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>,
	Peter Zijlstra <peterz@infradead.org>
Subject: Re: [PATCH 11/16] rcu: Check for spurious wakeup using return value
Date: Thu, 24 Jul 2014 13:27:55 -0700	[thread overview]
Message-ID: <20140724202755.GU11241@linux.vnet.ibm.com> (raw)
In-Reply-To: <53D165A5.6010206@gmail.com>

On Thu, Jul 24, 2014 at 03:59:33PM -0400, Pranith Kumar wrote:
> Adding peterz to CC as git blames him for wait_event code. :)
> 
> (original LKML link: https://lkml.org/lkml/2014/7/23/45)
> 
> On Thu, Jul 24, 2014 at 2:12 PM, Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:

[ . . . ]

> >> 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
> 
> In the grand scheme of things, I agree that minor optimizations may not seem
> to be worth much. But when the optimizationss are straight forward and 
> are _actually_ improving things, even by a small margin, I think they are
> worth considering.
> 
> Think of the billions of cycles we will save ;-)

If there are significant savings.  You have yet to demonstrate this.
In fact, you have yet to demonstrate that your change doesn't make
things worse.

							Thanx, Paul

> > -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?)
> >
> 
> 
> -- 
> Pranith
> 


  reply	other threads:[~2014-07-24 20:28 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
2014-07-24 19:59             ` Pranith Kumar
2014-07-24 20:27               ` Paul E. McKenney [this message]
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=20140724202755.GU11241@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=peterz@infradead.org \
    --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).