[PATCH v2 0/2] KVM: x86: read the PML log in the same order it was written

[PATCH v2 0/2] KVM: x86: read the PML log in the same order it was written

[Maxim Levitsky]

Reverse the order in which
the PML log is read to align more closely to the hardware. It should
not affect regular users of the dirty logging but it fixes a unit test
specific assumption in the dirty_log_test dirty-ring mode.

Best regards,
       Maxim Levitsky

Maxim Levitsky (2):
  KVM: VMX: refactor PML terminology
  KVM: VMX: read the PML log in the same order as it was written

 arch/x86/kvm/vmx/main.c   |  2 +-
 arch/x86/kvm/vmx/nested.c |  2 +-
 arch/x86/kvm/vmx/vmx.c    | 32 ++++++++++++++++++++------------
 arch/x86/kvm/vmx/vmx.h    |  5 ++++-
 4 files changed, 26 insertions(+), 15 deletions(-)

-- 
2.26.3

[PATCH v2 1/2] KVM: VMX: refactor PML terminology

[Maxim Levitsky]

Rename PML_ENTITY_NUM to PML_LOG_NR_ENTRIES
Add PML_HEAD_INDEX to specify the first entry that CPU writes.

No functional change intended.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
---
 arch/x86/kvm/vmx/main.c   |  2 +-
 arch/x86/kvm/vmx/nested.c |  2 +-
 arch/x86/kvm/vmx/vmx.c    | 10 +++++-----
 arch/x86/kvm/vmx/vmx.h    |  5 ++++-
 4 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kvm/vmx/main.c b/arch/x86/kvm/vmx/main.c
index 92d35cc6cd15..0c0ea786a947 100644
--- a/arch/x86/kvm/vmx/main.c
+++ b/arch/x86/kvm/vmx/main.c
@@ -126,7 +126,7 @@ struct kvm_x86_ops vt_x86_ops __initdata = {
 	.check_intercept = vmx_check_intercept,
 	.handle_exit_irqoff = vmx_handle_exit_irqoff,
 
-	.cpu_dirty_log_size = PML_ENTITY_NUM,
+	.cpu_dirty_log_size = PML_LOG_NR_ENTRIES,
 	.update_cpu_dirty_logging = vmx_update_cpu_dirty_logging,
 
 	.nested_ops = &vmx_nested_ops,
diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
index aa78b6f38dfe..71e0d887ff9c 100644
--- a/arch/x86/kvm/vmx/nested.c
+++ b/arch/x86/kvm/vmx/nested.c
@@ -3442,7 +3442,7 @@ static int nested_vmx_write_pml_buffer(struct kvm_vcpu *vcpu, gpa_t gpa)
 	if (!nested_cpu_has_pml(vmcs12))
 		return 0;
 
-	if (vmcs12->guest_pml_index >= PML_ENTITY_NUM) {
+	if (vmcs12->guest_pml_index >= PML_LOG_NR_ENTRIES) {
 		vmx->nested.pml_full = true;
 		return 1;
 	}
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 0f008f5ef6f0..30fc54eefeb4 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4828,7 +4828,7 @@ static void init_vmcs(struct vcpu_vmx *vmx)
 
 	if (enable_pml) {
 		vmcs_write64(PML_ADDRESS, page_to_phys(vmx->pml_pg));
-		vmcs_write16(GUEST_PML_INDEX, PML_ENTITY_NUM - 1);
+		vmcs_write16(GUEST_PML_INDEX, PML_HEAD_INDEX);
 	}
 
 	vmx_write_encls_bitmap(&vmx->vcpu, NULL);
@@ -6216,17 +6216,17 @@ static void vmx_flush_pml_buffer(struct kvm_vcpu *vcpu)
 	pml_idx = vmcs_read16(GUEST_PML_INDEX);
 
 	/* Do nothing if PML buffer is empty */
-	if (pml_idx == (PML_ENTITY_NUM - 1))
+	if (pml_idx == PML_HEAD_INDEX)
 		return;
 
 	/* PML index always points to next available PML buffer entity */
-	if (pml_idx >= PML_ENTITY_NUM)
+	if (pml_idx >= PML_LOG_NR_ENTRIES)
 		pml_idx = 0;
 	else
 		pml_idx++;
 
 	pml_buf = page_address(vmx->pml_pg);
-	for (; pml_idx < PML_ENTITY_NUM; pml_idx++) {
+	for (; pml_idx < PML_LOG_NR_ENTRIES; pml_idx++) {
 		u64 gpa;
 
 		gpa = pml_buf[pml_idx];
@@ -6235,7 +6235,7 @@ static void vmx_flush_pml_buffer(struct kvm_vcpu *vcpu)
 	}
 
 	/* reset PML index */
-	vmcs_write16(GUEST_PML_INDEX, PML_ENTITY_NUM - 1);
+	vmcs_write16(GUEST_PML_INDEX, PML_HEAD_INDEX);
 }
 
 static void vmx_dump_sel(char *name, uint32_t sel)
diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h
index 43f573f6ca46..83ca5c574704 100644
--- a/arch/x86/kvm/vmx/vmx.h
+++ b/arch/x86/kvm/vmx/vmx.h
@@ -330,7 +330,10 @@ struct vcpu_vmx {
 	bool ple_window_dirty;
 
 	/* Support for PML */
-#define PML_ENTITY_NUM		512
+#define PML_LOG_NR_ENTRIES	512
+	/* PML is written backwards: this is the first entry written by the CPU */
+#define PML_HEAD_INDEX		(PML_LOG_NR_ENTRIES-1)
+
 	struct page *pml_pg;
 
 	/* apic deadline value in host tsc */
-- 
2.26.3

[PATCH v2 2/2] KVM: VMX: read the PML log in the same order as it was written

[Maxim Levitsky]

Intel's PRM specifies that the CPU writes to the PML log 'backwards'
or in other words, it first writes entry 511, then entry 510 and so on.

I also confirmed on the bare metal that the CPU indeed writes the entries
in this order.

KVM on the other hand, reads the entries in the opposite order, from the
last written entry and towards entry 511 and dumps them in this order to
the dirty ring.

Usually this doesn't matter, except for one complex nesting case:

KVM reties the instructions that cause MMU faults.
This might cause an emulated PML log entry to be visible to L1's hypervisor
before the actual memory write was committed.

This happens when the L0 MMU fault is followed directly by the VM exit to
L1, for example due to a pending L1 interrupt or due to the L1's
'PML log full' event.

This problem doesn't have a noticeable real-world impact because this
write retry is not much different from the guest writing to the same page
multiple times, which is also not reflected in the dirty log. The users of
the dirty logging only rely on correct reporting of the clean pages, or
in other words they assume that if a page is clean, then no writes were
committed to it since the moment it was marked clean.

However KVM has a kvm_dirty_log_test selftest, a test that tests both
the clean and the dirty pages vs the memory contents, and can fail if it
detects a dirty page which has an old value at the offset 0 which the test
writes.

To avoid failure, the test has a workaround for this specific problem:

The test skips checking memory that belongs to the last dirty ring entry,
which it has seen, relying on the fact that as long as memory writes are
committed in-order, only the last entry can belong to a not yet committed
memory write.

However, since L1's KVM is reading the PML log in the opposite direction
that L0 wrote it, the last dirty ring entry often will be not the last
entry written by the L0.

To fix this, switch the order in which KVM reads the PML log.

Note that this issue is not present on the bare metal, because on the
bare metal, an update of the A/D bits of a present entry, PML logging and
the actual memory write are all done by the CPU without any hypervisor
intervention and pending interrupt evaluation, thus once a PML log and/or
vCPU kick happens, all memory writes that are in the PML log are
committed to memory.

The only exception to this rule is when the guest hits a not present EPT
entry, in which case KVM first reads (backward) the PML log, dumps it to
the dirty ring, and *then* sets up a SPTE entry with A/D bits set, and logs
this to the dirty ring, thus making the entry be the last one in the
dirty ring.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
---
 arch/x86/kvm/vmx/vmx.c | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 30fc54eefeb4..25ea43a8efb2 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -6211,25 +6211,33 @@ static void vmx_flush_pml_buffer(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u64 *pml_buf;
-	u16 pml_idx;
+	u16 pml_idx, pml_tail_index;
+	int i;
 
 	pml_idx = vmcs_read16(GUEST_PML_INDEX);
 
 	/* Do nothing if PML buffer is empty */
 	if (pml_idx == PML_HEAD_INDEX)
 		return;
+	/*
+	 * PML index always points to the next available PML buffer entity
+	 * unless PML log has just overflowed.
+	 */
+	pml_tail_index = (pml_idx >= PML_LOG_NR_ENTRIES) ? 0 : pml_idx + 1;
 
-	/* PML index always points to next available PML buffer entity */
-	if (pml_idx >= PML_LOG_NR_ENTRIES)
-		pml_idx = 0;
-	else
-		pml_idx++;
-
+	/*
+	 * PML log is written backwards: the CPU first writes the entry 511
+	 * then the entry 510, and so on.
+	 *
+	 * Read the entries in the same order they were written, to ensure that
+	 * the dirty ring is filled in the same order the CPU wrote them.
+	 */
 	pml_buf = page_address(vmx->pml_pg);
-	for (; pml_idx < PML_LOG_NR_ENTRIES; pml_idx++) {
+
+	for (i = PML_HEAD_INDEX; i >= pml_tail_index; i--) {
 		u64 gpa;
 
-		gpa = pml_buf[pml_idx];
+		gpa = pml_buf[i];
 		WARN_ON(gpa & (PAGE_SIZE - 1));
 		kvm_vcpu_mark_page_dirty(vcpu, gpa >> PAGE_SHIFT);
 	}
-- 
2.26.3

Re: [PATCH v2 0/2] KVM: x86: read the PML log in the same order it was written

[Sean Christopherson]

On Thu, 19 Dec 2024 17:10:32 -0500, Maxim Levitsky wrote:
> Reverse the order in which
> the PML log is read to align more closely to the hardware. It should
> not affect regular users of the dirty logging but it fixes a unit test
> specific assumption in the dirty_log_test dirty-ring mode.
> 
> Best regards,
>        Maxim Levitsky
> 
> [...]

Applied to kvm-x86 vmx, thanks!

[1/2] KVM: VMX: refactor PML terminology
      https://github.com/kvm-x86/linux/commit/ae81ce936ff9
[2/2] KVM: VMX: read the PML log in the same order as it was written
      https://github.com/kvm-x86/linux/commit/37c3ddfe5238

--
https://github.com/kvm-x86/linux/tree/next