linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Waiman Long <Waiman.Long@hp.com>, Ingo Molnar <mingo@elte.hu>,
	Andrew Morton <akpm@linux-foundation.org>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Alex Shi <alex.shi@linaro.org>, Andi Kleen <andi@firstfloor.org>,
	Michel Lespinasse <walken@google.com>,
	Davidlohr Bueso <davidlohr.bueso@hp.com>,
	Matthew R Wilcox <matthew.r.wilcox@intel.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Rik van Riel <riel@redhat.com>,
	Peter Hurley <peter@hurleysoftware.com>,
	linux-kernel@vger.kernel.org, linux-mm <linux-mm@kvack.org>
Subject: Re: [PATCH v6 5/6] MCS Lock: Restructure the MCS lock defines and locking code into its own file
Date: Fri, 27 Sep 2013 13:38:58 -0700	[thread overview]
Message-ID: <20130927203858.GB9093@linux.vnet.ibm.com> (raw)
In-Reply-To: <1380310733.3467.118.camel@schen9-DESK>

On Fri, Sep 27, 2013 at 12:38:53PM -0700, Tim Chen wrote:
> On Fri, 2013-09-27 at 08:29 -0700, Paul E. McKenney wrote:
> > On Wed, Sep 25, 2013 at 03:10:49PM -0700, Tim Chen wrote:
> > > We will need the MCS lock code for doing optimistic spinning for rwsem.
> > > Extracting the MCS code from mutex.c and put into its own file allow us
> > > to reuse this code easily for rwsem.
> > > 
> > > Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
> > > Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
> > > ---
> > >  include/linux/mcslock.h |   58 +++++++++++++++++++++++++++++++++++++++++++++++
> > >  kernel/mutex.c          |   58 +++++-----------------------------------------
> > >  2 files changed, 65 insertions(+), 51 deletions(-)
> > >  create mode 100644 include/linux/mcslock.h
> > > 
> > > diff --git a/include/linux/mcslock.h b/include/linux/mcslock.h
> > > new file mode 100644
> > > index 0000000..20fd3f0
> > > --- /dev/null
> > > +++ b/include/linux/mcslock.h
> > > @@ -0,0 +1,58 @@
> > > +/*
> > > + * MCS lock defines
> > > + *
> > > + * This file contains the main data structure and API definitions of MCS lock.
> > > + */
> > > +#ifndef __LINUX_MCSLOCK_H
> > > +#define __LINUX_MCSLOCK_H
> > > +
> > > +struct mcs_spin_node {
> > > +	struct mcs_spin_node *next;
> > > +	int		  locked;	/* 1 if lock acquired */
> > > +};
> > > +
> > > +/*
> > > + * We don't inline mcs_spin_lock() so that perf can correctly account for the
> > > + * time spent in this lock function.
> > > + */
> > > +static noinline
> > > +void mcs_spin_lock(struct mcs_spin_node **lock, struct mcs_spin_node *node)
> > > +{
> > > +	struct mcs_spin_node *prev;
> > > +
> > > +	/* Init node */
> > > +	node->locked = 0;
> > > +	node->next   = NULL;
> > > +
> > > +	prev = xchg(lock, node);
> > > +	if (likely(prev == NULL)) {
> > > +		/* Lock acquired */
> > > +		node->locked = 1;
> > > +		return;
> > > +	}
> > > +	ACCESS_ONCE(prev->next) = node;
> > > +	smp_wmb();
> 
> BTW, is the above memory barrier necessary?  It seems like the xchg
> instruction already provided a memory barrier.
> 
> Now if we made the changes that Jason suggested:
> 
> 
>         /* Init node */
> -       node->locked = 0;
>         node->next   = NULL;
> 
>         prev = xchg(lock, node);
>         if (likely(prev == NULL)) {
>                 /* Lock acquired */
> -               node->locked = 1;
>                 return;
>         }
> +       node->locked = 0;
>         ACCESS_ONCE(prev->next) = node;
>         smp_wmb();
> 
> We are probably still okay as other cpus do not read the value of
> node->locked, which is a local variable.

I don't immediately see the need for the smp_wmb() in either case.

> Tim
> 
> > > +	/* Wait until the lock holder passes the lock down */
> > > +	while (!ACCESS_ONCE(node->locked))
> > > +		arch_mutex_cpu_relax();

However, you do need a full memory barrier here in order to ensure that
you see the effects of the previous lock holder's critical section.

							Thanx, Paul

> > > +}
> > > +
> > > +static void mcs_spin_unlock(struct mcs_spin_node **lock, struct mcs_spin_node *node)
> > > +{
> > > +	struct mcs_spin_node *next = ACCESS_ONCE(node->next);
> > > +
> > > +	if (likely(!next)) {
> > > +		/*
> > > +		 * Release the lock by setting it to NULL
> > > +		 */
> > > +		if (cmpxchg(lock, node, NULL) == node)
> > > +			return;
> > > +		/* Wait until the next pointer is set */
> > > +		while (!(next = ACCESS_ONCE(node->next)))
> > > +			arch_mutex_cpu_relax();
> > > +	}
> > > +	ACCESS_ONCE(next->locked) = 1;
> > > +	smp_wmb();
> > 
> > Shouldn't the memory barrier precede the "ACCESS_ONCE(next->locked) = 1;"?
> > Maybe in an "else" clause of the prior "if" statement, given that the
> > cmpxchg() does it otherwise.
> > 
> > Otherwise, in the case where the "if" conditionn is false, the critical
> > section could bleed out past the unlock.
> > 
> > 							Thanx, Paul
> > 
> 
> 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2013-09-27 20:39 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <cover.1380144003.git.tim.c.chen@linux.intel.com>
2013-09-25 22:10 ` [PATCH v6 0/6] rwsem: performance optimizations Tim Chen
2013-09-25 22:10 ` [PATCH v6 1/6] rwsem: check the lock before cpmxchg in down_write_trylock Tim Chen
2013-09-25 22:10 ` [PATCH v6 2/6] rwsem: remove 'out' label in do_wake Tim Chen
2013-09-25 22:10 ` [PATCH v6 3/6] rwsem: remove try_reader_grant label do_wake Tim Chen
2013-09-25 22:10 ` [PATCH v6 4/6] rwsem/wake: check lock before do atomic update Tim Chen
2013-09-25 22:10 ` [PATCH v6 5/6] MCS Lock: Restructure the MCS lock defines and locking code into its own file Tim Chen
2013-09-26  6:46   ` Ingo Molnar
2013-09-26  8:40     ` Peter Zijlstra
2013-09-26  9:37       ` Ingo Molnar
2013-09-26 18:18       ` Tim Chen
2013-09-26 19:27   ` Jason Low
2013-09-26 20:06     ` Davidlohr Bueso
2013-09-26 20:23       ` Jason Low
2013-09-26 20:40         ` Davidlohr Bueso
2013-09-26 21:09           ` Jason Low
2013-09-26 21:41             ` Tim Chen
2013-09-26 22:42               ` Jason Low
2013-09-26 22:57                 ` Tim Chen
2013-09-27  6:02                   ` Ingo Molnar
2013-09-27  6:26                     ` Jason Low
2013-09-27 11:23                     ` Peter Zijlstra
2013-09-27 13:44                       ` Joe Perches
2013-09-27 13:48                         ` Peter Zijlstra
2013-09-27 14:05                           ` Joe Perches
2013-09-27 14:18                             ` Peter Zijlstra
2013-09-27 14:14                           ` [PATCH] checkpatch: Make the memory barrier test noisier Joe Perches
2013-09-27 14:26                             ` Peter Zijlstra
2013-09-27 14:34                               ` Joe Perches
2013-09-27 14:50                                 ` Peter Zijlstra
2013-09-27 15:17                                   ` Paul E. McKenney
2013-09-27 15:34                                     ` Peter Zijlstra
2013-09-27 16:04                                       ` Paul E. McKenney
2013-09-27 23:40                                   ` Oliver Neukum
2013-09-28  7:54                                     ` Peter Zijlstra
2013-09-27 16:12                     ` [PATCH v6 5/6] MCS Lock: Restructure the MCS lock defines and locking code into its own file Jason Low
2013-09-27 16:19                       ` Tim Chen
2013-10-02 19:19                 ` Waiman Long
2013-10-02 19:30                   ` Jason Low
2013-10-02 19:37                     ` Waiman Long
2013-09-26 22:22             ` Davidlohr Bueso
2013-09-27 15:29   ` Paul E. McKenney
2013-09-27 18:09     ` Tim Chen
2013-09-28  2:58       ` Waiman Long
2013-09-27 19:38     ` Tim Chen
2013-09-27 20:16       ` Jason Low
2013-09-27 20:38       ` Paul E. McKenney [this message]
2013-09-27 22:46         ` Tim Chen
2013-09-27 23:01           ` Paul E. McKenney
2013-09-27 23:54             ` Jason Low
2013-09-28  0:02               ` Davidlohr Bueso
2013-09-28  2:19               ` Paul E. McKenney
2013-09-28  4:34                 ` Jason Low
2013-09-30 15:51                   ` Waiman Long
2013-09-30 16:10                     ` Jason Low
2013-09-30 16:36                       ` Waiman Long
2013-10-01 16:48                         ` Tim Chen
2013-10-01 20:01                           ` Waiman Long
2013-10-01 21:16                             ` Tim Chen
2013-10-02  1:25                               ` Waiman Long
2013-10-02 18:43                                 ` Tim Chen
2013-10-02 19:32                                   ` Waiman Long
2013-09-30 16:28                 ` Tim Chen
2013-09-25 22:10 ` [PATCH v6 6/6] rwsem: do optimistic spinning for writer lock acquisition Tim Chen
2013-09-26  6:53   ` 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=20130927203858.GB9093@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=Waiman.Long@hp.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=alex.shi@linaro.org \
    --cc=andi@firstfloor.org \
    --cc=dave.hansen@intel.com \
    --cc=davidlohr.bueso@hp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matthew.r.wilcox@intel.com \
    --cc=mingo@elte.hu \
    --cc=peter@hurleysoftware.com \
    --cc=riel@redhat.com \
    --cc=tim.c.chen@linux.intel.com \
    --cc=walken@google.com \
    /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).