Merge tag 'perf-core-2024-05-13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf events updates from Ingo Molnar:

 - Combine perf and BPF for fast evalution of HW breakpoint
   conditions

 - Add LBR capture support outside of hardware events

 - Trigger IO signals for watermark_wakeup

 - Add RAPL support for Intel Arrow Lake and Lunar Lake

 - Optimize frequency-throttling

 - Miscellaneous cleanups & fixes

* tag 'perf-core-2024-05-13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
  perf/bpf: Mark perf_event_set_bpf_handler() and perf_event_free_bpf_handler() as inline too
  selftests/perf_events: Test FASYNC with watermark wakeups
  perf/ring_buffer: Trigger IO signals for watermark_wakeup
  perf: Move perf_event_fasync() to perf_event.h
  perf/bpf: Change the !CONFIG_BPF_SYSCALL stubs to static inlines
  selftest/bpf: Test a perf BPF program that suppresses side effects
  perf/bpf: Allow a BPF program to suppress all sample side effects
  perf/bpf: Remove unneeded uses_default_overflow_handler()
  perf/bpf: Call BPF handler directly, not through overflow machinery
  perf/bpf: Remove #ifdef CONFIG_BPF_SYSCALL from struct perf_event members
  perf/bpf: Create bpf_overflow_handler() stub for !CONFIG_BPF_SYSCALL
  perf/bpf: Reorder bpf_overflow_handler() ahead of __perf_event_overflow()
  perf/x86/rapl: Add support for Intel Lunar Lake
  perf/x86/rapl: Add support for Intel Arrow Lake
  perf/core: Reduce PMU access to adjust sample freq
  perf/core: Optimize perf_adjust_freq_unthr_context()
  perf/x86/amd: Don't reject non-sampling events with configured LBR
  perf/x86/amd: Support capturing LBR from software events
  perf/x86/amd: Avoid taking branches before disabling LBR
  perf/x86/amd: Ensure amd_pmu_core_disable_all() is always inlined
  ...

14 files changed, 525 insertions, 172 deletions

diff --git a/arch/arm/kernel/hw_breakpoint.c b/arch/arm/kernel/hw_breakpoint.c
index dc0fb7a81371..054e9199f30d 100644
--- a/arch/arm/kernel/hw_breakpoint.c
+++ b/arch/arm/kernel/hw_breakpoint.c
@@ -626,7 +626,7 @@ int hw_breakpoint_arch_parse(struct perf_event *bp,
 	hw->address &= ~alignment_mask;
 	hw->ctrl.len <<= offset;
 
-	if (uses_default_overflow_handler(bp)) {
+	if (is_default_overflow_handler(bp)) {
 		/*
 		 * Mismatch breakpoints are required for single-stepping
 		 * breakpoints.
@@ -798,7 +798,7 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,
 		 * Otherwise, insert a temporary mismatch breakpoint so that
 		 * we can single-step over the watchpoint trigger.
 		 */
-		if (!uses_default_overflow_handler(wp))
+		if (!is_default_overflow_handler(wp))
 			continue;
 step:
 		enable_single_step(wp, instruction_pointer(regs));
@@ -811,7 +811,7 @@ step:
 		info->trigger = addr;
 		pr_debug("watchpoint fired: address = 0x%x\n", info->trigger);
 		perf_bp_event(wp, regs);
-		if (uses_default_overflow_handler(wp))
+		if (is_default_overflow_handler(wp))
 			enable_single_step(wp, instruction_pointer(regs));
 	}
 
@@ -886,7 +886,7 @@ static void breakpoint_handler(unsigned long unknown, struct pt_regs *regs)
 			info->trigger = addr;
 			pr_debug("breakpoint fired: address = 0x%x\n", addr);
 			perf_bp_event(bp, regs);
-			if (uses_default_overflow_handler(bp))
+			if (is_default_overflow_handler(bp))
 				enable_single_step(bp, addr);
 			goto unlock;
 		}
diff --git a/arch/arm64/kernel/hw_breakpoint.c b/arch/arm64/kernel/hw_breakpoint.c
index 2f5755192c2b..722ac45f9f7b 100644
--- a/arch/arm64/kernel/hw_breakpoint.c
+++ b/arch/arm64/kernel/hw_breakpoint.c
@@ -655,7 +655,7 @@ static int breakpoint_handler(unsigned long unused, unsigned long esr,
 		perf_bp_event(bp, regs);
 
 		/* Do we need to handle the stepping? */
-		if (uses_default_overflow_handler(bp))
+		if (is_default_overflow_handler(bp))
 			step = 1;
 unlock:
 		rcu_read_unlock();
@@ -734,7 +734,7 @@ static u64 get_distance_from_watchpoint(unsigned long addr, u64 val,
 static int watchpoint_report(struct perf_event *wp, unsigned long addr,
 			     struct pt_regs *regs)
 {
-	int step = uses_default_overflow_handler(wp);
+	int step = is_default_overflow_handler(wp);
 	struct arch_hw_breakpoint *info = counter_arch_bp(wp);
 
 	info->trigger = addr;
diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c
index 985ef3b47919..1fc4ce44e743 100644
--- a/arch/x86/events/amd/core.c
+++ b/arch/x86/events/amd/core.c
@@ -647,7 +647,7 @@ static void amd_pmu_cpu_dead(int cpu)
 	}
 }
 
-static inline void amd_pmu_set_global_ctl(u64 ctl)
+static __always_inline void amd_pmu_set_global_ctl(u64 ctl)
 {
 	wrmsrl(MSR_AMD64_PERF_CNTR_GLOBAL_CTL, ctl);
 }
@@ -907,6 +907,37 @@ static int amd_pmu_handle_irq(struct pt_regs *regs)
 	return amd_pmu_adjust_nmi_window(handled);
 }
 
+/*
+ * AMD-specific callback invoked through perf_snapshot_branch_stack static
+ * call, defined in include/linux/perf_event.h. See its definition for API
+ * details. It's up to caller to provide enough space in *entries* to fit all
+ * LBR records, otherwise returned result will be truncated to *cnt* entries.
+ */
+static int amd_pmu_v2_snapshot_branch_stack(struct perf_branch_entry *entries, unsigned int cnt)
+{
+	struct cpu_hw_events *cpuc;
+	unsigned long flags;
+
+	/*
+	 * The sequence of steps to freeze LBR should be completely inlined
+	 * and contain no branches to minimize contamination of LBR snapshot
+	 */
+	local_irq_save(flags);
+	amd_pmu_core_disable_all();
+	__amd_pmu_lbr_disable();
+
+	cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	amd_pmu_lbr_read();
+	cnt = min(cnt, x86_pmu.lbr_nr);
+	memcpy(entries, cpuc->lbr_entries, sizeof(struct perf_branch_entry) * cnt);
+
+	amd_pmu_v2_enable_all(0);
+	local_irq_restore(flags);
+
+	return cnt;
+}
+
 static int amd_pmu_v2_handle_irq(struct pt_regs *regs)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
@@ -1443,6 +1474,10 @@ static int __init amd_core_pmu_init(void)
 		static_call_update(amd_pmu_branch_reset, amd_pmu_lbr_reset);
 		static_call_update(amd_pmu_branch_add, amd_pmu_lbr_add);
 		static_call_update(amd_pmu_branch_del, amd_pmu_lbr_del);
+
+		/* Only support branch_stack snapshot on perfmon v2 */
+		if (x86_pmu.handle_irq == amd_pmu_v2_handle_irq)
+			static_call_update(perf_snapshot_branch_stack, amd_pmu_v2_snapshot_branch_stack);
 	} else if (!amd_brs_init()) {
 		/*
 		 * BRS requires special event constraints and flushing on ctxsw.
diff --git a/arch/x86/events/amd/lbr.c b/arch/x86/events/amd/lbr.c
index 5149830c7c4f..19c7b76e21bc 100644
--- a/arch/x86/events/amd/lbr.c
+++ b/arch/x86/events/amd/lbr.c
@@ -310,10 +310,6 @@ int amd_pmu_lbr_hw_config(struct perf_event *event)
 {
 	int ret = 0;
 
-	/* LBR is not recommended in counting mode */
-	if (!is_sampling_event(event))
-		return -EINVAL;
-
 	ret = amd_pmu_lbr_setup_filter(event);
 	if (!ret)
 		event->attach_state |= PERF_ATTACH_SCHED_CB;
@@ -414,18 +410,11 @@ void amd_pmu_lbr_enable_all(void)
 void amd_pmu_lbr_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
-	u64 dbg_ctl, dbg_extn_cfg;
 
 	if (!cpuc->lbr_users || !x86_pmu.lbr_nr)
 		return;
 
-	rdmsrl(MSR_AMD_DBG_EXTN_CFG, dbg_extn_cfg);
-	wrmsrl(MSR_AMD_DBG_EXTN_CFG, dbg_extn_cfg & ~DBG_EXTN_CFG_LBRV2EN);
-
-	if (cpu_feature_enabled(X86_FEATURE_AMD_LBR_PMC_FREEZE)) {
-		rdmsrl(MSR_IA32_DEBUGCTLMSR, dbg_ctl);
-		wrmsrl(MSR_IA32_DEBUGCTLMSR, dbg_ctl & ~DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
-	}
+	__amd_pmu_lbr_disable();
 }
 
 __init int amd_pmu_lbr_init(void)
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index fb56518356ec..72b022a1e16c 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -1329,6 +1329,19 @@ void amd_pmu_lbr_enable_all(void);
 void amd_pmu_lbr_disable_all(void);
 int amd_pmu_lbr_hw_config(struct perf_event *event);
 
+static __always_inline void __amd_pmu_lbr_disable(void)
+{
+	u64 dbg_ctl, dbg_extn_cfg;
+
+	rdmsrl(MSR_AMD_DBG_EXTN_CFG, dbg_extn_cfg);
+	wrmsrl(MSR_AMD_DBG_EXTN_CFG, dbg_extn_cfg & ~DBG_EXTN_CFG_LBRV2EN);
+
+	if (cpu_feature_enabled(X86_FEATURE_AMD_LBR_PMC_FREEZE)) {
+		rdmsrl(MSR_IA32_DEBUGCTLMSR, dbg_ctl);
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, dbg_ctl & ~DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
+	}
+}
+
 #ifdef CONFIG_PERF_EVENTS_AMD_BRS
 
 #define AMD_FAM19H_BRS_EVENT 0xc4 /* RETIRED_TAKEN_BRANCH_INSTRUCTIONS */
diff --git a/arch/x86/events/rapl.c b/arch/x86/events/rapl.c
index fb2b1961e5a3..ca5f687fa420 100644
--- a/arch/x86/events/rapl.c
+++ b/arch/x86/events/rapl.c
@@ -675,10 +675,8 @@ static const struct attribute_group *rapl_attr_update[] = {
 static int __init init_rapl_pmus(void)
 {
 	int maxdie = topology_max_packages() * topology_max_dies_per_package();
-	size_t size;
 
-	size = sizeof(*rapl_pmus) + maxdie * sizeof(struct rapl_pmu *);
-	rapl_pmus = kzalloc(size, GFP_KERNEL);
+	rapl_pmus = kzalloc(struct_size(rapl_pmus, pmus, maxdie), GFP_KERNEL);
 	if (!rapl_pmus)
 		return -ENOMEM;
 
@@ -808,6 +806,9 @@ static const struct x86_cpu_id rapl_model_match[] __initconst = {
 	X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE_S,	&model_skl),
 	X86_MATCH_INTEL_FAM6_MODEL(METEORLAKE,		&model_skl),
 	X86_MATCH_INTEL_FAM6_MODEL(METEORLAKE_L,	&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(ARROWLAKE_H,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(ARROWLAKE,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(LUNARLAKE_M,		&model_skl),
 	{},
 };
 MODULE_DEVICE_TABLE(x86cpu, rapl_model_match);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index d2a15c0c6f8a..a5304ae8c654 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -809,11 +809,8 @@ struct perf_event {
 	u64				(*clock)(void);
 	perf_overflow_handler_t		overflow_handler;
 	void				*overflow_handler_context;
-#ifdef CONFIG_BPF_SYSCALL
-	perf_overflow_handler_t		orig_overflow_handler;
 	struct bpf_prog			*prog;
 	u64				bpf_cookie;
-#endif
 
 #ifdef CONFIG_EVENT_TRACING
 	struct trace_event_call		*tp_event;
@@ -883,6 +880,7 @@ struct perf_event_pmu_context {
 
 	unsigned int			nr_events;
 	unsigned int			nr_cgroups;
+	unsigned int			nr_freq;
 
 	atomic_t			refcount; /* event <-> epc */
 	struct rcu_head			rcu_head;
@@ -897,6 +895,11 @@ struct perf_event_pmu_context {
 	int				rotate_necessary;
 };
 
+static inline bool perf_pmu_ctx_is_active(struct perf_event_pmu_context *epc)
+{
+	return !list_empty(&epc->flexible_active) || !list_empty(&epc->pinned_active);
+}
+
 struct perf_event_groups {
 	struct rb_root	tree;
 	u64		index;
@@ -1342,8 +1345,10 @@ extern int perf_event_output(struct perf_event *event,
 			     struct pt_regs *regs);
 
 static inline bool
-__is_default_overflow_handler(perf_overflow_handler_t overflow_handler)
+is_default_overflow_handler(struct perf_event *event)
 {
+	perf_overflow_handler_t overflow_handler = event->overflow_handler;
+
 	if (likely(overflow_handler == perf_event_output_forward))
 		return true;
 	if (unlikely(overflow_handler == perf_event_output_backward))
@@ -1351,22 +1356,6 @@ __is_default_overflow_handler(perf_overflow_handler_t overflow_handler)
 	return false;
 }
 
-#define is_default_overflow_handler(event) \
-	__is_default_overflow_handler((event)->overflow_handler)
-
-#ifdef CONFIG_BPF_SYSCALL
-static inline bool uses_default_overflow_handler(struct perf_event *event)
-{
-	if (likely(is_default_overflow_handler(event)))
-		return true;
-
-	return __is_default_overflow_handler(event->orig_overflow_handler);
-}
-#else
-#define uses_default_overflow_handler(event) \
-	is_default_overflow_handler(event)
-#endif
-
 extern void
 perf_event_header__init_id(struct perf_event_header *header,
 			   struct perf_sample_data *data,
@@ -1697,6 +1686,14 @@ perf_event_addr_filters(struct perf_event *event)
 	return ifh;
 }
 
+static inline struct fasync_struct **perf_event_fasync(struct perf_event *event)
+{
+	/* Only the parent has fasync state */
+	if (event->parent)
+		event = event->parent;
+	return &event->fasync;
+}
+
 extern void perf_event_addr_filters_sync(struct perf_event *event);
 extern void perf_report_aux_output_id(struct perf_event *event, u64 hw_id);
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 724e6d7e128f..6b0a66ed2ae3 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -2302,8 +2302,10 @@ event_sched_out(struct perf_event *event, struct perf_event_context *ctx)
 
 	if (!is_software_event(event))
 		cpc->active_oncpu--;
-	if (event->attr.freq && event->attr.sample_freq)
+	if (event->attr.freq && event->attr.sample_freq) {
 		ctx->nr_freq--;
+		epc->nr_freq--;
+	}
 	if (event->attr.exclusive || !cpc->active_oncpu)
 		cpc->exclusive = 0;
 
@@ -2558,9 +2560,10 @@ event_sched_in(struct perf_event *event, struct perf_event_context *ctx)
 
 	if (!is_software_event(event))
 		cpc->active_oncpu++;
-	if (event->attr.freq && event->attr.sample_freq)
+	if (event->attr.freq && event->attr.sample_freq) {
 		ctx->nr_freq++;
-
+		epc->nr_freq++;
+	}
 	if (event->attr.exclusive)
 		cpc->exclusive = 1;
 
@@ -4123,30 +4126,14 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count, bo
 	}
 }
 
-/*
- * combine freq adjustment with unthrottling to avoid two passes over the
- * events. At the same time, make sure, having freq events does not change
- * the rate of unthrottling as that would introduce bias.
- */
-static void
-perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
+static void perf_adjust_freq_unthr_events(struct list_head *event_list)
 {
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
 	u64 now, period = TICK_NSEC;
 	s64 delta;
 
-	/*
-	 * only need to iterate over all events iff:
-	 * - context have events in frequency mode (needs freq adjust)
-	 * - there are events to unthrottle on this cpu
-	 */
-	if (!(ctx->nr_freq || unthrottle))
-		return;
-
-	raw_spin_lock(&ctx->lock);
-
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+	list_for_each_entry(event, event_list, active_list) {
 		if (event->state != PERF_EVENT_STATE_ACTIVE)
 			continue;
 
@@ -4154,18 +4141,17 @@ perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
 		if (!event_filter_match(event))
 			continue;
 
-		perf_pmu_disable(event->pmu);
-
 		hwc = &event->hw;
 
 		if (hwc->interrupts == MAX_INTERRUPTS) {
 			hwc->interrupts = 0;
 			perf_log_throttle(event, 1);
-			event->pmu->start(event, 0);
+			if (!event->attr.freq || !event->attr.sample_freq)
+				event->pmu->start(event, 0);
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
-			goto next;
+			continue;
 
 		/*
 		 * stop the event and update event->count
@@ -4187,8 +4173,41 @@ perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
 			perf_adjust_period(event, period, delta, false);
 
 		event->pmu->start(event, delta > 0 ? PERF_EF_RELOAD : 0);
-	next:
-		perf_pmu_enable(event->pmu);
+	}
+}
+
+/*
+ * combine freq adjustment with unthrottling to avoid two passes over the
+ * events. At the same time, make sure, having freq events does not change
+ * the rate of unthrottling as that would introduce bias.
+ */
+static void
+perf_adjust_freq_unthr_context(struct perf_event_context *ctx, bool unthrottle)
+{
+	struct perf_event_pmu_context *pmu_ctx;
+
+	/*
+	 * only need to iterate over all events iff:
+	 * - context have events in frequency mode (needs freq adjust)
+	 * - there are events to unthrottle on this cpu
+	 */
+	if (!(ctx->nr_freq || unthrottle))
+		return;
+
+	raw_spin_lock(&ctx->lock);
+
+	list_for_each_entry(pmu_ctx, &ctx->pmu_ctx_list, pmu_ctx_entry) {
+		if (!(pmu_ctx->nr_freq || unthrottle))
+			continue;
+		if (!perf_pmu_ctx_is_active(pmu_ctx))
+			continue;
+		if (pmu_ctx->pmu->capabilities & PERF_PMU_CAP_NO_INTERRUPT)
+			continue;
+
+		perf_pmu_disable(pmu_ctx->pmu);
+		perf_adjust_freq_unthr_events(&pmu_ctx->pinned_active);
+		perf_adjust_freq_unthr_events(&pmu_ctx->flexible_active);
+		perf_pmu_enable(pmu_ctx->pmu);
 	}
 
 	raw_spin_unlock(&ctx->lock);
@@ -6684,14 +6703,6 @@ static const struct file_operations perf_fops = {
  * to user-space before waking everybody up.
  */
 
-static inline struct fasync_struct **perf_event_fasync(struct perf_event *event)
-{
-	/* only the parent has fasync state */
-	if (event->parent)
-		event = event->parent;
-	return &event->fasync;
-}
-
 void perf_event_wakeup(struct perf_event *event)
 {
 	ring_buffer_wakeup(event);
@@ -9544,6 +9555,100 @@ static inline bool sample_is_allowed(struct perf_event *event, struct pt_regs *r
 	return true;
 }
 
+#ifdef CONFIG_BPF_SYSCALL
+static int bpf_overflow_handler(struct perf_event *event,
+				struct perf_sample_data *data,
+				struct pt_regs *regs)
+{
+	struct bpf_perf_event_data_kern ctx = {
+		.data = data,
+		.event = event,
+	};
+	struct bpf_prog *prog;
+	int ret = 0;
+
+	ctx.regs = perf_arch_bpf_user_pt_regs(regs);
+	if (unlikely(__this_cpu_inc_return(bpf_prog_active) != 1))
+		goto out;
+	rcu_read_lock();
+	prog = READ_ONCE(event->prog);
+	if (prog) {
+		perf_prepare_sample(data, event, regs);
+		ret = bpf_prog_run(prog, &ctx);
+	}
+	rcu_read_unlock();
+out:
+	__this_cpu_dec(bpf_prog_active);
+
+	return ret;
+}
+
+static inline int perf_event_set_bpf_handler(struct perf_event *event,
+					     struct bpf_prog *prog,
+					     u64 bpf_cookie)
+{
+	if (event->overflow_handler_context)
+		/* hw breakpoint or kernel counter */
+		return -EINVAL;
+
+	if (event->prog)
+		return -EEXIST;
+
+	if (prog->type != BPF_PROG_TYPE_PERF_EVENT)
+		return -EINVAL;
+
+	if (event->attr.precise_ip &&
+	    prog->call_get_stack &&
+	    (!(event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) ||
+	     event->attr.exclude_callchain_kernel ||
+	     event->attr.exclude_callchain_user)) {
+		/*
+		 * On perf_event with precise_ip, calling bpf_get_stack()
+		 * may trigger unwinder warnings and occasional crashes.
+		 * bpf_get_[stack|stackid] works around this issue by using
+		 * callchain attached to perf_sample_data. If the
+		 * perf_event does not full (kernel and user) callchain
+		 * attached to perf_sample_data, do not allow attaching BPF
+		 * program that calls bpf_get_[stack|stackid].
+		 */
+		return -EPROTO;
+	}
+
+	event->prog = prog;
+	event->bpf_cookie = bpf_cookie;
+	return 0;
+}
+
+static inline void perf_event_free_bpf_handler(struct perf_event *event)
+{
+	struct bpf_prog *prog = event->prog;
+
+	if (!prog)
+		return;
+
+	event->prog = NULL;
+	bpf_prog_put(prog);
+}
+#else
+static inline int bpf_overflow_handler(struct perf_event *event,
+				       struct perf_sample_data *data,
+				       struct pt_regs *regs)
+{
+	return 1;
+}
+
+static inline int perf_event_set_bpf_handler(struct perf_event *event,
+					     struct bpf_prog *prog,
+					     u64 bpf_cookie)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void perf_event_free_bpf_handler(struct perf_event *event)
+{
+}
+#endif
+
 /*
  * Generic event overflow handling, sampling.
  */
@@ -9564,6 +9669,9 @@ static int __perf_event_overflow(struct perf_event *event,
 
 	ret = __perf_event_account_interrupt(event, throttle);
 
+	if (event->prog && !bpf_overflow_handler(event, data, regs))
+		return ret;
+
 	/*
 	 * XXX event_limit might not quite work as expected on inherited
 	 * events
@@ -10422,97 +10530,6 @@ static void perf_event_free_filter(struct perf_event *event)
 	ftrace_profile_free_filter(event);
 }
 
-#ifdef CONFIG_BPF_SYSCALL
-static void bpf_overflow_handler(struct perf_event *event,
-				 struct perf_sample_data *data,
-				 struct pt_regs *regs)
-{
-	struct bpf_perf_event_data_kern ctx = {
-		.data = data,
-		.event = event,
-	};
-	struct bpf_prog *prog;
-	int ret = 0;
-
-	ctx.regs = perf_arch_bpf_user_pt_regs(regs);
-	if (unlikely(__this_cpu_inc_return(bpf_prog_active) != 1))
-		goto out;
-	rcu_read_lock();
-	prog = READ_ONCE(event->prog);
-	if (prog) {
-		perf_prepare_sample(data, event, regs);
-		ret = bpf_prog_run(prog, &ctx);
-	}
-	rcu_read_unlock();
-out:
-	__this_cpu_dec(bpf_prog_active);
-	if (!ret)
-		return;
-
-	event->orig_overflow_handler(event, data, regs);
-}
-
-static int perf_event_set_bpf_handler(struct perf_event *event,
-				      struct bpf_prog *prog,
-				      u64 bpf_cookie)
-{
-	if (event->overflow_handler_context)
-		/* hw breakpoint or kernel counter */
-		return -EINVAL;
-
-	if (event->prog)
-		return -EEXIST;
-
-	if (prog->type != BPF_PROG_TYPE_PERF_EVENT)
-		return -EINVAL;
-
-	if (event->attr.precise_ip &&
-	    prog->call_get_stack &&
-	    (!(event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) ||
-	     event->attr.exclude_callchain_kernel ||
-	     event->attr.exclude_callchain_user)) {
-		/*
-		 * On perf_event with precise_ip, calling bpf_get_stack()
-		 * may trigger unwinder warnings and occasional crashes.
-		 * bpf_get_[stack|stackid] works around this issue by using
-		 * callchain attached to perf_sample_data. If the
-		 * perf_event does not full (kernel and user) callchain
-		 * attached to perf_sample_data, do not allow attaching BPF
-		 * program that calls bpf_get_[stack|stackid].
-		 */
-		return -EPROTO;
-	}
-
-	event->prog = prog;
-	event->bpf_cookie = bpf_cookie;
-	event->orig_overflow_handler = READ_ONCE(event->overflow_handler);
-	WRITE_ONCE(event->overflow_handler, bpf_overflow_handler);
-	return 0;
-}
-
-static void perf_event_free_bpf_handler(struct perf_event *event)
-{
-	struct bpf_prog *prog = event->prog;
-
-	if (!prog)
-		return;
-
-	WRITE_ONCE(event->overflow_handler, event->orig_overflow_handler);
-	event->prog = NULL;
-	bpf_prog_put(prog);
-}
-#else
-static int perf_event_set_bpf_handler(struct perf_event *event,
-				      struct bpf_prog *prog,
-				      u64 bpf_cookie)
-{
-	return -EOPNOTSUPP;
-}
-static void perf_event_free_bpf_handler(struct perf_event *event)
-{
-}
-#endif
-
 /*
  * returns true if the event is a tracepoint, or a kprobe/upprobe created
  * with perf_event_open()
@@ -11971,13 +11988,11 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 		overflow_handler = parent_event->overflow_handler;
 		context = parent_event->overflow_handler_context;
 #if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_EVENT_TRACING)
-		if (overflow_handler == bpf_overflow_handler) {
+		if (parent_event->prog) {
 			struct bpf_prog *prog = parent_event->prog;
 
 			bpf_prog_inc(prog);
 			event->prog = prog;
-			event->orig_overflow_handler =
-				parent_event->orig_overflow_handler;
 		}
 #endif
 	}
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index 60ed43d1c29e..4013408ce012 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -22,6 +22,10 @@ static void perf_output_wakeup(struct perf_output_handle *handle)
 	atomic_set(&handle->rb->poll, EPOLLIN);
 
 	handle->event->pending_wakeup = 1;
+
+	if (*perf_event_fasync(handle->event) && !handle->event->pending_kill)
+		handle->event->pending_kill = POLL_IN;
+
 	irq_work_queue(&handle->event->pending_irq);
 }
 
diff --git a/tools/testing/selftests/bpf/prog_tests/perf_skip.c b/tools/testing/selftests/bpf/prog_tests/perf_skip.c
new file mode 100644
index 000000000000..37d8618800e4
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/perf_skip.c
@@ -0,0 +1,137 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+
+#include <test_progs.h>
+#include "test_perf_skip.skel.h"
+#include <linux/compiler.h>
+#include <linux/hw_breakpoint.h>
+#include <sys/mman.h>
+
+#ifndef TRAP_PERF
+#define TRAP_PERF 6
+#endif
+
+int sigio_count, sigtrap_count;
+
+static void handle_sigio(int sig __always_unused)
+{
+	++sigio_count;
+}
+
+static void handle_sigtrap(int signum __always_unused,
+			   siginfo_t *info,
+			   void *ucontext __always_unused)
+{
+	ASSERT_EQ(info->si_code, TRAP_PERF, "si_code");
+	++sigtrap_count;
+}
+
+static noinline int test_function(void)
+{
+	asm volatile ("");
+	return 0;
+}
+
+void serial_test_perf_skip(void)
+{
+	struct sigaction action = {};
+	struct sigaction previous_sigtrap;
+	sighandler_t previous_sigio = SIG_ERR;
+	struct test_perf_skip *skel = NULL;
+	struct perf_event_attr attr = {};
+	int perf_fd = -1;
+	int err;
+	struct f_owner_ex owner;
+	struct bpf_link *prog_link = NULL;
+
+	action.sa_flags = SA_SIGINFO | SA_NODEFER;
+	action.sa_sigaction = handle_sigtrap;
+	sigemptyset(&action.sa_mask);
+	if (!ASSERT_OK(sigaction(SIGTRAP, &action, &previous_sigtrap), "sigaction"))
+		return;
+
+	previous_sigio = signal(SIGIO, handle_sigio);
+	if (!ASSERT_NEQ(previous_sigio, SIG_ERR, "signal"))
+		goto cleanup;
+
+	skel = test_perf_skip__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel_load"))
+		goto cleanup;
+
+	attr.type = PERF_TYPE_BREAKPOINT;
+	attr.size = sizeof(attr);
+	attr.bp_type = HW_BREAKPOINT_X;
+	attr.bp_addr = (uintptr_t)test_function;
+	attr.bp_len = sizeof(long);
+	attr.sample_period = 1;
+	attr.sample_type = PERF_SAMPLE_IP;
+	attr.pinned = 1;
+	attr.exclude_kernel = 1;
+	attr.exclude_hv = 1;
+	attr.precise_ip = 3;
+	attr.sigtrap = 1;
+	attr.remove_on_exec = 1;
+
+	perf_fd = syscall(__NR_perf_event_open, &attr, 0, -1, -1, 0);
+	if (perf_fd < 0 && (errno == ENOENT || errno == EOPNOTSUPP)) {
+		printf("SKIP:no PERF_TYPE_BREAKPOINT/HW_BREAKPOINT_X\n");
+		test__skip();
+		goto cleanup;
+	}
+	if (!ASSERT_OK(perf_fd < 0, "perf_event_open"))
+		goto cleanup;
+
+	/* Configure the perf event to signal on sample. */
+	err = fcntl(perf_fd, F_SETFL, O_ASYNC);
+	if (!ASSERT_OK(err, "fcntl(F_SETFL, O_ASYNC)"))
+		goto cleanup;
+
+	owner.type = F_OWNER_TID;
+	owner.pid = syscall(__NR_gettid);
+	err = fcntl(perf_fd, F_SETOWN_EX, &owner);
+	if (!ASSERT_OK(err, "fcntl(F_SETOWN_EX)"))
+		goto cleanup;
+
+	/* Allow at most one sample. A sample rejected by bpf should
+	 * not count against this.
+	 */
+	err = ioctl(perf_fd, PERF_EVENT_IOC_REFRESH, 1);
+	if (!ASSERT_OK(err, "ioctl(PERF_EVENT_IOC_REFRESH)"))
+		goto cleanup;
+
+	prog_link = bpf_program__attach_perf_event(skel->progs.handler, perf_fd);
+	if (!ASSERT_OK_PTR(prog_link, "bpf_program__attach_perf_event"))
+		goto cleanup;
+
+	/* Configure the bpf program to suppress the sample. */
+	skel->bss->ip = (uintptr_t)test_function;
+	test_function();
+
+	ASSERT_EQ(sigio_count, 0, "sigio_count");
+	ASSERT_EQ(sigtrap_count, 0, "sigtrap_count");
+
+	/* Configure the bpf program to allow the sample. */
+	skel->bss->ip = 0;
+	test_function();
+
+	ASSERT_EQ(sigio_count, 1, "sigio_count");
+	ASSERT_EQ(sigtrap_count, 1, "sigtrap_count");
+
+	/* Test that the sample above is the only one allowed (by perf, not
+	 * by bpf)
+	 */
+	test_function();
+
+	ASSERT_EQ(sigio_count, 1, "sigio_count");
+	ASSERT_EQ(sigtrap_count, 1, "sigtrap_count");
+
+cleanup:
+	bpf_link__destroy(prog_link);
+	if (perf_fd >= 0)
+		close(perf_fd);
+	test_perf_skip__destroy(skel);
+
+	if (previous_sigio != SIG_ERR)
+		signal(SIGIO, previous_sigio);
+	sigaction(SIGTRAP, &previous_sigtrap, NULL);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_perf_skip.c b/tools/testing/selftests/bpf/progs/test_perf_skip.c
new file mode 100644
index 000000000000..7eb8b6de7a57
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_perf_skip.c
@@ -0,0 +1,15 @@
+// SPDX-License-Identifier: GPL-2.0
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+uintptr_t ip;
+
+SEC("perf_event")
+int handler(struct bpf_perf_event_data *data)
+{
+	/* Skip events that have the correct ip. */
+	return ip != PT_REGS_IP(&data->regs);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/perf_events/.gitignore b/tools/testing/selftests/perf_events/.gitignore
index 790c47001e77..ee93dc4969b8 100644
--- a/tools/testing/selftests/perf_events/.gitignore
+++ b/tools/testing/selftests/perf_events/.gitignore
@@ -1,3 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0-only
 sigtrap_threads
 remove_on_exec
+watermark_signal
diff --git a/tools/testing/selftests/perf_events/Makefile b/tools/testing/selftests/perf_events/Makefile
index db93c4ff081a..70e3ff211278 100644
--- a/tools/testing/selftests/perf_events/Makefile
+++ b/tools/testing/selftests/perf_events/Makefile
@@ -2,5 +2,5 @@
 CFLAGS += -Wl,-no-as-needed -Wall $(KHDR_INCLUDES)
 LDFLAGS += -lpthread
 
-TEST_GEN_PROGS := sigtrap_threads remove_on_exec
+TEST_GEN_PROGS := sigtrap_threads remove_on_exec watermark_signal
 include ../lib.mk
diff --git a/tools/testing/selftests/perf_events/watermark_signal.c b/tools/testing/selftests/perf_events/watermark_signal.c
new file mode 100644
index 000000000000..49dc1e831174
--- /dev/null
+++ b/tools/testing/selftests/perf_events/watermark_signal.c
@@ -0,0 +1,146 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/perf_event.h>
+#include <stddef.h>
+#include <sched.h>
+#include <signal.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+#include <sys/syscall.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#include "../kselftest_harness.h"
+
+#define __maybe_unused __attribute__((__unused__))
+
+static int sigio_count;
+
+static void handle_sigio(int signum __maybe_unused,
+			 siginfo_t *oh __maybe_unused,
+			 void *uc __maybe_unused)
+{
+	++sigio_count;
+}
+
+static void do_child(void)
+{
+	raise(SIGSTOP);
+
+	for (int i = 0; i < 20; ++i)
+		sleep(1);
+
+	raise(SIGSTOP);
+
+	exit(0);
+}
+
+TEST(watermark_signal)
+{
+	struct perf_event_attr attr;
+	struct perf_event_mmap_page *p = NULL;
+	struct sigaction previous_sigio, sigio = { 0 };
+	pid_t child = -1;
+	int child_status;
+	int fd = -1;
+	long page_size = sysconf(_SC_PAGE_SIZE);
+
+	sigio.sa_sigaction = handle_sigio;
+	EXPECT_EQ(sigaction(SIGIO, &sigio, &previous_sigio), 0);
+
+	memset(&attr, 0, sizeof(attr));
+	attr.size = sizeof(attr);
+	attr.type = PERF_TYPE_SOFTWARE;
+	attr.config = PERF_COUNT_SW_DUMMY;
+	attr.sample_period = 1;
+	attr.disabled = 1;
+	attr.watermark = 1;
+	attr.context_switch = 1;
+	attr.wakeup_watermark = 1;
+
+	child = fork();
+	EXPECT_GE(child, 0);
+	if (child == 0)
+		do_child();
+	else if (child < 0) {
+		perror("fork()");
+		goto cleanup;
+	}
+
+	if (waitpid(child, &child_status, WSTOPPED) != child ||
+	    !(WIFSTOPPED(child_status) && WSTOPSIG(child_status) == SIGSTOP)) {
+		fprintf(stderr,
+			"failed to sycnhronize with child errno=%d status=%x\n",
+			errno,
+			child_status);
+		goto cleanup;
+	}
+
+	fd = syscall(__NR_perf_event_open, &attr, child, -1, -1,
+		     PERF_FLAG_FD_CLOEXEC);
+	if (fd < 0) {
+		fprintf(stderr, "failed opening event %llx\n", attr.config);
+		goto cleanup;
+	}
+
+	if (fcntl(fd, F_SETFL, FASYNC)) {
+		perror("F_SETFL FASYNC");
+		goto cleanup;
+	}
+
+	if (fcntl(fd, F_SETOWN, getpid())) {
+		perror("F_SETOWN getpid()");
+		goto cleanup;
+	}
+
+	if (fcntl(fd, F_SETSIG, SIGIO)) {
+		perror("F_SETSIG SIGIO");
+		goto cleanup;
+	}
+
+	p = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
+	if (p == NULL) {
+		perror("mmap");
+		goto cleanup;
+	}
+
+	if (ioctl(fd, PERF_EVENT_IOC_ENABLE, 0)) {
+		perror("PERF_EVENT_IOC_ENABLE");
+		goto cleanup;
+	}
+
+	if (kill(child, SIGCONT) < 0) {
+		perror("SIGCONT");
+		goto cleanup;
+	}
+
+	if (waitpid(child, &child_status, WSTOPPED) != -1 || errno != EINTR)
+		fprintf(stderr,
+			"expected SIGIO to terminate wait errno=%d status=%x\n%d",
+			errno,
+			child_status,
+			sigio_count);
+
+	EXPECT_GE(sigio_count, 1);
+
+cleanup:
+	if (p != NULL)
+		munmap(p, 2 * page_size);
+
+	if (fd >= 0)
+		close(fd);
+
+	if (child > 0) {
+		kill(child, SIGKILL);
+		waitpid(child, NULL, 0);
+	}
+
+	sigaction(SIGIO, &previous_sigio, NULL);
+}
+
+TEST_HARNESS_MAIN