KVM: Dirty ring interface

Remove kvm_read_guest_atomic() because it's not used anywhere.

Signed-off-by Peter Xu <[email protected]>

---
 include/linux/kvm_host.h |  2 --
 virt/kvm/kvm_main.c      | 11 -----------
 2 files changed, 13 deletions(-)

include/linux/kvm_host.h +0 -2

index d41c521a39da..2ea1ea79befd 100644
@@ -730,8 +730,6 @@ void kvm_get_pfn(kvm_pfn_t pfn);
 
 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,
-			  unsigned long len);
 int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
 int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 			   void *data, unsigned long len);

virt/kvm/kvm_main.c +0 -4

index 13efc291b1c7..7ee28af9eb48 100644
@@ -2039,17 +2039,6 @@ static int __kvm_read_guest_atomic(struct kvm_memory_slot *slot, gfn_t gfn,
 	return 0;
 }
 
-int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
-			  unsigned long len)
-{
-	gfn_t gfn = gpa >> PAGE_SHIFT;

It would be clearer to dump the return value to know easily on whether did we go through the fast path for handling current page fault. Remove the old two last parameters because after all the old/new sptes were dumped in the same line.

Signed-off-by Peter Xu <[email protected]>

---
 arch/x86/kvm/mmutrace.h | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

arch/x86/kvm/mmutrace.h +0 -4

index 7ca8831c7d1a..09bdc5c91650 100644
@@ -244,9 +244,6 @@ TRACE_EVENT(
 		  __entry->access)
 );
 
-#define __spte_satisfied(__spte)				\
-	(__entry->retry && is_writable_pte(__entry->__spte))
-
 TRACE_EVENT(
 	fast_page_fault,
 	TP_PROTO(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code,

@@ -274,12 +271,10 @@ TRACE_EVENT(
 	),
 
 	TP_printk("vcpu %d gva %lx error_code %s sptep %p old %#llx"
-		  " new %llx spurious %d fixed %d", __entry->vcpu_id,

Originally, we have three code paths that can dirty a page without vcpu context for X86:

- init_rmode_identity_map
- init_rmode_tss
- kvmgt_rw_gpa

init_rmode_identity_map and init_rmode_tss will be setup on destination VM no matter what (and the guest cannot even see them), so it does not make sense to track them at all.

To do this, a new parameter is added to kvm_[write|clear]_guest_page() to show whether we would like to track dirty bits for the operations. With that, pass in "false" to this new parameter for any guest memory write of the ioctls (KVM_SET_TSS_ADDR, KVM_SET_IDENTITY_MAP_ADDR).

Signed-off-by Peter Xu <[email protected]>

---
 arch/x86/kvm/vmx/vmx.c   | 18 ++++++++++--------
 include/linux/kvm_host.h |  5 +++--
 virt/kvm/kvm_main.c      | 25 ++++++++++++++++---------
 3 files changed, 29 insertions(+), 19 deletions(-)

arch/x86/kvm/vmx/vmx.c +10 -8

index 04a8212704c1..1ff5a428f489 100644
@@ -3452,24 +3452,24 @@ static int init_rmode_tss(struct kvm *kvm)
 
 	idx = srcu_read_lock(&kvm->srcu);
 	fn = to_kvm_vmx(kvm)->tss_addr >> PAGE_SHIFT;
-	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
+	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE, false);
 	if (r < 0)
 		goto out;
 	data = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE;
 	r = kvm_write_guest_page(kvm, fn++, &data,
-			TSS_IOPB_BASE_OFFSET, sizeof(u16));
+				 TSS_IOPB_BASE_OFFSET, sizeof(u16), false);
 	if (r < 0)
 		goto out;
-	r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE);
+	r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE, false);
 	if (r < 0)
 		goto out;
-	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
+	r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE, false);
 	if (r < 0)
 		goto out;
 	data = ~0;
 	r = kvm_write_guest_page(kvm, fn, &data,
 				 RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
-				 sizeof(u8));
+				 sizeof(u8), false);
 out:
 	srcu_read_unlock(&kvm->srcu, idx);
 	return r;

@@ -3498,7 +3498,7 @@ static int init_rmode_identity_map(struct kvm *kvm)
 		goto out2;
 
 	idx = srcu_read_lock(&kvm->srcu);
-	r = kvm_clear_guest_page(kvm, identity_map_pfn, 0, PAGE_SIZE);
+	r = kvm_clear_guest_page(kvm, identity_map_pfn, 0, PAGE_SIZE, false);
 	if (r < 0)
 		goto out;
 	/* Set up identity-mapping pagetable for EPT in real mode */

@@ -3506,7 +3506,8 @@ static int init_rmode_identity_map(struct kvm *kvm)
 		tmp = (i << 22) + (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER |
 			_PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_PSE);
 		r = kvm_write_guest_page(kvm, identity_map_pfn,
-				&tmp, i * sizeof(tmp), sizeof(tmp));
+					 &tmp, i * sizeof(tmp),
+					 sizeof(tmp), false);
 		if (r < 0)
 			goto out;
 	}

@@ -7265,7 +7266,8 @@ static int vmx_write_pml_buffer(struct kvm_vcpu *vcpu)
 		dst = vmcs12->pml_address + sizeof(u64) * vmcs12->guest_pml_index;
 
 		if (kvm_write_guest_page(vcpu->kvm, gpa_to_gfn(dst), &gpa,
-					 offset_in_page(dst), sizeof(gpa)))
+					 offset_in_page(dst), sizeof(gpa),
+					 false))
 			return 0;
 
 		vmcs12->guest_pml_index--;

include/linux/kvm_host.h +3 -2

index 2ea1ea79befd..4e34cf97ca90 100644
@@ -734,7 +734,7 @@ int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
 int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 			   void *data, unsigned long len);
 int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
-			 int offset, int len);
+			 int offset, int len, bool track_dirty);
 int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
 		    unsigned long len);
 int kvm_write_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,

@@ -744,7 +744,8 @@ int kvm_write_guest_offset_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 				  unsigned long len);
 int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 			      gpa_t gpa, unsigned long len);
-int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len);
+int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len,
+			 bool track_dirty);
 int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len);
 struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn);
 bool kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn);

virt/kvm/kvm_main.c +16 -9

index 7ee28af9eb48..b1047173d78e 100644
@@ -2051,7 +2051,8 @@ int kvm_vcpu_read_guest_atomic(struct kvm_vcpu *vcpu, gpa_t gpa,
 EXPORT_SYMBOL_GPL(kvm_vcpu_read_guest_atomic);
 
 static int __kvm_write_guest_page(struct kvm_memory_slot *memslot, gfn_t gfn,
-			          const void *data, int offset, int len)
+			          const void *data, int offset, int len,
+				  bool track_dirty)
 {
 	int r;
 	unsigned long addr;

@@ -2062,16 +2063,19 @@ static int __kvm_write_guest_page(struct kvm_memory_slot *memslot, gfn_t gfn,
 	r = __copy_to_user((void __user *)addr + offset, data, len);
 	if (r)
 		return -EFAULT;
-	mark_page_dirty_in_slot(memslot, gfn);
+	if (track_dirty)
+		mark_page_dirty_in_slot(memslot, gfn);
 	return 0;
 }
 
 int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn,
-			 const void *data, int offset, int len)
+			 const void *data, int offset, int len,
+			 bool track_dirty)
 {
 	struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn);
 
-	return __kvm_write_guest_page(slot, gfn, data, offset, len);
+	return __kvm_write_guest_page(slot, gfn, data, offset, len,
+				      track_dirty);
 }
 EXPORT_SYMBOL_GPL(kvm_write_guest_page);
 

@@ -2080,7 +2084,8 @@ int kvm_vcpu_write_guest_page(struct kvm_vcpu *vcpu, gfn_t gfn,
 {
 	struct kvm_memory_slot *slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
 
-	return __kvm_write_guest_page(slot, gfn, data, offset, len);
+	return __kvm_write_guest_page(slot, gfn, data, offset,
+				      len, true);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_write_guest_page);
 

@@ -2093,7 +2098,7 @@ int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
 	int ret;
 
 	while ((seg = next_segment(len, offset)) != 0) {
-		ret = kvm_write_guest_page(kvm, gfn, data, offset, seg);
+		ret = kvm_write_guest_page(kvm, gfn, data, offset, seg, true);
 		if (ret < 0)
 			return ret;
 		offset = 0;

@@ -2232,11 +2237,13 @@ int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 }
 EXPORT_SYMBOL_GPL(kvm_read_guest_cached);
 
-int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len)
+int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len,
+			 bool track_dirty)
 {
 	const void *zero_page = (const void *) __va(page_to_phys(ZERO_PAGE(0)));
 
-	return kvm_write_guest_page(kvm, gfn, zero_page, offset, len);
+	return kvm_write_guest_page(kvm, gfn, zero_page, offset, len,
+				    track_dirty);
 }
 EXPORT_SYMBOL_GPL(kvm_clear_guest_page);
 

@@ -2248,7 +2255,7 @@ int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len)
 	int ret;
 
 	while ((seg = next_segment(len, offset)) != 0) {
-		ret = kvm_clear_guest_page(kvm, gfn, offset, seg);
+		ret = kvm_clear_guest_page(kvm, gfn, offset, seg, true);
 		if (ret < 0)
 			return ret;
 		offset = 0;

Quoted 11 lines · Peter Xu

Originally, we have three code paths that can dirty a page without vcpu context for X86:

- init_rmode_identity_map
- init_rmode_tss
- kvmgt_rw_gpa

init_rmode_identity_map and init_rmode_tss will be setup on destination VM no matter what (and the guest cannot even see them), so it does not make sense to track them at all.

Quoted 4 lines

To do this, a new parameter is added to kvm_[write|clear]_guest_page() to show whether we would like to track dirty bits for the operations. With that, pass in "false" to this new parameter for any guest memory write of the ioctls (KVM_SET_TSS_ADDR, KVM_SET_IDENTITY_MAP_ADDR).

We can also return the hva from x86_set_memory_region and __x86_set_memory_region.

Yes. Though it is a bit tricky in that then we'll also need to make sure to take slots_lock or srcu to protect that hva (say, we must drop that hva reference before we release the locks, otherwise the hva could gone under us, iiuc).

So if we want to do that we'd better
comment on that hva value very explicitly, just in case some future callers of __x86_set_memory_region could cache it somewhere.

Quoted 4 lines · Peter Xu

Yes. Though it is a bit tricky in that then we'll also need to make sure to take slots_lock or srcu to protect that hva (say, we must drop that hva reference before we release the locks, otherwise the hva could gone under us, iiuc).

Yes, kvm->slots_lock is taken by x86_set_memory_region. We need to move that to the callers, of which several are already taking the lock (all except vmx_set_tss_addr and kvm_arch_destroy_vm).

Quoted 3 lines

Yes, kvm->slots_lock is taken by x86_set_memory_region. We need to move that to the callers, of which several are already taking the lock (all except vmx_set_tss_addr and kvm_arch_destroy_vm).

OK, will do. I'll directly replace the x86_set_memory_region() calls in kvm_arch_destroy_vm() to be __x86_set_memory_region() since IIUC the slots_lock is helpless when destroying the vm... then drop the x86_set_memory_region() helper in the next version. Thanks,

Quoted 7 lines · Peter Xu

Quoted 3 lines

Yes, kvm->slots_lock is taken by x86_set_memory_region. We need to move that to the callers, of which several are already taking the lock (all except vmx_set_tss_addr and kvm_arch_destroy_vm).

OK, will do. I'll directly replace the x86_set_memory_region() calls in kvm_arch_destroy_vm() to be __x86_set_memory_region() since IIUC the slots_lock is helpless when destroying the vm... then drop the x86_set_memory_region() helper in the next version. Thanks,

Be careful because it may cause issues with lockdep. Better just take the lock.

But you seemed to have fixed that already? :)

3898da947bba ("KVM: avoid using rcu_dereference_protected", 2017-08-02)

And this path is after kvm_destroy_vm() so kvm->users_count should be 0. Or I feel like we need to have more places to take the lock..

Yeah, it should be okay assuming you test with lockdep.

Let's cache the address space ID just like the slot ID.

Suggested-by Paolo Bonzini <[email protected]>

Suggested-by Sean Christopherson <[email protected]>

Signed-off-by Peter Xu <[email protected]>

---
 include/linux/kvm_host.h | 1 +
 virt/kvm/kvm_main.c      | 2 ++
 2 files changed, 3 insertions(+)

include/linux/kvm_host.h +1 -0

index 4e34cf97ca90..24854c9e3717 100644
@@ -348,6 +348,7 @@ struct kvm_memory_slot {
 	unsigned long userspace_addr;
 	u32 flags;
 	short id;
+	u8 as_id;
 };
 
 static inline unsigned long kvm_dirty_bitmap_bytes(struct kvm_memory_slot *memslot)

virt/kvm/kvm_main.c +2 -0

index b1047173d78e..cea4b8dd4ac9 100644
@@ -1027,6 +1027,8 @@ int __kvm_set_memory_region(struct kvm *kvm,
 
 	new = old = *slot;
 
+	BUILD_BUG_ON(U8_MAX < KVM_ADDRESS_SPACE_NUM);
+	new.as_id = as_id;
 	new.id = id;
 	new.base_gfn = base_gfn;
 	new.npages = npages;

The context will be needed to implement the kvm dirty ring.

Signed-off-by Peter Xu <[email protected]>

---
 virt/kvm/kvm_main.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

virt/kvm/kvm_main.c +13 -9

index c80a363831ae..17969cf110dd 100644
@@ -144,7 +144,9 @@ static void hardware_disable_all(void);
 
 static void kvm_io_bus_destroy(struct kvm_io_bus *bus);
 
-static void mark_page_dirty_in_slot(struct kvm_memory_slot *memslot, gfn_t gfn);
+static void mark_page_dirty_in_slot(struct kvm *kvm,
+				    struct kvm_memory_slot *memslot,
+				    gfn_t gfn);
 
 __visible bool kvm_rebooting;
 EXPORT_SYMBOL_GPL(kvm_rebooting);

@@ -2053,8 +2055,9 @@ int kvm_vcpu_read_guest_atomic(struct kvm_vcpu *vcpu, gpa_t gpa,
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_read_guest_atomic);
 
-static int __kvm_write_guest_page(struct kvm_memory_slot *memslot, gfn_t gfn,
-			          const void *data, int offset, int len,
+static int __kvm_write_guest_page(struct kvm *kvm,
+				  struct kvm_memory_slot *memslot, gfn_t gfn,
+				  const void *data, int offset, int len,
 				  bool track_dirty)
 {
 	int r;

@@ -2067,7 +2070,7 @@ static int __kvm_write_guest_page(struct kvm_memory_slot *memslot, gfn_t gfn,
 	if (r)
 		return -EFAULT;
 	if (track_dirty)
-		mark_page_dirty_in_slot(memslot, gfn);
+		mark_page_dirty_in_slot(kvm, memslot, gfn);
 	return 0;
 }
 

@@ -2077,7 +2080,7 @@ int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn,
 {
 	struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn);
 
-	return __kvm_write_guest_page(slot, gfn, data, offset, len,
+	return __kvm_write_guest_page(kvm, slot, gfn, data, offset, len,
 				      track_dirty);
 }
 EXPORT_SYMBOL_GPL(kvm_write_guest_page);

@@ -2087,7 +2090,7 @@ int kvm_vcpu_write_guest_page(struct kvm_vcpu *vcpu, gfn_t gfn,
 {
 	struct kvm_memory_slot *slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
 
-	return __kvm_write_guest_page(slot, gfn, data, offset,
+	return __kvm_write_guest_page(vcpu->kvm, slot, gfn, data, offset,
 				      len, true);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_write_guest_page);

@@ -2202,7 +2205,7 @@ int kvm_write_guest_offset_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
 	r = __copy_to_user((void __user *)ghc->hva + offset, data, len);
 	if (r)
 		return -EFAULT;
-	mark_page_dirty_in_slot(ghc->memslot, gpa >> PAGE_SHIFT);
+	mark_page_dirty_in_slot(kvm, ghc->memslot, gpa >> PAGE_SHIFT);
 
 	return 0;
 }

@@ -2269,7 +2272,8 @@ int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len)
 }
 EXPORT_SYMBOL_GPL(kvm_clear_guest);
 
-static void mark_page_dirty_in_slot(struct kvm_memory_slot *memslot,
+static void mark_page_dirty_in_slot(struct kvm *kvm,
+				    struct kvm_memory_slot *memslot,
 				    gfn_t gfn)
 {
 	if (memslot && memslot->dirty_bitmap) {

@@ -2284,7 +2288,7 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn)
 	struct kvm_memory_slot *memslot;
 
 	memslot = gfn_to_memslot(kvm, gfn);
-	mark_page_dirty_in_slot(memslot, gfn);
+	mark_page_dirty_in_slot(kvm, memslot, gfn);
 }
 EXPORT_SYMBOL_GPL(mark_page_dirty);
 

From: Paolo Bonzini <[email protected]>

For ring-based dirty log tracking, it will be more efficient to account writes during schedule-out or schedule-in to the currently running VCPU. We would like to do it even if the write doesn't use the current VCPU's address space, as is the case for cached writes (see commit 4e335d9e7ddb, "Revert "KVM: Support vCPU-based gfn->hva cache"", 2017-05-02).

Therefore, add a mechanism to track the currently-loaded kvm_vcpu struct. There is already something similar in KVM/ARM; one important difference is that kvm_arch_vcpu_{load,put} have two callers in virt/kvm/kvm_main.c:
we have to update both the architecture-independent vcpu_{load,put} and the preempt notifiers.

Another change made in the process is to allow using kvm_get_running_vcpu() in preemptible code. This is allowed because preempt notifiers ensure that the value does not change even after the VCPU thread is migrated.

Signed-off-by Paolo Bonzini <[email protected]>

Signed-off-by Peter Xu <[email protected]>

---
 arch/arm/include/asm/kvm_host.h   |  2 --
 arch/arm64/include/asm/kvm_host.h |  2 --
 include/linux/kvm_host.h          |  3 +++
 virt/kvm/arm/arch_timer.c         |  2 +-
 virt/kvm/arm/arm.c                | 29 -----------------------------
 virt/kvm/arm/perf.c               |  6 +++---
 virt/kvm/arm/vgic/vgic-mmio.c     | 15 +++------------
 virt/kvm/kvm_main.c               | 25 ++++++++++++++++++++++++-
 8 files changed, 34 insertions(+), 50 deletions(-)

arch/arm/include/asm/kvm_host.h +0 -2

index 8a37c8e89777..40eff9cc3744 100644
@@ -274,8 +274,6 @@ int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
 int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 
-struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
-struct kvm_vcpu __percpu **kvm_get_running_vcpus(void);
 void kvm_arm_halt_guest(struct kvm *kvm);
 void kvm_arm_resume_guest(struct kvm *kvm);
 

arch/arm64/include/asm/kvm_host.h +0 -2

index f656169db8c3..df8d72f7c20e 100644
@@ -430,8 +430,6 @@ int kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 
-struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
-struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void);
 void kvm_arm_halt_guest(struct kvm *kvm);
 void kvm_arm_resume_guest(struct kvm *kvm);
 

include/linux/kvm_host.h +3 -0

index 24854c9e3717..b4f7bef38e0d 100644
@@ -1323,6 +1323,9 @@ static inline void kvm_vcpu_set_dy_eligible(struct kvm_vcpu *vcpu, bool val)
 }
 #endif /* CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */
 
+struct kvm_vcpu *kvm_get_running_vcpu(void);
+struct kvm_vcpu __percpu **kvm_get_running_vcpus(void);
+
 #ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
 bool kvm_arch_has_irq_bypass(void);
 int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *,

virt/kvm/arm/arch_timer.c +1 -1

index e2bb5bd60227..085e7fed850c 100644
@@ -1022,7 +1022,7 @@ static bool timer_irqs_are_valid(struct kvm_vcpu *vcpu)
 
 bool kvm_arch_timer_get_input_level(int vintid)
 {
-	struct kvm_vcpu *vcpu = kvm_arm_get_running_vcpu();
+	struct kvm_vcpu *vcpu = kvm_get_running_vcpu();
 	struct arch_timer_context *timer;
 
 	if (vintid == vcpu_vtimer(vcpu)->irq.irq)

virt/kvm/arm/arm.c +0 -29

index 86c6aa1cb58e..f7dbb94ec525 100644
@@ -47,9 +47,6 @@ __asm__(".arch_extension	virt");
 DEFINE_PER_CPU(kvm_host_data_t, kvm_host_data);
 static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
 
-/* Per-CPU variable containing the currently running vcpu. */
-static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_arm_running_vcpu);
-
 /* The VMID used in the VTTBR */
 static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1);
 static u32 kvm_next_vmid;

@@ -58,31 +55,8 @@ static DEFINE_SPINLOCK(kvm_vmid_lock);
 static bool vgic_present;
 
 static DEFINE_PER_CPU(unsigned char, kvm_arm_hardware_enabled);
-
-static void kvm_arm_set_running_vcpu(struct kvm_vcpu *vcpu)
-{
-	__this_cpu_write(kvm_arm_running_vcpu, vcpu);
-}
-
 DEFINE_STATIC_KEY_FALSE(userspace_irqchip_in_use);
 
-/**
- * kvm_arm_get_running_vcpu - get the vcpu running on the current CPU.
- * Must be called from non-preemptible context
- */
-struct kvm_vcpu *kvm_arm_get_running_vcpu(void)
-{
-	return __this_cpu_read(kvm_arm_running_vcpu);
-}
-
-/**
- * kvm_arm_get_running_vcpus - get the per-CPU array of currently running vcpus.
- */
-struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void)
-{
-	return &kvm_arm_running_vcpu;
-}
-
 int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
 {
 	return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE;

@@ -374,7 +348,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	vcpu->cpu = cpu;
 	vcpu->arch.host_cpu_context = &cpu_data->host_ctxt;
 
-	kvm_arm_set_running_vcpu(vcpu);
 	kvm_vgic_load(vcpu);
 	kvm_timer_vcpu_load(vcpu);
 	kvm_vcpu_load_sysregs(vcpu);

@@ -398,8 +371,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	kvm_vcpu_pmu_restore_host(vcpu);
 
 	vcpu->cpu = -1;
-
-	kvm_arm_set_running_vcpu(NULL);
 }
 
 static void vcpu_power_off(struct kvm_vcpu *vcpu)

virt/kvm/arm/perf.c +3 -3

index 918cdc3839ea..d45b8b9a4415 100644
@@ -13,14 +13,14 @@
 
 static int kvm_is_in_guest(void)
 {
-        return kvm_arm_get_running_vcpu() != NULL;
+        return kvm_get_running_vcpu() != NULL;
 }
 
 static int kvm_is_user_mode(void)
 {
 	struct kvm_vcpu *vcpu;
 
-	vcpu = kvm_arm_get_running_vcpu();
+	vcpu = kvm_get_running_vcpu();
 
 	if (vcpu)
 		return !vcpu_mode_priv(vcpu);

@@ -32,7 +32,7 @@ static unsigned long kvm_get_guest_ip(void)
 {
 	struct kvm_vcpu *vcpu;
 
-	vcpu = kvm_arm_get_running_vcpu();
+	vcpu = kvm_get_running_vcpu();
 
 	if (vcpu)
 		return *vcpu_pc(vcpu);

virt/kvm/arm/vgic/vgic-mmio.c +3 -12

index 0d090482720d..d656ebd5f9d4 100644
@@ -190,15 +190,6 @@ unsigned long vgic_mmio_read_pending(struct kvm_vcpu *vcpu,
  * value later will give us the same value as we update the per-CPU variable
  * in the preempt notifier handlers.
  */
-static struct kvm_vcpu *vgic_get_mmio_requester_vcpu(void)
-{
-	struct kvm_vcpu *vcpu;
-
-	preempt_disable();
-	vcpu = kvm_arm_get_running_vcpu();
-	preempt_enable();
-	return vcpu;
-}
 
 /* Must be called with irq->irq_lock held */
 static void vgic_hw_irq_spending(struct kvm_vcpu *vcpu, struct vgic_irq *irq,

@@ -221,7 +212,7 @@ void vgic_mmio_write_spending(struct kvm_vcpu *vcpu,
 			      gpa_t addr, unsigned int len,
 			      unsigned long val)
 {
-	bool is_uaccess = !vgic_get_mmio_requester_vcpu();
+	bool is_uaccess = !kvm_get_running_vcpu();
 	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
 	int i;
 	unsigned long flags;

@@ -274,7 +265,7 @@ void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu,
 			      gpa_t addr, unsigned int len,
 			      unsigned long val)
 {
-	bool is_uaccess = !vgic_get_mmio_requester_vcpu();
+	bool is_uaccess = !kvm_get_running_vcpu();
 	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
 	int i;
 	unsigned long flags;

@@ -335,7 +326,7 @@ static void vgic_mmio_change_active(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
 				    bool active)
 {
 	unsigned long flags;
-	struct kvm_vcpu *requester_vcpu = vgic_get_mmio_requester_vcpu();
+	struct kvm_vcpu *requester_vcpu = kvm_get_running_vcpu();
 
 	raw_spin_lock_irqsave(&irq->irq_lock, flags);
 

virt/kvm/kvm_main.c +17 -1

index 17969cf110dd..5c606d158854 100644
@@ -108,6 +108,7 @@ struct kmem_cache *kvm_vcpu_cache;
 EXPORT_SYMBOL_GPL(kvm_vcpu_cache);
 
 static __read_mostly struct preempt_ops kvm_preempt_ops;
+static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_running_vcpu);
 
 struct dentry *kvm_debugfs_dir;
 EXPORT_SYMBOL_GPL(kvm_debugfs_dir);

@@ -199,6 +200,8 @@ bool kvm_is_reserved_pfn(kvm_pfn_t pfn)
 void vcpu_load(struct kvm_vcpu *vcpu)
 {
 	int cpu = get_cpu();
+
+	__this_cpu_write(kvm_running_vcpu, vcpu);
 	preempt_notifier_register(&vcpu->preempt_notifier);
 	kvm_arch_vcpu_load(vcpu, cpu);
 	put_cpu();

@@ -210,6 +213,7 @@ void vcpu_put(struct kvm_vcpu *vcpu)
 	preempt_disable();
 	kvm_arch_vcpu_put(vcpu);
 	preempt_notifier_unregister(&vcpu->preempt_notifier);
+	__this_cpu_write(kvm_running_vcpu, NULL);
 	preempt_enable();
 }
 EXPORT_SYMBOL_GPL(vcpu_put);

@@ -4294,8 +4298,8 @@ static void kvm_sched_in(struct preempt_notifier *pn, int cpu)
 	WRITE_ONCE(vcpu->preempted, false);
 	WRITE_ONCE(vcpu->ready, false);
 
+	__this_cpu_write(kvm_running_vcpu, vcpu);
 	kvm_arch_sched_in(vcpu, cpu);
-
 	kvm_arch_vcpu_load(vcpu, cpu);
 }
 

@@ -4309,6 +4313,25 @@ static void kvm_sched_out(struct preempt_notifier *pn,
 		WRITE_ONCE(vcpu->ready, true);
 	}
 	kvm_arch_vcpu_put(vcpu);
+	__this_cpu_write(kvm_running_vcpu, NULL);
+}
+
+/**
+ * kvm_get_running_vcpu - get the vcpu running on the current CPU.
+ * Thanks to preempt notifiers, this can also be called from
+ * preemptible context.
+ */
+struct kvm_vcpu *kvm_get_running_vcpu(void)
+{
+        return __this_cpu_read(kvm_running_vcpu);
+}

This patch is heavily based on previous work from Lei Cao <[email protected]> and Paolo Bonzini <[email protected]>. [1]

KVM currently uses large bitmaps to track dirty memory. These bitmaps are copied to userspace when userspace queries KVM for its dirty page information. The use of bitmaps is mostly sufficient for live migration, as large parts of memory are be dirtied from one log-dirty pass to another. However, in a checkpointing system, the number of dirty pages is small and in fact it is often bounded---the VM is paused when it has dirtied a pre-defined number of pages. Traversing a large, sparsely populated bitmap to find set bits is time-consuming, as is copying the bitmap to user-space.

A similar issue will be there for live migration when the guest memory is huge while the page dirty procedure is trivial. In that case for each dirty sync we need to pull the whole dirty bitmap to userspace and analyse every bit even if it's mostly zeros.

The preferred data structure for above scenarios is a dense list of guest frame numbers (GFN). This patch series stores the dirty list in kernel memory that can be memory mapped into userspace to allow speedy harvesting.

This patch enables dirty ring for X86 only. However it should be easily extended to other archs as well.

[1] https://patchwork.kernel.org/patch/10471409/

Signed-off-by Lei Cao <[email protected]>

Signed-off-by Paolo Bonzini <[email protected]>

Signed-off-by Peter Xu <[email protected]>

---
  Documentation/virt/kvm/api.txt  |  89 ++++++++++++++
  arch/x86/include/asm/kvm_host.h |   3 +
  arch/x86/include/uapi/asm/kvm.h |   1 +
  arch/x86/kvm/Makefile           |   3 +-
  arch/x86/kvm/mmu.c              |   6 +
  arch/x86/kvm/vmx/vmx.c          |   7 ++
  arch/x86/kvm/x86.c              |   9 ++
  include/linux/kvm_dirty_ring.h  |  57 +++++++++
  include/linux/kvm_host.h        |  28 +++++
  include/trace/events/kvm.h      |  78 +++++++++++++
  include/uapi/linux/kvm.h        |  31 +++++
  virt/kvm/dirty_ring.c           | 201 ++++++++++++++++++++++++++++++++
  virt/kvm/kvm_main.c             | 172 ++++++++++++++++++++++++++-
  13 files changed, 682 insertions(+), 3 deletions(-)

create mode 100644 include/linux/kvm_dirty_ring.h create mode 100644 virt/kvm/dirty_ring.c

Documentation/virt/kvm/api.txt +89 -0

index 4833904d32a5..c141b285e673 100644
@@ -231,6 +231,7 @@ Based on their initialization different VMs may have different capabilities.
  It is thus encouraged to use the vm ioctl to query for capabilities (available
  with KVM_CAP_CHECK_EXTENSION_VM on the vm fd)
 
+
  4.5 KVM_GET_VCPU_MMAP_SIZE
 
  Capability: basic

@@ -243,6 +244,18 @@ The KVM_RUN ioctl (cf.) communicates with userspace via a shared
  memory region.  This ioctl returns the size of that region.  See the
  KVM_RUN documentation for details.
 
+Besides the size of the KVM_RUN communication region, other areas of
+the VCPU file descriptor can be mmap-ed, including:
+
+- if KVM_CAP_COALESCED_MMIO is available, a page at
+  KVM_COALESCED_MMIO_PAGE_OFFSET * PAGE_SIZE; for historical reasons,
+  this page is included in the result of KVM_GET_VCPU_MMAP_SIZE.
+  KVM_CAP_COALESCED_MMIO is not documented yet.
+
+- if KVM_CAP_DIRTY_LOG_RING is available, a number of pages at
+  KVM_DIRTY_LOG_PAGE_OFFSET * PAGE_SIZE.  For more information on
+  KVM_CAP_DIRTY_LOG_RING, see section 8.3.
+
 
  4.6 KVM_SET_MEMORY_REGION
 

@@ -5302,6 +5315,7 @@ CPU when the exception is taken. If this virtual SError is taken to EL1 using
  AArch64, this value will be reported in the ISS field of ESR_ELx.
 
  See KVM_CAP_VCPU_EVENTS for more details.
+
  8.20 KVM_CAP_HYPERV_SEND_IPI
 
  Architectures: x86

@@ -5309,6 +5323,7 @@ Architectures: x86
  This capability indicates that KVM supports paravirtualized Hyper-V IPI send
  hypercalls:
  HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx.
+
  8.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH
 
  Architecture: x86

@@ -5322,3 +5337,77 @@ handling by KVM (as some KVM hypercall may be mistakenly treated as TLB
  flush hypercalls by Hyper-V) so userspace should disable KVM identification
  in CPUID and only exposes Hyper-V identification. In this case, guest
  thinks it's running on Hyper-V and only use Hyper-V hypercalls.
+
+8.22 KVM_CAP_DIRTY_LOG_RING
+
+Architectures: x86
+Parameters: args[0] - size of the dirty log ring
+
+KVM is capable of tracking dirty memory using ring buffers that are
+mmaped into userspace; there is one dirty ring per vcpu.
+
+One dirty ring is defined as below internally:
+
+struct kvm_dirty_ring {
+	u32 dirty_index;
+	u32 reset_index;
+	u32 size;
+	u32 soft_limit;
+	struct kvm_dirty_gfn *dirty_gfns;
+	struct kvm_dirty_ring_indices *indices;
+	int index;
+};
+
+Dirty GFNs (Guest Frame Numbers) are stored in the dirty_gfns array.
+For each of the dirty entry it's defined as:
+
+struct kvm_dirty_gfn {
+        __u32 pad;
+        __u32 slot; /* as_id | slot_id */
+        __u64 offset;
+};
+
+Most of the ring structure is used by KVM internally, while only the
+indices are exposed to userspace:
+
+struct kvm_dirty_ring_indices {
+	__u32 avail_index; /* set by kernel */
+	__u32 fetch_index; /* set by userspace */
+};
+
+The two indices in the ring buffer are free running counters.
+
+Userspace calls KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM ioctl
+to enable this capability for the new guest and set the size of the
+rings.  It is only allowed before creating any vCPU, and the size of
+the ring must be a power of two.  The larger the ring buffer, the less
+likely the ring is full and the VM is forced to exit to userspace. The
+optimal size depends on the workload, but it is recommended that it be
+at least 64 KiB (4096 entries).
+
+Just like for dirty page bitmaps, the buffer tracks writes to
+all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was
+set in KVM_SET_USER_MEMORY_REGION.  Once a memory region is registered
+with the flag set, userspace can start harvesting dirty pages from the
+ring buffer.
+
+To harvest the dirty pages, userspace accesses the mmaped ring buffer
+to read the dirty GFNs up to avail_index, and sets the fetch_index
+accordingly.  This can be done when the guest is running or paused,
+and dirty pages need not be collected all at once.  After processing
+one or more entries in the ring buffer, userspace calls the VM ioctl
+KVM_RESET_DIRTY_RINGS to notify the kernel that it has updated
+fetch_index and to mark those pages clean.  Therefore, the ioctl
+must be called *before* reading the content of the dirty pages.
+
+However, there is a major difference comparing to the
+KVM_GET_DIRTY_LOG interface in that when reading the dirty ring from
+userspace it's still possible that the kernel has not yet flushed the
+hardware dirty buffers into the kernel buffer (which was previously
+done by the KVM_GET_DIRTY_LOG ioctl).  To achieve that, one needs to
+kick the vcpu out for a hardware buffer flush (vmexit) to make sure
+all the existing dirty gfns are flushed to the dirty rings.
+
+If one of the ring buffers is full, the guest will exit to userspace
+with the exit reason set to KVM_EXIT_DIRTY_LOG_FULL, and the KVM_RUN
+ioctl will return to userspace with zero.

arch/x86/include/asm/kvm_host.h +3 -0

index 4fc61483919a..7e5e2d3f0509 100644
@@ -1159,6 +1159,7 @@ struct kvm_x86_ops {
  					   struct kvm_memory_slot *slot,
  					   gfn_t offset, unsigned long mask);
  	int (*write_log_dirty)(struct kvm_vcpu *vcpu);
+	int (*cpu_dirty_log_size)(void);
 
  	/* pmu operations of sub-arch */
  	const struct kvm_pmu_ops *pmu_ops;

@@ -1641,4 +1642,6 @@ static inline int kvm_cpu_get_apicid(int mps_cpu)
  #define GET_SMSTATE(type, buf, offset)		\
  	(*(type *)((buf) + (offset) - 0x7e00))
 
+int kvm_cpu_dirty_log_size(void);
+
  #endif /* _ASM_X86_KVM_HOST_H */

arch/x86/include/uapi/asm/kvm.h +1 -0

index 503d3f42da16..b59bf356c478 100644
@@ -12,6 +12,7 @@
 
  #define KVM_PIO_PAGE_OFFSET 1
  #define KVM_COALESCED_MMIO_PAGE_OFFSET 2
+#define KVM_DIRTY_LOG_PAGE_OFFSET 64
 
  #define DE_VECTOR 0
  #define DB_VECTOR 1

arch/x86/kvm/Makefile +2 -1

index 31ecf7a76d5a..a66ddb552208 100644
@@ -5,7 +5,8 @@ ccflags-y += -Iarch/x86/kvm
  KVM := ../../../virt/kvm
 
  kvm-y			+= $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o \
-				$(KVM)/eventfd.o $(KVM)/irqchip.o $(KVM)/vfio.o
+				$(KVM)/eventfd.o $(KVM)/irqchip.o $(KVM)/vfio.o \
+				$(KVM)/dirty_ring.o
  kvm-$(CONFIG_KVM_ASYNC_PF)	+= $(KVM)/async_pf.o
 
  kvm-y			+= x86.o mmu.o emulate.o i8259.o irq.o lapic.o \

arch/x86/kvm/mmu.c +6 -0

index 2ce9da58611e..5f7d73730f73 100644
@@ -1818,7 +1818,13 @@ int kvm_arch_write_log_dirty(struct kvm_vcpu *vcpu)
  {
  	if (kvm_x86_ops->write_log_dirty)
  		return kvm_x86_ops->write_log_dirty(vcpu);
+	return 0;
+}
 
+int kvm_cpu_dirty_log_size(void)
+{
+	if (kvm_x86_ops->cpu_dirty_log_size)
+		return kvm_x86_ops->cpu_dirty_log_size();
  	return 0;
  }
 

arch/x86/kvm/vmx/vmx.c +7 -0

index 1ff5a428f489..c3565319b481 100644
@@ -7686,6 +7686,7 @@ static __init int hardware_setup(void)
  		kvm_x86_ops->slot_disable_log_dirty = NULL;
  		kvm_x86_ops->flush_log_dirty = NULL;
  		kvm_x86_ops->enable_log_dirty_pt_masked = NULL;
+		kvm_x86_ops->cpu_dirty_log_size = NULL;
  	}
 
  	if (!cpu_has_vmx_preemption_timer())

@@ -7750,6 +7751,11 @@ static __exit void hardware_unsetup(void)
  	free_kvm_area();
  }
 
+static int vmx_cpu_dirty_log_size(void)
+{
+	return enable_pml ? PML_ENTITY_NUM : 0;
+}
+
  static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
  	.cpu_has_kvm_support = cpu_has_kvm_support,
  	.disabled_by_bios = vmx_disabled_by_bios,

@@ -7873,6 +7879,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
  	.flush_log_dirty = vmx_flush_log_dirty,
  	.enable_log_dirty_pt_masked = vmx_enable_log_dirty_pt_masked,
  	.write_log_dirty = vmx_write_pml_buffer,
+	.cpu_dirty_log_size = vmx_cpu_dirty_log_size,
 
  	.pre_block = vmx_pre_block,
  	.post_block = vmx_post_block,

arch/x86/kvm/x86.c +9 -0

index 5d530521f11d..f93262025a61 100644
@@ -7965,6 +7965,15 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 
  	bool req_immediate_exit = false;
 
+	/* Forbid vmenter if vcpu dirty ring is soft-full */
+	if (unlikely(vcpu->kvm->dirty_ring_size &&
+		     kvm_dirty_ring_soft_full(&vcpu->dirty_ring))) {
+		vcpu->run->exit_reason = KVM_EXIT_DIRTY_RING_FULL;
+		trace_kvm_dirty_ring_exit(vcpu);
+		r = 0;
+		goto out;
+	}
+
  	if (kvm_request_pending(vcpu)) {
  		if (kvm_check_request(KVM_REQ_GET_VMCS12_PAGES, vcpu)) {
  			if (unlikely(!kvm_x86_ops->get_vmcs12_pages(vcpu))) {

include/linux/kvm_dirty_ring.h new file +12 -0

new file mode 100644
index 000000000000..06db2312b383
@@ -0,0 +1,57 @@
+#ifndef KVM_DIRTY_RING_H
+#define KVM_DIRTY_RING_H
+
+/**
+ * kvm_dirty_ring: KVM internal dirty ring structure
+ *
+ * @dirty_index: free running counter that points to the next slot in
+ *               dirty_ring->dirty_gfns, where a new dirty page should go
+ * @reset_index: free running counter that points to the next dirty page
+ *               in dirty_ring->dirty_gfns for which dirty trap needs to
+ *               be reenabled
+ * @size:        size of the compact list, dirty_ring->dirty_gfns

include/linux/kvm_dirty_ring.h new file +30 -0

+	int index;
+};
+
+u32 kvm_dirty_ring_get_rsvd_entries(void);
+int kvm_dirty_ring_alloc(struct kvm_dirty_ring *ring,
+			 struct kvm_dirty_ring_indices *indices,
+			 int index, u32 size);
+struct kvm_dirty_ring *kvm_dirty_ring_get(struct kvm *kvm);
+void kvm_dirty_ring_put(struct kvm *kvm,
+			struct kvm_dirty_ring *ring);
+
+/*
+ * called with kvm->slots_lock held, returns the number of
+ * processed pages.
+ */
+int kvm_dirty_ring_reset(struct kvm *kvm, struct kvm_dirty_ring *ring);
+
+/*
+ * returns =0: successfully pushed
+ *         <0: unable to push, need to wait
+ */
+int kvm_dirty_ring_push(struct kvm_dirty_ring *ring, u32 slot, u64 offset);
+
+/* for use in vm_operations_struct */
+struct page *kvm_dirty_ring_get_page(struct kvm_dirty_ring *ring, u32 offset);
+
+void kvm_dirty_ring_free(struct kvm_dirty_ring *ring);
+bool kvm_dirty_ring_soft_full(struct kvm_dirty_ring *ring);
+
+#endif

include/linux/kvm_host.h +28 -0

index b4f7bef38e0d..dff214ab72eb 100644
@@ -34,6 +34,7 @@
  #include <linux/kvm_types.h>
 
  #include <asm/kvm_host.h>
+#include <linux/kvm_dirty_ring.h>
 
  #ifndef KVM_MAX_VCPU_ID
  #define KVM_MAX_VCPU_ID KVM_MAX_VCPUS

@@ -321,6 +322,7 @@ struct kvm_vcpu {
  	bool ready;
  	struct kvm_vcpu_arch arch;
  	struct dentry *debugfs_dentry;
+	struct kvm_dirty_ring dirty_ring;
  };
 
  static inline int kvm_vcpu_exiting_guest_mode(struct kvm_vcpu *vcpu)

@@ -502,6 +504,9 @@ struct kvm {
  	struct srcu_struct srcu;
  	struct srcu_struct irq_srcu;
  	pid_t userspace_pid;
+	u32 dirty_ring_size;
+	struct spinlock dirty_ring_lock;
+	wait_queue_head_t dirty_ring_waitqueue;
  };
 
  #define kvm_err(fmt, ...) \

@@ -813,6 +818,8 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm,
  					gfn_t gfn_offset,
  					unsigned long mask);
 
+void kvm_reset_dirty_gfn(struct kvm *kvm, u32 slot, u64 offset, u64 mask);
+
  int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
  				struct kvm_dirty_log *log);
  int kvm_vm_ioctl_clear_dirty_log(struct kvm *kvm,

@@ -1392,4 +1399,25 @@ int kvm_vm_create_worker_thread(struct kvm *kvm, kvm_vm_thread_fn_t thread_fn,
  				uintptr_t data, const char *name,
  				struct task_struct **thread_ptr);
 
+/*
+ * This defines how many reserved entries we want to keep before we
+ * kick the vcpu to the userspace to avoid dirty ring full.  This
+ * value can be tuned to higher if e.g. PML is enabled on the host.
+ */
+#define  KVM_DIRTY_RING_RSVD_ENTRIES  64
+
+/* Max number of entries allowed for each kvm dirty ring */
+#define  KVM_DIRTY_RING_MAX_ENTRIES  65536
+
+/*
+ * Arch needs to define these macro after implementing the dirty ring
+ * feature.  KVM_DIRTY_LOG_PAGE_OFFSET should be defined as the
+ * starting page offset of the dirty ring structures, while
+ * KVM_DIRTY_RING_VERSION should be defined as >=1.  By default, this
+ * feature is off on all archs.
+ */
+#ifndef KVM_DIRTY_LOG_PAGE_OFFSET
+#define KVM_DIRTY_LOG_PAGE_OFFSET 0
+#endif
+
  #endif

include/trace/events/kvm.h +78 -0

index 2c735a3e6613..3d850997940c 100644
@@ -399,6 +399,84 @@ TRACE_EVENT(kvm_halt_poll_ns,
  #define trace_kvm_halt_poll_ns_shrink(vcpu_id, new, old) \
  	trace_kvm_halt_poll_ns(false, vcpu_id, new, old)
 
+TRACE_EVENT(kvm_dirty_ring_push,
+	TP_PROTO(struct kvm_dirty_ring *ring, u32 slot, u64 offset),
+	TP_ARGS(ring, slot, offset),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(u32, dirty_index)
+		__field(u32, reset_index)
+		__field(u32, slot)
+		__field(u64, offset)
+	),
+
+	TP_fast_assign(
+		__entry->index          = ring->index;
+		__entry->dirty_index    = ring->dirty_index;
+		__entry->reset_index    = ring->reset_index;
+		__entry->slot           = slot;
+		__entry->offset         = offset;
+	),
+
+	TP_printk("ring %d: dirty 0x%x reset 0x%x "
+		  "slot %u offset 0x%llx (used %u)",
+		  __entry->index, __entry->dirty_index,
+		  __entry->reset_index,  __entry->slot, __entry->offset,
+		  __entry->dirty_index - __entry->reset_index)
+);
+
+TRACE_EVENT(kvm_dirty_ring_reset,
+	TP_PROTO(struct kvm_dirty_ring *ring),
+	TP_ARGS(ring),
+
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(u32, dirty_index)
+		__field(u32, reset_index)
+	),
+
+	TP_fast_assign(
+		__entry->index          = ring->index;
+		__entry->dirty_index    = ring->dirty_index;
+		__entry->reset_index    = ring->reset_index;
+	),
+
+	TP_printk("ring %d: dirty 0x%x reset 0x%x (used %u)",
+		  __entry->index, __entry->dirty_index, __entry->reset_index,
+		  __entry->dirty_index - __entry->reset_index)
+);
+
+TRACE_EVENT(kvm_dirty_ring_waitqueue,
+	TP_PROTO(bool enter),
+	TP_ARGS(enter),
+
+	TP_STRUCT__entry(
+	    __field(bool, enter)
+	),
+
+	TP_fast_assign(
+	    __entry->enter = enter;
+	),
+
+	TP_printk("%s", __entry->enter ? "wait" : "awake")
+);
+
+TRACE_EVENT(kvm_dirty_ring_exit,
+	TP_PROTO(struct kvm_vcpu *vcpu),
+	TP_ARGS(vcpu),
+
+	TP_STRUCT__entry(
+	    __field(int, vcpu_id)
+	),
+
+	TP_fast_assign(
+	    __entry->vcpu_id = vcpu->vcpu_id;
+	),
+
+	TP_printk("vcpu %d", __entry->vcpu_id)
+);
+
  #endif /* _TRACE_KVM_MAIN_H */
 
  /* This part must be outside protection */

include/uapi/linux/kvm.h +1 -0

index 52641d8ca9e8..5ea98e35a129 100644
@@ -235,6 +235,7 @@ struct kvm_hyperv_exit {
  #define KVM_EXIT_S390_STSI        25
  #define KVM_EXIT_IOAPIC_EOI       26
  #define KVM_EXIT_HYPERV           27
+#define KVM_EXIT_DIRTY_RING_FULL  28
 
  /* For KVM_EXIT_INTERNAL_ERROR */
  /* Emulate instruction failed. */

include/uapi/linux/kvm.h +27 -0

+};
+
  /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
  struct kvm_run {
  	/* in */

@@ -415,6 +421,8 @@ struct kvm_run {
  		struct kvm_sync_regs regs;
  		char padding[SYNC_REGS_SIZE_BYTES];
  	} s;
+
+	struct kvm_dirty_ring_indices vcpu_ring_indices;
  };
 
  /* for KVM_REGISTER_COALESCED_MMIO / KVM_UNREGISTER_COALESCED_MMIO */

@@ -1000,6 +1008,7 @@ struct kvm_ppc_resize_hpt {
  #define KVM_CAP_PMU_EVENT_FILTER 173
  #define KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 174
  #define KVM_CAP_HYPERV_DIRECT_TLBFLUSH 175
+#define KVM_CAP_DIRTY_LOG_RING 176
 
  #ifdef KVM_CAP_IRQ_ROUTING
 

@@ -1461,6 +1470,9 @@ struct kvm_enc_region {
  /* Available with KVM_CAP_ARM_SVE */
  #define KVM_ARM_VCPU_FINALIZE	  _IOW(KVMIO,  0xc2, int)
 
+/* Available with KVM_CAP_DIRTY_LOG_RING */
+#define KVM_RESET_DIRTY_RINGS     _IO(KVMIO, 0xc3)
+
  /* Secure Encrypted Virtualization command */
  enum sev_cmd_id {
  	/* Guest initialization commands */

@@ -1611,4 +1623,23 @@ struct kvm_hyperv_eventfd {
  #define KVM_HYPERV_CONN_ID_MASK		0x00ffffff
  #define KVM_HYPERV_EVENTFD_DEASSIGN	(1 << 0)
 
+/*
+ * The following are the requirements for supporting dirty log ring
+ * (by enabling KVM_DIRTY_LOG_PAGE_OFFSET).
+ *
+ * 1. Memory accesses by KVM should call kvm_vcpu_write_* instead
+ *    of kvm_write_* so that the global dirty ring is not filled up
+ *    too quickly.
+ * 2. kvm_arch_mmu_enable_log_dirty_pt_masked should be defined for
+ *    enabling dirty logging.
+ * 3. There should not be a separate step to synchronize hardware
+ *    dirty bitmap with KVM's.
+ */
+
+struct kvm_dirty_gfn {
+	__u32 pad;
+	__u32 slot;
+	__u64 offset;
+};
+
  #endif /* __LINUX_KVM_H */

virt/kvm/dirty_ring.c new file +123 -0

new file mode 100644
index 000000000000..c614822493ff
@@ -0,0 +1,201 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * KVM dirty ring implementation
+ *
+ * Copyright 2019 Red Hat, Inc.
+ */
+#include <linux/kvm_host.h>
+#include <linux/kvm.h>
+#include <linux/vmalloc.h>
+#include <linux/kvm_dirty_ring.h>
+#include <trace/events/kvm.h>
+
+int __weak kvm_cpu_dirty_log_size(void)
+{
+	return 0;
+}
+
+u32 kvm_dirty_ring_get_rsvd_entries(void)
+{
+	return KVM_DIRTY_RING_RSVD_ENTRIES + kvm_cpu_dirty_log_size();
+}
+
+static u32 kvm_dirty_ring_used(struct kvm_dirty_ring *ring)
+{
+	return READ_ONCE(ring->dirty_index) - READ_ONCE(ring->reset_index);
+}
+
+bool kvm_dirty_ring_soft_full(struct kvm_dirty_ring *ring)
+{
+	return kvm_dirty_ring_used(ring) >= ring->soft_limit;
+}
+
+bool kvm_dirty_ring_full(struct kvm_dirty_ring *ring)
+{
+	return kvm_dirty_ring_used(ring) >= ring->size;
+}
+
+struct kvm_dirty_ring *kvm_dirty_ring_get(struct kvm *kvm)
+{
+	struct kvm_vcpu *vcpu = kvm_get_running_vcpu();
+
+        /*
+	 * TODO: Currently use vcpu0 as default ring.  Note that this
+	 * should not happen only if called by kvmgt_rw_gpa for x86.
+	 * After the kvmgt code refactoring we should remove this,
+	 * together with the kvm->dirty_ring_lock.
+	 */
+	if (!vcpu) {
+		pr_warn_once("Detected page dirty without vcpu context. "
+			     "Probably because kvm-gt is used. "
+			     "May expect unbalanced loads on vcpu0.");
+		vcpu = kvm->vcpus[0];
+	}
+
+	WARN_ON_ONCE(vcpu->kvm != kvm);
+
+	if (vcpu == kvm->vcpus[0])
+		spin_lock(&kvm->dirty_ring_lock);
+
+	return &vcpu->dirty_ring;
+}
+
+void kvm_dirty_ring_put(struct kvm *kvm,
+			struct kvm_dirty_ring *ring)
+{
+	struct kvm_vcpu *vcpu = kvm_get_running_vcpu();
+
+	if (!vcpu)
+		vcpu = kvm->vcpus[0];
+
+	WARN_ON_ONCE(vcpu->kvm != kvm);
+	WARN_ON_ONCE(&vcpu->dirty_ring != ring);
+
+	if (vcpu == kvm->vcpus[0])
+		spin_unlock(&kvm->dirty_ring_lock);
+}
+
+int kvm_dirty_ring_alloc(struct kvm_dirty_ring *ring,
+			 struct kvm_dirty_ring_indices *indices,
+			 int index, u32 size)
+{
+	ring->dirty_gfns = vmalloc(size);
+	if (!ring->dirty_gfns)
+		return -ENOMEM;
+	memset(ring->dirty_gfns, 0, size);
+
+	ring->size = size / sizeof(struct kvm_dirty_gfn);
+	ring->soft_limit = ring->size - kvm_dirty_ring_get_rsvd_entries();
+	ring->dirty_index = 0;
+	ring->reset_index = 0;
+	ring->index = index;
+	ring->indices = indices;
+
+	return 0;
+}
+
+int kvm_dirty_ring_reset(struct kvm *kvm, struct kvm_dirty_ring *ring)
+{
+	u32 cur_slot, next_slot;
+	u64 cur_offset, next_offset;
+	unsigned long mask;
+	u32 fetch;
+	int count = 0;
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+	bool first_round = true;
+
+	fetch = READ_ONCE(indices->fetch_index);
+
+	/*
+	 * Note that fetch_index is written by the userspace, which
+	 * should not be trusted.  If this happens, then it's probably
+	 * that the userspace has written a wrong fetch_index.
+	 */
+	if (fetch - ring->reset_index > ring->size)
+		return -EINVAL;
+
+	if (fetch == ring->reset_index)
+		return 0;
+
+	/* This is only needed to make compilers happy */
+	cur_slot = cur_offset = mask = 0;
+	while (ring->reset_index != fetch) {

virt/kvm/dirty_ring.c new file +40 -0

+			s64 delta = next_offset - cur_offset;
+
+			if (delta >= 0 && delta < BITS_PER_LONG) {
+				mask |= 1ull << delta;
+				continue;
+			}
+
+			/* Backwards visit, careful about overflows!  */
+			if (delta > -BITS_PER_LONG && delta < 0 &&
+			    (mask << -delta >> -delta) == mask) {
+				cur_offset = next_offset;
+				mask = (mask << -delta) | 1;
+				continue;
+			}
+		}
+		kvm_reset_dirty_gfn(kvm, cur_slot, cur_offset, mask);
+		cur_slot = next_slot;
+		cur_offset = next_offset;
+		mask = 1;
+		first_round = false;
+	}
+	kvm_reset_dirty_gfn(kvm, cur_slot, cur_offset, mask);
+
+	trace_kvm_dirty_ring_reset(ring);
+
+	return count;
+}
+
+int kvm_dirty_ring_push(struct kvm_dirty_ring *ring, u32 slot, u64 offset)
+{
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+
+	/*
+	 * Note: here we will start waiting even soft full, because we
+	 * can't risk making it completely full, since vcpu0 could use
+	 * it right after us and if vcpu0 context gets full it could
+	 * deadlock if wait with mmu_lock held.
+	 */
+	if (kvm_get_running_vcpu() == NULL &&

virt/kvm/dirty_ring.c new file +16 -0

+
+	trace_kvm_dirty_ring_push(ring, slot, offset);
+
+	return 0;
+}
+
+struct page *kvm_dirty_ring_get_page(struct kvm_dirty_ring *ring, u32 offset)
+{
+	return vmalloc_to_page((void *)ring->dirty_gfns + offset * PAGE_SIZE);
+}
+
+void kvm_dirty_ring_free(struct kvm_dirty_ring *ring)
+{
+	vfree(ring->dirty_gfns);
+	ring->dirty_gfns = NULL;
+}

virt/kvm/kvm_main.c +170 -2

index 5c606d158854..4050631d05f3 100644
@@ -64,6 +64,8 @@
  #define CREATE_TRACE_POINTS
  #include <trace/events/kvm.h>
 
+#include <linux/kvm_dirty_ring.h>
+
  /* Worst case buffer size needed for holding an integer. */
  #define ITOA_MAX_LEN 12
 

@@ -148,6 +150,9 @@ static void kvm_io_bus_destroy(struct kvm_io_bus *bus);
  static void mark_page_dirty_in_slot(struct kvm *kvm,
  				    struct kvm_memory_slot *memslot,
  				    gfn_t gfn);
+static void mark_page_dirty_in_ring(struct kvm *kvm,
+				    struct kvm_memory_slot *slot,
+				    gfn_t gfn);
 
  __visible bool kvm_rebooting;
  EXPORT_SYMBOL_GPL(kvm_rebooting);

@@ -357,11 +362,22 @@ int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
  	vcpu->preempted = false;
  	vcpu->ready = false;
 
+	if (kvm->dirty_ring_size) {
+		r = kvm_dirty_ring_alloc(&vcpu->dirty_ring,
+					 &vcpu->run->vcpu_ring_indices,
+					 id, kvm->dirty_ring_size);
+		if (r)
+			goto fail_free_run;
+	}
+
  	r = kvm_arch_vcpu_init(vcpu);
  	if (r < 0)
-		goto fail_free_run;
+		goto fail_free_ring;
  	return 0;
 
+fail_free_ring:
+	if (kvm->dirty_ring_size)
+		kvm_dirty_ring_free(&vcpu->dirty_ring);
  fail_free_run:
  	free_page((unsigned long)vcpu->run);
  fail:

@@ -379,6 +395,8 @@ void kvm_vcpu_uninit(struct kvm_vcpu *vcpu)
  	put_pid(rcu_dereference_protected(vcpu->pid, 1));
  	kvm_arch_vcpu_uninit(vcpu);
  	free_page((unsigned long)vcpu->run);
+	if (vcpu->kvm->dirty_ring_size)
+		kvm_dirty_ring_free(&vcpu->dirty_ring);
  }
  EXPORT_SYMBOL_GPL(kvm_vcpu_uninit);
 

@@ -693,6 +711,7 @@ static struct kvm *kvm_create_vm(unsigned long type)
  		return ERR_PTR(-ENOMEM);
 
  	spin_lock_init(&kvm->mmu_lock);
+	spin_lock_init(&kvm->dirty_ring_lock);
  	mmgrab(current->mm);
  	kvm->mm = current->mm;
  	kvm_eventfd_init(kvm);

@@ -700,6 +719,7 @@ static struct kvm *kvm_create_vm(unsigned long type)
  	mutex_init(&kvm->irq_lock);
  	mutex_init(&kvm->slots_lock);
  	INIT_LIST_HEAD(&kvm->devices);
+	init_waitqueue_head(&kvm->dirty_ring_waitqueue);
 
  	BUILD_BUG_ON(KVM_MEM_SLOTS_NUM > SHRT_MAX);
 

@@ -2283,7 +2303,10 @@ static void mark_page_dirty_in_slot(struct kvm *kvm,
  	if (memslot && memslot->dirty_bitmap) {
  		unsigned long rel_gfn = gfn - memslot->base_gfn;
 
-		set_bit_le(rel_gfn, memslot->dirty_bitmap);
+		if (kvm->dirty_ring_size)
+			mark_page_dirty_in_ring(kvm, memslot, gfn);
+		else
+			set_bit_le(rel_gfn, memslot->dirty_bitmap);
  	}
  }
 

@@ -2630,6 +2653,16 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me, bool yield_to_kernel_mode)
  }
  EXPORT_SYMBOL_GPL(kvm_vcpu_on_spin);
 
+static bool kvm_fault_in_dirty_ring(struct kvm *kvm, struct vm_fault *vmf)
+{
+	if (!KVM_DIRTY_LOG_PAGE_OFFSET)
+		return false;
+
+	return (vmf->pgoff >= KVM_DIRTY_LOG_PAGE_OFFSET) &&
+	    (vmf->pgoff < KVM_DIRTY_LOG_PAGE_OFFSET +
+	     kvm->dirty_ring_size / PAGE_SIZE);
+}
+
  static vm_fault_t kvm_vcpu_fault(struct vm_fault *vmf)
  {
  	struct kvm_vcpu *vcpu = vmf->vma->vm_file->private_data;

@@ -2645,6 +2678,10 @@ static vm_fault_t kvm_vcpu_fault(struct vm_fault *vmf)
  	else if (vmf->pgoff == KVM_COALESCED_MMIO_PAGE_OFFSET)
  		page = virt_to_page(vcpu->kvm->coalesced_mmio_ring);
  #endif
+	else if (kvm_fault_in_dirty_ring(vcpu->kvm, vmf))
+		page = kvm_dirty_ring_get_page(
+		    &vcpu->dirty_ring,
+		    vmf->pgoff - KVM_DIRTY_LOG_PAGE_OFFSET);
  	else
  		return kvm_arch_vcpu_fault(vcpu, vmf);
  	get_page(page);

@@ -3239,12 +3276,138 @@ static long kvm_vm_ioctl_check_extension_generic(struct kvm *kvm, long arg)
  #endif
  	case KVM_CAP_NR_MEMSLOTS:
  		return KVM_USER_MEM_SLOTS;
+	case KVM_CAP_DIRTY_LOG_RING:
+#ifdef CONFIG_X86
+		return KVM_DIRTY_RING_MAX_ENTRIES;
+#else
+		return 0;
+#endif
  	default:
  		break;
  	}
  	return kvm_vm_ioctl_check_extension(kvm, arg);
  }
 
+static void mark_page_dirty_in_ring(struct kvm *kvm,
+				    struct kvm_memory_slot *slot,
+				    gfn_t gfn)
+{
+	struct kvm_dirty_ring *ring;
+	u64 offset;
+	int ret;
+
+	if (!kvm->dirty_ring_size)
+		return;
+
+	offset = gfn - slot->base_gfn;
+
+	ring = kvm_dirty_ring_get(kvm);
+
+retry:
+	ret = kvm_dirty_ring_push(ring, (slot->as_id << 16) | slot->id,
+				  offset);
+	if (ret < 0) {
+		/* We must be without a vcpu context. */
+		WARN_ON_ONCE(kvm_get_running_vcpu());
+
+		trace_kvm_dirty_ring_waitqueue(1);
+		/*
+		 * Ring is full, put us onto per-vm waitqueue and wait
+		 * for another KVM_RESET_DIRTY_RINGS to retry
+		 */
+		wait_event_killable(kvm->dirty_ring_waitqueue,
+				    !kvm_dirty_ring_soft_full(ring));
+
+		trace_kvm_dirty_ring_waitqueue(0);
+
+		/* If we're killed, no worry on lossing dirty bits */
+		if (fatal_signal_pending(current))
+			return;
+
+		goto retry;
+	}
+
+	kvm_dirty_ring_put(kvm, ring);
+}
+
+void kvm_reset_dirty_gfn(struct kvm *kvm, u32 slot, u64 offset, u64 mask)
+{
+	struct kvm_memory_slot *memslot;
+	int as_id, id;
+
+	as_id = slot >> 16;
+	id = (u16)slot;
+	if (as_id >= KVM_ADDRESS_SPACE_NUM || id >= KVM_USER_MEM_SLOTS)
+		return;
+
+	memslot = id_to_memslot(__kvm_memslots(kvm, as_id), id);
+	if (offset >= memslot->npages)
+		return;
+
+	spin_lock(&kvm->mmu_lock);
+	kvm_arch_mmu_enable_log_dirty_pt_masked(kvm, memslot, offset, mask);
+	spin_unlock(&kvm->mmu_lock);
+}
+
+static int kvm_vm_ioctl_enable_dirty_log_ring(struct kvm *kvm, u32 size)
+{
+	int r;
+
+	/* the size should be power of 2 */
+	if (!size || (size & (size - 1)))
+		return -EINVAL;
+
+	/* Should be bigger to keep the reserved entries, or a page */
+	if (size < kvm_dirty_ring_get_rsvd_entries() *
+	    sizeof(struct kvm_dirty_gfn) || size < PAGE_SIZE)
+		return -EINVAL;
+
+	if (size > KVM_DIRTY_RING_MAX_ENTRIES *
+	    sizeof(struct kvm_dirty_gfn))
+		return -E2BIG;
+
+	/* We only allow it to set once */
+	if (kvm->dirty_ring_size)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+
+	if (kvm->created_vcpus) {
+		/* We don't allow to change this value after vcpu created */
+		r = -EINVAL;
+	} else {
+		kvm->dirty_ring_size = size;
+		r = 0;
+	}
+
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+
+static int kvm_vm_ioctl_reset_dirty_pages(struct kvm *kvm)
+{
+	int i;
+	struct kvm_vcpu *vcpu;
+	int cleared = 0;
+
+	if (!kvm->dirty_ring_size)
+		return -EINVAL;
+
+	mutex_lock(&kvm->slots_lock);
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		cleared += kvm_dirty_ring_reset(vcpu->kvm, &vcpu->dirty_ring);
+
+	mutex_unlock(&kvm->slots_lock);
+
+	if (cleared)
+		kvm_flush_remote_tlbs(kvm);
+
+	wake_up_all(&kvm->dirty_ring_waitqueue);
+
+	return cleared;
+}
+
  int __attribute__((weak)) kvm_vm_ioctl_enable_cap(struct kvm *kvm,
  						  struct kvm_enable_cap *cap)
  {

@@ -3262,6 +3425,8 @@ static int kvm_vm_ioctl_enable_cap_generic(struct kvm *kvm,
  		kvm->manual_dirty_log_protect = cap->args[0];
  		return 0;
  #endif
+	case KVM_CAP_DIRTY_LOG_RING:
+		return kvm_vm_ioctl_enable_dirty_log_ring(kvm, cap->args[0]);
  	default:
  		return kvm_vm_ioctl_enable_cap(kvm, cap);
  	}

@@ -3449,6 +3614,9 @@ static long kvm_vm_ioctl(struct file *filp,
  	case KVM_CHECK_EXTENSION:
  		r = kvm_vm_ioctl_check_extension_generic(kvm, arg);
  		break;
+	case KVM_RESET_DIRTY_RINGS:
+		r = kvm_vm_ioctl_reset_dirty_pages(kvm);
+		break;
  	default:
  		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
  	}

Review

Quoted 2 lines

(quoted diff) +2 -0

+struct kvm_dirty_ring {
+	u32 dirty_index;

Does this always equal to indices->avail_index?

Documentation/virt/kvm/api.txt handling by KVM (as some KVM hypercall may be mistakenly treated as TLB · line 5357

Quoted 7 lines

Documentation/virt/kvm/api.txt +5 -0

+	u32 reset_index;
+	u32 size;
+	u32 soft_limit;
+	struct kvm_dirty_gfn *dirty_gfns;
+	struct kvm_dirty_ring_indices *indices;

Any reason to keep dirty gfns and indices in different places? I guess it is because you want to map dirty_gfns as readonly page but I couldn't find such codes...

Documentation/virt/kvm/api.txt handling by KVM (as some KVM hypercall may be mistakenly treated as TLB · line 5375

Quoted 3 lines

Documentation/virt/kvm/api.txt +3 -0

+struct kvm_dirty_ring_indices {
+	__u32 avail_index; /* set by kernel */
+	__u32 fetch_index; /* set by userspace */

Is this better to make those two cacheline aligned?

Quoted 32 lines

virt/kvm/dirty_ring.c new file +32 -0

+int kvm_dirty_ring_reset(struct kvm *kvm, struct kvm_dirty_ring *ring)
+{
+	u32 cur_slot, next_slot;
+	u64 cur_offset, next_offset;
+	unsigned long mask;
+	u32 fetch;
+	int count = 0;
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+	bool first_round = true;
+
+	fetch = READ_ONCE(indices->fetch_index);
+
+	/*
+	 * Note that fetch_index is written by the userspace, which
+	 * should not be trusted.  If this happens, then it's probably
+	 * that the userspace has written a wrong fetch_index.
+	 */
+	if (fetch - ring->reset_index > ring->size)
+		return -EINVAL;
+
+	if (fetch == ring->reset_index)
+		return 0;
+
+	/* This is only needed to make compilers happy */
+	cur_slot = cur_offset = mask = 0;
+	while (ring->reset_index != fetch) {
+		entry = &ring->dirty_gfns[ring->reset_index & (ring->size - 1)];
+		next_slot = READ_ONCE(entry->slot);
+		next_offset = READ_ONCE(entry->offset);
+		ring->reset_index++;
+		count++;

virt/kvm/dirty_ring.c line 133

Quoted 5 lines

virt/kvm/dirty_ring.c new file +5 -0

+		/*
+		 * Try to coalesce the reset operations when the guest is
+		 * scanning pages in the same slot.
+		 */
+		if (!first_round && next_slot == cur_slot) {

initialize cur_slot to -1 then we can drop first_round here?

Quoted 17 lines

virt/kvm/dirty_ring.c new file +17 -0

+int kvm_dirty_ring_push(struct kvm_dirty_ring *ring, u32 slot, u64 offset)
+{
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+
+	/*
+	 * Note: here we will start waiting even soft full, because we
+	 * can't risk making it completely full, since vcpu0 could use
+	 * it right after us and if vcpu0 context gets full it could
+	 * deadlock if wait with mmu_lock held.
+	 */
+	if (kvm_get_running_vcpu() == NULL &&
+	    kvm_dirty_ring_soft_full(ring))
+		return -EBUSY;
+
+	/* It will never gets completely full when with a vcpu context */
+	WARN_ON_ONCE(kvm_dirty_ring_full(ring));

virt/kvm/dirty_ring.c line 183

Quoted 5 lines

virt/kvm/dirty_ring.c new file +5 -0

+
+	entry = &ring->dirty_gfns[ring->dirty_index & (ring->size - 1)];
+	entry->slot = slot;
+	entry->offset = offset;
+	smp_wmb();

Better to add comment to explain this barrier. E.g pairing.

virt/kvm/dirty_ring.c line 185

Quoted 4 lines

virt/kvm/dirty_ring.c new file +2 -0

+	ring->dirty_index++;
+	WRITE_ONCE(indices->avail_index, ring->dirty_index);

Is WRITE_ONCE() a must here?

Quoted 4 lines · Jason Wang

Review

Quoted 2 lines

(quoted diff) +2 -0

+struct kvm_dirty_ring {
+	u32 dirty_index;

Does this always equal to indices->avail_index?

Yes, but here we keep dirty_index as the internal one, so we never need to worry about illegal userspace writes to avail_index (then we never read it from kernel).

Quoted 11 lines

Documentation/virt/kvm/api.txt handling by KVM (as some KVM hypercall may be mistakenly treated as TLB · line 5357

Quoted 7 lines

Documentation/virt/kvm/api.txt +5 -0

+	u32 reset_index;
+	u32 size;
+	u32 soft_limit;
+	struct kvm_dirty_gfn *dirty_gfns;
+	struct kvm_dirty_ring_indices *indices;

Any reason to keep dirty gfns and indices in different places? I guess it is because you want to map dirty_gfns as readonly page but I couldn't find such codes...

That's a good point! We should actually map the dirty gfns as read only. I've added the check, something like this:

static int kvm_vcpu_mmap(struct file *file, struct vm_area_struct *vma)

{
	struct kvm_vcpu *vcpu = file->private_data;
	unsigned long pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;

	/* If to map any writable page within dirty ring, fail it */
	if ((kvm_page_in_dirty_ring(vcpu->kvm, vma->vm_pgoff) ||
	     kvm_page_in_dirty_ring(vcpu->kvm, vma->vm_pgoff + pages - 1)) &&
	    vma->vm_flags & VM_WRITE)
		return -EINVAL;

	vma->vm_ops = &kvm_vcpu_vm_ops;
	return 0;
}

I also changed the test code to cover this case.

[...]

Quoted 6 lines

Documentation/virt/kvm/api.txt handling by KVM (as some KVM hypercall may be mistakenly treated as TLB · line 5375

Quoted 4 lines

Documentation/virt/kvm/api.txt +3 -0

+struct kvm_dirty_ring_indices {
+	__u32 avail_index; /* set by kernel */
+	__u32 fetch_index; /* set by userspace */

Is this better to make those two cacheline aligned?

Yes, Paolo should have mentioned that but I must have missed it! I hope I didn't miss anything else.

[...]

Quoted 31 lines

Quoted 31 lines

virt/kvm/dirty_ring.c new file +31 -0

+int kvm_dirty_ring_reset(struct kvm *kvm, struct kvm_dirty_ring *ring)
+{
+	u32 cur_slot, next_slot;
+	u64 cur_offset, next_offset;
+	unsigned long mask;
+	u32 fetch;
+	int count = 0;
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+	bool first_round = true;
+
+	fetch = READ_ONCE(indices->fetch_index);
+
+	/*
+	 * Note that fetch_index is written by the userspace, which
+	 * should not be trusted.  If this happens, then it's probably
+	 * that the userspace has written a wrong fetch_index.
+	 */
+	if (fetch - ring->reset_index > ring->size)
+		return -EINVAL;
+
+	if (fetch == ring->reset_index)
+		return 0;
+
+	/* This is only needed to make compilers happy */
+	cur_slot = cur_offset = mask = 0;
+	while (ring->reset_index != fetch) {
+		entry = &ring->dirty_gfns[ring->reset_index & (ring->size - 1)];
+		next_slot = READ_ONCE(entry->slot);
+		next_offset = READ_ONCE(entry->offset);
+		ring->reset_index++;

Quoted 8 lines

virt/kvm/dirty_ring.c line 133

Quoted 6 lines

virt/kvm/dirty_ring.c new file +6 -0

+		count++;
+		/*
+		 * Try to coalesce the reset operations when the guest is
+		 * scanning pages in the same slot.
+		 */
+		if (!first_round && next_slot == cur_slot) {

initialize cur_slot to -1 then we can drop first_round here?

cur_slot is unsigned. We can force cur_slot to be s64 but maybe we can also simply keep the first_round to be clear from its name.

[...]

Quoted 17 lines

Quoted 17 lines

virt/kvm/dirty_ring.c new file +16 -0

+int kvm_dirty_ring_push(struct kvm_dirty_ring *ring, u32 slot, u64 offset)
+{
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;
+
+	/*
+	 * Note: here we will start waiting even soft full, because we
+	 * can't risk making it completely full, since vcpu0 could use
+	 * it right after us and if vcpu0 context gets full it could
+	 * deadlock if wait with mmu_lock held.
+	 */
+	if (kvm_get_running_vcpu() == NULL &&
+	    kvm_dirty_ring_soft_full(ring))
+		return -EBUSY;
+
+	/* It will never gets completely full when with a vcpu context */

Quoted 8 lines

virt/kvm/dirty_ring.c line 183

Quoted 6 lines

virt/kvm/dirty_ring.c new file +6 -0

+	WARN_ON_ONCE(kvm_dirty_ring_full(ring));
+
+	entry = &ring->dirty_gfns[ring->dirty_index & (ring->size - 1)];
+	entry->slot = slot;
+	entry->offset = offset;
+	smp_wmb();

Better to add comment to explain this barrier. E.g pairing.

Will do.

Quoted 1 line

Quoted 4 lines

virt/kvm/dirty_ring.c line 185

Quoted 2 lines

virt/kvm/dirty_ring.c new file +2 -0

+	ring->dirty_index++;
+	WRITE_ONCE(indices->avail_index, ring->dirty_index);

Is WRITE_ONCE() a must here?

I think not, but seems to be clearer that we're publishing something explicilty to userspace. Since asked, I'm actually curious on whether immediate memory writes like this could start to affect perf from any of your previous perf works?

Thanks,

virt/kvm/dirty_ring.c new file +4 -0

+int kvm_dirty_ring_push(struct kvm_dirty_ring *ring, u32 slot, u64 offset)
+{
+	struct kvm_dirty_gfn *entry;
+	struct kvm_dirty_ring_indices *indices = ring->indices;

virt/kvm/dirty_ring.c new file +15 -0

+
+	/* It will never gets completely full when with a vcpu context */
+	WARN_ON_ONCE(kvm_dirty_ring_full(ring));
+
+	entry = &ring->dirty_gfns[ring->dirty_index & (ring->size - 1)];
+	entry->slot = slot;
+	entry->offset = offset;
+	smp_wmb();
+	ring->dirty_index++;
+	WRITE_ONCE(indices->avail_index, ring->dirty_index);
+
+	trace_kvm_dirty_ring_push(ring, slot, offset);
+
+	return 0;
+}

here, which is still a bit tricky to makeup the kvmgt issue.

Now we still have the waitqueue but it'll only be used for no-vcpu-context dirtyings, so:

- For no-vcpu-context: thread could wait in the waitqueue if it makes vcpu0's ring soft-full (note, previously it was hard-full, so here we make it easier to wait so we make sure )

- For with-vcpu-context: we should never wait, guaranteed by the fact that KVM_RUN will return now if soft-full for that vcpu ring, and above waitqueue will make sure even vcpu0's waitqueue won't be filled up by kvmgt

Again this is still a workaround for kvmgt and I think it should not be needed after the refactoring. It's just a way to not depend on that work so this should work even with current kvmgt.

Quoted 14 lines · Peter Xu

here, which is still a bit tricky to makeup the kvmgt issue.

Now we still have the waitqueue but it'll only be used for no-vcpu-context dirtyings, so:

- For no-vcpu-context: thread could wait in the waitqueue if it makes vcpu0's ring soft-full (note, previously it was hard-full, so here we make it easier to wait so we make sure )

- For with-vcpu-context: we should never wait, guaranteed by the fact that KVM_RUN will return now if soft-full for that vcpu ring, and above waitqueue will make sure even vcpu0's waitqueue won't be filled up by kvmgt

Quoted 3 lines

Again this is still a workaround for kvmgt and I think it should not be needed after the refactoring. It's just a way to not depend on that work so this should work even with current kvmgt.

The kvmgt patches were posted, you could just include them in your next series and clean everything up. You can get them at
https://patchwork.kernel.org/cover/11316219/.

Quoted 3 lines

The kvmgt patches were posted, you could just include them in your next series and clean everything up. You can get them at
https://patchwork.kernel.org/cover/11316219/.

Good to know!

Maybe I'll simply drop all the redundants in the dirty ring series assuming it's there? Since these patchsets should not overlap with each other (so looks more like an ordering constraints for merging).

Quoted 8 lines · Peter Xu

Quoted 3 lines

The kvmgt patches were posted, you could just include them in your next series and clean everything up. You can get them at
https://patchwork.kernel.org/cover/11316219/.

Good to know!

Maybe I'll simply drop all the redundants in the dirty ring series assuming it's there? Since these patchsets should not overlap with each other (so looks more like an ordering constraints for merging).

Just include the patches, we'll make sure to get an ACK from Alex.

Quoted 6 lines · Paolo Bonzini

Quoted 5 lines · Peter Xu

Quoted 3 lines

The kvmgt patches were posted, you could just include them in your next series and clean everything up. You can get them at
https://patchwork.kernel.org/cover/11316219/.

Good to know!

Quoted 5 lines

Quoted 3 lines

Maybe I'll simply drop all the redundants in the dirty ring series assuming it's there? Since these patchsets should not overlap with each other (so looks more like an ordering constraints for merging).

Just include the patches, we'll make sure to get an ACK from Alex.

Sure. I can even wait for some more days until that consolidates (just in case we need to change back and forth for this series).