linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mikulas Patocka <mpatocka@redhat.com>
To: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jens Axboe <axboe@kernel.dk>,
	linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH] percpu-rwsem: use barrier in unlock path
Date: Thu, 18 Oct 2012 11:32:30 -0400 (EDT)	[thread overview]
Message-ID: <Pine.LNX.4.64.1210181127460.22996@file.rdu.redhat.com> (raw)
In-Reply-To: <507F66F6.20704@cn.fujitsu.com>



On Thu, 18 Oct 2012, Lai Jiangshan wrote:

> On 10/18/2012 04:28 AM, Steven Rostedt wrote:
> > On Wed, Oct 17, 2012 at 11:07:21AM -0400, Mikulas Patocka wrote:
> >>>
> >>> Even the previous patch is applied, percpu_down_read() still
> >>> needs mb() to pair with it.
> >>
> >> percpu_down_read uses rcu_read_lock which should guarantee that memory 
> >> accesses don't escape in front of a rcu-protected section.
> > 
> > You do realize that rcu_read_lock() does nothing more that a barrier(),
> > right?
> > 
> > Paul worked really hard to get rcu_read_locks() to not call HW barriers.
> > 
> >>
> >> If rcu_read_unlock has only an unlock barrier and not a full barrier, 
> >> memory accesses could be moved in front of rcu_read_unlock and reordered 
> >> with this_cpu_inc(*p->counters), but it doesn't matter because 
> >> percpu_down_write does synchronize_rcu(), so it never sees these accesses 
> >> halfway through.
> > 
> > Looking at the patch, you are correct. The read side doesn't need the
> > memory barrier as the worse thing that will happen is that it sees the
> > locked = false, and will just grab the mutex unnecessarily.
> 
> ---------------------
> A memory barrier can be added iff these two things are known:
> 	1) it disables the disordering between what and what.
> 	2) what is the corresponding mb() that it pairs with.
> 
> You tried to add a mb() in percpu_up_write(), OK, I know it disables the disordering
> between the writes to the protected data and the statement "p->locked = false",
> But I can't find out the corresponding mb() that it pairs with.

Or alternativelly, instead of barrier, we can do this.

Mikulas

---

percpu-rwsem: use barrier in unlock path

The lock is considered unlocked when p->locked is set to false.
Use barrier prevent reordering of operations around p->locked.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>

---
 include/linux/percpu-rwsem.h |    6 ++++++
 1 file changed, 6 insertions(+)

Index: linux-3.6.2-fast/include/linux/percpu-rwsem.h
===================================================================
--- linux-3.6.2-fast.orig/include/linux/percpu-rwsem.h	2012-10-17 20:48:40.000000000 +0200
+++ linux-3.6.2-fast/include/linux/percpu-rwsem.h	2012-10-18 17:19:24.000000000 +0200
@@ -66,6 +66,12 @@ static inline void percpu_down_write(str
 
 static inline void percpu_up_write(struct percpu_rw_semaphore *p)
 {
+	/*
+	 * Make sure that other processes that are in rcu section and that
+	 * may have seen partially modified state exit the rcu section and
+	 * try to grab the mutex.
+	 */
+	synchronize_rcu();
 	p->locked = false;
 	mutex_unlock(&p->mtx);
 }

  parent reply	other threads:[~2012-10-18 15:33 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Pine.LNX.4.64.1210151716310.10685@file.rdu.redhat.com>
     [not found] ` <Pine.LNX.4.64.1210161924350.20581@file.rdu.redhat.com>
2012-10-17  2:23   ` [PATCH] percpu-rwsem: use barrier in unlock path Linus Torvalds
2012-10-17  5:58     ` Lai Jiangshan
2012-10-17  5:58       ` Lai Jiangshan
2012-10-17 15:07       ` Mikulas Patocka
2012-10-17 20:28         ` Steven Rostedt
2012-10-17 20:28           ` Steven Rostedt
2012-10-18  2:18           ` Lai Jiangshan
2012-10-18  4:13             ` Steven Rostedt
2012-10-18 16:17               ` Mikulas Patocka
2012-10-18 15:32             ` Mikulas Patocka [this message]
2012-10-18 19:56             ` Mikulas Patocka
2012-10-18 16:05           ` Mikulas Patocka
2012-10-17  9:56     ` Alan Cox
2012-10-18 16:00     ` Mikulas Patocka
2012-10-19 18:48       ` Linus Torvalds

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=Pine.LNX.4.64.1210181127460.22996@file.rdu.redhat.com \
    --to=mpatocka@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=eric.dumazet@gmail.com \
    --cc=laijs@cn.fujitsu.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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).