From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-kernel@vger.kernel.org, laijs@cn.fujitsu.com,
dipankar@in.ibm.com, akpm@linux-foundation.org,
mathieu.desnoyers@polymtl.ca, josh@joshtriplett.org,
dvhltc@us.ibm.com, niv@us.ibm.com, tglx@linutronix.de,
peterz@infradead.org, rostedt@goodmis.org,
Valdis.Kletnieks@vt.edu, dhowells@redhat.com
Subject: Re: [PATCH tip/core/rcu 0/21] v6 add lockdep-based diagnostics to rcu_dereference()
Date: Thu, 25 Feb 2010 13:36:33 -0800 [thread overview]
Message-ID: <20100225213633.GA5936@linux.vnet.ibm.com> (raw)
In-Reply-To: <20100225181830.GC6771@linux.vnet.ibm.com>
On Thu, Feb 25, 2010 at 10:18:30AM -0800, Paul E. McKenney wrote:
> On Thu, Feb 25, 2010 at 01:04:44PM +0100, Ingo Molnar wrote:
> >
> > another, different warning is:
> >
> > PM: Adding info for No Bus:vcsa6
> > ------------[ cut here ]------------
> > WARNING: at kernel/softirq.c:143 local_bh_enable_ip+0xba/0xf0()
> > Hardware name: System Product Name
> > Modules linked in:
> > Pid: 0, comm: swapper Not tainted 2.6.33-tip-00730-gacec70d-dirty #18737
> > Call Trace:
> > [<ffffffff8104e0eb>] warn_slowpath_common+0x7b/0xc0
> > [<ffffffff8166cd60>] ? __dst_free+0x60/0xd0
> > [<ffffffff8104e144>] warn_slowpath_null+0x14/0x20
> > [<ffffffff81055c7a>] local_bh_enable_ip+0xba/0xf0
> > [<ffffffff817c16d9>] _raw_spin_unlock_bh+0x19/0x20
> > [<ffffffff8166cd60>] __dst_free+0x60/0xd0
> > [<ffffffff8169de14>] dst_rcu_free+0x34/0x40
> > [<ffffffff8109c80d>] rcu_do_batch+0xcd/0x290
> > [<ffffffff8109ca3e>] __rcu_process_callbacks+0x6e/0xe0
> > [<ffffffff8109cbca>] rcu_needs_cpu+0x11a/0x170
> > [<ffffffff8107c14e>] tick_nohz_stop_sched_tick+0x15e/0x440
> > [<ffffffff81001dc9>] cpu_idle+0x79/0x120
> > [<ffffffff817bb627>] start_secondary+0xa0/0xa2
> > ---[ end trace 155c62ea9b561096 ]---
> >
> > Config attached.
>
> Color me confused!
>
> rcu_needs_cpu() is supposed to be called with irqs disabled, and
> tick_nohz_stop_sched_tick() does in fact disable them with
> local_irq_save() near the beginning of the function. Doing a quick
> inspection, starting at that point in tick_nohz_stop_sched_tick():
>
> o smp_processor_id() does not mess with irq, nor does per_cpu().
>
> o tick_nohz_start_idle() calls a bunch of things.
> sched_clock_cpu() checks for irqs being disabled, but
> only if CONFIG_HAVE_UNSTABLE_SCHED_CLOCK. Which you have
> set. So we know irqs remained disabled at this point.
>
> And I don't see anything re-enabling irqs in the subsequent
> code path in this function.
>
> o need_resched() just checks the TIF_NEED_RESCHED flag.
>
> o Neither local_softirq_pending() and cpu_online() mess
> with irq enabling.
>
> o The code path containing the printk() was apparently not
> taken, as there is no message in your log.
>
> o read_seqbegin() and read_seqretry() leave irqs alone, as
> does timekeeping_max_deferment().
>
> And that puts us at the call to rcu_needs_cpu(). You have the
> new CONFIG_RCU_FAST_NO_HZ config variable set, and are not running
> preemptible RCU, so we are in the one at line 993 of rcutree_plugin.c.
> The fact that __rcu_process_callbacks() is on the stack means that all
> other CPUs were in dyntick-idle mode, so we went through the loop.
>
> o rcu_sched_qs() doesn't mess with irqs.
>
> o force_quiescent_state() does mess with irqs, but puts them
> back the way it found them.
>
> o Ditto for __rcu_process_callbacks().
>
> So I am reduced to putting together a diagnostic patch for you. :-/
-EICANTREAD
Commit 8bd93a2c ("Accelerate grace period...") is busted. I will work
out how to fix it.
Thanx, Paul
next prev parent reply other threads:[~2010-02-25 21:36 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-23 1:04 [PATCH tip/core/rcu 0/21] v6 add lockdep-based diagnostics to rcu_dereference() Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 01/21] rcu: introduce lockdep-based checking to RCU read-side primitives Paul E. McKenney
2010-02-25 10:09 ` [tip:core/rcu] rcu: Introduce " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 02/21] rcu: add lockdep-enabled variants of rcu_dereference() Paul E. McKenney
2010-02-25 10:09 ` [tip:core/rcu] rcu: Add " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 03/21] rcu: integrate rcu_dereference_check() message into lockdep Paul E. McKenney
2010-02-25 10:09 ` [tip:core/rcu] rcu: Integrate " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 04/21] rcu: disable lockdep checking in RCU list-traversal primitives Paul E. McKenney
2010-02-25 10:10 ` [tip:core/rcu] rcu: Disable " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 05/21] net: add checking to rcu_dereference() primitives Paul E. McKenney
2010-02-25 10:10 ` [tip:core/rcu] net: Add " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 06/21] sched: use lockdep-based checking on rcu_dereference() Paul E. McKenney
2010-02-25 10:11 ` [tip:core/rcu] sched: Use " tip-bot for Paul E. McKenney
2010-02-25 11:06 ` [tip:core/rcu] sched, cgroups: Fix module export tip-bot for Ingo Molnar
2010-02-23 1:04 ` [PATCH tip/core/rcu 07/21] sched: better name for for_each_domain_rd Paul E. McKenney
2010-02-25 10:11 ` [tip:core/rcu] sched: Better " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 08/21] vfs: apply lockdep-based checking to rcu_dereference() uses Paul E. McKenney
2010-02-25 10:11 ` [tip:core/rcu] vfs: Apply " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 09/21] vfs: abstract rcu_dereference_check for files-fdtable use Paul E. McKenney
2010-02-25 10:11 ` [tip:core/rcu] vfs: Abstract " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 10/21] radix-tree: disable RCU lockdep checking in radix tree Paul E. McKenney
2010-02-25 10:12 ` [tip:core/rcu] radix-tree: Disable " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 11/21] idr: apply lockdep-based diagnostics to rcu_dereference() uses Paul E. McKenney
2010-02-25 10:12 ` [tip:core/rcu] idr: Apply " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 12/21] security: apply lockdep-based checking " Paul E. McKenney
2010-02-25 10:12 ` [tip:core/rcu] security: Apply " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 13/21] rcu: documentation update for CONFIG_PROVE_RCU Paul E. McKenney
2010-02-25 10:12 ` [tip:core/rcu] rcu: Documentation " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 14/21] rcu: fix citation of Mathieu's dissertation Paul E. McKenney
2010-02-25 10:13 ` [tip:core/rcu] rcu: Fix " tip-bot for Paul E. McKenney
2010-02-23 1:04 ` [PATCH tip/core/rcu 15/21] rcu: accelerate grace period if last non-dynticked CPU Paul E. McKenney
2010-02-25 10:13 ` [tip:core/rcu] rcu: Accelerate " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 16/21] rcu: use canonical URL for Mathieu's dissertation Paul E. McKenney
2010-02-25 10:13 ` [tip:core/rcu] rcu: Use " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 17/21] rcu: stop overflowing signed integers Paul E. McKenney
2010-02-25 10:14 ` [tip:core/rcu] rcu: Stop " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 18/21] rcu: Convert to raw_spinlocks Paul E. McKenney
2010-02-25 10:14 ` [tip:core/rcu] " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 19/21] rcu: fix deadlock in TREE_PREEMPT_RCU CPU stall detection Paul E. McKenney
2010-02-25 10:14 ` [tip:core/rcu] rcu: Fix " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 20/21] rcu: fix rcutorture mod_timer argument to delay one jiffy Paul E. McKenney
2010-02-25 10:14 ` [tip:core/rcu] rcu: Fix " tip-bot for Paul E. McKenney
2010-02-23 1:05 ` [PATCH tip/core/rcu 21/21] rcu: add RCU_CPU_STALL_VERBOSE to dump detailed per-task information Paul E. McKenney
2010-02-25 10:15 ` [tip:core/rcu] rcu: Add " tip-bot for Paul E. McKenney
2010-02-23 12:59 ` [PATCH tip/core/rcu 0/21] v6 add lockdep-based diagnostics to rcu_dereference() Arnd Bergmann
2010-02-23 13:15 ` Mathieu Desnoyers
2010-02-23 14:35 ` Paul E. McKenney
2010-02-23 15:54 ` Arnd Bergmann
2010-02-23 16:16 ` Paul E. McKenney
2010-02-23 17:15 ` Arnd Bergmann
2010-02-23 18:01 ` Paul E. McKenney
2010-02-24 20:03 ` [PATCH 00/10] __rcu annotations, first draft Arnd Bergmann
2010-02-24 22:18 ` Paul E. McKenney
2010-02-25 8:37 ` Ingo Molnar
2010-02-24 20:03 ` [PATCH 01/10] rcu: define __rcu address space modifier for sparse Arnd Bergmann
2010-02-24 20:12 ` Mathieu Desnoyers
2010-02-24 20:22 ` Arnd Bergmann
2010-02-24 20:03 ` [PATCH 02/10] rcu: annotated list rcu code Arnd Bergmann
2010-02-24 20:15 ` Mathieu Desnoyers
2010-02-24 20:32 ` Arnd Bergmann
2010-02-24 20:03 ` [PATCH 03/10] cgroups: __rcu annotations Arnd Bergmann
2010-02-24 20:04 ` [PATCH 04/10] credentials: rcu annotation Arnd Bergmann
2010-02-24 20:04 ` [PATCH 05/10] perf_event: __rcu annotations Arnd Bergmann
2010-02-24 20:04 ` [PATCH 06/10] audit: " Arnd Bergmann
2010-02-24 20:04 ` [PATCH 07/10] module: " Arnd Bergmann
2010-02-24 20:13 ` Alexey Dobriyan
2010-02-24 20:26 ` Arnd Bergmann
2010-02-24 22:17 ` Paul E. McKenney
2010-02-24 23:07 ` Arnd Bergmann
2010-02-24 23:59 ` Paul E. McKenney
2010-02-25 17:06 ` Paul E. McKenney
2010-02-25 18:10 ` Arnd Bergmann
2010-02-25 20:05 ` Paul E. McKenney
2010-02-26 2:12 ` Paul E. McKenney
2010-02-24 20:04 ` [PATCH 08/10] pid: " Arnd Bergmann
2010-02-24 20:04 ` [PATCH 09/10] notifiers: " Arnd Bergmann
2010-02-24 20:04 ` [PATCH 10/10] scheduler: " Arnd Bergmann
2010-02-23 13:28 ` [PATCH tip/core/rcu 0/21] v6 add lockdep-based diagnostics to rcu_dereference() Paul E. McKenney
2010-02-25 10:00 ` Ingo Molnar
2010-02-25 10:01 ` Ingo Molnar
2010-02-25 12:04 ` Ingo Molnar
2010-02-25 18:18 ` Paul E. McKenney
2010-02-25 21:36 ` Paul E. McKenney [this message]
2010-02-25 21:22 ` Ingo Molnar
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=20100225213633.GA5936@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=dvhltc@us.ibm.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=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;
as well as URLs for NNTP newsgroup(s).