LKML Archive on lore.kernel.org
 help / color / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>, stable@vger.kernel.org
Subject: [FYI PATCH 3/5] x86/kvm: Cache gfn to pfn translation
Date: Thu, 30 Jan 2020 19:01:54 +0100
Message-ID: <1580407316-11391-4-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1580407316-11391-1-git-send-email-pbonzini@redhat.com>

From: Boris Ostrovsky <boris.ostrovsky@oracle.com>

__kvm_map_gfn()'s call to gfn_to_pfn_memslot() is
* relatively expensive
* in certain cases (such as when done from atomic context) cannot be called

Stashing gfn-to-pfn mapping should help with both cases.

This is part of CVE-2019-3016.

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  1 +
 arch/x86/kvm/x86.c              | 10 +++++
 include/linux/kvm_host.h        |  7 ++-
 include/linux/kvm_types.h       |  9 +++-
 virt/kvm/kvm_main.c             | 98 +++++++++++++++++++++++++++++++++--------
 5 files changed, 103 insertions(+), 22 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index b79cd6a..f48a306e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -689,6 +689,7 @@ struct kvm_vcpu_arch {
 		u64 last_steal;
 		struct gfn_to_hva_cache stime;
 		struct kvm_steal_time steal;
+		struct gfn_to_pfn_cache cache;
 	} st;
 
 	u64 tsc_offset;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 8c93691..0795bc8 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -9088,6 +9088,9 @@ static void fx_init(struct kvm_vcpu *vcpu)
 void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
 {
 	void *wbinvd_dirty_mask = vcpu->arch.wbinvd_dirty_mask;
+	struct gfn_to_pfn_cache *cache = &vcpu->arch.st.cache;
+
+	kvm_release_pfn(cache->pfn, cache->dirty, cache);
 
 	kvmclock_reset(vcpu);
 
@@ -9761,11 +9764,18 @@ int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 
 void kvm_arch_memslots_updated(struct kvm *kvm, u64 gen)
 {
+	struct kvm_vcpu *vcpu;
+	int i;
+
 	/*
 	 * memslots->generation has been incremented.
 	 * mmio generation may have reached its maximum value.
 	 */
 	kvm_mmu_invalidate_mmio_sptes(kvm, gen);
+
+	/* Force re-initialization of steal_time cache */
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vcpu_kick(vcpu);
 }
 
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 0cb78f5..71cb9cc 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -723,6 +723,7 @@ kvm_pfn_t __gfn_to_pfn_memslot(struct kvm_memory_slot *slot, gfn_t gfn,
 void kvm_set_pfn_accessed(kvm_pfn_t pfn);
 void kvm_get_pfn(kvm_pfn_t pfn);
 
+void kvm_release_pfn(kvm_pfn_t pfn, bool dirty, struct gfn_to_pfn_cache *cache);
 int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
 			int len);
 int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
@@ -775,10 +776,12 @@ int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 kvm_pfn_t kvm_vcpu_gfn_to_pfn_atomic(struct kvm_vcpu *vcpu, gfn_t gfn);
 kvm_pfn_t kvm_vcpu_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn);
 int kvm_vcpu_map(struct kvm_vcpu *vcpu, gpa_t gpa, struct kvm_host_map *map);
-int kvm_map_gfn(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map);
+int kvm_map_gfn(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map,
+		struct gfn_to_pfn_cache *cache, bool atomic);
 struct page *kvm_vcpu_gfn_to_page(struct kvm_vcpu *vcpu, gfn_t gfn);
 void kvm_vcpu_unmap(struct kvm_vcpu *vcpu, struct kvm_host_map *map, bool dirty);
-int kvm_unmap_gfn(struct kvm_vcpu *vcpu, struct kvm_host_map *map, bool dirty);
+int kvm_unmap_gfn(struct kvm_vcpu *vcpu, struct kvm_host_map *map,
+		  struct gfn_to_pfn_cache *cache, bool dirty, bool atomic);
 unsigned long kvm_vcpu_gfn_to_hva(struct kvm_vcpu *vcpu, gfn_t gfn);
 unsigned long kvm_vcpu_gfn_to_hva_prot(struct kvm_vcpu *vcpu, gfn_t gfn, bool *writable);
 int kvm_vcpu_read_guest_page(struct kvm_vcpu *vcpu, gfn_t gfn, void *data, int offset,
diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h
index 1c88e69..68e84cf 100644
--- a/include/linux/kvm_types.h
+++ b/include/linux/kvm_types.h
@@ -18,7 +18,7 @@
 
 enum kvm_mr_change;
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 /*
  * Address types:
@@ -51,4 +51,11 @@ struct gfn_to_hva_cache {
 	struct kvm_memory_slot *memslot;
 };
 
+struct gfn_to_pfn_cache {
+	u64 generation;
+	gfn_t gfn;
+	kvm_pfn_t pfn;
+	bool dirty;
+};
+
 #endif /* __KVM_TYPES_H__ */
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 9ef58a2..67eb302 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1821,27 +1821,72 @@ struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn)
 }
 EXPORT_SYMBOL_GPL(gfn_to_page);
 
+void kvm_release_pfn(kvm_pfn_t pfn, bool dirty, struct gfn_to_pfn_cache *cache)
+{
+	if (pfn == 0)
+		return;
+
+	if (cache)
+		cache->pfn = cache->gfn = 0;
+
+	if (dirty)
+		kvm_release_pfn_dirty(pfn);
+	else
+		kvm_release_pfn_clean(pfn);
+}
+
+static void kvm_cache_gfn_to_pfn(struct kvm_memory_slot *slot, gfn_t gfn,
+				 struct gfn_to_pfn_cache *cache, u64 gen)
+{
+	kvm_release_pfn(cache->pfn, cache->dirty, cache);
+
+	cache->pfn = gfn_to_pfn_memslot(slot, gfn);
+	cache->gfn = gfn;
+	cache->dirty = false;
+	cache->generation = gen;
+}
+
 static int __kvm_map_gfn(struct kvm_memslots *slots, gfn_t gfn,
-			 struct kvm_host_map *map)
+			 struct kvm_host_map *map,
+			 struct gfn_to_pfn_cache *cache,
+			 bool atomic)
 {
 	kvm_pfn_t pfn;
 	void *hva = NULL;
 	struct page *page = KVM_UNMAPPED_PAGE;
 	struct kvm_memory_slot *slot = __gfn_to_memslot(slots, gfn);
+	u64 gen = slots->generation;
 
 	if (!map)
 		return -EINVAL;
 
-	pfn = gfn_to_pfn_memslot(slot, gfn);
+	if (cache) {
+		if (!cache->pfn || cache->gfn != gfn ||
+			cache->generation != gen) {
+			if (atomic)
+				return -EAGAIN;
+			kvm_cache_gfn_to_pfn(slot, gfn, cache, gen);
+		}
+		pfn = cache->pfn;
+	} else {
+		if (atomic)
+			return -EAGAIN;
+		pfn = gfn_to_pfn_memslot(slot, gfn);
+	}
 	if (is_error_noslot_pfn(pfn))
 		return -EINVAL;
 
 	if (pfn_valid(pfn)) {
 		page = pfn_to_page(pfn);
-		hva = kmap(page);
+		if (atomic)
+			hva = kmap_atomic(page);
+		else
+			hva = kmap(page);
 #ifdef CONFIG_HAS_IOMEM
-	} else {
+	} else if (!atomic) {
 		hva = memremap(pfn_to_hpa(pfn), PAGE_SIZE, MEMREMAP_WB);
+	} else {
+		return -EINVAL;
 #endif
 	}
 
@@ -1856,20 +1901,25 @@ static int __kvm_map_gfn(struct kvm_memslots *slots, gfn_t gfn,
 	return 0;
 }
 
-int kvm_map_gfn(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map)
+int kvm_map_gfn(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map,
+		struct gfn_to_pfn_cache *cache, bool atomic)
 {
-	return __kvm_map_gfn(kvm_memslots(vcpu->kvm), gfn, map);
+	return __kvm_map_gfn(kvm_memslots(vcpu->kvm), gfn, map,
+			cache, atomic);
 }
 EXPORT_SYMBOL_GPL(kvm_map_gfn);
 
 int kvm_vcpu_map(struct kvm_vcpu *vcpu, gfn_t gfn, struct kvm_host_map *map)
 {
-	return __kvm_map_gfn(kvm_vcpu_memslots(vcpu), gfn, map);
+	return __kvm_map_gfn(kvm_vcpu_memslots(vcpu), gfn, map,
+		NULL, false);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_map);
 
 static void __kvm_unmap_gfn(struct kvm_memory_slot *memslot,
-			struct kvm_host_map *map, bool dirty)
+			struct kvm_host_map *map,
+			struct gfn_to_pfn_cache *cache,
+			bool dirty, bool atomic)
 {
 	if (!map)
 		return;
@@ -1877,34 +1927,44 @@ static void __kvm_unmap_gfn(struct kvm_memory_slot *memslot,
 	if (!map->hva)
 		return;
 
-	if (map->page != KVM_UNMAPPED_PAGE)
-		kunmap(map->page);
+	if (map->page != KVM_UNMAPPED_PAGE) {
+		if (atomic)
+			kunmap_atomic(map->hva);
+		else
+			kunmap(map->page);
+	}
 #ifdef CONFIG_HAS_IOMEM
-	else
+	else if (!atomic)
 		memunmap(map->hva);
+	else
+		WARN_ONCE(1, "Unexpected unmapping in atomic context");
 #endif
 
-	if (dirty) {
+	if (dirty)
 		mark_page_dirty_in_slot(memslot, map->gfn);
-		kvm_release_pfn_dirty(map->pfn);
-	} else {
-		kvm_release_pfn_clean(map->pfn);
-	}
+
+	if (cache)
+		cache->dirty |= dirty;
+	else
+		kvm_release_pfn(map->pfn, dirty, NULL);
 
 	map->hva = NULL;
 	map->page = NULL;
 }
 
-int kvm_unmap_gfn(struct kvm_vcpu *vcpu, struct kvm_host_map *map, bool dirty)
+int kvm_unmap_gfn(struct kvm_vcpu *vcpu, struct kvm_host_map *map, 
+		  struct gfn_to_pfn_cache *cache, bool dirty, bool atomic)
 {
-	__kvm_unmap_gfn(gfn_to_memslot(vcpu->kvm, map->gfn), map, dirty);
+	__kvm_unmap_gfn(gfn_to_memslot(vcpu->kvm, map->gfn), map,
+			cache, dirty, atomic);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(kvm_unmap_gfn);
 
 void kvm_vcpu_unmap(struct kvm_vcpu *vcpu, struct kvm_host_map *map, bool dirty)
 {
-	__kvm_unmap_gfn(kvm_vcpu_gfn_to_memslot(vcpu, map->gfn), map, dirty);
+	__kvm_unmap_gfn(kvm_vcpu_gfn_to_memslot(vcpu, map->gfn), map, NULL,
+			dirty, false);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_unmap);
 
-- 
1.8.3.1


  parent reply index

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-30 18:01 [FYI PATCH 0/5] Missing TLB flushes Paolo Bonzini
2020-01-30 18:01 ` [FYI PATCH 1/5] x86/kvm: Be careful not to clear KVM_VCPU_FLUSH_TLB bit Paolo Bonzini
2020-01-30 18:01 ` [FYI PATCH 2/5] x86/kvm: Introduce kvm_(un)map_gfn() Paolo Bonzini
2020-01-30 18:01 ` Paolo Bonzini [this message]
2020-01-30 18:01 ` [FYI PATCH 4/5] x86/KVM: Make sure KVM_VCPU_FLUSH_TLB flag is not missed Paolo Bonzini
2020-01-30 18:01 ` [FYI PATCH 5/5] x86/KVM: Clean up host's steal time structure Paolo Bonzini
2020-02-01  5:53 ` [FYI PATCH 0/5] Missing TLB flushes Wanpeng Li

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1580407316-11391-4-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=boris.ostrovsky@oracle.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git
	git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git
	git clone --mirror https://lore.kernel.org/lkml/9 lkml/git/9.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org
	public-inbox-index lkml

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git