     level, gfn, pfn, false, true);
 			++vcpu->stat.pf_fixed;
 			break;
 		}
arch/x86/kvm/paging_tmpl.h +8 -2
index d2fec9c..c9256ee 100644
@@ -273,9 +273,13 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
 	if (mmu_notifier_retry(vcpu, vcpu->arch.update_pte.mmu_seq))
 		return;
 	kvm_get_pfn(pfn);
+ 	/*
+ 	 * we call mmu_set_spte() with reset_host_protection = true beacuse that
+ 	 * vcpu->arch.update_pte.pfn was fetched from get_user_pages(write = 1).
+ 	 */
 	mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
 		     gpte & PT_DIRTY_MASK, NULL, PT_PAGE_TABLE_LEVEL,
-		     gpte_to_gfn(gpte), pfn, true);
+		     gpte_to_gfn(gpte), pfn, true, true);
 }
 
 /*
@@ -308,7 +312,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 				     user_fault, write_fault,
 				     gw->ptes[gw->level-1] & PT_DIRTY_MASK,
 				     ptwrite, level,
-				     gw->gfn, pfn, false);
+				     gw->gfn, pfn, false, true);
 			break;
 		}
 
@@ -558,6 +562,7 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
 static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 {
 	int i, offset, nr_present;
+        bool reset_host_protection;
 
 	offset = nr_present = 0;
 
@@ -595,9 +600,16 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
 
 		nr_present++;
 		pte_access = sp->role.access & FNAME(gpte_access)(vcpu, gpte);
+		if (!(sp->spt[i] & SPTE_HOST_WRITEABLE)) {

arch/x86/kvm/paging_tmpl.h static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) · line 612

Quoted 10 lines
arch/x86/kvm/paging_tmpl.h +7 -1
+ 			pte_access &= ~PT_WRITABLE_MASK;
+			reset_host_protection = 0;
+		} else {
+			reset_host_protection = 1;
+		}
 		set_spte(vcpu, &sp->spt[i], pte_access, 0, 0,
 			 is_dirty_gpte(gpte), PT_PAGE_TABLE_LEVEL, gfn,
-			 spte_to_pfn(sp->spt[i]), true, false);
+			 spte_to_pfn(sp->spt[i]), true, false,
+			 reset_host_protection);

Why can't you use the writable bit in the spte? So that you can only sync a writeable spte if it was writeable before, in sync_page?

Is there any other need for the extra bit?

Quoted 10 lines
arch/x86/kvm/paging_tmpl.h +0 -0
 	}
 
 	return !nr_present;
Signature · 6 lines
1.5.6.5

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#13 lore raw

Re: [PATCH 2/3] add SPTE_HOST_WRITEABLE flag to the shadow ptes

Quoted 2 lines · Marcelo Tosatti

Why can't you use the writable bit in the spte? So that you can only sync a writeable spte if it was writeable before, in sync_page?

I could, but there we will add overhead for read only gptes that become writable in the guest... If you prefer to fault on the syncing of the guest gpte readonly to gpte writeable I can change it...

What you prefer?

Quoted 2 lines

Is there any other need for the extra bit?

No