[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20200210122408.603857143@linuxfoundation.org>
Date: Mon, 10 Feb 2020 04:29:43 -0800
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: linux-kernel@...r.kernel.org
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
stable@...r.kernel.org,
"Steven Rostedt (VMware)" <rostedt@...dmis.org>,
"Joel Fernandes (Google)" <joel@...lfernandes.org>,
Peter Zijlstra <peterz@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>,
"Paul E. McKenney" <paulmck@...ux.ibm.com>,
syzbot+774fddf07b7ab29a1e55@...kaller.appspotmail.com,
Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
Subject: [PATCH 5.4 027/309] tracing: Fix sched switch start/stop refcount racy updates
From: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
commit 64ae572bc7d0060429e40e1c8d803ce5eb31a0d6 upstream.
Reading the sched_cmdline_ref and sched_tgid_ref initial state within
tracing_start_sched_switch without holding the sched_register_mutex is
racy against concurrent updates, which can lead to tracepoint probes
being registered more than once (and thus trigger warnings within
tracepoint.c).
[ May be the fix for this bug ]
Link: https://lore.kernel.org/r/000000000000ab6f84056c786b93@google.com
Link: http://lkml.kernel.org/r/20190817141208.15226-1-mathieu.desnoyers@efficios.com
Cc: stable@...r.kernel.org
CC: Steven Rostedt (VMware) <rostedt@...dmis.org>
CC: Joel Fernandes (Google) <joel@...lfernandes.org>
CC: Peter Zijlstra <peterz@...radead.org>
CC: Thomas Gleixner <tglx@...utronix.de>
CC: Paul E. McKenney <paulmck@...ux.ibm.com>
Reported-by: syzbot+774fddf07b7ab29a1e55@...kaller.appspotmail.com
Fixes: d914ba37d7145 ("tracing: Add support for recording tgid of tasks")
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@...dmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
kernel/trace/trace_sched_switch.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -89,8 +89,10 @@ static void tracing_sched_unregister(voi
static void tracing_start_sched_switch(int ops)
{
- bool sched_register = (!sched_cmdline_ref && !sched_tgid_ref);
+ bool sched_register;
+
mutex_lock(&sched_register_mutex);
+ sched_register = (!sched_cmdline_ref && !sched_tgid_ref);
switch (ops) {
case RECORD_CMDLINE:
Powered by blists - more mailing lists