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,
Jinrong Liang <cloudliang@tencent.com>,
Like Xu <likexu@tencent.com>
Subject: [PATCH v5 09/13] KVM: selftests: Test Intel PMU architectural events on fixed counters
Date: Mon, 23 Oct 2023 17:26:29 -0700 [thread overview]
Message-ID: <20231024002633.2540714-10-seanjc@google.com> (raw)
In-Reply-To: <20231024002633.2540714-1-seanjc@google.com>
From: Jinrong Liang <cloudliang@tencent.com>
Update test to cover Intel PMU architectural events on fixed counters.
Per Intel SDM, PMU users can also count architecture performance events
on fixed counters (specifically, FIXED_CTR0 for the retired instructions
and FIXED_CTR1 for cpu core cycles event). Therefore, if guest's CPUID
indicates that an architecture event is not available, the corresponding
fixed counter will also not count that event.
Co-developed-by: Like Xu <likexu@tencent.com>
Signed-off-by: Like Xu <likexu@tencent.com>
Signed-off-by: Jinrong Liang <cloudliang@tencent.com>
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
.../selftests/kvm/x86_64/pmu_counters_test.c | 54 ++++++++++++++++---
1 file changed, 46 insertions(+), 8 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 2a6336b994d5..410d09f788ef 100644
--- a/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c
+++ b/tools/testing/selftests/kvm/x86_64/pmu_counters_test.c
@@ -85,23 +85,44 @@ static void guest_measure_pmu_v1(struct kvm_x86_pmu_feature event,
GUEST_DONE();
}
+#define X86_PMU_FEATURE_NULL \
+({ \
+ struct kvm_x86_pmu_feature feature = {}; \
+ \
+ feature; \
+})
+
+static bool pmu_is_null_feature(struct kvm_x86_pmu_feature event)
+{
+ return !(*(u64 *)&event);
+}
+
static void guest_measure_loop(uint8_t idx)
{
const struct {
struct kvm_x86_pmu_feature gp_event;
+ struct kvm_x86_pmu_feature fixed_event;
} intel_event_to_feature[] = {
- [INTEL_ARCH_CPU_CYCLES] = { X86_PMU_FEATURE_CPU_CYCLES },
- [INTEL_ARCH_INSTRUCTIONS_RETIRED] = { X86_PMU_FEATURE_INSNS_RETIRED },
- [INTEL_ARCH_REFERENCE_CYCLES] = { X86_PMU_FEATURE_REFERENCE_CYCLES },
- [INTEL_ARCH_LLC_REFERENCES] = { X86_PMU_FEATURE_LLC_REFERENCES },
- [INTEL_ARCH_LLC_MISSES] = { X86_PMU_FEATURE_LLC_MISSES },
- [INTEL_ARCH_BRANCHES_RETIRED] = { X86_PMU_FEATURE_BRANCH_INSNS_RETIRED },
- [INTEL_ARCH_BRANCHES_MISPREDICTED] = { X86_PMU_FEATURE_BRANCHES_MISPREDICTED },
+ [INTEL_ARCH_CPU_CYCLES] = { X86_PMU_FEATURE_CPU_CYCLES, X86_PMU_FEATURE_CPU_CYCLES_FIXED },
+ [INTEL_ARCH_INSTRUCTIONS_RETIRED] = { X86_PMU_FEATURE_INSNS_RETIRED, X86_PMU_FEATURE_INSNS_RETIRED_FIXED },
+ /*
+ * Note, the fixed counter for reference cycles is NOT the same
+ * as the general purpose architectural event (because the GP
+ * event is garbage). The fixed counter explicitly counts at
+ * the same frequency as the TSC, whereas the GP event counts
+ * at a fixed, but uarch specific, frequency. Bundle them here
+ * for simplicity.
+ */
+ [INTEL_ARCH_REFERENCE_CYCLES] = { X86_PMU_FEATURE_REFERENCE_CYCLES, X86_PMU_FEATURE_REFERENCE_CYCLES_FIXED },
+ [INTEL_ARCH_LLC_REFERENCES] = { X86_PMU_FEATURE_LLC_REFERENCES, X86_PMU_FEATURE_NULL },
+ [INTEL_ARCH_LLC_MISSES] = { X86_PMU_FEATURE_LLC_MISSES, X86_PMU_FEATURE_NULL },
+ [INTEL_ARCH_BRANCHES_RETIRED] = { X86_PMU_FEATURE_BRANCH_INSNS_RETIRED, X86_PMU_FEATURE_NULL },
+ [INTEL_ARCH_BRANCHES_MISPREDICTED] = { X86_PMU_FEATURE_BRANCHES_MISPREDICTED, X86_PMU_FEATURE_NULL },
};
uint32_t nr_gp_counters = this_cpu_property(X86_PROPERTY_PMU_NR_GP_COUNTERS);
uint32_t pmu_version = this_cpu_property(X86_PROPERTY_PMU_VERSION);
- struct kvm_x86_pmu_feature gp_event;
+ struct kvm_x86_pmu_feature gp_event, fixed_event;
uint32_t counter_msr;
unsigned int i;
@@ -132,6 +153,23 @@ static void guest_measure_loop(uint8_t idx)
GUEST_ASSERT_EQ(this_pmu_has(gp_event), !!_rdpmc(i));
}
+ fixed_event = intel_event_to_feature[idx].fixed_event;
+ if (pmu_is_null_feature(fixed_event) || !this_pmu_has(fixed_event))
+ goto done;
+
+ i = fixed_event.f.bit;
+
+ wrmsr(MSR_CORE_PERF_FIXED_CTR0 + i, 0);
+ wrmsr(MSR_CORE_PERF_FIXED_CTR_CTRL, BIT_ULL(4 * i));
+
+ wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, BIT_ULL(PMC_IDX_FIXED + i));
+ __asm__ __volatile__("loop ." : "+c"((int){NUM_BRANCHES}));
+ wrmsr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+
+ if (pmu_is_intel_event_stable(idx))
+ GUEST_ASSERT_NE(_rdpmc(PMC_FIXED_RDPMC_BASE | i), 0);
+
+done:
GUEST_DONE();
}
--
2.42.0.758.gaed0368e0e-goog
next prev parent reply other threads:[~2023-10-24 0:27 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-24 0:26 [PATCH v5 00/13] KVM: x86/pmu: selftests: Fixes and new tests Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 01/13] KVM: x86/pmu: Don't allow exposing unsupported architectural events Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 02/13] KVM: x86/pmu: Don't enumerate support for fixed counters KVM can't virtualize Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 03/13] KVM: x86/pmu: Always treat Fixed counters as available when supported Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 04/13] KVM: selftests: Add vcpu_set_cpuid_property() to set properties Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 05/13] KVM: selftests: Drop the "name" param from KVM_X86_PMU_FEATURE() Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 06/13] KVM: selftests: Extend {kvm,this}_pmu_has() to support fixed counters Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 07/13] KVM: selftests: Add pmu.h and lib/pmu.c for common PMU assets Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 08/13] KVM: selftests: Test Intel PMU architectural events on gp counters Sean Christopherson
2023-10-24 19:49 ` Sean Christopherson
2023-10-24 21:00 ` Sean Christopherson
2023-10-25 3:17 ` JinrongLiang
2023-10-26 20:38 ` Mingwei Zhang
2023-10-26 20:54 ` Sean Christopherson
2023-10-26 22:10 ` Mingwei Zhang
2023-10-26 22:54 ` Sean Christopherson
2023-10-24 0:26 ` Sean Christopherson [this message]
2023-10-24 0:26 ` [PATCH v5 10/13] KVM: selftests: Test consistency of CPUID with num of " Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 11/13] KVM: selftests: Test consistency of CPUID with num of fixed counters Sean Christopherson
2023-10-24 11:40 ` JinrongLiang
2023-10-24 14:22 ` Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 12/13] KVM: selftests: Add functional test for Intel's fixed PMU counters Sean Christopherson
2023-10-24 0:26 ` [PATCH v5 13/13] KVM: selftests: Extend PMU counters test to permute on vPMU version Sean Christopherson
2023-10-24 11:49 ` JinrongLiang
2023-10-24 14:23 ` 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=20231024002633.2540714-10-seanjc@google.com \
--to=seanjc@google.com \
--cc=cloudliang@tencent.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