Merge Linus' tree to be be to apply submitted patches to newer code than

current trivial.git base

kernel/trace/ring_buffer.c

@@ -538,16 +538,18 @@ static void rb_wake_up_waiters(struct irq_work *work)
  * ring_buffer_wait - wait for input to the ring buffer
  * @buffer: buffer to wait on
  * @cpu: the cpu buffer to wait on
+ * @full: wait until a full page is available, if @cpu != RING_BUFFER_ALL_CPUS
  *
  * If @cpu == RING_BUFFER_ALL_CPUS then the task will wake up as soon
  * as data is added to any of the @buffer's cpu buffers. Otherwise
  * it will wait for data to be added to a specific cpu buffer.
  */
-int ring_buffer_wait(struct ring_buffer *buffer, int cpu)
+int ring_buffer_wait(struct ring_buffer *buffer, int cpu, bool full)
 {
-	struct ring_buffer_per_cpu *cpu_buffer;
+	struct ring_buffer_per_cpu *uninitialized_var(cpu_buffer);
 	DEFINE_WAIT(wait);
 	struct rb_irq_work *work;
+	int ret = 0;
 
 	/*
 	 * Depending on what the caller is waiting for, either any
@@ -564,36 +566,61 @@ int ring_buffer_wait(struct ring_buffer *buffer, int cpu)
 	}
 
 
-	prepare_to_wait(&work->waiters, &wait, TASK_INTERRUPTIBLE);
+	while (true) {
+		prepare_to_wait(&work->waiters, &wait, TASK_INTERRUPTIBLE);
 
-	/*
-	 * The events can happen in critical sections where
-	 * checking a work queue can cause deadlocks.
-	 * After adding a task to the queue, this flag is set
-	 * only to notify events to try to wake up the queue
-	 * using irq_work.
-	 *
-	 * We don't clear it even if the buffer is no longer
-	 * empty. The flag only causes the next event to run
-	 * irq_work to do the work queue wake up. The worse
-	 * that can happen if we race with !trace_empty() is that
-	 * an event will cause an irq_work to try to wake up
-	 * an empty queue.
-	 *
-	 * There's no reason to protect this flag either, as
-	 * the work queue and irq_work logic will do the necessary
-	 * synchronization for the wake ups. The only thing
-	 * that is necessary is that the wake up happens after
-	 * a task has been queued. It's OK for spurious wake ups.
-	 */
-	work->waiters_pending = true;
+		/*
+		 * The events can happen in critical sections where
+		 * checking a work queue can cause deadlocks.
+		 * After adding a task to the queue, this flag is set
+		 * only to notify events to try to wake up the queue
+		 * using irq_work.
+		 *
+		 * We don't clear it even if the buffer is no longer
+		 * empty. The flag only causes the next event to run
+		 * irq_work to do the work queue wake up. The worse
+		 * that can happen if we race with !trace_empty() is that
+		 * an event will cause an irq_work to try to wake up
+		 * an empty queue.
+		 *
+		 * There's no reason to protect this flag either, as
+		 * the work queue and irq_work logic will do the necessary
+		 * synchronization for the wake ups. The only thing
+		 * that is necessary is that the wake up happens after
+		 * a task has been queued. It's OK for spurious wake ups.
+		 */
+		work->waiters_pending = true;
+
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+
+		if (cpu == RING_BUFFER_ALL_CPUS && !ring_buffer_empty(buffer))
+			break;
+
+		if (cpu != RING_BUFFER_ALL_CPUS &&
+		    !ring_buffer_empty_cpu(buffer, cpu)) {
+			unsigned long flags;
+			bool pagebusy;
+
+			if (!full)
+				break;
+
+			raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags);
+			pagebusy = cpu_buffer->reader_page == cpu_buffer->commit_page;
+			raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
+
+			if (!pagebusy)
+				break;
+		}
 
-	if ((cpu == RING_BUFFER_ALL_CPUS && ring_buffer_empty(buffer)) ||
-	    (cpu != RING_BUFFER_ALL_CPUS && ring_buffer_empty_cpu(buffer, cpu)))
 		schedule();
+	}
 
 	finish_wait(&work->waiters, &wait);
-	return 0;
+
+	return ret;
 }
 
 /**
@@ -626,8 +653,22 @@ int ring_buffer_poll_wait(struct ring_buffer *buffer, int cpu,
 		work = &cpu_buffer->irq_work;
 	}
 
-	work->waiters_pending = true;
 	poll_wait(filp, &work->waiters, poll_table);
+	work->waiters_pending = true;
+	/*
+	 * There's a tight race between setting the waiters_pending and
+	 * checking if the ring buffer is empty.  Once the waiters_pending bit
+	 * is set, the next event will wake the task up, but we can get stuck
+	 * if there's only a single event in.
+	 *
+	 * FIXME: Ideally, we need a memory barrier on the writer side as well,
+	 * but adding a memory barrier to all events will cause too much of a
+	 * performance hit in the fast path.  We only need a memory barrier when
+	 * the buffer goes from empty to having content.  But as this race is
+	 * extremely small, and it's not a problem if another event comes in, we
+	 * will fix it later.
+	 */
+	smp_mb();
 
 	if ((cpu == RING_BUFFER_ALL_CPUS && !ring_buffer_empty(buffer)) ||
 	    (cpu != RING_BUFFER_ALL_CPUS && !ring_buffer_empty_cpu(buffer, cpu)))
@@ -1968,7 +2009,7 @@ rb_add_time_stamp(struct ring_buffer_event *event, u64 delta)
 
 /**
  * rb_update_event - update event type and data
- * @event: the even to update
+ * @event: the event to update
  * @type: the type of event
  * @length: the size of the event field in the ring buffer
  *
@@ -3341,21 +3382,16 @@ static void rb_iter_reset(struct ring_buffer_iter *iter)
 	struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer;
 
 	/* Iterator usage is expected to have record disabled */
-	if (list_empty(&cpu_buffer->reader_page->list)) {
-		iter->head_page = rb_set_head_page(cpu_buffer);
-		if (unlikely(!iter->head_page))
-			return;
-		iter->head = iter->head_page->read;
-	} else {
-		iter->head_page = cpu_buffer->reader_page;
-		iter->head = cpu_buffer->reader_page->read;
-	}
+	iter->head_page = cpu_buffer->reader_page;
+	iter->head = cpu_buffer->reader_page->read;
+
+	iter->cache_reader_page = iter->head_page;
+	iter->cache_read = cpu_buffer->read;
+
 	if (iter->head)
 		iter->read_stamp = cpu_buffer->read_stamp;
 	else
 		iter->read_stamp = iter->head_page->page->time_stamp;
-	iter->cache_reader_page = cpu_buffer->reader_page;
-	iter->cache_read = cpu_buffer->read;
 }
 
 /**
@@ -3748,12 +3784,14 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts)
 		return NULL;
 
 	/*
-	 * We repeat when a time extend is encountered.
-	 * Since the time extend is always attached to a data event,
-	 * we should never loop more than once.
-	 * (We never hit the following condition more than twice).
+	 * We repeat when a time extend is encountered or we hit
+	 * the end of the page. Since the time extend is always attached
+	 * to a data event, we should never loop more than three times.
+	 * Once for going to next page, once on time extend, and
+	 * finally once to get the event.
+	 * (We never hit the following condition more than thrice).
 	 */
-	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 2))
+	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 3))
 		return NULL;
 
 	if (rb_per_cpu_empty(cpu_buffer))

kernel/trace/ring_buffer_benchmark.c

@@ -205,7 +205,6 @@ static void ring_buffer_consumer(void)
 			break;
 
 		schedule();
-		__set_current_state(TASK_RUNNING);
 	}
 	reader_finish = 0;
 	complete(&read_done);
@@ -379,7 +378,6 @@ static int ring_buffer_consumer_thread(void *arg)
 			break;
 
 		schedule();
-		__set_current_state(TASK_RUNNING);
 	}
 	__set_current_state(TASK_RUNNING);
 
@@ -407,7 +405,6 @@ static int ring_buffer_producer_thread(void *arg)
 		trace_printk("Sleeping for 10 secs\n");
 		set_current_state(TASK_INTERRUPTIBLE);
 		schedule_timeout(HZ * SLEEP_TIME);
-		__set_current_state(TASK_RUNNING);
 	}
 
 	if (kill_test)

kernel/trace/trace.c

@@ -1076,13 +1076,14 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
 }
 #endif /* CONFIG_TRACER_MAX_TRACE */
 
-static int wait_on_pipe(struct trace_iterator *iter)
+static int wait_on_pipe(struct trace_iterator *iter, bool full)
 {
 	/* Iterators are static, they should be filled or empty */
 	if (trace_buffer_iter(iter, iter->cpu_file))
 		return 0;
 
-	return ring_buffer_wait(iter->trace_buffer->buffer, iter->cpu_file);
+	return ring_buffer_wait(iter->trace_buffer->buffer, iter->cpu_file,
+				full);
 }
 
 #ifdef CONFIG_FTRACE_STARTUP_TEST
@@ -4434,15 +4435,12 @@ static int tracing_wait_pipe(struct file *filp)
 
 		mutex_unlock(&iter->mutex);
 
-		ret = wait_on_pipe(iter);
+		ret = wait_on_pipe(iter, false);
 
 		mutex_lock(&iter->mutex);
 
 		if (ret)
 			return ret;
-
-		if (signal_pending(current))
-			return -EINTR;
 	}
 
 	return 1;
@@ -5372,16 +5370,12 @@ tracing_buffers_read(struct file *filp, char __user *ubuf,
 				goto out_unlock;
 			}
 			mutex_unlock(&trace_types_lock);
-			ret = wait_on_pipe(iter);
+			ret = wait_on_pipe(iter, false);
 			mutex_lock(&trace_types_lock);
 			if (ret) {
 				size = ret;
 				goto out_unlock;
 			}
-			if (signal_pending(current)) {
-				size = -EINTR;
-				goto out_unlock;
-			}
 			goto again;
 		}
 		size = 0;
@@ -5500,7 +5494,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 	};
 	struct buffer_ref *ref;
 	int entries, size, i;
-	ssize_t ret;
+	ssize_t ret = 0;
 
 	mutex_lock(&trace_types_lock);
 
@@ -5538,13 +5532,16 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 		int r;
 
 		ref = kzalloc(sizeof(*ref), GFP_KERNEL);
-		if (!ref)
+		if (!ref) {
+			ret = -ENOMEM;
 			break;
+		}
 
 		ref->ref = 1;
 		ref->buffer = iter->trace_buffer->buffer;
 		ref->page = ring_buffer_alloc_read_page(ref->buffer, iter->cpu_file);
 		if (!ref->page) {
+			ret = -ENOMEM;
 			kfree(ref);
 			break;
 		}
@@ -5582,19 +5579,19 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos,
 
 	/* did we read anything? */
 	if (!spd.nr_pages) {
+		if (ret)
+			goto out;
+
 		if ((file->f_flags & O_NONBLOCK) || (flags & SPLICE_F_NONBLOCK)) {
 			ret = -EAGAIN;
 			goto out;
 		}
 		mutex_unlock(&trace_types_lock);
-		ret = wait_on_pipe(iter);
+		ret = wait_on_pipe(iter, true);
 		mutex_lock(&trace_types_lock);
 		if (ret)
 			goto out;
-		if (signal_pending(current)) {
-			ret = -EINTR;
-			goto out;
-		}
+
 		goto again;
 	}
 

kernel/trace/trace_events.c

@@ -2513,8 +2513,11 @@ static __init int event_test_thread(void *unused)
 	kfree(test_malloc);
 
 	set_current_state(TASK_INTERRUPTIBLE);
-	while (!kthread_should_stop())
+	while (!kthread_should_stop()) {
 		schedule();
+		set_current_state(TASK_INTERRUPTIBLE);
+	}
+	__set_current_state(TASK_RUNNING);
 
 	return 0;
 }

kernel/trace/trace_selftest.c

@@ -382,6 +382,8 @@ static int trace_selftest_startup_dynamic_tracing(struct tracer *trace,
 
 	/* check the trace buffer */
 	ret = trace_test_buffer(&tr->trace_buffer, &count);
+
+	ftrace_enabled = 1;
 	tracing_start();
 
 	/* we should only have one item */
@@ -679,6 +681,8 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 
 	/* check the trace buffer */
 	ret = trace_test_buffer(&tr->trace_buffer, &count);
+
+	ftrace_enabled = 1;
 	trace->reset(tr);
 	tracing_start();
 
@@ -1025,6 +1029,12 @@ trace_selftest_startup_nop(struct tracer *trace, struct trace_array *tr)
 #endif
 
 #ifdef CONFIG_SCHED_TRACER
+
+struct wakeup_test_data {
+	struct completion	is_ready;
+	int			go;
+};
+
 static int trace_wakeup_test_thread(void *data)
 {
 	/* Make this a -deadline thread */
@@ -1034,51 +1044,56 @@ static int trace_wakeup_test_thread(void *data)
 		.sched_deadline = 10000000ULL,
 		.sched_period = 10000000ULL
 	};
-	struct completion *x = data;
+	struct wakeup_test_data *x = data;
 
 	sched_setattr(current, &attr);
 
 	/* Make it know we have a new prio */
-	complete(x);
+	complete(&x->is_ready);
 
 	/* now go to sleep and let the test wake us up */
 	set_current_state(TASK_INTERRUPTIBLE);
-	schedule();
+	while (!x->go) {
+		schedule();
+		set_current_state(TASK_INTERRUPTIBLE);
+	}
 
-	complete(x);
+	complete(&x->is_ready);
+
+	set_current_state(TASK_INTERRUPTIBLE);
 
 	/* we are awake, now wait to disappear */
 	while (!kthread_should_stop()) {
-		/*
-		 * This will likely be the system top priority
-		 * task, do short sleeps to let others run.
-		 */
-		msleep(100);
+		schedule();
+		set_current_state(TASK_INTERRUPTIBLE);
 	}
 
+	__set_current_state(TASK_RUNNING);
+
 	return 0;
 }
-
 int
 trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr)
 {
 	unsigned long save_max = tr->max_latency;
 	struct task_struct *p;
-	struct completion is_ready;
+	struct wakeup_test_data data;
 	unsigned long count;
 	int ret;
 
-	init_completion(&is_ready);
+	memset(&data, 0, sizeof(data));
+
+	init_completion(&data.is_ready);
 
 	/* create a -deadline thread */
-	p = kthread_run(trace_wakeup_test_thread, &is_ready, "ftrace-test");
+	p = kthread_run(trace_wakeup_test_thread, &data, "ftrace-test");
 	if (IS_ERR(p)) {
 		printk(KERN_CONT "Failed to create ftrace wakeup test thread ");
 		return -1;
 	}
 
 	/* make sure the thread is running at -deadline policy */
-	wait_for_completion(&is_ready);
+	wait_for_completion(&data.is_ready);
 
 	/* start the tracing */
 	ret = tracer_init(trace, tr);
@@ -1099,18 +1114,20 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr)
 		msleep(100);
 	}
 
-	init_completion(&is_ready);
+	init_completion(&data.is_ready);
+
+	data.go = 1;
+	/* memory barrier is in the wake_up_process() */
 
 	wake_up_process(p);
 
 	/* Wait for the task to wake up */
-	wait_for_completion(&is_ready);
+	wait_for_completion(&data.is_ready);
 
 	/* stop the tracing. */
 	tracing_stop();
 	/* check both trace buffers */
 	ret = trace_test_buffer(&tr->trace_buffer, NULL);
-	printk("ret = %d\n", ret);
 	if (!ret)
 		ret = trace_test_buffer(&tr->max_buffer, &count);
 

kernel/trace/trace_stack.c

@@ -13,7 +13,6 @@
 #include <linux/sysctl.h>
 #include <linux/init.h>
 #include <linux/fs.h>
-#include <linux/magic.h>
 
 #include <asm/setup.h>
 
@@ -171,8 +170,7 @@ check_stack(unsigned long ip, unsigned long *stack)
 			i++;
 	}
 
-	if ((current != &init_task &&
-		*(end_of_stack(current)) != STACK_END_MAGIC)) {
+	if (task_stack_end_corrupted(current)) {
 		print_max_stack();
 		BUG();
 	}

kernel/trace/trace_syscalls.c

@@ -313,7 +313,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 	int size;
 
 	syscall_nr = trace_get_syscall_nr(current, regs);
-	if (syscall_nr < 0)
+	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
 
 	/* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE) */
@@ -360,7 +360,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 	int syscall_nr;
 
 	syscall_nr = trace_get_syscall_nr(current, regs);
-	if (syscall_nr < 0)
+	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
 
 	/* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE()) */
@@ -425,7 +425,7 @@ static void unreg_event_syscall_enter(struct ftrace_event_file *file,
 		return;
 	mutex_lock(&syscall_trace_lock);
 	tr->sys_refcount_enter--;
-	rcu_assign_pointer(tr->enter_syscall_files[num], NULL);
+	RCU_INIT_POINTER(tr->enter_syscall_files[num], NULL);
 	if (!tr->sys_refcount_enter)
 		unregister_trace_sys_enter(ftrace_syscall_enter, tr);
 	mutex_unlock(&syscall_trace_lock);
@@ -463,7 +463,7 @@ static void unreg_event_syscall_exit(struct ftrace_event_file *file,
 		return;
 	mutex_lock(&syscall_trace_lock);
 	tr->sys_refcount_exit--;
-	rcu_assign_pointer(tr->exit_syscall_files[num], NULL);
+	RCU_INIT_POINTER(tr->exit_syscall_files[num], NULL);
 	if (!tr->sys_refcount_exit)
 		unregister_trace_sys_exit(ftrace_syscall_exit, tr);
 	mutex_unlock(&syscall_trace_lock);
@@ -567,7 +567,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 	int size;
 
 	syscall_nr = trace_get_syscall_nr(current, regs);
-	if (syscall_nr < 0)
+	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
 	if (!test_bit(syscall_nr, enabled_perf_enter_syscalls))
 		return;
@@ -641,7 +641,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 	int size;
 
 	syscall_nr = trace_get_syscall_nr(current, regs);
-	if (syscall_nr < 0)
+	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
 	if (!test_bit(syscall_nr, enabled_perf_exit_syscalls))
 		return;