diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index bae51ddfabd357c0c06a614b72b08aa5da5c9894..4b58cf1a11c2cee20ce44de69e77cda056e6d4c8 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -3,12 +3,23 @@
 
 #include <linux/trace_seq.h>
 #include <linux/ring_buffer.h>
-
+#include <linux/percpu.h>
 
 struct trace_array;
 struct tracer;
 struct dentry;
 
+DECLARE_PER_CPU(struct trace_seq, ftrace_event_seq);
+
+struct trace_print_flags {
+	unsigned long		mask;
+	const char		*name;
+};
+
+const char *ftrace_print_flags_seq(struct trace_seq *p, const char *delim,
+				   unsigned long flags,
+				   const struct trace_print_flags *flag_array);
+
 /*
  * The trace entry - the most basic unit of tracing. This is what
  * is printed in the end as a single line in the trace output, such as:
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index b5ff2e8229eced3e11edb0f6a8702ba83bccda68..22c94719c5695cf782c5ddbdc68d5533aba4ac9c 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -87,6 +87,7 @@
  *	struct trace_seq *s = &iter->seq;
  *	struct ftrace_raw_<call> *field; <-- defined in stage 1
  *	struct trace_entry *entry;
+ *	struct trace_seq *p;
  *	int ret;
  *
  *	entry = iter->ent;
@@ -98,7 +99,9 @@
  *
  *	field = (typeof(field))entry;
  *
+ *	p = get_cpu_var(ftrace_event_seq);
  *	ret = trace_seq_printf(s, <TP_printk> "\n");
+ *	put_cpu();
  *	if (!ret)
  *		return TRACE_TYPE_PARTIAL_LINE;
  *
@@ -119,6 +122,14 @@
 #undef __get_str
 #define __get_str(field)	((char *)__entry + __entry->__str_loc_##field)
 
+#undef __print_flags
+#define __print_flags(flag, delim, flag_array...)			\
+	({								\
+		static const struct trace_print_flags flags[] =		\
+			{ flag_array, { -1, NULL }};			\
+		ftrace_print_flags_seq(p, delim, flag, flags);		\
+	})
+
 #undef TRACE_EVENT
 #define TRACE_EVENT(call, proto, args, tstruct, assign, print)		\
 enum print_line_t							\
@@ -127,6 +138,7 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags)	\
 	struct trace_seq *s = &iter->seq;				\
 	struct ftrace_raw_##call *field;				\
 	struct trace_entry *entry;					\
+	struct trace_seq *p;						\
 	int ret;							\
 									\
 	entry = iter->ent;						\
@@ -138,7 +150,9 @@ ftrace_raw_output_##call(struct trace_iterator *iter, int flags)	\
 									\
 	field = (typeof(field))entry;					\
 									\
+	p = &get_cpu_var(ftrace_event_seq);				\
 	ret = trace_seq_printf(s, #call ": " print);			\
+	put_cpu();							\
 	if (!ret)							\
 		return TRACE_TYPE_PARTIAL_LINE;				\
 									\
diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
index 7136420603aa01ff997146509e9f2ec597fcd2c3..a4840c260c8919452590b643fe704368e32178a3 100644
--- a/kernel/trace/trace_output.c
+++ b/kernel/trace/trace_output.c
@@ -15,6 +15,9 @@
 #define EVENT_HASHSIZE	128
 
 static DECLARE_RWSEM(trace_event_mutex);
+
+DEFINE_PER_CPU(struct trace_seq, ftrace_event_seq);
+
 static struct hlist_head event_hash[EVENT_HASHSIZE] __read_mostly;
 
 static int next_event_type = __TRACE_LAST_TYPE + 1;
@@ -212,6 +215,42 @@ int trace_seq_path(struct trace_seq *s, struct path *path)
 	return 0;
 }
 
+const char *
+ftrace_print_flags_seq(struct trace_seq *p, const char *delim,
+		       unsigned long flags,
+		       const struct trace_print_flags *flag_array)
+{
+	unsigned long mask;
+	const char *str;
+	int i;
+
+	trace_seq_init(p);
+
+	for (i = 0;  flag_array[i].name && flags; i++) {
+
+		mask = flag_array[i].mask;
+		if ((flags & mask) != mask)
+			continue;
+
+		str = flag_array[i].name;
+		flags &= ~mask;
+		if (p->len && delim)
+			trace_seq_puts(p, delim);
+		trace_seq_puts(p, str);
+	}
+
+	/* check for left over flags */
+	if (flags) {
+		if (p->len && delim)
+			trace_seq_puts(p, delim);
+		trace_seq_printf(p, "0x%lx", flags);
+	}
+
+	trace_seq_putc(p, 0);
+
+	return p->buffer;
+}
+
 #ifdef CONFIG_KRETPROBES
 static inline const char *kretprobed(const char *name)
 {