All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org, mingo@kernel.org,
	laijs@cn.fujitsu.com, dipankar@in.ibm.com,
	akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
	josh@joshtriplett.org, tglx@linutronix.de, rostedt@goodmis.org,
	dhowells@redhat.com, edumazet@google.com, dvhart@linux.intel.com,
	fweisbec@gmail.com, oleg@redhat.com, bobby.prani@gmail.com
Subject: Re: [PATCH RFC tip/core/rcu 2/5] rcu: Short-circuit normal GPs via expedited GPs
Date: Wed, 1 Jul 2015 13:59:06 -0700	[thread overview]
Message-ID: <20150701205906.GQ3717@linux.vnet.ibm.com> (raw)
In-Reply-To: <20150701100352.GO19282@twins.programming.kicks-ass.net>

On Wed, Jul 01, 2015 at 12:03:52PM +0200, Peter Zijlstra wrote:
> On Tue, Jun 30, 2015 at 02:48:27PM -0700, Paul E. McKenney wrote:
> > @@ -2121,17 +2137,24 @@ static int __noreturn rcu_gp_kthread(void *arg)
> >  					       TPS("fqswait"));
> >  			rsp->gp_state = RCU_GP_WAIT_FQS;
> >  			ret = wait_event_interruptible_timeout(rsp->gp_wq,
> > -					((gf = READ_ONCE(rsp->gp_flags)) &
> > -					 RCU_GP_FLAG_FQS) ||
> > -					(!READ_ONCE(rnp->qsmask) &&
> > -					 !rcu_preempt_blocked_readers_cgp(rnp)),
> > -					j);
> > +				((gf = READ_ONCE(rsp->gp_flags)) &
> > +				 RCU_GP_FLAG_FQS) ||
> > +				(!READ_ONCE(rnp->qsmask) &&
> > +				 !rcu_preempt_blocked_readers_cgp(rnp)) ||
> > +				rcu_exp_gp_seq_done(rsp->exp_rsp,
> > +						    rsp->gp_exp_snap),
> > +				j);
> 
> How about using a helper function there?
> 
> static inline bool rcu_gp_done(rsp, rnp)
> {
> 	/* Forced Quiescent State complete */
> 	if (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS)
> 		return true;
> 
> 	/* QS not masked and not blocked by preempted readers */
> 	if (!READ_ONCE(rnp->qsmask) &&
> 	    !rcu_preempt_blocked_readers_cgp(rnp))
> 		return true;
> 
> 	/* Expedited Grace Period completed */
> 	if (rcu_exp_gp_seq_done(rsp))
> 		return true;
> 
> 	return false;
> }
> 
> 	ret = wait_event_interruptible_timeout(rsp->gp_wq,
> 		rcu_gp_done(rsp, rnp), j);

Fair point, and applies to the original as well, give or take comments
and naming.  Please see below for the corresponding patch.  Thoughts?

							Thanx, Paul

------------------------------------------------------------------------

commit a8bb9abbf3690e507d61efe8144cec091ce5fb02
Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Date:   Wed Jul 1 13:50:28 2015 -0700

    rcu: Pull out wait_event*() condition into helper function
    
    The condition for the wait_event_interruptible_timeout() that waits
    to do the next force-quiescent-state scan is a bit ornate:
    
    	((gf = READ_ONCE(rsp->gp_flags)) &
    	 RCU_GP_FLAG_FQS) ||
    	(!READ_ONCE(rnp->qsmask) &&
    	 !rcu_preempt_blocked_readers_cgp(rnp))
    
    This commit therefore pulls this condition out into a helper function
    and comments its component conditions.
    
    Reported-by: Peter Zijlstra <peterz@infradead.org>
    Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 4992bfd360b6..2afb8e8c5134 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -1904,6 +1904,26 @@ static int rcu_gp_init(struct rcu_state *rsp)
 }
 
 /*
+ * Helper function for wait_event_interruptible_timeout() wakeup
+ * at force-quiescent-state time.
+ */
+static bool rcu_gp_fqs_check_wake(struct rcu_state *rsp, int *gfp)
+{
+	struct rcu_node *rnp = rcu_get_root(rsp);
+
+	/* Someone like call_rcu() requested a force-quiescent-state scan. */
+	*gfp = READ_ONCE(rsp->gp_flags);
+	if (*gfp & RCU_GP_FLAG_FQS)
+		return true;
+
+	/* The current grace period has completed. */
+	if (!READ_ONCE(rnp->qsmask) && !rcu_preempt_blocked_readers_cgp(rnp))
+		return true;
+
+	return false;
+}
+
+/*
  * Do one round of quiescent-state forcing.
  */
 static int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in)
@@ -2067,11 +2087,7 @@ static int __noreturn rcu_gp_kthread(void *arg)
 					       TPS("fqswait"));
 			rsp->gp_state = RCU_GP_WAIT_FQS;
 			ret = wait_event_interruptible_timeout(rsp->gp_wq,
-					((gf = READ_ONCE(rsp->gp_flags)) &
-					 RCU_GP_FLAG_FQS) ||
-					(!READ_ONCE(rnp->qsmask) &&
-					 !rcu_preempt_blocked_readers_cgp(rnp)),
-					j);
+					rcu_gp_fqs_check_wake(rsp, &gf), j);
 			rsp->gp_state = RCU_GP_DONE_FQS;
 			/* Locking provides needed memory barriers. */
 			/* If grace period done, leave loop. */


  parent reply	other threads:[~2015-07-01 20:59 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-30 21:48 [PATCH RFC tip/core/rcu 0/5] Expedited grace periods encouraging normal ones Paul E. McKenney
2015-06-30 21:48 ` [PATCH RFC tip/core/rcu 1/5] rcu: Prepare for expedited GP driving normal GP Paul E. McKenney
2015-06-30 21:48   ` [PATCH RFC tip/core/rcu 2/5] rcu: Short-circuit normal GPs via expedited GPs Paul E. McKenney
2015-07-01 10:03     ` Peter Zijlstra
2015-07-01 13:42       ` Paul E. McKenney
2015-07-01 20:59       ` Paul E. McKenney [this message]
2015-07-01 10:05     ` Peter Zijlstra
2015-07-01 13:41       ` Paul E. McKenney
2015-07-01 13:48         ` Peter Zijlstra
2015-07-01 14:03           ` Paul E. McKenney
2015-07-02 12:03             ` Peter Zijlstra
2015-07-02 14:06               ` Paul E. McKenney
2015-07-02 16:48                 ` Peter Zijlstra
2015-07-02 19:35                   ` Paul E. McKenney
2015-07-06 14:52                     ` Peter Zijlstra
2015-06-30 21:48   ` [PATCH RFC tip/core/rcu 3/5] rcutorture: Ensure that normal GPs advance without " Paul E. McKenney
2015-06-30 21:48   ` [PATCH RFC tip/core/rcu 4/5] rcu: Wake grace-period kthread at end of expedited grace period Paul E. McKenney
2015-06-30 21:48   ` [PATCH RFC tip/core/rcu 5/5] rcu: Limit expedited helping to every 10 ms or every 4th GP Paul E. McKenney
2015-06-30 21:56     ` Eric Dumazet
2015-06-30 22:10       ` Paul E. McKenney
2015-07-01 10:07     ` Peter Zijlstra
2015-07-01 13:45       ` Paul E. McKenney
2015-07-01 19:30       ` Paul E. McKenney
2015-06-30 22:00 ` [PATCH RFC tip/core/rcu 0/5] Expedited grace periods encouraging normal ones josh
2015-06-30 22:12   ` Paul E. McKenney
2015-06-30 23:46     ` josh
2015-07-01  0:15       ` Paul E. McKenney
2015-07-01  0:42         ` Josh Triplett
2015-07-01  3:37           ` Paul E. McKenney
2015-07-01 10:12             ` Peter Zijlstra
2015-07-01 14:01               ` Paul E. McKenney
2015-07-01 14:08                 ` Eric Dumazet
2015-07-01 15:58                   ` Paul E. McKenney
2015-07-01 15:43             ` Josh Triplett
2015-07-01 15:59               ` Paul E. McKenney
2015-07-01 10:09       ` Peter Zijlstra
2015-07-01 10:55         ` Peter Zijlstra
2015-07-01 14:00           ` Paul E. McKenney
2015-07-01 14:17             ` Peter Zijlstra
2015-07-01 16:17               ` Paul E. McKenney
2015-07-01 17:02                 ` Peter Zijlstra
2015-07-01 20:09                   ` Paul E. McKenney
2015-07-01 21:20                     ` josh
2015-07-01 21:49                       ` Paul E. McKenney
2015-07-02  7:47                     ` Ingo Molnar
2015-07-02 13:58                       ` Paul E. McKenney
2015-07-02 18:35                         ` Ingo Molnar
2015-07-02 18:47                           ` Mathieu Desnoyers
2015-07-02 19:23                             ` Paul E. McKenney
2015-07-02 21:07                               ` Mathieu Desnoyers
2015-07-02 19:22                           ` Paul E. McKenney
2015-07-02  1:11                 ` Mike Galbraith
2015-07-02  1:34                   ` josh
2015-07-02  1:59                     ` Mike Galbraith
2015-07-02  2:18                       ` Paul E. McKenney
2015-07-02  2:50                         ` Mike Galbraith
2015-07-02  3:15                           ` 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=20150701205906.GQ3717@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=bobby.prani@gmail.com \
    --cc=dhowells@redhat.com \
    --cc=dipankar@in.ibm.com \
    --cc=dvhart@linux.intel.com \
    --cc=edumazet@google.com \
    --cc=fweisbec@gmail.com \
    --cc=josh@joshtriplett.org \
    --cc=laijs@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mingo@kernel.org \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.