Re: [PATCH] sched/fair: Fix RCU stall upon ENOMEM atsched_create_group().

From: Tetsuo Handa
Date: Tue Jan 06 2015 - 09:03:59 EST


Peter Zijlstra wrote:
> On Thu, Dec 25, 2014 at 10:10:45PM +0900, Tetsuo Handa wrote:
> > >From 052595ab1a1d1c5668d9de61395c9cc17694597e Mon Sep 17 00:00:00 2001
> > From: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
> > Date: Thu, 25 Dec 2014 15:51:21 +0900
> > Subject: [PATCH] sched/fair: Fix RCU stall upon ENOMEM at sched_create_group().
> >
> > When alloc_fair_sched_group() in sched_create_group() failed,
> > free_sched_group() is called, and free_fair_sched_group() is called by
> > free_sched_group(). Since destroy_cfs_bandwidth() is called by
> > free_fair_sched_group() without calling init_cfs_bandwidth(),
> > RCU stall occurs at hrtimer_cancel().
> >
>
> Thanks
>

Oops, I didn't notice this member depends on CONFIG_CFS_BANDWIDTH=y.

kernel/sched/fair.c: In function 'free_fair_sched_group':
>> kernel/sched/fair.c:7945:11: error: 'struct cfs_bandwidth' has no member named 'throttled_cfs_rq'
if (cfs_b->throttled_cfs_rq.next)
^

Here is updated patch.
----------
>From e812151aad03fb225211093987b575e690814fd3 Mon Sep 17 00:00:00 2001
From: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
Date: Tue, 6 Jan 2015 22:52:16 +0900
Subject: [PATCH] sched/fair: Fix RCU stall upon ENOMEM at sched_create_group().

When alloc_fair_sched_group() in sched_create_group() failed,
free_sched_group() is called, and free_fair_sched_group() is called by
free_sched_group(). Since destroy_cfs_bandwidth() is called by
free_fair_sched_group() without calling init_cfs_bandwidth(),
RCU stall occurs at hrtimer_cancel().

INFO: rcu_sched self-detected stall on CPU { 1} (t=60000 jiffies g=13074 c=13073 q=0)
Task dump for CPU 1:
(fprintd) R running task 0 6249 1 0x00000088
ffffffff81c47d40 ffff88007fc43d78 ffffffff81094988 0000000000000001
ffffffff81c47d40 ffff88007fc43d98 ffffffff81097acd ffff88007fc43dd8
0000000000000002 ffff88007fc43dc8 ffffffff810c3a80 ffff88007fc4d840
Call Trace:
<IRQ> [<ffffffff81094988>] sched_show_task+0xa8/0x110
[<ffffffff81097acd>] dump_cpu_task+0x3d/0x50
[<ffffffff810c3a80>] rcu_dump_cpu_stacks+0x90/0xd0
[<ffffffff810c7751>] rcu_check_callbacks+0x491/0x700
[<ffffffff810cbf2b>] update_process_times+0x4b/0x80
[<ffffffff810db046>] tick_sched_handle.isra.20+0x36/0x50
[<ffffffff810db0a2>] tick_sched_timer+0x42/0x70
[<ffffffff810ccb19>] __run_hrtimer+0x69/0x1a0
[<ffffffff810db060>] ? tick_sched_handle.isra.20+0x50/0x50
[<ffffffff810ccedf>] hrtimer_interrupt+0xef/0x230
[<ffffffff810452cb>] local_apic_timer_interrupt+0x3b/0x70
[<ffffffff8164a465>] smp_apic_timer_interrupt+0x45/0x60
[<ffffffff816485bd>] apic_timer_interrupt+0x6d/0x80
<EOI> [<ffffffff810cc588>] ? lock_hrtimer_base.isra.23+0x18/0x50
[<ffffffff81193cf1>] ? __kmalloc+0x211/0x230
[<ffffffff810cc9d2>] hrtimer_try_to_cancel+0x22/0xd0
[<ffffffff81193cf1>] ? __kmalloc+0x211/0x230
[<ffffffff810ccaa2>] hrtimer_cancel+0x22/0x30
[<ffffffff810a3cb5>] free_fair_sched_group+0x25/0xd0
[<ffffffff8108df46>] free_sched_group+0x16/0x40
[<ffffffff810971bb>] sched_create_group+0x4b/0x80
[<ffffffff810aa383>] sched_autogroup_create_attach+0x43/0x1c0
[<ffffffff8107dc9c>] sys_setsid+0x7c/0x110
[<ffffffff81647729>] system_call_fastpath+0x12/0x17

Check whether init_cfs_bandwidth() was called before calling
destroy_cfs_bandwidth(). Use #ifdef because tg_cfs_bandwidth() returns
NULL and destroy_cfs_bandwidth() is a no-op if CONFIG_CFS_BANDWIDTH=n.

Signed-off-by: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
---
kernel/sched/fair.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index df2cdf7..017fe22 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7938,8 +7938,14 @@ static void task_move_group_fair(struct task_struct *p, int queued)
void free_fair_sched_group(struct task_group *tg)
{
int i;
+#ifdef CONFIG_CFS_BANDWIDTH
+ struct cfs_bandwidth *cfs_b;

- destroy_cfs_bandwidth(tg_cfs_bandwidth(tg));
+ /* Check whether init_cfs_bandwidth() was called. */
+ cfs_b = tg_cfs_bandwidth(tg);
+ if (cfs_b->throttled_cfs_rq.next)
+ destroy_cfs_bandwidth(cfs_b);
+#endif

for_each_possible_cpu(i) {
if (tg->cfs_rq)
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/