mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Mingwei Zhang <mizhang@google.com>
To: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	 Arnaldo Carvalho de Melo <acme@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	 Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
	 Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>,  Ian Rogers <irogers@google.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Liang@google.com,  Kan <kan.liang@linux.intel.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	 linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	 kvm@vger.kernel.org, linux-kselftest@vger.kernel.org,
	 Mingwei Zhang <mizhang@google.com>,
	Yongwei Ma <yongwei.ma@intel.com>,
	 Xiong Zhang <xiong.y.zhang@linux.intel.com>,
	Dapeng Mi <dapeng1.mi@linux.intel.com>,
	 Jim Mattson <jmattson@google.com>,
	Sandipan Das <sandipan.das@amd.com>,
	 Zide Chen <zide.chen@intel.com>,
	Eranian Stephane <eranian@google.com>,
	 Das Sandipan <Sandipan.Das@amd.com>,
	Shukla Manali <Manali.Shukla@amd.com>,
	 Nikunj Dadhania <nikunj.dadhania@amd.com>
Subject: [PATCH v4 01/38] perf: Support get/put mediated PMU interfaces
Date: Mon, 24 Mar 2025 17:30:41 +0000	[thread overview]
Message-ID: <20250324173121.1275209-2-mizhang@google.com> (raw)
In-Reply-To: <20250324173121.1275209-1-mizhang@google.com>

From: Kan Liang <kan.liang@linux.intel.com>

Currently, the guest and host share the PMU resources when a guest is
running. KVM has to create an extra virtual event to simulate the
guest's event, which brings several issues, e.g., high overhead, not
accuracy and etc.

A new mediated PMU method is proposed to address the issue. It requires
that the PMU resources can be fully occupied by the guest while it's
running. Two new interfaces are implemented to fulfill the requirement.
The hypervisor should invoke the interface while creating a guest which
wants the mediated PMU capability.

The PMU resources should only be temporarily occupied as a whole when a
guest is running. When the guest is out, the PMU resources are still
shared among different users.

The exclude_guest event modifier is used to guarantee the exclusive
occupation of the PMU resources. When creating a guest, the hypervisor
should check whether there are !exclude_guest events in the system.
If yes, the creation should fail. Because some PMU resources have been
occupied by other users.
If no, the PMU resources can be safely accessed by the guest directly.
Perf guarantees that no new !exclude_guest events are created when a
guest is running.

Only the mediated PMU is affected, but not for other PMU e.g., uncore
and SW PMU. The behavior of those PMUs are not changed. The guest
enter/exit interfaces should only impact the supported PMUs.
Add a new PERF_PMU_CAP_MEDIATED_VPMU flag to indicate the PMUs that
support the feature.

Add nr_include_guest_events to track the !exclude_guest events of PMU
with PERF_PMU_CAP_MEDIATED_VPMU.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Mingwei Zhang <mizhang@google.com>
---
 include/linux/perf_event.h | 11 +++++++
 kernel/events/core.c       | 66 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 77 insertions(+)

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 8333f132f4a9..54018dd0b2a4 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -301,6 +301,8 @@ struct perf_event_pmu_context;
 #define PERF_PMU_CAP_AUX_OUTPUT			0x0080
 #define PERF_PMU_CAP_EXTENDED_HW_TYPE		0x0100
 #define PERF_PMU_CAP_AUX_PAUSE			0x0200
+/* Support to passthrough whole PMU resoure to guest */
+#define PERF_PMU_CAP_MEDIATED_VPMU		0x0400
 
 /**
  * pmu::scope
@@ -1811,6 +1813,8 @@ extern void perf_event_task_tick(void);
 extern int perf_event_account_interrupt(struct perf_event *event);
 extern int perf_event_period(struct perf_event *event, u64 value);
 extern u64 perf_event_pause(struct perf_event *event, bool reset);
+int perf_get_mediated_pmu(void);
+void perf_put_mediated_pmu(void);
 #else /* !CONFIG_PERF_EVENTS: */
 static inline void *
 perf_aux_output_begin(struct perf_output_handle *handle,
@@ -1901,6 +1905,13 @@ static inline int perf_exclude_event(struct perf_event *event, struct pt_regs *r
 {
 	return 0;
 }
+
+static inline int perf_get_mediated_pmu(void)
+{
+	return 0;
+}
+
+static inline void perf_put_mediated_pmu(void)			{ }
 #endif
 
 #if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_INTEL)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index bcb09e011e9e..be623701dc48 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -431,6 +431,20 @@ static atomic_t nr_bpf_events __read_mostly;
 static atomic_t nr_cgroup_events __read_mostly;
 static atomic_t nr_text_poke_events __read_mostly;
 static atomic_t nr_build_id_events __read_mostly;
+static atomic_t nr_include_guest_events __read_mostly;
+
+static atomic_t nr_mediated_pmu_vms;
+static DEFINE_MUTEX(perf_mediated_pmu_mutex);
+
+/* !exclude_guest event of PMU with PERF_PMU_CAP_MEDIATED_VPMU */
+static inline bool is_include_guest_event(struct perf_event *event)
+{
+	if ((event->pmu->capabilities & PERF_PMU_CAP_MEDIATED_VPMU) &&
+	    !event->attr.exclude_guest)
+		return true;
+
+	return false;
+}
 
 static LIST_HEAD(pmus);
 static DEFINE_MUTEX(pmus_lock);
@@ -5320,6 +5334,9 @@ static void _free_event(struct perf_event *event)
 
 	unaccount_event(event);
 
+	if (is_include_guest_event(event))
+		atomic_dec(&nr_include_guest_events);
+
 	security_perf_event_free(event);
 
 	if (event->rb) {
@@ -5877,6 +5894,36 @@ u64 perf_event_pause(struct perf_event *event, bool reset)
 }
 EXPORT_SYMBOL_GPL(perf_event_pause);
 
+/*
+ * Currently invoked at VM creation to
+ * - Check whether there are existing !exclude_guest events of PMU with
+ *   PERF_PMU_CAP_MEDIATED_VPMU
+ * - Set nr_mediated_pmu_vms to prevent !exclude_guest event creation on
+ *   PMUs with PERF_PMU_CAP_MEDIATED_VPMU
+ *
+ * No impact for the PMU without PERF_PMU_CAP_MEDIATED_VPMU. The perf
+ * still owns all the PMU resources.
+ */
+int perf_get_mediated_pmu(void)
+{
+	guard(mutex)(&perf_mediated_pmu_mutex);
+	if (atomic_inc_not_zero(&nr_mediated_pmu_vms))
+		return 0;
+
+	if (atomic_read(&nr_include_guest_events))
+		return -EBUSY;
+
+	atomic_inc(&nr_mediated_pmu_vms);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_get_mediated_pmu);
+
+void perf_put_mediated_pmu(void)
+{
+	atomic_dec(&nr_mediated_pmu_vms);
+}
+EXPORT_SYMBOL_GPL(perf_put_mediated_pmu);
+
 /*
  * Holding the top-level event's child_mutex means that any
  * descendant process that has inherited this event will block
@@ -12210,6 +12257,17 @@ static void account_event(struct perf_event *event)
 	account_pmu_sb_event(event);
 }
 
+static int perf_account_include_guest_event(void)
+{
+	guard(mutex)(&perf_mediated_pmu_mutex);
+
+	if (atomic_read(&nr_mediated_pmu_vms))
+		return -EOPNOTSUPP;
+
+	atomic_inc(&nr_include_guest_events);
+	return 0;
+}
+
 /*
  * Allocate and initialize an event structure
  */
@@ -12435,11 +12493,19 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 	if (err)
 		goto err_callchain_buffer;
 
+	if (is_include_guest_event(event)) {
+		err = perf_account_include_guest_event();
+		if (err)
+			goto err_security_alloc;
+	}
+
 	/* symmetric to unaccount_event() in _free_event() */
 	account_event(event);
 
 	return event;
 
+err_security_alloc:
+	security_perf_event_free(event);
 err_callchain_buffer:
 	if (!event->parent) {
 		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
-- 
2.49.0.395.g12beb8f557-goog


  reply	other threads:[~2025-03-24 17:32 UTC|newest]

Thread overview: 128+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-24 17:30 [PATCH v4 00/38] Mediated vPMU 4.0 for x86 Mingwei Zhang
2025-03-24 17:30 ` Mingwei Zhang [this message]
2025-05-14 22:48   ` [PATCH v4 01/38] perf: Support get/put mediated PMU interfaces Sean Christopherson
2025-05-15  1:31     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 02/38] perf: Skip pmu_ctx based on event_type Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 03/38] perf: Clean up perf ctx time Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 04/38] perf: Add a EVENT_GUEST flag Mingwei Zhang
2025-05-14 22:51   ` Sean Christopherson
2025-05-15  1:35     ` Mi, Dapeng
2025-05-19  6:58   ` Namhyung Kim
2025-05-20 16:09     ` Liang, Kan
2025-05-20 17:51       ` Namhyung Kim
2025-05-20 18:50         ` Liang, Kan
2025-05-21 19:46   ` Namhyung Kim
2025-03-24 17:30 ` [PATCH v4 05/38] perf: Add generic exclude_guest support Mingwei Zhang
2025-04-25 11:13   ` Peter Zijlstra
2025-05-14 23:19     ` Sean Christopherson
2025-05-15  1:37       ` Mi, Dapeng
2025-05-15 18:39       ` Liang, Kan
2025-05-15 19:25         ` Sean Christopherson
2025-05-15 20:18           ` Liang, Kan
2025-05-21 19:55   ` Namhyung Kim
2025-05-21 20:12     ` Liang, Kan
2025-03-24 17:30 ` [PATCH v4 06/38] x86/irq: Factor out common code for installing kvm irq handler Mingwei Zhang
2025-05-14 23:21   ` Sean Christopherson
2025-05-15  2:10     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 07/38] perf: core/x86: Register a new vector for KVM GUEST PMI Mingwei Zhang
2025-05-14 23:24   ` Sean Christopherson
2025-05-15  1:40     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 08/38] KVM: x86/pmu: Register KVM_GUEST_PMI_VECTOR handler Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 09/38] perf: Add switch_guest_ctx() interface Mingwei Zhang
2025-04-25 11:12   ` Peter Zijlstra
2025-05-14 23:30   ` Sean Christopherson
2025-05-15  1:45     ` Mi, Dapeng
2025-05-21 20:01   ` Namhyung Kim
2025-03-24 17:30 ` [PATCH v4 10/38] perf/x86: Support switch_guest_ctx interface Mingwei Zhang
2025-04-25 11:15   ` Peter Zijlstra
2025-04-25 13:06     ` Liang, Kan
2025-04-25 13:43       ` Peter Zijlstra
2025-04-25 13:56         ` Liang, Kan
2025-07-30  0:31           ` Sean Christopherson
2025-03-24 17:30 ` [PATCH v4 11/38] perf/x86: Forbid PMI handler when guest own PMU Mingwei Zhang
2025-05-15  0:00   ` Sean Christopherson
2025-05-15  1:52     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 12/38] perf/x86/core: Do not set bit width for unavailable counters Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 13/38] perf/x86/core: Plumb mediated PMU capability from x86_pmu to x86_pmu_cap Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 14/38] KVM: x86/pmu: Introduce enable_mediated_pmu global parameter Mingwei Zhang
2025-05-15  0:09   ` Sean Christopherson
2025-05-15  2:53     ` Mi, Dapeng
2025-05-21 18:43       ` Sean Christopherson
2025-05-22  1:36         ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 15/38] KVM: x86/pmu: Check PMU cpuid configuration from user space Mingwei Zhang
2025-05-15  0:12   ` Sean Christopherson
2025-05-15  3:00     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 16/38] KVM: x86: Rename vmx_vmentry/vmexit_ctrl() helpers Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 17/38] KVM: x86/pmu: Add perf_capabilities field in struct kvm_host_values{} Mingwei Zhang
2025-05-15  0:12   ` Sean Christopherson
2025-05-15  3:04     ` Mi, Dapeng
2025-03-24 17:30 ` [PATCH v4 18/38] KVM: x86/pmu: Move PMU_CAP_{FW_WRITES,LBR_FMT} into msr-index.h header Mingwei Zhang
2025-03-24 17:30 ` [PATCH v4 19/38] KVM: VMX: Add macros to wrap around {secondary,tertiary}_exec_controls_changebit() Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 20/38] KVM: x86/pmu: Check if mediated vPMU can intercept rdpmc Mingwei Zhang
2025-05-15  0:19   ` Sean Christopherson
2025-05-15  3:23     ` Mi, Dapeng
2025-05-26  6:15   ` Sandipan Das
2025-07-09 15:53     ` Sean Christopherson
2025-07-29  3:29       ` Mi, Dapeng
2025-07-30  0:38         ` Sean Christopherson
2025-07-30  2:25           ` Mi, Dapeng
2025-08-01 23:32             ` Sean Christopherson
2025-08-05  0:54               ` Sean Christopherson
2025-08-06  0:50                 ` Sean Christopherson
2025-03-24 17:31 ` [PATCH v4 21/38] KVM: x86/pmu/vmx: Save/load guest IA32_PERF_GLOBAL_CTRL with vm_exit/entry_ctrl Mingwei Zhang
2025-03-26 16:51   ` Chen, Zide
2025-03-26 20:09     ` Mingwei Zhang
2025-05-15  0:33       ` Sean Christopherson
2025-05-15  3:45         ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 22/38] KVM: x86/pmu: Optimize intel/amd_pmu_refresh() helpers Mingwei Zhang
2025-05-15  0:37   ` Sean Christopherson
2025-05-15  5:09     ` Mi, Dapeng
2025-05-15 19:22       ` Sean Christopherson
2025-05-16  1:03         ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 23/38] KVM: x86/pmu: Configure the interception of PMU MSRs Mingwei Zhang
2025-05-15  0:41   ` Sean Christopherson
2025-05-15  5:37     ` Mi, Dapeng
2025-05-15 19:06       ` Sean Christopherson
2025-05-16 13:34   ` Sean Christopherson
2025-05-19  5:18     ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 24/38] KVM: x86/pmu: Exclude PMU MSRs in vmx_get_passthrough_msr_slot() Mingwei Zhang
2025-05-16 13:35   ` Sean Christopherson
2025-05-16 14:45     ` Sean Christopherson
2025-05-19  5:21       ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 25/38] KVM: x86/pmu: Add AMD PMU registers to direct access list Mingwei Zhang
2025-05-16 13:36   ` Sean Christopherson
2025-03-24 17:31 ` [PATCH v4 26/38] KVM: x86/pmu: Introduce eventsel_hw to prepare for pmu event filtering Mingwei Zhang
2025-05-15  0:42   ` Sean Christopherson
2025-05-15  5:34     ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 27/38] KVM: x86/pmu: Handle PMU MSRs interception and " Mingwei Zhang
2025-05-15  0:43   ` Sean Christopherson
2025-05-15  5:38     ` Mi, Dapeng
2025-05-16  1:26   ` Mi, Dapeng
2025-05-16 20:54     ` Sean Christopherson
2025-05-19  4:16       ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 28/38] KVM: x86/pmu/svm: Set GuestOnly bit and clear HostOnly bit when guest writes to event selectors Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 29/38] KVM: x86/pmu: Switch host/guest PMU context at vm-exit/vm-entry Mingwei Zhang
2025-05-15 16:29   ` Sean Christopherson
2025-05-16  2:37     ` Mi, Dapeng
2025-05-16 13:26   ` Sean Christopherson
2025-05-19  5:07     ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 30/38] KVM: x86/pmu: Handle emulated instruction for mediated vPMU Mingwei Zhang
2025-05-16  1:10   ` Sean Christopherson
2025-03-24 17:31 ` [PATCH v4 31/38] KVM: nVMX: Add macros to simplify nested MSR interception setting Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 32/38] KVM: nVMX: Add nested virtualization support for mediated PMU Mingwei Zhang
2025-05-16 13:33   ` Sean Christopherson
2025-05-19  5:24     ` Mi, Dapeng
2025-03-24 17:31 ` [PATCH v4 33/38] perf/x86/intel: Support PERF_PMU_CAP_MEDIATED_VPMU Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 34/38] perf/x86/amd: Support PERF_PMU_CAP_MEDIATED_VPMU for AMD host Mingwei Zhang
2025-05-21 20:00   ` Namhyung Kim
2025-03-24 17:31 ` [PATCH v4 35/38] KVM: x86/pmu: Expose enable_mediated_pmu parameter to user space Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 36/38] KVM: selftests: Add mediated vPMU supported for pmu tests Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 37/38] KVM: Selftests: Support mediated vPMU for vmx_pmu_caps_test Mingwei Zhang
2025-03-24 17:31 ` [PATCH v4 38/38] KVM: Selftests: Fix pmu_counters_test error for mediated vPMU Mingwei Zhang
2025-04-16  7:22 ` [PATCH v4 00/38] Mediated vPMU 4.0 for x86 Mi, Dapeng
2025-04-25 12:27   ` Peter Zijlstra
2025-05-06  9:57 ` Mi, Dapeng
2025-05-06 19:45   ` Sean Christopherson
2025-05-07  0:46     ` Mi, Dapeng
2025-05-15  0:49 ` Sean Christopherson
2025-05-15  5:45   ` Mi, Dapeng

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=20250324173121.1275209-2-mizhang@google.com \
    --to=mizhang@google.com \
    --cc=Liang@google.com \
    --cc=Manali.Shukla@amd.com \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=dapeng1.mi@linux.intel.com \
    --cc=eranian@google.com \
    --cc=hpa@zytor.com \
    --cc=irogers@google.com \
    --cc=jmattson@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=nikunj.dadhania@amd.com \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=sandipan.das@amd.com \
    --cc=seanjc@google.com \
    --cc=xiong.y.zhang@linux.intel.com \
    --cc=yongwei.ma@intel.com \
    --cc=zide.chen@intel.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

Powered by JetHome