Linux-libre 4.3.2-gnu

1 files changed, 402 insertions, 362 deletions

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 6260717c1..fc347f8b1 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -400,6 +400,17 @@ struct rb_irq_work {
 };
 
 /*
+ * Structure to hold event state and handle nested events.
+ */
+struct rb_event_info {
+	u64			ts;
+	u64			delta;
+	unsigned long		length;
+	struct buffer_page	*tail_page;
+	int			add_timestamp;
+};
+
+/*
  * Used for which event context the event is in.
  *  NMI     = 0
  *  IRQ     = 1
@@ -1876,73 +1887,6 @@ rb_event_index(struct ring_buffer_event *event)
 	return (addr & ~PAGE_MASK) - BUF_PAGE_HDR_SIZE;
 }
 
-static inline int
-rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer,
-		   struct ring_buffer_event *event)
-{
-	unsigned long addr = (unsigned long)event;
-	unsigned long index;
-
-	index = rb_event_index(event);
-	addr &= PAGE_MASK;
-
-	return cpu_buffer->commit_page->page == (void *)addr &&
-		rb_commit_index(cpu_buffer) == index;
-}
-
-static void
-rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer)
-{
-	unsigned long max_count;
-
-	/*
-	 * We only race with interrupts and NMIs on this CPU.
-	 * If we own the commit event, then we can commit
-	 * all others that interrupted us, since the interruptions
-	 * are in stack format (they finish before they come
-	 * back to us). This allows us to do a simple loop to
-	 * assign the commit to the tail.
-	 */
- again:
-	max_count = cpu_buffer->nr_pages * 100;
-
-	while (cpu_buffer->commit_page != cpu_buffer->tail_page) {
-		if (RB_WARN_ON(cpu_buffer, !(--max_count)))
-			return;
-		if (RB_WARN_ON(cpu_buffer,
-			       rb_is_reader_page(cpu_buffer->tail_page)))
-			return;
-		local_set(&cpu_buffer->commit_page->page->commit,
-			  rb_page_write(cpu_buffer->commit_page));
-		rb_inc_page(cpu_buffer, &cpu_buffer->commit_page);
-		cpu_buffer->write_stamp =
-			cpu_buffer->commit_page->page->time_stamp;
-		/* add barrier to keep gcc from optimizing too much */
-		barrier();
-	}
-	while (rb_commit_index(cpu_buffer) !=
-	       rb_page_write(cpu_buffer->commit_page)) {
-
-		local_set(&cpu_buffer->commit_page->page->commit,
-			  rb_page_write(cpu_buffer->commit_page));
-		RB_WARN_ON(cpu_buffer,
-			   local_read(&cpu_buffer->commit_page->page->commit) &
-			   ~RB_WRITE_MASK);
-		barrier();
-	}
-
-	/* again, keep gcc from optimizing */
-	barrier();
-
-	/*
-	 * If an interrupt came in just after the first while loop
-	 * and pushed the tail page forward, we will be left with
-	 * a dangling commit that will never go forward.
-	 */
-	if (unlikely(cpu_buffer->commit_page != cpu_buffer->tail_page))
-		goto again;
-}
-
 static void rb_reset_reader_page(struct ring_buffer_per_cpu *cpu_buffer)
 {
 	cpu_buffer->read_stamp = cpu_buffer->reader_page->page->time_stamp;
@@ -1968,64 +1912,6 @@ static void rb_inc_iter(struct ring_buffer_iter *iter)
 	iter->head = 0;
 }
 
-/* Slow path, do not inline */
-static noinline struct ring_buffer_event *
-rb_add_time_stamp(struct ring_buffer_event *event, u64 delta)
-{
-	event->type_len = RINGBUF_TYPE_TIME_EXTEND;
-
-	/* Not the first event on the page? */
-	if (rb_event_index(event)) {
-		event->time_delta = delta & TS_MASK;
-		event->array[0] = delta >> TS_SHIFT;
-	} else {
-		/* nope, just zero it */
-		event->time_delta = 0;
-		event->array[0] = 0;
-	}
-
-	return skip_time_extend(event);
-}
-
-/**
- * rb_update_event - update event type and data
- * @event: the event to update
- * @type: the type of event
- * @length: the size of the event field in the ring buffer
- *
- * Update the type and data fields of the event. The length
- * is the actual size that is written to the ring buffer,
- * and with this, we can determine what to place into the
- * data field.
- */
-static void
-rb_update_event(struct ring_buffer_per_cpu *cpu_buffer,
-		struct ring_buffer_event *event, unsigned length,
-		int add_timestamp, u64 delta)
-{
-	/* Only a commit updates the timestamp */
-	if (unlikely(!rb_event_is_commit(cpu_buffer, event)))
-		delta = 0;
-
-	/*
-	 * If we need to add a timestamp, then we
-	 * add it to the start of the resevered space.
-	 */
-	if (unlikely(add_timestamp)) {
-		event = rb_add_time_stamp(event, delta);
-		length -= RB_LEN_TIME_EXTEND;
-		delta = 0;
-	}
-
-	event->time_delta = delta;
-	length -= RB_EVNT_HDR_SIZE;
-	if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT) {
-		event->type_len = 0;
-		event->array[0] = length;
-	} else
-		event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT);
-}
-
 /*
  * rb_handle_head_page - writer hit the head page
  *
@@ -2184,29 +2070,13 @@ rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer,
 	return 0;
 }
 
-static unsigned rb_calculate_event_length(unsigned length)
-{
-	struct ring_buffer_event event; /* Used only for sizeof array */
-
-	/* zero length can cause confusions */
-	if (!length)
-		length++;
-
-	if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT)
-		length += sizeof(event.array[0]);
-
-	length += RB_EVNT_HDR_SIZE;
-	length = ALIGN(length, RB_ARCH_ALIGNMENT);
-
-	return length;
-}
-
 static inline void
 rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
-	      struct buffer_page *tail_page,
-	      unsigned long tail, unsigned long length)
+	      unsigned long tail, struct rb_event_info *info)
 {
+	struct buffer_page *tail_page = info->tail_page;
 	struct ring_buffer_event *event;
+	unsigned long length = info->length;
 
 	/*
 	 * Only the event that crossed the page boundary
@@ -2276,13 +2146,14 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
  */
 static noinline struct ring_buffer_event *
 rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer,
-	     unsigned long length, unsigned long tail,
-	     struct buffer_page *tail_page, u64 ts)
+	     unsigned long tail, struct rb_event_info *info)
 {
+	struct buffer_page *tail_page = info->tail_page;
 	struct buffer_page *commit_page = cpu_buffer->commit_page;
 	struct ring_buffer *buffer = cpu_buffer->buffer;
 	struct buffer_page *next_page;
 	int ret;
+	u64 ts;
 
 	next_page = tail_page;
 
@@ -2368,74 +2239,120 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer,
 
  out_again:
 
-	rb_reset_tail(cpu_buffer, tail_page, tail, length);
+	rb_reset_tail(cpu_buffer, tail, info);
 
 	/* fail and let the caller try again */
 	return ERR_PTR(-EAGAIN);
 
  out_reset:
 	/* reset write */
-	rb_reset_tail(cpu_buffer, tail_page, tail, length);
+	rb_reset_tail(cpu_buffer, tail, info);
 
 	return NULL;
 }
 
-static struct ring_buffer_event *
-__rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer,
-		  unsigned long length, u64 ts,
-		  u64 delta, int add_timestamp)
+/* Slow path, do not inline */
+static noinline struct ring_buffer_event *
+rb_add_time_stamp(struct ring_buffer_event *event, u64 delta)
 {
-	struct buffer_page *tail_page;
-	struct ring_buffer_event *event;
-	unsigned long tail, write;
+	event->type_len = RINGBUF_TYPE_TIME_EXTEND;
 
-	/*
-	 * If the time delta since the last event is too big to
-	 * hold in the time field of the event, then we append a
-	 * TIME EXTEND event ahead of the data event.
-	 */
-	if (unlikely(add_timestamp))
-		length += RB_LEN_TIME_EXTEND;
+	/* Not the first event on the page? */
+	if (rb_event_index(event)) {
+		event->time_delta = delta & TS_MASK;
+		event->array[0] = delta >> TS_SHIFT;
+	} else {
+		/* nope, just zero it */
+		event->time_delta = 0;
+		event->array[0] = 0;
+	}
 
-	tail_page = cpu_buffer->tail_page;
-	write = local_add_return(length, &tail_page->write);
+	return skip_time_extend(event);
+}
 
-	/* set write to only the index of the write */
-	write &= RB_WRITE_MASK;
-	tail = write - length;
+static inline int rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer,
+				     struct ring_buffer_event *event);
+
+/**
+ * rb_update_event - update event type and data
+ * @event: the event to update
+ * @type: the type of event
+ * @length: the size of the event field in the ring buffer
+ *
+ * Update the type and data fields of the event. The length
+ * is the actual size that is written to the ring buffer,
+ * and with this, we can determine what to place into the
+ * data field.
+ */
+static void
+rb_update_event(struct ring_buffer_per_cpu *cpu_buffer,
+		struct ring_buffer_event *event,
+		struct rb_event_info *info)
+{
+	unsigned length = info->length;
+	u64 delta = info->delta;
+
+	/* Only a commit updates the timestamp */
+	if (unlikely(!rb_event_is_commit(cpu_buffer, event)))
+		delta = 0;
 
 	/*
-	 * If this is the first commit on the page, then it has the same
-	 * timestamp as the page itself.
+	 * If we need to add a timestamp, then we
+	 * add it to the start of the resevered space.
 	 */
-	if (!tail)
+	if (unlikely(info->add_timestamp)) {
+		event = rb_add_time_stamp(event, delta);
+		length -= RB_LEN_TIME_EXTEND;
 		delta = 0;
+	}
 
-	/* See if we shot pass the end of this buffer page */
-	if (unlikely(write > BUF_PAGE_SIZE))
-		return rb_move_tail(cpu_buffer, length, tail,
-				    tail_page, ts);
+	event->time_delta = delta;
+	length -= RB_EVNT_HDR_SIZE;
+	if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT) {
+		event->type_len = 0;
+		event->array[0] = length;
+	} else
+		event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT);
+}
 
-	/* We reserved something on the buffer */
+static unsigned rb_calculate_event_length(unsigned length)
+{
+	struct ring_buffer_event event; /* Used only for sizeof array */
 
-	event = __rb_page_index(tail_page, tail);
-	kmemcheck_annotate_bitfield(event, bitfield);
-	rb_update_event(cpu_buffer, event, length, add_timestamp, delta);
+	/* zero length can cause confusions */
+	if (!length)
+		length++;
 
-	local_inc(&tail_page->entries);
+	if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT)
+		length += sizeof(event.array[0]);
+
+	length += RB_EVNT_HDR_SIZE;
+	length = ALIGN(length, RB_ARCH_ALIGNMENT);
 
 	/*
-	 * If this is the first commit on the page, then update
-	 * its timestamp.
+	 * In case the time delta is larger than the 27 bits for it
+	 * in the header, we need to add a timestamp. If another
+	 * event comes in when trying to discard this one to increase
+	 * the length, then the timestamp will be added in the allocated
+	 * space of this event. If length is bigger than the size needed
+	 * for the TIME_EXTEND, then padding has to be used. The events
+	 * length must be either RB_LEN_TIME_EXTEND, or greater than or equal
+	 * to RB_LEN_TIME_EXTEND + 8, as 8 is the minimum size for padding.
+	 * As length is a multiple of 4, we only need to worry if it
+	 * is 12 (RB_LEN_TIME_EXTEND + 4).
 	 */
-	if (!tail)
-		tail_page->page->time_stamp = ts;
+	if (length == RB_LEN_TIME_EXTEND + RB_ALIGNMENT)
+		length += RB_ALIGNMENT;
 
-	/* account for these added bytes */
-	local_add(length, &cpu_buffer->entries_bytes);
+	return length;
+}
 
-	return event;
+#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
+static inline bool sched_clock_stable(void)
+{
+	return true;
 }
+#endif
 
 static inline int
 rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer,
@@ -2483,6 +2400,59 @@ static void rb_start_commit(struct ring_buffer_per_cpu *cpu_buffer)
 	local_inc(&cpu_buffer->commits);
 }
 
+static void
+rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer)
+{
+	unsigned long max_count;
+
+	/*
+	 * We only race with interrupts and NMIs on this CPU.
+	 * If we own the commit event, then we can commit
+	 * all others that interrupted us, since the interruptions
+	 * are in stack format (they finish before they come
+	 * back to us). This allows us to do a simple loop to
+	 * assign the commit to the tail.
+	 */
+ again:
+	max_count = cpu_buffer->nr_pages * 100;
+
+	while (cpu_buffer->commit_page != cpu_buffer->tail_page) {
+		if (RB_WARN_ON(cpu_buffer, !(--max_count)))
+			return;
+		if (RB_WARN_ON(cpu_buffer,
+			       rb_is_reader_page(cpu_buffer->tail_page)))
+			return;
+		local_set(&cpu_buffer->commit_page->page->commit,
+			  rb_page_write(cpu_buffer->commit_page));
+		rb_inc_page(cpu_buffer, &cpu_buffer->commit_page);
+		cpu_buffer->write_stamp =
+			cpu_buffer->commit_page->page->time_stamp;
+		/* add barrier to keep gcc from optimizing too much */
+		barrier();
+	}
+	while (rb_commit_index(cpu_buffer) !=
+	       rb_page_write(cpu_buffer->commit_page)) {
+
+		local_set(&cpu_buffer->commit_page->page->commit,
+			  rb_page_write(cpu_buffer->commit_page));
+		RB_WARN_ON(cpu_buffer,
+			   local_read(&cpu_buffer->commit_page->page->commit) &
+			   ~RB_WRITE_MASK);
+		barrier();
+	}
+
+	/* again, keep gcc from optimizing */
+	barrier();
+
+	/*
+	 * If an interrupt came in just after the first while loop
+	 * and pushed the tail page forward, we will be left with
+	 * a dangling commit that will never go forward.
+	 */
+	if (unlikely(cpu_buffer->commit_page != cpu_buffer->tail_page))
+		goto again;
+}
+
 static inline void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer)
 {
 	unsigned long commits;
@@ -2515,91 +2485,94 @@ static inline void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer)
 	}
 }
 
-static struct ring_buffer_event *
-rb_reserve_next_event(struct ring_buffer *buffer,
-		      struct ring_buffer_per_cpu *cpu_buffer,
-		      unsigned long length)
+static inline void rb_event_discard(struct ring_buffer_event *event)
 {
-	struct ring_buffer_event *event;
-	u64 ts, delta;
-	int nr_loops = 0;
-	int add_timestamp;
-	u64 diff;
+	if (event->type_len == RINGBUF_TYPE_TIME_EXTEND)
+		event = skip_time_extend(event);
 
-	rb_start_commit(cpu_buffer);
+	/* array[0] holds the actual length for the discarded event */
+	event->array[0] = rb_event_data_length(event) - RB_EVNT_HDR_SIZE;
+	event->type_len = RINGBUF_TYPE_PADDING;
+	/* time delta must be non zero */
+	if (!event->time_delta)
+		event->time_delta = 1;
+}
 
-#ifdef CONFIG_RING_BUFFER_ALLOW_SWAP
-	/*
-	 * Due to the ability to swap a cpu buffer from a buffer
-	 * it is possible it was swapped before we committed.
-	 * (committing stops a swap). We check for it here and
-	 * if it happened, we have to fail the write.
-	 */
-	barrier();
-	if (unlikely(ACCESS_ONCE(cpu_buffer->buffer) != buffer)) {
-		local_dec(&cpu_buffer->committing);
-		local_dec(&cpu_buffer->commits);
-		return NULL;
-	}
-#endif
+static inline int
+rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer,
+		   struct ring_buffer_event *event)
+{
+	unsigned long addr = (unsigned long)event;
+	unsigned long index;
 
-	length = rb_calculate_event_length(length);
- again:
-	add_timestamp = 0;
-	delta = 0;
+	index = rb_event_index(event);
+	addr &= PAGE_MASK;
+
+	return cpu_buffer->commit_page->page == (void *)addr &&
+		rb_commit_index(cpu_buffer) == index;
+}
+
+static void
+rb_update_write_stamp(struct ring_buffer_per_cpu *cpu_buffer,
+		      struct ring_buffer_event *event)
+{
+	u64 delta;
 
 	/*
-	 * We allow for interrupts to reenter here and do a trace.
-	 * If one does, it will cause this original code to loop
-	 * back here. Even with heavy interrupts happening, this
-	 * should only happen a few times in a row. If this happens
-	 * 1000 times in a row, there must be either an interrupt
-	 * storm or we have something buggy.
-	 * Bail!
+	 * The event first in the commit queue updates the
+	 * time stamp.
 	 */
-	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000))
-		goto out_fail;
+	if (rb_event_is_commit(cpu_buffer, event)) {
+		/*
+		 * A commit event that is first on a page
+		 * updates the write timestamp with the page stamp
+		 */
+		if (!rb_event_index(event))
+			cpu_buffer->write_stamp =
+				cpu_buffer->commit_page->page->time_stamp;
+		else if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) {
+			delta = event->array[0];
+			delta <<= TS_SHIFT;
+			delta += event->time_delta;
+			cpu_buffer->write_stamp += delta;
+		} else
+			cpu_buffer->write_stamp += event->time_delta;
+	}
+}
 
-	ts = rb_time_stamp(cpu_buffer->buffer);
-	diff = ts - cpu_buffer->write_stamp;
+static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer,
+		      struct ring_buffer_event *event)
+{
+	local_inc(&cpu_buffer->entries);
+	rb_update_write_stamp(cpu_buffer, event);
+	rb_end_commit(cpu_buffer);
+}
 
-	/* make sure this diff is calculated here */
-	barrier();
+static __always_inline void
+rb_wakeups(struct ring_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer)
+{
+	bool pagebusy;
 
-	/* Did the write stamp get updated already? */
-	if (likely(ts >= cpu_buffer->write_stamp)) {
-		delta = diff;
-		if (unlikely(test_time_stamp(delta))) {
-			int local_clock_stable = 1;
-#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
-			local_clock_stable = sched_clock_stable();
-#endif
-			WARN_ONCE(delta > (1ULL << 59),
-				  KERN_WARNING "Delta way too big! %llu ts=%llu write stamp = %llu\n%s",
-				  (unsigned long long)delta,
-				  (unsigned long long)ts,
-				  (unsigned long long)cpu_buffer->write_stamp,
-				  local_clock_stable ? "" :
-				  "If you just came from a suspend/resume,\n"
-				  "please switch to the trace global clock:\n"
-				  "  echo global > /sys/kernel/debug/tracing/trace_clock\n");
-			add_timestamp = 1;
-		}
+	if (buffer->irq_work.waiters_pending) {
+		buffer->irq_work.waiters_pending = false;
+		/* irq_work_queue() supplies it's own memory barriers */
+		irq_work_queue(&buffer->irq_work.work);
 	}
 
-	event = __rb_reserve_next(cpu_buffer, length, ts,
-				  delta, add_timestamp);
-	if (unlikely(PTR_ERR(event) == -EAGAIN))
-		goto again;
-
-	if (!event)
-		goto out_fail;
+	if (cpu_buffer->irq_work.waiters_pending) {
+		cpu_buffer->irq_work.waiters_pending = false;
+		/* irq_work_queue() supplies it's own memory barriers */
+		irq_work_queue(&cpu_buffer->irq_work.work);
+	}
 
-	return event;
+	pagebusy = cpu_buffer->reader_page == cpu_buffer->commit_page;
 
- out_fail:
-	rb_end_commit(cpu_buffer);
-	return NULL;
+	if (!pagebusy && cpu_buffer->irq_work.full_waiters_pending) {
+		cpu_buffer->irq_work.wakeup_full = true;
+		cpu_buffer->irq_work.full_waiters_pending = false;
+		/* irq_work_queue() supplies it's own memory barriers */
+		irq_work_queue(&cpu_buffer->irq_work.work);
+	}
 }
 
 /*
@@ -2672,6 +2645,178 @@ trace_recursive_unlock(struct ring_buffer_per_cpu *cpu_buffer)
 }
 
 /**
+ * ring_buffer_unlock_commit - commit a reserved
+ * @buffer: The buffer to commit to
+ * @event: The event pointer to commit.
+ *
+ * This commits the data to the ring buffer, and releases any locks held.
+ *
+ * Must be paired with ring_buffer_lock_reserve.
+ */
+int ring_buffer_unlock_commit(struct ring_buffer *buffer,
+			      struct ring_buffer_event *event)
+{
+	struct ring_buffer_per_cpu *cpu_buffer;
+	int cpu = raw_smp_processor_id();
+
+	cpu_buffer = buffer->buffers[cpu];
+
+	rb_commit(cpu_buffer, event);
+
+	rb_wakeups(buffer, cpu_buffer);
+
+	trace_recursive_unlock(cpu_buffer);
+
+	preempt_enable_notrace();
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit);
+
+static noinline void
+rb_handle_timestamp(struct ring_buffer_per_cpu *cpu_buffer,
+		    struct rb_event_info *info)
+{
+	WARN_ONCE(info->delta > (1ULL << 59),
+		  KERN_WARNING "Delta way too big! %llu ts=%llu write stamp = %llu\n%s",
+		  (unsigned long long)info->delta,
+		  (unsigned long long)info->ts,
+		  (unsigned long long)cpu_buffer->write_stamp,
+		  sched_clock_stable() ? "" :
+		  "If you just came from a suspend/resume,\n"
+		  "please switch to the trace global clock:\n"
+		  "  echo global > /sys/kernel/debug/tracing/trace_clock\n");
+	info->add_timestamp = 1;
+}
+
+static struct ring_buffer_event *
+__rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer,
+		  struct rb_event_info *info)
+{
+	struct ring_buffer_event *event;
+	struct buffer_page *tail_page;
+	unsigned long tail, write;
+
+	/*
+	 * If the time delta since the last event is too big to
+	 * hold in the time field of the event, then we append a
+	 * TIME EXTEND event ahead of the data event.
+	 */
+	if (unlikely(info->add_timestamp))
+		info->length += RB_LEN_TIME_EXTEND;
+
+	tail_page = info->tail_page = cpu_buffer->tail_page;
+	write = local_add_return(info->length, &tail_page->write);
+
+	/* set write to only the index of the write */
+	write &= RB_WRITE_MASK;
+	tail = write - info->length;
+
+	/*
+	 * If this is the first commit on the page, then it has the same
+	 * timestamp as the page itself.
+	 */
+	if (!tail)
+		info->delta = 0;
+
+	/* See if we shot pass the end of this buffer page */
+	if (unlikely(write > BUF_PAGE_SIZE))
+		return rb_move_tail(cpu_buffer, tail, info);
+
+	/* We reserved something on the buffer */
+
+	event = __rb_page_index(tail_page, tail);
+	kmemcheck_annotate_bitfield(event, bitfield);
+	rb_update_event(cpu_buffer, event, info);
+
+	local_inc(&tail_page->entries);
+
+	/*
+	 * If this is the first commit on the page, then update
+	 * its timestamp.
+	 */
+	if (!tail)
+		tail_page->page->time_stamp = info->ts;
+
+	/* account for these added bytes */
+	local_add(info->length, &cpu_buffer->entries_bytes);
+
+	return event;
+}
+
+static struct ring_buffer_event *
+rb_reserve_next_event(struct ring_buffer *buffer,
+		      struct ring_buffer_per_cpu *cpu_buffer,
+		      unsigned long length)
+{
+	struct ring_buffer_event *event;
+	struct rb_event_info info;
+	int nr_loops = 0;
+	u64 diff;
+
+	rb_start_commit(cpu_buffer);
+
+#ifdef CONFIG_RING_BUFFER_ALLOW_SWAP
+	/*
+	 * Due to the ability to swap a cpu buffer from a buffer
+	 * it is possible it was swapped before we committed.
+	 * (committing stops a swap). We check for it here and
+	 * if it happened, we have to fail the write.
+	 */
+	barrier();
+	if (unlikely(ACCESS_ONCE(cpu_buffer->buffer) != buffer)) {
+		local_dec(&cpu_buffer->committing);
+		local_dec(&cpu_buffer->commits);
+		return NULL;
+	}
+#endif
+
+	info.length = rb_calculate_event_length(length);
+ again:
+	info.add_timestamp = 0;
+	info.delta = 0;
+
+	/*
+	 * We allow for interrupts to reenter here and do a trace.
+	 * If one does, it will cause this original code to loop
+	 * back here. Even with heavy interrupts happening, this
+	 * should only happen a few times in a row. If this happens
+	 * 1000 times in a row, there must be either an interrupt
+	 * storm or we have something buggy.
+	 * Bail!
+	 */
+	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000))
+		goto out_fail;
+
+	info.ts = rb_time_stamp(cpu_buffer->buffer);
+	diff = info.ts - cpu_buffer->write_stamp;
+
+	/* make sure this diff is calculated here */
+	barrier();
+
+	/* Did the write stamp get updated already? */
+	if (likely(info.ts >= cpu_buffer->write_stamp)) {
+		info.delta = diff;
+		if (unlikely(test_time_stamp(info.delta)))
+			rb_handle_timestamp(cpu_buffer, &info);
+	}
+
+	event = __rb_reserve_next(cpu_buffer, &info);
+
+	if (unlikely(PTR_ERR(event) == -EAGAIN))
+		goto again;
+
+	if (!event)
+		goto out_fail;
+
+	return event;
+
+ out_fail:
+	rb_end_commit(cpu_buffer);
+	return NULL;
+}
+
+/**
  * ring_buffer_lock_reserve - reserve a part of the buffer
  * @buffer: the ring buffer to reserve from
  * @length: the length of the data to reserve (excluding event header)
@@ -2729,111 +2874,6 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length)
 }
 EXPORT_SYMBOL_GPL(ring_buffer_lock_reserve);
 
-static void
-rb_update_write_stamp(struct ring_buffer_per_cpu *cpu_buffer,
-		      struct ring_buffer_event *event)
-{
-	u64 delta;
-
-	/*
-	 * The event first in the commit queue updates the
-	 * time stamp.
-	 */
-	if (rb_event_is_commit(cpu_buffer, event)) {
-		/*
-		 * A commit event that is first on a page
-		 * updates the write timestamp with the page stamp
-		 */
-		if (!rb_event_index(event))
-			cpu_buffer->write_stamp =
-				cpu_buffer->commit_page->page->time_stamp;
-		else if (event->type_len == RINGBUF_TYPE_TIME_EXTEND) {
-			delta = event->array[0];
-			delta <<= TS_SHIFT;
-			delta += event->time_delta;
-			cpu_buffer->write_stamp += delta;
-		} else
-			cpu_buffer->write_stamp += event->time_delta;
-	}
-}
-
-static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer,
-		      struct ring_buffer_event *event)
-{
-	local_inc(&cpu_buffer->entries);
-	rb_update_write_stamp(cpu_buffer, event);
-	rb_end_commit(cpu_buffer);
-}
-
-static __always_inline void
-rb_wakeups(struct ring_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer)
-{
-	bool pagebusy;
-
-	if (buffer->irq_work.waiters_pending) {
-		buffer->irq_work.waiters_pending = false;
-		/* irq_work_queue() supplies it's own memory barriers */
-		irq_work_queue(&buffer->irq_work.work);
-	}
-
-	if (cpu_buffer->irq_work.waiters_pending) {
-		cpu_buffer->irq_work.waiters_pending = false;
-		/* irq_work_queue() supplies it's own memory barriers */
-		irq_work_queue(&cpu_buffer->irq_work.work);
-	}
-
-	pagebusy = cpu_buffer->reader_page == cpu_buffer->commit_page;
-
-	if (!pagebusy && cpu_buffer->irq_work.full_waiters_pending) {
-		cpu_buffer->irq_work.wakeup_full = true;
-		cpu_buffer->irq_work.full_waiters_pending = false;
-		/* irq_work_queue() supplies it's own memory barriers */
-		irq_work_queue(&cpu_buffer->irq_work.work);
-	}
-}
-
-/**
- * ring_buffer_unlock_commit - commit a reserved
- * @buffer: The buffer to commit to
- * @event: The event pointer to commit.
- *
- * This commits the data to the ring buffer, and releases any locks held.
- *
- * Must be paired with ring_buffer_lock_reserve.
- */
-int ring_buffer_unlock_commit(struct ring_buffer *buffer,
-			      struct ring_buffer_event *event)
-{
-	struct ring_buffer_per_cpu *cpu_buffer;
-	int cpu = raw_smp_processor_id();
-
-	cpu_buffer = buffer->buffers[cpu];
-
-	rb_commit(cpu_buffer, event);
-
-	rb_wakeups(buffer, cpu_buffer);
-
-	trace_recursive_unlock(cpu_buffer);
-
-	preempt_enable_notrace();
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit);
-
-static inline void rb_event_discard(struct ring_buffer_event *event)
-{
-	if (event->type_len == RINGBUF_TYPE_TIME_EXTEND)
-		event = skip_time_extend(event);
-
-	/* array[0] holds the actual length for the discarded event */
-	event->array[0] = rb_event_data_length(event) - RB_EVNT_HDR_SIZE;
-	event->type_len = RINGBUF_TYPE_PADDING;
-	/* time delta must be non zero */
-	if (!event->time_delta)
-		event->time_delta = 1;
-}
-
 /*
  * Decrement the entries to the page that an event is on.
  * The event does not even need to exist, only the pointer