Merge branch 'tip/perf/core' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace into perf/core

include/linux/ftrace.h

···
 
 typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
 
+/*
+ * FTRACE_OPS_FL_* bits denote the state of ftrace_ops struct and are
+ * set in the flags member.
+ *
+ * ENABLED - set/unset when ftrace_ops is registered/unregistered
+ * GLOBAL  - set manualy by ftrace_ops user to denote the ftrace_ops
+ *           is part of the global tracers sharing the same filter
+ *           via set_ftrace_* debugfs files.
+ * DYNAMIC - set when ftrace_ops is registered to denote dynamically
+ *           allocated ftrace_ops which need special care
+ * CONTROL - set manualy by ftrace_ops user to denote the ftrace_ops
+ *           could be controled by following calls:
+ *             ftrace_function_local_enable
+ *             ftrace_function_local_disable
+ */
 enum {
 	FTRACE_OPS_FL_ENABLED		= 1 << 0,
 	FTRACE_OPS_FL_GLOBAL		= 1 << 1,
 	FTRACE_OPS_FL_DYNAMIC		= 1 << 2,
+	FTRACE_OPS_FL_CONTROL		= 1 << 3,
 };
 
 struct ftrace_ops {
 	ftrace_func_t			func;
 	struct ftrace_ops		*next;
 	unsigned long			flags;
+	int __percpu			*disabled;
 #ifdef CONFIG_DYNAMIC_FTRACE
 	struct ftrace_hash		*notrace_hash;
 	struct ftrace_hash		*filter_hash;
···
 int register_ftrace_function(struct ftrace_ops *ops);
 int unregister_ftrace_function(struct ftrace_ops *ops);
 void clear_ftrace_function(void);
+
+/**
+ * ftrace_function_local_enable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_enable(struct ftrace_ops *ops)
+{
+	if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+		return;
+
+	(*this_cpu_ptr(ops->disabled))--;
+}
+
+/**
+ * ftrace_function_local_disable - enable controlled ftrace_ops on current cpu
+ *
+ * This function enables tracing on current cpu by decreasing
+ * the per cpu control variable.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline void ftrace_function_local_disable(struct ftrace_ops *ops)
+{
+	if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL)))
+		return;
+
+	(*this_cpu_ptr(ops->disabled))++;
+}
+
+/**
+ * ftrace_function_local_disabled - returns ftrace_ops disabled value
+ *                                  on current cpu
+ *
+ * This function returns value of ftrace_ops::disabled on current cpu.
+ * It must be called with preemption disabled and only on ftrace_ops
+ * registered with FTRACE_OPS_FL_CONTROL. If called without preemption
+ * disabled, this_cpu_ptr will complain when CONFIG_DEBUG_PREEMPT is enabled.
+ */
+static inline int ftrace_function_local_disabled(struct ftrace_ops *ops)
+{
+	WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_CONTROL));
+	return *this_cpu_ptr(ops->disabled);
+}
 
 extern void ftrace_stub(unsigned long a0, unsigned long a1);
 
···
 			int len, int reset);
 void ftrace_set_global_filter(unsigned char *buf, int len, int reset);
 void ftrace_set_global_notrace(unsigned char *buf, int len, int reset);
+void ftrace_free_filter(struct ftrace_ops *ops);
 
 int register_ftrace_command(struct ftrace_func_command *cmd);
 int unregister_ftrace_command(struct ftrace_func_command *cmd);
···
 #else
 static inline int skip_trace(unsigned long ip) { return 0; }
 static inline int ftrace_force_update(void) { return 0; }
-static inline void ftrace_set_filter(unsigned char *buf, int len, int reset)
-{
-}
 static inline void ftrace_disable_daemon(void) { }
 static inline void ftrace_enable_daemon(void) { }
 static inline void ftrace_release_mod(struct module *mod) {}
···
  */
 #define ftrace_regex_open(ops, flag, inod, file) ({ -ENODEV; })
 #define ftrace_set_early_filter(ops, buf, enable) do { } while (0)
+#define ftrace_set_filter(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_set_notrace(ops, buf, len, reset) ({ -ENODEV; })
+#define ftrace_free_filter(ops) do { } while (0)
 
 static inline ssize_t ftrace_filter_write(struct file *file, const char __user *ubuf,
 			    size_t cnt, loff_t *ppos) { return -ENODEV; }

include/linux/ftrace_event.h

···
 	TRACE_REG_UNREGISTER,
 	TRACE_REG_PERF_REGISTER,
 	TRACE_REG_PERF_UNREGISTER,
+	TRACE_REG_PERF_OPEN,
+	TRACE_REG_PERF_CLOSE,
+	TRACE_REG_PERF_ADD,
+	TRACE_REG_PERF_DEL,
 };
 
 struct ftrace_event_call;
···
 	void			*perf_probe;
 #endif
 	int			(*reg)(struct ftrace_event_call *event,
-				       enum trace_reg type);
+				       enum trace_reg type, void *data);
 	int			(*define_fields)(struct ftrace_event_call *);
 	struct list_head	*(*get_fields)(struct ftrace_event_call *);
 	struct list_head	fields;
···
 };
 
 extern int ftrace_event_reg(struct ftrace_event_call *event,
-			    enum trace_reg type);
+			    enum trace_reg type, void *data);
 
 enum {
 	TRACE_EVENT_FL_ENABLED_BIT,
···
 	FILTER_STATIC_STRING,
 	FILTER_DYN_STRING,
 	FILTER_PTR_STRING,
+	FILTER_TRACE_FN,
 };
 
 #define EVENT_STORAGE_SIZE 128

include/linux/perf_event.h

···
 #ifdef CONFIG_EVENT_TRACING
 	struct ftrace_event_call	*tp_event;
 	struct event_filter		*filter;
+#ifdef CONFIG_FUNCTION_TRACER
+	struct ftrace_ops               ftrace_ops;
+#endif
 #endif
 
 #ifdef CONFIG_CGROUP_PERF

kernel/trace/ftrace.c

···
 #define FTRACE_HASH_DEFAULT_BITS 10
 #define FTRACE_HASH_MAX_BITS 12
 
+#define FL_GLOBAL_CONTROL_MASK (FTRACE_OPS_FL_GLOBAL | FTRACE_OPS_FL_CONTROL)
+
 /* ftrace_enabled is a method to turn ftrace on or off */
 int ftrace_enabled __read_mostly;
 static int last_ftrace_enabled;
···
 };
 
 static struct ftrace_ops *ftrace_global_list __read_mostly = &ftrace_list_end;
+static struct ftrace_ops *ftrace_control_list __read_mostly = &ftrace_list_end;
 static struct ftrace_ops *ftrace_ops_list __read_mostly = &ftrace_list_end;
 ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub;
 static ftrace_func_t __ftrace_trace_function_delay __read_mostly = ftrace_stub;
 ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub;
 ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub;
 static struct ftrace_ops global_ops;
+static struct ftrace_ops control_ops;
 
 static void
 ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip);
···
 	__ftrace_trace_function(ip, parent_ip);
 }
 #endif
+
+static void control_ops_disable_all(struct ftrace_ops *ops)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		*per_cpu_ptr(ops->disabled, cpu) = 1;
+}
+
+static int control_ops_alloc(struct ftrace_ops *ops)
+{
+	int __percpu *disabled;
+
+	disabled = alloc_percpu(int);
+	if (!disabled)
+		return -ENOMEM;
+
+	ops->disabled = disabled;
+	control_ops_disable_all(ops);
+	return 0;
+}
+
+static void control_ops_free(struct ftrace_ops *ops)
+{
+	free_percpu(ops->disabled);
+}
 
 static void update_global_ops(void)
 {
···
 	return 0;
 }
 
+static void add_ftrace_list_ops(struct ftrace_ops **list,
+				struct ftrace_ops *main_ops,
+				struct ftrace_ops *ops)
+{
+	int first = *list == &ftrace_list_end;
+	add_ftrace_ops(list, ops);
+	if (first)
+		add_ftrace_ops(&ftrace_ops_list, main_ops);
+}
+
+static int remove_ftrace_list_ops(struct ftrace_ops **list,
+				  struct ftrace_ops *main_ops,
+				  struct ftrace_ops *ops)
+{
+	int ret = remove_ftrace_ops(list, ops);
+	if (!ret && *list == &ftrace_list_end)
+		ret = remove_ftrace_ops(&ftrace_ops_list, main_ops);
+	return ret;
+}
+
 static int __register_ftrace_function(struct ftrace_ops *ops)
 {
 	if (ftrace_disabled)
···
 	if (WARN_ON(ops->flags & FTRACE_OPS_FL_ENABLED))
 		return -EBUSY;
 
+	/* We don't support both control and global flags set. */
+	if ((ops->flags & FL_GLOBAL_CONTROL_MASK) == FL_GLOBAL_CONTROL_MASK)
+		return -EINVAL;
+
 	if (!core_kernel_data((unsigned long)ops))
 		ops->flags |= FTRACE_OPS_FL_DYNAMIC;
 
 	if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
-		int first = ftrace_global_list == &ftrace_list_end;
-		add_ftrace_ops(&ftrace_global_list, ops);
+		add_ftrace_list_ops(&ftrace_global_list, &global_ops, ops);
 		ops->flags |= FTRACE_OPS_FL_ENABLED;
-		if (first)
-			add_ftrace_ops(&ftrace_ops_list, &global_ops);
+	} else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+		if (control_ops_alloc(ops))
+			return -ENOMEM;
+		add_ftrace_list_ops(&ftrace_control_list, &control_ops, ops);
 	} else
 		add_ftrace_ops(&ftrace_ops_list, ops);
 
···
 		return -EINVAL;
 
 	if (ops->flags & FTRACE_OPS_FL_GLOBAL) {
-		ret = remove_ftrace_ops(&ftrace_global_list, ops);
-		if (!ret && ftrace_global_list == &ftrace_list_end)
-			ret = remove_ftrace_ops(&ftrace_ops_list, &global_ops);
+		ret = remove_ftrace_list_ops(&ftrace_global_list,
+					     &global_ops, ops);
 		if (!ret)
 			ops->flags &= ~FTRACE_OPS_FL_ENABLED;
+	} else if (ops->flags & FTRACE_OPS_FL_CONTROL) {
+		ret = remove_ftrace_list_ops(&ftrace_control_list,
+					     &control_ops, ops);
+		if (!ret) {
+			/*
+			 * The ftrace_ops is now removed from the list,
+			 * so there'll be no new users. We must ensure
+			 * all current users are done before we free
+			 * the control data.
+			 */
+			synchronize_sched();
+			control_ops_free(ops);
+		}
 	} else
 		ret = remove_ftrace_ops(&ftrace_ops_list, ops);
 
···
 	if (!hash || hash == EMPTY_HASH)
 		return;
 	call_rcu_sched(&hash->rcu, __free_ftrace_hash_rcu);
+}
+
+void ftrace_free_filter(struct ftrace_ops *ops)
+{
+	free_ftrace_hash(ops->filter_hash);
+	free_ftrace_hash(ops->notrace_hash);
 }
 
 static struct ftrace_hash *alloc_ftrace_hash(int size_bits)
···
 }
 
 #endif /* CONFIG_DYNAMIC_FTRACE */
+
+static void
+ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip)
+{
+	struct ftrace_ops *op;
+
+	if (unlikely(trace_recursion_test(TRACE_CONTROL_BIT)))
+		return;
+
+	/*
+	 * Some of the ops may be dynamically allocated,
+	 * they must be freed after a synchronize_sched().
+	 */
+	preempt_disable_notrace();
+	trace_recursion_set(TRACE_CONTROL_BIT);
+	op = rcu_dereference_raw(ftrace_control_list);
+	while (op != &ftrace_list_end) {
+		if (!ftrace_function_local_disabled(op) &&
+		    ftrace_ops_test(op, ip))
+			op->func(ip, parent_ip);
+
+		op = rcu_dereference_raw(op->next);
+	};
+	trace_recursion_clear(TRACE_CONTROL_BIT);
+	preempt_enable_notrace();
+}
+
+static struct ftrace_ops control_ops = {
+	.func = ftrace_ops_control_func,
+};
 
 static void
 ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip)

kernel/trace/trace.h

···
 #define F_STRUCT(args...)		args
 
 #undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print)	\
-	struct struct_name {					\
-		struct trace_entry	ent;			\
-		tstruct						\
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter)	\
+	struct struct_name {						\
+		struct trace_entry	ent;				\
+		tstruct							\
 	}
 
 #undef TP_ARGS
 #define TP_ARGS(args...)	args
 
 #undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk)
+#define FTRACE_ENTRY_DUP(name, name_struct, id, tstruct, printk, filter)
+
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print,	\
+			 filter, regfn) \
+	FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+		     filter)
 
 #include "trace_entries.h"
 
···
 /* for function tracing recursion */
 #define TRACE_INTERNAL_BIT		(1<<11)
 #define TRACE_GLOBAL_BIT		(1<<12)
+#define TRACE_CONTROL_BIT		(1<<13)
+
 /*
  * Abuse of the trace_recursion.
  * As we need a way to maintain state if we are tracing the function
···
 static inline int ftrace_is_dead(void) { return 0; }
 #endif
 
+int ftrace_event_is_function(struct ftrace_event_call *call);
+
 /*
  * struct trace_parser - servers for reading the user input separated by spaces
  * @cont: set if the input is not complete - no final space char was found
···
 	u64 			val;
 	struct regex		regex;
 	unsigned short		*ops;
-#ifdef CONFIG_FTRACE_STARTUP_TEST
 	struct ftrace_event_field *field;
-#endif
 	int 			offset;
 	int 			not;
 	int 			op;
···
 extern const char *__stop___trace_bprintk_fmt[];
 
 #undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, id, tstruct, print)		\
+#define FTRACE_ENTRY(call, struct_name, id, tstruct, print, filter)	\
 	extern struct ftrace_event_call					\
 	__attribute__((__aligned__(4))) event_##call;
 #undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print)		\
-	FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(call, struct_name, id, tstruct, print, filter)	\
+	FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+		     filter)
 #include "trace_entries.h"
+
+#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_FUNCTION_TRACER
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+			       enum trace_reg type, void *data);
+#else
+#define perf_ftrace_event_register NULL
+#endif /* CONFIG_FUNCTION_TRACER */
+#endif /* CONFIG_PERF_EVENTS */
 
 #endif /* _LINUX_KERNEL_TRACE_H */

kernel/trace/trace_entries.h

···
 /*
  * Function trace entry - function address and parent function address:
  */
-FTRACE_ENTRY(function, ftrace_entry,
+FTRACE_ENTRY_REG(function, ftrace_entry,
 
 	TRACE_FN,
 
···
 		__field(	unsigned long,	parent_ip	)
 	),
 
-	F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip)
+	F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
+
+	FILTER_TRACE_FN,
+
+	perf_ftrace_event_register
 );
 
 /* Function call entry */
···
 		__field_desc(	int,		graph_ent,	depth		)
 	),
 
-	F_printk("--> %lx (%d)", __entry->func, __entry->depth)
+	F_printk("--> %lx (%d)", __entry->func, __entry->depth),
+
+	FILTER_OTHER
 );
 
 /* Function return entry */
···
 	F_printk("<-- %lx (%d) (start: %llx  end: %llx) over: %d",
 		 __entry->func, __entry->depth,
 		 __entry->calltime, __entry->rettime,
-		 __entry->depth)
+		 __entry->depth),
+
+	FILTER_OTHER
 );
 
 /*
···
 	F_printk("%u:%u:%u  ==> %u:%u:%u [%03u]",
 		 __entry->prev_pid, __entry->prev_prio, __entry->prev_state,
 		 __entry->next_pid, __entry->next_prio, __entry->next_state,
-		 __entry->next_cpu
-		)
+		 __entry->next_cpu),
+
+	FILTER_OTHER
 );
 
 /*
···
 	F_printk("%u:%u:%u  ==+ %u:%u:%u [%03u]",
 		 __entry->prev_pid, __entry->prev_prio, __entry->prev_state,
 		 __entry->next_pid, __entry->next_prio, __entry->next_state,
-		 __entry->next_cpu
-		)
+		 __entry->next_cpu),
+
+	FILTER_OTHER
 );
 
 /*
···
 		 "\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
 		 __entry->caller[0], __entry->caller[1], __entry->caller[2],
 		 __entry->caller[3], __entry->caller[4], __entry->caller[5],
-		 __entry->caller[6], __entry->caller[7])
+		 __entry->caller[6], __entry->caller[7]),
+
+	FILTER_OTHER
 );
 
 FTRACE_ENTRY(user_stack, userstack_entry,
···
 		 "\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n\t=> (%08lx)\n",
 		 __entry->caller[0], __entry->caller[1], __entry->caller[2],
 		 __entry->caller[3], __entry->caller[4], __entry->caller[5],
-		 __entry->caller[6], __entry->caller[7])
+		 __entry->caller[6], __entry->caller[7]),
+
+	FILTER_OTHER
 );
 
 /*
···
 	),
 
 	F_printk("%08lx fmt:%p",
-		 __entry->ip, __entry->fmt)
+		 __entry->ip, __entry->fmt),
+
+	FILTER_OTHER
 );
 
 FTRACE_ENTRY(print, print_entry,
···
 	),
 
 	F_printk("%08lx %s",
-		 __entry->ip, __entry->buf)
+		 __entry->ip, __entry->buf),
+
+	FILTER_OTHER
 );
 
 FTRACE_ENTRY(mmiotrace_rw, trace_mmiotrace_rw,
···
 
 	F_printk("%lx %lx %lx %d %x %x",
 		 (unsigned long)__entry->phys, __entry->value, __entry->pc,
-		 __entry->map_id, __entry->opcode, __entry->width)
+		 __entry->map_id, __entry->opcode, __entry->width),
+
+	FILTER_OTHER
 );
 
 FTRACE_ENTRY(mmiotrace_map, trace_mmiotrace_map,
···
 
 	F_printk("%lx %lx %lx %d %x",
 		 (unsigned long)__entry->phys, __entry->virt, __entry->len,
-		 __entry->map_id, __entry->opcode)
+		 __entry->map_id, __entry->opcode),
+
+	FILTER_OTHER
 );
 
 
···
 
 	F_printk("%u:%s:%s (%u)",
 		 __entry->line,
-		 __entry->func, __entry->file, __entry->correct)
+		 __entry->func, __entry->file, __entry->correct),
+
+	FILTER_OTHER
 );
 

kernel/trace/trace_event_perf.c

···
 static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
 				 struct perf_event *p_event)
 {
+	/* The ftrace function trace is allowed only for root. */
+	if (ftrace_event_is_function(tp_event) &&
+	    perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
 	/* No tracing, just counting, so no obvious leak */
 	if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
 		return 0;
···
 	return 0;
 }
 
-static int perf_trace_event_init(struct ftrace_event_call *tp_event,
-				 struct perf_event *p_event)
+static int perf_trace_event_reg(struct ftrace_event_call *tp_event,
+				struct perf_event *p_event)
 {
 	struct hlist_head __percpu *list;
-	int ret;
+	int ret = -ENOMEM;
 	int cpu;
-
-	ret = perf_trace_event_perm(tp_event, p_event);
-	if (ret)
-		return ret;
 
 	p_event->tp_event = tp_event;
 	if (tp_event->perf_refcount++ > 0)
 		return 0;
-
-	ret = -ENOMEM;
 
 	list = alloc_percpu(struct hlist_head);
 	if (!list)
···
 		}
 	}
 
-	ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER);
+	ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER, NULL);
 	if (ret)
 		goto fail;
 
···
 	return ret;
 }
 
+static void perf_trace_event_unreg(struct perf_event *p_event)
+{
+	struct ftrace_event_call *tp_event = p_event->tp_event;
+	int i;
+
+	if (--tp_event->perf_refcount > 0)
+		goto out;
+
+	tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER, NULL);
+
+	/*
+	 * Ensure our callback won't be called anymore. The buffers
+	 * will be freed after that.
+	 */
+	tracepoint_synchronize_unregister();
+
+	free_percpu(tp_event->perf_events);
+	tp_event->perf_events = NULL;
+
+	if (!--total_ref_count) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
+			free_percpu(perf_trace_buf[i]);
+			perf_trace_buf[i] = NULL;
+		}
+	}
+out:
+	module_put(tp_event->mod);
+}
+
+static int perf_trace_event_open(struct perf_event *p_event)
+{
+	struct ftrace_event_call *tp_event = p_event->tp_event;
+	return tp_event->class->reg(tp_event, TRACE_REG_PERF_OPEN, p_event);
+}
+
+static void perf_trace_event_close(struct perf_event *p_event)
+{
+	struct ftrace_event_call *tp_event = p_event->tp_event;
+	tp_event->class->reg(tp_event, TRACE_REG_PERF_CLOSE, p_event);
+}
+
+static int perf_trace_event_init(struct ftrace_event_call *tp_event,
+				 struct perf_event *p_event)
+{
+	int ret;
+
+	ret = perf_trace_event_perm(tp_event, p_event);
+	if (ret)
+		return ret;
+
+	ret = perf_trace_event_reg(tp_event, p_event);
+	if (ret)
+		return ret;
+
+	ret = perf_trace_event_open(p_event);
+	if (ret) {
+		perf_trace_event_unreg(p_event);
+		return ret;
+	}
+
+	return 0;
+}
+
 int perf_trace_init(struct perf_event *p_event)
 {
 	struct ftrace_event_call *tp_event;
···
 	return ret;
 }
 
+void perf_trace_destroy(struct perf_event *p_event)
+{
+	mutex_lock(&event_mutex);
+	perf_trace_event_close(p_event);
+	perf_trace_event_unreg(p_event);
+	mutex_unlock(&event_mutex);
+}
+
 int perf_trace_add(struct perf_event *p_event, int flags)
 {
 	struct ftrace_event_call *tp_event = p_event->tp_event;
···
 	list = this_cpu_ptr(pcpu_list);
 	hlist_add_head_rcu(&p_event->hlist_entry, list);
 
-	return 0;
+	return tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event);
 }
 
 void perf_trace_del(struct perf_event *p_event, int flags)
 {
-	hlist_del_rcu(&p_event->hlist_entry);
-}
-
-void perf_trace_destroy(struct perf_event *p_event)
-{
 	struct ftrace_event_call *tp_event = p_event->tp_event;
-	int i;
-
-	mutex_lock(&event_mutex);
-	if (--tp_event->perf_refcount > 0)
-		goto out;
-
-	tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER);
-
-	/*
-	 * Ensure our callback won't be called anymore. The buffers
-	 * will be freed after that.
-	 */
-	tracepoint_synchronize_unregister();
-
-	free_percpu(tp_event->perf_events);
-	tp_event->perf_events = NULL;
-
-	if (!--total_ref_count) {
-		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
-			free_percpu(perf_trace_buf[i]);
-			perf_trace_buf[i] = NULL;
-		}
-	}
-out:
-	module_put(tp_event->mod);
-	mutex_unlock(&event_mutex);
+	hlist_del_rcu(&p_event->hlist_entry);
+	tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event);
 }
 
 __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
···
 	return raw_data;
 }
 EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
+
+#ifdef CONFIG_FUNCTION_TRACER
+static void
+perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip)
+{
+	struct ftrace_entry *entry;
+	struct hlist_head *head;
+	struct pt_regs regs;
+	int rctx;
+
+#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
+		    sizeof(u64)) - sizeof(u32))
+
+	BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
+
+	perf_fetch_caller_regs(&regs);
+
+	entry = perf_trace_buf_prepare(ENTRY_SIZE, TRACE_FN, NULL, &rctx);
+	if (!entry)
+		return;
+
+	entry->ip = ip;
+	entry->parent_ip = parent_ip;
+
+	head = this_cpu_ptr(event_function.perf_events);
+	perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0,
+			      1, &regs, head);
+
+#undef ENTRY_SIZE
+}
+
+static int perf_ftrace_function_register(struct perf_event *event)
+{
+	struct ftrace_ops *ops = &event->ftrace_ops;
+
+	ops->flags |= FTRACE_OPS_FL_CONTROL;
+	ops->func = perf_ftrace_function_call;
+	return register_ftrace_function(ops);
+}
+
+static int perf_ftrace_function_unregister(struct perf_event *event)
+{
+	struct ftrace_ops *ops = &event->ftrace_ops;
+	int ret = unregister_ftrace_function(ops);
+	ftrace_free_filter(ops);
+	return ret;
+}
+
+static void perf_ftrace_function_enable(struct perf_event *event)
+{
+	ftrace_function_local_enable(&event->ftrace_ops);
+}
+
+static void perf_ftrace_function_disable(struct perf_event *event)
+{
+	ftrace_function_local_disable(&event->ftrace_ops);
+}
+
+int perf_ftrace_event_register(struct ftrace_event_call *call,
+			       enum trace_reg type, void *data)
+{
+	switch (type) {
+	case TRACE_REG_REGISTER:
+	case TRACE_REG_UNREGISTER:
+		break;
+	case TRACE_REG_PERF_REGISTER:
+	case TRACE_REG_PERF_UNREGISTER:
+		return 0;
+	case TRACE_REG_PERF_OPEN:
+		return perf_ftrace_function_register(data);
+	case TRACE_REG_PERF_CLOSE:
+		return perf_ftrace_function_unregister(data);
+	case TRACE_REG_PERF_ADD:
+		perf_ftrace_function_enable(data);
+		return 0;
+	case TRACE_REG_PERF_DEL:
+		perf_ftrace_function_disable(data);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+#endif /* CONFIG_FUNCTION_TRACER */

kernel/trace/trace_events.c

···
 }
 EXPORT_SYMBOL_GPL(trace_event_raw_init);
 
-int ftrace_event_reg(struct ftrace_event_call *call, enum trace_reg type)
+int ftrace_event_reg(struct ftrace_event_call *call,
+		     enum trace_reg type, void *data)
 {
 	switch (type) {
 	case TRACE_REG_REGISTER:
···
 		tracepoint_probe_unregister(call->name,
 					    call->class->perf_probe,
 					    call);
+		return 0;
+	case TRACE_REG_PERF_OPEN:
+	case TRACE_REG_PERF_CLOSE:
+	case TRACE_REG_PERF_ADD:
+	case TRACE_REG_PERF_DEL:
 		return 0;
 #endif
 	}
···
 				tracing_stop_cmdline_record();
 				call->flags &= ~TRACE_EVENT_FL_RECORDED_CMD;
 			}
-			call->class->reg(call, TRACE_REG_UNREGISTER);
+			call->class->reg(call, TRACE_REG_UNREGISTER, NULL);
 		}
 		break;
 	case 1:
···
 				tracing_start_cmdline_record();
 				call->flags |= TRACE_EVENT_FL_RECORDED_CMD;
 			}
-			ret = call->class->reg(call, TRACE_REG_REGISTER);
+			ret = call->class->reg(call, TRACE_REG_REGISTER, NULL);
 			if (ret) {
 				tracing_stop_cmdline_record();
 				pr_info("event trace: Could not enable event "

kernel/trace/trace_events_filter.c

···
 	FILT_ERR_TOO_MANY_PREDS,
 	FILT_ERR_MISSING_FIELD,
 	FILT_ERR_INVALID_FILTER,
+	FILT_ERR_IP_FIELD_ONLY,
 };
 
 static char *err_text[] = {
···
 	"Too many terms in predicate expression",
 	"Missing field name and/or value",
 	"Meaningless filter expression",
+	"Only 'ip' field is supported for function trace",
 };
 
 struct opstack_op {
···
 	return FILTER_OTHER;
 }
 
+static bool is_function_field(struct ftrace_event_field *field)
+{
+	return field->filter_type == FILTER_TRACE_FN;
+}
+
 static bool is_string_field(struct ftrace_event_field *field)
 {
 	return field->filter_type == FILTER_DYN_STRING ||
···
 			fn = filter_pred_strloc;
 		else
 			fn = filter_pred_pchar;
+	} else if (is_function_field(field)) {
+		if (strcmp(field->name, "ip")) {
+			parse_error(ps, FILT_ERR_IP_FIELD_ONLY, 0);
+			return -EINVAL;
+		}
 	} else {
 		if (field->is_signed)
 			ret = strict_strtoll(pred->regex.pattern, 0, &val);
···
 
 	strcpy(pred.regex.pattern, operand2);
 	pred.regex.len = strlen(pred.regex.pattern);
-
-#ifdef CONFIG_FTRACE_STARTUP_TEST
 	pred.field = field;
-#endif
 	return init_pred(ps, field, &pred) ? NULL : &pred;
 }
 
···
 	__free_filter(filter);
 }
 
+struct function_filter_data {
+	struct ftrace_ops *ops;
+	int first_filter;
+	int first_notrace;
+};
+
+#ifdef CONFIG_FUNCTION_TRACER
+static char **
+ftrace_function_filter_re(char *buf, int len, int *count)
+{
+	char *str, *sep, **re;
+
+	str = kstrndup(buf, len, GFP_KERNEL);
+	if (!str)
+		return NULL;
+
+	/*
+	 * The argv_split function takes white space
+	 * as a separator, so convert ',' into spaces.
+	 */
+	while ((sep = strchr(str, ',')))
+		*sep = ' ';
+
+	re = argv_split(GFP_KERNEL, str, count);
+	kfree(str);
+	return re;
+}
+
+static int ftrace_function_set_regexp(struct ftrace_ops *ops, int filter,
+				      int reset, char *re, int len)
+{
+	int ret;
+
+	if (filter)
+		ret = ftrace_set_filter(ops, re, len, reset);
+	else
+		ret = ftrace_set_notrace(ops, re, len, reset);
+
+	return ret;
+}
+
+static int __ftrace_function_set_filter(int filter, char *buf, int len,
+					struct function_filter_data *data)
+{
+	int i, re_cnt, ret;
+	int *reset;
+	char **re;
+
+	reset = filter ? &data->first_filter : &data->first_notrace;
+
+	/*
+	 * The 'ip' field could have multiple filters set, separated
+	 * either by space or comma. We first cut the filter and apply
+	 * all pieces separatelly.
+	 */
+	re = ftrace_function_filter_re(buf, len, &re_cnt);
+	if (!re)
+		return -EINVAL;
+
+	for (i = 0; i < re_cnt; i++) {
+		ret = ftrace_function_set_regexp(data->ops, filter, *reset,
+						 re[i], strlen(re[i]));
+		if (ret)
+			break;
+
+		if (*reset)
+			*reset = 0;
+	}
+
+	argv_free(re);
+	return ret;
+}
+
+static int ftrace_function_check_pred(struct filter_pred *pred, int leaf)
+{
+	struct ftrace_event_field *field = pred->field;
+
+	if (leaf) {
+		/*
+		 * Check the leaf predicate for function trace, verify:
+		 *  - only '==' and '!=' is used
+		 *  - the 'ip' field is used
+		 */
+		if ((pred->op != OP_EQ) && (pred->op != OP_NE))
+			return -EINVAL;
+
+		if (strcmp(field->name, "ip"))
+			return -EINVAL;
+	} else {
+		/*
+		 * Check the non leaf predicate for function trace, verify:
+		 *  - only '||' is used
+		*/
+		if (pred->op != OP_OR)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int ftrace_function_set_filter_cb(enum move_type move,
+					 struct filter_pred *pred,
+					 int *err, void *data)
+{
+	/* Checking the node is valid for function trace. */
+	if ((move != MOVE_DOWN) ||
+	    (pred->left != FILTER_PRED_INVALID)) {
+		*err = ftrace_function_check_pred(pred, 0);
+	} else {
+		*err = ftrace_function_check_pred(pred, 1);
+		if (*err)
+			return WALK_PRED_ABORT;
+
+		*err = __ftrace_function_set_filter(pred->op == OP_EQ,
+						    pred->regex.pattern,
+						    pred->regex.len,
+						    data);
+	}
+
+	return (*err) ? WALK_PRED_ABORT : WALK_PRED_DEFAULT;
+}
+
+static int ftrace_function_set_filter(struct perf_event *event,
+				      struct event_filter *filter)
+{
+	struct function_filter_data data = {
+		.first_filter  = 1,
+		.first_notrace = 1,
+		.ops           = &event->ftrace_ops,
+	};
+
+	return walk_pred_tree(filter->preds, filter->root,
+			      ftrace_function_set_filter_cb, &data);
+}
+#else
+static int ftrace_function_set_filter(struct perf_event *event,
+				      struct event_filter *filter)
+{
+	return -ENODEV;
+}
+#endif /* CONFIG_FUNCTION_TRACER */
+
 int ftrace_profile_set_filter(struct perf_event *event, int event_id,
 			      char *filter_str)
 {
···
 		goto out_unlock;
 
 	err = create_filter(call, filter_str, false, &filter);
-	if (!err)
-		event->filter = filter;
+	if (err)
+		goto free_filter;
+
+	if (ftrace_event_is_function(call))
+		err = ftrace_function_set_filter(event, filter);
 	else
+		event->filter = filter;
+
+free_filter:
+	if (err || ftrace_event_is_function(call))
 		__free_filter(filter);
 
 out_unlock:

kernel/trace/trace_export.c

···
 #undef TRACE_SYSTEM
 #define TRACE_SYSTEM	ftrace
 
+/*
+ * The FTRACE_ENTRY_REG macro allows ftrace entry to define register
+ * function and thus become accesible via perf.
+ */
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(name, struct_name, id, tstruct, print, \
+			 filter, regfn) \
+	FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+		     filter)
+
 /* not needed for this file */
 #undef __field_struct
 #define __field_struct(type, item)
···
 #define F_printk(fmt, args...) fmt, args
 
 #undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print)	\
-struct ____ftrace_##name {					\
-	tstruct							\
-};								\
-static void __always_unused ____ftrace_check_##name(void)	\
-{								\
-	struct ____ftrace_##name *__entry = NULL;		\
-								\
-	/* force compile-time check on F_printk() */		\
-	printk(print);						\
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter)	\
+struct ____ftrace_##name {						\
+	tstruct								\
+};									\
+static void __always_unused ____ftrace_check_##name(void)		\
+{									\
+	struct ____ftrace_##name *__entry = NULL;			\
+									\
+	/* force compile-time check on F_printk() */			\
+	printk(print);							\
 }
 
 #undef FTRACE_ENTRY_DUP
-#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print)	\
-	FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print))
+#define FTRACE_ENTRY_DUP(name, struct_name, id, tstruct, print, filter)	\
+	FTRACE_ENTRY(name, struct_name, id, PARAMS(tstruct), PARAMS(print), \
+		     filter)
 
 #include "trace_entries.h"
 
···
 	ret = trace_define_field(event_call, #type, #item,		\
 				 offsetof(typeof(field), item),		\
 				 sizeof(field.item),			\
-				 is_signed_type(type), FILTER_OTHER);	\
+				 is_signed_type(type), filter_type);	\
 	if (ret)							\
 		return ret;
 
···
 				 offsetof(typeof(field),		\
 					  container.item),		\
 				 sizeof(field.container.item),		\
-				 is_signed_type(type), FILTER_OTHER);	\
+				 is_signed_type(type), filter_type);	\
 	if (ret)							\
 		return ret;
 
···
 		ret = trace_define_field(event_call, event_storage, #item, \
 				 offsetof(typeof(field), item),		\
 				 sizeof(field.item),			\
-				 is_signed_type(type), FILTER_OTHER);	\
+				 is_signed_type(type), filter_type);	\
 		mutex_unlock(&event_storage_mutex);			\
 		if (ret)						\
 			return ret;					\
···
 				 offsetof(typeof(field),		\
 					  container.item),		\
 				 sizeof(field.container.item),		\
-				 is_signed_type(type), FILTER_OTHER);	\
+				 is_signed_type(type), filter_type);	\
 	if (ret)							\
 		return ret;
 
···
 #define __dynamic_array(type, item)					\
 	ret = trace_define_field(event_call, #type, #item,		\
 				 offsetof(typeof(field), item),		\
-				 0, is_signed_type(type), FILTER_OTHER);\
+				 0, is_signed_type(type), filter_type);\
 	if (ret)							\
 		return ret;
 
 #undef FTRACE_ENTRY
-#define FTRACE_ENTRY(name, struct_name, id, tstruct, print)		\
+#define FTRACE_ENTRY(name, struct_name, id, tstruct, print, filter)	\
 int									\
 ftrace_define_fields_##name(struct ftrace_event_call *event_call)	\
 {									\
 	struct struct_name field;					\
 	int ret;							\
+	int filter_type = filter;					\
 									\
 	tstruct;							\
 									\
···
 #undef F_printk
 #define F_printk(fmt, args...) #fmt ", "  __stringify(args)
 
-#undef FTRACE_ENTRY
-#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print)		\
+#undef FTRACE_ENTRY_REG
+#define FTRACE_ENTRY_REG(call, struct_name, etype, tstruct, print, filter,\
+			 regfn)						\
 									\
 struct ftrace_event_class event_class_ftrace_##call = {			\
 	.system			= __stringify(TRACE_SYSTEM),		\
 	.define_fields		= ftrace_define_fields_##call,		\
 	.fields			= LIST_HEAD_INIT(event_class_ftrace_##call.fields),\
+	.reg			= regfn,				\
 };									\
 									\
 struct ftrace_event_call __used event_##call = {			\
···
 };									\
 struct ftrace_event_call __used						\
 __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
+
+#undef FTRACE_ENTRY
+#define FTRACE_ENTRY(call, struct_name, etype, tstruct, print, filter)	\
+	FTRACE_ENTRY_REG(call, struct_name, etype,			\
+			 PARAMS(tstruct), PARAMS(print), filter, NULL)
+
+int ftrace_event_is_function(struct ftrace_event_call *call)
+{
+	return call == &event_function;
+}
 
 #include "trace_entries.h"

kernel/trace/trace_kprobe.c

···
 #endif	/* CONFIG_PERF_EVENTS */
 
 static __kprobes
-int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
+int kprobe_register(struct ftrace_event_call *event,
+		    enum trace_reg type, void *data)
 {
 	struct trace_probe *tp = (struct trace_probe *)event->data;
 
···
 		return enable_trace_probe(tp, TP_FLAG_PROFILE);
 	case TRACE_REG_PERF_UNREGISTER:
 		disable_trace_probe(tp, TP_FLAG_PROFILE);
+		return 0;
+	case TRACE_REG_PERF_OPEN:
+	case TRACE_REG_PERF_CLOSE:
+	case TRACE_REG_PERF_ADD:
+	case TRACE_REG_PERF_DEL:
 		return 0;
 #endif
 	}

kernel/trace/trace_output.c

···
 	unsigned long mask;
 	const char *str;
 	const char *ret = p->buffer + p->len;
-	int i;
+	int i, first = 1;
 
 	for (i = 0;  flag_array[i].name && flags; i++) {
 
···
 
 		str = flag_array[i].name;
 		flags &= ~mask;
-		if (p->len && delim)
+		if (!first && delim)
 			trace_seq_puts(p, delim);
+		else
+			first = 0;
 		trace_seq_puts(p, str);
 	}
 
 	/* check for left over flags */
 	if (flags) {
-		if (p->len && delim)
+		if (!first && delim)
 			trace_seq_puts(p, delim);
 		trace_seq_printf(p, "0x%lx", flags);
 	}
···
 		break;
 	}
 
-	if (!p->len)
+	if (ret == (const char *)(p->buffer + p->len))
 		trace_seq_printf(p, "0x%lx", val);
 		
 	trace_seq_putc(p, 0);
···
 		break;
 	}
 
-	if (!p->len)
+	if (ret == (const char *)(p->buffer + p->len))
 		trace_seq_printf(p, "0x%llx", val);
 
 	trace_seq_putc(p, 0);

kernel/trace/trace_syscalls.c

···
 static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
 
 static int syscall_enter_register(struct ftrace_event_call *event,
-				 enum trace_reg type);
+				 enum trace_reg type, void *data);
 static int syscall_exit_register(struct ftrace_event_call *event,
-				 enum trace_reg type);
+				 enum trace_reg type, void *data);
 
 static int syscall_enter_define_fields(struct ftrace_event_call *call);
 static int syscall_exit_define_fields(struct ftrace_event_call *call);
···
 #endif /* CONFIG_PERF_EVENTS */
 
 static int syscall_enter_register(struct ftrace_event_call *event,
-				 enum trace_reg type)
+				 enum trace_reg type, void *data)
 {
 	switch (type) {
 	case TRACE_REG_REGISTER:
···
 	case TRACE_REG_PERF_UNREGISTER:
 		perf_sysenter_disable(event);
 		return 0;
+	case TRACE_REG_PERF_OPEN:
+	case TRACE_REG_PERF_CLOSE:
+	case TRACE_REG_PERF_ADD:
+	case TRACE_REG_PERF_DEL:
+		return 0;
 #endif
 	}
 	return 0;
 }
 
 static int syscall_exit_register(struct ftrace_event_call *event,
-				 enum trace_reg type)
+				 enum trace_reg type, void *data)
 {
 	switch (type) {
 	case TRACE_REG_REGISTER:
···
 		return perf_sysexit_enable(event);
 	case TRACE_REG_PERF_UNREGISTER:
 		perf_sysexit_disable(event);
+		return 0;
+	case TRACE_REG_PERF_OPEN:
+	case TRACE_REG_PERF_CLOSE:
+	case TRACE_REG_PERF_ADD:
+	case TRACE_REG_PERF_DEL:
 		return 0;
 #endif
 	}