From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: linux-kernel@vger.kernel.org
Cc: mingo@elte.hu, laijs@cn.fujitsu.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@polymtl.ca,
josh@joshtriplett.org, niv@us.ibm.com, tglx@linutronix.de,
peterz@infradead.org, rostedt@goodmis.org,
Valdis.Kletnieks@vt.edu, dhowells@redhat.com,
eric.dumazet@gmail.com, darren@dvhart.com, patches@linaro.org,
greearb@candelatech.com, edt@aei.ca,
"Paul E. McKenney" <paul.mckenney@linaro.org>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Subject: [PATCH tip/core/urgent 7/7] signal: align __lock_task_sighand() irq disabling and RCU
Date: Wed, 20 Jul 2011 11:26:23 -0700 [thread overview]
Message-ID: <1311186383-24819-7-git-send-email-paulmck@linux.vnet.ibm.com> (raw)
In-Reply-To: <20110720182512.GA22946@linux.vnet.ibm.com>
From: Paul E. McKenney <paul.mckenney@linaro.org>
The __lock_task_sighand() function calls rcu_read_lock() with interrupts
and preemption enabled, but later calls rcu_read_unlock() with interrupts
disabled. It is therefore possible that this RCU read-side critical
section will be preempted and later RCU priority boosted, which means that
rcu_read_unlock() will call rt_mutex_unlock() in order to deboost itself, but
with interrupts disabled. This results in lockdep splats, so this commit
nests the RCU read-side critical section within the interrupt-disabled
region of code. This prevents the RCU read-side critical section from
being preempted, and thus prevents the attempt to deboost with interrupts
disabled.
It is quite possible that a better long-term fix is to make rt_mutex_unlock()
disable irqs when acquiring the rt_mutex structure's ->wait_lock.
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
kernel/signal.c | 19 +++++++++++++------
1 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/kernel/signal.c b/kernel/signal.c
index ff76786..415d85d 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1178,18 +1178,25 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
{
struct sighand_struct *sighand;
- rcu_read_lock();
for (;;) {
+ local_irq_save(*flags);
+ rcu_read_lock();
sighand = rcu_dereference(tsk->sighand);
- if (unlikely(sighand == NULL))
+ if (unlikely(sighand == NULL)) {
+ rcu_read_unlock();
+ local_irq_restore(*flags);
break;
+ }
- spin_lock_irqsave(&sighand->siglock, *flags);
- if (likely(sighand == tsk->sighand))
+ spin_lock(&sighand->siglock);
+ if (likely(sighand == tsk->sighand)) {
+ rcu_read_unlock();
break;
- spin_unlock_irqrestore(&sighand->siglock, *flags);
+ }
+ spin_unlock(&sighand->siglock);
+ rcu_read_unlock();
+ local_irq_restore(*flags);
}
- rcu_read_unlock();
return sighand;
}
--
1.7.3.2
prev parent reply other threads:[~2011-07-20 18:26 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-07-20 0:17 [PATCH rcu/urgent 0/6] Fixes for RCU/scheduler/irq-threads trainwreck Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 1/7] rcu: decrease rcu_report_exp_rnp coupling with scheduler Paul E. McKenney
2011-07-20 2:40 ` Peter Zijlstra
2011-07-20 4:54 ` Paul E. McKenney
2011-07-20 11:23 ` Ed Tomlinson
2011-07-20 11:31 ` Ed Tomlinson
2011-07-20 12:35 ` Peter Zijlstra
2011-07-20 13:23 ` Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 2/7] rcu: Fix RCU_BOOST race handling current->rcu_read_unlock_special Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 3/7] rcu: Streamline code produced by __rcu_read_unlock() Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 4/7] rcu: protect __rcu_read_unlock() against scheduler-using irq handlers Paul E. McKenney
2011-07-20 12:54 ` Peter Zijlstra
2011-07-20 13:25 ` Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 5/7] sched: Add irq_{enter,exit}() to scheduler_ipi() Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 6/7] softirq,rcu: Inform RCU of irq_exit() activity Paul E. McKenney
2011-07-20 0:18 ` [PATCH tip/core/urgent 7/7] signal: align __lock_task_sighand() irq disabling and RCU Paul E. McKenney
2011-07-20 1:10 ` [PATCH rcu/urgent 0/6] Fixes for RCU/scheduler/irq-threads trainwreck Ben Greear
2011-07-20 1:30 ` Ed Tomlinson
2011-07-20 2:07 ` Ed Tomlinson
2011-07-20 4:44 ` Paul E. McKenney
2011-07-20 5:03 ` Linus Torvalds
2011-07-20 13:34 ` Paul E. McKenney
2011-07-20 17:02 ` Ben Greear
2011-07-20 17:15 ` Paul E. McKenney
2011-07-20 18:44 ` Ingo Molnar
2011-07-20 18:52 ` Peter Zijlstra
2011-07-20 19:01 ` Paul E. McKenney
2011-07-20 19:25 ` Peter Zijlstra
2011-07-20 20:06 ` Paul E. McKenney
2011-07-20 19:02 ` Linus Torvalds
2011-07-20 19:29 ` Paul E. McKenney
2011-07-20 19:39 ` Ingo Molnar
2011-07-20 19:57 ` Ingo Molnar
2011-07-20 20:33 ` Paul E. McKenney
2011-07-20 20:54 ` Ben Greear
2011-07-20 21:12 ` Paul E. McKenney
2011-07-21 3:25 ` Ben Greear
2011-07-21 16:04 ` Paul E. McKenney
2011-07-20 21:04 ` [GIT PULL] RCU fixes for v3.0 Ingo Molnar
2011-07-20 21:55 ` Ed Tomlinson
2011-07-20 22:06 ` Paul E. McKenney
2011-07-20 20:08 ` [PATCH rcu/urgent 0/6] Fixes for RCU/scheduler/irq-threads trainwreck Paul E. McKenney
2011-07-20 21:05 ` Peter Zijlstra
2011-07-20 21:39 ` Paul E. McKenney
2011-07-20 10:49 ` Ed Tomlinson
2011-07-20 18:25 ` [PATCH rcu/urgent 0/7 v2] " Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 1/7] rcu: decrease rcu_report_exp_rnp coupling with scheduler Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 2/7] rcu: Fix RCU_BOOST race handling current->rcu_read_unlock_special Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 3/7] rcu: Streamline code produced by __rcu_read_unlock() Paul E. McKenney
2011-07-20 22:44 ` Linus Torvalds
2011-07-21 5:09 ` Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 4/7] rcu: protect __rcu_read_unlock() against scheduler-using irq handlers Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 5/7] sched: Add irq_{enter,exit}() to scheduler_ipi() Paul E. McKenney
2011-07-20 18:26 ` [PATCH tip/core/urgent 6/7] softirq,rcu: Inform RCU of irq_exit() activity Paul E. McKenney
2011-07-20 18:26 ` Paul E. McKenney [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1311186383-24819-7-git-send-email-paulmck@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=darren@dvhart.com \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edt@aei.ca \
--cc=eric.dumazet@gmail.com \
--cc=greearb@candelatech.com \
--cc=josh@joshtriplett.org \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=niv@us.ibm.com \
--cc=patches@linaro.org \
--cc=paul.mckenney@linaro.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®