public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	 Kan Liang <kan.liang@linux.intel.com>,
	Dapeng Mi <dapeng1.mi@linux.intel.com>,
	 Jim Mattson <jmattson@google.com>,
	Jinrong Liang <cloudliang@tencent.com>,
	 Aaron Lewis <aaronlewis@google.com>,
	Like Xu <likexu@tencent.com>
Subject: [PATCH v10 21/29] KVM: selftests: Expand PMU counters test to verify LLC events
Date: Tue,  9 Jan 2024 15:02:41 -0800	[thread overview]
Message-ID: <20240109230250.424295-22-seanjc@google.com> (raw)
In-Reply-To: <20240109230250.424295-1-seanjc@google.com>

Expand the PMU counters test to verify that LLC references and misses have
non-zero counts when the code being executed while the LLC event(s) is
active is evicted via CFLUSH{,OPT}.  Note, CLFLUSH{,OPT} requires a fence
of some kind to ensure the cache lines are flushed before execution
continues.  Use MFENCE for simplicity (performance is not a concern).

Suggested-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 .../selftests/kvm/x86_64/pmu_counters_test.c  | 59 +++++++++++++------
 1 file changed, 40 insertions(+), 19 deletions(-)

diff --git a/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c b/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c
index f5dedd112471..4c7133ddcda8 100644
--- a/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c
+++ b/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c
@@ -14,9 +14,9 @@
 /*
  * Number of "extra" instructions that will be counted, i.e. the number of
  * instructions that are needed to set up the loop and then disabled the
- * counter.  2 MOV, 2 XOR, 1 WRMSR.
+ * counter.  1 CLFLUSH/CLFLUSHOPT/NOP, 1 MFENCE, 2 MOV, 2 XOR, 1 WRMSR.
  */
-#define NUM_EXTRA_INSNS		5
+#define NUM_EXTRA_INSNS		7
 #define NUM_INSNS_RETIRED	(NUM_BRANCHES + NUM_EXTRA_INSNS)
 
 static uint8_t kvm_pmu_version;
@@ -107,6 +107,12 @@ static void guest_assert_event_count(uint8_t idx,
 	case INTEL_ARCH_BRANCHES_RETIRED_INDEX:
 		GUEST_ASSERT_EQ(count, NUM_BRANCHES);
 		break;
+	case INTEL_ARCH_LLC_REFERENCES_INDEX:
+	case INTEL_ARCH_LLC_MISSES_INDEX:
+		if (!this_cpu_has(X86_FEATURE_CLFLUSHOPT) &&
+		    !this_cpu_has(X86_FEATURE_CLFLUSH))
+			break;
+		fallthrough;
 	case INTEL_ARCH_CPU_CYCLES_INDEX:
 	case INTEL_ARCH_REFERENCE_CYCLES_INDEX:
 		GUEST_ASSERT_NE(count, 0);
@@ -123,29 +129,44 @@ static void guest_assert_event_count(uint8_t idx,
 	GUEST_ASSERT_EQ(_rdpmc(pmc), 0xdead);
 }
 
+/*
+ * Enable and disable the PMC in a monolithic asm blob to ensure that the
+ * compiler can't insert _any_ code into the measured sequence.  Note, ECX
+ * doesn't need to be clobbered as the input value, @pmc_msr, is restored
+ * before the end of the sequence.
+ *
+ * If CLFUSH{,OPT} is supported, flush the cacheline containing (at least) the
+ * start of the loop to force LLC references and misses, i.e. to allow testing
+ * that those events actually count.
+ */
+#define GUEST_MEASURE_EVENT(_msr, _value, clflush)				\
+do {										\
+	__asm__ __volatile__("wrmsr\n\t"					\
+			     clflush "\n\t"					\
+			     "mfence\n\t"					\
+			     "1: mov $" __stringify(NUM_BRANCHES) ", %%ecx\n\t"	\
+			     "loop .\n\t"					\
+			     "mov %%edi, %%ecx\n\t"				\
+			     "xor %%eax, %%eax\n\t"				\
+			     "xor %%edx, %%edx\n\t"				\
+			     "wrmsr\n\t"					\
+			     :: "a"((uint32_t)_value), "d"(_value >> 32),	\
+				"c"(_msr), "D"(_msr)				\
+	);									\
+} while (0)
+
 static void __guest_test_arch_event(uint8_t idx, struct kvm_x86_pmu_feature event,
 				    uint32_t pmc, uint32_t pmc_msr,
 				    uint32_t ctrl_msr, uint64_t ctrl_msr_value)
 {
 	wrmsr(pmc_msr, 0);
 
-	/*
-	 * Enable and disable the PMC in a monolithic asm blob to ensure that
-	 * the compiler can't insert _any_ code into the measured sequence.
-	 * Note, ECX doesn't need to be clobbered as the input value, @pmc_msr,
-	 * is restored before the end of the sequence.
-	 */
-	__asm__ __volatile__("wrmsr\n\t"
-			     "mov $" __stringify(NUM_BRANCHES) ", %%ecx\n\t"
-			     "loop .\n\t"
-			     "mov %%edi, %%ecx\n\t"
-			     "xor %%eax, %%eax\n\t"
-			     "xor %%edx, %%edx\n\t"
-			     "wrmsr\n\t"
-			     :: "a"((uint32_t)ctrl_msr_value),
-				"d"(ctrl_msr_value >> 32),
-				"c"(ctrl_msr), "D"(ctrl_msr)
-			     );
+	if (this_cpu_has(X86_FEATURE_CLFLUSHOPT))
+		GUEST_MEASURE_EVENT(ctrl_msr, ctrl_msr_value, "clflushopt 1f");
+	else if (this_cpu_has(X86_FEATURE_CLFLUSH))
+		GUEST_MEASURE_EVENT(ctrl_msr, ctrl_msr_value, "clflush 1f");
+	else
+		GUEST_MEASURE_EVENT(ctrl_msr, ctrl_msr_value, "nop");
 
 	guest_assert_event_count(idx, event, pmc, pmc_msr);
 }
-- 
2.43.0.472.g3155946c3a-goog


  parent reply	other threads:[~2024-01-09 23:03 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-09 23:02 [PATCH v10 00/29] KVM: x86/pmu: selftests: Fixes and new tests Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 01/29] KVM: x86/pmu: Always treat Fixed counters as available when supported Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 02/29] KVM: x86/pmu: Allow programming events that match unsupported arch events Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 03/29] KVM: x86/pmu: Remove KVM's enumeration of Intel's architectural encodings Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 04/29] KVM: x86/pmu: Setup fixed counters' eventsel during PMU initialization Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 05/29] KVM: x86/pmu: Get eventsel for fixed counters from perf Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 06/29] KVM: x86/pmu: Don't ignore bits 31:30 for RDPMC index on AMD Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 07/29] KVM: x86/pmu: Prioritize VMX interception over #GP on RDPMC due to bad index Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 08/29] KVM: x86/pmu: Apply "fast" RDPMC only to Intel PMUs Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 09/29] KVM: x86/pmu: Disallow "fast" RDPMC for architectural " Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 10/29] KVM: x86/pmu: Treat "fixed" PMU type in RDPMC as index as a value, not flag Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 11/29] KVM: x86/pmu: Explicitly check for RDPMC of unsupported Intel PMC types Sean Christopherson
2024-01-12  3:50   ` Mi, Dapeng
2024-01-09 23:02 ` [PATCH v10 12/29] KVM: selftests: Add vcpu_set_cpuid_property() to set properties Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 13/29] KVM: selftests: Drop the "name" param from KVM_X86_PMU_FEATURE() Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 14/29] KVM: selftests: Extend {kvm,this}_pmu_has() to support fixed counters Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 15/29] KVM: selftests: Add pmu.h and lib/pmu.c for common PMU assets Sean Christopherson
2024-01-10  9:21   ` Andrew Jones
2024-01-10 13:58     ` Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 16/29] KVM: selftests: Test Intel PMU architectural events on gp counters Sean Christopherson
2024-01-12  9:14   ` Mi, Dapeng
2024-01-12 21:37     ` Sean Christopherson
2024-01-15  2:03       ` Mi, Dapeng
2024-01-30 23:27         ` Sean Christopherson
2024-01-31  2:11           ` Mi, Dapeng
2024-01-31 15:31             ` Sean Christopherson
2024-02-01  1:51               ` Mi, Dapeng
2024-01-09 23:02 ` [PATCH v10 17/29] KVM: selftests: Test Intel PMU architectural events on fixed counters Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 18/29] KVM: selftests: Test consistency of CPUID with num of gp counters Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 19/29] KVM: selftests: Test consistency of CPUID with num of fixed counters Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 20/29] KVM: selftests: Add functional test for Intel's fixed PMU counters Sean Christopherson
2024-01-09 23:02 ` Sean Christopherson [this message]
2024-01-09 23:02 ` [PATCH v10 22/29] KVM: selftests: Add a helper to query if the PMU module param is enabled Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 23/29] KVM: selftests: Add helpers to read integer module params Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 24/29] KVM: selftests: Query module param to detect FEP in MSR filtering test Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 25/29] KVM: selftests: Move KVM_FEP macro into common library header Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 26/29] KVM: selftests: Test PMC virtualization with forced emulation Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 27/29] KVM: selftests: Add a forced emulation variation of KVM_ASM_SAFE() Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 28/29] KVM: selftests: Add helpers for safe and safe+forced RDMSR, RDPMC, and XGETBV Sean Christopherson
2024-01-09 23:02 ` [PATCH v10 29/29] KVM: selftests: Extend PMU counters test to validate RDPMC after WRMSR Sean Christopherson
2024-01-12  9:17 ` [PATCH v10 00/29] KVM: x86/pmu: selftests: Fixes and new tests Mi, Dapeng
2024-01-31  0:59 ` Sean Christopherson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240109230250.424295-22-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=aaronlewis@google.com \
    --cc=cloudliang@tencent.com \
    --cc=dapeng1.mi@linux.intel.com \
    --cc=jmattson@google.com \
    --cc=kan.liang@linux.intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=likexu@tencent.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox