Merge tag 'trace-v4.14' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

Pull tracing updates from Steven Rostedt:
 "Nothing new in development for this release. These are mostly fixes
  that were found during development of changes for the next merge
  window and fixes that were sent to me late in the last cycle"

* tag 'trace-v4.14' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Apply trace_clock changes to instance max buffer
  tracing: Fix clear of RECORDED_TGID flag when disabling trace event
  tracing: Add barrier to trace_printk() buffer nesting modification
  ftrace: Fix memleak when unregistering dynamic ops when tracing disabled
  ftrace: Fix selftest goto location on error
  ftrace: Zero out ftrace hashes when a module is removed
  tracing: Only have rmmod clear buffers that its events were active in
  ftrace: Fix debug preempt config name in stack_tracer_{en,dis}able

include/linux/ftrace.h

@@ -307,7 +307,7 @@ DECLARE_PER_CPU(int, disable_stack_tracer);
 static inline void stack_tracer_disable(void)
 {
 	/* Preemption or interupts must be disabled */
-	if (IS_ENABLED(CONFIG_PREEMPT_DEBUG))
+	if (IS_ENABLED(CONFIG_DEBUG_PREEMPT))
 		WARN_ON_ONCE(!preempt_count() || !irqs_disabled());
 	this_cpu_inc(disable_stack_tracer);
 }
@@ -320,7 +320,7 @@ static inline void stack_tracer_disable(void)
  */
 static inline void stack_tracer_enable(void)
 {
-	if (IS_ENABLED(CONFIG_PREEMPT_DEBUG))
+	if (IS_ENABLED(CONFIG_DEBUG_PREEMPT))
 		WARN_ON_ONCE(!preempt_count() || !irqs_disabled());
 	this_cpu_dec(disable_stack_tracer);
 }

include/linux/trace_events.h

@@ -217,7 +217,6 @@ enum {
 	TRACE_EVENT_FL_CAP_ANY_BIT,
 	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	TRACE_EVENT_FL_IGNORE_ENABLE_BIT,
-	TRACE_EVENT_FL_WAS_ENABLED_BIT,
 	TRACE_EVENT_FL_TRACEPOINT_BIT,
 	TRACE_EVENT_FL_KPROBE_BIT,
 	TRACE_EVENT_FL_UPROBE_BIT,
@@ -229,9 +228,6 @@ enum {
  *  CAP_ANY	  - Any user can enable for perf
  *  NO_SET_FILTER - Set when filter has error and is to be ignored
  *  IGNORE_ENABLE - For trace internal events, do not enable with debugfs file
- *  WAS_ENABLED   - Set and stays set when an event was ever enabled
- *                    (used for module unloading, if a module event is enabled,
- *                     it is best to clear the buffers that used it).
  *  TRACEPOINT    - Event is a tracepoint
  *  KPROBE        - Event is a kprobe
  *  UPROBE        - Event is a uprobe
@@ -241,7 +237,6 @@ enum {
 	TRACE_EVENT_FL_CAP_ANY		= (1 << TRACE_EVENT_FL_CAP_ANY_BIT),
 	TRACE_EVENT_FL_NO_SET_FILTER	= (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
 	TRACE_EVENT_FL_IGNORE_ENABLE	= (1 << TRACE_EVENT_FL_IGNORE_ENABLE_BIT),
-	TRACE_EVENT_FL_WAS_ENABLED	= (1 << TRACE_EVENT_FL_WAS_ENABLED_BIT),
 	TRACE_EVENT_FL_TRACEPOINT	= (1 << TRACE_EVENT_FL_TRACEPOINT_BIT),
 	TRACE_EVENT_FL_KPROBE		= (1 << TRACE_EVENT_FL_KPROBE_BIT),
 	TRACE_EVENT_FL_UPROBE		= (1 << TRACE_EVENT_FL_UPROBE_BIT),
@@ -306,6 +301,7 @@ enum {
 	EVENT_FILE_FL_TRIGGER_MODE_BIT,
 	EVENT_FILE_FL_TRIGGER_COND_BIT,
 	EVENT_FILE_FL_PID_FILTER_BIT,
+	EVENT_FILE_FL_WAS_ENABLED_BIT,
 };
 
 /*
@@ -321,6 +317,7 @@ enum {
  *  TRIGGER_MODE  - When set, invoke the triggers associated with the event
  *  TRIGGER_COND  - When set, one or more triggers has an associated filter
  *  PID_FILTER    - When set, the event is filtered based on pid
+ *  WAS_ENABLED   - Set when enabled to know to clear trace on module removal
  */
 enum {
 	EVENT_FILE_FL_ENABLED		= (1 << EVENT_FILE_FL_ENABLED_BIT),
@@ -333,6 +330,7 @@ enum {
 	EVENT_FILE_FL_TRIGGER_MODE	= (1 << EVENT_FILE_FL_TRIGGER_MODE_BIT),
 	EVENT_FILE_FL_TRIGGER_COND	= (1 << EVENT_FILE_FL_TRIGGER_COND_BIT),
 	EVENT_FILE_FL_PID_FILTER	= (1 << EVENT_FILE_FL_PID_FILTER_BIT),
+	EVENT_FILE_FL_WAS_ENABLED	= (1 << EVENT_FILE_FL_WAS_ENABLED_BIT),
 };
 
 struct trace_event_file {

kernel/trace/ftrace.c

@@ -2828,13 +2828,14 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command)
 
 	if (!command || !ftrace_enabled) {
 		/*
-		 * If these are per_cpu ops, they still need their
-		 * per_cpu field freed. Since, function tracing is
+		 * If these are dynamic or per_cpu ops, they still
+		 * need their data freed. Since, function tracing is
 		 * not currently active, we can just free them
 		 * without synchronizing all CPUs.
 		 */
-		if (ops->flags & FTRACE_OPS_FL_PER_CPU)
-			per_cpu_ops_free(ops);
+		if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU))
+			goto free_ops;
+
 		return 0;
 	}
 
@@ -2900,6 +2901,7 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command)
 		if (IS_ENABLED(CONFIG_PREEMPT))
 			synchronize_rcu_tasks();
 
+ free_ops:
 		arch_ftrace_trampoline_free(ops);
 
 		if (ops->flags & FTRACE_OPS_FL_PER_CPU)
@@ -5690,10 +5692,51 @@ static int referenced_filters(struct dyn_ftrace *rec)
 	return cnt;
 }
 
+static void
+clear_mod_from_hash(struct ftrace_page *pg, struct ftrace_hash *hash)
+{
+	struct ftrace_func_entry *entry;
+	struct dyn_ftrace *rec;
+	int i;
+
+	if (ftrace_hash_empty(hash))
+		return;
+
+	for (i = 0; i < pg->index; i++) {
+		rec = &pg->records[i];
+		entry = __ftrace_lookup_ip(hash, rec->ip);
+		/*
+		 * Do not allow this rec to match again.
+		 * Yeah, it may waste some memory, but will be removed
+		 * if/when the hash is modified again.
+		 */
+		if (entry)
+			entry->ip = 0;
+	}
+}
+
+/* Clear any records from hashs */
+static void clear_mod_from_hashes(struct ftrace_page *pg)
+{
+	struct trace_array *tr;
+
+	mutex_lock(&trace_types_lock);
+	list_for_each_entry(tr, &ftrace_trace_arrays, list) {
+		if (!tr->ops || !tr->ops->func_hash)
+			continue;
+		mutex_lock(&tr->ops->func_hash->regex_lock);
+		clear_mod_from_hash(pg, tr->ops->func_hash->filter_hash);
+		clear_mod_from_hash(pg, tr->ops->func_hash->notrace_hash);
+		mutex_unlock(&tr->ops->func_hash->regex_lock);
+	}
+	mutex_unlock(&trace_types_lock);
+}
+
 void ftrace_release_mod(struct module *mod)
 {
 	struct dyn_ftrace *rec;
 	struct ftrace_page **last_pg;
+	struct ftrace_page *tmp_page = NULL;
 	struct ftrace_page *pg;
 	int order;
 
@@ -5723,14 +5766,25 @@ void ftrace_release_mod(struct module *mod)
 
 			ftrace_update_tot_cnt -= pg->index;
 			*last_pg = pg->next;
-			order = get_count_order(pg->size / ENTRIES_PER_PAGE);
-			free_pages((unsigned long)pg->records, order);
-			kfree(pg);
+
+			pg->next = tmp_page;
+			tmp_page = pg;
 		} else
 			last_pg = &pg->next;
 	}
  out_unlock:
 	mutex_unlock(&ftrace_lock);
+
+	for (pg = tmp_page; pg; pg = tmp_page) {
+
+		/* Needs to be called outside of ftrace_lock */
+		clear_mod_from_hashes(pg);
+
+		order = get_count_order(pg->size / ENTRIES_PER_PAGE);
+		free_pages((unsigned long)pg->records, order);
+		tmp_page = pg->next;
+		kfree(pg);
+	}
 }
 
 void ftrace_module_enable(struct module *mod)

kernel/trace/trace.c

@@ -1702,6 +1702,9 @@ void tracing_reset_all_online_cpus(void)
 	struct trace_array *tr;
 
 	list_for_each_entry(tr, &ftrace_trace_arrays, list) {
+		if (!tr->clear_trace)
+			continue;
+		tr->clear_trace = false;
 		tracing_reset_online_cpus(&tr->trace_buffer);
 #ifdef CONFIG_TRACER_MAX_TRACE
 		tracing_reset_online_cpus(&tr->max_buffer);
@@ -2799,11 +2802,17 @@ static char *get_trace_buf(void)
 	if (!buffer || buffer->nesting >= 4)
 		return NULL;
 
-	return &buffer->buffer[buffer->nesting++][0];
+	buffer->nesting++;
+
+	/* Interrupts must see nesting incremented before we use the buffer */
+	barrier();
+	return &buffer->buffer[buffer->nesting][0];
 }
 
 static void put_trace_buf(void)
 {
+	/* Don't let the decrement of nesting leak before this */
+	barrier();
 	this_cpu_dec(trace_percpu_buffer->nesting);
 }
 
@@ -6220,7 +6229,7 @@ static int tracing_set_clock(struct trace_array *tr, const char *clockstr)
 	tracing_reset_online_cpus(&tr->trace_buffer);
 
 #ifdef CONFIG_TRACER_MAX_TRACE
-	if (tr->flags & TRACE_ARRAY_FL_GLOBAL && tr->max_buffer.buffer)
+	if (tr->max_buffer.buffer)
 		ring_buffer_set_clock(tr->max_buffer.buffer, trace_clocks[i].func);
 	tracing_reset_online_cpus(&tr->max_buffer);
 #endif

kernel/trace/trace.h

@@ -245,6 +245,7 @@ struct trace_array {
 	int			stop_count;
 	int			clock_id;
 	int			nr_topts;
+	bool			clear_trace;
 	struct tracer		*current_trace;
 	unsigned int		trace_flags;
 	unsigned char		trace_flags_index[TRACE_FLAGS_MAX_SIZE];

kernel/trace/trace_events.c

@@ -406,7 +406,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file,
 
 			if (file->flags & EVENT_FILE_FL_RECORDED_TGID) {
 				tracing_stop_tgid_record();
-				clear_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags);
+				clear_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags);
 			}
 
 			call->class->reg(call, TRACE_REG_UNREGISTER, file);
@@ -466,7 +466,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file,
 			set_bit(EVENT_FILE_FL_ENABLED_BIT, &file->flags);
 
 			/* WAS_ENABLED gets set but never cleared. */
-			call->flags |= TRACE_EVENT_FL_WAS_ENABLED;
+			set_bit(EVENT_FILE_FL_WAS_ENABLED_BIT, &file->flags);
 		}
 		break;
 	}
@@ -2058,6 +2058,10 @@ static void event_remove(struct trace_event_call *call)
 	do_for_each_event_file(tr, file) {
 		if (file->event_call != call)
 			continue;
+
+		if (file->flags & EVENT_FILE_FL_WAS_ENABLED)
+			tr->clear_trace = true;
+
 		ftrace_event_enable_disable(file, 0);
 		/*
 		 * The do_for_each_event_file() is
@@ -2396,15 +2400,11 @@ static void trace_module_add_events(struct module *mod)
 static void trace_module_remove_events(struct module *mod)
 {
 	struct trace_event_call *call, *p;
-	bool clear_trace = false;
 
 	down_write(&trace_event_sem);
 	list_for_each_entry_safe(call, p, &ftrace_events, list) {
-		if (call->mod == mod) {
-			if (call->flags & TRACE_EVENT_FL_WAS_ENABLED)
-				clear_trace = true;
+		if (call->mod == mod)
 			__trace_remove_event_call(call);
-		}
 	}
 	up_write(&trace_event_sem);
 
@@ -2416,8 +2416,7 @@ static void trace_module_remove_events(struct module *mod)
 	 * over from this module may be passed to the new module events and
 	 * unexpected results may occur.
 	 */
-	if (clear_trace)
-		tracing_reset_all_online_cpus();
+	tracing_reset_all_online_cpus();
 }
 
 static int trace_module_notify(struct notifier_block *self,

kernel/trace/trace_selftest.c

@@ -273,7 +273,7 @@ static int trace_selftest_ops(struct trace_array *tr, int cnt)
 		goto out_free;
 	if (cnt > 1) {
 		if (trace_selftest_test_global_cnt == 0)
-			goto out;
+			goto out_free;
 	}
 	if (trace_selftest_test_dyn_cnt == 0)
 		goto out_free;