* [PATCH urgent v2] sched: Add missing rcu protection to wake_up_all_idle_cpus
@ 2014-11-29 16:13 Andy Lutomirski
2014-12-01 3:13 ` Liu, Chuansheng
2014-12-09 10:17 ` [tip:sched/core] " tip-bot for Andy Lutomirski
0 siblings, 2 replies; 3+ messages in thread
From: Andy Lutomirski @ 2014-11-29 16:13 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, linux-kernel
Cc: Peter Zijlstra, Andy Lutomirski, Chuansheng Liu
Locklessly doing is_idle_task(rq->curr) is only okay because of RCU
protection. The older variant of the broken code checked
rq->curr == rq->idle instead and therefore didn't need RCU.
Fixes: f6be8af1c95d sched: Add new API wake_up_if_idle() to wake up the idle cpu
Cc: Chuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
---
This fixes a 3.18 regression, I think. This code seems to only be called
inside preempt_disable, so I think it's unlikely to be a problem in practice.
Compile-tested only, although the kbuild bot is chewing on it.
Changes from v1:
- Cc the right people (whoops)
- Add rcu_read_lock and rcu_read_unlock.
kernel/sched/core.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 240157c13ddc..b7937512965f 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1623,8 +1623,10 @@ void wake_up_if_idle(int cpu)
struct rq *rq = cpu_rq(cpu);
unsigned long flags;
- if (!is_idle_task(rq->curr))
- return;
+ rcu_read_lock();
+
+ if (!is_idle_task(rcu_dereference(rq->curr)))
+ goto out;
if (set_nr_if_polling(rq->idle)) {
trace_sched_wake_idle_without_ipi(cpu);
@@ -1635,6 +1637,9 @@ void wake_up_if_idle(int cpu)
/* Else cpu is not in idle, do nothing here */
raw_spin_unlock_irqrestore(&rq->lock, flags);
}
+
+out:
+ rcu_read_unlock();
}
bool cpus_share_cache(int this_cpu, int that_cpu)
--
1.9.3
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: [PATCH urgent v2] sched: Add missing rcu protection to wake_up_all_idle_cpus
2014-11-29 16:13 [PATCH urgent v2] sched: Add missing rcu protection to wake_up_all_idle_cpus Andy Lutomirski
@ 2014-12-01 3:13 ` Liu, Chuansheng
2014-12-09 10:17 ` [tip:sched/core] " tip-bot for Andy Lutomirski
1 sibling, 0 replies; 3+ messages in thread
From: Liu, Chuansheng @ 2014-12-01 3:13 UTC (permalink / raw)
To: Andy Lutomirski, Ingo Molnar, Thomas Gleixner, linux-kernel
Cc: Peter Zijlstra
> -----Original Message-----
> From: Andy Lutomirski [mailto:luto@amacapital.net]
> Sent: Sunday, November 30, 2014 12:14 AM
> To: Ingo Molnar; Thomas Gleixner; linux-kernel@vger.kernel.org
> Cc: Peter Zijlstra; Andy Lutomirski; Liu, Chuansheng
> Subject: [PATCH urgent v2] sched: Add missing rcu protection to
> wake_up_all_idle_cpus
>
> Locklessly doing is_idle_task(rq->curr) is only okay because of RCU
> protection. The older variant of the broken code checked
> rq->curr == rq->idle instead and therefore didn't need RCU.
>
> Fixes: f6be8af1c95d sched: Add new API wake_up_if_idle() to wake up the idle
> cpu
> Cc: Chuansheng Liu <chuansheng.liu@intel.com>
> Signed-off-by: Andy Lutomirski <luto@amacapital.net>
> ---
Reviewed-by: Chuansheng Liu <chuansheng.liu@intel.com>
Thanks Andy.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [tip:sched/core] sched: Add missing rcu protection to wake_up_all_idle_cpus
2014-11-29 16:13 [PATCH urgent v2] sched: Add missing rcu protection to wake_up_all_idle_cpus Andy Lutomirski
2014-12-01 3:13 ` Liu, Chuansheng
@ 2014-12-09 10:17 ` tip-bot for Andy Lutomirski
1 sibling, 0 replies; 3+ messages in thread
From: tip-bot for Andy Lutomirski @ 2014-12-09 10:17 UTC (permalink / raw)
To: linux-tip-commits
Cc: hpa, linux-kernel, chuansheng.liu, peterz, luto, tglx, mingo
Commit-ID: fd7de1e8d5b2b2b35e71332fafb899f584597150
Gitweb: http://git.kernel.org/tip/fd7de1e8d5b2b2b35e71332fafb899f584597150
Author: Andy Lutomirski <luto@amacapital.net>
AuthorDate: Sat, 29 Nov 2014 08:13:51 -0800
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Mon, 8 Dec 2014 11:44:19 +0100
sched: Add missing rcu protection to wake_up_all_idle_cpus
Locklessly doing is_idle_task(rq->curr) is only okay because of
RCU protection. The older variant of the broken code checked
rq->curr == rq->idle instead and therefore didn't need RCU.
Fixes: f6be8af1c95d ("sched: Add new API wake_up_if_idle() to wake up the idle cpu")
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Chuansheng Liu <chuansheng.liu@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/729365dddca178506dfd0a9451006344cd6808bc.1417277372.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/sched/core.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d44d0c5..88f49bc 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1628,8 +1628,10 @@ void wake_up_if_idle(int cpu)
struct rq *rq = cpu_rq(cpu);
unsigned long flags;
- if (!is_idle_task(rq->curr))
- return;
+ rcu_read_lock();
+
+ if (!is_idle_task(rcu_dereference(rq->curr)))
+ goto out;
if (set_nr_if_polling(rq->idle)) {
trace_sched_wake_idle_without_ipi(cpu);
@@ -1640,6 +1642,9 @@ void wake_up_if_idle(int cpu)
/* Else cpu is not in idle, do nothing here */
raw_spin_unlock_irqrestore(&rq->lock, flags);
}
+
+out:
+ rcu_read_unlock();
}
bool cpus_share_cache(int this_cpu, int that_cpu)
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2014-12-09 10:18 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-11-29 16:13 [PATCH urgent v2] sched: Add missing rcu protection to wake_up_all_idle_cpus Andy Lutomirski
2014-12-01 3:13 ` Liu, Chuansheng
2014-12-09 10:17 ` [tip:sched/core] " tip-bot for Andy Lutomirski
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome