Received-SPF: pass (google.com: domain of linux-kernel+bounces-108286-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Reply-To: Sean Christopherson <seanjc@google.com>
Date: Tue, 19 Mar 2024 17:50:24 -0700
In-Reply-To: <20240320005024.3216282-1-seanjc@google.com>
Precedence: bulk
Mime-Version: 1.0
References: <20240320005024.3216282-1-seanjc@google.com>
Message-ID: <20240320005024.3216282-5-seanjc@google.com>
Subject: [RFC PATCH 4/4] KVM: x86/mmu: Don't force flush if SPTE update clears
 Accessed bit
From: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>, Sean Christopherson <seanjc@google.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, 
	David Hildenbrand <david@redhat.com>, David Matlack <dmatlack@google.com>, 
	David Stevens <stevensd@chromium.org>, Matthew Wilcox <willy@infradead.org>
Content-Type: text/plain; charset="UTF-8"

Don't force a TLB if mmu_spte_update() clears Accessed bit, as access
tracking tolerates false negatives, as evidenced by the mmu_notifier hooks
that explicit test and age SPTEs without doing a TLB flush.

In practice, this is very nearly a nop.  spte_write_protect() and
spte_clear_dirty() never clear the Accessed bit.  make_spte() always
sets the Accessed bit for !prefetch scenarios.  FNAME(sync_spte) only sets
SPTE if the protection bits are changing, i.e. if a flush will be needed
regardless of the Accessed bits.  And FNAME(pte_prefetch) sets SPTE if and
only if the old SPTE is !PRESENT.

That leaves kvm_arch_async_page_ready() as the one path that will generate
a !ACCESSED SPTE *and* overwrite a PRESENT SPTE.  And that's very arguably
a bug, as clobbering a valid SPTE in that case is nonsensical.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/mmu/mmu.c | 31 +++++++++----------------------
 1 file changed, 9 insertions(+), 22 deletions(-)

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 0a6c6619d213..77d1072b130d 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -515,37 +515,24 @@ static u64 mmu_spte_update_no_track(u64 *sptep, u64 new_spte)
  * TLBs must be flushed. Otherwise rmap_write_protect will find a read-only
  * spte, even though the writable spte might be cached on a CPU's TLB.
  *
+ * Remote TLBs also need to be flushed if the Dirty bit is cleared, as false
+ * negatives are not acceptable, e.g. if KVM is using D-bit based PML on VMX.
+ *
+ * Don't flush if the Accessed bit is cleared, as access tracking tolerates
+ * false negatives, and the one path that does care about TLB flushes,
+ * kvm_mmu_notifier_clear_flush_young(), uses mmu_spte_update_no_track().
+ *
  * Returns true if the TLB needs to be flushed
  */
 static bool mmu_spte_update(u64 *sptep, u64 new_spte)
 {
-	bool flush = false;
 	u64 old_spte = mmu_spte_update_no_track(sptep, new_spte);
 
 	if (!is_shadow_present_pte(old_spte))
 		return false;
 
-	/*
-	 * For the spte updated out of mmu-lock is safe, since
-	 * we always atomically update it, see the comments in
-	 * spte_has_volatile_bits().
-	 */
-	if (is_mmu_writable_spte(old_spte) &&
-	      !is_writable_pte(new_spte))
-		flush = true;
-
-	/*
-	 * Flush TLB when accessed/dirty states are changed in the page tables,
-	 * to guarantee consistency between TLB and page tables.
-	 */
-
-	if (is_accessed_spte(old_spte) && !is_accessed_spte(new_spte))
-		flush = true;
-
-	if (is_dirty_spte(old_spte) && !is_dirty_spte(new_spte))
-		flush = true;
-
-	return flush;
+	return (is_mmu_writable_spte(old_spte) && !is_writable_pte(new_spte)) ||
+	       (is_dirty_spte(old_spte) && !is_dirty_spte(new_spte));
 }
 
 /*
-- 
2.44.0.291.gc1ea87d7ee-goog