KVM: Fully serialize gfn=>pfn cache refresh via mutex

Protect gfn=>pfn cache refresh with a mutex to fully serialize refreshes.
The refresh logic doesn't protect against

- concurrent unmaps, or refreshes with different GPAs (which may or may not
  happen in practice, for example if a cache is only used under vcpu->mutex;
  but it's allowed in the code)

- a false negative on the memslot generation.  If the first refresh sees
  a stale memslot generation, it will refresh the hva and generation before
  moving on to the hva=>pfn translation.  If it then drops gpc->lock, a
  different user of the cache can come along, acquire gpc->lock, see that
  the memslot generation is fresh, and skip the hva=>pfn update due to the
  userspace address also matching (because it too was updated).

The refresh path can already sleep during hva=>pfn resolution, so wrap
the refresh with a mutex to ensure that any given refresh runs to
completion before other callers can start their refresh.

Cc: stable@vger.kernel.org
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-7-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h
index ac1ebb3..f328a01 100644 (file)
--- a/include/linux/kvm_types.h
+++ b/include/linux/kvm_types.h
@@ -19,6 +19,7 @@ struct kvm_memslots;
 enum kvm_mr_change;
 
 #include <linux/bits.h>
+#include <linux/mutex.h>
 #include <linux/types.h>
 #include <linux/spinlock_types.h>
 
@@ -69,6 +70,7 @@ struct gfn_to_pfn_cache {
        struct kvm_vcpu *vcpu;
        struct list_head list;
        rwlock_t lock;
+       struct mutex refresh_lock;
        void *khva;
        kvm_pfn_t pfn;
        enum pfn_cache_usage usage;

diff --git a/virt/kvm/pfncache.c b/virt/kvm/pfncache.c
index 05cb0bc..f610d39 100644 (file)
--- a/virt/kvm/pfncache.c
+++ b/virt/kvm/pfncache.c
@@ -157,6 +157,13 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
        if (page_offset + len > PAGE_SIZE)
                return -EINVAL;
 
+       /*
+        * If another task is refreshing the cache, wait for it to complete.
+        * There is no guarantee that concurrent refreshes will see the same
+        * gpa, memslots generation, etc..., so they must be fully serialized.
+        */
+       mutex_lock(&gpc->refresh_lock);
+
        write_lock_irq(&gpc->lock);
 
        old_pfn = gpc->pfn;
@@ -248,6 +255,8 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
  out:
        write_unlock_irq(&gpc->lock);
 
+       mutex_unlock(&gpc->refresh_lock);
+
        gpc_release_pfn_and_khva(kvm, old_pfn, old_khva);
 
        return ret;
@@ -259,6 +268,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct gfn_to_pfn_cache *gpc)
        void *old_khva;
        kvm_pfn_t old_pfn;
 
+       mutex_lock(&gpc->refresh_lock);
        write_lock_irq(&gpc->lock);
 
        gpc->valid = false;
@@ -274,6 +284,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct gfn_to_pfn_cache *gpc)
        gpc->pfn = KVM_PFN_ERR_FAULT;
 
        write_unlock_irq(&gpc->lock);
+       mutex_unlock(&gpc->refresh_lock);
 
        gpc_release_pfn_and_khva(kvm, old_pfn, old_khva);
 }
@@ -288,6 +299,7 @@ int kvm_gfn_to_pfn_cache_init(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 
        if (!gpc->active) {
                rwlock_init(&gpc->lock);
+               mutex_init(&gpc->refresh_lock);
 
                gpc->khva = NULL;
                gpc->pfn = KVM_PFN_ERR_FAULT;