[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110121174547.GA8796@redhat.com>
Date: Fri, 21 Jan 2011 18:45:47 +0100
From: Oleg Nesterov <oleg@...hat.com>
To: Ingo Molnar <mingo@...e.hu>
Cc: mingo@...hat.com, hpa@...or.com, acme@...hat.com, paulus@...ba.org,
linux-kernel@...r.kernel.org, stern@...land.harvard.edu,
a.p.zijlstra@...llo.nl, fweisbec@...il.com, roland@...hat.com,
tglx@...utronix.de, prasad@...ux.vnet.ibm.com,
linux-tip-commits@...r.kernel.org,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Subject: [PATCH] perf: perf_event_exit_task_context:
s/rcu_dereference/rcu_dereference_raw/
In theory, almost every user of task->child->perf_event_ctxp[]
is wrong. find_get_context() can install the new context at any
moment, we need read_barrier_depends().
dbe08d82ce3967ccdf459f7951d02589cf967300 "perf: Fix
find_get_context() vs perf_event_exit_task() race" added
rcu_dereference() into perf_event_exit_task_context() to make
the precedent, but this makes __rcu_dereference_check() unhappy.
Use rcu_dereference_raw() to shut up the warning.
Reported-by: Ingo Molnar <mingo@...e.hu>
Signed-off-by: Oleg Nesterov <oleg@...hat.com>
---
kernel/perf_event.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- git/kernel/perf_event.c~6_rcu_check 2011-01-19 18:49:23.000000000 +0100
+++ git/kernel/perf_event.c 2011-01-21 18:41:02.000000000 +0100
@@ -6133,7 +6133,7 @@ static void perf_event_exit_task_context
* scheduled, so we are now safe from rescheduling changing
* our context.
*/
- child_ctx = rcu_dereference(child->perf_event_ctxp[ctxn]);
+ child_ctx = rcu_dereference_raw(child->perf_event_ctxp[ctxn]);
task_ctx_sched_out(child_ctx, EVENT_ALL);
/*
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists