1 files changed, 47 insertions, 36 deletions

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 834b361a4a66..78502d4c7214 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -523,6 +523,8 @@ struct ring_buffer_per_cpu {
 	rb_time_t			before_stamp;
 	u64				event_stamp[MAX_NEST];
 	u64				read_stamp;
+	/* pages removed since last reset */
+	unsigned long			pages_removed;
 	/* ring buffer pages to update, > 0 to add, < 0 to remove */
 	long				nr_pages_to_update;
 	struct list_head		new_pages; /* new pages to add */
@@ -536,6 +538,7 @@ struct trace_buffer {
 	unsigned			flags;
 	int				cpus;
 	atomic_t			record_disabled;
+	atomic_t			resizing;
 	cpumask_var_t			cpumask;
 
 	struct lock_class_key		*reader_lock_key;
@@ -558,6 +561,7 @@ struct ring_buffer_iter {
 	struct buffer_page		*head_page;
 	struct buffer_page		*cache_reader_page;
 	unsigned long			cache_read;
+	unsigned long			cache_pages_removed;
 	u64				read_stamp;
 	u64				page_stamp;
 	struct ring_buffer_event	*event;
@@ -688,10 +692,7 @@ static void rb_time_set(rb_time_t *t, u64 val)
 static inline bool
 rb_time_read_cmpxchg(local_t *l, unsigned long expect, unsigned long set)
 {
-	unsigned long ret;
-
-	ret = local_cmpxchg(l, expect, set);
-	return ret == expect;
+	return local_try_cmpxchg(l, &expect, set);
 }
 
 static bool rb_time_cmpxchg(rb_time_t *t, u64 expect, u64 set)
@@ -748,9 +749,7 @@ static void rb_time_set(rb_time_t *t, u64 val)
 
 static bool rb_time_cmpxchg(rb_time_t *t, u64 expect, u64 set)
 {
-	u64 val;
-	val = local64_cmpxchg(&t->time, expect, set);
-	return val == expect;
+	return local64_try_cmpxchg(&t->time, &expect, set);
 }
 #endif
 
@@ -946,6 +945,7 @@ static void rb_wake_up_waiters(struct irq_work *work)
 /**
  * ring_buffer_wake_waiters - wake up any waiters on this ring buffer
  * @buffer: The ring buffer to wake waiters on
+ * @cpu: The CPU buffer to wake waiters on
  *
  * In the case of a file that represents a ring buffer is closing,
  * it is prudent to wake up any waiters that are on this.
@@ -1489,14 +1489,11 @@ static bool rb_head_page_replace(struct buffer_page *old,
 {
 	unsigned long *ptr = (unsigned long *)&old->list.prev->next;
 	unsigned long val;
-	unsigned long ret;
 
 	val = *ptr & ~RB_FLAG_MASK;
 	val |= RB_PAGE_HEAD;
 
-	ret = cmpxchg(ptr, val, (unsigned long)&new->list);
-
-	return ret == val;
+	return try_cmpxchg(ptr, &val, (unsigned long)&new->list);
 }
 
 /*
@@ -1956,6 +1953,8 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages)
 		to_remove = rb_list_head(to_remove)->next;
 		head_bit |= (unsigned long)to_remove & RB_PAGE_HEAD;
 	}
+	/* Read iterators need to reset themselves when some pages removed */
+	cpu_buffer->pages_removed += nr_removed;
 
 	next_page = rb_list_head(to_remove)->next;
 
@@ -1977,12 +1976,6 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages)
 		cpu_buffer->head_page = list_entry(next_page,
 						struct buffer_page, list);
 
-	/*
-	 * change read pointer to make sure any read iterators reset
-	 * themselves
-	 */
-	cpu_buffer->read = 0;
-
 	/* pages are removed, resume tracing and then free the pages */
 	atomic_dec(&cpu_buffer->record_disabled);
 	raw_spin_unlock_irq(&cpu_buffer->reader_lock);
@@ -2167,7 +2160,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size,
 
 	/* prevent another thread from changing buffer sizes */
 	mutex_lock(&buffer->mutex);
-
+	atomic_inc(&buffer->resizing);
 
 	if (cpu_id == RING_BUFFER_ALL_CPUS) {
 		/*
@@ -2322,6 +2315,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size,
 		atomic_dec(&buffer->record_disabled);
 	}
 
+	atomic_dec(&buffer->resizing);
 	mutex_unlock(&buffer->mutex);
 	return 0;
 
@@ -2342,6 +2336,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size,
 		}
 	}
  out_err_unlock:
+	atomic_dec(&buffer->resizing);
 	mutex_unlock(&buffer->mutex);
 	return err;
 }
@@ -3000,7 +2995,6 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer,
 {
 	unsigned long new_index, old_index;
 	struct buffer_page *bpage;
-	unsigned long index;
 	unsigned long addr;
 	u64 write_stamp;
 	u64 delta;
@@ -3057,8 +3051,9 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer,
 		 */
 		old_index += write_mask;
 		new_index += write_mask;
-		index = local_cmpxchg(&bpage->write, old_index, new_index);
-		if (index == old_index) {
+
+		/* caution: old_index gets updated on cmpxchg failure */
+		if (local_try_cmpxchg(&bpage->write, &old_index, new_index)) {
 			/* update counters */
 			local_sub(event_length, &cpu_buffer->entries_bytes);
 			return true;
@@ -3373,7 +3368,6 @@ void ring_buffer_nest_end(struct trace_buffer *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.
  *
@@ -4392,6 +4386,7 @@ static void rb_iter_reset(struct ring_buffer_iter *iter)
 
 	iter->cache_reader_page = iter->head_page;
 	iter->cache_read = cpu_buffer->read;
+	iter->cache_pages_removed = cpu_buffer->pages_removed;
 
 	if (iter->head) {
 		iter->read_stamp = cpu_buffer->read_stamp;
@@ -4846,12 +4841,13 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts)
 	buffer = cpu_buffer->buffer;
 
 	/*
-	 * Check if someone performed a consuming read to
-	 * the buffer. A consuming read invalidates the iterator
-	 * and we need to reset the iterator in this case.
+	 * Check if someone performed a consuming read to the buffer
+	 * or removed some pages from the buffer. In these cases,
+	 * iterator was invalidated and we need to reset it.
 	 */
 	if (unlikely(iter->cache_read != cpu_buffer->read ||
-		     iter->cache_reader_page != cpu_buffer->reader_page))
+		     iter->cache_reader_page != cpu_buffer->reader_page ||
+		     iter->cache_pages_removed != cpu_buffer->pages_removed))
 		rb_iter_reset(iter);
 
  again:
@@ -5242,28 +5238,34 @@ unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu)
 }
 EXPORT_SYMBOL_GPL(ring_buffer_size);
 
+static void rb_clear_buffer_page(struct buffer_page *page)
+{
+	local_set(&page->write, 0);
+	local_set(&page->entries, 0);
+	rb_init_page(page->page);
+	page->read = 0;
+}
+
 static void
 rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer)
 {
+	struct buffer_page *page;
+
 	rb_head_page_deactivate(cpu_buffer);
 
 	cpu_buffer->head_page
 		= list_entry(cpu_buffer->pages, struct buffer_page, list);
-	local_set(&cpu_buffer->head_page->write, 0);
-	local_set(&cpu_buffer->head_page->entries, 0);
-	local_set(&cpu_buffer->head_page->page->commit, 0);
-
-	cpu_buffer->head_page->read = 0;
+	rb_clear_buffer_page(cpu_buffer->head_page);
+	list_for_each_entry(page, cpu_buffer->pages, list) {
+		rb_clear_buffer_page(page);
+	}
 
 	cpu_buffer->tail_page = cpu_buffer->head_page;
 	cpu_buffer->commit_page = cpu_buffer->head_page;
 
 	INIT_LIST_HEAD(&cpu_buffer->reader_page->list);
 	INIT_LIST_HEAD(&cpu_buffer->new_pages);
-	local_set(&cpu_buffer->reader_page->write, 0);
-	local_set(&cpu_buffer->reader_page->entries, 0);
-	local_set(&cpu_buffer->reader_page->page->commit, 0);
-	cpu_buffer->reader_page->read = 0;
+	rb_clear_buffer_page(cpu_buffer->reader_page);
 
 	local_set(&cpu_buffer->entries_bytes, 0);
 	local_set(&cpu_buffer->overrun, 0);
@@ -5289,6 +5291,7 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer)
 	cpu_buffer->last_overrun = 0;
 
 	rb_head_page_activate(cpu_buffer);
+	cpu_buffer->pages_removed = 0;
 }
 
 /* Must have disabled the cpu buffer then done a synchronize_rcu */
@@ -5347,7 +5350,6 @@ EXPORT_SYMBOL_GPL(ring_buffer_reset_cpu);
 /**
  * ring_buffer_reset_online_cpus - reset a ring buffer per CPU buffer
  * @buffer: The ring buffer to reset a per cpu buffer of
- * @cpu: The CPU buffer to be reset
  */
 void ring_buffer_reset_online_cpus(struct trace_buffer *buffer)
 {
@@ -5535,6 +5537,15 @@ int ring_buffer_swap_cpu(struct trace_buffer *buffer_a,
 	if (local_read(&cpu_buffer_b->committing))
 		goto out_dec;
 
+	/*
+	 * When resize is in progress, we cannot swap it because
+	 * it will mess the state of the cpu buffer.
+	 */
+	if (atomic_read(&buffer_a->resizing))
+		goto out_dec;
+	if (atomic_read(&buffer_b->resizing))
+		goto out_dec;
+
 	buffer_a->buffers[cpu] = cpu_buffer_b;
 	buffer_b->buffers[cpu] = cpu_buffer_a;