From: Sasha Levin <levinsasha928@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: paulmck@linux.vnet.ibm.com, Dave Jones <davej@redhat.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: rcu: BUG: spinlock recursion on CPU#3, trinity-child19/5970
Date: Mon, 02 Jul 2012 15:12:53 +0200 [thread overview]
Message-ID: <1341234773.2958.4.camel@lappy> (raw)
In-Reply-To: <1341230352.23484.7.camel@twins>
On Mon, 2012-07-02 at 13:59 +0200, Peter Zijlstra wrote:
> It only needs moving back before that spin_release() thing. But unless
> there's a reason to do it later, something like the below might be the
> right thing.
>
> ---
> kernel/sched/core.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 9bb7d28..7e20e6c 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -1914,6 +1914,7 @@ prepare_task_switch(struct rq *rq, struct task_struct *prev,
> sched_info_switch(prev, next);
> perf_event_task_sched_out(prev, next);
> fire_sched_out_preempt_notifiers(prev, next);
> + rcu_switch_from(prev);
> prepare_lock_switch(rq, next);
> prepare_arch_switch(next);
> trace_sched_switch(prev, next);
> @@ -2082,7 +2083,6 @@ context_switch(struct rq *rq, struct task_struct *prev,
> #endif
>
> /* Here we just switch the register state and the stack. */
> - rcu_switch_from(prev);
> switch_to(prev, next, prev);
>
> barrier();
>
With this patch applied, I'm seeing the following:
[ 37.047889] ======================================================
[ 37.048018] [ INFO: possible circular locking dependency detected ]
[ 37.048018] 3.5.0-rc5-next-20120702-sasha-00007-gefea299 #493 Tainted: G W
[ 37.048018] -------------------------------------------------------
[ 37.048018] rcu_torture_rea/2545 is trying to acquire lock:
[ 37.048018] (&rsp->gp_wq){..-...}, at: [<ffffffff81128b5d>] __wake_up+0x2d/0x70
[ 37.048018]
[ 37.048018] but task is already holding lock:
[ 37.048018] (&rq->lock){-.-.-.}, at: [<ffffffff83738420>] __schedule+0x1a0/0x870
[ 37.048018]
[ 37.048018] which lock already depends on the new lock.
[ 37.048018]
[ 37.048018]
[ 37.048018] the existing dependency chain (in reverse order) is:
[ 37.048018]
-> #2 (&rq->lock){-.-.-.}:
[ 37.048018] [<ffffffff81163813>] validate_chain+0x6a3/0x790
[ 37.048018] [<ffffffff81163d13>] __lock_acquire+0x413/0x4b0
[ 37.048018] [<ffffffff81163f3a>] lock_acquire+0x18a/0x1e0
[ 37.048018] [<ffffffff83739c1b>] _raw_spin_lock+0x3b/0x70
[ 37.048018] [<ffffffff81133401>] wake_up_new_task+0x131/0x270
[ 37.048018] [<ffffffff810eadea>] do_fork+0x27a/0x300
[ 37.048018] [<ffffffff81071e71>] kernel_thread+0x71/0x80
[ 37.048018] [<ffffffff83646701>] rest_init+0x21/0x140
[ 37.048018] [<ffffffff850d8091>] start_kernel+0x3e1/0x3ee
[ 37.048018] [<ffffffff850d7324>] x86_64_start_reservations+0xff/0x104
[ 37.048018] [<ffffffff850d747e>] x86_64_start_kernel+0x155/0x164
[ 37.048018]
-> #1 (&p->pi_lock){-.-.-.}:
[ 37.048018] [<ffffffff81163813>] validate_chain+0x6a3/0x790
[ 37.048018] [<ffffffff81163d13>] __lock_acquire+0x413/0x4b0
[ 37.048018] [<ffffffff81163f3a>] lock_acquire+0x18a/0x1e0
[ 37.048018] [<ffffffff83739db1>] _raw_spin_lock_irqsave+0x81/0xc0
[ 37.048018] [<ffffffff81133034>] try_to_wake_up+0x34/0x290
[ 37.048018] [<ffffffff8113329d>] default_wake_function+0xd/0x10
[ 37.048018] [<ffffffff8111ad48>] autoremove_wake_function+0x18/0x40
[ 37.048018] [<ffffffff81125b92>] __wake_up_common+0x52/0x90
[ 37.048018] [<ffffffff81128b73>] __wake_up+0x43/0x70
[ 37.048018] [<ffffffff811a4715>] rcu_start_gp+0x85/0xb0
[ 37.048018] [<ffffffff811ab1c2>] __rcu_process_callbacks+0x142/0x170
[ 37.048018] [<ffffffff811ab270>] rcu_process_callbacks+0x80/0xb0
[ 37.048018] [<ffffffff810f6061>] __do_softirq+0x221/0x460
[ 37.048018] [<ffffffff810f639d>] run_ksoftirqd+0xfd/0x230
[ 37.048018] [<ffffffff8111a502>] kthread+0xb2/0xc0
[ 37.048018] [<ffffffff8373c8c4>] kernel_thread_helper+0x4/0x10
[ 37.048018]
-> #0 (&rsp->gp_wq){..-...}:
[ 37.048018] [<ffffffff81162dbf>] check_prev_add+0x11f/0x4d0
[ 37.048018] [<ffffffff81163813>] validate_chain+0x6a3/0x790
[ 37.048018] [<ffffffff81163d13>] __lock_acquire+0x413/0x4b0
[ 37.048018] [<ffffffff81163f3a>] lock_acquire+0x18a/0x1e0
[ 37.048018] [<ffffffff83739db1>] _raw_spin_lock_irqsave+0x81/0xc0
[ 37.048018] [<ffffffff81128b5d>] __wake_up+0x2d/0x70
[ 37.048018] [<ffffffff811a47b3>] rcu_report_qs_rsp+0x73/0x80
[ 37.048018] [<ffffffff811a7c9d>] rcu_report_qs_rnp+0x26d/0x2c0
[ 37.048018] [<ffffffff811a8c2e>] rcu_report_unblock_qs_rnp+0x7e/0x90
[ 37.048018] [<ffffffff811a9530>] rcu_read_unlock_special+0x330/0x400
[ 37.048018] [<ffffffff811a982a>] rcu_preempt_note_context_switch+0x22a/0x300
[ 37.048018] [<ffffffff8373870c>] __schedule+0x48c/0x870
[ 37.048018] [<ffffffff83739074>] preempt_schedule_irq+0x94/0xd0
[ 37.048018] [<ffffffff8373aee6>] retint_kernel+0x26/0x30
[ 37.048018] [<ffffffff811a337d>] rcu_torture_read_unlock+0x5d/0x60
[ 37.048018] [<ffffffff811a361d>] rcu_torture_reader+0x29d/0x380
[ 37.048018] [<ffffffff8111a502>] kthread+0xb2/0xc0
[ 37.048018] [<ffffffff8373c8c4>] kernel_thread_helper+0x4/0x10
[ 37.048018]
[ 37.048018] other info that might help us debug this:
[ 37.048018]
[ 37.048018] Chain exists of:
&rsp->gp_wq --> &p->pi_lock --> &rq->lock
[ 37.048018] Possible unsafe locking scenario:
[ 37.048018]
[ 37.048018] CPU0 CPU1
[ 37.048018] ---- ----
[ 37.048018] lock(&rq->lock);
[ 37.048018] lock(&p->pi_lock);
[ 37.048018] lock(&rq->lock);
[ 37.048018] lock(&rsp->gp_wq);
[ 37.048018]
[ 37.048018] *** DEADLOCK ***
[ 37.048018]
[ 37.048018] 1 lock held by rcu_torture_rea/2545:
[ 37.048018] #0: (&rq->lock){-.-.-.}, at: [<ffffffff83738420>] __schedule+0x1a0/0x870
[ 37.048018]
[ 37.048018] stack backtrace:
[ 37.048018] Pid: 2545, comm: rcu_torture_rea Tainted: G W 3.5.0-rc5-next-20120702-sasha-00007-gefea299 #493
[ 37.048018] Call Trace:
[ 37.048018] [<ffffffff81160da5>] print_circular_bug+0x105/0x120
[ 37.048018] [<ffffffff81162dbf>] check_prev_add+0x11f/0x4d0
[ 37.048018] [<ffffffff81163813>] validate_chain+0x6a3/0x790
[ 37.048018] [<ffffffff81135488>] ? sched_clock_cpu+0x108/0x120
[ 37.048018] [<ffffffff81163d13>] __lock_acquire+0x413/0x4b0
[ 37.048018] [<ffffffff81163f3a>] lock_acquire+0x18a/0x1e0
[ 37.048018] [<ffffffff81128b5d>] ? __wake_up+0x2d/0x70
[ 37.048018] [<ffffffff83739db1>] _raw_spin_lock_irqsave+0x81/0xc0
[ 37.048018] [<ffffffff81128b5d>] ? __wake_up+0x2d/0x70
[ 37.048018] [<ffffffff81128b5d>] __wake_up+0x2d/0x70
[ 37.048018] [<ffffffff811a47b3>] rcu_report_qs_rsp+0x73/0x80
[ 37.048018] [<ffffffff811a7c9d>] rcu_report_qs_rnp+0x26d/0x2c0
[ 37.048018] [<ffffffff811a8c2e>] rcu_report_unblock_qs_rnp+0x7e/0x90
[ 37.048018] [<ffffffff811a9530>] rcu_read_unlock_special+0x330/0x400
[ 37.048018] [<ffffffff811604a4>] ? __lock_acquired+0x2a4/0x2e0
[ 37.048018] [<ffffffff811a982a>] rcu_preempt_note_context_switch+0x22a/0x300
[ 37.048018] [<ffffffff8373870c>] __schedule+0x48c/0x870
[ 37.048018] [<ffffffff83738abd>] ? __schedule+0x83d/0x870
[ 37.048018] [<ffffffff83739074>] preempt_schedule_irq+0x94/0xd0
[ 37.048018] [<ffffffff8373aee6>] retint_kernel+0x26/0x30
[ 37.048018] [<ffffffff811a9200>] ? rcu_preempt_qs+0x120/0x120
[ 37.048018] [<ffffffff811161dd>] ? __rcu_read_unlock+0x4d/0xa0
[ 37.048018] [<ffffffff811a337d>] rcu_torture_read_unlock+0x5d/0x60
[ 37.048018] [<ffffffff811a361d>] rcu_torture_reader+0x29d/0x380
[ 37.048018] [<ffffffff811a2160>] ? T.861+0x50/0x50
[ 37.048018] [<ffffffff811a3380>] ? rcu_torture_read_unlock+0x60/0x60
[ 37.048018] [<ffffffff8111a502>] kthread+0xb2/0xc0
[ 37.048018] [<ffffffff8373c8c4>] kernel_thread_helper+0x4/0x10
[ 37.048018] [<ffffffff8373adb4>] ? retint_restore_args+0x13/0x13
[ 37.048018] [<ffffffff8111a450>] ? __init_kthread_worker+0x70/0x70
[ 37.048018] [<ffffffff8373c8c0>] ? gs_change+0x13/0x13
next prev parent reply other threads:[~2012-07-02 13:12 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-29 10:09 rcu: BUG: spinlock recursion on CPU#3, trinity-child19/5970 Sasha Levin
2012-06-29 17:23 ` Paul E. McKenney
2012-06-29 21:40 ` Sasha Levin
2012-06-29 22:27 ` Paul E. McKenney
2012-06-29 22:40 ` Sasha Levin
2012-06-29 23:01 ` Paul E. McKenney
2012-07-02 10:32 ` Peter Zijlstra
2012-07-02 11:35 ` Paul E. McKenney
2012-07-02 11:59 ` Peter Zijlstra
2012-07-02 13:12 ` Sasha Levin [this message]
2012-07-02 13:33 ` Paul E. McKenney
2012-07-02 14:22 ` Paul E. McKenney
2012-07-02 14:36 ` Sasha Levin
2012-07-02 14:59 ` Paul E. McKenney
2012-07-04 14:54 ` Sasha Levin
2012-07-04 19:07 ` Paul E. McKenney
2012-06-30 11:36 ` Sasha Levin
2012-06-30 13:14 ` Paul E. McKenney
2012-06-30 13:28 ` Sasha Levin
2012-06-30 13:58 ` Paul E. McKenney
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=1341234773.2958.4.camel@lappy \
--to=levinsasha928@gmail.com \
--cc=davej@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.