Merge tag 'trace-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

Pull tracing updates from Steven Rostedt:
 "This includes two new updates for the ftrace infrastructure.

   - With the changing of the code for filtering events by pid, from a
     list of pids to a bitmask, we can now easily implement following
     forks.  With a new tracing option "event-fork" which, when set,
     will have tasks with pids in set_event_pid, when they fork, to have
     their child pids added to set_event_pid and the child will be
     traced as well.

     Note, if "event-fork" is set and a task with its pid in
     set_event_pid exits, its pid will be removed from set_event_pid

   - The addition of Tom Zanussi's hist triggers.  This includes a very
     thorough documentatino on how to use the hist triggers with events.
     This introduces a quick and easy way to get histogram data from
     events and their fields.

  Some other cleanups and updates were added as well.  Like Masami
  Hiramatsu added test cases for the event trigger and hist triggers.
  Also I added a speed up of filtering by using a temp buffer when
  filters are set"

* tag 'trace-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (45 commits)
  tracing: Use temp buffer when filtering events
  tracing: Remove TRACE_EVENT_FL_USE_CALL_FILTER logic
  tracing: Remove unused function trace_current_buffer_lock_reserve()
  tracing: Remove one use of trace_current_buffer_lock_reserve()
  tracing: Have trace_buffer_unlock_commit() call the _regs version with NULL
  tracing: Remove unused function trace_current_buffer_discard_commit()
  tracing: Move trace_buffer_unlock_commit{_regs}() to local header
  tracing: Fold filter_check_discard() into its only user
  tracing: Make filter_check_discard() local
  tracing: Move event_trigger_unlock_commit{_regs}() to local header
  tracing: Don't use the address of the buffer array name in copy_from_user
  tracing: Handle tracing_map_alloc_elts() error path correctly
  tracing: Add check for NULL event field when creating hist field
  tracing: checking for NULL instead of IS_ERR()
  tracing: Do not inherit event-fork option for instances
  tracing: Fix unsigned comparison to zero in hist trigger code
  kselftests/ftrace: Add a test for log2 modifier of hist trigger
  tracing: Add hist trigger 'log2' modifier
  kselftests/ftrace: Add hist trigger testcases
  kselftests/ftrace : Add event trigger testcases
  ...

Documentation/trace/ftrace.txt

@@ -210,6 +210,11 @@ of ftrace. Here is a list of some of the key files:
 	Note, sched_switch and sched_wake_up will also trace events
 	listed in this file.
 
+	To have the PIDs of children of tasks with their PID in this file
+	added on fork, enable the "event-fork" option. That option will also
+	cause the PIDs of tasks to be removed from this file when the task
+	exits.
+
   set_graph_function:
 
 	Set a "trigger" function where tracing should start
@@ -725,16 +730,14 @@ noraw
 nohex
 nobin
 noblock
-nostacktrace
 trace_printk
-noftrace_preempt
 nobranch
 annotate
 nouserstacktrace
 nosym-userobj
 noprintk-msg-only
 context-info
-latency-format
+nolatency-format
 sleep-time
 graph-time
 record-cmd
@@ -742,7 +745,10 @@ overwrite
 nodisable_on_free
 irq-info
 markers
+noevent-fork
 function-trace
+nodisplay-graph
+nostacktrace
 
 To disable one of the options, echo in the option prepended with
 "no".
@@ -796,11 +802,6 @@ Here are the available options:
 
   block - When set, reading trace_pipe will not block when polled.
 
-  stacktrace - This is one of the options that changes the trace
-	       itself. When a trace is recorded, so is the stack
-	       of functions. This allows for back traces of
-	       trace sites.
-
   trace_printk - Can disable trace_printk() from writing into the buffer.
 
   branch - Enable branch tracing with the tracer.
@@ -897,6 +898,10 @@ x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]
   	    When disabled, the trace_marker will error with EINVAL
 	    on write.
 
+  event-fork - When set, tasks with PIDs listed in set_event_pid will have
+	       the PIDs of their children added to set_event_pid when those
+	       tasks fork. Also, when tasks with PIDs in set_event_pid exit,
+	       their PIDs will be removed from the file.
 
   function-trace - The latency tracers will enable function tracing
   	    if this option is enabled (default it is). When
@@ -904,8 +909,17 @@ x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]
 	    functions. This keeps the overhead of the tracer down
 	    when performing latency tests.
 
- Note: Some tracers have their own options. They only appear
-       when the tracer is active.
+  display-graph - When set, the latency tracers (irqsoff, wakeup, etc) will
+	          use function graph tracing instead of function tracing.
+
+  stacktrace - This is one of the options that changes the trace
+	       itself. When a trace is recorded, so is the stack
+	       of functions. This allows for back traces of
+	       trace sites.
+
+ Note: Some tracers have their own options. They only appear in this
+       file when the tracer is active. They always appear in the
+       options directory.
 
 
 

include/linux/trace_events.h

@@ -154,21 +154,6 @@ trace_event_buffer_lock_reserve(struct ring_buffer **current_buffer,
 				struct trace_event_file *trace_file,
 				int type, unsigned long len,
 				unsigned long flags, int pc);
-struct ring_buffer_event *
-trace_current_buffer_lock_reserve(struct ring_buffer **current_buffer,
-				  int type, unsigned long len,
-				  unsigned long flags, int pc);
-void trace_buffer_unlock_commit(struct trace_array *tr,
-				struct ring_buffer *buffer,
-				struct ring_buffer_event *event,
-				unsigned long flags, int pc);
-void trace_buffer_unlock_commit_regs(struct trace_array *tr,
-				     struct ring_buffer *buffer,
-				     struct ring_buffer_event *event,
-				     unsigned long flags, int pc,
-				     struct pt_regs *regs);
-void trace_current_buffer_discard_commit(struct ring_buffer *buffer,
-					 struct ring_buffer_event *event);
 
 void tracing_record_cmdline(struct task_struct *tsk);
 
@@ -229,7 +214,6 @@ enum {
 	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	TRACE_EVENT_FL_IGNORE_ENABLE_BIT,
 	TRACE_EVENT_FL_WAS_ENABLED_BIT,
-	TRACE_EVENT_FL_USE_CALL_FILTER_BIT,
 	TRACE_EVENT_FL_TRACEPOINT_BIT,
 	TRACE_EVENT_FL_KPROBE_BIT,
 	TRACE_EVENT_FL_UPROBE_BIT,
@@ -244,7 +228,6 @@ enum {
  *  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).
- *  USE_CALL_FILTER - For trace internal events, don't use file filter
  *  TRACEPOINT    - Event is a tracepoint
  *  KPROBE        - Event is a kprobe
  *  UPROBE        - Event is a uprobe
@@ -255,7 +238,6 @@ enum {
 	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_USE_CALL_FILTER	= (1 << TRACE_EVENT_FL_USE_CALL_FILTER_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),
@@ -407,16 +389,12 @@ enum event_trigger_type {
 	ETT_SNAPSHOT		= (1 << 1),
 	ETT_STACKTRACE		= (1 << 2),
 	ETT_EVENT_ENABLE	= (1 << 3),
+	ETT_EVENT_HIST		= (1 << 4),
+	ETT_HIST_ENABLE		= (1 << 5),
 };
 
 extern int filter_match_preds(struct event_filter *filter, void *rec);
 
-extern int filter_check_discard(struct trace_event_file *file, void *rec,
-				struct ring_buffer *buffer,
-				struct ring_buffer_event *event);
-extern int call_filter_check_discard(struct trace_event_call *call, void *rec,
-				     struct ring_buffer *buffer,
-				     struct ring_buffer_event *event);
 extern enum event_trigger_type event_triggers_call(struct trace_event_file *file,
 						   void *rec);
 extern void event_triggers_post_call(struct trace_event_file *file,
@@ -450,100 +428,6 @@ trace_trigger_soft_disabled(struct trace_event_file *file)
 	return false;
 }
 
-/*
- * Helper function for event_trigger_unlock_commit{_regs}().
- * If there are event triggers attached to this event that requires
- * filtering against its fields, then they wil be called as the
- * entry already holds the field information of the current event.
- *
- * It also checks if the event should be discarded or not.
- * It is to be discarded if the event is soft disabled and the
- * event was only recorded to process triggers, or if the event
- * filter is active and this event did not match the filters.
- *
- * Returns true if the event is discarded, false otherwise.
- */
-static inline bool
-__event_trigger_test_discard(struct trace_event_file *file,
-			     struct ring_buffer *buffer,
-			     struct ring_buffer_event *event,
-			     void *entry,
-			     enum event_trigger_type *tt)
-{
-	unsigned long eflags = file->flags;
-
-	if (eflags & EVENT_FILE_FL_TRIGGER_COND)
-		*tt = event_triggers_call(file, entry);
-
-	if (test_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags))
-		ring_buffer_discard_commit(buffer, event);
-	else if (!filter_check_discard(file, entry, buffer, event))
-		return false;
-
-	return true;
-}
-
-/**
- * event_trigger_unlock_commit - handle triggers and finish event commit
- * @file: The file pointer assoctiated to the event
- * @buffer: The ring buffer that the event is being written to
- * @event: The event meta data in the ring buffer
- * @entry: The event itself
- * @irq_flags: The state of the interrupts at the start of the event
- * @pc: The state of the preempt count at the start of the event.
- *
- * This is a helper function to handle triggers that require data
- * from the event itself. It also tests the event against filters and
- * if the event is soft disabled and should be discarded.
- */
-static inline void
-event_trigger_unlock_commit(struct trace_event_file *file,
-			    struct ring_buffer *buffer,
-			    struct ring_buffer_event *event,
-			    void *entry, unsigned long irq_flags, int pc)
-{
-	enum event_trigger_type tt = ETT_NONE;
-
-	if (!__event_trigger_test_discard(file, buffer, event, entry, &tt))
-		trace_buffer_unlock_commit(file->tr, buffer, event, irq_flags, pc);
-
-	if (tt)
-		event_triggers_post_call(file, tt, entry);
-}
-
-/**
- * event_trigger_unlock_commit_regs - handle triggers and finish event commit
- * @file: The file pointer assoctiated to the event
- * @buffer: The ring buffer that the event is being written to
- * @event: The event meta data in the ring buffer
- * @entry: The event itself
- * @irq_flags: The state of the interrupts at the start of the event
- * @pc: The state of the preempt count at the start of the event.
- *
- * This is a helper function to handle triggers that require data
- * from the event itself. It also tests the event against filters and
- * if the event is soft disabled and should be discarded.
- *
- * Same as event_trigger_unlock_commit() but calls
- * trace_buffer_unlock_commit_regs() instead of trace_buffer_unlock_commit().
- */
-static inline void
-event_trigger_unlock_commit_regs(struct trace_event_file *file,
-				 struct ring_buffer *buffer,
-				 struct ring_buffer_event *event,
-				 void *entry, unsigned long irq_flags, int pc,
-				 struct pt_regs *regs)
-{
-	enum event_trigger_type tt = ETT_NONE;
-
-	if (!__event_trigger_test_discard(file, buffer, event, entry, &tt))
-		trace_buffer_unlock_commit_regs(file->tr, buffer, event,
-						irq_flags, pc, regs);
-
-	if (tt)
-		event_triggers_post_call(file, tt, entry);
-}
-
 #ifdef CONFIG_BPF_EVENTS
 unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx);
 #else

kernel/trace/Kconfig

@@ -528,6 +528,32 @@ config MMIOTRACE
 	  See Documentation/trace/mmiotrace.txt.
 	  If you are not helping to develop drivers, say N.
 
+config TRACING_MAP
+	bool
+	depends on ARCH_HAVE_NMI_SAFE_CMPXCHG
+	help
+	  tracing_map is a special-purpose lock-free map for tracing,
+	  separated out as a stand-alone facility in order to allow it
+	  to be shared between multiple tracers.  It isn't meant to be
+	  generally used outside of that context, and is normally
+	  selected by tracers that use it.
+
+config HIST_TRIGGERS
+	bool "Histogram triggers"
+	depends on ARCH_HAVE_NMI_SAFE_CMPXCHG
+	select TRACING_MAP
+	default n
+	help
+	  Hist triggers allow one or more arbitrary trace event fields
+	  to be aggregated into hash tables and dumped to stdout by
+	  reading a debugfs/tracefs file.  They're useful for
+	  gathering quick and dirty (though precise) summaries of
+	  event activity as an initial guide for further investigation
+	  using more advanced tools.
+
+	  See Documentation/trace/events.txt.
+	  If in doubt, say N.
+
 config MMIOTRACE_TEST
 	tristate "Test module for mmiotrace"
 	depends on MMIOTRACE && m

kernel/trace/Makefile

@@ -31,6 +31,7 @@ obj-$(CONFIG_TRACING) += trace_output.o
 obj-$(CONFIG_TRACING) += trace_seq.o
 obj-$(CONFIG_TRACING) += trace_stat.o
 obj-$(CONFIG_TRACING) += trace_printk.o
+obj-$(CONFIG_TRACING_MAP) += tracing_map.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
@@ -53,6 +54,7 @@ obj-$(CONFIG_EVENT_TRACING) += trace_event_perf.o
 endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_EVENT_TRACING) += trace_events_trigger.o
+obj-$(CONFIG_HIST_TRIGGERS) += trace_events_hist.o
 obj-$(CONFIG_BPF_EVENTS) += bpf_trace.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_TRACEPOINTS) += power-traces.o

kernel/trace/trace.c

@@ -253,6 +253,9 @@ unsigned long long ns2usecs(cycle_t nsec)
 #define TOP_LEVEL_TRACE_FLAGS (TRACE_ITER_PRINTK |			\
 	       TRACE_ITER_PRINTK_MSGONLY | TRACE_ITER_RECORD_CMD)
 
+/* trace_flags that are default zero for instances */
+#define ZEROED_TRACE_FLAGS \
+	TRACE_ITER_EVENT_FORK
 
 /*
  * The global_trace is the descriptor that holds the tracing
@@ -303,33 +306,18 @@ void trace_array_put(struct trace_array *this_tr)
 	mutex_unlock(&trace_types_lock);
 }
 
-int filter_check_discard(struct trace_event_file *file, void *rec,
-			 struct ring_buffer *buffer,
-			 struct ring_buffer_event *event)
-{
-	if (unlikely(file->flags & EVENT_FILE_FL_FILTERED) &&
-	    !filter_match_preds(file->filter, rec)) {
-		ring_buffer_discard_commit(buffer, event);
-		return 1;
-	}
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(filter_check_discard);
-
 int call_filter_check_discard(struct trace_event_call *call, void *rec,
 			      struct ring_buffer *buffer,
 			      struct ring_buffer_event *event)
 {
 	if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
 	    !filter_match_preds(call->filter, rec)) {
-		ring_buffer_discard_commit(buffer, event);
+		__trace_event_discard_commit(buffer, event);
 		return 1;
 	}
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(call_filter_check_discard);
 
 static cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu)
 {
@@ -1672,6 +1660,16 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
 }
 EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
 
+static __always_inline void
+trace_event_setup(struct ring_buffer_event *event,
+		  int type, unsigned long flags, int pc)
+{
+	struct trace_entry *ent = ring_buffer_event_data(event);
+
+	tracing_generic_entry_update(ent, flags, pc);
+	ent->type = type;
+}
+
 struct ring_buffer_event *
 trace_buffer_lock_reserve(struct ring_buffer *buffer,
 			  int type,
@@ -1681,34 +1679,137 @@ trace_buffer_lock_reserve(struct ring_buffer *buffer,
 	struct ring_buffer_event *event;
 
 	event = ring_buffer_lock_reserve(buffer, len);
-	if (event != NULL) {
-		struct trace_entry *ent = ring_buffer_event_data(event);
+	if (event != NULL)
+		trace_event_setup(event, type, flags, pc);
+
+	return event;
+}
+
+DEFINE_PER_CPU(struct ring_buffer_event *, trace_buffered_event);
+DEFINE_PER_CPU(int, trace_buffered_event_cnt);
+static int trace_buffered_event_ref;
+
+/**
+ * trace_buffered_event_enable - enable buffering events
+ *
+ * When events are being filtered, it is quicker to use a temporary
+ * buffer to write the event data into if there's a likely chance
+ * that it will not be committed. The discard of the ring buffer
+ * is not as fast as committing, and is much slower than copying
+ * a commit.
+ *
+ * When an event is to be filtered, allocate per cpu buffers to
+ * write the event data into, and if the event is filtered and discarded
+ * it is simply dropped, otherwise, the entire data is to be committed
+ * in one shot.
+ */
+void trace_buffered_event_enable(void)
+{
+	struct ring_buffer_event *event;
+	struct page *page;
+	int cpu;
 
-		tracing_generic_entry_update(ent, flags, pc);
-		ent->type = type;
+	WARN_ON_ONCE(!mutex_is_locked(&event_mutex));
+
+	if (trace_buffered_event_ref++)
+		return;
+
+	for_each_tracing_cpu(cpu) {
+		page = alloc_pages_node(cpu_to_node(cpu),
+					GFP_KERNEL | __GFP_NORETRY, 0);
+		if (!page)
+			goto failed;
+
+		event = page_address(page);
+		memset(event, 0, sizeof(*event));
+
+		per_cpu(trace_buffered_event, cpu) = event;
+
+		preempt_disable();
+		if (cpu == smp_processor_id() &&
+		    this_cpu_read(trace_buffered_event) !=
+		    per_cpu(trace_buffered_event, cpu))
+			WARN_ON_ONCE(1);
+		preempt_enable();
 	}
 
-	return event;
+	return;
+ failed:
+	trace_buffered_event_disable();
 }
 
-void
-__buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event)
+static void enable_trace_buffered_event(void *data)
 {
-	__this_cpu_write(trace_cmdline_save, true);
-	ring_buffer_unlock_commit(buffer, event);
+	/* Probably not needed, but do it anyway */
+	smp_rmb();
+	this_cpu_dec(trace_buffered_event_cnt);
 }
 
-void trace_buffer_unlock_commit(struct trace_array *tr,
-				struct ring_buffer *buffer,
-				struct ring_buffer_event *event,
-				unsigned long flags, int pc)
+static void disable_trace_buffered_event(void *data)
 {
-	__buffer_unlock_commit(buffer, event);
+	this_cpu_inc(trace_buffered_event_cnt);
+}
 
-	ftrace_trace_stack(tr, buffer, flags, 6, pc, NULL);
-	ftrace_trace_userstack(buffer, flags, pc);
+/**
+ * trace_buffered_event_disable - disable buffering events
+ *
+ * When a filter is removed, it is faster to not use the buffered
+ * events, and to commit directly into the ring buffer. Free up
+ * the temp buffers when there are no more users. This requires
+ * special synchronization with current events.
+ */
+void trace_buffered_event_disable(void)
+{
+	int cpu;
+
+	WARN_ON_ONCE(!mutex_is_locked(&event_mutex));
+
+	if (WARN_ON_ONCE(!trace_buffered_event_ref))
+		return;
+
+	if (--trace_buffered_event_ref)
+		return;
+
+	preempt_disable();
+	/* For each CPU, set the buffer as used. */
+	smp_call_function_many(tracing_buffer_mask,
+			       disable_trace_buffered_event, NULL, 1);
+	preempt_enable();
+
+	/* Wait for all current users to finish */
+	synchronize_sched();
+
+	for_each_tracing_cpu(cpu) {
+		free_page((unsigned long)per_cpu(trace_buffered_event, cpu));
+		per_cpu(trace_buffered_event, cpu) = NULL;
+	}
+	/*
+	 * Make sure trace_buffered_event is NULL before clearing
+	 * trace_buffered_event_cnt.
+	 */
+	smp_wmb();
+
+	preempt_disable();
+	/* Do the work on each cpu */
+	smp_call_function_many(tracing_buffer_mask,
+			       enable_trace_buffered_event, NULL, 1);
+	preempt_enable();
+}
+
+void
+__buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event)
+{
+	__this_cpu_write(trace_cmdline_save, true);
+
+	/* If this is the temp buffer, we need to commit fully */
+	if (this_cpu_read(trace_buffered_event) == event) {
+		/* Length is in event->array[0] */
+		ring_buffer_write(buffer, event->array[0], &event->array[1]);
+		/* Release the temp buffer */
+		this_cpu_dec(trace_buffered_event_cnt);
+	} else
+		ring_buffer_unlock_commit(buffer, event);
 }
-EXPORT_SYMBOL_GPL(trace_buffer_unlock_commit);
 
 static struct ring_buffer *temp_buffer;
 
@@ -1719,8 +1820,23 @@ trace_event_buffer_lock_reserve(struct ring_buffer **current_rb,
 			  unsigned long flags, int pc)
 {
 	struct ring_buffer_event *entry;
+	int val;
 
 	*current_rb = trace_file->tr->trace_buffer.buffer;
+
+	if ((trace_file->flags &
+	     (EVENT_FILE_FL_SOFT_DISABLED | EVENT_FILE_FL_FILTERED)) &&
+	    (entry = this_cpu_read(trace_buffered_event))) {
+		/* Try to use the per cpu buffer first */
+		val = this_cpu_inc_return(trace_buffered_event_cnt);
+		if (val == 1) {
+			trace_event_setup(entry, type, flags, pc);
+			entry->array[0] = len;
+			return entry;
+		}
+		this_cpu_dec(trace_buffered_event_cnt);
+	}
+
 	entry = trace_buffer_lock_reserve(*current_rb,
 					 type, len, flags, pc);
 	/*
@@ -1738,17 +1854,6 @@ trace_event_buffer_lock_reserve(struct ring_buffer **current_rb,
 }
 EXPORT_SYMBOL_GPL(trace_event_buffer_lock_reserve);
 
-struct ring_buffer_event *
-trace_current_buffer_lock_reserve(struct ring_buffer **current_rb,
-				  int type, unsigned long len,
-				  unsigned long flags, int pc)
-{
-	*current_rb = global_trace.trace_buffer.buffer;
-	return trace_buffer_lock_reserve(*current_rb,
-					 type, len, flags, pc);
-}
-EXPORT_SYMBOL_GPL(trace_current_buffer_lock_reserve);
-
 void trace_buffer_unlock_commit_regs(struct trace_array *tr,
 				     struct ring_buffer *buffer,
 				     struct ring_buffer_event *event,
@@ -1760,14 +1865,6 @@ void trace_buffer_unlock_commit_regs(struct trace_array *tr,
 	ftrace_trace_stack(tr, buffer, flags, 0, pc, regs);
 	ftrace_trace_userstack(buffer, flags, pc);
 }
-EXPORT_SYMBOL_GPL(trace_buffer_unlock_commit_regs);
-
-void trace_current_buffer_discard_commit(struct ring_buffer *buffer,
-					 struct ring_buffer_event *event)
-{
-	ring_buffer_discard_commit(buffer, event);
-}
-EXPORT_SYMBOL_GPL(trace_current_buffer_discard_commit);
 
 void
 trace_function(struct trace_array *tr,
@@ -3571,6 +3668,9 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled)
 	if (mask == TRACE_ITER_RECORD_CMD)
 		trace_event_enable_cmd_record(enabled);
 
+	if (mask == TRACE_ITER_EVENT_FORK)
+		trace_event_follow_fork(tr, enabled);
+
 	if (mask == TRACE_ITER_OVERWRITE) {
 		ring_buffer_change_overwrite(tr->trace_buffer.buffer, enabled);
 #ifdef CONFIG_TRACER_MAX_TRACE
@@ -3658,7 +3758,7 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf,
 	if (cnt >= sizeof(buf))
 		return -EINVAL;
 
-	if (copy_from_user(&buf, ubuf, cnt))
+	if (copy_from_user(buf, ubuf, cnt))
 		return -EFAULT;
 
 	buf[cnt] = 0;
@@ -3804,11 +3904,18 @@ static const char readme_msg[] =
 	"\t   trigger: traceon, traceoff\n"
 	"\t            enable_event:<system>:<event>\n"
 	"\t            disable_event:<system>:<event>\n"
+#ifdef CONFIG_HIST_TRIGGERS
+	"\t            enable_hist:<system>:<event>\n"
+	"\t            disable_hist:<system>:<event>\n"
+#endif
 #ifdef CONFIG_STACKTRACE
 	"\t\t    stacktrace\n"
 #endif
 #ifdef CONFIG_TRACER_SNAPSHOT
 	"\t\t    snapshot\n"
+#endif
+#ifdef CONFIG_HIST_TRIGGERS
+	"\t\t    hist (see below)\n"
 #endif
 	"\t   example: echo traceoff > events/block/block_unplug/trigger\n"
 	"\t            echo traceoff:3 > events/block/block_unplug/trigger\n"
@@ -3825,6 +3932,56 @@ static const char readme_msg[] =
 	"\t   To remove a trigger with a count:\n"
 	"\t     echo '!<trigger>:0 > <system>/<event>/trigger\n"
 	"\t   Filters can be ignored when removing a trigger.\n"
+#ifdef CONFIG_HIST_TRIGGERS
+	"      hist trigger\t- If set, event hits are aggregated into a hash table\n"
+	"\t    Format: hist:keys=<field1[,field2,...]>\n"
+	"\t            [:values=<field1[,field2,...]>]\n"
+	"\t            [:sort=<field1[,field2,...]>]\n"
+	"\t            [:size=#entries]\n"
+	"\t            [:pause][:continue][:clear]\n"
+	"\t            [:name=histname1]\n"
+	"\t            [if <filter>]\n\n"
+	"\t    When a matching event is hit, an entry is added to a hash\n"
+	"\t    table using the key(s) and value(s) named, and the value of a\n"
+	"\t    sum called 'hitcount' is incremented.  Keys and values\n"
+	"\t    correspond to fields in the event's format description.  Keys\n"
+	"\t    can be any field, or the special string 'stacktrace'.\n"
+	"\t    Compound keys consisting of up to two fields can be specified\n"
+	"\t    by the 'keys' keyword.  Values must correspond to numeric\n"
+	"\t    fields.  Sort keys consisting of up to two fields can be\n"
+	"\t    specified using the 'sort' keyword.  The sort direction can\n"
+	"\t    be modified by appending '.descending' or '.ascending' to a\n"
+	"\t    sort field.  The 'size' parameter can be used to specify more\n"
+	"\t    or fewer than the default 2048 entries for the hashtable size.\n"
+	"\t    If a hist trigger is given a name using the 'name' parameter,\n"
+	"\t    its histogram data will be shared with other triggers of the\n"
+	"\t    same name, and trigger hits will update this common data.\n\n"
+	"\t    Reading the 'hist' file for the event will dump the hash\n"
+	"\t    table in its entirety to stdout.  If there are multiple hist\n"
+	"\t    triggers attached to an event, there will be a table for each\n"
+	"\t    trigger in the output.  The table displayed for a named\n"
+	"\t    trigger will be the same as any other instance having the\n"
+	"\t    same name.  The default format used to display a given field\n"
+	"\t    can be modified by appending any of the following modifiers\n"
+	"\t    to the field name, as applicable:\n\n"
+	"\t            .hex        display a number as a hex value\n"
+	"\t            .sym        display an address as a symbol\n"
+	"\t            .sym-offset display an address as a symbol and offset\n"
+	"\t            .execname   display a common_pid as a program name\n"
+	"\t            .syscall    display a syscall id as a syscall name\n\n"
+	"\t            .log2       display log2 value rather than raw number\n\n"
+	"\t    The 'pause' parameter can be used to pause an existing hist\n"
+	"\t    trigger or to start a hist trigger but not log any events\n"
+	"\t    until told to do so.  'continue' can be used to start or\n"
+	"\t    restart a paused hist trigger.\n\n"
+	"\t    The 'clear' parameter will clear the contents of a running\n"
+	"\t    hist trigger and leave its current paused/active state\n"
+	"\t    unchanged.\n\n"
+	"\t    The enable_hist and disable_hist triggers can be used to\n"
+	"\t    have one event conditionally start and stop another event's\n"
+	"\t    already-attached hist trigger.  The syntax is analagous to\n"
+	"\t    the enable_event and disable_event triggers.\n"
+#endif
 ;
 
 static ssize_t
@@ -4474,7 +4631,7 @@ tracing_set_trace_write(struct file *filp, const char __user *ubuf,
 	if (cnt > MAX_TRACER_SIZE)
 		cnt = MAX_TRACER_SIZE;
 
-	if (copy_from_user(&buf, ubuf, cnt))
+	if (copy_from_user(buf, ubuf, cnt))
 		return -EFAULT;
 
 	buf[cnt] = 0;
@@ -5264,7 +5421,7 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf,
 	if (cnt >= sizeof(buf))
 		return -EINVAL;
 
-	if (copy_from_user(&buf, ubuf, cnt))
+	if (copy_from_user(buf, ubuf, cnt))
 		return -EFAULT;
 
 	buf[cnt] = 0;
@@ -6650,7 +6807,7 @@ static int instance_mkdir(const char *name)
 	if (!alloc_cpumask_var(&tr->tracing_cpumask, GFP_KERNEL))
 		goto out_free_tr;
 
-	tr->trace_flags = global_trace.trace_flags;
+	tr->trace_flags = global_trace.trace_flags & ~ZEROED_TRACE_FLAGS;
 
 	cpumask_copy(tr->tracing_cpumask, cpu_all_mask);
 
@@ -6724,6 +6881,12 @@ static int instance_rmdir(const char *name)
 
 	list_del(&tr->list);
 
+	/* Disable all the flags that were enabled coming in */
+	for (i = 0; i < TRACE_FLAGS_MAX_SIZE; i++) {
+		if ((1 << i) & ZEROED_TRACE_FLAGS)
+			set_tracer_flag(tr, 1 << i, 0);
+	}
+
 	tracing_set_nop(tr);
 	event_trace_del_tracer(tr);
 	ftrace_destroy_function_files(tr);

kernel/trace/trace.h

@@ -177,9 +177,8 @@ struct trace_options {
 };
 
 struct trace_pid_list {
-	unsigned int			nr_pids;
-	int				order;
-	pid_t				*pids;
+	int				pid_max;
+	unsigned long			*pids;
 };
 
 /*
@@ -656,6 +655,7 @@ static inline void __trace_stack(struct trace_array *tr, unsigned long flags,
 extern cycle_t ftrace_now(int cpu);
 
 extern void trace_find_cmdline(int pid, char comm[]);
+extern void trace_event_follow_fork(struct trace_array *tr, bool enable);
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 extern unsigned long ftrace_update_tot_cnt;
@@ -967,6 +967,7 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf,
 		C(STOP_ON_FREE,		"disable_on_free"),	\
 		C(IRQ_INFO,		"irq-info"),		\
 		C(MARKERS,		"markers"),		\
+		C(EVENT_FORK,		"event-fork"),		\
 		FUNCTION_FLAGS					\
 		FGRAPH_FLAGS					\
 		STACK_FLAGS					\
@@ -1064,6 +1065,137 @@ struct trace_subsystem_dir {
 	int				nr_events;
 };
 
+extern int call_filter_check_discard(struct trace_event_call *call, void *rec,
+				     struct ring_buffer *buffer,
+				     struct ring_buffer_event *event);
+
+void trace_buffer_unlock_commit_regs(struct trace_array *tr,
+				     struct ring_buffer *buffer,
+				     struct ring_buffer_event *event,
+				     unsigned long flags, int pc,
+				     struct pt_regs *regs);
+
+static inline void trace_buffer_unlock_commit(struct trace_array *tr,
+					      struct ring_buffer *buffer,
+					      struct ring_buffer_event *event,
+					      unsigned long flags, int pc)
+{
+	trace_buffer_unlock_commit_regs(tr, buffer, event, flags, pc, NULL);
+}
+
+DECLARE_PER_CPU(struct ring_buffer_event *, trace_buffered_event);
+DECLARE_PER_CPU(int, trace_buffered_event_cnt);
+void trace_buffered_event_disable(void);
+void trace_buffered_event_enable(void);
+
+static inline void
+__trace_event_discard_commit(struct ring_buffer *buffer,
+			     struct ring_buffer_event *event)
+{
+	if (this_cpu_read(trace_buffered_event) == event) {
+		/* Simply release the temp buffer */
+		this_cpu_dec(trace_buffered_event_cnt);
+		return;
+	}
+	ring_buffer_discard_commit(buffer, event);
+}
+
+/*
+ * Helper function for event_trigger_unlock_commit{_regs}().
+ * If there are event triggers attached to this event that requires
+ * filtering against its fields, then they wil be called as the
+ * entry already holds the field information of the current event.
+ *
+ * It also checks if the event should be discarded or not.
+ * It is to be discarded if the event is soft disabled and the
+ * event was only recorded to process triggers, or if the event
+ * filter is active and this event did not match the filters.
+ *
+ * Returns true if the event is discarded, false otherwise.
+ */
+static inline bool
+__event_trigger_test_discard(struct trace_event_file *file,
+			     struct ring_buffer *buffer,
+			     struct ring_buffer_event *event,
+			     void *entry,
+			     enum event_trigger_type *tt)
+{
+	unsigned long eflags = file->flags;
+
+	if (eflags & EVENT_FILE_FL_TRIGGER_COND)
+		*tt = event_triggers_call(file, entry);
+
+	if (test_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags) ||
+	    (unlikely(file->flags & EVENT_FILE_FL_FILTERED) &&
+	     !filter_match_preds(file->filter, entry))) {
+		__trace_event_discard_commit(buffer, event);
+		return true;
+	}
+
+	return false;
+}
+
+/**
+ * event_trigger_unlock_commit - handle triggers and finish event commit
+ * @file: The file pointer assoctiated to the event
+ * @buffer: The ring buffer that the event is being written to
+ * @event: The event meta data in the ring buffer
+ * @entry: The event itself
+ * @irq_flags: The state of the interrupts at the start of the event
+ * @pc: The state of the preempt count at the start of the event.
+ *
+ * This is a helper function to handle triggers that require data
+ * from the event itself. It also tests the event against filters and
+ * if the event is soft disabled and should be discarded.
+ */
+static inline void
+event_trigger_unlock_commit(struct trace_event_file *file,
+			    struct ring_buffer *buffer,
+			    struct ring_buffer_event *event,
+			    void *entry, unsigned long irq_flags, int pc)
+{
+	enum event_trigger_type tt = ETT_NONE;
+
+	if (!__event_trigger_test_discard(file, buffer, event, entry, &tt))
+		trace_buffer_unlock_commit(file->tr, buffer, event, irq_flags, pc);
+
+	if (tt)
+		event_triggers_post_call(file, tt, entry);
+}
+
+/**
+ * event_trigger_unlock_commit_regs - handle triggers and finish event commit
+ * @file: The file pointer assoctiated to the event
+ * @buffer: The ring buffer that the event is being written to
+ * @event: The event meta data in the ring buffer
+ * @entry: The event itself
+ * @irq_flags: The state of the interrupts at the start of the event
+ * @pc: The state of the preempt count at the start of the event.
+ *
+ * This is a helper function to handle triggers that require data
+ * from the event itself. It also tests the event against filters and
+ * if the event is soft disabled and should be discarded.
+ *
+ * Same as event_trigger_unlock_commit() but calls
+ * trace_buffer_unlock_commit_regs() instead of trace_buffer_unlock_commit().
+ */
+static inline void
+event_trigger_unlock_commit_regs(struct trace_event_file *file,
+				 struct ring_buffer *buffer,
+				 struct ring_buffer_event *event,
+				 void *entry, unsigned long irq_flags, int pc,
+				 struct pt_regs *regs)
+{
+	enum event_trigger_type tt = ETT_NONE;
+
+	if (!__event_trigger_test_discard(file, buffer, event, entry, &tt))
+		trace_buffer_unlock_commit_regs(file->tr, buffer, event,
+						irq_flags, pc, regs);
+
+	if (tt)
+		event_triggers_post_call(file, tt, entry);
+}
+
 #define FILTER_PRED_INVALID	((unsigned short)-1)
 #define FILTER_PRED_IS_RIGHT	(1 << 15)
 #define FILTER_PRED_FOLD	(1 << 15)
@@ -1161,6 +1293,15 @@ extern struct mutex event_mutex;
 extern struct list_head ftrace_events;
 
 extern const struct file_operations event_trigger_fops;
+extern const struct file_operations event_hist_fops;
+
+#ifdef CONFIG_HIST_TRIGGERS
+extern int register_trigger_hist_cmd(void);
+extern int register_trigger_hist_enable_disable_cmds(void);
+#else
+static inline int register_trigger_hist_cmd(void) { return 0; }
+static inline int register_trigger_hist_enable_disable_cmds(void) { return 0; }
+#endif
 
 extern int register_trigger_cmds(void);
 extern void clear_event_triggers(struct trace_array *tr);
@@ -1174,9 +1315,41 @@ struct event_trigger_data {
 	char				*filter_str;
 	void				*private_data;
 	bool				paused;
+	bool				paused_tmp;
 	struct list_head		list;
+	char				*name;
+	struct list_head		named_list;
+	struct event_trigger_data	*named_data;
+};
+
+/* Avoid typos */
+#define ENABLE_EVENT_STR	"enable_event"
+#define DISABLE_EVENT_STR	"disable_event"
+#define ENABLE_HIST_STR		"enable_hist"
+#define DISABLE_HIST_STR	"disable_hist"
+
+struct enable_trigger_data {
+	struct trace_event_file		*file;
+	bool				enable;
+	bool				hist;
 };
 
+extern int event_enable_trigger_print(struct seq_file *m,
+				      struct event_trigger_ops *ops,
+				      struct event_trigger_data *data);
+extern void event_enable_trigger_free(struct event_trigger_ops *ops,
+				      struct event_trigger_data *data);
+extern int event_enable_trigger_func(struct event_command *cmd_ops,
+				     struct trace_event_file *file,
+				     char *glob, char *cmd, char *param);
+extern int event_enable_register_trigger(char *glob,
+					 struct event_trigger_ops *ops,
+					 struct event_trigger_data *data,
+					 struct trace_event_file *file);
+extern void event_enable_unregister_trigger(char *glob,
+					    struct event_trigger_ops *ops,
+					    struct event_trigger_data *test,
+					    struct trace_event_file *file);
 extern void trigger_data_free(struct event_trigger_data *data);
 extern int event_trigger_init(struct event_trigger_ops *ops,
 			      struct event_trigger_data *data);
@@ -1189,7 +1362,18 @@ extern void unregister_trigger(char *glob, struct event_trigger_ops *ops,
 extern int set_trigger_filter(char *filter_str,
 			      struct event_trigger_data *trigger_data,
 			      struct trace_event_file *file);
+extern struct event_trigger_data *find_named_trigger(const char *name);
+extern bool is_named_trigger(struct event_trigger_data *test);
+extern int save_named_trigger(const char *name,
+			      struct event_trigger_data *data);
+extern void del_named_trigger(struct event_trigger_data *data);
+extern void pause_named_trigger(struct event_trigger_data *data);
+extern void unpause_named_trigger(struct event_trigger_data *data);
+extern void set_named_trigger_data(struct event_trigger_data *data,
+				   struct event_trigger_data *named_data);
 extern int register_event_command(struct event_command *cmd);
+extern int unregister_event_command(struct event_command *cmd);
+extern int register_trigger_hist_enable_disable_cmds(void);
 
 /**
  * struct event_trigger_ops - callbacks for trace event triggers

kernel/trace/trace_events.c

@@ -15,7 +15,7 @@
 #include <linux/kthread.h>
 #include <linux/tracefs.h>
 #include <linux/uaccess.h>
-#include <linux/bsearch.h>
+#include <linux/vmalloc.h>
 #include <linux/module.h>
 #include <linux/ctype.h>
 #include <linux/sort.h>
@@ -381,6 +381,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file,
 {
 	struct trace_event_call *call = file->event_call;
 	struct trace_array *tr = file->tr;
+	unsigned long file_flags = file->flags;
 	int ret = 0;
 	int disable;
 
@@ -463,6 +464,15 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file,
 		break;
 	}
 
+	/* Enable or disable use of trace_buffered_event */
+	if ((file_flags & EVENT_FILE_FL_SOFT_DISABLED) !=
+	    (file->flags & EVENT_FILE_FL_SOFT_DISABLED)) {
+		if (file->flags & EVENT_FILE_FL_SOFT_DISABLED)
+			trace_buffered_event_enable();
+		else
+			trace_buffered_event_disable();
+	}
+
 	return ret;
 }
 
@@ -489,24 +499,26 @@ static void ftrace_clear_events(struct trace_array *tr)
 	mutex_unlock(&event_mutex);
 }
 
-static int cmp_pid(const void *key, const void *elt)
+/* Shouldn't this be in a header? */
+extern int pid_max;
+
+/* Returns true if found in filter */
+static bool
+find_filtered_pid(struct trace_pid_list *filtered_pids, pid_t search_pid)
 {
-	const pid_t *search_pid = key;
-	const pid_t *pid = elt;
+	/*
+	 * If pid_max changed after filtered_pids was created, we
+	 * by default ignore all pids greater than the previous pid_max.
+	 */
+	if (search_pid >= filtered_pids->pid_max)
+		return false;
 
-	if (*search_pid == *pid)
-		return 0;
-	if (*search_pid < *pid)
-		return -1;
-	return 1;
+	return test_bit(search_pid, filtered_pids->pids);
 }
 
 static bool
-check_ignore_pid(struct trace_pid_list *filtered_pids, struct task_struct *task)
+ignore_this_task(struct trace_pid_list *filtered_pids, struct task_struct *task)
 {
-	pid_t search_pid;
-	pid_t *pid;
-
 	/*
 	 * Return false, because if filtered_pids does not exist,
 	 * all pids are good to trace.
@@ -514,15 +526,68 @@ check_ignore_pid(struct trace_pid_list *filtered_pids, struct task_struct *task)
 	if (!filtered_pids)
 		return false;
 
-	search_pid = task->pid;
+	return !find_filtered_pid(filtered_pids, task->pid);
+}
+
+static void filter_add_remove_task(struct trace_pid_list *pid_list,
+				   struct task_struct *self,
+				   struct task_struct *task)
+{
+	if (!pid_list)
+		return;
+
+	/* For forks, we only add if the forking task is listed */
+	if (self) {
+		if (!find_filtered_pid(pid_list, self->pid))
+			return;
+	}
 
-	pid = bsearch(&search_pid, filtered_pids->pids,
-		      filtered_pids->nr_pids, sizeof(pid_t),
-		      cmp_pid);
-	if (!pid)
-		return true;
+	/* Sorry, but we don't support pid_max changing after setting */
+	if (task->pid >= pid_list->pid_max)
+		return;
 
-	return false;
+	/* "self" is set for forks, and NULL for exits */
+	if (self)
+		set_bit(task->pid, pid_list->pids);
+	else
+		clear_bit(task->pid, pid_list->pids);
+}
+
+static void
+event_filter_pid_sched_process_exit(void *data, struct task_struct *task)
+{
+	struct trace_pid_list *pid_list;
+	struct trace_array *tr = data;
+
+	pid_list = rcu_dereference_sched(tr->filtered_pids);
+	filter_add_remove_task(pid_list, NULL, task);
+}
+
+static void
+event_filter_pid_sched_process_fork(void *data,
+				    struct task_struct *self,
+				    struct task_struct *task)
+{
+	struct trace_pid_list *pid_list;
+	struct trace_array *tr = data;
+
+	pid_list = rcu_dereference_sched(tr->filtered_pids);
+	filter_add_remove_task(pid_list, self, task);
+}
+
+void trace_event_follow_fork(struct trace_array *tr, bool enable)
+{
+	if (enable) {
+		register_trace_prio_sched_process_fork(event_filter_pid_sched_process_fork,
+						       tr, INT_MIN);
+		register_trace_prio_sched_process_exit(event_filter_pid_sched_process_exit,
+						       tr, INT_MAX);
+	} else {
+		unregister_trace_sched_process_fork(event_filter_pid_sched_process_fork,
+						    tr);
+		unregister_trace_sched_process_exit(event_filter_pid_sched_process_exit,
+						    tr);
+	}
 }
 
 static void
@@ -535,8 +600,8 @@ event_filter_pid_sched_switch_probe_pre(void *data, bool preempt,
 	pid_list = rcu_dereference_sched(tr->filtered_pids);
 
 	this_cpu_write(tr->trace_buffer.data->ignore_pid,
-		       check_ignore_pid(pid_list, prev) &&
-		       check_ignore_pid(pid_list, next));
+		       ignore_this_task(pid_list, prev) &&
+		       ignore_this_task(pid_list, next));
 }
 
 static void
@@ -549,7 +614,7 @@ event_filter_pid_sched_switch_probe_post(void *data, bool preempt,
 	pid_list = rcu_dereference_sched(tr->filtered_pids);
 
 	this_cpu_write(tr->trace_buffer.data->ignore_pid,
-		       check_ignore_pid(pid_list, next));
+		       ignore_this_task(pid_list, next));
 }
 
 static void
@@ -565,7 +630,7 @@ event_filter_pid_sched_wakeup_probe_pre(void *data, struct task_struct *task)
 	pid_list = rcu_dereference_sched(tr->filtered_pids);
 
 	this_cpu_write(tr->trace_buffer.data->ignore_pid,
-		       check_ignore_pid(pid_list, task));
+		       ignore_this_task(pid_list, task));
 }
 
 static void
@@ -582,7 +647,7 @@ event_filter_pid_sched_wakeup_probe_post(void *data, struct task_struct *task)
 
 	/* Set tracing if current is enabled */
 	this_cpu_write(tr->trace_buffer.data->ignore_pid,
-		       check_ignore_pid(pid_list, current));
+		       ignore_this_task(pid_list, current));
 }
 
 static void __ftrace_clear_event_pids(struct trace_array *tr)
@@ -620,7 +685,7 @@ static void __ftrace_clear_event_pids(struct trace_array *tr)
 	/* Wait till all users are no longer using pid filtering */
 	synchronize_sched();
 
-	free_pages((unsigned long)pid_list->pids, pid_list->order);
+	vfree(pid_list->pids);
 	kfree(pid_list);
 }
 
@@ -964,11 +1029,32 @@ static void t_stop(struct seq_file *m, void *p)
 	mutex_unlock(&event_mutex);
 }
 
+static void *
+p_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct trace_array *tr = m->private;
+	struct trace_pid_list *pid_list = rcu_dereference_sched(tr->filtered_pids);
+	unsigned long pid = (unsigned long)v;
+
+	(*pos)++;
+
+	/* pid already is +1 of the actual prevous bit */
+	pid = find_next_bit(pid_list->pids, pid_list->pid_max, pid);
+
+	/* Return pid + 1 to allow zero to be represented */
+	if (pid < pid_list->pid_max)
+		return (void *)(pid + 1);
+
+	return NULL;
+}
+
 static void *p_start(struct seq_file *m, loff_t *pos)
 	__acquires(RCU)
 {
 	struct trace_pid_list *pid_list;
 	struct trace_array *tr = m->private;
+	unsigned long pid;
+	loff_t l = 0;
 
 	/*
 	 * Grab the mutex, to keep calls to p_next() having the same
@@ -981,10 +1067,18 @@ static void *p_start(struct seq_file *m, loff_t *pos)
 
 	pid_list = rcu_dereference_sched(tr->filtered_pids);
 
-	if (!pid_list || *pos >= pid_list->nr_pids)
+	if (!pid_list)
+		return NULL;
+
+	pid = find_first_bit(pid_list->pids, pid_list->pid_max);
+	if (pid >= pid_list->pid_max)
 		return NULL;
 
-	return (void *)&pid_list->pids[*pos];
+	/* Return pid + 1 so that zero can be the exit value */
+	for (pid++; pid && l < *pos;
+	     pid = (unsigned long)p_next(m, (void *)pid, &l))
+		;
+	return (void *)pid;
 }
 
 static void p_stop(struct seq_file *m, void *p)
@@ -994,25 +1088,11 @@ static void p_stop(struct seq_file *m, void *p)
 	mutex_unlock(&event_mutex);
 }
 
-static void *
-p_next(struct seq_file *m, void *v, loff_t *pos)
-{
-	struct trace_array *tr = m->private;
-	struct trace_pid_list *pid_list = rcu_dereference_sched(tr->filtered_pids);
-
-	(*pos)++;
-
-	if (*pos >= pid_list->nr_pids)
-		return NULL;
-
-	return (void *)&pid_list->pids[*pos];
-}
-
 static int p_show(struct seq_file *m, void *v)
 {
-	pid_t *pid = v;
+	unsigned long pid = (unsigned long)v - 1;
 
-	seq_printf(m, "%d\n", *pid);
+	seq_printf(m, "%lu\n", pid);
 	return 0;
 }
 
@@ -1561,11 +1641,6 @@ show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos)
 	return r;
 }
 
-static int max_pids(struct trace_pid_list *pid_list)
-{
-	return (PAGE_SIZE << pid_list->order) / sizeof(pid_t);
-}
-
 static void ignore_task_cpu(void *data)
 {
 	struct trace_array *tr = data;
@@ -1579,7 +1654,7 @@ static void ignore_task_cpu(void *data)
 					     mutex_is_locked(&event_mutex));
 
 	this_cpu_write(tr->trace_buffer.data->ignore_pid,
-		       check_ignore_pid(pid_list, current));
+		       ignore_this_task(pid_list, current));
 }
 
 static ssize_t
@@ -1589,7 +1664,7 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 	struct seq_file *m = filp->private_data;
 	struct trace_array *tr = m->private;
 	struct trace_pid_list *filtered_pids = NULL;
-	struct trace_pid_list *pid_list = NULL;
+	struct trace_pid_list *pid_list;
 	struct trace_event_file *file;
 	struct trace_parser parser;
 	unsigned long val;
@@ -1597,7 +1672,7 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 	ssize_t read = 0;
 	ssize_t ret = 0;
 	pid_t pid;
-	int i;
+	int nr_pids = 0;
 
 	if (!cnt)
 		return 0;
@@ -1610,10 +1685,43 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 		return -ENOMEM;
 
 	mutex_lock(&event_mutex);
+	filtered_pids = rcu_dereference_protected(tr->filtered_pids,
+					     lockdep_is_held(&event_mutex));
+
 	/*
-	 * Load as many pids into the array before doing a
-	 * swap from the tr->filtered_pids to the new list.
+	 * Always recreate a new array. The write is an all or nothing
+	 * operation. Always create a new array when adding new pids by
+	 * the user. If the operation fails, then the current list is
+	 * not modified.
 	 */
+	pid_list = kmalloc(sizeof(*pid_list), GFP_KERNEL);
+	if (!pid_list) {
+		read = -ENOMEM;
+		goto out;
+	}
+	pid_list->pid_max = READ_ONCE(pid_max);
+	/* Only truncating will shrink pid_max */
+	if (filtered_pids && filtered_pids->pid_max > pid_list->pid_max)
+		pid_list->pid_max = filtered_pids->pid_max;
+	pid_list->pids = vzalloc((pid_list->pid_max + 7) >> 3);
+	if (!pid_list->pids) {
+		kfree(pid_list);
+		read = -ENOMEM;
+		goto out;
+	}
+	if (filtered_pids) {
+		/* copy the current bits to the new max */
+		pid = find_first_bit(filtered_pids->pids,
+				     filtered_pids->pid_max);
+		while (pid < filtered_pids->pid_max) {
+			set_bit(pid, pid_list->pids);
+			pid = find_next_bit(filtered_pids->pids,
+					    filtered_pids->pid_max,
+					    pid + 1);
+			nr_pids++;
+		}
+	}
+
 	while (cnt > 0) {
 
 		this_pos = 0;
@@ -1631,92 +1739,35 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 		ret = -EINVAL;
 		if (kstrtoul(parser.buffer, 0, &val))
 			break;
-		if (val > INT_MAX)
+		if (val >= pid_list->pid_max)
 			break;
 
 		pid = (pid_t)val;
 
-		ret = -ENOMEM;
-		if (!pid_list) {
-			pid_list = kmalloc(sizeof(*pid_list), GFP_KERNEL);
-			if (!pid_list)
-				break;
+		set_bit(pid, pid_list->pids);
+		nr_pids++;
 
-			filtered_pids = rcu_dereference_protected(tr->filtered_pids,
-							lockdep_is_held(&event_mutex));
-			if (filtered_pids)
-				pid_list->order = filtered_pids->order;
-			else
-				pid_list->order = 0;
-
-			pid_list->pids = (void *)__get_free_pages(GFP_KERNEL,
-								  pid_list->order);
-			if (!pid_list->pids)
-				break;
-
-			if (filtered_pids) {
-				pid_list->nr_pids = filtered_pids->nr_pids;
-				memcpy(pid_list->pids, filtered_pids->pids,
-				       pid_list->nr_pids * sizeof(pid_t));
-			} else
-				pid_list->nr_pids = 0;
-		}
-
-		if (pid_list->nr_pids >= max_pids(pid_list)) {
-			pid_t *pid_page;
-
-			pid_page = (void *)__get_free_pages(GFP_KERNEL,
-							    pid_list->order + 1);
-			if (!pid_page)
-				break;
-			memcpy(pid_page, pid_list->pids,
-			       pid_list->nr_pids * sizeof(pid_t));
-			free_pages((unsigned long)pid_list->pids, pid_list->order);
-
-			pid_list->order++;
-			pid_list->pids = pid_page;
-		}
-
-		pid_list->pids[pid_list->nr_pids++] = pid;
 		trace_parser_clear(&parser);
 		ret = 0;
 	}
 	trace_parser_put(&parser);
 
 	if (ret < 0) {
-		if (pid_list)
-			free_pages((unsigned long)pid_list->pids, pid_list->order);
+		vfree(pid_list->pids);
 		kfree(pid_list);
-		mutex_unlock(&event_mutex);
-		return ret;
-	}
-
-	if (!pid_list) {
-		mutex_unlock(&event_mutex);
-		return ret;
+		read = ret;
+		goto out;
 	}
 
-	sort(pid_list->pids, pid_list->nr_pids, sizeof(pid_t), cmp_pid, NULL);
-
-	/* Remove duplicates */
-	for (i = 1; i < pid_list->nr_pids; i++) {
-		int start = i;
-
-		while (i < pid_list->nr_pids &&
-		       pid_list->pids[i - 1] == pid_list->pids[i])
-			i++;
-
-		if (start != i) {
-			if (i < pid_list->nr_pids) {
-				memmove(&pid_list->pids[start], &pid_list->pids[i],
-					(pid_list->nr_pids - i) * sizeof(pid_t));
-				pid_list->nr_pids -= i - start;
-				i = start;
-			} else
-				pid_list->nr_pids = start;
-		}
+	if (!nr_pids) {
+		/* Cleared the list of pids */
+		vfree(pid_list->pids);
+		kfree(pid_list);
+		read = ret;
+		if (!filtered_pids)
+			goto out;
+		pid_list = NULL;
 	}
-
 	rcu_assign_pointer(tr->filtered_pids, pid_list);
 
 	list_for_each_entry(file, &tr->events, list) {
@@ -1726,7 +1777,7 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 	if (filtered_pids) {
 		synchronize_sched();
 
-		free_pages((unsigned long)filtered_pids->pids, filtered_pids->order);
+		vfree(filtered_pids->pids);
 		kfree(filtered_pids);
 	} else {
 		/*
@@ -1763,10 +1814,12 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 	 */
 	on_each_cpu(ignore_task_cpu, tr, 1);
 
+ out:
 	mutex_unlock(&event_mutex);
 
 	ret = read;
-	*ppos += read;
+	if (read > 0)
+		*ppos += read;
 
 	return ret;
 }
@@ -2121,6 +2174,10 @@ event_create_dir(struct dentry *parent, struct trace_event_file *file)
 		trace_create_file("trigger", 0644, file->dir, file,
 				  &event_trigger_fops);
 
+#ifdef CONFIG_HIST_TRIGGERS
+	trace_create_file("hist", 0444, file->dir, file,
+			  &event_hist_fops);
+#endif
 	trace_create_file("format", 0444, file->dir, call,
 			  &ftrace_event_format_fops);
 
@@ -3368,7 +3425,7 @@ static __init void event_trace_self_tests(void)
 
 static DEFINE_PER_CPU(atomic_t, ftrace_test_event_disable);
 
-static struct trace_array *event_tr;
+static struct trace_event_file event_trace_file __initdata;
 
 static void __init
 function_test_events_call(unsigned long ip, unsigned long parent_ip,
@@ -3392,17 +3449,17 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip,
 
 	local_save_flags(flags);
 
-	event = trace_current_buffer_lock_reserve(&buffer,
-						  TRACE_FN, sizeof(*entry),
-						  flags, pc);
+	event = trace_event_buffer_lock_reserve(&buffer, &event_trace_file,
+						TRACE_FN, sizeof(*entry),
+						flags, pc);
 	if (!event)
 		goto out;
 	entry	= ring_buffer_event_data(event);
 	entry->ip			= ip;
 	entry->parent_ip		= parent_ip;
 
-	trace_buffer_unlock_commit(event_tr, buffer, event, flags, pc);
-
+	event_trigger_unlock_commit(&event_trace_file, buffer, event,
+				    entry, flags, pc);
  out:
 	atomic_dec(&per_cpu(ftrace_test_event_disable, cpu));
 	preempt_enable_notrace();
@@ -3417,9 +3474,11 @@ static struct ftrace_ops trace_ops __initdata  =
 static __init void event_trace_self_test_with_function(void)
 {
 	int ret;
-	event_tr = top_trace_array();
-	if (WARN_ON(!event_tr))
+
+	event_trace_file.tr = top_trace_array();
+	if (WARN_ON(!event_trace_file.tr))
 		return;
+
 	ret = register_ftrace_function(&trace_ops);
 	if (WARN_ON(ret < 0)) {
 		pr_info("Failed to enable function tracer for event tests\n");

kernel/trace/trace_events_filter.c

@@ -689,10 +689,7 @@ static void append_filter_err(struct filter_parse_state *ps,
 
 static inline struct event_filter *event_filter(struct trace_event_file *file)
 {
-	if (file->event_call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		return file->event_call->filter;
-	else
-		return file->filter;
+	return file->filter;
 }
 
 /* caller must hold event_mutex */
@@ -826,12 +823,12 @@ static void __free_preds(struct event_filter *filter)
 
 static void filter_disable(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
+	unsigned long old_flags = file->flags;
 
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		call->flags &= ~TRACE_EVENT_FL_FILTERED;
-	else
-		file->flags &= ~EVENT_FILE_FL_FILTERED;
+	file->flags &= ~EVENT_FILE_FL_FILTERED;
+
+	if (old_flags != file->flags)
+		trace_buffered_event_disable();
 }
 
 static void __free_filter(struct event_filter *filter)
@@ -883,13 +880,8 @@ static int __alloc_preds(struct event_filter *filter, int n_preds)
 
 static inline void __remove_filter(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
 	filter_disable(file);
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		remove_filter_string(call->filter);
-	else
-		remove_filter_string(file->filter);
+	remove_filter_string(file->filter);
 }
 
 static void filter_free_subsystem_preds(struct trace_subsystem_dir *dir,
@@ -906,15 +898,8 @@ static void filter_free_subsystem_preds(struct trace_subsystem_dir *dir,
 
 static inline void __free_subsystem_filter(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) {
-		__free_filter(call->filter);
-		call->filter = NULL;
-	} else {
-		__free_filter(file->filter);
-		file->filter = NULL;
-	}
+	__free_filter(file->filter);
+	file->filter = NULL;
 }
 
 static void filter_free_subsystem_filters(struct trace_subsystem_dir *dir,
@@ -1718,69 +1703,43 @@ static int replace_preds(struct trace_event_call *call,
 
 static inline void event_set_filtered_flag(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
+	unsigned long old_flags = file->flags;
 
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		call->flags |= TRACE_EVENT_FL_FILTERED;
-	else
-		file->flags |= EVENT_FILE_FL_FILTERED;
+	file->flags |= EVENT_FILE_FL_FILTERED;
+
+	if (old_flags != file->flags)
+		trace_buffered_event_enable();
 }
 
 static inline void event_set_filter(struct trace_event_file *file,
 				    struct event_filter *filter)
 {
-	struct trace_event_call *call = file->event_call;
-
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		rcu_assign_pointer(call->filter, filter);
-	else
-		rcu_assign_pointer(file->filter, filter);
+	rcu_assign_pointer(file->filter, filter);
 }
 
 static inline void event_clear_filter(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		RCU_INIT_POINTER(call->filter, NULL);
-	else
-		RCU_INIT_POINTER(file->filter, NULL);
+	RCU_INIT_POINTER(file->filter, NULL);
 }
 
 static inline void
 event_set_no_set_filter_flag(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
-	else
-		file->flags |= EVENT_FILE_FL_NO_SET_FILTER;
+	file->flags |= EVENT_FILE_FL_NO_SET_FILTER;
 }
 
 static inline void
 event_clear_no_set_filter_flag(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
-	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
-		call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
-	else
-		file->flags &= ~EVENT_FILE_FL_NO_SET_FILTER;
+	file->flags &= ~EVENT_FILE_FL_NO_SET_FILTER;
 }
 
 static inline bool
 event_no_set_filter_flag(struct trace_event_file *file)
 {
-	struct trace_event_call *call = file->event_call;
-
 	if (file->flags & EVENT_FILE_FL_NO_SET_FILTER)
 		return true;
 
-	if ((call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) &&
-	    (call->flags & TRACE_EVENT_FL_NO_SET_FILTER))
-		return true;
-
 	return false;
 }
 

tools/testing/selftests/ftrace/test.d/functions

@@ -14,3 +14,12 @@ enable_tracing() { # start trace recording
 reset_tracer() { # reset the current tracer
     echo nop > current_tracer
 }
+
+reset_trigger() { # reset all current setting triggers
+    grep -v ^# events/*/*/trigger |
+    while read line; do
+        cmd=`echo $line | cut -f2- -d: | cut -f1 -d" "`
+	echo "!$cmd" > `echo $line | cut -f1 -d:`
+    done
+}
+

tools/testing/selftests/ftrace/test.d/trigger/trigger-eventonoff.tc

+#!/bin/sh
+# description: event trigger - test event enable/disable trigger
+
+do_reset() {
+    reset_trigger
+    echo > set_event
+    clear_trace
+}
+
+fail() { #msg
+    do_reset
+    echo $1
+    exit $FAIL
+}
+
+if [ ! -f set_event -o ! -d events/sched ]; then
+    echo "event tracing is not supported"
+    exit_unsupported
+fi
+
+if [ ! -f events/sched/sched_process_fork/trigger ]; then
+    echo "event trigger is not supported"
+    exit_unsupported
+fi
+
+reset_tracer
+do_reset
+
+FEATURE=`grep enable_event events/sched/sched_process_fork/trigger`
+if [ -z "$FEATURE" ]; then
+    echo "event enable/disable trigger is not supported"
+    exit_unsupported
+fi
+
+echo "Test enable_event trigger"
+echo 0 > events/sched/sched_switch/enable
+echo 'enable_event:sched:sched_switch' > events/sched/sched_process_fork/trigger
+( echo "forked")
+if [ `cat events/sched/sched_switch/enable` != '1*' ]; then
+    fail "enable_event trigger on sched_process_fork did not work"
+fi
+
+reset_trigger
+
+echo "Test disable_event trigger"
+echo 1 > events/sched/sched_switch/enable
+echo 'disable_event:sched:sched_switch' > events/sched/sched_process_fork/trigger
+( echo "forked")
+if [ `cat events/sched/sched_switch/enable` != '0*' ]; then
+    fail "disable_event trigger on sched_process_fork did not work"
+fi
+
+reset_trigger
+
+echo "Test semantic error for event enable/disable trigger"
+! echo 'enable_event:nogroup:noevent' > events/sched/sched_process_fork/trigger
+! echo 'disable_event+1' > events/sched/sched_process_fork/trigger
+echo 'enable_event:sched:sched_switch' > events/sched/sched_process_fork/trigger
+! echo 'enable_event:sched:sched_switch' > events/sched/sched_process_fork/trigger
+! echo 'disable_event:sched:sched_switch' > events/sched/sched_process_fork/trigger
+
+do_reset
+
+exit 0