Discussions of the Parallel Programming book
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Akira Yokosawa <akiyks@gmail.com>
Cc: Junchang Wang <junchangwang@gmail.com>, perfbook@vger.kernel.org
Subject: Re: [PATCH 1/2] count_stat_eventual: Switch from ACCESS_ONCE() to READ_ONCE()/WRITE_ONCE()
Date: Sat, 13 May 2017 05:45:56 -0700	[thread overview]
Message-ID: <20170513124556.GJ3956@linux.vnet.ibm.com> (raw)
In-Reply-To: <61acaa88-abd6-58f5-4d1f-80a2ef58d53e@gmail.com>

On Sat, May 13, 2017 at 09:04:38PM +0900, Akira Yokosawa wrote:
> Hi Jason & Paul,
> 
> although this has already been applied, I have a comment.
> 
> On 2017/05/11 23:03:41 +0800, Junchang Wang wrote:
> > Signed-off-by: Junchang Wang <junchangwang@gmail.com>
> > ---
> >  CodeSamples/count/count_stat_eventual.c | 8 ++++----
> >  1 file changed, 4 insertions(+), 4 deletions(-)
> > 
> > diff --git a/CodeSamples/count/count_stat_eventual.c b/CodeSamples/count/count_stat_eventual.c
> > index 059ab8b..cbde4aa 100644
> > --- a/CodeSamples/count/count_stat_eventual.c
> > +++ b/CodeSamples/count/count_stat_eventual.c
> > @@ -27,12 +27,12 @@ int stopflag;
> >  
> >  void inc_count(void)
> >  {
> > -	ACCESS_ONCE(__get_thread_var(counter))++;
> > +	READ_ONCE(__get_thread_var(counter))++;
> 
> This is OK because READ_ONCE() is defined as the same as ACCESS_ONCE()
> in CodeSamples. However, the definition in the current Linux kernel
> would not permit this.
> 
> A read-modify-write access would need both READ_ONCE() and WRITE_ONCE().
> However, since "counter" is thread local and updated only by its owner,
> we don't need READ_ONCE() here. So:
> 
> +	WRITE_ONCE(__get_thread_var(counter), __get_thread_var(counter) + 1);
> 
> should have been sufficient.
> 
> Problem with this change is that the line gets too wide when applied to
> the code snippet in 2-column layout.

Good point -- though renumbering the code is not all -that- hard.

I clearly should have made a better READ_ONCE() that enforced the same
constraints as does the Linux kernel, perhaps something like this:

	#define READ_ONCE(x) ({ ACCESS_ONCE(x) })

Thoughts?

							Thanx, Paul

> Hmm...
> 
>                                  Thanks, Akira
> 
> >  }
> >  
> >  unsigned long read_count(void)
> >  {
> > -	return ACCESS_ONCE(global_count);
> > +	return READ_ONCE(global_count);
> >  }
> >  
> >  void *eventual(void *arg)
> > @@ -43,8 +43,8 @@ void *eventual(void *arg)
> >  	while (stopflag < 3) {
> >  		sum = 0;
> >  		for_each_thread(t)
> > -			sum += ACCESS_ONCE(per_thread(counter, t));
> > -		ACCESS_ONCE(global_count) = sum;
> > +			sum += READ_ONCE(per_thread(counter, t));
> > +		WRITE_ONCE(global_count, sum);
> >  		poll(NULL, 0, 1);
> >  		if (stopflag) {
> >  			smp_mb();
> > 
> 


  reply	other threads:[~2017-05-13 12:46 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-11 15:03 [PATCH 0/2] Use READ_ONCE() and WRITE_ONCE in count_stat_eventual.c Junchang Wang
2017-05-11 15:03 ` [PATCH 1/2] count_stat_eventual: Switch from ACCESS_ONCE() to READ_ONCE()/WRITE_ONCE() Junchang Wang
2017-05-13 12:04   ` Akira Yokosawa
2017-05-13 12:45     ` Paul E. McKenney [this message]
2017-05-13 13:31       ` Junchang Wang
2017-05-13 22:57         ` Paul E. McKenney
2017-05-13 14:37       ` Akira Yokosawa
2017-05-13 22:56         ` Paul E. McKenney
2017-05-14  0:58           ` Akira Yokosawa
2017-05-14  1:31             ` Akira Yokosawa
2017-05-14  4:20               ` Paul E. McKenney
2017-05-14 13:56                 ` Junchang Wang
2017-05-14 15:15                 ` Akira Yokosawa
2017-05-15  0:31                   ` Paul E. McKenney
2017-05-15 14:35                     ` Akira Yokosawa
2017-05-16  4:16                       ` Paul E. McKenney
2017-05-11 15:03 ` [PATCH 2/2] count_stat_eventual: Add READ_ONCE() to protect global shared variable stopflag Junchang Wang
2017-05-14 10:57   ` Akira Yokosawa
2017-05-14 13:28     ` Junchang Wang
2017-05-14 22:57       ` Akira Yokosawa
2017-05-16 13:14         ` Junchang Wang
2017-05-11 16:51 ` [PATCH 0/2] Use READ_ONCE() and WRITE_ONCE in count_stat_eventual.c 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=20170513124556.GJ3956@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=akiyks@gmail.com \
    --cc=junchangwang@gmail.com \
    --cc=perfbook@vger.kernel.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