perf, intel: Use GO/HO bits in perf-ctr

Intel does not have guest/host-only bit in perf counters like AMD
does.  To support GO/HO bits KVM needs to switch EVENTSELn values
(or PERF_GLOBAL_CTRL if available) at a guest entry. If a counter is
configured to count only in a guest mode it stays disabled in a host,
but VMX is configured to switch it to enabled value during guest entry.

This patch adds GO/HO tracking to Intel perf code and provides interface
for KVM to get a list of MSRs that need to be switched on a guest entry.

Only cpus with architectural PMU (v1 or later) are supported with this
patch.  To my knowledge there is not p6 models with VMX but without
architectural PMU and p4 with VMX are rare and the interface is general
enough to support them if need arise.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1317816084-18026-7-git-send-email-gleb@redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

3 files changed, 112 insertions, 3 deletions

diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h
index ce2bfb33538..e47cb6167e8 100644
--- a/arch/x86/include/asm/perf_event.h
+++ b/arch/x86/include/asm/perf_event.h
@@ -162,7 +162,19 @@ extern unsigned long perf_misc_flags(struct pt_regs *regs);
 	);							\
 }
 
+struct perf_guest_switch_msr {
+	unsigned msr;
+	u64 host, guest;
+};
+
+extern struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr);
 #else
+static inline perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+{
+	*nr = 0;
+	return NULL;
+}
+
 static inline void perf_events_lapic_init(void)	{ }
 #endif
 
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index fb330b0a816..b9698d40ac4 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -131,6 +131,13 @@ struct cpu_hw_events {
 	struct perf_branch_entry	lbr_entries[MAX_LBR_ENTRIES];
 
 	/*
+	 * Intel host/guest exclude bits
+	 */
+	u64				intel_ctrl_guest_mask;
+	u64				intel_ctrl_host_mask;
+	struct perf_guest_switch_msr	guest_switch_msrs[X86_PMC_IDX_MAX];
+
+	/*
 	 * manage shared (per-core, per-cpu) registers
 	 * used on Intel NHM/WSM/SNB
 	 */
@@ -295,6 +302,11 @@ struct x86_pmu {
 	 */
 	struct extra_reg *extra_regs;
 	unsigned int er_flags;
+
+	/*
+	 * Intel host/guest support (KVM)
+	 */
+	struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr);
 };
 
 #define ERF_NO_HT_SHARING	1
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 61fa35750b9..e09ca20e86e 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -749,7 +749,8 @@ static void intel_pmu_enable_all(int added)
 
 	intel_pmu_pebs_enable_all();
 	intel_pmu_lbr_enable_all();
-	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL,
+			x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask);
 
 	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
 		struct perf_event *event =
@@ -872,6 +873,7 @@ static void intel_pmu_disable_fixed(struct hw_perf_event *hwc)
 static void intel_pmu_disable_event(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
 	if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
 		intel_pmu_disable_bts();
@@ -879,6 +881,9 @@ static void intel_pmu_disable_event(struct perf_event *event)
 		return;
 	}
 
+	cpuc->intel_ctrl_guest_mask &= ~(1ull << hwc->idx);
+	cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx);
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_disable_fixed(hwc);
 		return;
@@ -924,6 +929,7 @@ static void intel_pmu_enable_fixed(struct hw_perf_event *hwc)
 static void intel_pmu_enable_event(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
 	if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
 		if (!__this_cpu_read(cpu_hw_events.enabled))
@@ -933,6 +939,11 @@ static void intel_pmu_enable_event(struct perf_event *event)
 		return;
 	}
 
+	if (event->attr.exclude_host)
+		cpuc->intel_ctrl_guest_mask |= (1ull << hwc->idx);
+	if (event->attr.exclude_guest)
+		cpuc->intel_ctrl_host_mask |= (1ull << hwc->idx);
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_enable_fixed(hwc);
 		return;
@@ -1302,12 +1313,84 @@ static int intel_pmu_hw_config(struct perf_event *event)
 	return 0;
 }
 
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+{
+	if (x86_pmu.guest_get_msrs)
+		return x86_pmu.guest_get_msrs(nr);
+	*nr = 0;
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(perf_guest_get_msrs);
+
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+
+	arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
+	arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
+	arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask;
+
+	*nr = 1;
+	return arr;
+}
+
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++)  {
+		struct perf_event *event = cpuc->events[idx];
+
+		arr[idx].msr = x86_pmu_config_addr(idx);
+		arr[idx].host = arr[idx].guest = 0;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		arr[idx].host = arr[idx].guest =
+			event->hw.config | ARCH_PERFMON_EVENTSEL_ENABLE;
+
+		if (event->attr.exclude_host)
+			arr[idx].host &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+		else if (event->attr.exclude_guest)
+			arr[idx].guest &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+	}
+
+	*nr = x86_pmu.num_counters;
+	return arr;
+}
+
+static void core_pmu_enable_event(struct perf_event *event)
+{
+	if (!event->attr.exclude_host)
+		x86_pmu_enable_event(event);
+}
+
+static void core_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+
+		if (!test_bit(idx, cpuc->active_mask) ||
+				cpuc->events[idx]->attr.exclude_host)
+			continue;
+
+		__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+	}
+}
+
 static __initconst const struct x86_pmu core_pmu = {
 	.name			= "core",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= x86_pmu_disable_all,
-	.enable_all		= x86_pmu_enable_all,
-	.enable			= x86_pmu_enable_event,
+	.enable_all		= core_pmu_enable_all,
+	.enable			= core_pmu_enable_event,
 	.disable		= x86_pmu_disable_event,
 	.hw_config		= x86_pmu_hw_config,
 	.schedule_events	= x86_schedule_events,
@@ -1325,6 +1408,7 @@ static __initconst const struct x86_pmu core_pmu = {
 	.get_event_constraints	= intel_get_event_constraints,
 	.put_event_constraints	= intel_put_event_constraints,
 	.event_constraints	= intel_core_event_constraints,
+	.guest_get_msrs		= core_guest_get_msrs,
 };
 
 struct intel_shared_regs *allocate_shared_regs(int cpu)
@@ -1431,6 +1515,7 @@ static __initconst const struct x86_pmu intel_pmu = {
 	.cpu_prepare		= intel_pmu_cpu_prepare,
 	.cpu_starting		= intel_pmu_cpu_starting,
 	.cpu_dying		= intel_pmu_cpu_dying,
+	.guest_get_msrs		= intel_guest_get_msrs,
 };
 
 static void intel_clovertown_quirks(void)