[PATCH 2/2] kvm: x86/mmu: Add TDP MMU SPTE changed trace point

From: Ben Gardon <bgardon@google.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: Paolo Bonzini <pbonzini@redhat.com>, Peter Xu <peterx@redhat.com>,
	Sean Christopherson <sean.j.christopherson@intel.com>,
	Peter Shier <pshier@google.com>,
	Junaid Shahid <junaids@google.com>,
	Peter Feiner <pfeiner@google.com>,
	Ben Gardon <bgardon@google.com>
Subject: [PATCH 2/2] kvm: x86/mmu: Add TDP MMU SPTE changed trace point
Date: Tue, 27 Oct 2020 10:59:44 -0700	[thread overview]
Message-ID: <20201027175944.1183301-2-bgardon@google.com> (raw)
In-Reply-To: <20201027175944.1183301-1-bgardon@google.com>

Add an extremely verbose trace point to the TDP MMU to log all SPTE
changes, regardless of callstack / motivation. This is useful when a
complete picture of the paging structure is needed or a change cannot be
explained with the other, existing trace points.

Tested: ran the demand paging selftest on an Intel Skylake machine with
	all the trace points used by the TDP MMU enabled and observed
	them firing with expected values.

This patch can be viewed in Gerrit at:
https://linux-review.googlesource.com/c/virt/kvm/kvm/+/3813

Signed-off-by: Ben Gardon <bgardon@google.com>
---
 arch/x86/kvm/mmu/mmutrace.h | 29 +++++++++++++++++++++++++++++
 arch/x86/kvm/mmu/tdp_mmu.c  |  2 ++
 2 files changed, 31 insertions(+)

diff --git a/arch/x86/kvm/mmu/mmutrace.h b/arch/x86/kvm/mmu/mmutrace.h
index 213699b27b448..e798489b56b55 100644
--- a/arch/x86/kvm/mmu/mmutrace.h
+++ b/arch/x86/kvm/mmu/mmutrace.h
@@ -381,6 +381,35 @@ TRACE_EVENT(
 	)
 );
 
+TRACE_EVENT(
+	kvm_tdp_mmu_spte_changed,
+	TP_PROTO(int as_id, gfn_t gfn, int level, u64 old_spte, u64 new_spte),
+	TP_ARGS(as_id, gfn, level, old_spte, new_spte),
+
+	TP_STRUCT__entry(
+		__field(u64, gfn)
+		__field(u64, old_spte)
+		__field(u64, new_spte)
+		/* Level cannot be larger than 5 on x86, so it fits in a u8. */
+		__field(u8, level)
+		/* as_id can only be 0 or 1 x86, so it fits in a u8. */
+		__field(u8, as_id)
+	),
+
+	TP_fast_assign(
+		__entry->gfn = gfn;
+		__entry->old_spte = old_spte;
+		__entry->new_spte = new_spte;
+		__entry->level = level;
+		__entry->as_id = as_id;
+	),
+
+	TP_printk("as id %d gfn %llx level %d old_spte %llx new_spte %llx",
+		  __entry->as_id, __entry->gfn, __entry->level,
+		  __entry->old_spte, __entry->new_spte
+	)
+);
+
 #endif /* _TRACE_KVMMMU_H */
 
 #undef TRACE_INCLUDE_PATH
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 047e2d966abef..5820c36ccfdca 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -241,6 +241,8 @@ static void __handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn,
 	if (old_spte == new_spte)
 		return;
 
+	trace_kvm_tdp_mmu_spte_changed(as_id, gfn, level, old_spte, new_spte);
+
 	/*
 	 * The only times a SPTE should be changed from a non-present to
 	 * non-present state is when an MMIO entry is installed/modified/
-- 
2.29.0.rc2.309.g374f81d7ae-goog

