KVM: x86/mmu: Use an rwlock for the x86 MMU

Add a read / write lock to be used in place of the MMU spinlock on x86.
The rwlock will enable the TDP MMU to handle page faults, and other
operations in parallel in future commits.

Reviewed-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Ben Gardon <bgardon@google.com>

Message-Id: <20210202185734.1680553-19-bgardon@google.com>
[Introduce virt/kvm/mmu_lock.h - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

3 files changed, 46 insertions, 19 deletions

diff --git a/virt/kvm/dirty_ring.c b/virt/kvm/dirty_ring.c
index 790f17325f8d..7aafefc50aa7 100644
--- a/virt/kvm/dirty_ring.c
+++ b/virt/kvm/dirty_ring.c
@@ -9,6 +9,7 @@
 #include <linux/vmalloc.h>
 #include <linux/kvm_dirty_ring.h>
 #include <trace/events/kvm.h>
+#include "mmu_lock.h"
 
 int __weak kvm_cpu_dirty_log_size(void)
 {
@@ -60,9 +61,9 @@ static void kvm_reset_dirty_gfn(struct kvm *kvm, u32 slot, u64 offset, u64 mask)
 	if (!memslot || (offset + __fls(mask)) >= memslot->npages)
 		return;
 
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	kvm_arch_mmu_enable_log_dirty_pt_masked(kvm, memslot, offset, mask);
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 }
 
 int kvm_dirty_ring_alloc(struct kvm_dirty_ring *ring, int index, u32 size)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 335a1a2b8edc..48ccdf4e3d04 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -58,6 +58,7 @@
 
 #include "coalesced_mmio.h"
 #include "async_pf.h"
+#include "mmu_lock.h"
 #include "vfio.h"
 
 #define CREATE_TRACE_POINTS
@@ -459,13 +460,15 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn,
 	int idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	spin_lock(&kvm->mmu_lock);
+
+	KVM_MMU_LOCK(kvm);
+
 	kvm->mmu_notifier_seq++;
 
 	if (kvm_set_spte_hva(kvm, address, pte))
 		kvm_flush_remote_tlbs(kvm);
 
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 	srcu_read_unlock(&kvm->srcu, idx);
 }
 
@@ -476,7 +479,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 	int need_tlb_flush = 0, idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	/*
 	 * The count increase must become visible at unlock time as no
 	 * spte can be established without taking the mmu_lock and
@@ -489,7 +492,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 	if (need_tlb_flush || kvm->tlbs_dirty)
 		kvm_flush_remote_tlbs(kvm);
 
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 	srcu_read_unlock(&kvm->srcu, idx);
 
 	return 0;
@@ -500,7 +503,7 @@ static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
 {
 	struct kvm *kvm = mmu_notifier_to_kvm(mn);
 
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	/*
 	 * This sequence increase will notify the kvm page fault that
 	 * the page that is going to be mapped in the spte could have
@@ -514,7 +517,7 @@ static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
 	 * in conjunction with the smp_rmb in mmu_notifier_retry().
 	 */
 	kvm->mmu_notifier_count--;
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 
 	BUG_ON(kvm->mmu_notifier_count < 0);
 }
@@ -528,13 +531,13 @@ static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn,
 	int young, idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 
 	young = kvm_age_hva(kvm, start, end);
 	if (young)
 		kvm_flush_remote_tlbs(kvm);
 
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 	srcu_read_unlock(&kvm->srcu, idx);
 
 	return young;
@@ -549,7 +552,7 @@ static int kvm_mmu_notifier_clear_young(struct mmu_notifier *mn,
 	int young, idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	/*
 	 * Even though we do not flush TLB, this will still adversely
 	 * affect performance on pre-Haswell Intel EPT, where there is
@@ -564,7 +567,7 @@ static int kvm_mmu_notifier_clear_young(struct mmu_notifier *mn,
 	 * more sophisticated heuristic later.
 	 */
 	young = kvm_age_hva(kvm, start, end);
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 	srcu_read_unlock(&kvm->srcu, idx);
 
 	return young;
@@ -578,9 +581,9 @@ static int kvm_mmu_notifier_test_young(struct mmu_notifier *mn,
 	int young, idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	young = kvm_test_age_hva(kvm, address);
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 	srcu_read_unlock(&kvm->srcu, idx);
 
 	return young;
@@ -745,7 +748,7 @@ static struct kvm *kvm_create_vm(unsigned long type)
 	if (!kvm)
 		return ERR_PTR(-ENOMEM);
 
-	spin_lock_init(&kvm->mmu_lock);
+	KVM_MMU_LOCK_INIT(kvm);
 	mmgrab(current->mm);
 	kvm->mm = current->mm;
 	kvm_eventfd_init(kvm);
@@ -1525,7 +1528,7 @@ static int kvm_get_dirty_log_protect(struct kvm *kvm, struct kvm_dirty_log *log)
 		dirty_bitmap_buffer = kvm_second_dirty_bitmap(memslot);
 		memset(dirty_bitmap_buffer, 0, n);
 
-		spin_lock(&kvm->mmu_lock);
+		KVM_MMU_LOCK(kvm);
 		for (i = 0; i < n / sizeof(long); i++) {
 			unsigned long mask;
 			gfn_t offset;
@@ -1541,7 +1544,7 @@ static int kvm_get_dirty_log_protect(struct kvm *kvm, struct kvm_dirty_log *log)
 			kvm_arch_mmu_enable_log_dirty_pt_masked(kvm, memslot,
 								offset, mask);
 		}
-		spin_unlock(&kvm->mmu_lock);
+		KVM_MMU_UNLOCK(kvm);
 	}
 
 	if (flush)
@@ -1636,7 +1639,7 @@ static int kvm_clear_dirty_log_protect(struct kvm *kvm,
 	if (copy_from_user(dirty_bitmap_buffer, log->dirty_bitmap, n))
 		return -EFAULT;
 
-	spin_lock(&kvm->mmu_lock);
+	KVM_MMU_LOCK(kvm);
 	for (offset = log->first_page, i = offset / BITS_PER_LONG,
 		 n = DIV_ROUND_UP(log->num_pages, BITS_PER_LONG); n--;
 	     i++, offset += BITS_PER_LONG) {
@@ -1659,7 +1662,7 @@ static int kvm_clear_dirty_log_protect(struct kvm *kvm,
 								offset, mask);
 		}
 	}
-	spin_unlock(&kvm->mmu_lock);
+	KVM_MMU_UNLOCK(kvm);
 
 	if (flush)
 		kvm_arch_flush_remote_tlbs_memslot(kvm, memslot);
diff --git a/virt/kvm/mmu_lock.h b/virt/kvm/mmu_lock.h
new file mode 100644
index 000000000000..9e1308f9734c
--- /dev/null
+++ b/virt/kvm/mmu_lock.h
@@ -0,0 +1,23 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#ifndef KVM_MMU_LOCK_H
+#define KVM_MMU_LOCK_H 1
+
+/*
+ * Architectures can choose whether to use an rwlock or spinlock
+ * for the mmu_lock.  These macros, for use in common code
+ * only, avoids using #ifdefs in places that must deal with
+ * multiple architectures.
+ */
+
+#ifdef KVM_HAVE_MMU_RWLOCK
+#define KVM_MMU_LOCK_INIT(kvm) rwlock_init(&(kvm)->mmu_lock)
+#define KVM_MMU_LOCK(kvm)      write_lock(&(kvm)->mmu_lock)
+#define KVM_MMU_UNLOCK(kvm)    write_unlock(&(kvm)->mmu_lock)
+#else
+#define KVM_MMU_LOCK_INIT(kvm) spin_lock_init(&(kvm)->mmu_lock)
+#define KVM_MMU_LOCK(kvm)      spin_lock(&(kvm)->mmu_lock)
+#define KVM_MMU_UNLOCK(kvm)    spin_unlock(&(kvm)->mmu_lock)
+#endif /* KVM_HAVE_MMU_RWLOCK */
+
+#endif