From: Frederic Weisbecker <fweisbec@gmail.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Eric Dumazet <eric.dumazet@gmail.com>,
linux-kernel@vger.kernel.org, mingo@elte.hu,
laijs@cn.fujitsu.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
josh@joshtriplett.org, niv@us.ibm.com, tglx@linutronix.de,
dhowells@redhat.com, edumazet@google.com, darren@dvhart.com,
sbw@mit.edu
Subject: Re: [PATCH] rcu: Is it safe to enter an RCU read-side critical section?
Date: Mon, 9 Sep 2013 14:45:49 +0200 [thread overview]
Message-ID: <20130909124547.GB16280@somewhere> (raw)
In-Reply-To: <20130909083926.3eceebef@gandalf.local.home>
On Mon, Sep 09, 2013 at 08:39:26AM -0400, Steven Rostedt wrote:
> On Mon, 9 Sep 2013 14:13:31 +0200
> Frederic Weisbecker <fweisbec@gmail.com> wrote:
>
>
> > > In any case the preempt_disable/enable pair there is just plain wrong as
> > > Eric pointed out.
> >
> > Check this:
> >
> > 34240697d619c439c55f21989680024dcb604aab "rcu: Disable preemption in rcu_is_cpu_idle()"
>
>
> Ug, and that patch does nothing to fix the bug that it reported!
>
> 1. Task A on CPU 1 enters rcu_is_cpu_idle() and picks up the
> pointer to CPU 1's per-CPU variables.
>
> 2. Task B preempts Task A and starts running on CPU 1.
>
> Let's say that B preempts Task A here:
>
> preempt_disable();
> ret = (atomic_read(&__get_cpu_var(rcu_dynticks).dynticks) & 0x1) == 0;
> preempt_enable();
> <preempt>
> return ret;
>
>
> 3. Task A migrates to CPU 2.
>
> 4. Task B blocks, leaving CPU 1 idle.
>
> 5. Task A continues execution on CPU 2, accessing CPU 1's
> dyntick-idle information using the pointer fetched in step 1 above,
> and finds that CPU 1 is idle.
>
> Yeah, and Task A is using the "ret" from CPU 1!
>
> 6. Task A therefore incorrectly concludes that it is executing in
> an extended quiescent state, possibly issuing a spurious splat.
>
> Therefore, this commit disables preemption within the
> rcu_is_cpu_idle() function.
>
> Where this commit is totally bogus. Sorry, but it is.
>
> This just proves that the caller of rcu_is_cpu_idle() must disable
> preemption itself for the entire time that it needs to use the result
> of rcu_is_cpu_idle().
Sorry, I don't understand your point here. What's wrong with checking the
ret from another CPU?
next prev parent reply other threads:[~2013-09-09 12:45 UTC|newest]
Thread overview: 98+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-05 19:52 [PATCH] rcu: Is it safe to enter an RCU read-side critical section? Paul E. McKenney
2013-09-05 20:25 ` Steven Rostedt
2013-09-05 20:59 ` Paul E. McKenney
2013-09-05 21:05 ` Paul E. McKenney
2013-09-05 23:40 ` Steven Rostedt
2013-09-06 10:59 ` Frederic Weisbecker
2013-09-06 15:18 ` Paul E. McKenney
2013-09-06 15:33 ` Steven Rostedt
2013-09-06 16:40 ` Frederic Weisbecker
2013-09-06 16:52 ` Steven Rostedt
2013-09-06 16:58 ` Paul E. McKenney
2013-09-06 17:00 ` Frederic Weisbecker
2013-09-06 17:16 ` Steven Rostedt
2013-09-06 17:52 ` Paul E. McKenney
2013-09-06 17:56 ` Paul E. McKenney
2013-09-06 18:21 ` Steven Rostedt
2013-09-07 0:49 ` Paul E. McKenney
2013-09-07 1:19 ` Mathieu Desnoyers
2013-09-08 1:55 ` Paul E. McKenney
2013-09-09 10:56 ` Peter Zijlstra
2013-09-06 17:21 ` Eric Dumazet
2013-09-06 17:41 ` Paul E. McKenney
2013-09-06 18:59 ` Frederic Weisbecker
2013-09-06 20:38 ` Paul E. McKenney
2013-09-09 10:53 ` Peter Zijlstra
2013-09-09 12:13 ` Frederic Weisbecker
2013-09-09 12:39 ` Steven Rostedt
2013-09-09 12:45 ` Frederic Weisbecker [this message]
2013-09-09 12:55 ` Steven Rostedt
2013-09-09 13:08 ` Frederic Weisbecker
2013-09-09 13:21 ` Steven Rostedt
2013-09-09 13:29 ` Paul E. McKenney
2013-09-09 13:29 ` Steven Rostedt
2013-09-09 13:37 ` Peter Zijlstra
2013-09-09 13:48 ` Paul E. McKenney
2013-09-09 14:40 ` Frederic Weisbecker
2013-09-09 15:20 ` Steven Rostedt
2013-09-09 15:39 ` Steven Rostedt
2013-09-09 16:03 ` Frederic Weisbecker
2013-09-09 16:09 ` Paul E. McKenney
2013-09-09 16:30 ` Steven Rostedt
2013-09-09 16:56 ` Paul E. McKenney
2013-09-09 16:21 ` Peter Zijlstra
2013-09-09 13:45 ` Frederic Weisbecker
2013-09-09 13:56 ` Paul E. McKenney
2013-09-09 14:16 ` Steven Rostedt
2013-09-09 16:17 ` Paul E. McKenney
2013-09-09 16:34 ` Steven Rostedt
2013-09-09 16:58 ` Paul E. McKenney
2013-09-09 17:06 ` Steven Rostedt
2013-09-09 17:45 ` Paul E. McKenney
2013-09-09 17:29 ` Mathieu Desnoyers
2013-09-09 17:56 ` Paul E. McKenney
2013-09-09 18:36 ` Steven Rostedt
2013-09-09 18:50 ` Paul E. McKenney
2013-09-09 21:40 ` Mathieu Desnoyers
2013-09-09 21:59 ` Steven Rostedt
2013-09-09 22:34 ` Paul E. McKenney
2013-09-11 14:13 ` Paul E. McKenney
2013-09-11 14:26 ` Steven Rostedt
2013-09-11 15:23 ` Paul E. McKenney
2013-09-11 15:49 ` Steven Rostedt
2013-09-11 16:03 ` Paul E. McKenney
2013-09-09 13:14 ` Peter Zijlstra
2013-09-09 13:29 ` Frederic Weisbecker
2013-09-09 13:41 ` Steven Rostedt
2013-09-09 13:49 ` Frederic Weisbecker
2013-09-09 13:50 ` Paul E. McKenney
2013-09-09 13:46 ` Paul E. McKenney
2013-09-09 13:55 ` Steven Rostedt
2013-09-09 16:22 ` Paul E. McKenney
2013-09-09 16:40 ` Steven Rostedt
2013-09-09 17:45 ` Paul E. McKenney
2013-09-09 13:23 ` Paul E. McKenney
2013-09-09 13:36 ` Peter Zijlstra
2013-09-09 13:53 ` Paul E. McKenney
2013-09-09 16:18 ` Peter Zijlstra
2013-09-09 14:49 ` Christoph Lameter
2013-09-09 15:08 ` Peter Zijlstra
2013-09-09 15:24 ` Christoph Lameter
2013-09-09 15:41 ` Steven Rostedt
2013-09-09 15:47 ` Steven Rostedt
2013-09-09 16:00 ` Ingo Molnar
2013-09-09 16:03 ` Steven Rostedt
2013-09-09 16:11 ` Ingo Molnar
2013-09-10 21:37 ` Christoph Lameter
2013-09-12 6:39 ` Ingo Molnar
2013-09-12 14:20 ` Christoph Lameter
2013-09-10 21:28 ` Christoph Lameter
2013-09-12 6:38 ` Ingo Molnar
2013-09-12 14:43 ` Christoph Lameter
2013-09-09 16:15 ` Peter Zijlstra
2013-09-10 4:07 ` Mike Galbraith
2013-09-09 13:36 ` Steven Rostedt
2013-09-09 14:21 ` Peter Zijlstra
2013-09-09 16:26 ` Paul E. McKenney
2013-09-09 16:42 ` Steven Rostedt
2013-09-09 16:59 ` 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=20130909124547.GB16280@somewhere \
--to=fweisbec@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=darren@dvhart.com \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edumazet@google.com \
--cc=eric.dumazet@gmail.com \
--cc=josh@joshtriplett.org \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@elte.hu \
--cc=niv@us.ibm.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=sbw@mit.edu \
--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