[PATCH v7 bpf-next 1/3] perf: enable branch record for software events

[Song Liu]

The typical way to access branch record (e.g. Intel LBR) is via hardware
perf_event. For CPUs with FREEZE_LBRS_ON_PMI support, PMI could capture
reliable LBR. On the other hand, LBR could also be useful in non-PMI
scenario. For example, in kretprobe or bpf fexit program, LBR could
provide a lot of information on what happened with the function. Add API
to use branch record for software use.

Note that, when the software event triggers, it is necessary to stop the
branch record hardware asap. Therefore, static_call is used to remove some
branch instructions in this process.

Suggested-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Acked-by: John Fastabend <john.fastabend@xxxxxxxxx>
Acked-by: Andrii Nakryiko <andrii@xxxxxxxxxx>
Signed-off-by: Song Liu <songliubraving@xxxxxx>
---
 arch/x86/events/intel/core.c | 67 ++++++++++++++++++++++++++++++++----
 arch/x86/events/intel/ds.c   |  2 +-
 arch/x86/events/intel/lbr.c  | 20 +++--------
 arch/x86/events/perf_event.h | 19 ++++++++++
 include/linux/perf_event.h   | 23 +++++++++++++
 kernel/events/core.c         |  2 ++
 6 files changed, 111 insertions(+), 22 deletions(-)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 7011e87be6d03..1248fc1937f82 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -2143,19 +2143,19 @@ static __initconst const u64 knl_hw_cache_extra_regs
  * However, there are some cases which may change PEBS status, e.g. PMI
  * throttle. The PEBS_ENABLE should be updated where the status changes.
  */
-static void __intel_pmu_disable_all(void)
+static __always_inline void __intel_pmu_disable_all(bool bts)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
 
-	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask))
+	if (bts && test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask))
 		intel_pmu_disable_bts();
 }
 
-static void intel_pmu_disable_all(void)
+static __always_inline void intel_pmu_disable_all(void)
 {
-	__intel_pmu_disable_all();
+	__intel_pmu_disable_all(true);
 	intel_pmu_pebs_disable_all();
 	intel_pmu_lbr_disable_all();
 }
@@ -2186,6 +2186,49 @@ static void intel_pmu_enable_all(int added)
 	__intel_pmu_enable_all(added, false);
 }
 
+static noinline int
+__intel_pmu_snapshot_branch_stack(struct perf_branch_entry *entries,
+				  unsigned int cnt, unsigned long flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	intel_pmu_lbr_read();
+	cnt = min_t(unsigned int, cnt, x86_pmu.lbr_nr);
+
+	memcpy(entries, cpuc->lbr_entries, sizeof(struct perf_branch_entry) * cnt);
+	intel_pmu_enable_all(0);
+	local_irq_restore(flags);
+	return cnt;
+}
+
+static int
+intel_pmu_snapshot_branch_stack(struct perf_branch_entry *entries, unsigned int cnt)
+{
+	unsigned long flags;
+
+	/* must not have branches... */
+	local_irq_save(flags);
+	__intel_pmu_disable_all(false); /* we don't care about BTS */
+	__intel_pmu_pebs_disable_all();
+	__intel_pmu_lbr_disable();
+	/*            ... until here */
+	return __intel_pmu_snapshot_branch_stack(entries, cnt, flags);
+}
+
+static int
+intel_pmu_snapshot_arch_branch_stack(struct perf_branch_entry *entries, unsigned int cnt)
+{
+	unsigned long flags;
+
+	/* must not have branches... */
+	local_irq_save(flags);
+	__intel_pmu_disable_all(false); /* we don't care about BTS */
+	__intel_pmu_pebs_disable_all();
+	__intel_pmu_arch_lbr_disable();
+	/*            ... until here */
+	return __intel_pmu_snapshot_branch_stack(entries, cnt, flags);
+}
+
 /*
  * Workaround for:
  *   Intel Errata AAK100 (model 26)
@@ -2929,7 +2972,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs)
 		apic_write(APIC_LVTPC, APIC_DM_NMI);
 	intel_bts_disable_local();
 	cpuc->enabled = 0;
-	__intel_pmu_disable_all();
+	__intel_pmu_disable_all(true);
 	handled = intel_pmu_drain_bts_buffer();
 	handled += intel_bts_interrupt();
 	status = intel_pmu_get_status();
@@ -6283,9 +6326,21 @@ __init int intel_pmu_init(void)
 			x86_pmu.lbr_nr = 0;
 	}
 
-	if (x86_pmu.lbr_nr)
+	if (x86_pmu.lbr_nr) {
 		pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr);
 
+		/* only support branch_stack snapshot for perfmon >= v2 */
+		if (x86_pmu.disable_all == intel_pmu_disable_all) {
+			if (boot_cpu_has(X86_FEATURE_ARCH_LBR)) {
+				static_call_update(perf_snapshot_branch_stack,
+						   intel_pmu_snapshot_arch_branch_stack);
+			} else {
+				static_call_update(perf_snapshot_branch_stack,
+						   intel_pmu_snapshot_branch_stack);
+			}
+		}
+	}
+
 	intel_pmu_check_extra_regs(x86_pmu.extra_regs);
 
 	/* Support full width counters using alternative MSR range */
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index 8647713276a73..ac5991fea9ee3 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -1301,7 +1301,7 @@ void intel_pmu_pebs_disable_all(void)
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 
 	if (cpuc->pebs_enabled)
-		wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
+		__intel_pmu_pebs_disable_all();
 }
 
 static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs)
diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c
index 9e6d6eaeb4cb6..6b72e9b55c69c 100644
--- a/arch/x86/events/intel/lbr.c
+++ b/arch/x86/events/intel/lbr.c
@@ -228,20 +228,6 @@ static void __intel_pmu_lbr_enable(bool pmi)
 		wrmsrl(MSR_ARCH_LBR_CTL, lbr_select | ARCH_LBR_CTL_LBREN);
 }
 
-static void __intel_pmu_lbr_disable(void)
-{
-	u64 debugctl;
-
-	if (static_cpu_has(X86_FEATURE_ARCH_LBR)) {
-		wrmsrl(MSR_ARCH_LBR_CTL, 0);
-		return;
-	}
-
-	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
-	debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
-	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
-}
-
 void intel_pmu_lbr_reset_32(void)
 {
 	int i;
@@ -779,8 +765,12 @@ void intel_pmu_lbr_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 
-	if (cpuc->lbr_users && !vlbr_exclude_host())
+	if (cpuc->lbr_users && !vlbr_exclude_host()) {
+		if (static_cpu_has(X86_FEATURE_ARCH_LBR))
+			return __intel_pmu_arch_lbr_disable();
+
 		__intel_pmu_lbr_disable();
+	}
 }
 
 void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc)
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index e3ac05c97b5e5..0e3e596e33cdd 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -1240,6 +1240,25 @@ static inline bool intel_pmu_has_bts(struct perf_event *event)
 	return intel_pmu_has_bts_period(event, hwc->sample_period);
 }
 
+static __always_inline void __intel_pmu_pebs_disable_all(void)
+{
+	wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
+}
+
+static __always_inline void __intel_pmu_arch_lbr_disable(void)
+{
+	wrmsrl(MSR_ARCH_LBR_CTL, 0);
+}
+
+static __always_inline void __intel_pmu_lbr_disable(void)
+{
+	u64 debugctl;
+
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+	debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+}
+
 int intel_pmu_save_and_restart(struct perf_event *event);
 
 struct event_constraint *
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index fe156a8170aa3..0cbc5dfe11102 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -57,6 +57,7 @@ struct perf_guest_info_callbacks {
 #include <linux/cgroup.h>
 #include <linux/refcount.h>
 #include <linux/security.h>
+#include <linux/static_call.h>
 #include <asm/local.h>
 
 struct perf_callchain_entry {
@@ -1612,4 +1613,26 @@ extern void __weak arch_perf_update_userpage(struct perf_event *event,
 extern __weak u64 arch_perf_get_page_size(struct mm_struct *mm, unsigned long addr);
 #endif
 
+/*
+ * Snapshot branch stack on software events.
+ *
+ * Branch stack can be very useful in understanding software events. For
+ * example, when a long function, e.g. sys_perf_event_open, returns an
+ * errno, it is not obvious why the function failed. Branch stack could
+ * provide very helpful information in this type of scenarios.
+ *
+ * On software event, it is necessary to stop the hardware branch recorder
+ * fast. Otherwise, the hardware register/buffer will be flushed with
+ * entries of the triggering event. Therefore, static call is used to
+ * stop the hardware recorder.
+ */
+
+/*
+ * cnt is the number of entries allocated for entries.
+ * Return number of entries copied to .
+ */
+typedef int (perf_snapshot_branch_stack_t)(struct perf_branch_entry *entries,
+					   unsigned int cnt);
+DECLARE_STATIC_CALL(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t);
+
 #endif /* _LINUX_PERF_EVENT_H */
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 744e8726c5b2f..349f80aa9e7d8 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -13435,3 +13435,5 @@ struct cgroup_subsys perf_event_cgrp_subsys = {
 	.threaded	= true,
 };
 #endif /* CONFIG_CGROUP_PERF */
+
+DEFINE_STATIC_CALL_RET0(perf_snapshot_branch_stack, perf_snapshot_branch_stack_t);
-- 
2.30.2