linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Pranith Kumar <bobby.prani@gmail.com>
Cc: Josh Triplett <josh@joshtriplett.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Lai Jiangshan <laijs@cn.fujitsu.com>,
	"open list:READ-COPY UPDATE..." <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 11/16] rcu: Check for spurious wakeup using return value
Date: Thu, 24 Jul 2014 11:12:23 -0700	[thread overview]
Message-ID: <20140724181223.GO11241@linux.vnet.ibm.com> (raw)
In-Reply-To: <CAJhHMCDhwnKF5tiWvHyGrz3KhCab+QUnAfEmkzjsBrNn5O3gyQ@mail.gmail.com>

On Thu, Jul 24, 2014 at 12:03:34AM -0400, Pranith Kumar wrote:
> On Wed, Jul 23, 2014 at 11:43 PM, Paul E. McKenney
> <paulmck@linux.vnet.ibm.com> wrote:
> > On Wed, Jul 23, 2014 at 10:36:19PM -0400, Pranith Kumar wrote:
> >> On Wed, Jul 23, 2014 at 8:26 AM, Paul E. McKenney
> >> <paulmck@linux.vnet.ibm.com> wrote:
> >> > On Wed, Jul 23, 2014 at 01:09:48AM -0400, Pranith Kumar wrote:
> >> >> When the gp_kthread wakes up from the wait event, it returns 0 if the wake up is
> >> >> due to the condition having been met. This commit checks this return value
> >> >> for a spurious wake up before calling rcu_gp_init().
> >> >>
> >> >> Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
> >> >
> >> > How does this added check help?  I don't see that it does.  If the flag
> >> > is set, we want to wake up.  If we get a spurious wakeup, but then the
> >> > flag gets set before we actually wake up, we still want to wake up.
> >>
> >> So I took a look at the docs again, and using the return value is the
> >> recommended way to check for spurious wakeups.
> >>
> >> The condition in wait_event_interruptible() is checked when the task
> >> is woken up (either due to stray signals or explicitly) and it returns
> >> true if condition evaluates to true.
> 
> this should be returns '0' if the condition evaluates to true.

Ah, but if the condition changes while wait_event_interruptible() is in
the process of returning, it is quite possible that the answer will be
different afterwards.  For example, consider this scenario:

o	wait_event_interruptible() decides to return spuriously for
	whatever reason, and thus returns a non-zero value.

o	An invocation of (say) rcu_start_gp_advanced() sets ->gp_flags
	to RCU_GP_FLAG_INIT, thus requesting that a new grace period
	start.

o	At this point, the return value says that we should not start
	a new grace period, but the ->gp_flags value says that we
	should.

Because it is the ->gp_flags value that really knows, the current code
ignores wait_event_interruptible()'s return value and just checks
the ->gp_flags value.

> >> In the current scenario, if we get a spurious wakeup, we take the
> >> costly path of checking this condition again (with a barrier and lock)
> >> before going back to wait.
> >>
> >> The scenario of getting an actual wakeup after getting a spurious
> >> wakeup exists even today, this is the window after detecting a
> >> spurious wakeup and before going back to wait. I am not sure if using
> >> the return value enlarges that window as we are going back to sleep
> >> immediately.
> >>
> >> Thoughts?
> >
> > If the flag is set, why should we care whether or not the wakeup was
> > spurious?  If the flag is not set, why should we care whether or not
> > wait_event_interruptible() thought that the wakeup was not spurious?
> 
> A correction about the return value above: return will be 0 if the
> condition is true, in this case if the flag is set.
> 
> If the flag is set, ret will be 0 and we will go ahead with
> rcu_gp_init(). (no change wrt current behavior)

Sorry, this is not always correct.  RCU is highly concurrent, so you do
need to start thinking in terms of concurrency.  Your reasoning above
is a symptom of single-threaded thinking.  Please see my scenario above
showing how the return can be non-zero even though ->gp_flags is set.

> If the flag is not set, currently we go ahead and call rcu_gp_init()
> from where we check if the flag is set (after a lock+barrier) and
> return.

True enough.  Is that really a problem?  If so, exactly why is it a
problem?

> If we care about what wait_event_interruptible() returns, we can go
> back and wait for an actual wakeup much earlier without the additional
> overhead of calling rcu_gp_init().

The key phrase here is "If we care".  Should we care?  If so, why?

I suggest running some random benchmark and counting how many times
rcu_gp_init() is called and how many times rcu_gp_init() returns
because ->gp_flags is not set.  If rcu_gp_init() returns because
->gp_flags is not set a significant fraction of the time, then this
-might- be worth worrying about.  (Extra credit: Under what conditions
would it be worth worrying about, and how would you go about checking
to see whether those conditions hold?)

							Thanx, Paul


  reply	other threads:[~2014-07-24 18:12 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-23  5:09 [PATCH 00/16] rcu: Some minor fixes and cleanups Pranith Kumar
2014-07-23  5:09 ` [PATCH 01/16] rcu: Use rcu_num_nodes instead of NUM_RCU_NODES Pranith Kumar
2014-07-23  5:09 ` [PATCH 02/16] rcu: Check return value for cpumask allocation Pranith Kumar
2014-07-23 12:06   ` Paul E. McKenney
2014-07-23 12:49     ` Pranith Kumar
2014-07-23 17:14     ` Pranith Kumar
2014-07-23 18:01       ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 03/16] rcu: Fix comment for gp_state field values Pranith Kumar
2014-07-23  5:09 ` [PATCH 04/16] rcu: Remove redundant check for an online CPU Pranith Kumar
2014-07-23 12:09   ` Paul E. McKenney
2014-07-23 13:23     ` Pranith Kumar
2014-07-23 13:41       ` Paul E. McKenney
2014-07-23 14:01         ` Pranith Kumar
2014-07-23 14:14           ` Paul E. McKenney
2014-07-23 15:07             ` Pranith Kumar
2014-07-23 15:21               ` Pranith Kumar
2014-07-23  5:09 ` [PATCH 05/16] rcu: Add noreturn attribute to boost kthread Pranith Kumar
2014-07-23  5:09 ` [PATCH 06/16] rcu: Clear gp_flags only when actually starting new gp Pranith Kumar
2014-07-23 12:13   ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 07/16] rcu: Save and restore irq flags in rcu_gp_cleanup() Pranith Kumar
2014-07-23 12:16   ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 08/16] rcu: Clean up rcu_spawn_one_boost_kthread() Pranith Kumar
2014-07-23  5:09 ` [PATCH 09/16] rcu: Remove redundant check for online cpu Pranith Kumar
2014-07-23 12:21   ` Paul E. McKenney
2014-07-23 12:59     ` Pranith Kumar
2014-07-23 13:50       ` Paul E. McKenney
2014-07-23 14:12         ` Pranith Kumar
2014-07-23 14:23           ` Paul E. McKenney
2014-07-23 15:11             ` Pranith Kumar
2014-07-23 15:30               ` Paul E. McKenney
2014-07-23 15:44                 ` Pranith Kumar
2014-07-23 19:15                   ` Paul E. McKenney
2014-07-23 20:01                     ` Pranith Kumar
2014-07-23 20:16                     ` Pranith Kumar
2014-07-23 20:23                       ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 10/16] rcu: Check for RCU_FLAG_GP_INIT bit in gp_flags for spurious wakeup Pranith Kumar
2014-07-23 12:23   ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 11/16] rcu: Check for spurious wakeup using return value Pranith Kumar
2014-07-23 12:26   ` Paul E. McKenney
2014-07-24  2:36     ` Pranith Kumar
2014-07-24  3:43       ` Paul E. McKenney
2014-07-24  4:03         ` Pranith Kumar
2014-07-24 18:12           ` Paul E. McKenney [this message]
2014-07-24 19:59             ` Pranith Kumar
2014-07-24 20:27               ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 12/16] rcu: Rename rcu_spawn_gp_kthread() to rcu_spawn_kthreads() Pranith Kumar
2014-07-23  5:09 ` [PATCH 13/16] rcu: Spawn nocb kthreads from rcu_prepare_kthreads() Pranith Kumar
2014-07-23  5:09 ` [PATCH 14/16] rcu: Remove redundant checks for rcu_scheduler_fully_active Pranith Kumar
2014-07-23 12:27   ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 15/16] rcu: Check for a nocb cpu before trying to spawn nocb threads Pranith Kumar
2014-07-23 12:28   ` Paul E. McKenney
2014-07-23 13:14     ` Pranith Kumar
2014-07-23 13:42       ` Paul E. McKenney
2014-07-23  5:09 ` [PATCH 16/16] rcu: kvm.sh: Fix error when you pass --cpus argument Pranith Kumar
2014-07-23 12:31   ` Paul E. McKenney
2014-07-23 14:45 ` [PATCH 00/16] rcu: Some minor fixes and cleanups Paul E. McKenney
2014-08-27  1:10   ` Pranith Kumar
2014-08-27  3:20     ` 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=20140724181223.GO11241@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=bobby.prani@gmail.com \
    --cc=josh@joshtriplett.org \
    --cc=laijs@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=rostedt@goodmis.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 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).