Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1258597

[for-next][PATCH 4/5] tracing: Check all tasks on each CPU when filtering pids

Show all headers | View raw

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

My tests found that if a task is running but not filtered when set_event_pid
is modified, then it can still be traced.

Call on_each_cpu() to check if the current running task should be filtered
and update the per cpu flags of tr->data appropriately.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 kernel/trace/trace_events.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index ab07058e27c1..2b7fccd499c6 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -1549,6 +1549,22 @@ 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;
+	struct trace_pid_list *pid_list;
+
+	/*
+	 * This function is called by on_each_cpu() while the
+	 * event_mutex is held.
+	 */
+	pid_list = rcu_dereference_protected(tr->filtered_pids,
+					     mutex_is_locked(&event_mutex));
+
+	this_cpu_write(tr->trace_buffer.data->ignore_pid,
+		       check_ignore_pid(pid_list, current));
+}
+
 static ssize_t
 ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 		       size_t cnt, loff_t *ppos)
@@ -1711,6 +1727,12 @@ ftrace_event_pid_write(struct file *filp, const char __user *ubuf,
 						 tr, INT_MAX);
 		register_trace_prio_sched_wakeup(event_filter_pid_sched_wakeup_probe_post,
 						 tr, 0);
+
+		/*
+		 * Ignoring of pids is done at task switch. But we have to
+		 * check for those tasks that are currently running.
+		 */
+		on_each_cpu(ignore_task_cpu, tr, 1);
 	}
 
 	mutex_unlock(&event_mutex);
-- 
2.6.1


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[for-next][PATCH 0/5] tracing: Addition of set_event_pid Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100
  [for-next][PATCH 1/5] tracepoint: Give priority to probes of tracepoints Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100
    Re: [for-next][PATCH 1/5] tracepoint: Give priority to probes of  tracepoints Mathieu Desnoyers <mathieu.desnoyers@efficios.com> - 2015-10-29 19:00 +0100
      Re: [for-next][PATCH 1/5] tracepoint: Give priority to probes of  tracepoints Steven Rostedt <rostedt@goodmis.org> - 2015-10-30 21:30 +0100
        Re: [for-next][PATCH 1/5] tracepoint: Give priority to probes of  tracepoints Mathieu Desnoyers <mathieu.desnoyers@efficios.com> - 2015-10-30 21:50 +0100
  [for-next][PATCH 4/5] tracing: Check all tasks on each CPU when filtering pids Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100
    Re: [for-next][PATCH 4/5] tracing: Check all tasks on each CPU when  filtering pids Jiaxing Wang <hello.wjx@gmail.com> - 2015-10-30 07:20 +0100
      Re: [for-next][PATCH 4/5] tracing: Check all tasks on each CPU when  filtering pids Steven Rostedt <rostedt@goodmis.org> - 2015-10-30 21:30 +0100
  [for-next][PATCH 3/5] tracing: Implement event pid filtering Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100
  [for-next][PATCH 2/5] tracing: Add set_event_pid directory for future use Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100
  [for-next][PATCH 5/5] tracing: Fix sparse RCU warning Steven Rostedt <rostedt@goodmis.org> - 2015-10-29 08:20 +0100

csiph-web